Huge spike in GitLab imports after Microsoft reportedly acquires GitHub


Ah yes, I'm pretty sure the private repo part is what interest microsoft. And integrating github in the VSStudio suite is certainly something they have in mind. And they may probably do the same for VSCode. Even for the free part probably. The objective beeing converting as much company as possible to the VSStudio/github ecosystem. And make more benefit with it. I'm pretty sure MS bought github for profit, not to help the opensource world. But that doesn't mean the opensource part of github is doomed, that is just my point.
 
Any one self hosting?

I have been self hosting my own repos from the start, so now I don't have to move anything :)

If anyone else is interested in self hosting there own, try Gogs, unlike git-labs it has the github style interface!
https://gogs.io/
 
There was a comment on a Reg Article about staying away from the "YCombinator mafia"... citing GitHub, but interesting GitLab is also listed as a YCombinator startup too.

When I realised you didn't actually need server-side software to host git repos, I basically just started making repos on one of my home computers, and just clone across to my other machines.
 

yeah and even worse, they are migrating from azure to gcloud services... not really better if you ask me. Our company also can't use gitlabs hosting for our stuff, as gitlab apparently is not GDPR compliant... we are probably going to host our own git repos and leave the wiki for our framework on github, so we don't lose the community.

https://venturebeat.com/2018/04/06/why-and-how-gitlab-abandoned-microsoft-azure-for-google-cloud/
 
I did set up a Raspberry home server for fun during the last weeks.
I have gitea installed on it as well now.

All I need now is a second or third machine on other places for redundancy/availability.
I can not trust my internet connection to be reliable nor do I trust one single machine.
 
that sounds interesting, I'll have a look at running my own instance on a pi at home for my private stuff. I tried running gitlab on my webserver a while ago, but it was way too heavy and the server became unresponsive (I probably misconfigured gitlab horribly in it's defense though. that might have been a big factor).
 
It's really simple to do.
Download, unzip and run.
Start it and open the web setup.
That's it.
 
It depends what you're trying to do really. If you just want people to be able to clone your repo then you can just splat it onto a webserver, and write the relevant instrcutions to download. If you want people to be able to browse it before downloading, you'll need something server side I think. Where github really wins is in the community, where anyone logged in can be notified of any issues raised changes made they choose of, and add their own comments, tests or patches under the same login. That last bit there's no real alternative to yet that I'm aware of - I don't yet have a login for any of these other development communities, for example.

For instance, I get mailed whenever someone raises an issue on the FLIF project, which really gives me an insight as to who's using it and hopefully where it's going to pop up in released projects.

But not many projects are like that, and a lot just sit there. I'm thinking of pulling my code and just replacing it all with a link to wherever I finally put it.
 
IMO Microsoft is certainly not interested in Github's business - even if they were profitable. With their $150M in annual revenue it would take a loooong time for that $7.5G investment to pay of. And how much could they grow? They are already number one!

No - the real value to Microsoft is the user base of 25M or so developers on there - developers who's mindshare MS desperately wants back.

To me this mirrors the Whatsapp acquisition by Facebook - again a company bought at 50 times their annual revenue. Whatsapp *was* already profitable with their subscription model - but Facebook did not even want that profit - they made the service free to use! Again, the real value was access to the user base (500M at the time, and growing fast - in contrast to FB).
 
IMO Microsoft is certainly not interested in Github's business - even if they were profitable. With their $150M in annual revenue it would take a loooong time for that $7.5G investment to pay of. And how much could they grow? They are already number one!

No - the real value to Microsoft is the user base of 25M or so developers on there - developers who's mindshare MS desperately wants back.

To me this mirrors the Whatsapp acquisition by Facebook - again a company bought at 50 times their annual revenue. Whatsapp *was* already profitable with their subscription model - but Facebook did not even want that profit - they made the service free to use! Again, the real value was access to the user base (500M at the time, and growing fast - in contrast to FB).

Insightful post.

More of 'the user is the product and the customer is the advertiser' model.
 
and to all people saying microsuck has changed since ballmer left: no they haven't, in 2015 they tried to force the UK gov't not to switch to open source software and threatened they would leave the UK with their RnD department if they did switch to open source. Luckily they did not give in to MS bullying:
https://www.theinquirer.net/inquire...er-government-switch-to-open-source-standards

or this list of shit right here:
http://www.catb.org/esr/halloween/

or this recent bullying of a US PC refurbisher:
https://www.theverge.com/2018/4/27/...-microsoft-eric-lundgren-prison-restore-discs

I refuse to believe that they have suddenly pivoted out of altruistic motifs. They are still the same bunch of assholes, it's just become more profitable for them to support open source than actively combating it. Also let's not forget, that we are talking about open source here and not about FOSS, which they will likely never ever support, their whole business is set up around making and keeping software non-free, non-libre and non-transparent.
 
I'm self-hosting a GitLab Community Instance since 2014 and I'm pretty happy with it.
The features grew over the year and it is running rockstable.

It's a bit of work to get it up and running. And it needs 1-2h per month for maintenance (security updates etc.)

https://about.gitlab.com/installation/
 
Sorry to respond over three weeks later, but I have two relevant things nobody even touched on:

First, the reason I'm researching bi-directional sync between GitHub and GitLab is that Microsoft already locked me out of one of my previous accounts. Specifically, I can't access my old Hotmail account because I have no SMS number to add to it.

If I set up bi-directional sync between GitHub and GitLab, I can continue to use GitHub-specific stuff if necessary but, if I get locked out by Microsoft for lack of a smartphone, I can push an update to README.md or README.rst on GitLab saying that the GitHub side is a deprecated mirror and have it added to the GitHub repo despite my being locked out from direct modification.

Any one self hosting?

I have been self hosting my own repos from the start, so now I don't have to move anything :)

If anyone else is interested in self hosting there own, try Gogs, unlike git-labs it has the github style interface!
https://gogs.io/

Self-hosting is generally a bad idea. You have no idea how many times I've been frustrated to find that old Pascal resources I want to for writing utilities for my retro-hobby PC have fallen off the web because the site were they were hosted is gone and the Wayback Machine hadn't gotten around to archiving them. Git as opposed to bog-standard files on an HTTP site will just make that loss of history worse. (If the Wayback Machine doesn't archive something as significant to the late 90s and early 2000s as its FTP sites, I seriously doubt it's going to archive git repositories.)

Likewise, according to the DejaNews archive Google Groups began with, Plug-ins for Program Manager by Plannet Crafters got freeware'd before they went out of business, but it wasn't Archive'd before the site went away, so all I have access to are the shareware versions from various shovelware CDs.

Sure, my blog is self-hosted, but I've been prioritizing preventing URL breakage, so I'm running the same WordPress configuration I originally chose back in 2004. When I find time to take another shot at reverse-engineering the index format for Sphinx's JavaScript-based site search (to avoid adding a dependency on Google CSE) and write the tooling to verify 0% link breakage and catch incompatible syntax extensions in posts, I want to migrate from WordPress to a static site engine to avoid having to babysit security updates. In doing so, I'd originally planned to also switch away from self-hosting by moving it to GitHub Pages.
 
Last edited:
old Pascal resources
I used sintelnet a lot (had a CD of it's data, but now it's unreadable) and the resources are gone too.
*sintel used to host zip files of sourcecode for various languages. Nice things like quick 3D rotation, libraries for handling reading/writing to COM1/2 ports, mathematical algorithms, keyboard handling (read those weird keys like play, pause) in all type of languages like assembler, C, Pascal, etc.
 
I used sintelnet a lot (had a CD of it's data, but now it's unreadable) and the resources are gone too.
*sintel used to host zip files of sourcecode for various languages. Nice things like quick 3D rotation, libraries for handling reading/writing to COM1/2 ports, mathematical algorithms, keyboard handling (read those weird keys like play, pause) in all type of languages like assembler, C, Pascal, etc.

I'm lucky in that I've never had a professionally-made CD go unreadable on me and all of my CD-Rs got migrated to DVD+R pretty early on. (Plus, I've always been pretty good about researching media before I bought it, aside from my earliest purchases being DVD-R because I didn't know it was technically inferior to DVD+R and was somewhat offended by the conflation of the hyphen and minus sign that DVD+R represented.)

As for Simtel.net, search the Internet Archive. People have uploaded ISOs of tons of those CDs to their catalogue, but they don't show up via Google.

https://archive.org/search.php?query=simtel

That said, I have been working to archive surviving FTP mirrors that I run across, just in case they hold something that never made it to those CDs. (And apparently I'm not the only one. The Archive community also has people who mirror FTP sites and upload them.)


https://archive.org/details/cd-roms (lists many more specific lists of shareware/simtel/magazine/etc. CDs)
https://archive.org/details/ftpsites
 
Last edited:
@sskolow: Ah... siMtel... thanks! Wow... lots of CD's!
 
Also, on the topic of dying media, one piece of advice: EA Kids apparently used garbage stock for their floppies, because my Scooter's Magic Castle and Peter Pan: A Story Painting Adventure floppies are in far worse shape than anything else in my collection. If you have EA Kids floppies you haven't backed up, point GNU ddrescue at them ASAP.

(And, if you've got the rarer pack-in CD versions, back them up and treasure them.)
 
Last edited:
Back
Top