Hacker News Reader

To be fair Ive been using for git for 8 years and I still dont quite understand it beyond the basics.

You're not alone. My experience is that about 20% of devs deeply understand Git and are very comfortable with it, and about 80% know the the basics and hope nothing goes wrong.

I'm somewhere in the middle myself. That's part of why I designed Grace to be much easier to understand. I can teach it to you in about 15 minutes, not the days and weeks it takes all of us to feel like we understand Git.

> To be fair Ive been using for git for 8 years and I still dont quite understand it beyond the basics.

I don't see why this is supposed to be a problem. What constitutes "the basics" is what you use in your everyday routine, and if that works perfectly well then there is absolutely no need to do something you never need to do.

while I don't judge someone for not knowing past the basics of git because, as you point out, if it works, it works, the very valid fear is that they'll somehow get into a funky state and have to find a git expert to fix it for them, or painfully muddle through it, with the very real fear that their work will get lost somehow. if you know what you're doing, that doesn't happen, but if you're not an expert, it's a very real thing that can happen, so it's that fear that constitutes a problem for some.

it's this black box that saves all my hard work, and if it accidentally hit the wrong button, it'll delete all my data and find my kids and scare them as well.

I was fortunate enough to dive deep into git professionally so I'm good enough with it to get myself out of trouble, but watching others use it, I can understand their worry.

Its gotten me along so I havent bothered, but occasionally I will fall into a mess and I find some improper/inefficient ways around it. Every time I try interwctive rebase I get into a huge mess where it cant apply some updates for some reason and I say f it and just do git reset hard and aoply the commits I want and force push.

if it makes you feel better i’ve been using it ~20 years and still learning…

> Not a problem I thought I needed to solve, but okay. Also this means I can't easily run it locally?

I would add that this sounds like a big step backwards, as it conveys the idea of a svn-like version control system designed for the service provider to hold your project hostage.

>> easy to use, easy to understand

> I already understand git, so does everyone on my team, and everyone that interviews...

Really? Because every team I've ever met could use git but the moment anything left the golden path they had to either 1. delete everything, reclone, and manually fix things up, or 2. turn to the one greybeard who actually did understand git. Either your team is the 99th percentile, or your definition of "understand" is rather generous.

> Really? Because every team I've ever met could use git but the moment anything left the golden path they had to either (...)

I've been using Git for over a decade and I never had the need to "delete everything, reclone".

The only time I screwed up a Git repo was when I was experimenting with storing Git repos in USB pens and one of them got corrupted. I have no idea what might lead anyone to screw up a Git repo, because that's simply unrealistic.

The easiest way is to change a commit you've already pushed, and that something somewhere else references, and then force push it.

> (...) and then force push it.

I don't think this is a good example. Forcing a push means that the repository will lose commits, but you still keep yours in your local branch. This means the repo is not broken, but at best you have a perfectly valid local repository that just happens to be out of sync.

If you rename your local branch and set it to not track the remote one, and afterwards you fetch changes from the remote branch, then you're done.

> Also this means I can't easily run it locally?

It's not meant to be a local version control system, unless you enjoy running local Kubernetes clusters (which I have to do, but don't enjoy).

It's meant to be the next big thing in version control - no reason not to go for it - which means that it would have to be picked up by the major source control hosters, and since I know what it takes for GitHub to run its infrastructure, I know that it makes much more sense to build something new on PaaS services, not on file servers. Not anymore.

> I already understand git, so does everyone on my team, and everyone that interviews...

Yeah, but do they? That's not my experience, and it's not the experience of most people I talk to about it. Most devs I've asked about it understand the basics of how to use Git, but they're still afraid of it if anything goes wrong. My guess is that the ratio is 20% deeply understand it, and 80% only know what they need to and hope nothing bad happens.

Maybe your team are all a bunch of reflog wizards... that's awesome. And uncommon.

And I almost always get laughs and head nods when I talk about the problems with Git's UX.

> Is large files the main problem this solves?

No, but it's a big problem for gaming companies, who are mostly stuck on Perforce. And Git can't handle them well without the bolted-on LFS. And with the rise in monorepos, more and more enterprises want to be able to store more and bigger files than ever before.

> And maybe requires an internet connection?

Yes, absolutely, it does. So does Git if you expect to push anything anywhere. And if you happen to be doing dev using Azure or GCP or AWS you need one too.

Building something that would become popular in the late 2020's, and assuming that users will have solid Internet connections (don't forget satellite) is what makes sense. If you're still in a situation where you need offline VCS then, Git will still be there.

> Maybe this is for pair programming?

You could use it for that, but pair programming is not a direct design intent.

> It's not meant to be a local version control system, unless you enjoy running local Kubernetes clusters (which I have to do, but don't enjoy).

You should be clear about what is a major design trait, and arguably a major design flaw.

Also, there is already standard terminology for this: centralized VCS. I don't understand why you decided to avoid objective descriptions of your project's single most important design trait and instead resort to vague meaningless buzzwords like "cloud-native" or "real-time". In fact, in light of this those terms start to sound like weasel words used to deceive the reader.

When I hear "cloud-native" I think "built to scale up well". As opposed to "built to run on file servers" which means "doesn't scale well at all".

Is that just me?

Also, [1]. I start by saying it's centralized. I'm proud of it. It's the right direction for moving version control forward. And modern use of Git isn't really distributed anyway; it's centralized. We don't push to production from our dev boxes.

[1] https://github.com/ScottArbeit/Grace/blob/main/docs/Frequent....

Hm. To me, cloud-native screams and PaaS completely out of my control and completely subject to the whims of some company that does not have my best interest in mind. It implies the impossibility of firing up some local instance for experimenting without having fear of leaving traces. In short, something to be avoided if possible.

> It's meant to be the next big thing in version control

I wish you the best but kind of hope it isn't. I want my vcs to be local and conceptually simple. I definitely don't want a client-server architecture!

> And I almost always get laughs and head nods when I talk about the problems with Git's UX.

Yes, the UX is bad. But it's conceptually simple: blobs, trees, commits, pointers (branches etc). I really fear someone will replace Git with something having a better UX but conceptually much more complex.

Complexity bad.

We've gone over this so many times as an industry and we haven't learned yet.

Complexity bad.

I agree, complexity bad. So why do you like Git? :-)

Git is _incredibly_ complex to understand, as proven for almost 20 years by the vast majority of people who have been forced to use it. And by quite a bit of academic and industry research, for instance, [1].

I can teach you Grace in about 15 minutes. How many days and weeks does it take most devs to start to understand Git? And even when they do, for most, it's only the basics, and please don't let anything go wrong. I mean, there were people for over a decade who made their living running week-long workshops on learning Git. I don't see how you could run a half-day-long workshop teaching Grace, unless you go really slowly.

If you're one of the probably 20% or so who really feels like they understand Git and are in control of it, that's awesome. But you're projecting your experience more widely if you think that's the norm. It's not.

As for local, well, if you're working with a team on GitHub or GitLab or Azure DevOps or some other hoster, you're already doing centralized VCS, you're just using a decentralized VCS to do it. Most shops don't let you push to production from your dev box, right?

[1]: https://static.googleusercontent.com/media/research.google.c...

> How many days and weeks does it take most devs to start to understand Git?

A few weeks to understand a technology you’re gonna be working with for years to come is nothing.

> And by quite a bit of academic and industry research, for instance, [1].

Isn’t that a positive aspect? It’s well studied and there are wealths of info about it for just about anything you need to do.

I see Grace less as a git replacement and more as its own niche. I certainly see the benefits of easier onboarding and centralization for companies and education but those who grew up with git will likely keep using it

Personally, I avoid anything “cloud-native”

> I agree, complexity bad. So why do you like Git? :-)

I think you're trying to fabricate problems where there are none.

Git's UX problem lies in the way it's CLI is not intuitive for those unfamiliar with it, but a) using GUI frontends like SourceTree allows newbies to eliminate that issue, b) with time you onboard to the CLI and everything just works.

At best, your suggestion to use another user interface is equivalent to suggesting Git users to adopt a new GUI frontend that's polished in a different way.

> Git is _incredibly_ complex to understand,

I don't know what you can possibly mean by "incredibly complex".

For end users, your mental model can be limited to a) you clone a repository, b) you commit and your changes, c) you push your changes to make them available to everyone, d) you pull everyone's changes to have access to them.

This is hardly rocket science. I mean, why do you think Git managed to become the world's de facto standard VCS and sets the gold standard for VCSs?

> I think you're trying to fabricate problems where there are none.

No, I'm not. The problems with Git's UX are well-documented, and have spawned many projects over the last 10+ years trying to deliver "Git, but easier" or "Git, but better", so it's not just me who sees this.

I'm happy for you that you're comfortable with Git, or so indoctrinated to the workarounds required to use Git well that you're used to them. I believe it's time for something very different, and much easier to understand.

> why do you think Git managed to become the world's de facto standard VCS

I think it was because Git has lightweight branches, and an ephemeral working directory, both of which made it nicer to use than the older, slower, centralized VCS's. I've kept both of those features in Grace.

I also think it was because of GitHub wrapping a lightweight social network around Git and popularizing it, at the same moment that shared open-source dev really started to catch on as an idea. Without GitHub, Git wouldn't have won.

I do not think it was because Git is easy to use, overall. Again, maybe 20% of devs really get it, and the rest don't and just hope nothing bad happens. It was better in some important axes, and we've all paid the bad-UX tax to get those better parts, but 2005 was a long time ago, with a very different set of network and hardware conditions, and we can do better.

> I believe it's time for something very different, and much easier to understand.

I'd like to reiterate my request for clarification of the concepts behind Grace. If it's as easy to understand as blobs, trees, commits, and refs, I'm sold!

> Without GitHub, Git wouldn't have won.

True, but git is good not because of GitHub, git is good because it's so simple.

I'm scared you will replace git with something easier but a lot more complex. I don't want easy, I want simple.

> Git is _incredibly_ complex to understand

No it isn't. Git is just blobs, trees, commits, and refs. Git isn't easy but it's conceptually simple. I'll take simple over easy anytime.

If you could explain the concepts Grace is built on, that'd be great!

> If you're one of the probably 20% or so who really feels like they understand Git

Again, blobs, trees, commits, and refs. I don't know all of git's crazy commands, but they can be explained in terms of these four simple concepts.

> As for local, well, if you're working with a team on GitHub or GitLab or Azure DevOps or some other hoster, you're already doing centralized VCS, you're just using a decentralized VCS to do it.

No, that is still fully decentralized. Each team member has a full copy of the repository, which, if GitHub or GitLab or Azure DevOps or whatever suddenly disappeared could be promoted to be the new shared source of truth.

At my $job-2 we were using GitLab but it often went down. I just set up a git repository on one of my servers and authorized everyone's ssh keys: it took me ten minutes and we had a way to collaborate even with GitLab down. Yes there weren't pull requests or anything, it was just a dumb repo used over ssh. But that was the whole point!

> Git isn't easy but it's conceptually simple

That may be, but the state of your repo, expressed as a combination of the 4 things listed after doing an arcane and globally unique sequence of git commands, is in no way conceptually simple. If it was, the implicit lurking horror that every programmer knows lies inside git would not be a shared traumatic developer coming of age story. You are the exception here.

> No, that is still fully decentralized.

The word decentralized does not really apply here. Is Figma decentralized? Do you ever do peer-to-peer git? Do you really? Or do you kind of just have a single source of truth with a lot of local copies, that allow offline-first workflows that you rarely need.

> if GitHub or GitLab or Azure DevOps or whatever suddenly disappeared could be promoted to be the new shared source of truth.

This is not a selling point you think will be taken seriously, right?

> Do you ever do peer-to-peer git? Do you really? Or do you kind of just have a single source of truth with a lot of local copies, that allow offline-first workflows that you rarely need.

Have you read my comment? Yes I did. I use GitHub to sync my git things, but if it were to disappear I could easily start using something else. Sometimes I push between other remotes too. Each of the repos is self-contained and whole by itself.

> > if GitHub or GitLab or Azure DevOps or whatever suddenly disappeared could be promoted to be the new shared source of truth.

> This is not a selling point you think will be taken seriously, right?

Again, have you even read my comment? This is not a theoretical scenario, it's a thing that happened to me in the past. Thanks to git's distributed nature it was very easy to work around.

> Git is just blobs, trees, commits, and refs

By this logic the entire computing industry is so simple a toddler could understand it - “It’s just ones and zeroes”

Unfortunately the complexity of a system is greater than the number of its components...

No. Vast majority of software does not consist of four simple and easy to grok concepts.

Vast majority of software consists of badly designed abstractions full of various hacked on workarounds for exceptional cases. Such as Subversion, what a horror that was!

> Yes, absolutely, it does. So does Git if you expect to push anything anywhere. And if you happen to be doing dev using Azure or GCP or AWS you need one too.

Sure if you want to push, but only maybe 10% of my Git commands relate to pushing/internet related stuff. The majority of my work is local-only commands that can be run on an airplane without wifi. Git lets me defer the internet required stuff to a later time. Its not clear Grace will allow me to do that at all.

Also I once had a case of working on an air-gapped network. That was an interesting case that I'm not sure Grace would be suitable for at all? Granted that's super niche.

The fact that most of your Git commands are local-only is an artifact of how Git works, but I expect that ~100% of the time, you have an internet connection, so the fact that Grace needs to be connected to the cloud just isn't a thing I worry about.

I'm not writing a new VCS based on the 0.00000001% "but I'm on an airplane without WiFi" case.

There's ~0% reason in 2024 to build software for offline use cases, and even less reason in 2026 and 2028. I'm happy to cede that to Git if you really need it.

As an industry, we fetishize offline for version control only because Git sort-of does that. Again, it doesn't really... you still have to push to do real work with your team, but we need to stop pretending that's a hard requirement. It's totally not, it's just a "feature" of Git that gets in our way today more than it helps us.

> Also I once had a case of working on an air-gapped network.

Coming from Microsoft, and being familiar with the air-gapped Azure instances for government, I designed Grace to be able to run on those Azure clouds. In other words, all of the PaaS services that Grace would use on Azure are present in those clouds.

Even the air-gapped world isn't "offline", it's totally networked, just on a network that's not connected to the Internet.

I haven't specifically looked at similar AWS instances, but I have to believe it's possible there, too.