I have no gripe with gh specifically, but it's very disappointing that the first thing people do is flock to centralized services on top of the brilliant p2p scm that is git. You really don't need to look further to get a grasp of how things for the dweb will work out.
Also of note: github blocks indie search crawlers, contributing to search engine monoculture. Moreover, your user's clicks are captured by a company with an incentive to push for privacy invasion as a business model going forward. So, indeed, F/OSS projects using gh should definitely reconsider, if they haven't already done so when MS bought gh.
If I want to contribute to a project, it’s super helpful to know what the canonical location for this is (“canonical repository” vs “pull from some random persons repo”)
If I want to run a FOSS project, I might not be interested in the details of managing distribution and security of some server (one-click software rather than setting up some server myself)
I also .... kinda don’t care about account management (=> using something that people already use is nice)
Peer to peer isn’t actually interesting for almost any aspect of a project. The main advantage is availability, but it doesn’t help (and can hinder) other objectives of running a FOSS project.
I think that there’s a way to get most of the goodies (see GitLab... kinda), though it’s a hell of a lot more work than “just send email patches around!” I think sourcehut is getting there though
Centralized services on top of decentralized things are mostly used for visibility.
Pirate Bay for torrents, Napster, GitHub for git, ... They do fix a problem.
I actually think that Microsoft is the lesser evil in this case. They will add azure flows, but not enforce them. I believe they will also embrace other services ( not the embrace and extend way though. The Enterprise culture makes them support things for a very long term ( even codeplex still let's you download old archives)
They reinvented themselves with the mantra: developers.
Making them pissed off for using something else is not their mantra any more ( them = developers outside the dot Net/ windows ecosystem)
And yes, as each business, they also want profits. But the way to do it has changed.
Ps. Google code also let's you download the old repos ( just checked it)
Edit 2: the only thing I can think of is that the evolution of Windows forms didn't go as planned and got changed to "more web". As a result, WPF ( I think) is in maintenance mode and considered as feature complete untill it's fully integrated in. Net core ( which was a hard change because of differences in graphic rendering cross platform). The only thing that got deprecated here was Silverlight, which was a correct and hard decision.
The same thing is happening with blockchains. A good chunk of developers and companies are working on creating monopolies and walled gardens, instead of focusing on open market schemes and infrastructures. It's the old, known Zero to One approach. No one wants to be a player on a competitive market if there's a chance to avoid it.
I personally can't stand Gitlab's interface and general slowness.
sh.rt (which kind of became https://sourcehut.org) looks much better but it looks like you need to be logged in to see most of the content.
About abuse on Github, I only had issues with posting email addresses in public gists (even if they are public ones), and saw repo owners editing other people's comments, which should not be a thing.
About Github's source code, it's not open-source but you can host it with Github Enterprise and actually look at the source code (it's a very simple encryption of the source code as far as I remember).
Anyway, since Microsoft's acquisition of Github I started to worry of the general direction Github is going to but so far they haven't messed up too much IMHO.
Hello! Thanks for mentioning GitLab. I would like to ask if you could provide more details about what should be improved regarding the interface. Feel free to open an issue about it via https://gitlab.com/gitlab-org/gitlab-ce/issues.
We are heavily focusing on performance improvements - both in the product (so that features run better)[1] and in our GitLab.com infrastructure (so that GitLab.com runs more reliably)[2].
There was a time when I thought places like GitHub are the future of Open Science [1], which will replace paper articles by data+programs. The problem is now well-known: at any moment somebody will buy the whole place and will upgrade it, so to say.
[1] https://chorasimilarity.wordpress.com/2017/04/02/the-price-o...
Firstly, I don't think it is hypocritical to run a service that isn't FOSS, but that also supports FOSS. Secondly, nobody is under the illusion that Github is open source. Thirdly, Gitlab is only partly open source. Many of its features aren't.
> Github pull requests suck, why not just use email?
Maybe because email doesn't: a) Give you a convenient list of open pull requests, b) Give you a list of pull requests that other people can see at all c) Let you do nice line-by-line code reviews, d) Let you integrate CI tools easily, etc. etc. etc.
> You can't host malware on Github
Ok?
> I set up gitolite for repository access control, cgit for a simple Web frontend, and I plan to include an issue tracker that treats E-mail as first-class rather than forcing users to create accounts (possibly Bugzilla but I'm open to suggestions)
Riiiight. I'm sure everyone will love going back to the good old days of Bugzilla.
We are an open core company. We ship two editions - Community Edition and Enterprise Edition. CE is completely open source and licensed under MIT license. EE is proprietary, closed source code but we try to work in a way similar to GitLab CE: the issue tracker is publicly viewable and the EE license allows modifications.
In conclusion (TLDR), GitLab has an open core business model and ships both open and closed source software.
They already started to botch it up anyway, you can see the "copy-pasted-from-competitors-we-wanna-be" feature creep already arriving. That notification timeline lifted from the manyfacedbook, where they want you to look at the adds they will shove into it soon enough.
That maintenance lack for the open source users and the lack of new really relevant features.
Actually, GitLab started as an open source feature-by-feature copy of GitHub. Since incorporating they have actually introduced many original functionalities, including their own CI pipeline.
Additionally, there is nothing wrong with implementing a better version of what the competition offers.
GitLab.com was originally announced with free private repos. When you’re starting to program and aren't ready to share your code with the world yet, you don't have to have a paid account to keep it private.
Now, we're focusing on making a single application for the entire DevOps lifecycle that can replace a lot of other tools[1]. So instead of a version control system that lets people try out different integrations, GitLab provides an opinionated (yet flexible with key integrations and the option to opt out of anything you don't want) way to run the entire software development and deployment lifecycle.
Or, as Stavros Korokithakis phrased it: "My move to GitLab was basically 'Come for the free repos, stay for the rest of the amazing features.' I will not be moving off it, and my new repos will keep being on GitLab."
Agreed - could have been more precise.
Just had a hard time holding down that bile that builds up- once some clueless productmanager jams down useless obsticle-"features" in my workflow, to meet some delusional growth projection.
This stuff doesn’t matter ... it really boils down to the extreme FOSS ideological stance.
Fact is GitHub is the site that won this winner take all race. If you put your open source project anywhere else you do your project and it’s users a disservice. Everyone knows and is familiar with github and uses it every day ... taking your project elsewhere takes it out of this daily workflow.
And frankly who cares if GitHub itself is not open source... honestly it really doesn’t matter in any meaningful practical way, it matters only from an ideological standpoint.
I like it that the modern world of OSS isn’t controlled by the extreme ideology because these people tend to choose worse options to get the job done because that worse option meets an ideological position.
The commercial option tends to look like the better option at first, because companies invest a lot of money upfront to polish the product and buy market share by giving the product away.
Later, when the company is fully grown, customers are locked-in, and investors want to cash out, the commercial option often becomes the worse option.
This was a choice that kept me on board without panicking and jumping ship like many others. I'm pretty comfortable with github, and find its search engine options and overall site responsiveness (speed) to literally quash any chance of me joining the fleeing masses. The private repos was a very happy addition.
Out of curiosity, how does this lock-in manifest itself? I'm pretty sure all I'd have to do to move to a competitor is upload my local directories to that site. Is it the loss of revision history, or ??
> I like it that the modern world of OSS isn’t controlled by the extreme ideology because these people tend to choose worse options to get the job done because that worse option meets an ideological position.
I dislike it for the very same reason. A blind pragmaticism leads to a worse world for everybody. I would prefer if developers were more conscious (and yes, ideological) of their fundamental role in the future of general-purpose computing.
Also of note: github blocks indie search crawlers, contributing to search engine monoculture. Moreover, your user's clicks are captured by a company with an incentive to push for privacy invasion as a business model going forward. So, indeed, F/OSS projects using gh should definitely reconsider, if they haven't already done so when MS bought gh.