I think he maybe clicked at "fork button", just because he was playing with github .. and he did not search about how to delete his fork.

Well, `fork` is simply GitHub's on-site `clone`.

If you think about it, people clone repositories all the time - having a copy in case someone deletes the repository being one of the reasons.

In a non-GitHub workflow you'd normally get a `patch` emailed to you instead of a `pull request`. How many do people usually get? I suppose the ratio of GitHub's `pull requests` might be slightly higher than non-GitHub `patches` but not by a huge margin.

The upside of a `fork` as a "backup", is that one does need to `clone` any data locally.

Other than than that, there's not much of a difference me thinks.

I often github-fork repo's just to track them, when I find them interesting enough to eventually have a further look at them use them. Which also could just never happen. Did I mention I'd like 48 hours days?

This morning I was looking for alternatives to Trello, and managed to find a reference to Wunderlist, which is supposed to be located at https://github.com/6wunderkinder/wunderlist. Well go there and see. You get a 404.

Now go to http://archive.is/8WQQ or seek it up on ohloh and you will find a snapshot of that Github repository. Seems the six wonderkids decided to pull their application from the net and go commercial.

Many of the forks on Github made by the anti-socials were much older than the one on Ohloh and, hence, useless as backup. Searching for wunderlist on Github gives quite a few anti-social forks.

I just have a problem deciding who is being anti-social in this case: 6wunderkinder or the users forking a backup on Github.

I can understand (possibly) why he did this. If he wants to make a permanent reference to code (like if you're using hg subrepos for instance) it's possible he'd like more control over the source that he refers to. For all he knows you could decide to take your code down tomorrow and his reference to your http://github.com/xxx/xxx repo could get broken. You might not be about to do that, but he has no way to guarantee that.

I deliberately picked the ambiguous name "Robin" which can be male or female, but then spoiled it by writing it "Robyn" which is a typically female-spelling.

Interesting to see two "he" responses; although I can appreciate some gendered names don't translate to different countries/locations.

That said I suddenly see legitimate reasons for cloning that don't boil down to paranoia; so thank you all :)

I've seen that before and hoped they'd bring something new and exciting with a merge request. Sometimes that happens but other times, well nothing happens that I can see.

It's not that they're doing anything wrong, I was just hoping for some good stuff.

I have at least once forked a repository accidentially.

I forked another github project because I strongly disagreed with the direction they were taking the project. Although I am a developer, at the time I was completely 'committed' to other projects so could not do much about the direction they chose but I was not alone -- there was quite a few people who were worried about that particular commit. In the end they have continued to spend developer hours and the resulting change hasn't really done as much damage to the ecosystem as I figured it would but these 'empty' forks, at least in cases like mine can be interpreted as dissent from both skilled and unskilled developers who have different ideas of what the code should do, even if they aren't fully fleshed out yet.

Perhaps one day I'll go back to that fork and resync as many changesets from the 'main' branch as I can to my branch and make sure that the change I was trying to avoid gets removed. But until then it sits just as a placeholder for intent to do so, given skill, resources and time constraints I have to deal with.