Slashdot Mirror

← Back to Stories (view on slashdot.org)

GitLab Acquires Software Chat Startup Gitter, Will Open-Source the Code (venturebeat.com)

Posted by BeauHD on from the if-you-can't-beat-'em-join-'em dept.

According to VentureBeat, "GitLab, a startup that provides open source and premium source code repository software that people use to collaborate on software, is announcing today that it has acquired Gitter, a startup that provides chat rooms that are attached to repositories of code so that collaborators can exchange messages." From the report: GitLab won't bundle it in its community edition or its enterprise edition yet, but it will open-source the Gitter code for others to build on, GitLab cofounder and CEO Sid Sijbrandij told VentureBeat in an interview. What's happening now, though, is that as part of GitLab, Gitter is launching a new feature called Topics, where people will be able to ask and answer questions -- sort of like Stack Overflow. "Although Gitter is best in class with indexing things, it's still sometimes hard to find things," Sijbrandij said. "In this Q&A product, it's a lot easier to structure the Q&A. You're not dealing so much with a chronological timeline where people have different conversations that cross each other. There's a location for every piece of knowledge, and it can grow over time." That technology is already available in beta in Gitter rooms on GitHub, and it will become available on GitLab's Gitter pages over time, Sijbrandij said.

28 comments

  1. What programming language is used? by Anonymous Coward · · Score: 0

    What language is this software written in? Go? Rust? Erlang?

    1. Re: What programming language is used? by Anonymous Coward · · Score: 0

      It's written in C and designed to be compiled using the GNAA compiler collection (gcc).

  2. waste of money by Anonymous Coward · · Score: 0

    There's no money in open source. Nobody wants to pay for free software. Any company with a business model built around open source is inherently failing. This is basically one sinking ship taking over another sinking ship, and both are still destined to fail.

    1. Re: waste of money by Anonymous Coward · · Score: 0

      They have investor money to spend and it was either acquisitions, or hookers and blow. They took the high road this time.

    2. Re:waste of money by The+Raven · · Score: 2

      That's a silly comment. GitLab is buying Gitter so it can be incorporated into their own platform eventually. They're buying a well-made codebase to improve own product. They could have written it themselves, but they decided that their time to create an equal feature would have been longer or cost more that buying it.

      GitLab needs to compete on features with GitHub, who is winning the popularity contest by a wide margin.

      Also, "will not be adding it directly now" is not the same as never. I still think the long term goal is to improve GitLab using features borrowed from Gitter.

      --
      "I will trust Google to 'do no evil' until the founders no longer run it." Hello Alphabet.
    3. Re:waste of money by Anonymous Coward · · Score: 0

      Can I host GitHub on my own server? Can I use it when my Internet inevitably fails (just about guaranteed to happen out here in the woods)? As far as I am concerned GitLab is a totally different product for a different audience.

    4. Re:waste of money by Anonymous Coward · · Score: 0

      Where's the fucking APPS guy!?

      gitAPPS!

    5. Re:waste of money by Aighearach · · Score: 1

      The trick to understanding it is that in most sectors there is little money in software anyways, and most of the money is in support. This is true for the proprietary software, too.

      If you think open source companies are failing, you might want to check that one in a search engine. Because you're so embarrassingly wrong that I'm not going to spell it out.

  3. Can't wait for the Git fad to die out. by Anonymous Coward · · Score: 5, Insightful

    I can't wait until the Git fad finally dies out. I've worked with a number of teams that use Git, and they all use it like a centralized VCS, except it's more awkward to set up and use than a VCS like SVN or Perforce or even CVS is. Then they spend more time arguing about whether or not to rebase than they spend actually developing software.

    It's a real shame that Mercurial didn't win out. It's a superior DVCS in every way, except for not having as much mindless hype surrounding it. But I suppose in some ways that's one of its best features, too. It hasn't attracted all of the fools that Git has.

    1. Re:Can't wait for the Git fad to die out. by Anonymous Coward · · Score: 0

      I like the Subversion commit model better, each commit is sequentially numbered so if M > N then M happened after N. Period. And the numbers are relatively small.

      I learned in the early '90s (from Microsoft COM, etc) that UUIDs were an ugly hack that were people should never have to deal with directly, unless they had to put out a fire.

      One thing nice about Git is that it's usually pretty easy to switch to a different branch in the same sandbox. SVN makes you copy everything over; while that can be convenient, the copy is slow and takes a lot of disk space.

    2. Re:Can't wait for the Git fad to die out. by Anonymous Coward · · Score: 0

      Some systems, like MongoDB for instance, prepend a timestamp to the UUID so that they sort chronologically when string-sorted. It would probably be nice if Git did that. I imagine it wouldn't be too hard to prototype, but getting it finessed, accepted and updated in mainstream distros would take months.

    3. Re:Can't wait for the Git fad to die out. by Anonymous Coward · · Score: 0

      If you knew what you were talking about, you'd realize that even Torvolds says that Mercurial is "good". But that git was just easier in his personal workflow. (Watch the video he gave at google HQ on this)

      Of course git is used as a central repository, it _makes sense_ to do it that way... but only because humans don't think "distributively" but instead LIKE having a central repo & repo master. (plus the inherent chain of responsibility that gives the repo master in a business). But just because git isn't USED dist. as much as you would like, doesn't mean anything. It has the feature... but "normal people" don't want to run a gitlab server install (or something equivalent) on each of their machines... Its easier to use gitlab/github/bitbucket/etc... as a central distribution center for repo hits...

       

    4. Re:Can't wait for the Git fad to die out. by drinkypoo · · Score: 1

      Of course git is used as a central repository, it _makes sense_ to do it that way... but only because humans don't think "distributively" but instead LIKE having a central repo & repo master. (plus the inherent chain of responsibility that gives the repo master in a business).

      If humans have to be able to understand commits, and they do, then it makes sense to make commits work in a way that humans understand.

      But just because git isn't USED dist. as much as you would like, doesn't mean anything. It has the feature...

      If you're not using it, then git is the wrong tool, because it's harder to make sense of than the things that aren't distributed. So in those cases, those people absolutely are wrong, and git is absolutely a problem.

      --
      "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
    5. Re:Can't wait for the Git fad to die out. by Kergan · · Score: 1

      Using SVN day in day out is far too slow with non-trivial repos.

      (I hear git and hg can be slow too when used with very large repos. But, as I understood, this primarily is when first cloning a repo or during large fetches. In stark contrast with SVN, you don't end up with enough time to make coffee or walk your dog each time you commit.)

    6. Re:Can't wait for the Git fad to die out. by Anonymous Coward · · Score: 0

      Using SVN day in day out is far too slow with non-trivial repos.

      Funny. Lots of people say this -- but I work with very large SVN and GIT repositories. SVN seems to beat GIT by a wide margin. I think for an active code base where lots of people are digging into history -- SVN wins. By quite a bit. One other nice thing with SVN is that you can check out only the portions of a large codebase you need. Where as with GIT it's an all-or-nothing pull.

    7. Re:Can't wait for the Git fad to die out. by Anonymous Coward · · Score: 0

      One thing nice about Git is that it's usually pretty easy to switch to a different branch in the same sandbox. SVN makes you copy everything over; while that can be convenient, the copy is slow and takes a lot of disk space.

      It doesn't -- there is a command svn switch that can repoint/update your working copy. I, personally, like to have quite a few checked-out copies of a large repo checked-out and compiled. This way when I switch one of those to a branch I don't have to recompile everything either.

      I agree 100% -- SVN's revision model is far better than GIT's useless hashes. Most of git seems useless and poorly designed; the exact opposite of SVN.

    8. Re:Can't wait for the Git fad to die out. by afranke · · Score: 1

      I've worked with a number of teams that use Git, and they all use it like a centralized VCS

      The D in DVCS stands for distributed, not decentralized. The model is up to you, but even with a centralized one there are many benefits. Try bisecting with SVN.

      except it's more awkward to set up and use than a VCS like SVN or Perforce or even CVS is.

      Then you must be doing something wrong. I’ve had to handle both SVN and git repo sharing and git was a breeze compared to SVN. Not to mention starting a project on your own is as easy as `git init`.

      Then they spend more time arguing about whether or not to rebase than they spend actually developing software.

      Sure, the tool is to blame for people who like bikeshedding.

      It's a real shame that Mercurial didn't win out. It's a superior DVCS in every way, except for not having as much mindless hype surrounding it. But I suppose in some ways that's one of its best features, too. It hasn't attracted all of the fools that Git has.

      My first DVCS experience was with Mercurial, and I’m glad git won the race. The only features I missed when moving to git were `hg incoming` and `hg outgoing` and I kinda managed to implement them with aliases using fetch and log, so I’m happy now.

    9. Re:Can't wait for the Git fad to die out. by Anonymous Coward · · Score: 0

      Yeah, git is the problem. /s

    10. Re:Can't wait for the Git fad to die out. by Anonymous Coward · · Score: 0

      no, that would require actually trying to learn instead of hoping the new tool works like you except, bad habits included.

    11. Re:Can't wait for the Git fad to die out. by Aighearach · · Score: 2

      I think it is funny, too. When I used svn, I tried to use it in the ways that it was good at, and if you set up the directories right it is simple and painless even in complex settings. Actual code collisions shouldn't be frequent anyways, so people can like one or the other better for that but it is really a different problem.

      I switched to git for offline commits, which svn added but too late to prevent people switching. Once you switch all your practices to the ones that are best with git, then git becomes better. But if you were already using practices best for svn, it didn't start out better.

      What is obvious is that a lot of people want to do things their own way instead of the way that is a best practice using the tools that they're using. So then they'll prefer whichever software is least painful using their own weird system. But sharing code with them will be painful for others, and no version control system is going to help with that.

  4. Startup? by 0100010001010011 · · Score: 1
    • Launched 2011; 6 years ago
    • As of December 2016, the company has 150 team members and more than 1400 open source contributors.

    I think it's safe to say they've outgrown "startup".

    1. Re:Startup? by XXeR · · Score: 1

      I think it's safe to say they've outgrown "startup".

      Gitter launched in 2014 and was seeded in 2015...that's a startup in my book.

    2. Re:Startup? by 0100010001010011 · · Score: 1

      GitLab, a startup that provides open source

    3. Re:Startup? by XXeR · · Score: 2

      GitLab, a startup that provides open source

      Sorry, be specific then, it's in the title of the article for Gitter:

      GitLab Acquires Software Chat Startup Gitter, Will Open-Source the Code

  5. Git by Anonymous Coward · · Score: 0

    https://www.youtube.com/watch?v=HtyO4tmpPdk

  6. Very soon after by Anonymous Coward · · Score: 0

    last incident ....