A year ago, I taught a course in which students did team projects that they submitted to me through Bitbucket. Changes to their Bitbucket repositories still show up in my Bitbucket dashboard.

One student just modified the project report he submitted with two classmates, to remove their names and leave only his own. Not sure what to make of this...

@ff524 Do nothing for this, but for the future I'd advise against using a non-institutional, possibly public, repository to which students maintain access for projects that should be graded and that can be possibly part of an official record.

@MassimoOrtolano it's not public unless students make it public. (Which many students do, because they find it useful for job searches.)

My general thoughts on this are here:

@ff524 The issue, I think, is that in this way there is no clear cut between what had been submitted for the exam and what's happened after it.

@MassimoOrtolano I'm not sure what you mean

My main concern with this is that it looks like the student is misrepresenting to potential employers that he did the project by himself. But I'd have the same concern if I hadn't used git in class at all, and noticed that he uploaded his project without his collaborators names afterward.

@ff524 You cannot follow what your students will do with their material for the rest of their life. That's why I said there should be a clear distinction between what they delivered for the exam and what they publish afterward. If the repository used for the exam simply becomes public, readers might think that that was the work delivered and accepted for the exam by a single person.

@MassimoOrtolano Readers really can't tell the difference. There is no indication on the repository that it is used to submit work, it looks exactly the same as it does when students upload the whole thing after the class.

(and also the same as it does when students use git because they like it, not because I ask them to)

@ff524 In that case, there's really nothing you can do to prevent that behaviour, because those students would do the same whatever the mean used to deliver their work. My idea is that there should be a track of the reports that you had graded, in case of appeal or whatever, but you cannot really follow what happens afterward.

@MassimoOrtolano oh, I keep an offline copy of all the reports that I graded. That's not really the issue.

I'm just wondering whether there's anything I can say pre-emptively next time I teach a course with a team project, to discourage students from trying this kind of thing.

Those who do that would not be discouraged whatever you can say ;-)

At most, they'd use another repository for going public.

I was thinking more along the lines of explaining to them that employers like to see that they're able to work in a team, which is part of the reason we do group projects

Make them think it's not in their best interest to claim something was an individual project when it wasn't

And git preserves individual contributions, so they can show exactly what part they did while still showing that they can work as a team

That is certainly a sensible thing to say, but my feeling is that such kind of behaviour is not only driven by the desire of appearing better before a possible employer, but also because of frictions within the group, egoistic personality etc. And they would do it anyway. But take my comments with a grain of salt: sometimes I'm slightly pessimistic when it comes to human behaviour ;-)