Move from Github to Gitlab?

  • I am against any move until we see a negative impact for us. We got over 300 repos hosted on Github now, and Infrastructure is already now under pressure with keeping up with open work items. Moving now, not even when the deal is dry, without seeing any policy changes is overreacted IMHO. We want to finish OTA-4, and when we start moving to Gitlab now you can forget any release at end of June, or even longer.

    After OTA-4 we can review this again. But I strictly disagree on any fast reaction now.

  • @flohack I agree, of course OTA 4 is the top priority now.

  • @advocatux it isn't just "now".

    Even after 16.04 ships do we really want to spend a month moving everything instead of...

    • Porting to new phone(s)?
    • Getting anbox support released?
    • Getting Unity8 desktop working again?
    • Delivering other cool stuff?
    • Fixing bugs?
    • Promoting the project?

    Really, unless we realistically expect things to be better after the move this has to be a bad idea. (Getting 50% more work done afterwards would be a good argument - that's break-even in 3 months.)

  • This M$-github thingy is unfortunate, but we should keep OTA4 (and critical bugs!) at priority one. If there are going to be some pay-to-use $$, maybe then we should look some replacement for github..

  • For me the workflow with git and github ist totaly new. I try to work on this and want to contribute to the project. For me it's better to stay on github for now. Otherwise i need to learn how gitlab works

  • I agree with @alan_g , I think the project should be more important than the fact that MS bought a website UBports, lots of Linux and FOSS developers are using.
    Especially since we don't if that affects the project in any way, yet we know that the move would probably cost the project time.

  • I strongly dislike Microsoft and all their stand for - based on what they did in the past the company is not to be trusted no matter how well they may disguise their actions nowadays. But as some people realise there should be no rush on this - take the time. Just be ready in case Microsoft is about to introduce their "better" new services that will limit everything and cost money as a bonus.

  • Be like the "boy scouts" be prepared ,in case the move is needed at short notice.

  • Move to GitLab, UBports should make at least a firm commitment to move to GitLab in the next two to three months.
    Community will forgive if this delays other plans for UBports a bit. What they won't do is contribute to a project that doesn't make it clear they do not support Microsoft's agenda and wont stand for their embrace extend extinguish.
    Let us not forget that Unity failed to GNOME mostly because of Ubuntu CLA and similar stuff that didn't align with the foss philosophy in the minds of many Linux devs and users.

    This is more than just a PR move, this is about how we do this things in the future too. I say let us not do compromises in face of events such as this one. Make a firm statement that UBports does not do compromises and that UBports will move to other service even if Microsoft doesn't have an agenda, do not wait to see the negative impact, make it clear that UBports will be paranoid so that you don't need to be, so that you could have a peace of mind.

    A piece of software that is free and doesn't have any proprietary attachments because UBports is not Microsoft and wont stand for their agenda even if they buy GitHub and try to shove it down our throats by their corporate force...

    This move will most surely strengthen the bond and the trust between the project and it's community.

    @alan_g has some good points and I would agree with him if we were together in a company but sometimes, like in this case, the most economic and the most efficient thing isn't the right thing to do.

    [edit]: I feel like this is appropriate: click for the video

  • @flohack I fully agree with you, @alan_g, @advocatux @NeoTheThird and almost everyone here, right now is not the moment but the infrastructure should be there ready when the time comes in the future.
    Thanks for everything.

  • I fully agree with @aresminos I think a move to gitlab is more a statement for our community as well as for the opensource community in general.

    Also, like other said no rush is needed. A move after OTA4 will be OK I guess, but maybe an official announcement for when the move will be.

  • Even after 16.04 ships do we really want to spend a month moving everything instead of...

    @alan_g the answer to your question is pretty obvious if that is the only possible scenario, but I don't think those are the only parameters we should take into consideration.

    A realistic ETA for the move could be a good first step. I don't say you're wrong stating "one month" but I honestly don't know if it would take one week, one month, or perhaps even way more.

    On the other hand, I see one of the main developers (@mariogrip) showing an intense discomfort to the current platform, and I'd like to hear the rest of the UBports developers about how they feel about all of this. Because in a volunteer based project personal motivation is very important and that can also be affected when developers feel like they're betraying their principles, ideology, etc or just hate for whatever reason the tool they're using.

    I mean, developers feeling bad or uncomfortable using Github can affect the project too.

    The opinion of the community as a whole is important and has its weight, but the final decision has to be made by those doing the heavy lifting: developers & sysadmins.

  • Somehow this makes me upset... I think because it makes me feel that you can almost buy anything if you just have enough money...
    While I really don't like the thought of spending one month just for moving things from github to somewhere, I think on the other hand all the open source projects should take the opportunity to show them that they cannot buy everything...

  • @hummlbach said :

    all the open source projects should take the opportunity to show them that they cannot buy everything...


  • Personally speaking being attached to the great micro will feel wrong. But, there's time to plan, consider all the implications to people, projects, UBports in general and the time scale to move evertything smoothly to wherever (git lab looks the most obvious) oh and add in half the time again just to cover the inevitable issues (Do you use TSB !!).
    But if at all possible (which seems most likely) lets get OTA 4 out there first and have a stable base to move from.

  • To be clear of what was the among problems of Github:

    • Having operational costs higher than income (loss in the range of 60 mil $$ in 2016)
    • Having no CEO, and no one obviously wanting to have this job
    • Probably not getting more money from the banks

    So, in this situation, you got 3 choices: Cease to exist, or get a partner, or sell. Github is free for public code, but obviously not enough people wanted to pay for it. File bankruptcy is not an option, it would hurt the planet a lot, and you will be forever in the books as "the guys who destroyed FOSS", ignoring the fact that FOSS is not Github.

    Getting a partner as second option needs a strong leadership which was obviously not there. Unable to find a CEO for such an attractive company seems very odd to me. Whats wrong in their owners and board level, if they are looking for almost one year? So probably also partnering up was out of question.

    So sell it. Especially for that price. Of course money can buy anything on this planet, if it is just enough of it.

    The time of all-free services on the internet is over, and we should rather ask, what is it worth to have an independent provider of services? And what everyone would pay for it?

  • I'm in favor of moving but I think we should wait and see what the merge of github and Microsoft look like...

  • @flohack said:

    Of course money can buy anything on this planet

    Precisely not the free softwares; that is why it is important to be as consistent as possible in this matter.

  • Videolan words are so right (except for the verse on M$):"
    And the @github acquisition (potential) explains why we self-host all sourcecode at @videolan with git or with our @gitlab instance.
    @Microsoft has been a great citizen for open source lately, but it shows you can never know what will happen with a non-open-source SaaS." (

  • I suggest waiting a few months (preferably at least six months) to gauge whether GitHub will change for the worse in any way. If you notice changes for the worse, then plan an orderly exit at that time. But if you don’t notice any changes for the worse, no need to divert precious time from development.

Log in to reply