The Changelog
Postgres at PlanetScale (Interview)
This episode of The Changelog features PlanetScale CEO Sam Lambert discussing PlanetScale's expansion into Postgres, its Metal/NVMe-based database infrastructure, and lessons from Vitess, GitHub, and open source. The conversation covers developer experience, technical marketing, workload isolation, security, and how AI is changing software development and database demand. Lambert also addresses competition with large cloud providers and PlanetScale's approach to building scalable, reliable database services.
本期 The Changelog 的嘉宾是 PlanetScale CEO Sam Lambert。他介绍了 PlanetScale 从 MySQL/Vitess 扩展到 Postgres 的计划，以及基于 Metal 和 NVMe 的数据库基础设施。对话重点包括面向开发者的体验、技术营销、工作负载隔离与安全，以及大规模数据库的可用性和扩展性挑战。Sam Lambert 还谈到开源、与 Amazon/Google 等云巨头的竞争，以及 AI 对软件开发、prompt 生成和数据基础设施的影响。整体上，他强调产品应围绕真实工程需求和开发者反馈演进，而不是只靠营销。
00:001:42:17
00:00
Well friends, we'll go back.
好了朋友们，我们回来了。
00:02
Summers over.
夏天结束了。
00:03
The show is back.
节目回归了。
00:05
Some of you just can't get enough.
有些人就是听不够。
00:07
Let's just say, you can't get enough.
这么说吧，你就是听不够。
00:09
And I'm here to bring it to you.
而我就在这里给你带来这一切。
00:11
This week I'm joined by a good old friend,
这周和我一起的是一位老友，
00:13
Sam Lambert, CEO of PlanetSkale.
Sam Lambert，PlanetSkale 的 CEO。
00:15
One of our first conversations happened back in January of 2022.
我们的第一次对话之一发生在2022年1月。
00:19
So we had a lot to cover in this plot.
所以我们这次要聊的内容很多。
00:21
We talk about the new $5 PlanetSkale Postgres,
我们谈到了新的5美元PlanetSkale Postgres，
00:25
the story behind Metal, spoiler they are not racking
Metal背后的故事，剧透一下，他们并不是在自行上架
00:29
and stacking their own MVMEs.
和堆叠他们自己的MVMEs。
00:31
They build on the hyperscalers bare metal machines.
他们构建在hyperscalers的bare metal机器上。
00:35
This is more a way.
这更多是一种方式。
00:36
The cursor demo they have where agents optimized a database
他们做的那个cursor demo，agents优化了一个database。
00:41
and shipped it safely through deploy requests.
并通过 deploy requests 安全地把它部署上线了。
00:44
Why you can roll back a 500 terabyte schema change
为什么你能 roll back 一个 500 terabyte 的 schema change
00:48
as fast as a one megabyte change
和 one megabyte change 一样快，
00:50
and Sam's genuinely unresolved wrestling match
以及 Sam 真正还没解决的内心挣扎，
00:53
over whether to open source Niki,
关于要不要 open source Niki，
00:56
their new Sharded Postgres product.
他们的新 Sharded Postgres 产品。
00:58
A huge, massive thank you to our friends,
超级超级感谢我们的朋友们，
01:02
and our partners over at fly.io.
以及我们在 fly.io 的合作伙伴们。
01:05
That is the home of changelaw.com.
那是changelaw.com的网站。
01:07
Learn more at fly.io.
更多信息，请访问fly.io。
01:11
Okay, let's get into it.
好了，我们进入正题吧。
01:23
Well friends, I'm here with the CTO of BuildKite
好吧，朋友们，我身边是BuildKite的CTO，
01:25
and one of the most challenging problems
以及最具挑战性的问题之一：
01:28
of modern era software development
现代软件开发中的
01:30
is continuous integration and continuous delivery.
持续集成和持续交付。
01:33
And so, Lockland Donald, BuildKite CTO,
所以，Lockland Donald，BuildKite的CTO，
01:37
what are you thinking about today's teams?
你怎么看待现在的团队？
01:39
The challenges they face, the speeds at which
他们面临的挑战，以及他们开发新 features、新 code 的速度。
01:42
they're developing new features, new code.
真的让人应接不暇。
01:45
It is just overwhelming.
你们大家怎么看？
01:47
How do you all think about that?
这个问题问得特别好，这正是现在每个人都在问的问题。
01:48
Such a good question is the question everyone's asking right now.
我们所有大客户现在都在问我们，比如说，
01:51
All of our big customers are asking us at the minute like,
你知道，如果我们今年能把 throughput 做到 5x 或 10x。
01:54
you know, if we five or 10X are three put this year
你知道吗，如果我们今年把 throughput 提高 5 倍或 10 倍
01:57
or 1,000X it, what breaks and when.
或者把它放大1000倍，什么会坏，什么时候坏。
02:00
And you know, my answer is kind of same as it's being
你知道，我的答案跟过去20年一样，瓶颈就在于
02:03
for the past 20 years, which is that the bottleneck
仍然是试着把那些 code changes 集成进来，然后 deploy，再检查它们能不能正常工作
02:06
is still trying to integrate those code changes in
然后在你不断往里面堆更多 code 的时候，让它们一直保持正常。
02:09
and then deploy them and check they work
我觉得很多 fundamentals 都是没变的。
02:12
and then keep them working as you keep throwing more
然后让它们保持运行，同时你不断往里扔更多
02:15
and more code at it.
更多的代码。
02:16
I think a lot of the fundamentals are the same
我觉得很多基本原理都是一样的。
02:18
but we're just 1,000Xing the speed of it
我们只是把它的速度给 1,000X 了。
02:21
and you know, that changes nearly every variable.
你知道，这几乎改变了每一个变量。
02:23
Yeah, for sure.
对，确实。
02:24
Okay, so where does BuildKite thrive?
好，那 BuildKite 到底在哪儿最吃得开？
02:27
What particular type of team or enterprise do you thrive in?
你们特别适合哪种类型的团队或企业？
02:33
The area that BuildKite has always thrived in
BuildKite 一直以来最吃得开的领域
02:35
is like this like fastest moving tech companies of the world.
就是全球迭代速度最快的那些科技公司。
02:38
So like we've been disproportionately successful
所以说，我们在这个领域里的成功是有点不成比例的。
02:41
in that small niche.
在那个小众领域里。
02:42
They're kind of Shopify class, Uber class, you know,
他们算是Shopify级别、Uber级别的那种，你懂的，
02:45
OpenAI class of folks that have this key problem
OpenAI级别的那群人，都有这个核心问题，
02:49
around iterating really, really fast.
就是迭代得非常非常快。
02:51
And you know, the thing about all of those
而且你知道，所有这些
02:53
fixes they all have subtly different needs,
修复方案，它们都有细微不同的需求，
02:56
subtly different problems.
细微不同的问题。
02:58
And so we've tended historically towards building
所以我们历来倾向于构建
03:01
like really well engineered LEGO blocks
就像是设计得非常精良的 LEGO 积木，
03:04
that scale like orders of magnitude
它们的 scaling 能力高出好几个数量级，
03:07
more than what our nearest competitor does.
比我们最接近的竞争对手还要高。
03:09
So you know, I think that that puts our system
所以，你知道，我觉得这让我们的系统
03:11
in this tension where, you know,
处于一种两难的境地，你知道，
03:13
you've got a spend some time assembling those building blocks
你得花些时间把这些积木块组装起来，
03:16
that those LEGO blocks to get the thing that you want
也就是那些 LEGO 积木，才能得到你想要的东西，
03:18
but the end of results is far and away
但最终结果却远远胜出。
03:21
more performant and scalable
更 performant 和 scalable
03:24
and the experience is better
而且体验更好
03:25
than what you get from something that's off the shelf.
比那些现成的方案要好。
03:27
So I think we've started from a position
所以我觉得我们的起点
03:29
of really well engineered LEGO blocks
是真正精心打造的 LEGO 积木
03:32
and then are kind of working backwards
然后我们有点像在倒推
03:34
towards kind of creating the thing that scales down
去创建一种能 scale down 到
03:37
to a startup that starts with one person
一个从一个人开始的创业公司。
03:39
and 10 agents next week.
然后下周10个agents。
03:41
Well friends, go to buildkite.com
好吧朋友们，去 buildkite.com
03:44
that's buildkitekyte.com.
就是 buildkitekyte.com。
03:48
You deserve better CI.
你值得更好的CI。
03:50
Engineer for the frontier, we are all facing
为前沿而工程，我们都在面对
03:52
trusted by the teams setting the pace.
受到那些设定节奏的团队信任。
03:55
Again, buildkite.com.
再次强调，buildkite.com。
03:58
Once again, buildkite.com.
再一次，buildkite.com。
04:18
Friends, I am back with an old friend.
朋友们，我回来了，带着一位老朋友。
04:20
It's been several years, about four years, Sam.
已经好几年了，大概四年了吧，Sam。
04:25
So did you build the pod?
所以你搞出那个pod了吗？
04:26
Sam Lambert, CEO of PlanetSkill,
Sam Lambert，PlanetSkill的CEO，
04:27
one of my most favorite databases that I don't use
是我最喜欢但没在用的数据库之一，
04:30
and I'm so sad, Sam, I am on the list
我真的很遗憾，Sam，我在名单上，
04:33
of trying out metal because it doesn't love
想试试metal，因为它对我这种
04:36
a good envy on me, right?
容易眼红的人不太友好，对吧？
04:37
Oh yeah, go have that performance,
哦对，就要那种性能，
04:39
especially with databases.
尤其是 databases。
04:41
Inside the server, disk, magic.
在 server 里面，disk，简直是魔法。
04:43
Yeah, I always felt like PlanetSkill was out of my realm
是啊，我一直觉得 PlanetSkill 超出了我的领域，
04:48
in terms of what I personally needed
就我个人需要来说的话，
04:50
because it's just so good.
因为它实在太强了。
04:51
Is that a good thing or a bad thing?
这是好事还是坏事？
04:52
What do you think, Sam?
你怎么看，Sam？
04:55
Luckily, for folks like yourself,
幸运的是，对于像你这样的朋友，
04:57
we now have a $5 postgres option,
我们现在有一个5美元的postgres选项，
05:00
planned scale postgres, very accessible.
planned scale postgres，非常容易上手。
05:02
The price of a kind of diet coke
价格大概是一杯diet coke，
05:04
in the fine but expensive city of San Francisco
在美好但昂贵的旧金山市，
05:07
and so like very well priced for people
所以对人们来说价格非常划算，
05:10
they just want to get going
他们只是想赶紧用起来，
05:11
and have some simple postgres
并且有一个简单的postgres。
05:12
but all the great features of PlanetSkill bundled in.
但 PlanetSkill 的所有强大功能都打包在里面了。
05:15
It's there, we're ready.
它就在那儿，我们准备好了。
05:17
Come through.
过来吧。
05:18
Give me a landscape of what PlanetSkill is today.
给我讲讲如今 PlanetSkill 的全貌吧。
05:21
The last time we talked was in January, 2022.
我们上次聊是在 2022 年 1 月。
05:24
Now I know the audience pays attention to you
我知道观众们一直都在关注你，
05:26
but just for those who may not have caught up since then
但为了那些从那之后就没跟上的人，
05:30
or they just want to lay off the land.
或者他们只是想先了解一下情况。
05:31
What is the landscape of PlanetSkill these days?
现在 PlanetSkill 的格局是什么样的？
05:34
I know you're in postgres world
我知道你是在 postgres 圈子里
05:35
and you've got sharding, you've got metal,
你那边有 sharding，也有 metal
05:37
I've mentioned that already
这个我之前提过了
05:38
but give me a zoom in to what you're doing.
但你再具体讲讲你现在在做什么吧
05:41
Yes, so a lot has changed since 2022.
是的，从 2022 年以来变化很大
05:44
That was like the very early days.
那时候还算是非常早期的阶段
05:47
Far earlier.
还要早得多
05:48
Yeah, I mean things have gone really well.
是的，我是说，一切都发展得很顺利。
05:52
Well, right now we have entered into the world of postgres
嗯，现在我们已经进入了 postgres 的世界，
06:00
which we announced mid last year.
这是我们去年年中发布的。
06:03
We have announced our sharded postgres product, Niki,
我们已经发布了我们的 sharded postgres 产品 Niki，
06:07
which is coming very soon
它很快就会推出，
06:09
and is already ramping up with customers.
而且已经在和客户快速推进了。
06:12
And for tests keeps growing
而且 for tests 还在持续增长，
06:14
and hosts now some of the world's largest databases.
并且现在托管着一些世界上最大的数据库。
06:17
We run very, very large sharded workloads
我们跑非常非常大的 sharded workloads，
06:20
for tons of very large AI companies.
为很多大型 AI 公司服务。
06:24
The AI era has been phenomenal for PlanetSkill.
AI 时代对 PlanetSkill 来说堪称现象级。
06:27
We've always presented ourselves,
我们一直把自己定位成，
06:30
especially in the latter years,
尤其是后期，
06:33
as just a big engineering team
就是一支大型 engineering team，
06:35
as a company that has a very thin abstraction over the top.
一家在顶层只有很薄 abstraction 的公司。
06:38
So our marketing is very technical.
所以我们的 marketing 非常 technical。
06:40
We're very well known for our blog posts
我们以我们的博客文章而闻名
06:42
and our technical content
以及我们的技术内容
06:43
and our incredible marketing team,
还有我们超棒的营销团队
06:45
all engineers that spend their time kind of curating
都是工程师，他们花时间算是做内容筛选
06:47
for our engineering audience.
为我们工程领域的读者
06:50
And none of us really foresaw
而且我们谁也没真正预见到
06:51
this kind of agent, agent like revolution, AI revolution.
这种 agent，agent 式的革命，AI 革命
06:55
But it turns out if you're just kind of hanging out
但结果就是，如果你只是随便待着
06:58
and building a database that scales
以及构建一个能 scale 的 database
07:01
to extreme scales for customers, AI companies need this.
为客户扩展到 extreme scales，AI 公司需要这个。
07:06
And we've been very fortunate to have
而我们非常幸运地有
07:09
a load of them kind of come onto the platform
一大批这样的公司来到了 platform 上
07:11
and grow incredibly quickly.
并且以惊人的速度增长。
07:13
And so it's been an amazing time here at the moment.
所以现在在这里真的是一段很棒的时光。
07:16
You get the, we're kind of in that era
你知道，我们有点像处于那个时代，
07:18
of like you build, you build, you want to get successful
就是你不断构建、构建，然后想获得成功。
07:20
and then you get all of the sort of success problems
然后你就会遇到所有那种成功带来的问题，压力和成功。
07:22
and stress and success.
我们在处理这些，同时也在 scaling 我们自己，
07:23
And we're dealing with that and scaling ourselves
scaling 我们的各种产品。
07:25
and scaling our various products.
现在公司真的是一个非常令人兴奋的时刻。
07:27
And it's just a very exciting time at the company right now.
我就是 24/7 到处跑，见客户，
07:30
I'm just running around 24, 7 meeting customers,
帮客户 migrate，聊天，你懂的，很有趣。
07:34
getting customers migrated, talking, you know, it's fun.
你是这么说的，对吧？
07:38
So you say that, right?
所以你是这么说的，对吧？
07:40
But then I go on your website to your homepage
但后来我去了你的网站，打开主页，用 command F 搜 AI，结果哪儿都找不到。怎么回事？你们在这个时代里，利用、构建、发展都做得这么好，但似乎并没有把自己定位成一个 AI database？
07:43
and I just search command F, AI.
然后我就用 command+F 搜了一下“AI”。
07:46
And I don't see it anywhere.
结果我哪儿都找不到它。
07:48
What is going on?
这到底是怎么回事？
07:48
How are you leveraging and building
你是如何在这样一个时代里运用、构建并发展得这么好，但看起来却并没有真正把自己定位成一个 AI database 呢？
07:51
and growing so well in this era,
如果你还记得我们在2022年聊过，那时候他们就是把 AI 硬贴到所有东西上。真的不怎么样。
07:53
but yet you're not really seemingly positioned
那不是销售说辞，对吧？
07:56
as an AI database?
作为一个AI数据库？
07:58
How was that?
那怎么样？
08:00
I don't know if there is a specific definition
我不知道现在有没有一个具体的定义
08:04
of an AI database right now.
关于 AI database 的。
08:06
There's definitely AI database platforms
确实有 AI database 平台
08:08
and we certainly have that.
而且我们当然有。
08:10
I just gave a talk at Kerser's conference yesterday
我昨天刚在 Kerser 的会议上做了个演讲
08:13
where I showed off 10 Kerser agents working
当时我展示了 10 个 Kerser agents 正在
08:16
on the plan scale platform.
plan scale platform 上工作。
08:18
The whole point of the talk is they went
这次演讲的核心内容是，他们做了大量的性能优化，
08:20
through a bunch of performance optimizations,
并且非常安全地部署上线了。
08:23
deployed them very safely.
而且你知道，关于Plan Scale最酷的一点就是，
08:25
And you know, this is the thing that's really cool
我们是为人类打造这个系统的。
08:26
about plan scale is we built this for humans
我在Twitter上被转发挺多的一张幻灯片里写的是，
08:29
and one of my slides that got shared quite a bit on Twitter
我们是为所有人类中那些flakies agents打造的。
08:31
was that we built this for the flakies agents of all humans.
你还记得我们在2022年聊过的那次吗？
08:34
If you remember when we spoke in 2022,
如果你还记得我们在2022年聊过的时候，
08:36
we were all about allowing people to ship
我们当时就是想让人们能够发布，
08:38
really, really quickly, safely,
非常非常快速且安全地。
08:39
like we have database branching online schema changes.
比如我们有 database branching、online schema changes。
08:43
We have rollbacks for those schema changes.
这些 schema changes 我们都有 rollbacks。
08:45
We block agents from pushing schemas into prod
我们会阻止 agents 把 schemas 推送到 prod，
08:49
that would break queries.
那些 schemas 会破坏 queries。
08:51
We have traffic control that isolates workloads
我们有 traffic control，可以隔离 workloads，
08:53
and does QOS for databases.
并为 databases 做 QOS。
08:55
Like we have all these like incredible features
就是我们有这么多超棒的功能
08:56
to make databases extremely safe for humans
让 databases 对人类来说极其安全
08:59
or their agents.
或者对他们的 agents。
09:02
The thing that communicates that we're there for AI
能说明我们是为 AI 而存在的东西
09:04
is the fact that the best AI companies already work
就是最好的 AI 公司已经在我们的产品上
09:06
on our product and use it.
工作并使用它。
09:08
We use AI behind the scenes to deliver insights
我们在幕后使用 AI 来提供 insights
09:12
to our customers and really good things for our customers.
给我们的客户，以及为我们的客户带来真正好的东西。
09:15
We don't need to necessarily scream the words,
我们不一定非得把“AI”这个词挂在嘴边反复喊。
09:17
AI over and over again.
你会看到那些其实根本不懂自己在干嘛的公司，
09:18
You see these companies that don't really have a clue
他们的营销团队被CEO告知，我们需要搞个AI的故事线，
09:20
what they're doing and their marketing team
我们就是需要，呃，
09:23
has been told by the CEO, we need like an AI narrative
然后他们就把AI贴到所有东西上。
09:25
we need like to just,
这真的不太好。
09:27
and they just slap AI on everything.
他们就是把AI到处乱贴。
09:31
And it's just like not good.
而且这样真的不行。
09:33
It's just insulting to the users
对用户来说简直是侮辱。
09:35
whether they're AI companies or not.
不管他们是不是AI公司。
09:37
Like things did not change that dramatically,
就像事情并没有发生那么大的变化，
09:39
but we do use AI and the AI companies use us
但我们的确在用AI，而AI公司也在用我们。
09:41
and that's the important thing.
这才是最重要的。
09:43
Can you give me a glimpse into what's required
你能让我大概了解一下需要什么吗？
09:47
behind the scenes to do what you said with cursor?
在幕后做你之前说的用Cursor那些事？
09:50
I think you said you had several agents provisioning, deploying
我记得你说过你有几个agent在provisioning、deploying。
09:54
to a point of skill.
到了一个熟练的程度。
09:55
Can you give me a zoom into that?
你能给我深入讲讲那个吗？
09:57
Yeah, and that's actually another part of the talk
是啊，那其实是演讲里另一个关键部分。
09:58
that was key.
比如我不希望这次演讲只是 plan scale 的一个大演示。
09:59
Like I didn't want the talk to be just one big demo
它不是销售宣传，对吧？
10:02
of plan scale.
但我真正想说的关键点是
10:03
It wasn't a sales pitch, right?
这又不是推销，对吧？
10:05
But the key points I was actually making
但我实际上想说的关键点是
10:08
is that infrastructure platforms are going to change now
就是基础设施平台现在要变了
10:10
to give insights to AI, right?
来给 AI 提供 insights，对吧？
10:12
Like, you know, there's so many database companies out there
就像，你知道，外面有那么多 database 公司
10:15
talking about the day one.
都在谈论第一天。
10:17
Oh wow, we're the fastest to spin up a database
哦哇，我们是最快把 database 跑起来的
10:19
and then like put it to sleep and never use it
然后就像让它休眠，再也不用了
10:21
because it's agentic or whatever.
因为它是 agentic 还是什么的。
10:22
That's like really not how software's built.
这真的不是软件构建的方式。
10:24
Like day one is day one.
就像，第一天就是第一天。
10:25
Great, like it's the point of creation, it's fine.
很好，就像它是创造的起点，没问题。
10:27
You pick any stack, it doesn't matter.
你选哪个 stack 都无所谓。
10:29
Plan scale, although you should start your startup
Plan scale，不过你应该创立你的初创公司，
10:32
on plan scale at $5 because then you don't have to do a switch
在 plan scale 的 $5 套餐上，因为这样你就不用切换了，
10:35
and like we have an unbelievable amount of startups
而且我们有数量惊人的初创公司，
10:37
and the new YC batch building on plan scale
而新的 YC batch 正在 plan scale 上构建，
10:40
for this reason.
正是因为这个原因。
10:41
They know the ones that you used to default to
他们知道你过去会 default 到的那几个
10:43
just fall apart at basically no scale.
基本上还没有什么 scale 就崩了。
10:46
And so we really focus though on like day one
不过我们真的很关注，像第一天的时候
10:53
you gotta get going, you gotta build.
你得先动起来，你得先构建。
10:54
But it's like day one thousand
但就像到了第一千天
10:55
is where the real problems show up
才是真正问题出现的时候
10:56
when it comes to maintaining, right?
尤其是在维护的时候，对吧？
10:58
Like any infrastructure is like this,
就像任何 infrastructure 都是这样，
11:00
is like a big living city
就像一个巨大的活城市
11:03
and it's complicated.
而且它很复杂。
11:07
There is so many interacting services
有太多互相交互的 services
11:11
and like you'll get your most productivity from AI
还有，你会发现从 AI 获得最大的生产力
11:14
working with existing code bases
是在已有的 code bases 上工作
11:17
that need constant pruning and evolution,
这些 code bases 需要不断的修剪和演化、
11:20
maintenance and care which humans have to do.
维护和照料，而这些必须由人类来做。
11:22
Agents can go and do this.
Agents 可以去做这些。
11:23
And so what we do now is like we generate prompts
所以我们现在做的，就是生成 prompts
11:28
and instructions for agents
以及给 agents 的 instructions
11:30
and it is us scaling our expertise
也就是我们在扩展自己的专业能力
11:32
through infrastructure services
通过 infrastructure services
11:34
and like giving it to the customer very directly
然后非常直接地把它交给客户
11:36
and giving it to their agents very directly
也非常直接地交给他们的 agents
11:38
and that takes an immense amount
而这需要大量的
11:40
of infrastructure behind the scenes.
幕后的 infrastructure。
11:41
So we take traces and log the performance characteristics
所以我们抓取 traces，并记录每个查询的 performance characteristics。
11:48
of every query that hits our platform
这些查询打到我们平台上，
11:51
which is tens of millions of queries a second
每秒有数千万个 queries。
11:53
and we have to stream those data warehouse them.
我们必须 stream 这些数据，然后 data warehouse 它们。
11:57
We have to like apply our various algorithms
我们还得应用各种 algorithms，
12:02
to detect whether they're using indexes or not.
来检测它们是否使用了 indexes。
12:04
Like we've had to edit Postgres core,
就像我们不得不去修改 Postgres core，
12:07
my SQL core to go and do this
还有 MySQL core，才做到这一点。
12:09
all to like generate a three line prompt for an agent.
就好像是给 agent 生成一个三行的 prompt。
12:13
And so you turn this like depth
然后你把这个深度
12:15
of infrastructure and expertise
的 infrastructure 和专业知识
12:17
into this insight for agents
转化成给 agent 的洞察
12:19
and these emergent behaviors then get observed
然后这些 emergent behaviors 被观察到
12:22
and become the next instruction for agents.
就变成了 agent 的下一个指令。
12:25
And it's an incredible amount of infrastructure
而且这背后有大量的 infrastructure
12:26
and plumbing behind the scenes
以及幕后的 plumbing。
12:29
that's required our platform to be
这就要求我们的platform非常复杂且管理良好，而我们也做到了。然后它就变成一个非常简单的流程：我只需要上台说“optimize this database”，然后这些agents当然就会去处理，并从system那里得到推荐。
12:32
very complicated and well managed
非常复杂而且管理得很好
12:34
and we've achieved that
而且我们已经做到了
12:34
and it gets this very simple flow
然后它就变成了一个非常简单的流程
12:36
where I could just go on stage
我只需要走上台
12:38
and say optimize this database
然后说优化这个database
12:40
and curse this agents went away
然后当然这些agents就消失了
12:42
and got recommendations from the system.
然后从system那里得到了建议
12:44
We are very early to this.
我们做这个非常早。
12:46
Like there's very few infrastructure providers
就像，很少有 infrastructure provider 是这样构建的。
12:49
that have built this way.
这是因为我们很早就意识到，人类会把数据库搞坏，而我们必须让它们变得合理。
12:50
It's because we were early to understanding
我们的 developer experience 优于大多数公司，
12:52
that humans break databases
就是说人类会把数据库搞坏，
12:53
and we have to make them make sense.
而我们得让它们变得合理。
12:55
Our developer experience is superior
我们的developer experience更优越。
12:57
to most companies out there,
对大多数公司来说，
12:59
especially in the database space
尤其是在 database 领域
13:01
and so it's not hard for people
所以对于那些
13:02
that have built developer experience for 20 years
已经构建了20年 developer experience 的人来说
13:05
to figure out how to get agents
要弄清楚怎么让 agents
13:08
that same experience.
获得同样的体验并不难。
13:09
Yeah, and so that's where we go.
是的，所以这就是我们的方向。
13:11
Take me to the technicals of this.
带我去看看这其中的技术细节。
13:13
So you have cursor in this demo
所以，在这个 demo 里你有 cursor。
13:16
is just saying in your talk,
只是在说，在你的演讲中，
13:18
optimize this database.
optimize这个database。
13:20
What happens?
会发生什么？
13:22
Is cursor talking to MCP?
Cursor是在和MCP对话吗？
13:24
Is it an API?
它是一个API吗？
13:25
Is it a CLI installed into cursor?
它是安装到Cursor里的一个CLI吗？
13:28
Help me understand.
帮我理解一下。
13:29
What enabled that query and that question happen?
是什么让那个query和那个问题发生的？
13:32
Yeah, it loads up the plan scale scales
嗯，它加载了 Plan Scale scales
13:35
and the plan scale MCP
还有 Plan Scale MCP
13:37
and asks plan scale, what do you have for me?
然后问 Plan Scale：你有什么给我的？
13:39
What insights do you actually have for me?
你实际上有什么见解给我？
13:42
Rather than just,
而不是只是，
13:43
like we're in the stage of most people
就像我们处在大多数人
13:45
are just like pointing agents
只是把 agents 指向
13:47
at like massive streams of logs or just a code base
大量 logs 流或者一个 code base 那样的阶段
13:50
and it's like very difficult
而且这非常困难
13:53
to get the agents to sift through that
让 agents 去筛选那些内容
13:55
and find the right thing.
然后找到正确的东西
13:56
And so we make sure that it gets very narrow instruction,
所以我们确保它会得到非常具体的 instruction
13:59
make this, add this index, remove this index,
做这个，添加这个 index，移除这个 index
14:03
scale this cluster, like all these various things
scale 这个 cluster，诸如此类的各种事情
14:05
that the agents can actually act on
agents 可以真正据此行动
14:08
and they will take the MCP and they're going to do the code
然后他们会拿着 MCP 去写 code
14:11
and then they will also hit the API
然后他们也会去调用 API
14:13
to go and spin up a dev branch, to go and do the work on.
去启动一个 dev branch，然后在这个分支上开展工作。
14:16
And that's where the next chapter comes in, I guess,
而这就是下一章的内容了，我想，
14:18
which is they get the instruction,
也就是他们收到指令，
14:21
they figure out what to do with the code base
他们弄清楚要对 code base 做什么，
14:23
and then they need a really safe isolated way
然后他们需要一个非常安全且隔离的方式
14:25
to test their work.
来测试他们的工作成果。
14:27
And so they will use plan scale branches
所以他们会使用 plan scale branches。
14:29
to have an isolated dev-like environment,
要有一个隔离的 dev-like environment，
14:33
sorry, production-like environment for dev
抱歉，是 production-like environment for dev
14:34
where they can't break anything.
在那里他们不会弄坏任何东西。
14:36
And then the thing that we truly know
然后我们真正知道的事情是
14:38
when our stars is we get that change into production
当我们的 stars 是，我们把那个 change 放到 production
14:41
in an incredibly safe way.
以极其安全的方式。
14:43
So there's like a gate, it's the deploy request, right?
所以有一个 gate，就是 deploy request，对吧？
14:46
Like it's like the pull request,
它就像 pull request 一样，
14:47
the same people that designed the pull request
就是设计 pull request 的那些人
14:50
where you are doing the deploy request here.
也就是你在这里发起 deploy request 的地方。
14:53
So like there's various things,
所以就像有很多事情，
14:55
there's like, first of all,
首先吧，
14:56
we stop agents screwing things up before they get to.
我们在 agents 搞砸之前就阻止它们。
14:59
So we block you, I'll give you a example, right?
所以我们阻止你，我给你举个例子，对吧？
15:02
Like an agent has screwed up and they want to,
就像一个 agent 搞砸了，然后他们想要，
15:04
and I demoed like we poison some agent instructions
然后我演示过，比如我们会 poison 一些 agent instructions。
15:07
to go and do the wrong thing.
去做错事。
15:09
And they got rejected from doing it.
然后他们被拒绝了。
15:11
We made it then let one through
我们故意让一个通过了。
15:12
so we could show what happens
这样我们就能展示会发生什么。
15:13
even if the worst of the worst happened.
即使最最糟糕的情况发生了。
15:15
And I'll tell you that's not in a second.
而且我告诉你，那可不是一秒钟的事。
15:16
But basically like, if you go to drop a column in prod,
但基本上，如果你要去 prod 里 drop 一个 column，
15:20
which is a common thing.
这是很常见的事。
15:21
Like when you're doing a refattering,
比如当你在做 refactoring 的时候，
15:22
you want to get this like rid of this data,
你想要去掉这些 data，
15:24
you don't want it in the database.
你不想让它留在 database 里。
15:26
If you have referenced that column
如果你在过去 24 小时内用 query 引用过那个 column，
15:27
within the last 24 hours with a query,
我们就会阻止你 deploy 它，
15:29
we block you from deploying it
并且告诉你哪些 queries 引用了那个 column。
15:30
and we show you what queries have referenced that column.
这是人们非常常见的 outage 来源。
15:32
It is a very common source of outages for people.
这是人们遇到 outage 的一个非常常见的来源。
15:36
It is to think they are no longer accessing that column
就是说，你以为他们不再 access 那个 column 或者 table 或者什么之类的了，但其实他们还在 access。
15:39
or table or whatever and they are.
然后我们会阻止这种情况发生，而且我们每小时都在为客户预防 outage，就像我们 block 了很多很多坏 change 进去。
15:41
And we will stop that happening
然后我们就遇到了一个场景：我们以为它会通过，结果它真的把 column drop 掉了。
15:43
and we prevent outages every hour for customers
然后我们还有另一个 feature 来修复这个问题。
15:46
when they are like we block many, many bad changes going in.
当他们说“我们阻止了很多很多不好的 change 进入”的时候。
15:51
We then have the scenario where we thought
然后我们就遇到了这样的场景：我们原本以为
15:52
loud that to go through and it did drop the column.
放行了它，结果它确实 drop 了那个 column。
15:55
And then we have another feature for fixing that,
然后我们还有另一个 feature 用来修复那个问题，
15:57
which is we fire anomalies to say,
也就是说，我们触发 anomalies 来表示，
15:59
this is broken, this you have broken
这个坏了，这是你搞坏的，
16:02
your database queries and our failing.
你的 database queries 和我们的失败。
16:04
And we can roll that change back instantly
而且我们可以瞬间 roll back 那个变更，
16:07
and it brings the old version of the table back
然后把旧版本的 table 带回来，
16:11
with zero data loss.
实现 zero data loss。
16:12
Like even if we've done intermediate changes
比如即使我们做了中间变更，
16:15
to the new table,
到新 table 上，
16:16
they are still going to be in the old table
它们仍然会在旧的 table 里面。
16:18
and we can bring it back.
而且我们可以把它恢复回来。
16:19
And that is again, nobody else does this in a platform.
而且再说一遍，没有别的 platform 能做到这一点。
16:21
We are the only people,
我们是唯一一家，
16:22
only platform that has this capability for databases.
唯一对 databases 有这种能力的 platform。
16:25
And so you can see it's these small sharp tools
所以你看，就是这些短小精悍的 tools，
16:29
of a tool chain that get an agent
来自一个 tool chain，能让 agent
16:31
from dev to production safely can undo changes.
从 dev 安全地到 production，并且可以 undo changes。
16:36
These changes are all attributed to the agent.
这些变化都归因于这个 agent。
16:40
Like you go into a partscale graph
比如你进入一个 partscale graph，
16:42
and you'll see a line that goes down the graph
你会看到一条线在图上往下走，
16:44
that shows when it was deployed.
那就是它部署的时间点。
16:46
And you can then like see exactly what's gone wrong.
然后你就能很清楚地看到到底哪里出了问题。
16:49
And then once you've gone and made like changes
等你做了一些改动之后，
16:52
and it looks good and it's like fine,
看起来不错，也挺好的，
16:54
we generate more for you to do if it needs to do more.
如果它还需要更多处理，我们就会再生成一些任务给你。
16:56
And so it's this constant loop
所以这就是一个不断的循环
16:58
that goes round and round.
一直转来转去。
17:00
And so now like agents are building planet scale for us
那么现在呢，agents 正在为我们构建 planet scale
17:04
and deploying their own work for us
并且为我们部署它们自己的工作
17:06
in extremely safe environments
在极其安全的环境中
17:09
against like we see customers doing schema changes
比如说我们看到客户做 schema changes
17:12
on 500 terabyte tables
在 500 terabyte 的表上
17:15
and they just get it done.
然后它们就搞定了。
17:16
And by the way,
而且，顺便说一句，
17:18
you can roll back a 500 terabyte schema change
你可以 roll back 一个 500 terabyte 的 schema change，
17:21
in exactly the same time as a one megabyte schema change.
所需的时间和一个 1 megabyte 的 schema change 完全一样。
17:24
Like it will be exactly the same.
真的会完全一样。
17:29
That's pretty astounding
这真的很惊人，
17:31
to even think about this kind of changes.
光是想想这种变化都让人觉得不可思议。
17:32
And I think what I think about
然后我在想的是，
17:36
is the primitives you've probably built upon.
就是你可能已经基于的那些 primitives。
17:38
So you've been doing this for a while.
所以你做这个已经有一段时间了。
17:40
Developer experience focused for so long.
我们专注于 developer experience 已经这么久了。
17:43
And so that's not just magic wand waving.
所以那不只是挥挥魔杖而已。
17:46
That's literally hey, we're developers,
这真的是——嘿，我们是开发者，
17:47
we use this every single day.
我们每天都在用这个东西。
17:49
We're improving it based upon our developer needs.
我们基于自己的开发者需求来改进它。
17:52
Just even your data warehouse
甚至你的 data warehouse，
17:54
where you pull logs into to gain insights
就是你把 logs 拉进去获取 insights 的地方。
17:56
is a clue to that for me.
对我来说，这是一个线索。
17:59
But you have primitives you've built on.
但是你有这些 primitives，你是基于它们构建的。
18:00
So branching, you've got these safeguards,
所以 branching，你有这些保障措施，
18:02
you call them gates.
你称它们为 gates。
18:03
Help me understand how those gates come into play
帮我理解这些 gates 是如何发挥作用的，
18:07
to give that insight but also block you
既能给你那种洞察力，又能阻止你，
18:12
if you're going the wrong direction.
如果你走错了方向。
18:13
How those gates appear to the developer
这些 gates 对开发者来说是如何呈现的。
18:16
when it comes to DX?
那说到 DX 呢？
18:17
They look very tactile
它们看起来很有实感
18:18
and they look like we expect from other products, right?
而且它们看起来就像我们从其他产品里期待看到的那样，对吧？
18:22
Like the deploy request is really meaningfully meant
比如 deploy request 真的刻意做成它所对应的 pull request 的样子
18:24
to look like the pull request it communicates
它会展示你 schema 的 diff
18:26
it shows you a diff of your schema
即使你并没有给我们发 diff
18:28
even though you didn't send us a diff.
你只是在一个 branch 上做了 schema 变更
18:30
You did schema changes on a branch
你在一个 branch 上做了 schema changes，
18:32
by using the normal DDL that my SQL Postgres supports
通过使用我们的 SQL Postgres 支持的普通 DDL
18:36
and then we show you a diff.
然后我们给你展示一个 diff。
18:38
It should feel like your other tools.
它应该感觉像你其他的工具。
18:41
And then for an agent, the same thing
然后对于 agent，同样的事情
18:44
just looks and feels native to them
对他们来说看起来和感觉都很原生
18:46
through MCP servers or APIs.
通过 MCP servers 或 APIs。
18:48
And like we have passed remote for our CLI,
而且就像我们已经为 CLI 提供了 remote 一样，
18:51
so agents can arbitrarily call API endpoints
所以 agents 可以任意调用 API endpoints。
18:55
that we have set up for our own use
我们自己用的那些东西
18:57
so they get these like, these ability,
所以他们就有了这些，这些能力
19:00
like in the talk I talk about,
就像我在演讲里谈到的
19:02
the unix philosophy of these independent tool sets
就是这些独立 tool sets 的 Unix 哲学
19:05
that humans or agents can gather together
人类或 agents 可以把它们集合起来
19:09
to get what they want and what they need
去获得他们想要和需要的东西
19:12
and compose their own journey.
然后组合出自己的旅程
19:13
Like you might not be doing a schema change
比如说你可能不是在做一个 schema change
19:16
but people spin up a planned scale branch
但是人们每次做 CI 的时候都会启动一个 planned scale branch
19:18
every time they do CI so that CI runs against the branch
每次做 CI 的时候，就让 CI 跑在那个 branch 上
19:22
and so it's basically running against a copy of production
所以它基本上就是跑在 production 的一个副本上
19:25
or a system that behaves just like production.
或者一个行为和 production 完全一样的系统
19:29
And it's just about getting out of the way
它就是为了不挡你的路
19:33
but blocking you from doing the wrong things.
但是会阻止你去做错误的事
19:35
It does another way of thinking about this
这是另一种思考方式
19:36
which is I've been in databases for a very long time
就是我在 databases 这一行已经待了很久了
19:39
on very large products and services.
在非常大的产品和服务上。
19:42
And I basically took every outage that I'd ever seen
然后我基本上把我见过的每一个 outage 都拿来，
19:46
and we're undoing the ability for that to happen.
然后我们正在让这种事情不再可能发生。
19:50
I think for me, I'm curious what
对我来说，我很好奇的是
19:53
is the most exciting thing happening right now.
现在最令人兴奋的事情是什么。
19:55
Obviously you've got a boom
显然，现在有一波热潮，
19:57
and we've got a boom in the industry with AI
而我们行业因为 AI 也迎来了一波热潮，
20:00
and there's a lot of speculation of bubble, bursting,
而且有很多关于泡沫破裂的猜测。
20:03
just a lot of what we call a fear uncertainty
就是很多我们所谓的恐惧、不确定性
20:07
and doubt, fud, obviously.
和怀疑，也就是 FUD，显然。
20:09
But aside from what might be the obvious answer,
但除了那个显而易见的答案之外，
20:14
what is the most exciting thing happening
对你个人而言，在 PlanetScale 中正在发生的最令人兴奋的事情
20:16
for you personally in planet scale?
是什么？
20:20
So the thing I'm most excited about right now
所以我现在最兴奋的事情
20:23
is watching Niki get built, which is our sharded postgres solution
是看着 Niki 建起来，这是我们的 sharded Postgres 解决方案
20:26
like the test for postgres.
就像是 Postgres 的测试。
20:28
In the last five years, we have spent so much time and work
在过去的五年里，我们投入了大量的时间和精力
20:33
like evolving for tests from something
就像是为此而从某个东西演进
20:35
that was a very raw open source project
那是一个非常原始的 open source 项目
20:37
that was being used at very few companies
只有很少的公司在使用
20:39
to now something that I can generally use to cloud product.
到如今，我通常可以把它用作 cloud product
20:42
And people interact, it's like impossible now
而且，现在人们几乎不可能
20:45
to be a consumer in the US or a knowledge worker in the US
在美国做一个消费者或知识工作者
20:48
without interacting with products that run on planet scale.
而不与运行在 planet scale 上的产品互动
20:51
We have cursor, they use us to build,
我们有 cursor，他们用我们来构建，
20:54
to just run gigantic sharded databases
就是为了运行庞大的 sharded databases，
20:58
that back all of the product interactions that cursor
来支撑 cursor 所有的 product interactions，
21:02
builds including origin.
包括 origin。
21:03
Then you products running on us already.
然后你们的产品已经跑在我们上面了。
21:06
We have kick.com,
我们有 kick.com，
21:08
pokes interaction, lime scooters,
pokes interaction、lime scooters，
21:11
every square swipe, whenever people are buying coffee
每一次 square swipe，每当有人买咖啡时。
21:14
or on square terminals, that's going through planet scale,
或者在 Square 终端上，那就是通过 PlanetScale 来做的，
21:18
cash app, just extremely large consumer products
Cash App，就是那种体量极大的消费产品，
21:23
that need a very scalable database
它们需要一个非常 scalable 的 database，
21:25
and they need sharding, run on planet scale.
而且它们需要 sharding，跑在 PlanetScale 上。
21:28
And now we're bringing that to postgres.
现在我们要把这个带到 Postgres。
21:31
It's been phenomenal to watch a team of experts
看着这支专家团队，真是让人惊叹，
21:33
like the world's foremost engineers on this subject
就像在这个课题上世界最顶尖的工程师，
21:37
and field that have the experience
以及在这个领域里经验丰富的人。
21:39
of running these gigantic workloads in production
在 production 中运行这些巨型 workloads
21:41
on our cloud.
在我们的 cloud 上。
21:42
Now bringing the same to postgres through Niki.
现在通过 Niki 把同样的方案带到 postgres。
21:45
Not just how quickly they're building
不仅仅是他们构建得有多快
21:48
and who is rolling out.
以及谁在推出。
21:49
Like right now we are moving another set
比如现在我们正在迁移另一批
21:52
of just huge postgres workloads
超大的 postgres workloads
21:54
at very, very well-known companies on Niki.
在非常非常知名的公司里，基于 Niki。
21:57
And then watching that go is just phenomenal.
1. 然后看着它发展起来，真是令人惊叹。
21:59
So like bringing, sharding,
2. 所以就像把 sharding，
22:01
and the sort of traditional practices of database scaling
3. 以及 database scaling 的传统实践，
22:05
to the new audience is the postgres audience.
4. 带给新的受众，也就是 Postgres 的用户群体。
22:08
That is full of very dynamic startups.
5. 那里充满了非常有活力的初创公司。
22:10
Postgres is definitely the database of choice
6. Postgres 绝对是新公司和 AI 公司的
22:12
for new companies and AI companies.
7. 首选 database。
22:14
And so I'm just extremely excited to see the demand.
8. 所以我看到这样的需求，真是无比兴奋。
22:17
Like we actually, there's so much demand for this product
就是，实际上这个产品的需求量特别大
22:20
that we have companies just,
以至于我们有一些公司，就是
22:23
I had to just very gently tell the CEO of a startup.
我只好非常委婉地告诉一个startup的CEO。
22:25
Like they told me it was unfair
他们跟我说这不公平
22:27
that they weren't being given first access to this product
因为他们没有被给予这个产品的优先使用权。
22:30
and I just had to say, you know,
然后我只好说，你懂的，
22:31
we have a queue and fortunately we're working
我们有一个排队名单，而且幸运的是我们现在正在
22:33
with a lot of people right now
和很多人合作。
22:34
and we're trying to rate limit the demand.
我们在尝试对需求做 rate limit。
22:35
It's, I've never seen anything like it.
这，我从来没见过这种情况。
22:37
And so it isn't it?
所以，不是吗？
22:39
Yeah, it's an effort in like technical mastery.
是啊，这算是一种技术掌控上的努力。
22:43
And it's also incredible to see
而且看到这一点也特别不可思议。
22:46
that we've landed in this part of the market
我们竟然进入了这个市场领域。
22:47
and there's so much need for the thing
而且对这个东西的需求量巨大。
22:48
that we are so good at.
这正是我们最擅长的。
22:53
So you have quite a queue for Niki.
所以 Niki 的 queue 挺长的。
22:55
And would you, I'm coming from the outside
还有，你愿不愿意——我是从外面来的，
22:59
to help me shape a lens here,
来帮我建立一个视角，
23:01
but is Niki synonymous with Votes
但 Niki 和 Votes 是同义的吗？
23:05
in that Votes is my sequel?
即 Votes 是我的续集？
23:08
Niki is postgres.
Niki 是 postgres。
23:09
Is that, are they?
是这样吗，它们是同一个吗？
23:10
Correct.
没错。
23:11
I would say that Niki is more evolved
我会说，Niki 更进化
23:14
in the sense that if you're doing a do-over
意思就是如果你要重来一次
23:16
on a 50, like Votes is 15 years old now,
在 50 上，比如 Votes 现在已经 15 年了
23:19
getting a do-over.
要重来一次
23:20
Yeah.
对
23:21
What do we do now?
我们现在怎么办？
23:22
I don't want to look too much of the architecture right now,
我现在不想过多地看 architecture
23:24
but like, I mean, we certainly,
但就像，我是说，我们当然，
23:26
we learned there was like at YouTube,
我们了解到，就像在YouTube，
23:28
there was, YouTube did not design it
就是说，YouTube并没有把它设计成
23:29
to be a general purpose system.
一个通用系统。
23:31
It was designed purely for YouTube.
它纯粹是为了YouTube自身而设计的。
23:34
And so there was mistakes that were made
所以确实犯了一些错误，
23:36
or not even mistakes.
或者甚至不能说是错误。
23:37
Actually, I will give that I will say
实际上，我要承认的是，
23:38
they were not in any way mistakes.
它们根本不是错误。
23:40
No, I think it was more, I think it's actually,
不，我觉得更多的是，我觉得实际上，如果他们真的去做了那项工作，我反而会更不尊重他们，因为在当时那会是一种 overoptimization，他们做的事情只是对 YouTube 有效，也就那样了。测试出来的东西会随着时间演变，后来被很多其他公司使用。
23:42
I would have less respect
那我就会少一些尊重，
23:44
if they'd have gone and done that work,
要是他们当时真去做了那个工作的话，
23:45
because at the time that would have been an overoptimization,
因为那时候那样做就是 overoptimization 了，
23:48
they did things that worked for YouTube
他们做的那些事对 YouTube 是有效的，
23:50
and that was about it.
也就仅此而已。
23:51
Votes tested evolve over time
经过测试的 votes 会随时间演变。
23:54
to be used by a bunch of other companies,
供一堆其他公司用，
23:56
but they were still hyperscale
但她们仍然是 hyperscale
23:57
and they were pretty custom.
而且它们相当 custom。
24:00
But Niki is built with 15 years of a test knowledge,
但 Niki 是建立在 15 年的测试知识之上，
24:04
five years of planet scale knowledge,
五年的 planet scale 知识，
24:05
running it directly on the cloud
直接在 cloud 上运行，
24:07
and learning everything we'd learned
并且学习我们学到的一切，
24:09
and building general purpose systems
构建通用系统，
24:10
and then just getting all of the different things
然后就是把所有不同的东西整合起来。
24:13
we want to add new features, new user experience on top
我们想在此基础上添加新功能、新的用户体验。
24:17
and built by the union of all those things.
并且是由所有这些事物的总和构建而成的。
24:22
It's an amazing piece of software.
这是一款非常了不起的软件。
24:28
Where does, I understand the test
那么，我理解，在那里测试和我的SQL处于什么位置？
24:30
and my SQL in the relationship there.
帮我理解一下，Niki，
24:32
Help me understand, Niki,
这是专有的东西吗？
24:34
is this something that is proprietary
我们到底能不能亲手用上这个软件，
24:37
will we ever have our own hands on the software,
我们到底能不能亲手用上这个软件，
24:40
the, I mean, it's not open source, is it?
那个，我是说，它不是 open source 的，对吧？
24:43
We have not decided yet.
我们还没有决定。
24:45
I'll be very transparent.
我会非常透明。
24:46
We have not yet settled on whether we will open source
我们还没决定要不要 open source Niki。
24:49
Niki or not.
不幸的是，我们看到一些人，
24:50
Unfortunately, we've seen some people,
一些公司，在 database 领域非常知名的，
24:53
some companies, very popular ones in the database space
表现得非常非常糟糕，尤其是在
24:57
be very, very misbehaving when it comes
在涉及到……的时候会非常非常不守规矩，
25:00
to other companies open source projects,
对其他公司的 open source 项目，
25:02
claiming credit for them, forking them,
声称功劳是自己的，直接 fork 它们，
25:06
like hiring someone that's 15th on the retainer list
比如雇一个在 retainer 名单上排第15的人，
25:09
and saying they now maintain that project,
然后说他们现在维护那个项目。
25:11
just things that are very bad
这些都是非常恶劣的行为，
25:13
and that didn't really used to happen
这种事在 open source 早期基本不会发生，
25:14
in the early days of open source,
但现在有一代新人。
25:16
but there's a new generation of people
但有一代新人，
25:17
that don't really understand these things and...
那些并不真正理解这些事情的人，而且...
25:20
I think it's so cutthroat.
我觉得这太残酷了。
25:21
It is true cutthroat, yeah, they use their position
这确实很残酷，对，他们利用自己的地位
25:23
and they use their leverage to try and do these things
而且他们利用自己的影响力来试图做这些事情
25:25
and their voice and, you know,
还有他们的话语权，你知道的，
25:27
I don't want you to want to see that happening
我不希望你想要看到这种情况发生
25:29
and then there's obviously Amazon
然后显然还有亚马逊
25:30
who are an abysmal player in the world of open source
在 open source 领域，他们是个非常糟糕的角色
25:32
no matter how much they try and whitewash that
不管他们怎么试图洗白这一点
25:34
and a terrible partner.
而且是个糟糕的合作伙伴。
25:35
And so working with them and seeing them,
所以，一边和他们合作，一边看着他们的时候，
25:39
well, I mean, like they're saying, they have,
嗯，我是说，就像他们说的，他们有，
25:42
they have a charted product called Limitless.
他们有一个在规划中的产品，叫 Limitless。
25:44
It's...
就是...
25:46
Amazing name.
名字真棒。
25:47
Yes, I saw this amazing tweet once where someone said,
是的，我见过一条很棒的推文，有个人说，
25:50
we have had so much fun finding Limitless's limits
我们找Limitless的极限找得超开心
25:53
in production, it was great.
在production里跑的时候，简直太棒了
25:55
So the largest database that ever ran on Limitless
所以，曾经跑在Limitless上最大的database
25:58
runs on planet scale,
是在planet scale上跑的
26:00
they had an average two outages a day on Limitless.
他们在Limitless上平均每天有两次outage
26:03
And so now we have Amazon telling people
然后现在Amazon就跟大家说
26:06
that the new version will be much more like for tests
新版本会更像是一个用于测试的东西
26:09
and it's like, well, it's open source
然后就是，嗯，它是open source的
26:10
so they can go and try and do that
所以他們可以去試試看
26:11
and so Amazon, the bounds at which Amazon will try
所以 Amazon 試圖
26:15
and screw over small companies
搞垮小公司
26:16
or the world of open source is unlimited.
或整個 open source 世界，是沒有界限的。
26:19
And so when it's been proven that we are better engineers
所以當我們已經證明自己是更好的工程師
26:23
in databases than they are now,
在 databases 方面比他們現在更強，
26:24
it wasn't the case back in the day,
以前可不是這樣，
26:26
we weren't around then,
那時候我們還不在。
26:27
but the early teams that built the early databases
但是最初构建那些早期 database 的团队
26:30
at Amazon were phenomenal engineers,
在 Amazon 的都是非常出色的工程师，
26:32
they just don't work there anymore.
只是他们已经不在那里工作了。
26:34
Also, the largest database that ever ran on
还有，那个曾经运行在
26:36
uncharted Aurora now runs on planet scale.
uncharted Aurora 上的最大 database 现在运行在 planet scale 上。
26:38
So planet scale, like it's just our bread and butter
所以 planet scale 就像我们的家常便饭，
26:41
all day long, we move databases off of Amazon's products.
我们整天都在把 database 从 Amazon 的产品上迁移走。
26:46
And they behave very poorly.
而且它们表现得很差。
26:47
As partners with that, they tell people
作为那方面的合作伙伴，他们告诉人们他们会让自己的 software 变得像我们一样。
26:49
they're gonna make their software like us.
所以我们不知道 Niki 的 open source 未来会怎样，这需要一些深思熟虑，
26:51
And so we do not know what the future of open source
但这是因为像那样的人以及 database 领域的其他一些 startups 所造成的。
26:54
for Niki will be, it will take some deliberation,
你知道，我们到处都能看到，
26:57
but it is because of people like that
但正是因为那样的人，
26:59
and some other startups in the database space
以及数据库领域的一些其他创业公司，
27:01
that it's causing.
它正在造成的。
27:02
You know, we're seeing everywhere,
你知道，我们到处都看到，
27:03
people are removing tests, removing their stuff,
人们在删 tests，删他们的东西，
27:06
it's just because the same bad players are behaving badly
只是因为还是那帮坏家伙在使坏。
27:08
now, you know, it used to be much harder to copy pre-AI,
现在，你知道，在 pre-AI 时代，复制要难得多，
27:13
but now people are gonna like slot fork,
但现在人们就会，比如，直接 fork，
27:14
but even like they can just kind of move
但即便如此，他们也能更快地
27:17
a lot faster in interpreting the code.
去解读 code。
27:19
Like back in the day, you had defensibility
就像以前，你是有 defensibility 的，
27:21
because most people couldn't just read the code
因为大多数人根本读不懂 code。
27:24
and understand really what it was doing
并且真正理解它当时在做什么
27:25
and what the trade-offs were
以及有哪些权衡
27:26
and how it could be done better.
以及怎么才能做得更好
27:27
Now AI can explain it to them.
现在AI可以给他们解释这些了
27:29
So it's desperately sad that these people do this
所以这些人这么做，真的让人特别难过
27:32
and that companies are even willing to work with them,
而且公司居然还愿意跟他们合作
27:35
but they're the kind of casino you have to work with them.
但他们就是你不得不合作的那种赌场
27:37
And so it just means that I don't have a good answer.
所以这只能说明我真没什么好答案
27:39
I'm not gonna pretend to know what to do.
我不会假装我知道该怎么做。
27:40
We may still open source it.
我们可能还是会开源它。
27:42
It's just a world where you have to kind of play
现在的世界就是你必须得藏着掖着，
27:44
with your hand closed nowadays because of,
因为，你知道，如果你是一家新兴公司，
27:47
you know, if you are a rising company
威胁到了那些已经站稳脚跟的现有企业，
27:49
that's threatening others that are already more incumbent,
他们会使阴招，
27:52
they will do dirty tricks
而我绝不想给他们任何可乘之机。
27:53
and I would never want to give them any advantage.
而且我绝不想给他们任何优势。
27:56
Yeah, well, you're touching on a very,
嗯，好吧，你触及到一个非常——
28:02
I would say a very interesting hot button for me.
对我来说，一个非常有趣的敏感话题。
28:04
I'm not sure how much it's out there, I think it is.
我不确定这东西有多少，但我觉得是有的。
28:07
I'm actually all the enough in this day and my age
我其实在这个年代更是这样，
28:11
less slurping up what's on X
不太去刷X上的东西，
28:14
and less slurping up what's on LinkedIn in places like that.
也不太去刷LinkedIn那种地方的内容。
28:17
I'm just sort of like in my own world in a way
我有点像是活在自己的世界里，
28:20
which is kind of good because I've needed a mental buffer
这样其实挺好，因为我一直需要一个心理缓冲。
28:22
for this reason, but I've been in this thought process
因为这个原因，但我一直在这种思考过程里
28:25
of like, you know, the primitives you build on,
比如，你知道，你构建所依赖的那些 primitives，
28:29
the seemingly crown jewels that you would normally open source.
那些看似皇冠明珠、你通常会 open source 的东西。
28:33
Because hey, let's make this primitive available for others.
因为嘿，让我们把这个 primitive 开放给其他人。
28:36
We built on it.
我们是建立在它之上的。
28:37
Let's make this, let's be cool, right?
让我们这么做，让我们显得很酷，对吧？
28:38
Let's be cool.
让我们酷一点。
28:39
We're just a good thing to do.
这只是件好事。
28:40
I just, it's just a good thing to do.
我只是觉得，这是件好事。
28:41
Earlier in this room, I interviewed a friend of my Jacob
之前在这个房间，我采访了我朋友 Jacob，
28:43
who runs a company called Pierre.
他经营一家叫 Pierre 的公司。
28:45
He created Twitter Bootstrap and we were kind of talking
他创建了 Twitter Bootstrap，我们当时聊到
28:47
about how open source used to be used to just do it
过去 open source 往往就是单纯地做这件事，
28:49
because it was a good thing to be.
因为那是件好事。
28:51
And now it's got a fair, you know,
而现在它已经变得有点，你知道，
28:52
I dedicated my life to open source.
我把一生都献给了 open source。
28:53
I worked at GitHub for seven years, right?
我在GitHub工作了七年，对吧？
28:56
Like I supported the world of open source
就像我支持open source的世界
28:59
and I watched it from the inside and the nuances of it.
并且我从内部观察了它，以及它的细微之处。
29:01
It's a very fragile ecosystem.
这是一个非常脆弱的生态系统。
29:03
As a collective, the open source world is one
作为一个集体，open source的世界是
29:05
of the greatest things humanity has ever pulled off, right?
人类有史以来最伟大的成就之一，对吧？
29:08
Sharing our work for the betterment of humanity.
为了人类的进步，分享我们的工作。
29:11
It's why we have AI today.
这就是我们今天拥有AI的原因。
29:14
It is very sad to see the cynicism
看到那种愤世嫉俗的态度，真的很难过
29:17
that has overlaid open source
那种态度笼罩着 open source
29:19
because of the way commercial companies have behaved,
因为商业公司的所作所为，
29:22
especially ones like Amazon.
尤其是像 Amazon 这样的公司。
29:24
You know, I'm not sure, I know that your lens
你知道，我不太确定，我知道你的视角
29:27
and I'm curious if you feel it's just simply human nature.
我很好奇你是否觉得这只是人性使然。
29:31
I think it's human nature.
我觉得这就是人性。
29:32
We're, especially here in America,
我们，尤其是在美国这里，
29:34
we're our capitalists for the most part, right?
我们大部分都是资本家，对吧？
29:36
We're running companies.
我们在经营公司。
29:37
We're booting companies, we're...
我们在创办公司，我们……
29:38
Well, we can talk about that for a second while we're on capitalism.
好吧，既然聊到资本主义，我们可以顺便谈一下这个。
29:41
I love capitalism.
我热爱资本主义。
29:42
I, the capitalism is the greatest vehicle for him.
我，资本主义对他来说是最伟大的工具。
29:44
But I am so capitalistic
但我是如此信奉资本主义，
29:46
that I think these large tech companies
以至于我认为这些大型科技公司……
29:47
should be broken up, broken up.
应该拆分，拆分。
29:49
I should not be competing with my own company
我不应该和自己所在的公司竞争，
29:51
by buying a rotisserie chicken at Whole Foods.
通过去Whole Foods买烤鸡。
29:54
That should not be allowed.
这就不应该被允许。
29:55
Like that, that, that is just fundamentally anti-competitive.
就像这样，这从根本上就是反竞争的。
29:59
Let's look at AI.
我们来看看AI。
30:01
Google saw the early versions of AI very early
Google很早就看到了AI的早期版本。
30:04
and Peter Till is completely correct
而Peter Till是完全正确的。
30:06
that Google have run out of imagination
就凭他们银行里的钱，Google已经没想象力了。
30:07
based on how much money they have in the bank.
他们不再创新了。
30:09
They've stopped innovating.
他们利用，利用他们拥有Google search这个事实，而那是他们挣来的。
30:10
They use, they use the fact they have Google search,
他们靠Google search赚到了钱，也建成了史上最伟大的企业之一。
30:13
which is they earned that.
但他们看到了AI的早期版本。
30:14
They earned and they built one of the greatest businesses
他们应得这一切，他们打造了有史以来最伟大的
30:16
of all time with Google search.
企业之一，就是靠 Google search。
30:18
But they saw the early version of AI
但他们看到了 AI 的早期版本，
30:20
and they had the research and they just sat doing nothing,
他们有研究，却只是坐在那里什么都不做，
30:24
just wasting the opportunity.
白白浪费了机会。
30:26
Bam.
砰。
30:27
And then others came up, open AI.
然后其他人出现了，OpenAI。
30:29
Thank God for open AI that woke up and built something
感谢上帝，OpenAI 醒了，做出了东西，
30:33
and lit a fire under Google.
给Google点了一把火，逼它动起来。
30:35
And then Google get to wake up in the world
然后Google才不得不在这个世界上醒来，
30:37
like the speed AI is moving right now, an age later.
而那时候AI发展的速度已经像是过了一个时代。
30:43
And there's almost no consequence for them
而且他们几乎不用承担什么后果
30:45
because they still have a massive search business.
因为他们还有庞大的搜索业务
30:48
They still have so much capital to throw into AI.
他们仍有巨额资本投入到 AI 中
30:51
This is not good for innovation.
这对创新不利
30:53
I am not, I am not this kind of socialist
我不是，我不是那种社会主义者
30:56
kind of break up tech companies, anti-billionaire.
那种要拆分科技公司、反亿万富翁的人
30:58
I think it's phenomenal that people put rockets in space
我觉得人们把火箭送上太空并获得巨大回报，这很了不起
31:01
and get gigantic rewards for doing so.
并因此获得了巨大的回报。
31:04
I think we should be making more billionaires
我觉得我们应该造出更多亿万富翁，
31:06
that get to make more things and build things
让他们去创造更多东西、建设更多东西，
31:08
and that should trickle down.
而这应该会涓滴下去。
31:08
I think it is the true spirit of capitalism.
我认为这才是资本主义的真正精神。
31:11
If you look at the forefathers of capitalism,
如果你看看资本主义的先驱们，
31:13
the people, the way that in the young America
那些人，在年轻的America，
31:15
people talked about opportunity,
人们谈论机遇的方式，
31:17
they did not foresee the kind of leverage
他们没有预见到那种力量，
31:20
that these giant tech companies get.
这些大型科技公司得到的。
31:22
And it's herty innovation.
而且这正在损害创新。
31:23
These sleeping lazy giants sit on innovation
这些沉睡的懒惰巨人坐拥创新，
31:27
and it doesn't hurt them.
而这对它们并无损害。
31:28
That is not good for the general innovation of humanity.
这对人类的整体创新不利。
31:31
I like that.
我喜欢这个说法。
31:32
I want to coin that sleeping lazy giants.
我想发明“沉睡的懒惰巨人”这个词。
31:33
So, at the same time, you see forefathers of capitalism,
所以，与此同时，你看到资本主义的先驱们，
31:38
they were cutthroat.
他们可是心狠手辣。
31:40
Oh, you should be cutthroat.
哦，你也该心狠手辣。
31:41
There was probably murder even.
甚至可能还闹出过人命。
31:42
Like this is back, back, original OG business days, right?
就像这是很久很久以前，最原始的商业时代，对吧？
31:47
I could only speculate.
我也只能瞎猜。
31:47
I'm not speaking from a biography by any means,
我可不是在引用什么传记，
31:51
but I imagine there was a lot of cutthroat
但我觉得那会儿肯定有很多尔虞我诈，
31:53
and if there was a version of open source,
而且如果当时有 open source 这种东西的话，
31:56
they would certainly glean from it.
他们肯定会从中有所收获。
31:58
Like they're not benign, right?
他们可没那么好心，对吧？
32:00
Just because we got early.
就因为我们早了一步。
32:01
Right.
对。
32:02
But also open source came from a capital.
但 open source 也来自资本。
32:05
Like I mean, they tried, right?
我是说，他们试过了，对吧？
32:06
It was like Microsoft.
就像 Microsoft。
32:08
Let's not let them, let's not remember the October letters.
别让他们……我们别记住 October letters。
32:10
Let's forget the October letters.
咱们忘掉十月份的那些信吧。
32:12
Not so forget that they tried to kill
但别忘得那么彻底，忘了他们曾试图扼杀和压制 open source。
32:13
and suppress open source.
而且你知道，他们收购 GitHub 就是为了扭转这个局面。
32:15
And you know, they bought GitHub to undo this
他们确实做得不错。
32:18
and they have done a good job
他们也做出了一些很棒的 open source。
32:19
and they have done some great open source.
但资本主义，像 Microsoft 这样的大组织，
32:21
But capitalism, big organizations like Microsoft
确实试图根除 open source。
32:24
did try to stamp out open source.
确实试图扼杀open source。
32:26
And it did not work.
而且这没成功。
32:27
And there is still a very rebellious,
而且 open source 仍然有一股非常叛逆的、几乎政治性的运动。
32:29
almost political movement of open source.
再说，他们在某些地方走得太极端了。
32:31
Again, they take it too far in some areas.
但在资本主义下它也确实繁荣了。
32:33
But it did thrive in capitalism too.
有，有净收益。
32:36
There was a, there was a net good.
但我同意你。
32:39
But I agree with you.
但我不，早期的先驱们确实，你知道。
32:39
But I don't, the early forefathers did, you know,
但我不认为，早期的先驱们确实，你知道，
32:44
the crony capitalism is not what they wanted.
裙带资本主义不是他们想要的。
32:46
They, some of the early decisions around contract law
他们早期在合同法上的一些决定
32:49
are why America has such an incredible economy now
是美国现在经济如此惊人的原因
32:51
because they, because you could get a fair shaking court.
因为你能在法庭上得到公平的对待。
32:54
You can still sue and get your day in court and win.
你依然可以起诉，有上法庭的机会，并且能赢。
32:59
And that is extremely important.
这一点极其重要。
33:01
So I don't think it's worth viewing capitalism
所以我不认为资本主义值得被看作
33:03
as always having to end up into in a very cynical,
总是最终会演变成一种非常愤世嫉俗的境地。
33:06
everyone's gonna screw it.
每个人都会搞砸的。
33:07
But I do think you have to be cutthroat.
但我确实认为你必须要心狠手辣。
33:09
I do absolutely think.
我绝对这么认为。
33:10
I don't mean to say that.
我不是那个意思。
33:11
I just mean that I think it's human nature
我只是说我觉得这是人的本性。
33:13
and I'm with you.
我也站在你这边。
33:14
So I'm in a way narratively lamenting the scenario
所以我某种程度上是在以叙述的方式哀叹我们现在所处的境况，从一个立场。
33:20
we're in right now from a position
我们现在所处的位置。
33:22
that the change log is by and large
那个 changelog 基本上
33:25
been supporting open source since the year after
从第二年开始就一直在支持 open source
33:27
GitHub was accepted, like at the same year.
GitHub 诞生的时候，就好像在同一年。
33:30
Like I think GitHub was born in 2008, we were born in 2009.
就是我觉得 GitHub 是 2008 年诞生的，我们是 2009 年诞生的。
33:33
And we've been riding the wave of open source moves fast
而我们一直在乘着 open source 的浪潮快速前进
33:36
keep up since the beginning.
从一开始就跟上了。
33:39
Well friends, this episode is brought to you by our friends
好吧朋友们，本期节目由我们的朋友
33:41
at coder.com secure environments where developers
在 coder.com 的安全环境中，开发者
33:45
and agents work in parallel.
而且 agents 是并行工作的。
33:47
And I'm joined by Nikki Pike, field CTO, forecoder.
和我一起的是 Nikki Pike，field CTO，Forecoder。
33:51
Nikki, what is the field CTO?
Nikki，field CTO 是什么？
33:52
So I get that question a lot and it's, you know,
这个问题我被问过很多次，而且，你知道，
33:55
half the people understand it, half the people don't.
一半人理解，一半人不理解。
33:57
So a field CTO, I describe it very simply
所以 field CTO，我描述得非常简单，
33:59
as we're dev realm for the seed suite.
就是，我们是 Seed Suite 的 dev rel。
34:01
So we provide a bridge between the customer voice,
所以我们提供了一个桥梁，连接客户的声音。
34:04
between the seed suite and the managers
在 seed suite 和经理们
34:06
and the leadership teams of our customers
以及我们客户的领导团队之间
34:08
back into our product.
再回到我们的产品中。
34:10
And then we go through and we help enable our teams
然后我们会帮助我们的团队
34:12
to have the same message to make sure
统一信息，确保
34:14
that the message is correct and that we're building
信息准确，并且我们构建的
34:16
on something that people actually want,
是大家真正想要的东西，
34:17
not just something that we think they want.
而不是我们以为他们想要的东西。
34:19
Okay, so we're taking the laptop away from the developer.
好的，所以我们正要把开发者的笔记本电脑拿走。
34:22
Not really though, we're putting them in a cloud
不过也不是真的那样，我们是把他们放到一个 cloud
34:24
development environment, a secure environment
development environment 里，一个安全的环境
34:26
where they can work with their agents
在那里他们可以和自己的 agents 一起工作
34:28
in parallel.
并行地工作。
34:29
These are blessed environments.
这些是经过批准的环境。
34:30
What's wrong with the laptop?
笔记本电脑有什么问题？
34:31
The laptop is the, is the trap here.
笔记本电脑就是，就是这里的陷阱。
34:34
And not only because the fact that it could be stolen,
而且不仅仅是因为它可能被偷，
34:36
you could lose it at breaks and you're out of work
你可能在休息时把它丢了，然后就没法工作了，
34:38
while you're waiting for a new one.
还得等着换新的。
34:39
But there's also just the consistency that you got there.
但这样做也有个好处，就是一致性。
34:41
We all know developers, developers are going to be looking
我们都知道，开发者们总是会去追求
34:44
for some of the latest and greatest.
最新最好的东西。
34:45
And if you're not really controlling how they get out there,
如果你不好好控制他们怎么得到这些东西，
34:48
that's where you get this.
那你就会遇到这种情况。
34:49
It works on my machine.
在我的机器上能跑。
34:49
It doesn't work on production.
在 production 上就跑不起来。
34:51
It doesn't work anywhere else because you don't have that
它在别的地方都跑不了，因为你没有那种
34:53
consistency.
一致性。
34:53
You don't have that ability to really standardize
你没有那种能力去真正 standardize
34:56
what that environment looks like.
那个 environment 长什么样。
34:57
And this is a problem not only for new people coming in,
这个问题不仅影响新加入的人，
35:00
you know, the onboarding statement as average,
你知道，onboarding statement 一般般。
35:02
I think is like four to five weeks for a new employee
我觉得一个新员工大概要四到五周
35:05
to really get their local laptops set up and ready to start
才能真正把他们的 local 笔记本电脑设置好，准备好开始
35:07
doing their first time of code.
写他们的第一段代码。
35:09
And you know, the time to first commit is a metric
而且你知道，time to first commit 是一个 metric
35:11
that almost everybody knows.
几乎每个人都知道的。
35:12
And the reason they can't do that is because there's a lot
他们做不到的原因，是因为有太多
35:14
of tribal knowledge out there.
tribal knowledge 散落在各处。
35:16
They got to go talk to other developers.
他们得去跟其他开发者聊。
35:17
What are we using?
我们在用什么？
35:18
Where do we get our dependencies?
我们的依赖从哪来？
35:19
Are we getting them from public?
是从公开渠道获取的吗？
35:20
Are we getting them from private repositories?
是从私有仓库获取的吗？
35:22
But there's also the security and the supply chain aspect
但这里面还有安全和供应链的问题，尤其是当你有很多本地机器在外面跑的时候。
35:25
of this when you have local machines out there.
就像那个 shy halad，你知道吧，
35:28
Look at like the shy halad, you know,
就是不久前爆出来的那个病毒。
35:29
that virus that went out not long ago.
那个不久前出现的 virus。
35:31
This was a compromise of the MPM public repositories.
这是对 MPM public repositories 的一次 compromise。
35:35
They went and downloaded things.
他们去下载了一些东西。
35:37
MPM did what it did.
MPM 该干嘛就干嘛。
35:38
Next thing you know, you're compromised.
一转眼你就被 compromise 了。
35:40
But when you use something like what we're doing
但是当你用类似我们正在做的这种方案，
35:42
with cloud development environments,
就是 cloud development environments 的时候，
35:44
then you can mandate and you can put restrictions
那你就能够强制规定，也可以加一些限制，
35:46
on there to say, hey, you can only go get your packages
比如说，嘿，你只能去获取你的 packages。
35:49
from our private repo.
从我们的 private repo 里。
35:50
Those packages are expected to have been thoroughly vetted.
这些 packages 应该都经过了彻底的审查。
35:53
We know that they're clean.
我们知道它们是干净的。
35:54
Now does this stop everything like shy halad?
那这能阻止一切吗，比如像 shy halad 那样？
35:56
No, if that compromised package gets into your private repo,
不，如果那个被入侵的 package 进入了你的 private repo，
35:59
you can still have that,
你仍然可能遇到那种情况，
36:00
but it really reduces the surface area of the attack.
但它确实能减少 attack surface。
36:03
And it also reduces the blast area
而且也能缩小 blast area。
36:06
of the compromise should it happen
这种 compromise 如果真的发生
36:07
because if your laptop gets compromised
因为如果你的 laptop 被 compromise 了
36:10
and you have to kill the laptop for whatever reason,
然后不管什么原因你都得把 laptop kill 掉
36:12
that's weeks out of work
那就是好几周没法工作
36:14
while you're either fixing that
期间你要么在修它
36:15
or you're getting a new laptop in,
要么在等新 laptop 到位
36:17
the cloud development environments allows you to kill that,
而 cloud development environments 可以让你直接 kill 掉它
36:19
start back up fresh and you're back
重新启动一个全新的环境，然后你就能继续工作了
36:21
and running in five minutes.
并且五分钟内就能跑起来。
36:22
You don't have to wait all that time.
你不必等那么久。
36:24
Well friends, the first step is to go to coder.com,
好了朋友们，第一步是去 coder.com，
36:27
install coder, self-hosted environments for your teams
安装 coder，为你的团队提供 self-hosted 环境，
36:31
to enjoy, to standardize around,
让大家用得开心，并围绕它来标准化，
36:33
and it's open source.
而且它是 open source 的。
36:34
So you can try it out today.
所以你今天就可以试试看。
36:35
Once again, coder.com.
再说一次，coder.com。
36:44
And so as that same person I'm sitting here today saying,
所以，作为同一个人，我今天坐在这里说，
36:47
okay, Niki, is it gonna be open source?
好，Niki，这会是 open source 吗？
36:50
I don't mean that, like, hey Sam,
我不是说，嘿 Sam，
36:51
give away your crown jewels.
把你的家底都交出来。
36:52
It's been like, how do you battle this struggle?
而是说，你怎么面对这种挣扎？
36:56
How do you deal with this struggle of like,
你怎么应对这种，就是，
36:58
gosh, we're in the scenario where my most fierce competitors
天哪，我们现在的处境是，我最强的竞争对手，
37:01
that have more than me, they can kebash me,
他们比我拥有的多，他们能碾压我。
37:04
maybe not technically,
也许不是技术上的，
37:05
but maybe with a purse potentially or a war chest.
但也许是靠一笔资金，或者一个储备金。
37:09
How can I give them my 15 years of a test,
我怎么能把我15年的这场考验交给他们，
37:13
my five years of running, playing this scale
我五年来运营、玩转这个规模，
37:14
and just hand that over in open source.
然后就这样开源交出去。
37:17
It's a very challenging position to be in,
这是一个非常艰难的处境，
37:19
but yet we come from this past thread of lovers
但尽管如此，我们来自这条爱好者和支持者的脉络，
37:23
and supporters of open source.
是开源的支持者。
37:24
Yeah, it's hard.
是啊，很难。
37:25
I don't, I don't love thinking about it this way.
我不，我不太喜欢这么想。
37:27
And like I said, I used to say we would never do
而且就像我说的，我以前总说我们绝不会做
37:29
purchase grids here we are.
purchase grids，结果现在我们就在做。
37:30
I don't, I just don't know.
我不，我真的不知道。
37:32
You did say that when we talked in the first time.
你确实在我们第一次聊的时候说过。
37:34
I mean, you're like, I might as go for life.
我是说，你好像在说，我可能还不如去追求生活。
37:36
Right, but test for life.
对，但那是生活的考验。
37:38
Right, and I changed my mind on that.
对，而且我改变了主意。
37:40
And I'm still, I do not mind being insistent.
而且我还是，我不介意坚持己见。
37:44
Like, I mean, I, you know, my public stance
就像，我的意思是，我，你懂的，我的公开立场
37:47
is I think about things very openly.
就是我会非常开放地思考事情。
37:48
And I express my feelings and I have absolutely.
而且我表达我的感受，我绝对这样做了。
37:52
Like people, when I tweet about stuff,
就像人们，当我发tweet说事情的时候，
37:54
people love to dig up old tweets.
人们就爱翻出我以前的tweet。
37:55
It's like some sort of gotcha.
这就像某种gotcha。
37:58
As if like, I'm like, people change their minds.
就好像，我是说，人们会改变主意。
38:00
Yeah, correct.
嗯，对。
38:01
I will change my mind faster than you.
我会比你更快改变主意。
38:02
No changes.
没有变化。
38:03
Correct.
没错。
38:04
I get no embarrassment and no shame
我不会感到尴尬或羞愧
38:06
from not being logically insistent.
因为我不在逻辑上固执己见。
38:07
In fact, my by-law, one of my things I say
事实上，我的准则，我常说的一句话
38:10
in my like kind of description on Twitter
我在Twitter上的简介大概就是
38:12
is I'm making more mistakes than you.
我犯的错误比你多
38:14
And I'm actually very proud of that.
而且我其实非常自豪
38:15
Like people, again, they read that and they get it wrong.
但人们，他们读到这个就理解错了
38:17
They're like, oh, it's a guy's proud of making mistakes.
他们会觉得，哦，这人以犯错为荣
38:19
Stupid.
真蠢
38:20
I genuinely deliberately make mistakes.
我确实是故意犯错的
38:24
I move extremely quickly.
我行动极其迅速
38:25
I believe life is like a spread bet.
我觉得生活就像一场分散下注。
38:28
You have to make mistakes.
你必须犯错。
38:30
If I said to you, there's a target down range.
如果我跟你说，靶场那头有个目标。
38:33
I'll give you a million dollars if you hit it.
你打中我就给你一百万美元。
38:35
Choose a sniper eye for a shotgun.
选狙击枪还是霰弹枪。
38:36
What are you choosing?
你会选哪个？
38:37
Choose a shotgun because you're going to get,
选霰弹枪，因为你会有，
38:39
I don't know how many shot they go in,
我不知道里面有多少颗弹丸，
38:40
put in a shotgun.
那就用霰弹枪。
38:41
But you're going to get more chances
但你会得到更多机会，
38:42
than the one sniper shot.
比那一次狙击手的一枪还多。
38:44
If you bundle everything into one big decision,
如果你把所有东西都捆绑在同一个大决定上，
38:47
you better not miss.
你最好别打偏。
38:49
Well, with a shotgun blast.
嗯，霰弹枪一枪出去就是一大片。
38:50
And I learned this in the early days of GitHub.
而我在 GitHub 早期就学到了这一点。
38:52
That culture was incredibly iterative.
那种文化极其 iterative。
38:53
We would put something small out, learn, iterate, build.
我们会先放出一个小东西，然后学习、迭代、构建。
38:57
I just will not, I don't, I refuse to ever stick to anything
我就是不会，我不，我拒绝永远固守任何
39:02
in the past.
过去的东西。
39:03
I have almost to a fault zero respect for the past
我对过去几乎到了过分的程度，零尊重，
39:07
or time for the past.
或者说，零时间。
39:09
I just want to see what's tomorrow.
我就想看看明天会怎样。
39:10
And if the most offensive, horrible thing
而如果我能对竞争对手做的最冒犯、最可怕的事，
39:14
I can do to my competitors is open source Niki.
就是 open source Niki。
39:17
You bet you'll be done in days.
你当然觉得几天就能搞定。
39:19
Until then, I think the best thing I can do is
在那之前，我觉得我能做的最好的事就是
39:21
I'm watching some being built in the open.
我看着有些东西在公开场合被造出来。
39:24
They are years behind us.
他们落后我们好几年。
39:27
I'm watching us take Amazon's best swing
我看着我们扛住亚马逊最猛的一击，
39:30
and moving customers off constantly.
同时不断把客户转移走。
39:32
So right now, I'm keeping that for ourselves.
所以现在，我把那个留给我们自己。
39:34
But who knows?
但谁知道呢？
39:35
We'll see.
再看吧。
39:37
Yeah, you know, I would, I would struggle
是啊，你知道，我也会，我也会跟你现在一样，面对这个选择很纠结。而且如果我是你，我就直说了，我可能会先藏着，因为你图什么呢？
39:43
just as much as you are right now with that choice.
没错。
39:47
And if I were you, just speaking out loud here,
比如说，你 open source Niki 到底对谁有好处？
39:50
I'd probably keep it on the inside for now
我目前可能会把它保持在内部。
39:53
just because what do you gain?
因为你能得到什么好处呢？
39:57
Yeah.
对。
39:58
Like who do you help by open sourcing Niki?
比如说，把 Niki open source 出来能帮到谁？
40:02
That's the question, yeah.
这就是问题所在，是啊。
40:03
It's sad.
挺难过的。
40:04
And then even with open source, there's so much change.
而且就算是在开源领域，变化也太大了。
40:06
Maybe you could speak to this too potentially
也许你也能聊聊这个，
40:08
as you, as you weigh these pros and cons.
当你在权衡这些利弊的时候。
40:12
As you see a lot of change in open source,
你看到开源社区有太多变化，
40:13
you mentioned it pulling tests back, all sorts of things,
你提到把测试撤回来之类的各种事情，
40:18
not accepting pull requests, not accepting contributions.
不接受 pull request，不接受贡献。
40:21
So we're open source.
所以我们是 open source 的。
40:23
Sure, you can use it.
当然，你可以用。
40:23
It's still the free gift.
这仍然是免费的礼物。
40:24
It's still the USB stick on the front curve.
它仍然是前面曲线上的 USB stick。
40:27
Yeah, thank you very much.
是的，非常感谢。
40:28
But we don't accept any contributions.
但我们不接受任何贡献。
40:31
It's just in-house only.
它只是内部使用。
40:32
So there's no, and then you've got the Mitchell Hashimoto
所以没有，然后你还有 Mitchell Hashimoto。
40:36
direction, which is like, let my current contributors,
方向，就像是，让我现在的contributors，
40:41
you know, I think I'm sort of, you know,
你知道，我觉得我有点，你懂的，
40:44
TLD are in a way what they're doing,
TLD在某种程度上就是他们在做的事情，
40:46
but it's sort of like a merit based
但这有点像是一种merit-based的
40:48
if you've done some good things
如果你做过一些好事
40:50
or somebody suggests you can kind of come in kind of thing.
或者有人推荐你，你就可以加入之类的。
40:53
And I could be wrong on that.
而在这一点上我可能是错的。
40:55
Again, I'm sort of out of the loop
再说一次，我有点跟不上情况了。
40:56
to some degree on certain things,
在某种程度上，在某些事情上，
40:57
but I feel like the scoreboard
但我感觉记分牌
41:00
and the way we participate in certain open source,
以及我们参与某些 open source 的方式，
41:03
not all open source, but certainly high stakes open source,
不是所有的 open source，但肯定是高风险的 open source，
41:06
like something like Niki being open source, for example.
比如像 Niki 是 open source 这样的例子。
41:09
There's a different game being played.
这里玩的是另一种游戏。
41:12
And it's just a little unfair to release your crown jewel
而把你自己的皇冠明珠发布出来就有点不公平了，
41:16
that helps you do so much.
它帮你做了那么多事。
41:18
And then what do you really gain from giving it away?
那把它送出去，你真正能得到什么？
41:20
I don't know.
我不知道。
41:21
Yeah, I think about this.
是啊，我也在想这个问题。
41:23
And I don't want to criticize anyone who's done,
而且我不想批评任何这样做过的人，
41:26
like there are people that you have to find your own path
就像有些人，你得找到自己的路，
41:28
in this world, and people do amazing.
在这个世界上，而且人们做的事情特别了不起。
41:29
Like open source maintainers are saints.
比如 open source 的维护者，他们都是圣人。
41:32
Like the things they put up were this hard for this year.
就像他们今年忍受的那些事情，真的很难。
41:34
And now we, I just want to say that like,
而現在我們，我只想說，就是說，
41:37
you are all incredible people.
你們都是非常了不起的人。
41:38
And I hope that we all find a better path forward for this.
我希望我們都能為此找到更好的出路。
41:43
Maybe it's licensing.
也許是 licensing。
41:45
I don't know.
我不知道。
41:46
But like I really hope so.
但我真的希望如此。
41:47
So I would never want to seem like I'm critical.
所以我絕不想讓自己看起來像是在批評。
41:50
We're losing from this too, right?
我們也在這之中失去一些東西，對吧？
41:51
Like we're still not the size of Amazon.
就像我们还没有达到Amazon的规模。
41:54
So still when we're working with these gigantic businesses,
所以当我们和这些巨型企业合作的时候，
41:57
they're like, well, I know we know.
他们就会说，好吧，我知道，我们也知道。
41:59
Plant scale is profitable and has lots of money in the bank
Planet scale 是赚钱的，银行里有很多钱，
42:01
and is going to be around forever.
而且会一直存在下去。
42:03
But you know, it would be easier if like we knew,
但你知道，如果我们知道会更容易，
42:06
because, you know, we've never had an enterprise customer
因为，你懂的，我们从来没有任何一个企业客户
42:10
turn, right?
流失过，对吧？
42:10
And each we've never had a sharded customer leave us.
而且我们从来没有一个 sharded 客户离开过我们。
42:12
So we've retained every dollar of enterprise business.
所以我们 enterprise business 的每一分钱都留住了。
42:15
And they know that, right?
他们也知道这一点，对吧？
42:16
And they know that it's like not incredibly easy to go from.
他们也知道，要从这里换走并不是那么容易的事。
42:18
If plants go as the best most scalable database,
如果 PlanetScale 是最好的、最 scalable 的数据库，
42:21
if we hit the limits of that, like they know
就算我们达到了它的极限，他们也知道
42:23
that they're not going to turn around and do it tomorrow.
他们不会第二天就马上换掉。
42:25
But if a lot of the code was open source,
但如果很多代码都是 open source 的，
42:27
it does give them more of a sense and a feeling
这确实让他们更有一种感觉和体会
42:29
that they have a plan if something was ever to happen
就是万一发生什么事，他们至少有个计划
42:31
or we got acquired or whatever, right?
或者我们被收购了之类的，对吧？
42:33
So we're losing from this too.
所以我们也失去了这个
42:34
And maybe that's why Amazon behaved the way.
也许这就是亚马逊那样做的原因
42:36
It's because this way, it's because it's easier for them
是因为这样，是因为对他们来说更容易
42:39
to stymie us if we know that we can't do that.
如果我们知道我们不能那样做，他们就能阻挠我们
42:41
I don't know.
我不知道。
42:42
It's pretty sad and it's not an easy situation.
这挺让人难过的，而且情况并不容易。
42:45
And I will always come back and update you
而且我会一直回来跟你更新，
42:47
on as my thoughts change.
随着我的想法改变。
42:48
But right now, I speak to like a lot of the people
但现在，我确实和很多人聊过，
42:51
you've mentioned as being like shining stars
就是你提到的那些耀眼明星一样的人。
42:53
where it was also I've had this conversation with people.
我也和他们聊过这个话题。
42:55
And we're all trying to find our way
而我们都在试图找到自己的路，
42:58
toward like figuring this out.
去把这件事弄清楚。
43:03
Well, I mean, if you can give a demo with cursor,
嗯，我的意思是，如果你能用 Cursor 做个 demo，
43:07
have it do some cool stuff with insights
让它用 insights 做一些很酷的事情，
43:09
and you could do the same thing with the code base.
你也能对 code base 做同样的事。
43:11
You can look at somebody's code base,
你可以看某人的 code base，
43:12
whether it's even from a license perspective,
甚至从 license 的角度来看，
43:15
you can graph intent.
你可以 graph intent。
43:18
You can, you know, you don't have to copy code these days
你可以，你懂的，现在你不用复制代码
43:21
to copy the idea or even to, I didn't want to say copy
就能复制想法，甚至——我本来不想说“复制”这个词。
43:25
is like a bad thing.
就像是件坏事。
43:26
You know, we've been all been copying.
你知道，我们一直都在复制。
43:27
We've been standing on the shoulders of giants
我们一直站在巨人的肩膀上，
43:29
for a long time.
已经很久了。
43:31
But it's really just an interesting space
但这确实只是一个有趣的空间，
43:34
where you have a super intelligence available
在那里，super intelligence 对几乎任何人都是可用的，
43:37
for most anybody with 20 bucks on most of the planet
在这个星球的大多数地方，只要20美元，
43:42
that has access to these tooling
并且能接触到这些 tooling。
43:45
and to learn, to examine, to replicate.
去学习，去检验，去复制。
43:49
You can do it nefariously.
你可以用心怀不轨的方式去做。
43:51
You can do it altruistically, you know,
你也可以出于利他目的去做，你知道的。
43:53
you can do it from somebody different angles.
你也可以从不同的角度去做。
43:55
But the point is the primitive, the ability is there.
但关键是，这种原始的能力就在那里。
44:01
Take me, take me back to this unfair advantage of Niki.
让我，让我回到Niki的这种不公平优势上。
44:04
I don't want to, I don't want to blaze past that
我不想，我不想就这么跳过那个，
44:06
with how deep you win in the, the feelings you have
以及你有多深地陷入那些，那些感受之中。
44:10
on today's open source.
关于今天的 open source。
44:11
But what is this unfair advantage?
但这个不公平的优势是什么？
44:13
How long is this list and why do people feel,
这个列表有多长？为什么人们会觉得，
44:18
take me further into that, that remark?
再跟我深入聊聊那个说法？
44:21
Well, there's two sides to this.
嗯，这件事有两面性。
44:23
And you've just touched on AI.
你刚才提到了 AI。
44:25
One, one unfair advantage is is that we have,
一个，一个不公平的优势就是，我们有，
44:28
some of us have spent the last 15 years
我们中的一些人花了过去 15 年
44:32
in our careers working on very large websites
在我们的职业生涯中，处理过非常大型的网站
44:35
or very large database deployments.
或者非常大型的 database deployment。
44:37
And that is knowledge that has not made it
而这些知识并没有进入
44:39
into any single model, right?
任何一个单一的 model，对吧？
44:41
Like the code that we wrote to solve these problems may do.
就像我们为了解决这些问题写的 code 可能已经做到了这一点。
44:45
But the brains that that code was formulated are still
但构思出那些 code 的大脑仍然
44:48
in our heads and precisely none of that has gone in.
在我们脑子里，而且确切地说，其中没有任何一点被纳入。
44:51
And the reason we have very, very large AI companies
而这就是我们拥有非常非常大的 AI 公司的原因
44:54
building up of us is because they're buying
我们做起来，是因为他们买
44:57
for our expertise, they're buying for what my engineers
我们的专业能力，以及我的工程师们
45:01
will do at two in the morning,
在凌晨两点会做什么，
45:02
paged with a behavior that has never happened before.
被一个从未出现过的行为page。
45:06
That active human brain instructing the agents
那个活跃的人类大脑指挥着agents
45:10
to fix things and building the pipelines
去修复问题，并构建pipelines，
45:13
that the agents use is what people pay for.
这些agents所使用的正是人们付费购买的东西。
45:16
So that gives us one advantage, which is,
所以这给了我们一个优势，那就是，
45:19
yeah, maybe I'm not 22 anymore doing all nighters
是啊，也许我不再是22岁了，不能老是熬夜
45:21
all the time, but agents can do all nighters.
但 agents 可以一直通宵。
45:25
So we've got, so not only has Niki been
所以，我们有了——所以 Niki 不仅仅是
45:28
the distillation of so many years of experience
多年经验的结晶
45:32
of incredibly talented people.
来自那些极其有才华的人。
45:34
It's been built at a breakneck pace
它是以惊人的速度建成的
45:37
because those people turned out exceptional engineers
因为那些人成了杰出的工程师，
45:39
can do exceptional engineering with AI
能用 AI 做出非凡的工程。
45:41
and we're taking our experience and using it.
然后我们正在运用我们的经验。
45:44
And so there is potential for a lot of tech companies
所以有很多 tech 公司有这个潜力。
45:48
that can you hear them?
那个——你们能听到吗？
45:53
Yeah.
嗯。
45:54
What in the world?
什么情况？
45:56
They're doing construction outside.
他们外面在施工。
45:58
That is likely a drill.
那大概是电钻。
45:59
It's okay.
没事儿。
46:00
Let's see if it passes.
看看能不能通过。
46:02
I hope so.
但愿如此。
46:03
Give me a second.
给我一点时间。
46:04
That was a great question.
这个问题问得好。
46:05
You asked me, I want to make sure I don't forget that.
你问我的那个，我想确保自己别忘了。
46:08
Man, you're so good at this, by the way.
哇，你真是太厉害了，顺便说一句。
46:10
I literally, you know, I've never,
我真的，你知道，我从来没有……
46:13
I asked you to come on this.
是我请你来上这个的。
46:14
I've never asked a podcast to have me back.
我从来没要求过哪个播客再请我回来。
46:16
No.
不。
46:17
Nope.
没有。
46:18
Wow.
哇。
46:19
You stood out.
你当时特别突出。
46:21
I still have the t-shirt.
我还有那件T恤。
46:22
I also don't keep my arch either.
我也不留着我的 arch。
46:23
Like I really thought you did an amazing job.
我真的觉得你做得非常出色。
46:25
So like I would love to make sure it was what I did.
所以说，我特别想确认那确实是我做的。
46:30
No, seriously, I've been through some stuff, man.
不，说真的，我经历了不少事，兄弟。
46:32
It's been, it's been a journey.
这，这真是一段旅程。
46:33
I hope you're okay.
我希望你还好。
46:35
It's while we're waiting for this to go.
就在我们等着开录的时候。
46:37
I'll give you a little snapshot.
我给你们简单说一点。
46:39
So behind the scenes, this last 2025 was a challenging year
所以在幕后，刚刚过去的2025年对我来说是充满挑战的一年，
46:43
for me and my business partner, Jared.
对我的生意伙伴Jared和我来说。
46:44
He's a great dude.
他人很好。
46:45
I love him the death.
我爱死他了。
46:46
But we just had differing opinions
但我们只是意见不合。
46:49
on how the business should run.
在公司该如何运营上。
46:51
And I just got tired of the partnership.
我只是厌倦了这种合伙关系。
46:55
And I really did.
我确实厌倦了。
46:57
And I'm sure he got tired of the partnership as well.
我肯定他也厌倦了合伙关系。
46:58
It's not like it's a one-sided scenario here.
这也不是单方面的事。
47:01
But we were together as partners for a decade, basically.
但基本上，我们作为合伙人一起干了十年。
47:05
And that's a challenging change, really.
这确实是个很难的转变。
47:08
And so right now, in December of last year,
所以在去年十二月，
47:12
I asked him if I could buy him out.
我问他，能不能我把他那份买下来。
47:14
He agreed.
他同意了。
47:15
And so in February of 26th this year,
然后在今年2月26号，
47:19
we inked that deal and it's done.
我们签了协议，这就搞定了。
47:21
So he's known for that business.
所以他是因为那个生意出名的。
47:23
So that's the change.
所以这就是变化。
47:24
But at the same time, we have this AI revolution happening.
但与此同时，我们正在经历这场 AI 革命。
47:29
I've been building a lot of cool stuff.
我一直在做很多很酷的东西。
47:31
Although while also podcasting and maintaining this business,
虽然同时也在做播客、维持这个业务，
47:34
but it's been challenging because the world has just changed drastically.
但确实很有挑战性，因为世界刚刚发生了翻天覆地的变化。
47:39
And I've had some imposter syndrome, to be honest with you.
说实话，我也有些 imposter syndrome。
47:43
Well, you should know it.
嗯，你应该知道。
47:44
And you should know it.
而且你应该知道。
47:45
But see here, you say that is what makes me tell you
但你看，你说这就是为什么我会告诉你这些幕后的、看似TMI的事情，我平时不会跟人分享这么深的。
47:48
that seemingly TMI behind the scenes that I would not normally
但我也很感激，因为我希望你真心接受我的鼓励，继续努力做下去。
47:52
share so deeply with someone.
由于多种原因，计划有变，我们裁了员。
47:56
But so I appreciate that because I want you
我们一度差点活不下去了。
47:57
to wholeheartedly take my encouragement
要全心全意地接受我的鼓励，
47:59
that you should keep doing going forward.
你今后应该继续这样做。
48:02
For multiple reasons, plants go, we did layoffs.
由于多种原因，工厂倒闭了，我们进行了裁员。
48:04
We were nearly dead at one point.
我们一度差点完蛋。
48:06
Things are better than ever.
现在比以往任何时候都好。
48:07
We are now more alive, ripping business.
我们更有活力了，业务做得风生水起。
48:12
We got features like Brex did a report that we're in the top 25
我们上了很多榜单，比如Brex发过一份报告，说我们在top 25里面。
48:16
of things that start up suspending on now.
就像我们一路走过来，当时真的他妈的暗无天日，
48:19
Like we are going, when it was so f***ing dark
如果你不撑过去，就永远到不了今天。
48:22
and if you don't push through that, you will never get to that.
这真的很难。
48:25
And it's really hard.
你也是创始人，你懂的。
48:26
You're founder too.
你也是创始人。
48:27
And literally the story you just told me, another creator told me the exact same thing.
而且真的，你刚跟我说的那个故事，另一个创作者也跟我说了完全一样的。
48:31
Business partner, breakup, buying them out, pain.
生意伙伴，分手，买断他们，痛苦。
48:35
It's very common.
这很常见。
48:36
But I'm serious, like deadly serious.
但我是认真的，认真得要命。
48:40
You are P99, amazing at what you do.
你是 P99，特别厉害。
48:43
You should not feel it.
你不该有这种感觉。
48:44
But all the best people feel imposter than TMI do too.
但所有最优秀的人都会觉得自己是冒牌货，TMI 也一样。
48:47
But you shouldn't.
但你不该。
48:49
It's also a long journey to having done this.
做到这件事也是一段很长的旅程。
48:53
Like my whole career has been a podcaster for the most part.
就像我整个职业生涯大部分时间都是在当播客主。
48:56
And I can honestly say, I haven't built anything technologically
而且说实话，我在技术上并没有做出任何真正有价值的东西。
48:59
that's worth really anything.
所以说，身在科技行业却没有什么拿得出手的技术，某种程度上挺尴尬的。
49:02
So being in the tech industry and not really having any tech to show for it
不，但现在媒体就是我。
49:06
is kind of embarrassing in a way.
我——你知道，媒体现在比以往任何时候都更火，对吧？
49:08
No, but like media now is me.
不，但现在媒体就是我。
49:10
I, you know, media is hotter than ever now, right?
我，你知道，现在媒体比以往任何时候都火，对吧？
49:14
People want to tell, because there's now, like,
人们想要讲述，因为现在，就像，
49:17
I remember back in the day, there was like two companies doing the same thing, right?
我记得当年，大概只有两家公司在做同样的事情，对吧？
49:22
Like you basically just like in 2013, you had to be the first person with the idea
就像在2013年，你基本上只要成为第一个有这个想法的人
49:25
and you could make a big company.
就能做出一家很大的公司。
49:27
Now there's 10 people, 10 startups in San Francisco going after every bad idea.
现在旧金山有10个人、10家startup在追每一个烂点子。
49:33
And people are differentiating now through storytelling, craft, design, experiences.
而现在人们通过讲故事、手艺、设计和体验来让自己与众不同。
49:40
I think you're about to enter an age where your skills are more important than ever.
我觉得你即将进入一个技能比以往任何时候都重要的时代。
49:43
I genuinely believe, I look at how we build our marketing team
我真心相信，看看我们如何组建我们的营销团队就知道了。
49:46
and our buying behaviors or going in that direction.
以及我们的购买行为，或者说是朝着那个方向发展。
49:48
Yeah.
嗯。
49:49
Well, I appreciate that.
嗯，我感谢你这么说。
49:50
I mean, I think the business is definitely changing.
我的意思是，我觉得这个行业确实在变。
49:55
I'm still excited about podcasting.
我依然对播客充满热情。
49:59
I'm just a little bummed to be alone, even though it was my choice.
我只是有点沮丧，因为只有我一个人，尽管这是我自己的选择。
50:02
Yeah.
嗯。
50:03
And so I'm just navigating some personal motivation, I would say.
所以，我也就是在摸索一些个人动力方面的事情吧，可以这么说。
50:08
And then also I'd think a little bit of burnout and then also some imposter syndrome.
然后我也觉得会有点 burnout，再加上一些 imposter syndrome。
50:11
So you mix all that together.
把这些全混在一起。
50:12
You get a really weird suit that just tastes kind of nasty sometimes now.
最后变成一碗很奇怪的汤，有时候喝起来就是很难受。
50:15
Well, you should come out San Francisco and fill the energy out here.
嗯，你真应该来 San Francisco 感受一下这边的能量。
50:17
Like, it will really pick you up.
肯定会让你振作起来。
50:20
I think it will really make you feel like go to events, see people.
我觉得真的会让你想去参加活动、见见人。
50:24
I think you'll really enjoy it.
我想你会很享受的。
50:25
You'll meet some new generation who found us to interview and chat too.
你会遇到一些新一代的人，他们也会来找我们做采访、聊聊天。
50:28
I think it'd be really good.
我觉得那会特别好。
50:29
Man, I got a ton of friends out there too.
天哪，我在外面也有好多朋友。
50:30
I mean, I would consider you as a friend as well.
我是说，我也会把你当朋友的。
50:32
But I would come through, come to the office, hang out.
但我会过来，去办公室坐坐，聊聊天。
50:34
You're more welcome.
你太欢迎了。
50:35
I should just do like a tour.
我该来个tour之类的。
50:37
Yep.
嗯。
50:38
Go see Adam Jacob.
去看看Adam Jacob。
50:39
Come see you.
来看你。
50:40
Exactly.
没错。
50:41
There's a few others that I would so reach out to.
还有几个人我很想联系一下。
50:44
My buddy at Render on a rag.
我在 Render 的哥们，搞 RAG 的。
50:47
Great.
太好了。
50:48
There's a small handful folks I'd just like going hang out with if they would let me of
还有一小部分人，我很想跟他们一起玩，当然前提是他们愿意。
50:51
course.
过来吧。
50:52
Come through.
挺过去。
50:53
Right.
对。
50:54
I think we're going to drill this window.
我觉得我们要深入探讨这个窗口期。
50:55
So I'll carry on.
所以我继续。
50:56
So I'll start.
那我开始了。
50:57
Edit.
剪辑。
50:58
I didn't pick up the edges.
我没注意到那些边边角角。
50:59
So we're like in this very interesting time right now.
所以我们现在就像处在一个非常有趣的时期。
51:01
There's a bunch of companies being built by people who've got deep expertise from working
有一大批公司，是由那些从工作中获得深厚专业知识的人创办的。
51:07
in other places.
在其他地方。
51:08
They know they're domain really well and they can build for the next generation and
他们非常了解自己的领域，并且能够为下一代构建东西，
51:12
using AI to do that really rapidly like AI scales expertise phenomenally.
用AI来快速实现，因为AI对专业知识的放大效果非常惊人。
51:17
And I do speak to startups where they're kind of, there's an arc of time and this is
我确实和创业公司聊过，他们有点像是，有一条时间弧线，这就是
51:22
how I view us via Amazon right, which is like five years ago VCs used to tell me you're
我对我们的看法，就好比亚马逊，对吧，就像五年前，VC们总是跟我说，你
51:27
never going to be Amazon.
永远成不了亚马逊。
51:28
Well, five years later, we've got the largest databases from multiple their products running
好吧，五年后，我们拥有了最大的 databases，他们的多个产品都运行在
51:32
on them and the people advocate for us over them constantly is the kind of arc of time
上面，并且人们不断为我们而不是他们发声，这就是那种时间弧线。
51:37
when it comes to startups.
说到创业公司的时候。
51:38
There's an incumbent and there's the rising ones.
有老牌玩家，也有新兴的崛起者。
51:40
And the things that mean they are struggling right now to respond to demand, they're doing
而他们现在难以应对需求，这其实说明他们的业务现在
51:46
great as a business right now.
做得非常好。
51:47
They're the ones with GPUs, but it's difficult for a 50,000 person organization and we have
他们才是拥有 GPU 的人，但对一个五万人的组织来说这很难，而我们只有
51:54
a 50 person engineering team, they have 5,000 people engineering on the competing products
一个50人的工程团队，他们却有5000人在开发与我们竞争的产品，
51:59
to us and we still beat them.
而我们依然打败了他们。
52:01
This is because we can move faster.
这是因为我们行动更快。
52:02
We are nimble.
我们很灵活。
52:03
We are involved.
我们参与其中。
52:04
We get there, like large organizations thanks AI, although they are trying to keep everyone
我们做到了，就像大型组织一样，多亏了AI，尽管他们试图把所有人
52:09
out like they are at risk because they can't move fast enough.
挡在外面，好像他们处于风险之中，因为他们动作不够快。
52:12
You just like the only masses of people to do stuff anymore.
你们就像是仅剩的那群还在做事的人。
52:15
And so it's interesting, like I'm seeing some young companies that come with a different
所以这很有趣，比如我看到一些年轻公司，它们带着一种不同的
52:19
angle of seeing things and they're going to be truly disruptive, but I'm seeing some
看问题的角度，而且它们会真正具有颠覆性，但我也看到一些
52:22
wash against the waves of just the facts of life and the facts of how some software dynamics
在生活现实的浪潮和某些 software dynamics 的实际情况面前碰壁。
52:29
are.
是。
52:30
Again, that may or wash out in 10 years time, it might be completely irrelevant to have
再说，那些可能在十年后就会消失，完全无关紧要的是去拥有
52:34
anything we've built.
我们建立的任何东西。
52:35
We might have a God computer that can just do everything.
我们可能会有一台神级计算机，什么事都能做。
52:38
We're not there yet.
我们还没到那一步。
52:39
We've got one more cycle maybe before that, maybe a few who knows, where some great companies
在那之前我们可能还有一个周期，也许几个，谁知道呢，在那个周期里，一些伟大的公司
52:43
will be built by very experienced people, scaling their expertise through AI.
将由非常有经验的人建立，通过AI scaling他们的专长。
52:46
So it's actually an amazing, I've never had more fun.
所以这真的很奇妙，我从来没有这么开心过。
52:51
We have a customer who they are just phenomenal at what they do and they're young and they're
我们有一个客户，他们在自己做的事情上简直太厉害了，而且他们很年轻，正在为他们的受众打造产品，他们真的非常出色，而且他们在消费领域扩张得极快——在消费领域，体验真的会要你的命。
52:56
building for their audience and they are just absolutely outstanding and they're scaling
你不能在这些事情上太老成，但说到 infrastructure，就是有一些做 infrastructure 的方式会持续相当长一段时间，而且有规则来决定怎么做，我觉得在一段时间内就是这样。
52:59
incredibly rapidly when it comes to consumer, when all these places experience actually
帮我理解一下你的 hardware store，因为这是我自己比较头疼的事。
53:04
kills you.
会把你搞死。
53:05
You can't be too old for these things, but when it comes to infrastructure, there are
做这些事情你永远不会太老，但谈到 infrastructure 时，就是有一些
53:10
just ways of doing infrastructure that will persist for quite a while and there's rules
做 infrastructure 的方式会持续相当久，而且还有规则。
53:14
to how that gets done and that's how it's going to be for a while, I think.
至于这是怎么做到的，我觉得短期内都会是这样。
53:20
Help me understand your hardware store because this is something that I'm personally challenged
帮我理解一下你们的硬件库，因为这是我个人觉得比较棘手的地方。
53:23
with is like how do you compete and innovate against Amazon or even Google whenever you
这就好比说，你怎么和 Amazon 甚至 Google 竞争和创新，每当你要
53:31
deliver the tests or Niki or planet scale to GCP or AWS?
把 tests 或 Niki 或 PlanetScale 部署到 GCP 或 AWS 上？
53:37
How do you compete and innovate in their playground so to speak, their hardware, their world?
你怎么在他们所谓的游乐场里竞争和创新？在他们的硬件、他们的世界里？
53:44
Do you build your own machines?
你们会自己造机器吗？
53:45
What is your story?
你们的故事呢？
53:46
We do not build our own machines.
我们不自己造机器。
53:48
I think when we named metal, I think it was a good name at the time.
我觉得当我们命名 Metal 的时候，那在当时是个不错的名字。
53:51
I think it was kind of obvious what it meant, but people have conflated it with us running
我觉得它的意思挺明显的，但人们把它和我们在运行...混为一谈了。
53:56
our own data centers.
我们自己的 data centers。
53:57
That is not true.
那不是真的。
53:58
I thought you did.
我还以为你们有呢。
53:59
You bought some MVMEs and wrapped them in stacked them.
你们买了一些 MVMEs，然后把它们包起来堆叠在一起。
54:02
I'm glad you then thought that so I can correct that, which is we use the bare metal machines
我很高兴你当时这么想，所以我可以纠正一下：我们用的是 bare metal machines，
54:07
that Google and Amazon give us and we build on top of that, so that's how we compete.
就是 Google 和 Amazon 给我们的，然后我们在此基础上构建，这就是我们竞争的方式。
54:11
We take the same machine they do and build our entire software stack on top of it, which
我们采用和他们一样的机器，并在其上构建我们的整个 software stack，
54:15
means the profile at the cost profile of the machine is low because they don't really
这意味着机器的 cost profile 很低，因为他们并不真的...
54:19
put much margin on top because they don't do anything, but the most impressive thing is
在上面加了很多利润，因为他们什么都不做，但最令人印象深刻的是
54:22
those machines are a femoral.
那些机器就是命脉。
54:24
If we lose one of those machines or we give one back by accident and we miss schedule
如果我们丢了一台机器，或者不小心退回去一台，然后在 Kubernetes 里错过了某个调度或者别的什么，我们就会丢掉客户的数据。
54:28
something in Kubernetes or whatever, we would lose a customer's data.
我们有差异化，没有其他初创公司这样做。
54:33
We differentiate and no other startups are doing this.
有少数几家认为自己是这么做的。
54:36
There's a few that think they are.
有少数人觉得自己是这样。
54:38
It's hilarious.
太搞笑了。
54:39
We can see the code and they're going to lose data.
我们可以直接看到 code，然后他们就要丢数据了。
54:44
We differentiate by running databases in state and we run more state on Kubernetes than
我们的差异化在于运行 state 数据库，我们在 Kubernetes 上运行的 state 比
54:49
anyone on Earth.
地球上的任何人都要多。
54:50
We just actually released a really amazing Fatty on our team released a great blog post
我们刚刚实际上发布了一个非常棒的——我们团队的 Fatty 发布了一篇很棒的博客文章
54:54
about Kubernetes.
关于 Kubernetes 的。
54:55
It's just a wonderful, masterful piece of writing.
这真是一篇精彩、大师级的文章。
54:57
I recommend anyone read it.
我推荐每个人都去读一读。
55:00
We build from the base up, so it means we can perform as fast or better than their solutions.
我们从底层开始构建，所以这意味着我们的性能可以与他们一样快，甚至超过他们的解决方案。
55:06
With all of our dx, agentic ax on top of it, the higher up time we are, we compete with
有了我们所有的 dx、agentic ax 加持，我们的 uptime 越高，我们就与
55:14
them by having higher up time and just generally a better product while inside the cloud.
通过更高的 uptime 和总体上更好的产品，在云内部实现这一点。
55:19
This is the thing.
关键就在这里。
55:20
There's a few Neo clouds that have built really well outside of the major clouds, but
有一些 Neo clouds 在主流云之外做得很好，但是
55:27
if you want to sell to large companies and companies that are far beyond startups, you
如果你想卖给大公司，以及那些远超初创阶段的公司，你
55:33
have to be in AWS and Google because of latency issues and all of these things, we have
就必须在 AWS 和 Google 里，因为延迟问题等等这些原因，我们得
55:39
to be there.
在那里。
55:40
The fact we found the perfect hybrid between on-prem performance and scalability, well, no,
事实上，我们找到了 on-prem 性能和 scalability 之间的完美结合，嗯，不，
55:47
we add the scalability on-prem performance.
我们是在 on-prem 性能上加上 scalability。
55:50
The hybrid, the hype scares cloud, is why we get to do what we do.
Hybrid 和炒作把 cloud 吓到了，所以我们才能做我们现在做的事。
55:54
We also run in the customers account too.
我们也在客户的 account 里运行。
55:56
It is not on-prem.
这不是 on-prem。
55:58
You still go to planscale.com, it's like, bring your own cloud, which I think you and I
你还是去 planscale.com，就像是 bring your own cloud，我觉得这个你我在 2022 年就聊过。
56:02
will have talked about in 2022.
我们在这方面起步很早。
56:03
We were early to that.
我们很早就那么做了。
56:04
It was one of the greatest things we did.
那是我们做过的最棒的事情之一。
56:06
We have gigantic customers running it because they are customers of Google and Amazon.
我们有超大型客户在运行它，因为他们是Google和Amazon的客户。
56:11
We run inside their accounts, and so it means that they secure it, they pay the bill,
我们是在他们的账户里运作的，所以意思是他们负责安全，他们付账单。
56:16
they get great discounts when you're committing $1 billion to Amazon, you get a decent discount
当你向 Amazon 承诺投入 10 亿美元时，你会得到很大的折扣，相当可观的折扣。
56:22
and so they can use that as an option.
所以他们可以把那当作一种选择。
56:24
We can get the total cost like our cost plus the discount on those hardware costs.
我们算总成本的时候，就是我们的成本再加上那些 hardware 成本的折扣。
56:29
We can often be Amazon on price, and so that's why that's how we compete.
我们经常能在价格上赢过 Amazon，所以这就是我们竞争的方式。
56:38
It's interesting because it's almost hypocritical, but you can't help it, right?
这很有意思，因为几乎有点虚伪，但你又没办法，对吧？
56:46
If you just said in one second, they're terrible partners alluding to Amazon, but yet you
如果你一秒钟前还说他们是很糟糕的合作伙伴，暗指 Amazon，但你还是
56:54
have to play in their world, so it's like a masked party and maybe a masquerading kind
不得不在他们的世界里玩，所以这就像一场化装舞会，也许是一种伪装。
57:01
of thing.
之类的东西。
57:02
A party where you have to go with a mask on, but you're still kind to people.
一个你必须戴着面具去参加的派对，但你仍然对人们很友善。
57:07
It's this really weird scenario where you have to play nice but not quite so nice.
这是一种非常奇怪的场景：你得表现得友好，但又不能太友好。
57:12
I massively respect them for the position of leverage they've put themselves in.
我非常佩服他们，因为他们让自己处于一个很有筹码的位置。
57:16
AWS is a phenomenal business built by phenomenal people, and we should have nothing but respect
AWS 是一项了不起的业务，由一群了不起的人打造，我们对他们只有尊敬。
57:21
for them.
与此同时，他们也必须在与运行在自己平台上的那些公司竞争之间取得微妙的平衡，对吧？
57:22
At the same time, they have to strike a fine balance of competing with the people that run
与此同时，他们还得跟那些跑在他们平台上的人竞争，这中间得有个微妙的平衡，对吧？
57:29
on them, right?
各位，我认识其他一些CEO，他们和Amazon有很好的合作，因为他们竞争的是Amazon不太重要的服务。比如说，你肯定——我是说，我刚遇到你的一位客户。
57:30
And how long people will tolerate that?
那人们能忍受多久呢？
57:33
Now, luckily there's that undoing as well because they've built like what, 350 services
现在呢，幸好也有那种抵消作用，因为他们已经搞了大概350个services。
57:37
and now there's 10 competitors for every one of those services, and it hurts for them
而现在每一个service都有10个竞争对手，这对他们来说很痛苦。
57:42
because they have to keep up and people want the best in class tools.
因为他们必须跟上节奏，而且人们想要的是同类最佳的工具。
57:45
Unfortunately, the database is some of the services that make the absolute most money
不幸的是，database是Amazon最赚钱的services之一。
57:49
for Amazon.
各位，我认识其他CEO，他们和Amazon有很好的合作关系，因为他们竞争的是Amazon不那么重要的services。比如，你肯定，我是说，我刚见到了你的一个客户。
57:51
People, I have other CEOs who have great partnerships with Amazon because they compete
这不应该成为问题。
57:55
with less important Amazon services, say, you must, I mean, I just met one of your customers
其实特别有意思，有个创始人——我忘了是哪家公司了——在Twitter上发了一张截图。
58:00
and they told me how much they spend on Amazon, how much they spend on you, like Amazon
然后他们告诉我他们在 Amazon 上花了多少钱，在你身上花了多少钱，看来
58:04
must love you.
Amazon 肯定很喜欢你。
58:05
And I was like, we can have a terrible time because we're competing with the software
然后我就想，我们可能会很惨，因为我们在和 software 竞争，
58:08
that they're trying to put the services they're trying to protect the most, and so yeah,
他们试图用这些 software 来保护他们最看重的服务，所以，
58:12
they behave really badly in that world.
他们在那个世界里表现得非常糟糕。
58:15
And so they just have to be careful.
所以他们必须小心。
58:16
And I hope we don't make the same with some of the large model companies that constantly
而且我希望我们不会对某些 large model 公司犯同样的错误，那些公司总是
58:21
talk about how they're going to replace everyone.
说要取代所有人。
58:22
It's like, I would caution anyone from allowing that to happen again.
就是说，我要告诫任何人，不能再让这种事情发生了。
58:27
It was a, with AI, it could be dangerous.
这是……有了AI，可能会很危险。
58:29
Like it is really bad that you can be an Amazon customer and that they can see you working
比如，作为Amazon的客户，他们能看到你在和别的商家合作，还能看到他们的销售情况，然后就用这些信息去联系对方，这真的很糟糕。
58:35
with someone else and their sales, and then they reach, use that to reach out to that.
而且他们确实会这么做。
58:39
And they do do that.
而且这极其糟糕。
58:40
And it is extremely bad.
而且对此完全没办法。
58:42
And there's nothing with it.
这种事根本不该存在。
58:43
There should not be a thing.
这就不应该成为一件事。
58:44
It happens.
这种事确实有。
58:45
Our customers tell us constantly.
我们的客户经常这么说。
58:47
And so they have to be careful because, like, you don't get to rain forever.
所以他们得小心，因为，怎么说呢，你不可能永远下雨。
58:52
There's other clouds, and Microsoft actually is not like that.
还有其他的 clouds，而 Microsoft 其实不是那样。
58:56
On Azure, their sales team, the way they operate, is much more friendly to the ecosystem
在 Azure，他们的销售团队，他们的运作方式，对 ecosystem 要友好得多。
58:59
of people that work on them.
也就是那些在它们上面工作的人。
59:01
Amazon particularly really do not do that.
Amazon 尤其真的不会那样。
59:05
They pretend to, they suck you in.
他们假装会，然后把你吸进去。
59:07
Like, they have released, it's a, it's a, it's of a company that I think most people
就，他们发布了一个产品，是一家公司的，我觉得大多数人认为这家公司能和我们比一比，但其实并不是。
59:12
think is a compared to us, not isn't really.
但 Amazon 今天刚发布了一个和他们竞争的产品，而去年还在 re-invent 上让那家公司上台了。
59:15
But Amazon just released a competing product to them today after having that company on stage
所以他们就会把你拉进来。
59:19
at re-invent last year.
其实挺搞笑的，有个创始人——我忘了是哪家公司——发推贴了一张他们收到的来自 Amazon Corp 的邮件截图。
59:21
So they will pull you in.
他们大概就是说：“哦，嘿，你知道吗，看起来你们做得很好。”
59:22
It's so funny actually, a founder, I can't remember what company it was, tweeted a screenshot
其实特别搞笑，有个创始人，我忘了是哪个公司了，他发了一条推文，配上截图。
59:28
of an email they got from Amazon Corp.
说起他们从 Amazon Corp. 收到的一封邮件
59:31
They've been like, oh, hey, you know, it looks like you're doing really well.
他们就说，哦，嘿，你懂的，看起来你干得挺不错的
59:33
We'd love to reach out and see how we can help.
我们很乐意主动联系，看看有什么能帮上忙的。
59:36
That was sent like six months before Amazon released a competitive product.
那是在 Amazon 发布竞品大约六个月前发的。
59:39
When I reached out to that founder and showed him an identically email from me, I just
当我联系那位创始人，给他看我发过的一封一模一样的邮件时，我就是
59:43
didn't pick up the phone because I've been tired of this for so long, there's no way
没接电话，因为我对这种事已经厌倦太久了，绝不可能
59:46
I'm talking to any of them.
我会跟他们任何人谈。
59:48
Yeah, it's like, are you a fan of Silicon Valley money chance?
对，就好像，你是 Silicon Valley money chance 的粉丝吗？
59:51
The TV show?
那个美剧？
59:52
It's too close to.
太贴近了。
59:53
I've watched, I've enjoyed watching it and I've met the producers, they came to GitHub
我看过，也很喜欢看，而且我见过制片人，他们来 GitHub 做过调研，我有个朋友是他们的顾问之一。
59:57
when they were doing research and I know a friend of mine is one of their advisors.
他们把硅谷模仿得太像了，以至于我自己就生活在 Silicon Valley，看着都觉得难受。
60:00
They did a great job of parroting it to the degree where I find it a struggle to watch
很难看得下去。
60:04
when I live in it in Silicon Valley.
而且有些真实道具就是从 GitHub 来的，甚至有一场他们收到放在门口的周边商品的戏，那件道具就是 GitHub 提供的。
60:06
It's hard to watch.
不过嘛，反正我是看不下去，这就是绕了一大圈在说“不”。
60:07
And some of the actual props were from GitHub and even from the scene where they get delivered
而且有些实际的道具来自 GitHub，甚至包括那个在门口收到周边商品的场景，GitHub 也提供了那个道具
60:12
merch on the doorstep, GitHub provided that as a prop.
不过，是啊，对我来说真的有点看不下去，那就是一个绕一大圈、最后等于说不的回答
60:17
But yeah, like, it's hard to watch for me, it's a long-winding answer of saying no.
我们的观众可能不了解 Work OS，但开发者们会面临哪些挑战呢？比如开始一个新项目，选择合适的工具，选择合适的 database，选择合适的 off
60:21
I've seen some of it and it is good.
我看过一些，挺不错的。
60:22
Mike judges are genius, but when you kind of know people that are just exactly the same
Mike Judge 是个天才，但当你认识的人跟那些角色一模一样的时候，就觉得有点过了，而且我也不太看电视。
60:27
as the characters, it's a bit much for you, plus I don't really watch TV.
所以他们用极其冗长的方式跟你解释那件事。
60:29
So they go extremely long-winded way of explaining that to you.
答案是不。
60:32
The answer's no.
朋友们，我回来了，和我的一位好朋友 Michael Grinich 一起。
60:37
My friends, I'm back with a good friend of mine, Michael Grinich.
Michael，我知道我很喜欢 Work OS。
60:40
Michael, I know that I love Work OS.
我们的观众可能不了解 Work OS，但 developers 面临的挑战是什么，
60:42
Our audience may not know about Work OS, but what are the challenges developers face,
我们的听众可能不知道 Work OS，但开发者面临的挑战是什么，
60:46
starting a new project, choosing the right tools, choosing the right database, choosing
启动新项目、选择合适的工具、选择合适的数据库、选择
60:50
the right off.
合适的 off。
60:51
Take me there.
带我过去吧。
60:52
When a developer starts a new project, the decisions that they make at the very beginning
当一个开发者开始一个新项目时，他们最开始做的决定会产生非常持久的影响。
60:56
end up having long-lasting consequences.
你用什么 language 构建，在什么 platform 上构建，选择什么 database。
60:59
What language you build in, what platform you build on top of, what database you choose.
这些都是之后很难改变的事情。
61:03
These are things that are very hard to change later on.
所以它们会产生重大影响，尤其是如果它们限制了你 scale 的能力。
61:05
So they have major consequences, and especially if they limit your ability to grow in scale,
也就是要确保你选了一个能和你一起 scale 的 platform。
61:10
at some point, as the product starts to take off, you're going to have to stop developing
到了某个阶段，随着产品开始起飞，你就得停止开发
61:15
new product features and go re-architect to rebuild your system.
新功能，然后去 re-architect，重建你的系统。
61:19
And that might be a killing blow right at the moment you need to accelerate.
而那可能在你需要加速的时候成为致命一击。
61:23
So these decisions are really, really important, and I think that's why developers gravitate
所以这些决定真的非常非常重要，我觉得这就是为什么开发者们会倾向于
61:27
toward solutions that are mature, things that they know that will scale, even things that
那些成熟的解决方案，那些他们知道能 scale 的东西，甚至那些
61:31
are open source.
open source 的东西。
61:33
You're going to pick something like planet scale for your database provider, not because
你会选像 PlanetScale 这样的作为你的 database provider，不是因为
61:37
it's the cheapest, or because it's the most fun to use, but because you know it's going
它最便宜，也不是因为它最好玩，而是因为你知道它将
61:41
to be a durable provider that you can scale on for years.
成为稳定可靠、能让你 scale 很多年的提供商。
61:45
And WorkOS is like that for Auth.
而 WorkOS 在 Auth 领域就是这样的存在。
61:47
At the earliest, earliest days, if you look across all these different services, they kind
在最早最早的时候，如果你看所有这些不同的 services，它们看起来都差不多，但到了 Day 1,000 或 Day 2,000 或 Day 10,000，你会希望自己当初确保选的是一个能和你一起 scale 的 platform。
61:50
of look very similar, but at Day 1,000 or Day 2,000 or Day 10,000, you're going to want
而今天，WorkOS 正在为所有这些 AI 公司提供 Auth、Identity、Security 和 Permissions，它们 literally 是世界上增长最快的公司，比如 OpenAI、Traffic 和 Persever for Plexi，WorkOS 就在它们背后。
61:56
to have made sure that you picked a platform that could scale with you.
而今天，WorkOS 正在为所有这些提供 Auth、Identity、Security 和 Permissions 的支持。
61:59
And today, WorkOS is powering Auth and Identity and Security and Permissions for all of these
而今天，WorkOS 正在为所有这些提供 Auth、Identity、Security 和 Permissions 的支持。
62:04
AI companies, literally the fastest growing companies in the world, like OpenAI and
AI公司，真的是世界上增长最快的公司，比如OpenAI和
62:07
Traffic and Persever for Plexi, WorkOS is under the hood there.
Traffic和Persever for Plexi，WorkOS就在那背后。
62:11
So I think when people pick WorkOS early on, really what they're doing is trying to pick
所以我觉得，当人们早期选择 WorkOS 时，他们真正在做的是试图选择
62:15
the defaults to allow them to grow in rapidly scale.
那些 defaults，让他们能够快速 scaling、不断成长。
62:19
And there's no platform other than us that's done that at the same level.
而且除了我们之外，没有其他平台能做到同样水平。
62:22
Well, friends, the next step is to go to WorkOS.com.
好，朋友们，下一步就是去 WorkOS.com。
62:27
Find it today, check it out, free for a million active users, try it today.
今天就找到它，看看吧，对一百万活跃用户免费，今天就试试吧。
62:33
There's no excuse not to, it is your default.
没有借口不选它，它就是你的 default。
62:35
You should choose it.
你应该选择它。
62:37
So do so.
所以，就这么做吧。
62:38
WorkOS.com.
WorkOS.com。
62:39
Once again, workOS.com.
再说一遍，WorkOS.com。
62:48
That's okay because I won't ruin the plot by any means, but I'll just give a glimpse.
没关系，我绝不会剧透，只是稍微透露一点。
62:53
I think it was in season two, it may have been in season one, they were brought in by a venture
我记得是在第二季，也可能是第一季，他们被一家风投公司引入，也就是Pied Piper那帮人。
63:01
capitalist firm, they being pie-piper.
他们在创新这个压缩算法。
63:04
And they were innovating on this compression algorithm.
显然他们在找钱，因为任何创业公司都需要资金来发展。
63:08
And they were hunting for money, obviously, because like any startup you need money to grow.
我想他们是被“brain reaped”了，这大概才是最合适的说法。
63:14
And they were brain reaped, I guess, is probably the proper term.
然后他们被挖角了，我想这大概是正确的说法。
63:18
I don't know if that's the proper term or not.
我不知道那算不算正确的说法。
63:21
So similar.
挺类似的。
63:22
It's like, let's email you, let's get you into the room and ask you how your thing
就像是，给你发邮件，把你叫进房间，然后问你的东西
63:25
works with no intention to invest because they were not intending to invest.
是怎么运作的，但他们完全没有投资意图，因为本来就没打算投资。
63:30
And because they're naive and they're early in their careers, they gave away information.
而且因为他们很天真，又刚入行，所以就把信息都透露了。
63:36
They were happy to like, gosh, in these venture capitalist meetings, no one ever asks
他们很高兴，像是，天哪，在这些 venture capitalist 会议上，从来没人会问
63:41
us how the platform works.
我们 platform 是怎么运作的。
63:43
Let's just whiteboard the thing for you.
干脆把这个东西 whiteboard 给你看吧。
63:45
So they literally whiteboard their crown jewels.
所以他们真的把自己的核心机密直接画在白板上了。
63:50
And now they have competitor, let's just say, before they did not now middle out
而现在他们有竞争对手了，可以这么说，以前他们没有。现在 middle out。
63:55
that was pretty much the crux of pie-piper was now given to a funded team who is capable
那基本上就是 pie-piper 的核心，现在却给了一个有能力、有工程师的获投团队。
64:05
with engineers and they go and build the same thing.
有了工程师，他们就去把同样的东西做出来。
64:07
Now they have competitor.
现在他们有竞争对手了。
64:08
This happens all of the time.
这种事一直都在发生。
64:09
Yeah.
是啊。
64:10
This is a real thing.
这是真的。
64:11
You see a story once a week of someone on Twitter being like, oh yeah, actually this
你每周都会在 Twitter 上看到有人说，哦对，实际上这家
64:16
company offered to try and buy us and do due diligence and they use that really poorly.
公司提出想收购我们，做 due diligence，但他们搞得很糟糕。
64:21
And it's very bad.
这非常糟糕。
64:25
I chat to young founders and founders that are new in this and it's amazing how naive
我和年轻创始人、还有刚入行的创始人聊天，真的很惊讶人们有多天真，
64:33
people can be to thinking that these giants are here to help and that it's not immensely
竟然会认为这些巨头是来帮忙的，以为这不是极其
64:39
cut.
残酷的。
64:40
And it's getting worse.
而且情况越来越糟。
64:41
There is so much on the line.
这关系到太多东西了。
64:43
Like people are terrified by AI.
就像人们对AI感到非常恐惧。
64:45
They are terrified of what it's doing to them, even though they should not be.
他们害怕AI会对他们做什么，虽然这种担心其实没必要。
64:50
It's causing this to happen much, much more frequently.
这正在让这种情况发生的频率高了很多很多。
64:54
And yeah, you have to be very careful and very secretive about what you do.
而且，你得非常小心，对自己做的事情要高度保密。
64:58
We just don't, we try, like we have learned to the lessons and we try not to fall for these
我们就是不会去做——我们尽量，比如已经吸取了教训，努力不上这些当。
65:02
things.
是啊，光是来播客让我问你可能涉及某些领域的问题，就已经很有挑战性了。
65:03
Yeah, it must be challenging to even come up on a podcast to have me ask you questions
是啊，光是来上播客让我问你问题就很有挑战性，
65:08
that might go into an area.
而且问题可能会触及某个领域。
65:09
I'm sure you're protective of it, but nonetheless, there's some anxiety to some degree like
我相信你肯定会保护它，但尽管如此，还是有一定程度的焦虑，比如说
65:14
I've got to be in the public because I want to promote what we're doing and I'm the CEO.
我必须出现在公众面前，因为我想推广我们正在做的事情，而我是CEO。
65:18
You have to do your job and there's a job.
你必须履行你的职责，而这就是一份工作。
65:21
One of the jobs at the CEO is to be some version of a figurehead to the company and the spokesperson.
CEO的工作之一就是以某种形式担任公司的门面和发言人。
65:26
And obviously speak eloquently and have clear thoughts and share your ideas as well.
而且显然要口才出众、思路清晰，并且分享你的想法。
65:31
But then there's the part that's like, well, now I'm at risk because like if I share too
但还有一部分是，好吧，现在我有风险了，因为如果我分享太多，
65:35
much, you know, how much do I share, how, how do I navigate this, this new world we're
你知道，我该分享多少，怎么，怎么去应对这个，这个我们身处的新
65:43
in?
世界？
65:45
How do you, how do you personally handle that yourself with being in the public because
你自己是怎么应对这件事的？作为公众人物，因为
65:49
you're pretty, you're pretty well spoken in two ways, well spoken, but then also I hear
你确实、确实很会说话，两种意义上都很会讲，但我也听
65:54
you and I see you on X and I'm like, dang, there's Sam going toe to toe with like benchmarking
你在X上说的，我一看到就想，天哪，Sam又在那儿跟人家就benchmarking针锋相对，
66:00
and like calling them out and like, you're in the most positive way, ruthless, you
还直接点名批评，而且是以一种最正面的方式，毫不留情，你
66:05
know?
懂吗？
66:06
Like I love that about you.
我真的很喜欢你这样。
66:07
I love the way you come at folks in a way, but how do you, how do you manage that the ability
我喜欢你跟人交锋的方式，但你是怎么做到既能
66:12
to sort of show up somewhere like this, but then also not give it all away?
像这样出现在某个场合，又不会把所有东西都展现出来？
66:17
I think about this very often is a constant evolving debate.
我经常想这个问题，这是一个不断演变的辩论。
66:21
So we're very open.
所以我们非常开放。
66:23
We're known for our technical content is a great example, right?
我们以我们的技术内容而闻名，这就是一个很好的例子，对吧？
66:25
So how do you share really deep insightful things that have a meaning?
那么，你如何分享那些真正有深度、有意义的见解呢？
66:30
You can tell when tech content is just like fluff, right?
你能分辨出tech content什么时候只是空洞的废话，对吧？
66:33
How do you talk about really deep things about giving away proprietary things?
你怎么谈论那些真正深入的、关于分享proprietary东西的事？
66:38
I talk about this all the time at the beginning of every one of our all hands.
我在每次all hands的开头都会讲这个。
66:42
There's a giant slide that says confidential on it and we're very clear and I remind the
有一张巨大的幻灯片，上面写着confidential，我们非常清楚，而且我提醒……
66:50
employees that we are incredibly open at that scale.
在那个规模下，我们对员工是非常透明的。
66:53
Like if you're an employer, you know, you can look at how our finances are.
比如说如果你是雇主，你可以看看我们的财务状况。
66:58
If when we do board meetings, we present the board deck for the entire company.
我们开董事会的时候，会把董事会报告展示给全公司。
67:03
I believe in hiring exception, good people, I believe, I mean, I cannot self assign this
我相信招募优秀的人，我确实相信，我是说，我不能自己给自己贴这个标签。
67:10
but it's something I'm aware of that people would view plan scales having quite a high
但我知道，人们会认为 plan scales 有很高的
67:14
talent bar and being a small engineering team that's clearly boxing above our weight.
人才门槛，而且作为一个很小的工程团队，明显是在超水平发挥。
67:23
I do not believe in hiring incredible people and then hiding anything from like every business
我不相信招来优秀的人然后对他们隐瞒任何事情，就像每个商业
67:28
problem can be engineered.
问题都可以用工程手段解决。
67:29
We are only engineers to engineer ourselves towards building a phenomenal business.
我们只是工程师，用工程的方式打磨自己，来打造一家非凡的企业。
67:33
So I share very, very liberally internally, but we have to remind people that you you
所以我在内部分享得非常非常开放，但我们得提醒大家，你你
67:39
might not like, it's very, unless you are essentially myself or like the sea level at the company,
可能不会喜欢，这很难，除非你基本上是我本人，或者像公司的C-level一样，
67:46
it is very hard to know whether the thing you are doing or working on or you might just
真的很难知道你正在做或着手的事情，或者你可能只是
67:50
say it plays an important role in a very big picture that we don't want to get out there.
说它在某个很大的图景中扮演重要角色，而这个图景我们不想公开。
67:57
So we guard very, you know, we're very protective over what we do, our roadmap, what we're working
所以我们非常小心，你知道，我们对自己所做的、我们的路线图、我们正在做的
68:05
on.
事情非常保护。
68:06
Like when we built Postgres, we have to work in secret like you had to have a code name.
就像我们当初构建Postgres的时候，我们必须秘密进行，还得有个代号。
68:08
So that when we could ship front iron end code because we're iterative, it wouldn't be
这样，当我们因为迭代而发布 front end code 时，就不会被提到。
68:12
referred to.
这样，当我们在 South Park 时，就可以谈论我们在做什么，而他们不会知道。
68:13
So that when we're in South Park, we can talk about what we're like working on without
我们在这方面非常保密，但我们必须把握好一条微妙的线：既要保持开放，比如让我们的市场团队、我自己和其他人代表公司发言，又不能泄露、不能搞砸事情。
68:18
them knowing.
所以这需要打磨，需要文化，也需要把这些问题想清楚。
68:19
Like we are very secretive in that way, but we have to do a fine line of like being open
我们在这方面非常保密，但我们必须把握好分寸，既要开放，
68:23
and like allowing our marketing team and myself and folks to go and speak on behalf of
又要允许我们的市场团队、我自己和其他人代表公司去发言，
68:28
the company and not leak and mess things up.
同时不泄露任何信息，也不把事情搞砸。
68:30
And so it just takes refinement and it takes culture and it takes thinking that through
所以这需要打磨，需要文化，也需要把事情想透。
68:34
like I could, like I could be as opposed to we did, but that is amazing.
就好像我可以——我是说，我可以做到，而不是我们做到，但那太厉害了。
68:37
It goes very far into showing you how well we think about systems without exactly kind
它在很大程度上展示了我们对系统的思考有多深入，同时又不完全算是把王国的钥匙交给你。
68:43
of giving you the keys to the kingdom.
所以这是一种持续的平衡，而且这种平衡——再说一遍——会随着时间改变，但我会努力。
68:44
And so it's this constant balance that will, again, it will change over time, but I try.
说到 Twitter，Twitter 对我来说挺有意思的。
68:49
When it comes to Twitter, Twitter's a bit of fun for me.
差不多一半的时候我并不当真，我就是不介意说一些我知道会让大家炸锅的话，而且我觉得这太容易了，尤其是当这对我没有任何影响的时候，真的很好笑。
68:52
Like half the time I don't take things seriously, I just don't mind saying things that
比如很多时候我都不当回事，我就是不介意说一些
68:56
I know is going to make people all these their minds and it's like, it's just funny
我知道会让人失去理智，而且就是觉得很好笑
68:59
to me how easy it is, especially also when it just doesn't impact me.
对我来说这太容易了，尤其是当这根本不影响我的时候。
69:03
Like I don't get anxiety about this.
就像我对这个一点都不焦虑。
69:05
I can have a fight with someone online and then just close my phone and never think about
我可以在网上跟人吵一架，然后关掉手机，再也不会去想它。
69:10
it.
我照样睡得跟婴儿一样香。
69:11
I can sleep like a baby.
所以这真的不会困扰我，但有趣的是，当它真的激怒了你的一些竞争对手，他们会因为我的所作所为而失态，然后在LinkedIn上发帖。
69:12
So it doesn't really bother me, but it's funny when it really lights up some of your
但还有一部分原因是，我在社交媒体上的这种态度，是因为我不相信让那些不实之词或关于你公司的说法流传开来，而不立刻去反驳。
69:15
competitors and they lose it and what I've done, do some LinkedIn posting.
竞争对手就会失态，而我所做的只是发些LinkedIn帖子。
69:18
But also like partly the way I am on social media is because I do not believe in letting
但还有一部分是，我在社交媒体上这样，是因为我不认为应该让
69:25
in false truths or things come be said about your company and not jump on them immediately.
不实之词或关于你公司的话出现，却不立刻反驳。
69:30
I've worked at places where I've watched other companies that just get drowned on social
我在一些地方工作过，见过别的公司在社交媒体上被不实信息淹没，而他们太克制、太礼貌，根本不去压制这些不实说法。我会以极快的速度去处理这些东西，在公开记录里纠正。但这次不同。我说得好像这家公司有很多drama似的，但实际上真没有。
69:35
media with things that are not true about that company and they are too reserved and polite
媒体上关于那家公司的不实消息，而他们太克制、太礼貌了。
69:40
about it.
关于这件事。
69:41
They don't squash these mis-truths.
他们不会打压这些不实之词。
69:44
I go at that stuff incredibly quickly to correct it in the public record.
我处理那些东西快得惊人，就为了在公开记录里纠正。
69:47
Not this.
但这个不是。
69:48
I make it sound like there's tons of drama around this company.
我说得好像这家公司戏很多。
69:51
There really is not.
其实真没有。
69:52
But people come up and say things and they try, they do silly things and you just call
但人们会过来说些话，他们尝试，他们做傻事，你就直接指出来，直说就是这样。
69:57
it out and just say it fully is and it's just a way of handling PR and public relations
这只是在处理PR和公共关系而已。
70:01
and sometimes companies misstep and you can get out there and correct them really quickly.
有时候公司会犯错，你可以迅速站出来纠正。
70:05
And I watch companies just getting absolutely destroyed in the public narrative and their CEO
我看到有些公司在公众舆论里被彻底摧毁，而他们的CEO就应该出来纠正这些事，说出来，直面他们，诸如此类。
70:10
should just be out there correcting these things, saying it, confronting them, whatever.
比如有人对我说过恶意刻薄的话。
70:14
Like I've had people say hateful mean things.
我也收到过几次死亡威胁。
70:18
I've had the odd death threat too.
我经常和这些人见面。
70:21
I often meet the people.
我经常见到这些人。
70:22
You know, like you just go and say hi and talk to them and make friends.
你知道，就是直接过去打个招呼，跟他们聊聊，交个朋友。
70:26
And then I have a few people that are now talking to you very right.
然后我有几个人，现在跟你聊得特别来。
70:29
It's a bit of a meme internally that I've had people that have been real haters.
在内部有个梗，就是我遇到过一些真正的黑粉。
70:33
Like just absolutely hated us and me and now are customers.
就是曾经特别恨我们、恨我的人，现在成了我们的客户。
70:38
And you just, I just reach out to them and have a conversation with them.
而你就，我就直接联系他们，跟他们聊一聊。
70:40
Like I'm, if someone, if someone is mad and hateful at me and makes a good point,
比如如果有个人对我很生气、很有敌意，但他说得有道理，
70:45
I like you can really flip that by going, yeah, actually you're right.
我觉得你可以通过说“对，你说得没错”来扭转局面。
70:48
I was completely wrong with what I said.
我之前说的完全错了。
70:49
Like I genuinely do try and put controversial things out there, see how people react
就是，我确实会故意抛一些有争议的观点出来，看看大家什么反应。
70:55
and then change my mind if I do believe.
然后如果我真的信了，我就会改。
70:56
And I very often change my mind.
而且我经常改主意。
70:57
I have zero fear changing my mind.
我一点都不怕改主意。
70:59
Like it's my job to do that.
这本来就是我的工作。
71:01
Yeah, a lot can change.
是啊，很多东西都会变。
71:03
Well, a lot can happen in relationships whenever you can be transparent,
嗯，在关系里，只要你能坦诚，很多事都可能发生，
71:09
but then also be willing to have a conversation.
但同时也要愿意去对话。
71:11
Like somebody people just don't have the conversation.
就像有些人根本不会进行那样的对话。
71:14
Correct.
没错。
71:15
It would much rather sub-tweet you or they would much rather, you know, talk about you
他们更愿意发帖暗讽你，或者更愿意，你懂的，谈论你
71:19
behind your back or, you know, round about, so to speak.
在背后，或者，你懂的，拐弯抹角地，这么说吧。
71:22
And I'd be like negative or positive, just sort of like just about you.
然后我就会像，不管是负面还是正面，反正就是关于你。
71:27
But you know, something changes whenever you actually have a conversation with somebody
但你知道，当你真的和某人进行对话时，情况就变了。
71:31
and there's clarity and then there's also just transparency in terms of, you know what?
然后就有了清晰度，也有了透明度，你知道吗？
71:35
I was not actually telling you right here, I've changed my mind here and I don't feel
我其实不是在这儿告诉你，我在这儿已经改主意了，而且我不觉得...
71:38
that way anymore.
不再是那样了。
71:39
Or you were right and I was wrong, you know, that is the way it happens there.
或者你是对的，我错了，你知道，在那里就是这么发生的。
71:43
Absolutely.
绝对。
71:44
I completely agree.
我完全同意。
71:45
That is the way to deal with sub-tweeting by the way.
顺便说一下，处理sub-tweet就应该这样。
71:46
If people sub-tweet you, just immediately go at them, just say, I know that's about
如果有人sub-tweet你，就直接怼回去，就说，我知道那说的是我，然后直接回应。
71:49
me and then just like react to it.
他们会停下来的，因为他们sub-tweet是因为懦弱，你要是挑明了，但是...
71:51
They stop because they sub-tweet because they're cowards and if you call it out, but
他们停下来是因为他们发sub-tweet，因为他们是懦夫，而且如果你指出来，但
71:54
people are too polite, they won't call it out and so it lets all this stuff happen.
人们都太客气了，不会直接指出来，所以就让这些事情发生了。
71:58
It's silly.
这很蠢。
71:59
For those that are listening that maybe be aware, I'm like, I have a zero expectation
对于那些在听、可能有印象的人，我想说，我完全不指望
72:02
that a single person knows who I am that is listening to you.
在听你节目的人里，有哪怕一个人知道我是谁。
72:07
And maybe they do now, obviously that's why I'm here, but you know, maybe you've interacted
也许他们现在知道了，显然这就是我来的原因，但你知道，也许你和我互动过
72:10
or seen my Twitter or whatever.
或者在Twitter上看到过我什么的。
72:12
And there might be people that really think this guy, like this guy, his piece of shit
而且可能真的有人认为这家伙，就像这家伙，他就是个垃圾
72:17
online or whatever, I just would love to let you know that it is one of our largest
网上还是什么的，我只想让你知道，这是我们最大的之一。
72:22
sources of leads.
leads 的来源。
72:24
There is a drop down in our sales force to designate their source being my Twitter.
我们的销售团队里有一个 dropdown，让他们把来源标成我的 Twitter。
72:29
One of the hottest AI companies in Europe came inbound two weeks ago.
欧洲最火的 AI 公司之一，两周前 inbound 找上门来了。
72:35
Whenever I see, I love meeting customers.
每当我看到，我就喜欢见客户。
72:37
If you read blog posts from our customers, it's out there very many times that I will
如果你读过客户的博客，里面很多次都会提到我会
72:43
be the first person they've spoken to and I will also babysit their migrations, I'll
是第一个他们交谈过的人，我也会照看他们的 migration，我会
72:47
be in the snatcher.
在 snatcher 里。
72:48
I love it.
我喜欢这样。
72:49
Talking to customers is one of my favorite.
和客户聊天是我最喜欢的事之一。
72:50
I don't know.
我也不知道。
72:51
People are like, why are you talking to me?
大家就会说，你为什么要跟我说话？
72:52
I'm like, what else should I see you'll be doing?
我就会说，那不然我该做什么？
72:53
I'm talking to customers, work on the prime.
我在和客户交流，做最重要的事。
72:55
Right.
对。
72:56
But they're very surprised.
但他们非常惊讶。
72:57
Anyway, I joined.
总之，我加入了。
72:58
I saw them come.
我看到他们过来了。
72:59
One of their engineers filled out a contact us form and I was like, that's a great company.
他们有个工程师填了一个 contact us 表单，我当时就想，这公司真不错。
73:02
I wanted to talk to them.
我想和他们聊聊。
73:03
I added myself to the meeting.
我把自己加进了会议里。
73:06
I added them to the meeting, the founder showed up and I said, wow, and this is a great
我把他们加进会议，创始人来了，我说，哇，这公司真棒。
73:10
company.
他们会付我们几百万美元。
73:11
They're going to pay us a few million dollars.
我说，你们怎么听说 plans 的？说说看。
73:13
I said, how did you hear of plans go?
我说，你是怎么听到计划进展的？
73:15
And he said, I love your Twitter, and I was like, that's great.
他说，“我喜欢你的Twitter”，我就说，“那太好了”。
73:19
And I said, you can't tell.
我说，“你根本看不出来”。
73:20
And I said, unfortunately, that's going to just make me worse.
我说，“不幸的是，那只会让我变本加厉”。
73:23
But it works.
但这招管用。
73:24
And I get told all the time that people just want, they don't hear genuine takes
我总听人说，人们只是想要——他们已经听不到真诚的观点了。
73:27
anymore.
科技圈变得太拐弯抹角了，而且到处都是骗子。
73:28
Tech has become so mealy mouthed and there's fakers there.
从来没有过这么多根本不懂电脑是怎么运作的骗子。
73:31
It's never been full of more full of fakers that don't know how computers work.
从来没有像现在这样满是那种根本不懂电脑的骗子。
73:35
I think people will feel refreshed.
我觉得大家会感到耳目一新。
73:37
Tech got a little bit too friendly to the non-technical ones that can yap all day long.
科技圈对那些整天能瞎叨叨的非技术人士有点太友好了。
73:41
We should drive them out and get back to talking about actual engineering.
我们应该把他们赶出去，回去谈真正的 engineering。
73:46
And I think there's a void of that and I think people miss it and I'm just saying my
而且我觉得现在缺少这种东西，大家也怀念它，我只是在说我的想法。
73:50
stuff.
这不是故意的。
73:51
It's not deliberate.
我可能明天就不干了。
73:52
I might quit tomorrow.
我不在乎。
73:53
I don't care.
我不在乎。
73:54
I don't think so much about it.
我不太想这个。
73:55
I'm not obsessed with my father account.
我并不执着于我的父亲账户。
73:56
I just say the things I care about and believe and maybe don't believe I want to test
我只是说出我在乎和相信的事情，也许有些我并不确信，我想检验
74:01
the belief in.
这些信念。
74:02
And I just use it.
而我就用它。
74:03
And it's that.
而且就是这样。
74:04
There's a lot of that.
这种事很多。
74:05
It's like, people mention my, I get people mention my to it to me constantly.
就像，人们提到我的，我老是听别人跟我提这个。
74:09
Maybe I should change.
也许我该改变一下。
74:10
I don't know.
我不知道。
74:11
Why follow you?
为什么关注你？
74:12
So I pay attention.
所以我注意。
74:13
I see.
我明白了。
74:14
I'm not involved in the conversations, but like anybody else, I lurk, and I'm here for
我没有参与那些对话，但和其他人一样，我潜水，我来这里就是为了
74:19
the stuff.
那些东西。
74:20
I'm here for the good stuff.
我来这里是为了好东西。
74:21
Sometimes it's drama.
有时候是drama。
74:22
Sometimes it's not.
有时候不是。
74:23
And I can't really recall the ones that the posts that I'm talking about, but I recall
我其实想不起我说的那些帖子具体是哪些，但我记得其他 database 公司吹嘘它们在速度和 migrations 上有多厉害。
74:31
other database companies touting their superiority in terms of their speed and their migrations.
比如，我连具体内容都不记得了，但大概就是那样。
74:38
Like, I don't even remember what, but it was something like that.
然后你就说，不对，不是那样。
74:41
And you were like, no, not true.
比如，这才是真正的 benchmarks。
74:43
Like, here's the real benchmarks.
这就是它们在细节上不真实的地方，因为你可以造假。
74:45
That's how they're being not true with their details because you can fudge that.
这就是他们在细节上不真实的地方，因为你可以糊弄过去。
74:50
Just like calling them out.
就像直接点名批评他们。
74:51
I'm like dang.
我当时就想，哎呀。
74:52
You know why not?
你知道为什么不行吗？
74:53
Why not?
为什么不行？
74:54
I love that.
我喜欢这样。
74:55
I mean, like, I guess because some people would be maybe too busy, not that you're not too
我的意思是，我觉得有些人可能太忙了，不是说你不忙，但就是有人会用太忙当借口不去怼他们。
75:01
busy, but like somebody might use the excuse of being too busy to come at them.
但我觉得最让我印象深刻的是，我喜欢你性格里的那一点就是……
75:06
But that's what I think stood out to me was like that I liked about your personality was
但我觉得最突出的是，就好像我喜欢你性格的地方是
75:10
that you're not, you're not unwilling to push back.
你不是，你不是不愿意反驳的人。
75:15
When you know, you know, maybe of an untruth or a, a fudging of a number, so to speak,
当你知道的时候，你心里清楚，也许是不实的地方，或者说，某个数字被做了手脚，可以说，
75:21
or some sort of benchmark detail that just didn't really fit, that you have a reputation
或者某个 benchmark 细节就是不太对劲，而且你有一个名声
75:25
to maintain with playing a skill, obviously, but I would say that you are definitely edgy
要维护，显然是在发挥技能的时候，但我要说你绝对很有锋芒，
75:33
and I like that.
我喜欢这一点。
75:34
Like, it's not a bad thing.
就是说，这不是坏事。
75:36
It's not how I would respond because I'm just not quite that way, but I respect it.
我不会那样回应，因为我确实不是那种人，但我尊重这一点。
75:40
I like that.
我喜欢这一点。
75:41
Yeah.
对。
75:42
Maybe it's just that I'm oddly broken and have brain damage.
也许只是我比较奇怪，脑子有点问题。
75:45
I didn't know.
我不知道。
75:46
I also don't, but I don't believe that people are too busy.
我也不这么认为，但我不相信人们会太忙。
75:49
If someone is comparing your product technically and lying about it, you should stamp on that
如果有人从技术上对比你的产品，并且撒谎，你就应该立刻狠狠地制止。
75:53
immediately.
对。
75:54
Yeah.
比如，如果你在乎你的手艺，而我非常在乎engineering这门手艺。
75:55
Like, if you care about your craft and I care massively about the craft of engineering,
就像，如果你在乎你的手艺，而我非常在乎工程这门手艺，
76:00
I love the database space because it is just engineering.
我喜欢数据库这个领域，因为它纯粹就是工程。
76:03
There are so many companies that have out-talked us, out-yapped us, and now that they're, now
有太多公司比我们会说、能吹，而如今他们很可能成为我们最大的客户，因为他们用 Postgres。
76:08
they're the largest customers that might go into us because they use Postgres.
我们正在赢，但根本上这是工程能力的较量，而我对此非常非常认真。
76:12
We are winning, but it's fundamentally a game of engineering and I take it very, very
我喜欢的是，我们内部并不是特别严肃的文化，从某种意义上说，我们很有幽默感。
76:15
seriously.
我喜欢开玩笑。
76:16
I love, like, we are not a very serious culture internally, in a sense, we have a sense
我喜欢的是，我们内部文化并不是很严肃，某种意义上，我们很有幽默感。
76:21
of humor.
的幽默。
76:22
I like to make jokes.
我喜欢开玩笑。
76:23
I like to be funny.
我喜欢搞笑。
76:24
I like to be silly, but when it comes to engineering and we take our job of engineering
我喜欢犯傻，但说到工程，我们对自己的工程工作
76:30
very seriously, everyone accidentally lets down customers and things are now and then
非常认真。每个人都会偶尔不小心让客户失望，这种事
76:34
that happens with us and we take that very seriously as well.
也会发生在我们身上，我们也非常认真对待。
76:38
We are here to be a database product and I feel like one of the best things I can do
我们是来做 database 产品的，我觉得我能做的最好的事情之一
76:42
to attack my competitors is out-engineer them and then have a lot of fun doing it publicly
就是在工程上超越竞争对手，然后公开地享受这个过程
76:47
and build an narrative around doing that and they can come for it.
并围绕这件事构建一个叙事，他们可以随时来挑战。
76:50
When we announced our Postgres product, it was a very simple blog post, it just benchmarked
当我们发布 Postgres 产品时，那只是一篇非常简单的博客，只是跑了个 benchmark。
76:55
us against all the ones who are on the market.
我们对抗市场上所有的那些家伙。
76:57
We smashed some of them into the floor, we had to give two of them, we had to give them
我们把其中一些打得满地找牙，我们不得不给其中两个，我们不得不给他们
77:02
double the resources just so they could show up on the graph next to us.
双倍的资源，就只是为了让他们能出现在图表上，就在我们旁边。
77:06
That was the way.
当时就是这样。
77:07
We showed up, humble because we said we weren't going to go into Postgres because we didn't
我们出现了，表现得很谦虚，因为我们说过我们不会搞 Postgres，因为我们确实没
77:11
go into Postgres.
去搞 Postgres。
77:12
I didn't know if we could do anything better, I didn't want to be arrogant enough and believe
我不知道我们能不能做得更好，我不想自大到去相信
77:17
that we can just grandiocially show up and do it at a different database better.
我们可以大张旗鼓地出现在一个不同的 database 上，然后做得更好。
77:21
Once we'd shown that we could, though, we made that the point of what we're doing.
不过，一旦我们证明了我们能做到，我们就把这当成了我们正在做的事的核心。
77:25
All of the companies that we benchmarked, they got very triggered.
所有我们 benchmark 过的公司，全都反应特别激烈。
77:29
One, actually, the CTO came up to our people, we released it while we had a booth at a conference
有一家，实际上，他们的 CTO 直接走到我们的人面前——我们那天是在一个会议上发布的，当时我们有个展位——他们走过来，气得唾沫星子直飞。
77:35
that day, they walked up to them and got so angry that spit was flat.
另外一家公司的人——是我朋友，他公司的展位就在我们旁边——说：“你得来看看你的团队，他们刚被你们 benchmark 的其中一家公司的 CTO 吼了一顿。”
77:39
Someone else from another company, a friend of mine who's company's booth is next to us,
我会说出他们的名字，能在播客里被提到，对他们来说可是天大的好事。
77:42
you need to check on your team, they've just been screamed at by the CTO of one of the
你得去看看你的团队，他们刚被你 benchmark 过的某个人的 CTO 吼了一顿。
77:45
people you benchmarked.
我就说出他们的名字吧，能在播客节目里被提到，对他们来说可是大事。
77:46
I'll say their name, it would be so big for them to get mentioned on a podcast night
你不会冒犯到我的，我是在 Google Doc 里做的，所以成本很低。
77:51
for this, so I won't, I won't, but they, yeah, it really made people extremely mad.
關於這個，所以我不會，我不會，但他們，是啊，這真的讓大家氣炸了。
77:56
They all promised follow-ups.
他們全都承諾會跟進。
77:57
We watched them all benchmarks, it's been a year, not a single follow-up because they couldn't
我們看了他們所有的 benchmarks，已經一年了，一個後續都沒有，因為他們根本無法
78:00
do it, so they can sit back down and try and write code and do all their yapping, they
做到，所以他們可以坐回去，試著寫程式，繼續他們所有的嘴砲，他們
78:05
can hire their big brand designers to spend 90 grand on a website and we all just do
可以雇用他們的大品牌設計師花九萬美金做一個網站，而我們全都只做
78:09
engineering and we'll take that bit very seriously and then we'll have fun messing with them
engineering，而且我們會非常認真地對待這部分，然後我們會很開心地跟他們糾纏
78:13
publicly.
公開地。
78:14
Yeah, that's wild.
是啊，真是瘋了。
78:18
Let's go to, to maybe an angle, I wasn't really going to mention, but you mentioned $90,000
让我们转到另一个角度，我本来不太想提，但你提到了那些九万美元的网站。我不是说你的网站不值九万美元，我也没打算用猜测你花了多少钱来做这个网站来冒犯你，但这非常简单。
78:24
websites and I'm not saying that your website is not $90,000 and I'm not going to try and
你不会冒犯我的，我是在 Google Doc 里做的，所以非常便宜。
78:29
offend you by telling you how much you did it and spend on your website, but it's very
你说你做的，是什么意思？
78:34
simple.
那个网站的第一个版本，还有我们出色的 CMO Holly，她有一条关于这个的 Twitter thread。
78:35
You're not going to offend me, I did it in a Google Doc, so it was very cheap.
你说你做了，是什么意思？
78:40
How do you mean you did it?
那个网站的第一版，还有我们超棒的 CMO Holly，她有一条关于这个的 Twitter thread。
78:42
The first version of that website and Holly our incredible CMO, she has a Twitter thread
那个网站的第一个版本，还有我们超棒的 CMO Holly，她有一条 Twitter thread
78:46
about this.
讲的就是这个。
78:47
I was on a trip to see the Redwoods and I was up late at night and I was like, we need
我当时在去看 Redwoods 的旅途中，深夜没睡，心里想，我们需要更简单的方式，我只是想说，我们有很棒的技术，就让我们这样沟通，我的意思是，然后我打开了一个 Google Doc，在里面画出草图、写字，然后 Jason，我们那个特别棒的设计师，接手之后让它看起来不像精神病人的乱涂乱画，又把它放到了网上，结果立刻走红。我也因此被邀请上了那些有几十万听众的 podcast，比如 marketing podcast。
78:53
to go way simpler, I just want to, we've got great tech, let's just communicate this like
说得更简单点，我只是想说，我们有很棒的 tech，就让我们这么沟通，比如
78:57
I read me and so I opened a Google Doc and just started drawing asky out into it and
我的意思是，于是我打开了一个 Google Doc，然后就开始在里面画 ASCII，
79:04
writing words and then Jason, our phenomenal designer, took it, made it look like less
写字，然后 Jason，我们超棒的设计师，接手后让它看起来没那么像
79:09
like the let, like sort of scribblings of a mental patient and put it up on the internet
那种，就像精神病患者的涂鸦，然后把它放到了网上
79:14
and it went viral immediately.
然后它马上就火了。
79:15
I got invited onto podcasts that have like hundreds of thousands of listeners like marketing
我被邀请上了一些有几十万听众的 podcast，像 marketing
79:19
podcasts.
类的 podcast。
79:20
I didn't go on those, but people were fascinated by it and these things that get spoken about
我没去参与那些，但人们对它和这些经常被谈论的事情很着迷
79:23
all the time, it will change, we are going to work on a new version of the website, it's
一直如此，它会改变，我们将要做一个新版的website，它
79:27
too simple, now it's that format and that medium is too simple to communicate the complexities
太简单了，现在的这种format和medium太简单了，无法传达复杂性
79:34
of three separate products and all this sort of stuff, so it will evolve, but it was very
三个独立产品以及诸如此类的东西的复杂性，所以它会进化，但那非常
79:38
fun to run in the other direction.
有趣，反其道而行之。
79:41
Now by the way, there is some beautiful $90,000 websites out there, some companies produce
不过顺便说一下，外面有一些漂亮的$90,000的website，有些公司做出来
79:46
having same taste and produce absolutely incredible websites.
有着相同的品味，做出绝对令人难以置信的website。
79:52
We just went very simple for now and we're going to, you know, never going to go to these
我们现在只是做得很简单，而且我们，你知道的，永远不会去做那些
79:56
like, I also just don't like it, it's bad engineering to me when you open a website and
比如说，我就是不喜欢那样。对我来说，打开一个网站，花里胡哨的，然后你能听到你的 MacBook 风扇狂转，Chrome 卡顿，那就是 bad engineering，并不能传达更多东西，而我是想要传达信息的。但那就是个有点傻的小事，你懂的。再说，我就是喜欢尝试一些蠢东西，有效就行，不行你可以撤销，管他呢，你知道吗，效果还挺好的。
80:00
it's all flashy and like you can hear your MacBook spin up and Chrome is stuttering, like
我觉得当年你第一次来播客的时候，以及那之后的几年里，你的网站更接近那种看起来好像值九万美元的网站，你懂的。然后我觉得大概一两年前，你就转向了你现在的方向，所以你没有...
80:05
that's bad engineering, that does not communicate more, I want to communicate, but it was just
那是糟糕的 engineering，它并不能传达更多，我想要沟通，但它只是
80:10
a little silly thing, you know, this, again, I just like to try dumb stuff, it works
有点傻的小事，你懂的，这个，又来了，我就是喜欢尝试些蠢东西，结果还真的有用。
80:15
so it doesn't, you can undo it, whatever, it, you know, it worked pretty well.
所以没关系，你可以撤销它，随便啦，反正效果还挺好的。
80:20
I think back in the day when we had your first in the pod and for several years after
我想当年你第一次上播客的时候，以及之后好几年里。
80:25
that you had closer to what would seemingly look like a $90,000 website, you know, and
你的网站更接近那种看起来像九万美元做的样子，你懂的。
80:30
I think it's about a year or two ago, you've gone the direction you've gone, so you haven't
我想大概是一两年前，你转向了你现在的方向，所以你已经——
80:34
always had this, you know, you've had this minimalistic opinion, but it didn't seem like
一直有这个，你知道，你一直有这种minimalistic的观点，但感觉你并不是这种plain——大概是这么说的——你知道，这非常接近plain text，这不是贬义。
80:38
you were this plane, there's probably the way it's said, you know, this is very close
我们之前的网站也极其出色，是一群极有才华的人做的，然后被抄袭了。真的有人在open source库里模仿我们的gradient hero。最近有人跟我说，他们还记得我们有那个网站的那一年，到处都能看到它的山寨版。确实如此，大家就是直接剽窃了那个旧网站。
80:44
to plain text, it's not, it's not a pejorative.
变成 plain text 了，这不是贬义。
80:47
Our previous website was also extremely good, done by an extremely talented set of people,
我们之前的网站也非常好，是由一群非常有才华的人做的。
80:52
and it got copied, there was actually people at open source libraries to copy our like
而且它被抄袭了，确实有 open source libraries 的人来模仿我们的东西。
80:56
gradient hero and someone said to me recently, like they remember that year when we had
gradient hero，最近有人跟我说，他们还记得那年我们有那个网站，到处都能看到它的仿冒版，确实如此，人们把那个旧网站抄了个遍。
81:02
that website, they saw copies of it everywhere, and we did, like people ripped that old
汤底。
81:08
website off.
嗯，那是个很好的给你钱的机会，所以我没能成为你的天使投资人。
81:09
So now I'm saying, what we're doing right now is figuring out the new one, and I'm like,
所以现在我要说的是，我们现在在做的事情是在摸索新的方向，然后我就觉得，
81:15
when we tried really hard, people loved it, and like it was good.
当我们特别努力去做的时候，大家很喜欢，效果也很好。
81:20
When we didn't try at all, people also loved it, and copied both versions.
当我们完全没怎么费劲的时候，大家也喜欢，而且两个版本都被拿去抄了。
81:25
Now I kind of feel like the pressure is on to do something else that's a bit crazy.
现在我会觉得有点压力，得再做点有点疯狂的事情。
81:29
I'm not trying to pretend I'm some giant design genius, and we're going to come up with
我不是想装作自己是什么设计天才，然后我们就能搞出
81:32
anything amazing, but you do feel that kind of like, huh, like we also didn't try to do
什么惊艳的东西，但你确实会有那种感觉，嗯，就像我们当初也没刻意去设计那些，
81:39
those things, we just did things that made sense to us, but it gives you that pressure
只是做了些对我们来说合理的事，但这会让你有压力，
81:42
to like think, what can we do this time that's special, maybe we've found it, I don't
去想这次我们能做出什么特别的东西，也许我们已经找到了，我也不确定。
81:47
know, we'll see, we'll see what people think.
嗯，我们等着瞧，看看大家怎么想。
81:49
I'm sure they won't be afraid to tell me on one.
我肯定他们不会害怕在一件事上告诉我。
81:53
Sinceress, spicy nature to fellow, that's probably the sweetest negative price to say.
真诚，对同伴的火辣本性，那大概是最好听的负面评价了。
82:00
That's cool.
那很酷。
82:01
I like that.
我喜欢这样。
82:02
I'm enjoying my own words.
我在享受自己说的话。
82:03
I'm sorry to say him.
我很抱歉对他说。
82:04
Say something meaner.
说点更狠的话。
82:05
I'd like to do it.
我想这么做。
82:06
Keep this in as well.
这个也保留吧。
82:07
Don't cut it.
别剪掉。
82:08
Say something mean.
说点难听的。
82:09
What do you think the worst means?
你觉得最糟糕的是什么意思？
82:10
No, no, no.
不不不。
82:11
I really meant that in that way.
我真的就是那个意思。
82:12
I liked how it sounded.
我喜欢它听起来的样子。
82:13
I had to laugh.
我忍不住笑了。
82:14
It's like a comedian on the stage laughing their own joke.
就像喜剧演员在台上讲自己的笑话自己笑一样。
82:16
That's like you shouldn't do that.
那感觉就像你不该那么做。
82:17
So I did that.
所以我就那么做了。
82:18
No, it's good.
不，挺好的。
82:19
I like that.
我喜欢这样。
82:20
You should enjoy it.
你应该享受它。
82:21
I may be I'm spicy.
我可能有点冲。
82:22
I don't know.
我不知道。
82:23
Like you are spicy.
比如你很辣。
82:24
Maybe everyone's bland.
也许其他人都是平平淡淡的。
82:25
Another role of that, man.
这是那个角色的另一面，兄弟。
82:26
Yeah.
没错。
82:27
Listen, I live in Texas, and it's just fine to be spicy, okay?
听着，我住在Texas，辣一点完全没问题，好吗？
82:30
Just fine.
完全没问题。
82:31
That's true.
确实如此。
82:32
I want to go to this level because I this is something that I've had we've had podcasts
我想聊到这个层面，因为这是我——我们曾经在播客里聊过的。
82:38
on this.
我们聊过multigress。
82:39
We talked about multigress.
我们采访过Suga。
82:40
We've talked to Suga.
我们聊过……我甚至是soup base的投资人，还是soup base的小天使投资人。
82:42
We've talked to I'm even an investor in soup base and the small angel investor in
嗯，那是个给你投钱的好机会，所以我没法做你的天使投资人。
82:46
soup base.
你能否想象这样一个世界：我把一个巨大的 sharded post-grows 客户置于我们的对立面？
82:47
Well, that was a very good chance to give you money so I couldn't be an angel investor
嗯，那是一个非常给你钱的好机会，所以我没法成为你的天使投资人。
82:50
for you.
你能否想象这样一个世界：我坐在一个巨大的 sharded post-grows 客户对面，与我们对抗？
82:51
Well, that was an exception investment because they've built a very valuable business.
嗯，那是一次例外投资，因为他们建立了一个非常有价值的企业。
82:54
Yeah.
嗯。
82:55
Yeah.
嗯。
82:56
It's my I've only had two angel investments.
这是……我只有两笔angel investments。
82:59
That's my best one.
那是我最好的一笔。
83:00
Okay.
好的。
83:01
Good.
很好。
83:02
So, not very many angel investments on my side, but that one I think is paying off very
所以，我这边没太多angel investments，但我觉得那一个的回报非常。
83:07
well.
好吧。
83:08
And I'm happy about that.
而且我对此很高兴。
83:09
I've got friends who use the base and I'm like, why do you use it?
我有朋友用 base，我就说，你为什么要用它？
83:12
And these are obviously, well, I shouldn't say that.
而这些人显然是，嗯，我不该这么说。
83:16
They're newer to software development, a vibe coding, which is just fine.
他们比较新入软件开发，是 vibe coding，这完全没问题。
83:21
You know, everybody's got to start somewhere in the online.
你知道，每个人在网上都得从某个地方开始。
83:22
That is their audience.
那就是他们的受众。
83:23
And they've done a phenomenal job of building that audience.
而且他们在建立受众方面做得非常出色。
83:25
They have millions of people that have I could do with that platform.
他们有数百万人，我可以利用那个平台。
83:28
I'm like, why are you using soup base?
我就说，你为什么要用 Supabase？
83:30
That's where my database is.
我的数据库就在那里。
83:31
I'm like, is that it?
我就说，就这？
83:32
Is that all you're using them for is just a postgres database?
你拿它就只当一个 Postgres 数据库用？
83:35
Okay.
好吧。
83:36
Fine.
行吧。
83:37
But whatever from this.
不过也无所谓了。
83:38
They've written this wave and they had great experience.
他们书写了这一波浪潮，而且体验非常棒。
83:42
I think they should feel very proud of themselves and they've created an amazing business.
我觉得他们应该为自己感到非常自豪，他们创造了一个了不起的事业。
83:46
Most people have not done that.
大多数人没有做到这点。
83:47
It's amazing.
真了不起。
83:50
What are your thoughts on, you know, what you're doing with Niki and multi-gress?
你对，呃，你和 Niki 以及 multi-gress 在做的事情有什么看法？
83:55
How do those two rub, I suppose, in the marketplace?
我想，这两者在市场上是怎么相互作用的？
84:00
It's their attempt.
这是他们的尝试。
84:01
There is like you've just referred to.
就像你刚才提到的那个。
84:03
There is one of the people that created for tests at YouTube
有一位在 YouTube 做测试开发的人，
84:07
and was a co-founder of PlantScale that is a soup base doing this
同时也是 PlantScale 的联合创始人，PlantScale 是一家 database 公司，也在做这件事。
84:11
and a couple of ex-employees, you know, they are.
还有几位前员工，你懂的，他们也在做。
84:16
The competition is good.
竞争是好事。
84:17
They're welcome to go and do this.
欢迎他们去做这件事。
84:19
It's great.
这很好。
84:20
It's different.
但这不一样。
84:21
I'll tell you how it's different.
我来告诉你有什么不同。
84:22
Well, first of all, I'll address that.
嗯，首先，我先回应一下这个问题。
84:24
Sue left the company before we had a single-sharded customer in production.
Sue 在我们还没有任何 single-sharded 客户进入 production 之前就离开了公司。
84:29
And so the thing that I don't like about this situation
所以，关于这件事，我不喜欢的一点是
84:33
is the perception that someone who made PlantScale what it is today
有人觉得，那个让 PlantScale 成为今天这个样子的人
84:37
has gone to soup base.
已经变成了汤底。
84:38
That is completely untrue.
这完全不是事实。
84:40
If you look at the version of the company pre-2020 and 2021,
如果你看看 2020 年和 2021 年之前那个版本的公司，
84:46
that is the counterfactual to what PlantScale could have been
那就是 PlantScale 本可能成为的样子的反事实。
84:49
and it was not and now PlantScale is what it is now.
当时不是，现在PlantScale才是现在这个样子。
84:52
He played no role in what it is now.
PlantScale现在这样跟他一点关系都没有。
84:54
And so when people online make takes about,
所以当网上的人发表观点时，
84:57
oh, you know, they're going to soup bases
哦，你知道，他们会去soup bases，
84:59
as a shot of becoming anything like PlantScale.
当作是成为任何像PlantScale一样公司的机会。
85:01
That is just not true.
这完全不是真的。
85:01
None of the PlantScale DNA has gone to soup base.
PlantScale的DNA一点都没流入soup base。
85:04
So I can correct that, like leaving that early in a company
所以我可以纠正这一点，因为我很早就在一家公司发展初期离开了。
85:08
that has achieved so much since that period
自从那个时期以来取得了如此多的成就。
85:10
of that person leaving is hardly the boon people think it is.
那个人的离开很难说是人们以为的那种好事。
85:14
But that's fine.
但那也没关系。
85:15
And I wish them well.
我祝他们一切顺利。
85:17
I don't like talking about individuals in their histories,
我不喜欢谈论个人的过往，
85:20
but they have had no problem trying to claim the protest name
但他们却毫无顾忌地试图宣称那个抗议名称，
85:22
and claim an association with PlantScale.
并声称与PlantScale有关联。
85:26
I don't like that.
我不喜欢那样。
85:28
So what they're building, I have no doubt,
所以对于他们正在构建的东西，我毫不怀疑，
85:32
will be of quality and the people that are building it
质量会很高，而且构建它的人
85:35
are not unserious and they are not bad engineers.
不是不严肃，也不是糟糕的工程师。
85:40
However, they're in a business that does different things.
然而，他们做的业务是另一回事。
85:44
It is not the union of the 15 years experience
这并不是我们15年经验、
85:46
of attest the five years of Building PlantScale
至少五年打造PlantScale的经验、
85:48
and all the other things that we do.
以及我们所做其他所有事情的简单相加。
85:51
And so it will be interesting to see how it goes.
所以看看会怎么发展，会很有意思。
85:54
I wish everyone well.
祝大家一切都好。
85:55
I think it's great building software.
我觉得做软件是很棒的事。
85:58
However, when nearly, like there's customers
然而，当差不多，比如说有些客户
86:03
that are still on their homepage
还在他们首页上
86:05
that have migrated to us already when
就已经迁移到我们这儿了，当
86:07
now we do Postgres, the logos that are on our page
现在我们做 Postgres 的时候，我们页面上的那些 logo
86:11
and you consider the long-term sustainability
而且你要考虑长期可持续性
86:13
of what it takes to build a sharded product
构建一个 sharded 产品需要什么。
86:16
and the companies that run on sharding.
还有那些依赖 sharding 的公司。
86:20
A company that's really only service-fied coders
一家真的只是为程序员提供服务的公司
86:23
is not going to compete with a company
是没法跟一个公司竞争的，
86:25
that has done the opposite of that
这个公司做了相反的事，
86:26
and is running some of the world's largest databases.
并且运行着世界上最大的几个 databases。
86:29
And that will prove out.
这一点最终会被证明。
86:30
But at the end of the day, I would not
但说到底，我不会
86:33
try and go after their audience
试图去抢他们的用户。
86:35
because I don't believe I could, right?
因为我不相信我能做到，对吧？
86:37
I do not believe.
我不相信。
86:38
And until I maybe will believe that one day,
而且也许有一天我会相信，
86:40
but I try and stay humble and say I would not
但我试着保持谦逊，说我现在不会
86:43
try and compete with super base right now
尝试去和 super base 竞争
86:46
on five coding, on back ends of service.
在 five coding、在服务的 back end 上。
86:49
They've done a really good job at this.
他们在这方面做得非常好。
86:51
They are welcome to try and compete on what we do,
欢迎他们来尝试和我们竞争我们所做的。
86:57
which is what they think they're doing with multi-grows.
这就是他们以为自己在用 multi-grows 做的事情。
86:59
But as someone has been through this,
但作为一个经历过这些的人，
87:02
there is many more things that are not just,
还有很多东西不仅仅是，
87:04
you just don't just have an open source
你并不是仅仅有一个 open source，
87:06
like sharding tool is not, it is, their record of security
比如 sharding tool 并不是，它是，他们的安全记录
87:12
is very publicly quite bad.
在公开场合是非常糟糕的。
87:14
We do not have that, right?
我们没有这些，对吧？
87:15
These are all things that everyone is care about.
这些都是大家关心的事情。
87:19
At the end of the day, when I've seen the largest database
归根结底，当我看到他们曾拥有过的最大 database 迁移到我们这里时，
87:23
that they ever had that moved to us,
它就像是，我的意思是，相比之下简直微不足道，远远低于我们的平均客户规模，
87:25
and it would like, I mean, minuscule in comparison
那将是一场持久战，当他们不得不——
87:29
to it was far below our average customer size,
你是否设想这样一个世界：我面对着一个庞大的 sharded Postgres 客户，正与我们竞争？
87:33
it's going to be a long battle when they have to,
那你们最大的客户是谁？
87:35
do you envision a world where I've set opposite a giant
去打造 Niki 吗？
87:37
sharded post-grows customer versus us?
它会及时做到吗？
87:41
And what's your largest customer?
那你们最大的客户是谁？
87:45
And ours is four orders of magnitude larger, I mean.
而且我们的大了四个 orders of magnitude，我是说。
87:49
I don't know, we'll see.
我不知道，再说吧。
87:50
I like the conversation, but I don't really think about,
我喜欢这种对话，但我不太去想，
87:54
like there's other ones that are trying other sharding ones.
比如有其他人在尝试其他 sharding 的方案。
87:56
It's like, what if I do it?
就像，如果我来做呢？
87:57
I just don't think about it.
我就不去想它。
87:58
We are on schedule, we are on our schedule,
我们在按计划进行，按我们自己的计划，
88:01
when it comes to database scaling,
在 database scaling 方面。
88:02
we set that standard industry wide.
我们把这个标准定成了全行业的标准。
88:08
You have to focus in other places
你得把注意力放在其他地方。
88:10
if you want to win that way.
如果你想那样赢的话。
88:12
But I do wish them well,
但我真心希望他们一切都好。
88:13
and I wouldn't hate to communicate any disrespect
我也不想表达任何不敬。
88:16
for the incredible business that they have built.
因为他们打造的了不起的企业。
88:19
Yeah.
嗯。
88:21
So you mentioned having extraordinary engineers,
所以你提到你拥有非常出色的工程师，
88:23
and you mentioned having the ability
而且你提到说你有能力
88:26
to give those extraordinary engineers artificial intelligence,
给那些杰出的工程师人工智能，
88:29
AI, right?
AI，对吧？
88:30
Models, LLMs, Anthropic, Cloud, Codex,
Models、LLMs、Anthropic、Cloud、Codex，
88:34
pick your flavors.
挑你喜欢的口味。
88:35
I know what my flavors are.
我知道我的口味。
88:37
I say use them all, of course, right?
我说当然全都用，对吧？
88:38
Use the ones that work for you.
用那些对你管用的。
88:41
You've got Niki out there,
你现在有了 Niki，你不确定要不要 open sourcing 它，原因我们都聊过了。但把那个出色的 engineering team——50人，不管实际需要多少人，或者你从那一大群人里切了多少人出来——用来 build out Niki，那是什么感觉？
88:42
you're not sure about open sourcing it
你不确定要不要 open sourcing 它，
88:44
for all the reasons we've already talked about,
就因为我们之前讨论过的那些原因，
88:45
but what was it like to take
但带着那个卓越的工程团队，50 人，
88:48
that exceptional engineering team, 50,
不管实际需要多少，
88:52
how many ever was necessary,
或者你从那一大群人里分出多少，
88:54
or how many ever you sliced off from that larger swath of folks
来打造 Niki？
88:57
to build out Niki?
是的。
88:59
What was it like to kind of have a redo?
重新来一次是什么感觉？
89:02
I would call it like I've been enjoying the redoes
我想说我一直很享受这种重来，
89:05
in this AI world,
在这个 AI 世界里，
89:06
because there's so much more possible.
因为有太多可能了。
89:09
I'm almost happy to fail at trying to implement something
我甚至有点乐于在尝试实现某个东西时失败，
89:13
because I learned,
因为我学到了东西，
89:14
and that my next version would just be that much better
这样我的下一个版本就会更好，
89:16
and more closer to what I actually want to deliver.
更接近我真正想要交付的东西。
89:19
So what was it like having had that many years
那么，拥有那么多年的
89:21
of experience with the test,
与这个测试相关的经验，
89:22
this brand new superpower,
这个全新的超能力，
89:24
some of the most amazing engineers out there anyways,
总之，就是一些最了不起的工程师们，
89:27
and then here you go and you're building Niki.
然后你就去打造 Niki 了。
89:28
What was I like?
我是什么感觉？
89:29
For normal, I like it's just been amazing.
对我来说，我觉得真的就是太棒了。
89:32
Like to watch people that are so good
就像看着那些那么厉害的人
89:34
at what they do and they're craft,
看他们做的事情和他们的手艺，
89:36
to have the superpowers to build.
拥有那种去构建（build）的超能力。
89:39
Now, we are not just five coding this thing, right?
不过，我们也不只是 vibe coding 这个东西，对吧？
89:42
I sure am.
我当然是啊。
89:44
To be generous to the multi-cress team as well,
公平地说，multi-cress 团队也一样，
89:46
they did a really good post about how they used AI to build
他们发了一篇很好的帖子，讲他们怎么用 AI 来 build。
89:48
and people also interpreted that as five coding,
大家也把那解读成了 vibe coding，
89:51
which I don't think that's what they were doing either, right?
但我觉得他们也不是在 vibe coding，对吧？
89:53
Like some exceptional engineers are building really,
就像一些出色的工程师在构建，真的
89:55
really quickly,
真的很快，
89:57
but I'm watching phenomenal people build
但我看到的是非凡的人在构建，
89:59
in phenomenal ways with AI.
以非凡的方式使用AI。
90:01
And it is like,
然后就像是，
90:03
I thought Niki would take two to three years, right?
我原以为Niki需要两到三年，对吧？
90:07
We're a year in and we're already moving customers now.
我们才一年，就已经在迁移客户了。
90:12
That is like,
这就像是，
90:13
and I thought three years with like earlier AI, right?
我当时想的是，三年时间，跟早期的AI比，对吧？
90:17
Like how it was a year ago when we started this.
就像一年前我们刚开始的时候那样。
90:19
And then I also thought,
然后我还在想，
90:23
you know, just doing 15 years of work that quickly, right?
你知道，就是这么快就把15年的工作做完了，对吧？
90:26
Like now, is it got feature parity with the test?
那现在，它跟测试集实现功能对齐了吗？还没有。
90:29
No.
它会很快达到吗？会的。
90:30
Will it get there soon enough?
酷吗？
90:32
Yes.
对。
90:33
It is amazing.
这太神奇了。
90:35
It's actually a great kind of meditation
这其实是一种很好的冥想，
90:37
on the fact that code was just a thing you had to get through
关于这样一个事实：code 曾经只是你不得不搞定的事情，
90:42
and that taste and now that's a terribly overused word,
而品味——这个词现在已经被用烂了，
90:45
so whatever, but like,
所以随便吧，但就像，
90:46
but your taste, your biases, your experience,
但你的品味、你的偏好、你的经验，
90:49
fed into things that will write the code
被注入到那些将要写 code 的东西里，
90:52
is truly where it was.
那才是真正的核心所在。
90:53
Like and it gives me like,
而且这让我觉得，
90:55
there's a lot of people feeling despair about AI.
有很多人对AI感到很绝望。
90:58
We are ingenious creatures.
我们是很聪明的生物。
91:00
We have engineered our way off of this planet
我们用工程能力飞离了地球，
91:02
to the depths of the ocean.
也潜入了海洋深处。
91:04
We've engineered our environment around us.
我们已经改造了我们周围的环境。
91:06
AI will be a tool to do that.
AI将是实现这一点的工具。
91:07
And when AI takes over more
而当AI接管更多。
91:10
and our ingenuity,
而我们的聪明才智，
91:11
it's flowed into that machine.
已经注入了那台机器。
91:13
We will gain more as a species.
作为一个物种，我们将获得更多。
91:15
So it's very exciting to me to see this
所以看到这一切，我特别兴奋，
91:17
and see how we engineer using these tools.
也看到我们如何用这些工具进行工程实践。
91:20
And the productivity is there
而生产力确实在那里，
91:21
and it does truly gel with the belief
它与一个信念真正契合，
91:24
that I have and other people have
这个信念我和其他人都持有。
91:25
that this stuff will scale the expertise that companies have.
这些东西会 scale 公司所拥有的专业知识。
91:29
We are proving that right now.
我们现在就在证明这一点。
91:31
Like the knowledge that things we know,
就像我们所知道的知识，
91:33
the proprietary knowledge we know
我们掌握的专有知识，
91:35
about running databases of this scale
关于运行这种 scale 的 databases 的知识，
91:38
can influence how we build
可以影响我们如何构建，
91:40
and produce something very quickly.
并非常快速地生产出东西。
91:41
And that shows that there is an incredible
而这表明有一种难以置信的
91:44
mode to experience and intelligence
体验和智能的模式
91:46
and your company's culture
以及你们公司的文化
91:48
and the way you build
以及你构建的方式
91:49
and the targets you set for yourself.
以及你为自己设定的目标。
91:51
And code was always just syntax that you have to build
而 code 始终只是你必须构建的 syntax
91:55
and you had to be smart of reading it
而且你必须很聪明地阅读它
91:56
and you had to be really good at coding
而且你必须非常擅长 coding
91:59
and there's a massive power law
而且存在一个巨大的 power law
92:01
to the distributed like ability to code
对于像编码这样的分布式能力
92:03
has been unfairly distributed across the industry.
在整个行业中分布得不公平。
92:08
But if you're listening to this
但如果你正在听这个
92:10
and you'll skip tool to AI,
而且你会跳过工具，直接使用 AI
92:12
I hate to tell you that it's here
我很遗憾地告诉你，它已经来了
92:16
and that it's happening
而且它正在发生
92:17
and the best engineers are engineering in phenomenal ways.
最优秀的工程师们正在以惊人的方式进行工程创新。
92:20
I'm also happy to tell you
我也很高兴地告诉你
92:20
it's nowhere near too late to adopt these tools.
现在用这些工具一点也不晚。
92:24
The hype slopplords on Twitter
Twitter 上那些炒作垃圾内容的大佬们，
92:25
that are just saying random stupid things
就只会随口说些没头没脑的蠢话，
92:28
and making everyone get triggered and upset are wrong.
把大家都搞得火大又心烦——这些人错了。
92:30
They've always been here.
他们一直都在。
92:31
They've existed throughout the industry.
整个行业里一直都有这样的人。
92:33
Done people have done takes on tech.
蠢人一直在对科技发表看法。
92:36
The good people that are good at engineering
而那些擅长工程的能人。
92:37
look at this is the thing to engineer
你看，这就是需要工程化的东西。
92:38
and they just engineer it.
他们就直接把它工程化了。
92:40
And so it has been amazing to have that do over
所以能有这么一次重来的机会，真的很棒。
92:42
and it taught me so many lessons
它让我学到了很多。
92:44
that I knew in an abstract way
那些我以前只是抽象地知道的。
92:46
after building software for 20 years
在写了20年软件之后。
92:47
but now I really know it.
但现在我是真的懂了。
92:50
Yeah.
嗯。
92:52
Do you think there's one part of my question
你觉得我问题的某一部分
92:55
that you can answer very well
你能回答得很好吗？
92:56
and I'm just curious about this and that's fine
我只是好奇这个，那也没关系
92:58
is the redo part of it?
就是那个 redo 的部分？
93:00
Do you feel like,
你觉得，
93:02
I don't want to say that necessarily
我不一定非得这么说
93:04
because I'm not sure of the actual code I haven't read it
因为我不确定实际的代码，我没读过
93:07
because it's not open source and that's cool.
因为它不是开源的，那也没关系。
93:11
But when we look at Niki and the opportunity to redo
但当我们审视Niki和重来的机会
93:15
you mentioned it's not quite where Vitas is at.
你提到它还没到Vitas现在的程度。
93:17
Will it be eventually of course?
最终当然会吗？
93:19
Yeah.
嗯。
93:20
But was it a joy for you and for others
但对你和其他人来说，这是否是一种快乐
93:22
that were involved in writing this code
那些参与编写这段code的人
93:24
or operating the AI that wrote the code
或操作编写这段code的AI
93:28
and all the wisdom that came from that?
以及从中获得的所有智慧？
93:31
Was it cool?
是不是很酷？
93:32
Was it fun to do a redo?
重做一次好玩吗？
93:34
Oh yeah.
哦，当然。
93:35
Take two on what Vitas was for my sequel
再来一遍，看看 Vitas 对我的续集来说是什么
93:38
to do at Niki's for Postgres.
就是要在 Niki's 为 Postgres 做的那个续集。
93:40
Members of that team have told me
那个团队的成员告诉过我
93:43
that they're having the most fun of that career.
那是他们职业生涯里最开心的一段。
93:46
Like just start with an empty repo.
比如就从一个空的 repo 开始。
93:49
Heck yeah.
那必须的！
93:50
Like we had this meeting, we all got together
就像我们开了个会，所有人都聚到一起
93:51
and we just talked about how we're going to build this.
然后就开始聊我们打算怎么把这事儿做出来。
93:53
Like what are the main lessons we learned
比如说，我们学到的主要教训是什么？
93:55
from taking Vitas?
从搞定Vitas这件事里？
93:57
And like Vitas was built by very smart people at Google, right?
而且Vitas是谷歌一帮特别聪明的人做的，对吧？
94:00
And they did an incredible thing breaking out of Google
他们从谷歌独立出来这件事本身就超厉害
94:03
and did all these things.
还做了那么多事情。
94:04
And like I said, Vitas has grown
就像我说的，Vitas 已经成长了。
94:07
and become a real mature product in the last five years.
并且在过去五年里，它变成了一个真正成熟的产品。
94:09
In the last five years since 2021,
从2021年起的这五年里，
94:12
its usage has gone up 65,000%.
它的使用量增长了65,000%。
94:15
It's pretty good.
这相当不错。
94:16
It's used by more people now than it ever has been.
现在使用它的人比以往任何时候都多。
94:20
But like there's nothing like that get in it
但就像，没有什么比得上那种全身心投入的感觉。
94:23
on all of your like washing all your sins away
就像把你所有的罪过都洗掉一样。
94:26
and trying again.
然后再试一次。
94:28
And Vitas is an amazing piece of software.
而 Vitas 是一个非常棒的 software。
94:30
But of course, if you can just start from like,
但当然，如果你可以直接从比如说，
94:32
you do it some days, you do it just anyway, right?
你有些日子就这么做，反正就做了，对吧？
94:34
You like take one crack at something and go,
你就像，对某件事试一次，然后就说，
94:36
okay, I learned like I felt around.
好吧，我学到了，就像我摸索了一下。
94:38
I learned all the things.
我学到了所有东西。
94:41
I'm gonna just refresh and just like do the first shot again
我要直接 refresh，然后就像第一次那样再来一次。
94:43
and it works and it goes well.
而且它管用，也运行得很好。
94:45
And so doing that a decade scale
所以这么做是十年量级的，
94:47
with massive customers running on it,
大量客户都跑在上面，
94:50
you just make things differently.
你就会换一种方式来做。
94:52
Like Vitas is still a phenomenal piece of engineering
比如 Vitas 依然是一个现象级的 engineering 杰作，
94:55
and it is far superior to Niki is right now
而且它比现在的 Niki 要强得多，
94:58
or anything else out there
也比市面上任何其他东西都好，
94:59
in terms of database scalability.
在 database scalability 方面。
95:01
But over time, Niki may outdo it just from the sheer nature
但时间久了，Niki 可能就会超过它，光是架构不同这一点就让我们特别特别兴奋。
95:06
that it's like the architecture is different
同时我们也很期待继续把 Vitas 往前推。
95:10
and we're very, very excited about it.
Vitas 会持续不断有新东西加进来，因为 MySQL 能做到一些事。
95:12
And we are excited to keep pushing Vitas.
其实我不觉得哪个真的会比另一个好。
95:14
Vitas is getting new things added continually
Vitas 一直在不停地加新东西。
95:17
because of some of the things that my SQL can do.
因为 MySQL 能做的一些事情。
95:20
Like I wouldn't say one's gonna be better
比如我不会说哪个会更好，
95:21
actually than the other.
实际上比另一个。
95:22
It's gonna be different.
这会不一样的。
95:24
But taking a database system that's similar
但是拿一个 database system，它类似
95:27
to my SQL in architecture is sort of a SQL database.
于 MySQL 的 architecture，那就是一种 SQL database。
95:30
And then doing it again, incredible fun.
然后再做一次，简直太好玩了。
95:34
Like the real do-over was incredible fun.
就像是真正的推倒重来，特别有意思。
95:36
Everyone's enjoyed it.
大家都乐在其中。
95:37
And I feel very lucky to watch that teamwork.
我觉得很幸运能见证那样的团队合作。
95:41
Yeah, both sides of the major database coin,
是啊，主流 database 这枚硬币的两面。
95:45
my SQL and Postgres will forever be in some version of a war.
my SQL 和 Postgres 永远都会处于某种战争状态。
95:52
We've always been around here Postgres for life.
我们一直在这儿，终身 Postgres。
95:54
So thank you for coming around Sam.
所以谢谢你过来，Sam。
95:57
Even though you feel that Postgres is in fact
尽管你觉得 Postgres 实际上
96:00
inferior to my SQL and we'll leave that for a different day.
不如 my SQL，这个我们改天再聊。
96:02
Oh, can I hit that nuance there?
哦，我能点一下那个细微差别吗？
96:04
Sure, sure.
当然，当然。
96:05
It is just different, right?
就是不同而已，对吧？
96:07
My SQL lost the crown
MySQL 丢了王冠
96:09
because they stopped listening to their users.
因为他们不再听用户的了。
96:11
They stopped building for developers.
他们不再为开发者打造东西了。
96:13
It is a very good database to operate.
它是一个非常好运维的 database。
96:15
And I think it is still easier to operate than Postgres, right?
而且我觉得它还是比 Postgres 更好运维，对吧？
96:18
Like high availability is easier, all of these things.
比如 high availability 这些方面都更容易。
96:21
Like we run into so many issues with customers
比如我们在客户那儿遇到好多问题
96:23
and you just have to explain to them,
然后你就得跟他们解释，
96:25
this is how Postgres does it.
Postgres就是这么做的。
96:26
Vacuuming is bad because it's a core thing.
Vacuuming不好，因为它是核心的东西。
96:30
Like it's a core decision similar to the replication
就像是一个和replication类似的核心决策。
96:34
is not as how I would like it to be, but it's different.
它不是我希望的样子，但确实不同。
96:37
They have a different philosophy
他们的理念不同。
96:38
and they've achieved something phenomenal.
而且他们做出了很了不起的成就。
96:40
It is the default database now
它现在已经是默认的数据库了。
96:42
because people have done a good job
因为人们做得很出色。
96:44
of helping you operate it.
帮助你操作它的。
96:45
Haroku, the major clouds have taken that pain away.
Haroku，各大 clouds 已经把那种痛苦消除了。
96:51
And so now you're left with a dev interface
所以现在你剩下的就是一个 dev interface
96:53
and the extensions ecosystem so much better, so much better.
而且 extensions ecosystem 也好太多了，好太多了。
96:57
So I love Postgres now.
所以我现在太爱 Postgres 了。
96:58
Like getting to work with that audience
比如能和那样的用户群体一起工作
97:00
and seeing these brilliant and amazing companies
看到这些才华横溢又了不起的公司
97:03
and using it and then to bring something truly differentiated
然后使用它，再去带来一些真正差异化的东西。
97:06
from our biases into my SQL world, it's phenomenal.
从我们的偏见进入我的 SQL 世界，这太棒了。
97:09
So I wouldn't say one's better or worse,
所以我不会说哪个更好或更差，
97:10
they're just different.
它们只是不同而已。
97:12
But like they will get better over time
但它们会随着时间变得更好，
97:16
or like they all wear them.
或者就像它们都带着这些偏见。
97:17
I hope we can play a role in making both better
我希望我们能在让两者变得更好上发挥作用，
97:19
continually at this time.
持续地，在此时。
97:21
Well you're a minute or two away from a customer column.
好了，你离 customer column 还有一两分钟。
97:24
I appreciate you going literally to the one yard line
我真的很感激你陪我走到了最后一码线。
97:27
with me on this podcast.
就在这个播客节目里。
97:30
Is there anything you want to say in closing
最后你有什么想说的吗？
97:31
just as we tail out and to show anything whatsoever?
就在我们慢慢收尾的时候，有没有任何想展示的东西？
97:36
I loved the episode we recorded all those years ago.
我很喜欢我们很多年前录的那一期节目。
97:39
It was one of my favorite I've ever done.
那是我做过的最喜欢的节目之一。
97:41
It's one of the ones that have been,
它是那种多年以来
97:43
people have referred to over the years
人们一直在提到的节目。
97:45
and people still, you know,
而且人们仍然，你知道，
97:46
that was maybe how they heard a plant scale
那也许就是他们听说一个工厂的规模的方式，
97:48
and they're very complimentary about how we both did
而且他们对我们俩的表现赞不绝口，
97:51
on that show.
在那个节目上。
97:52
And so I'm really glad to be back.
所以我很高兴能回来。
97:54
And it's amazing to see these enduring things
看到这些持久的东西真是令人惊叹。
97:57
that go through the industry has changed a lot
它们一直延续着，而行业已经变化很大，
98:00
while you do what you do.
而你在做你所做的事情的时候。
98:01
And so it's amazing to have something enduring
所以，能有一个持久的东西让我们寄托，真的很棒。
98:03
that we anchor on and I enjoy listening
我喜欢听，也很享受能参与其中。
98:06
and I've enjoyed being part of it.
我只想说声谢谢，感谢你为我们所有听众所做的一切。
98:08
So I just want to say thank you
也感谢你打造了这个你值得骄傲、我们也愿意订阅的节目。
98:10
and for everything you've done for all of us
以及你为我们所有人做的一切，
98:11
that are listeners.
也就是我们这些听众。
98:13
And for building something that you should be very proud of
而且说到打造一个你应该引以为傲、我们也愿意订阅的东西，我敢打赌你身边有一群很棒的人在为它增光。
98:17
and that we like subscribing to
而且我们喜欢订阅那个。
98:20
and that I really enjoyed being part of too.
而且我也真的很享受参与其中。
98:22
And I'm very happy to be back
我很高兴能回来，
98:24
with a company that's defaulted live
回到一家默认能存活的（公司），
98:26
with hope for the users listening now
对现在正在收听节目的用户怀抱希望，
98:27
and potential users in the future.
也对未来的潜在用户怀抱希望。
98:30
So thank you to yourself and your audience,
所以感谢你、你的听众，
98:32
and everyone involved.
以及所有参与其中的每一个人。
98:33
So you've got something enduring and really amazing.
你们打造出的是持久且真正了不起的东西。
98:36
Well, thank you.
嗯，谢谢。
98:37
Thank you very much.
非常感谢。
98:38
It's been a lot of work as you know over the years.
你知道，这些年来我们付出了很多努力。
98:40
I mentioned the pod we were,
我刚才提到过我们的播客，
98:42
our inception was our birthday was November 2009.
我们的创办日，也就是我们的生日，是2009年11月。
98:47
So we've been here for a very long time.
所以我们在这里已经很久了。
98:48
We're coming up on our next birthday here later this year
今年晚些时候我们就要迎来下一个生日了。
98:51
and I forget how many years has been,
我都忘了已经多少年了。
98:53
but as long as I care about this community,
但只要我还在乎这个社区，
98:57
which I think will be till the day I die,
我觉得这会一直持续到我死的那天，
98:59
I'll be producing some version of a podcast.
我就会一直做某种形式的 podcast。
99:01
Hopefully is always the change law
希望它永远都是 change law。
99:04
because I feel like the name we chose,
因为我感觉我们选的名字，
99:07
changelaw.com even is like,
changelaw.com 甚至就像是，
99:09
you can't erase it.
你根本抹不掉它。
99:10
It's an institution.
它已经是一种传统了。
99:12
It may change and have different variations over the years,
这些年可能会有变化，也有各种不同的形式，
99:14
but I feel like we're,
但我觉得我们，
99:16
we've always struck the right balance
我们一直都能找到合适的平衡
99:17
of who to talk to, why to talk to them, et cetera.
关于和谁聊、为什么聊，等等。
99:21
And obviously having you a part of that
显然，有你的参与
99:23
has been amazing.
真是太棒了。
99:24
So thank you, Sam.
所以谢谢你，Sam。
99:24
And thank you for all the work you do
也感谢你所做的所有工作。
99:26
of playing the scale and the whole team there.
发挥规模效应，以及整个团队在那里。
99:27
No, thank you.
不，谢谢。
99:28
And just go to our customers.
然后就归功于我们的客户吧。
99:29
It's not me.
这不是我。
99:30
It's our customers.
是我们的客户。
99:31
We're only here because our customers.
我们在这里完全是因为我们的客户。
99:32
Thank you to all of them.
感谢他们所有人。
99:34
We love our customers.
我们爱我们的客户。
99:35
They're very special people.
他们是非常特别的人。
99:38
But like you have truly achieved the change log
但是说真的，你已经成了整个行业的 change log。
99:40
for the industry.
你干这行很久了。
99:41
You've been doing this long.
你一直在追踪变化。
99:41
You've tracked the change.
你知道你该做的就是回头
99:43
You know what you should do is you should go back
然后回顾所有的节目。
99:44
and like reflect on all episodes.
我打赌你有很多了不起的人在向它致敬。
99:46
I bet you've got some amazing people that honor it.
我敢打赌你有一些很棒的人会尊重它。
99:48
Like there's like, you could do a super car.
就像，比如说，你可以做一辆超级跑车。
99:49
So many.
太多了。
99:50
Yeah, I mean, like it has really been
是啊，我是说，它真的已经
99:52
a change log for the industry and you should keep going
成为了行业的 change log，而且你应该继续下去
99:54
because we're about to enter a new era
因为我们将要进入一个新时代
99:55
that will be fun and different and really cool.
那会很有趣、很不同，而且真的很酷。
99:58
I think so too.
我也这么觉得。
99:59
This next era is going to be interesting.
接下来的这个时代会很有意思。
100:01
And I sure hope we all endure it.
我当然希望我们都能扛过去。
100:04
As you've said, I sure hope we can and we do.
正如你说的，我当然希望我们能，而且我们也确实做到了。
100:08
Because things change seemingly every single day.
因为事情好像每天都在变化。
100:11
But Sam, thank you so much for being a friend,
但是 Sam，真的非常感谢你作为朋友，
100:13
coming back on the pod, sharing all your wisdom.
回到播客，分享你所有的智慧。
100:15
And really just sharing the struggle
也真正分享了那份挣扎，
100:17
of what it takes to last in this industry
要在这个行业坚持下去需要什么，
100:20
and what it takes to be a great player in this industry.
以及要成为这个行业里的顶尖玩家需要什么。
100:22
I appreciate your perspective, your wisdom and your team,
我欣赏你的观点、你的智慧和你的团队，
100:25
your product and your customers.
还有你的产品和你的客户。
100:26
Thank you.
谢谢。
100:27
Thank you.
谢谢。
100:31
You better keep that glazing in.
你最好把那些吹捧的话留着。
100:33
If I hear that edited out, I will be furious at you.
如果我发现那段被剪掉了，我会气疯的。
100:35
You better sit there and have other people go,
你就坐在那儿，听别人说，
100:38
wow, yeah, actually.
哇，对，确实。
100:40
Now I think about it, Adam is amazing
现在想想，Adam 真的太厉害了。
100:42
and we're great.
而且我们也很棒。
100:42
Oh man.
哎呀。
100:43
You better keep it in.
你最好忍住别说出来。
100:44
I will be so mad.
我会很生气的。
100:45
You'll get my Twitter eye if you don't.
不然你会得到我的 Twitter 白眼。
100:47
Well, friends, thanks for tuning in to this episode
好了，朋友们，感谢收听这一期
100:50
of the change log joined by Sam Lambert,
The Changelog，Sam Lambert 也加入了。
100:53
CEO of PlanetSkill, big fan of what they're doing over there.
PlanetSkill 的 CEO，我特别欣赏他们那边在做的事情。
100:56
Love the way Sam has led this company.
我很喜欢 Sam 领导这家公司的方式。
100:59
You know, love the almost died,
你知道，我喜欢那种差点就死了，
101:02
but not default dead.
但不是 default dead。
101:04
Now they are default alive and thriving, which is cool.
现在他们 default alive 而且蒸蒸日上，真酷。
101:08
Databases have been all the rage for a very long time
Databases 已经火了好久了，
101:12
for obvious reasons.
原因显而易见。
101:12
And now in this agentic era we're in
而到了现在我们身处的 agentic era，
101:15
with agent skill being slung around everywhere,
既然 agent skill 被到处提及，
101:18
we need PlanetSkill.
我们需要 PlanetSkill。
101:19
We need PlanetSkills to focus on my sequel
我们需要 PlanetSkills 来专注于 my sequel
101:22
to focus on Postgres, to focus on Niki,
专注于 Postgres，专注于 Niki，
101:24
to focus on Metal, to focus on NVME based databases.
专注于 Metal，专注于 NVME based databases。
101:29
And that's cool stuff.
这些都是很酷的东西。
101:30
And so I hope you enjoyed this show.
所以，我希望你喜欢这期节目。
101:31
A massive thank you to our friends
非常感谢我们的朋友们。
101:33
who have sponsored this POP.
感谢赞助本期POP的朋友们。
101:34
We got our good friends over at BuildKite,
我们有在BuildKite的好朋友们，
101:37
our good friends obviously in partners over at Fly
显然还有我们在Fly的合作伙伴们，
101:41
and our good friends over at coder.com.
以及我们在coder.com的好朋友们。
101:43
And of course, to the beat freak in residence,
当然，还有驻场节拍狂人，
101:46
break master cylinder.
break master cylinder。
101:49
All right friends, this shows done.
好了朋友们，这期节目就到这里。
101:50
We'll see you again very soon.
我们很快再见。