Gitly: A light and fast GitHub/GitLab alternative written in V lang (pre-alpha)

123 126 5 years ago

lbotos• 5 years ago

Unfortunately this will probably run afoul of Git's trademark:

"In addition, you may not use any of the Marks as a syllable in a new word or as part of a portmanteau (e.g., "Gitalicious", "Gitpedia") used as a mark for a third-party product or service without Conservancy's written permission. For the avoidance of doubt, this provision applies even to third-party marks that use the Marks as a syllable or as part of a portmanteau to refer to a product or service's use of Git code."

How enforceable that is... I'm not a lawyer.

https://git-scm.com/about/trademark (section 2.3)

edgyquant• 5 years ago

How do gitlab, github, gitea, etc get away with it then?

john_cogs• 5 years ago

GitLab team member here.

Git is a trademark of Software Freedom Conservancy and our use of 'GitLab' is under license.

orra• 5 years ago

GitLab and GitHub predate the trademark, and were explicitly granted an exception when the trademark was approved.

soperj• 5 years ago

Could they keep the same domain and change the name to gitly.org?

rvz3• 5 years ago

So I should be seeing incoming lawsuits for GitLab (Who also acquired Gitorious and Gitter) GitHub, Gitea, GitKraken, Gitcoin, Gitpod, GitSense and a VSCode tool called GitLens?

Why haven't they happened yet?

Downvoters: So you have ALL sources to your claims that these aforementioned Git-related services have the necessary permissions to use the Git trademark from the Software Freedom Conservancy?

I don't want 'probablys' or 'maybes' but full sources to ALL these services that explicitly show the permission granted for using the Git trademark under license.

GitLab is covered. Evidence for the rest?

sdesol3• 5 years ago

You can add my startup GitSense as well

https://public-001.gitsense.com/insights/github/repos?q=wind...

The thing is, from what I've read, if you were using Git<plus something> before Git started cracking down, they will let it slide. I also think they don't care until you hit it big and cause issues for them.

ezfe• 5 years ago

They almost certainly have permission to use the phrases that way.

shakna• 5 years ago

They're used under license. [0][1]

Git is a trademark of Software Freedom Conservancy and our use of 'GitLab' is under license

[0] https://news.ycombinator.com/item?id=28062305

[1] https://about.gitlab.com/handbook/marketing/corporate-market...

dragonwriter• 5 years ago

Unless someone has agreed to a contract binding them to the Git trademark policy, the policy would seem to serve as nothing but a statement as to the Conservancy’s preferences (except where it explicitly permits something, where it may constitute a gratuitous license); what is enforceable is trademark law.

But, for a git-related service offered in commercial trade, there would seem likely to be a trademark violation independent of the policy, so the Conservancy is probably within their rights to enforce the policy in this case.

aorth• 5 years ago

Strange. What about GitLab and Gitea? And probably others.

abdusco• 5 years ago

In the case of Gitea, an argument could be made that it's actually `gi•tea`, not `git•ea` :)

dr_kiszonka• 5 years ago

It would be a stretch but git has another meaning, which some find consistent with their perception of SWEs ;)

https://www.merriam-webster.com/dictionary/git

ketralnis• 5 years ago

So does Apple but you'd have a hard time defending your Apple Phone in court on that basis

GoblinSlayer6• 5 years ago

EcmaScript all over again. As I understand, letter reversal doesn't count?

psychoslave• 5 years ago

Is that kind of extension to trademark even legal anywhere?

That's so aweful how pissing territory behaviors are driving mindset of people involved in intellectual works.

encryptluks24• 5 years ago

I like the approach that sourcehut took overall. I would like to find a way to implement issue tracking into Git, but the mailing list approach isn't a bad idea.

q3k2• 5 years ago

I would like to find a way to implement issue tracking into Git

Maybe git-bug would work for you? https://github.com/MichaelMure/git-bug

encryptluks24• 5 years ago

Thank you. I am interested, but am concerned by this statement:

This is now more than a proof of concept, but still not fully stable. Expect dragons and unfinished business

throwaway889900• 5 years ago

Isn't that what Fossil does? https://www.fossil-scm.org/

encryptluks24• 5 years ago

Yes, but I honestly hate the idea of having to learn yet another SCM. I may reconsider at some point though, especially if more implementations emerge.

analognoise• 5 years ago

Fossil does all kinds of awesome stuff out of the box, I remember looking at it and being blown away.

algorithm3142• 5 years ago

Or https://github.com/dspinellis/git-issue

algorithm3142• 5 years ago

Don't like that it tries to mimic Github UI.

lights0123• 5 years ago

What would you do instead? They popularized that interface, and I think it's very usable.

adamrezich• 5 years ago

anyone here keep up with V and can give a quick rundown of how development of the language has been going? I used to keep up with it out of curiosity but haven't checked in in awhile.

lazypenguin• 5 years ago

V and it's creator get a lot of flack here on HN and the criticisms are probably warranted. However, I personally enjoy seeing them out here hustling and really trying to make their language succeed despite all the shortcomings. As someone who skews more apathetic, I enjoy seeing people driven by so much passion.

Now, it's time for me to be one of those critics! It's easy to promote the speed and size of your github alternative when it has a fraction of the features. I'll be curious to see where this goes and I think it has some value but only as a way to dogfood V itself.

rvz3• 5 years ago

It's easy to promote the speed and size of your github alternative when it has a fraction of the features.

I could have said that about Gogs years ago since a high school student made that in their spare time.

Once projects become open-source, the community usually takes over that work. Just like many others, the same could happen to this project, just like how interested contributors did so with the V language as soon as that was open-sourced.

junon5• 5 years ago

Can you perhaps link to some of said criticism? I'd be interested to read it.

kangalioo• 5 years ago

Here is a nice mini series of blog posts about V https://christine.website/blog/series/v

tleb_• 5 years ago

Some can be found in every thread about the project: https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que...

pityJuke• 5 years ago

This HN User has been quite outspoken on his criticism: https://news.ycombinator.com/threads?id=vlang1dot0

penagwin2• 5 years ago

V and it's creator get a lot of flack here on HN and the criticisms are probably warranted.

Hi fellow Penguin :D!

I'm impressed by the creator's dedication here too, but he's clearly biting off more then he can chew - he's been saying that he's making a

* Web framework * GUI framework * SQL ORM * A language that has a lot of promises - seriously checkout their list. * Github alternative * VIM alternative

This stuff is awesome as a project, but each one of these have teams of dedicated engineers sponsored by many companies to hammer out issues and make them as robust as possible. And while he certainly has working demo's of each, they're demos. There's a huge difference between a POC and a robust framework.

A half baked framework is useless sadly, I think he's spreading himself too thin while have a roadmap that looks like a programmer's christmas list.

hobo_mark• 5 years ago

Also

* package manager

* operating system

amedvednikov9• 5 years ago

I'm not working on these alone :)

machiaweliczny2• 5 years ago

I think it's good test of lang features

sdesol3• 5 years ago

As the author states, he is not working alone as the following shows:

https://public-001.gitsense.com/insights/github/repos?p=impa...

When you have a large number of people in Zone 2 and nobody in Zone 1, this is a very good sign that effort is being fairly distributed. And if you look at the activity for the last year:

https://public-001.gitsense.com/insights/github/repos?q=comm...

it is extremely active with 98% days active. And if I had to guess, I would say there are at least 5-7 people working on this full time based on the following:

https://public-001.gitsense.com/insights/github/repos?p=cont...

Full Disclosure: I'm the creator of the tool with the insights above.

agumonkey4• 5 years ago

i was annoyed by the hyperbolic ambitious claims but if the guy and his friends pour out solid code then there's nothing else to say :)

also they seem to aim at frugal fat free programming, i cannot not like that

crazy_horse3• 5 years ago

Those that lie often either don't respect other people or they don't have the ability to know when they are lying. Which is scarier?

Go, Nim, and Zig have similar focuses and their devs have put in years of public, good work.

Some of the take downs at the time he was making his initial claims are absolutely brutal and make it clear that he not only couldn't back up his crazy claims, he wasn't getting the small stuff right.

https://christine.website/blog/v-vaporware-2019-06-23

He still hasn't learned.

turminal• 5 years ago

if the guy and his friends pour out solid code then there's nothing else to say :)

Unfortunately, they don't.

GoblinSlayer6• 5 years ago

It's easy to promote the speed and size of your github alternative when it has a fraction of the features.

At leat this fraction of features works, what can't be said about gitlab and is only partially true for github. It shows repos, files, commits and what not, what features are missing?

udia• 5 years ago

Good to see another open source git development platform. I am currently using sourcehut and am very happy with it.

https://sourcehut.org/

yewenjie• 5 years ago

They claim to be the fastest among all such platforms too.

https://forgeperf.org/

Wowfunhappy• 5 years ago

Wow—they weren't kidding when they said fast! It's interesting how I've been conditioned to expect delays on websites—so when I come across something like this which loads seemingly-instantly, it's super noticeable!

sparaker• 5 years ago

amedvednikov9• 5 years ago

Hi,

Author of V and Gitly here.

Gitly is currently in alpha, and works more like an example/documentation of the V web framework and V ORM.

You can't even sign in at the moment.

It will be launched this month. We're actually going to switch to Gitly CI, since Github Actions has been really slow for us.

NeckBeardPrince• 5 years ago

You might have trademark issues with Gitlab's Gitaly https://gitlab.com/gitlab-org/gitaly

zerr• 5 years ago

Saw your hiring post - may I ask, what funding did you get?

Zababa• 5 years ago

Good luck on your projects! You should probably add Gitea to your comparaisons, at it seems to be the closest to your project (written in Go, lightweight).

WhyNotHugo• 5 years ago

Have you considered integrating with git’s pull request model (eg: sending an email) rather than just cloning GitHub web-based fork-based model?

svnpenn• 5 years ago

Please add "contributions in the last year", thats the one thing keeping me from using GitLab:

https://gitlab.com/gitlab-org/gitlab/-/issues/22578

kvark• 5 years ago

Looks like V author/community is on the track to rewrite the world. Expect operating system stuff next? I'm shocked by how quickly it goes along, though. Gitly is quite usable already, and I appreciate the swiftness. I'd think that Rust community is rewriting the world, but it's happening at much slower pace. Is V's basis strong enough to do this? Seeing the issues like [1] (about the basic memory management stuff) open makes me suspicious.

[1] https://github.com/vlang/v/issues/6456

penagwin2• 5 years ago

Already beat you to it https://github.com/vlang/vinix

crazy_horse3• 5 years ago

Could someone explain to me why, after seeing how many exaggerated claims the author has made, they have invested their time into building up a community around it?

I'd be straight up terrified to trust anything serious with this guy's work.

In this instance, the demo won't even load for me and the documentation page can't be bothered to put the most basic of padding in it. Why even release publicly?

Promises, promises.

andrewnc• 5 years ago

My honest opinion (as someone with no dog in this fight) is that the author was naive and really bad at PR, but is improving overtime.

I generally encourage new programming language efforts, so I'm willing to be more generous with forgiveness in the hopes that this succeeds.

nemo1618• 5 years ago

I confess that I've found it both enraging and depressing to see that the "bootstrapped-by-hype" model works even in programming language development, which I had (naively, I guess) assumed would be more immune to that sort of thing. I suppose the field is composed of fallible humans just like any other.

TeddyDD• 5 years ago

Medvednikov seems to be good at marketing ¯\_(ツ)_/¯

Given the over-prommise-under-deliver launch of V, it boggles my mind that it gained so much traction on GH.

machiaweliczny2• 5 years ago

Same could be said about Elon Musk but he might become first trillionere. Sometimes you have to try bold stuff. I'm rooting for them.

vlang1dot06• 5 years ago

Yeah, it's absolutely wild. Even the main contributors don't really know what they're talking about, they just regurgitate the party line.

http://imgur.com/a/hUXR6s3

For instance, when asked how the memory management system works, here you see a team member give the standard response: the compiler inserts calls to free when it detects the variable is no longer used. This seems reasonable to anyone who hasn't thought much about PL/compilers before. But when pressed further, you see Alex admit "most stuff is simply cloned ATM" which is in stark contrast to the "autofree handles 90-100% of objects for you" on their homepage.

Or this where Alex shows he has no idea what UB is?

http://imgur.com/a/mbLtASH

UB is a property of the language's abstract machine. Common C UB like dereferencing a null pointer or signed integer overflow has well defined semantics on basically any commodity hardware manufactured in three decades and yet it is still UB not because of the runtime properties but because the C abstract machine says it can't happen and compilers are free to optimize based on that assumption. For a language that compiles to C and claims no UB, the author should have a much better understanding of this.

k__• 5 years ago

People are always talking about the 10x engineer. Nobody imagined that someone would invent the 10x programming language.

ylluminate2• 5 years ago

Jobs' engineering team did it when compared with other libs at the time with NextStep. I know devs who claimed they were 3x (or more) more productive than their Windows and other Unix counterparts during that time. Rails arguably brought a similar revolution in the webdev world.

pokstad• 5 years ago

Only 128MB of RAM? What happens when you try to GC a huge repo? This must mean only the non-git requirements. Under the covers either libgit2 or the git executable are being used, and those will use many resources depending on the operation.

amedvednikov9• 5 years ago

Well this is for average repos, you can't expect to run a huge multi gigabyte repo on the most basic instance. I'll make it clear.

shrubble• 5 years ago

One thing that is positive is that writing an application in a new language is a 'rubber meets the road' situation where the language is being tested in the real world.

Garlef• 5 years ago

Any documentation on the CI/CD features?

CrazyPyroLinux• 5 years ago

I'm a big fan of GitBucket - It's like GitHub-in-a-WARfile https://gitbucket.github.io/

Cavallium2• 5 years ago

Looking at gitly repository it almost has no features, only a fraction of the basic git features are present, and they are barely working. Why is it being advertised here on HN as a fast alternative of github/gitlab?

P.S.: Probably the website crashed, it's giving me 502 bad gateway from cloudflare

amedvednikov9• 5 years ago

502 was fixed.

I'm actually surprised it handled a front page HN traffic spike on a $3 VPS.

That was a good test of V/vweb.

Cavallium2• 5 years ago

Now it's online

tiffanyh• 5 years ago

“ Works without JS”
“ Open source“
“ Minimum RAM needed is only 128 MB”

Curious to see how this compares to @drewdevault great Sourcehut.org alternative and to get his take on Gitly

For performance comparison, maybe Drew will add Gitly to https://forgeperf.org/

73572• 5 years ago

Can what this guy is doing be done without making mistakes? Is this thread not good publicity? Does crazy_horse not sound pathetic and amedvednikov comparatively heroic? Could we not dream a little, please?

maleldil• 5 years ago

amedvednikov only sounds heroic if this is first time you've heard about him or the criticisms towards him over the years.

To those who've heard of him before, he's just a conman.

Make no mistake, this is just one of his many "ambitious" projects that he creates for publicity and then abandons after a while. Again, this isn't the first time he does this. He promises the world and under-delivers.

terafo• 5 years ago

Can you give few examples?

73572• 5 years ago

I choose to give this guy the benefit of the doubt.

robinshen• 5 years ago

Checkout https://github.com/theonedev/onedev for a serious self-hosted git server, with CI/CD baked in.

From a user of OneDev:

> As an open source project, it is too good to be true

Gitly: A light and fast GitHub/GitLab alternative written in V lang (pre-alpha)

Related Stories