Ole Bulbuk

Back-end engineer since the nineties,
co-organising GDG Berlin Golang,
developing open source software.

[Mitchell Hashimoto @mitchellh]

I strongly believe there are entire companies right now under heavy AI psychosis and its impossible to have rational conversations about it with them. I can't name any specific people because they include personal friends I deeply respect, but I worry about how this plays out.

I lived through the great MTBF vs MTTR (mean-time-between-failure vs. mean-time-to-recovery) reckoning of infrastructure during the transition to cloud and cloud automation. All those arguments are rearing their ugly heads again but now its... the whole software development industry (maybe the whole world, really).

It's frightening, because the psychosis folks operate under an almost absolute "MTTR is all you need" mentality: "its fine to ship bugs because the agents will fix them so quickly and at a scale humans can't do!" We learned in infrastructure that MTTR is great but you can't yeet resilient systems entirely.

The main issue is I don't even know how to bring this up to people I know personally, because bringing this topic up leads to immediately dismissals like "no no, it has full test coverage" or "bug reports are going down" or something, which just don't paint the whole picture.

We already learned this lesson once in infrastructure: you can automate yourself into a very resilient catastrophe machine. Systems can appear healthy by local metrics while globally becoming incomprehensible. Bug reports can go down while latent risk explodes. Test coverage can rise while semantic understanding falls. Changes happens so fast that nobody notices the underlying architecture decaying.

I worry.

[Ole Bulbuk @flowdev_org]

@tsenart I think it's more about sec teams now being *forced* to learn and use these tools next to everything they have on their plate already.
And even much more that the time to roll out patches is reduced to minutes.
This essentially removes any chance to test patches first.

[Ole Bulbuk @flowdev_org]

@goinggodotnet Not in Germany either. 😉

[Casey Handmer @CJHandmer]

A rare personal note. My dear friend Ivo (the dad) has a rare genetic liver disease (PSC) and has, after concerted persuasion from those of us who love him dearly, finally consented to a public appeal for a living donor.

Liver donation is quite miraculous in that one can give half and both halves grow back.

Fwiw, none of the AI models yet understand how to cure PSC, though I would bet that the next generation of humans will be more lucky.

I've never asked for RTs etc before but I think it would help in this case!

Link below.

[Ole Bulbuk @flowdev_org]

@duginabox @ibuildthecloud If they are your (de)serializers, you can build a custom way to hand any data to them. 🤷‍♂️

[Ole Bulbuk @flowdev_org]

@VaughnVernon And what are those better ways?
I would love to learn more about it

[Ole Bulbuk @flowdev_org]

@JasonThorsness I'm finishing a parser combinator library that can report multiple errors.
Only parsing (mathematical) expressions isn't finished yet.
https://t.co/Q30LKFMxQu

[Ole Bulbuk @flowdev_org]

@nzachow_ I think it's a good decision.
A better one might be something aspect oriented. That could help to centralize errors, logging, metrics, and other technical aspects.

[Ole Bulbuk @flowdev_org]

@zajacr There are already a lot of testing libraries. And many Gophers are happy with the standard library.
I'm not sure the world is waiting for yours. 🤷‍♂️

[Ole Bulbuk @flowdev_org]

@zajacr Clearly B and D.
want has to be the edge value. The actual one has to be great or later.

[Ole Bulbuk @flowdev_org]

@bernoussama It's good for many in-house projects.
But I wouldn't use it for open-source libraries.
I don't like to force dependencies that aren't strictly necessary on others.
And it's one more thing to keep safe/up to date.

[Ole Bulbuk @flowdev_org]

@CAFxX I would like to see an example of that, too.

[Ole Bulbuk @flowdev_org]

@VipinJangi3398 So it becomes easy to understand. 😀
It depends mostly on the context.

[Ole Bulbuk @flowdev_org]

@thehumanmaskot Idiomatic Go code. It often gets the most optimizations from the compiler, too. 😀
Unless you can prove that it's a bottleneck.
Then you write efficient Go code. 😁

[Ole Bulbuk @flowdev_org]

@CAFxX Your statement is so abstract. So what exactly are you talking about then?

[Ole Bulbuk @flowdev_org]

@Rajshri0987 Usually web APIs.
I'm using GORM only for a desktop APP.
ORMs are great for simple cases. But if things get complex, they become a bottleneck, IMHO.

[Ole Bulbuk @flowdev_org]

@sinore69 I am not a fan of try/catch. It's a lot of extra keywords and behavior for very little or even no gain.
I would love to have good aspect support in Go.
That could centralize error handling, logging, instrumentation,...
Much more orthogonal and bang for the buck.

[Ole Bulbuk @flowdev_org]

@sinore69 What kind of project interests you?
I would start with a rather small project.
Things like Kubernetes are really cool, but most people need years of training and experience before they can provide value to such a project.

[Ole Bulbuk @flowdev_org]

@EngComp269 C++ is great for training your mind because it's so complex.
In practice, most professional programmers use simpler languages.
Many aren't able to work on a big C++ code base.
So, no. You don't need C++ at all to become professional.