Facebook uses Mercurial, but when people praise their developer tooling it's not just that. They're using their CLI which is built on top of Mercurial but cleans up its errors and commands further, it's all running on their own virtual filesystem (EdenFS), their dev testing in a customized version of chromium, and they sync code using their own in-house equivalent of GitHub, and all of it connects super nicely into their own customized version of VS Codium.
this post was submitted on 18 Jul 2024
104 points (96.4% liked)
Programming
17419 readers
49 users here now
Welcome to the main community in programming.dev! Feel free to post anything relating to programming here!
Cross posting is strongly encouraged in the instance. If you feel your post or another person's post makes sense in another community cross post into it.
Hope you enjoy the instance!
Rules
Rules
- Follow the programming.dev instance rules
- Keep content related to programming in some way
- If you're posting long videos try to add in some form of tldr for those who don't want to watch videos
Wormhole
Follow the wormhole through a path of communities [email protected]
founded 1 year ago
MODERATORS
Damn that sounds sick
The source control was so smooth and pleasant that it convinced me that git isn't the be all end all, and the general developer focus was super nice, but some of that tooling was pretty janky, poorly documented, and you had no stack overflow to fall back on. And some of it (like EdenFS), really felt like it was the duct tape holding that overloaded monorepo together (complete with all the jankiness of a duct tape solution).
What you can do with 84000 employees
And some good management. Probably not a common opinion around here, but my company is not a tenth of that size, with a hundredth the number of devs, yet different teams still end up copy pasting libraries. Because it's faster than convincing management DevOps is important.
And kinda horrifying. If something goes wrong, no Google, it's straight to IT
There's probably specific ticket queues and wiki/doc spaces for each support team.
Problem with an app? Send it to the internal dev/support team. Then if needed it gets routed.
The inhouse tooling from the massive tech companies is very cool but I always wonder how that impacts transferrable skills. I work in a much smaller shop but intentionally make tech decisions that will give our engineers a highly transferrable skill set. If someone wants to leave it should be easy to bring their knowledge to bear elsewhere.
Speaking from my own experience and a few other seniors I work with, you try to recreate solutions you like at those smaller shops. It may not be identical, but you know what's possible.
I came into a company that didn't have a system to manage errors. At my old job, errors would get grouped automatically and work can be prioritized through the groupings. The new company only handled errors when they saw it, by word of mouth.
Immediately went to work setting up a similar system.
There’s also a whole industry of ex-Googlers reimplementing Google tooling as SaaS services to sell to other ex-Googlers at other companies.
There’s even a lookup table: https://github.com/jhuangtw/xg2xg
(some of those are open source projects, some are SaaS services)
The inhouse tooling from the massive tech companies is very cool
I agree. I personally know nothing about tooling like this but I went through the tooling used at rockstar for example GTA V and it was very cool to how much they have automated and made tools easier to use.
Oh, it impacts indeed. And I would expect that to be partially to keep the devs from hopping away, as they will have a hard time transferring
On the other hand, onboarding is longer and wastes more time and money of the company ¯\_(ツ)_/¯
I'm pleased to report that git has made significant strides, and git submodule can now be easily used to achieve a mono-repo-like level of painful jankiness.
My best VCS experience so far was when working with Plastic SCM. I like how it can track merges, the code review workflow is also nice, and in general it was pretty nice to work with.
Fuck Unity, who paywalled it into unusability, though. Another amazing project that was bought and killed by absurd monetization by Unity, same as Parsec.
How was Parsec before the acquisition?
I only really have experience after, and it's the only Unity product I've actually found that I like. My only major complaint is that it's not compatible with the base configuration of Palo Alto, but that's really more of a Palo Alto problem than a Parsec problem.
I still use Parsec for remote, and I don't have any issue with it, it works great and I like it. However, they also did offer a free SDK (Unity plugin) to integrate remote play into your game natively (just like you can have "Invite to Steam Remote Play" button from Steam SDK), which was exactly what we needed - and Steam Remote was never working without issues for us, in comparison to Parsec which worked amazingly well every time we tried it.
I found numerous mentions of Parsec SDK and how easy it is to integrate, but after Unity bought it, I couldn't find it anywhere. Only mention was that if you need it, you should contact them.
So I did that, mentioning that we are a small team of students working on a offline co-op only 2 player game in our free time, and that since Steam Remote wasn't working for us and I have great experience with Parsec, I asked what we have to do to get access to the SDK/Unity plugin.
Unity's answer? Sure, no problem, they will be happy to give us access, with first step being that we pay them 1 000 000$ for it.
Like, wtf? Did they even read the email? How out of touch you have to be, to casually ask a small student team to pay 1 000 000$?
Okay that's fair. Their pricing is awful in general, and that's especially egregious for something that used to be free
As far as performance goes, Microsoft did manage to make git work for them later on (...with many contributions upstreamed and homegrown solutions developed—but then, Facebook is the same, isn't it?).
Mercurial does have a few things going for it, though for most use-cases it's behind Git in almost all metrics.
I really do like the fact that it keeps a commit number counter, it's a lot easier to know if "commit 405572" is newer than "commit 405488" after all, instead of Git's "commit ea43f56" vs "commit ab446f1". (Though Git does have the describe format, which helps somewhat in this regard. E.g. "0.95b-4204-g1e97859fb" being the 4204th commit after tag 0.95b)
I suspect rebasing makes sequential commit IDs not really work in practice.
Rebasing updates the commit ids. It's fine. Commit IDs are only local anyway.
One thing that makes mercurial better for rebase based flows is obsolescence markers. The old version of the commits still exist after a rebases and are marked as being made obsolete by the new commits. This means somebody you've shared those old commits with isn't left in hyperspace when they fetch your new commits. There's history about what happened being shared.
Commit IDs are only local anyway.
Whay do you mean by that?
That's exactly the same in git. The old commits are still there, they just don't show up in git log because nothing points to them.
Old, unreachable commits will be garbage collected.
Does that not happen with Mercurial? If not that seems like a point against it.
I'm confused, the behavior you just said was "exactly the same in git" is now a problem for Mercurial?
Because Facebook is a terrible company that can't even build a functional website. They think they know better than the entire industry, yet can't get basic features like browser history, link sharing, back buttons, or even comments and zooming working. Fuckin idiots.
I use git daily and still wonder why I had fewer merge issues on a larger team in the 1990s with command line rcs on Solaris. Maybe we were just more disciplined then. I know we were less likely to work on the same file concurrently. I feel like I spend more time fighting the tools than I ever used to. Some of that is because of the dumb decisions that were made on our project a decade or more ago.
What kind of RCS is used always depends on the organisation. We are actually using GIT and SVN, and both make sense for the departments that are using them.
Serious question, why do they use SVN, as in what does SVN better than Git for the department using it?
The manager likes it.
While I'm not using it, since we started our small-team hobby project in git and moving away from it would be a bother, there is one use-case of SVN that would save us a lot of headaches.
SVN being centralized means you can lock files. Merging Unity scenes together is really pain, the tooling mostly doesn't work properly and you have no way how to quickly check that nothing was lost. Usually, with several people working on a scene, it resulted in us having to decide whose work we will scratch and he will do it again, because merging it wouldn't work properly and you end up in a situation where two people each did hundreds or thousands of changes to a scene, you know that the Unity mergetool is wonky at best, and checking that all of those changes merged properly would take longer and be more error prone than simply copying one persons work over the other.
We resorted to simply asking in chat if anyone has any uncommited work, but with SVN (or any other centralized VSC, I suppose) we wouldn't have to bother with that - you simply lock the scene file and be safe.
Git LFS does actually support file locking. But in general I find LFS to be hackily pasted onto Git and not very good (as with submodules).
Right, completely forgot that locking exists in SVN, and I guess it definitely makes sense if you're collaboratively editing unmergeable files.
Thanks!
Because we always used it!
SVN has the big advantage of serialized revision numbers. Which is essential for out build- and release-system.
Split up the monorepo?
That brings more problems. Despite the scaling challenges monorepos are clearly the way to go for company code in most cases.
Unfortunately my company heavily uses submodules and it is a complete mess. People duplicating work all over the place, updates in submodules breaking their super-modules because testing becomes intractable. Tons of duplicate submodules because of transitive dependencies. Making cross-repo changes becomes extremely difficult.
But if not for using submodules, how can one share code between (mono-)repos, which rely on the same common "module" / library / etc.? Is it a matter of "not letting submodules usage get out of hand", sticking to an "upper limit of submodules", or are submodules to be avoided entirely for monorepos of a certain scale and there's a better option?
You don't share code between monorepos, the whole point of a monorepo is you only have one repo where all code goes. Want to share a library, just start using it as it is just in a different directory.
Submodules are a poor way to share code between lots of small separate repos. IMO they should never be used as I have never seen them work well.
If you don't want a mono repo then have your repos publish code to artifact stores/registries that can be reused by other projects. But IMO that just adds more complexities and problems then having everything in a single repo does.
So AFAIU, if a company had:
- frontend
- backend
- desktop apps
- mobile apps
... and all those apps would share some smaller, self developed libraries / components with the frontend and/or backend, then the "no submodules, but one big monorepo" approach would be to just put all those apps into that monorepo as well and simply reference whatever shared code there might be via relative paths, effectively tracking "latest", or maybe some distinct "stable version folders" (not sure if that's a thing).
Anyway, certainly never thought to go that far, because having an app that's "mostly independant" from a codebase perspective be in it's own repo seemed beneficial. But yeah, it seems to me this is a matter of scale and at some point the cost of not having everything in a monorepo would become too great.
Thanks!