Re: [rdiff-backup-users] Plans?

[Edward Ned Harvey (rdiff-backup)]

> From: Kevin Fenzi [mailto:address@hidden
>  
> I'd be willing to try and help out though... perhaps I could assist
> with bug triage? Many of the bugs I get on the fedora side are just
> user error or issues with filesystems, etc and can be closed out once
> people fix the underlying issue. Looks like there's 69 bugs in
> savannah. I don't know if I need any privs there, but would be happy to
> try and go close obvious ones and ask for info on others, etc...

That sounds extremely helpful.  For me, time is of course, also the issue.  
(And I probably shouldn't be writing this at work.)   ;-)  But I'm self 
employed, so I'm not at risk of punishing myself.  I'm just not spending my 
time billing somebody right now, when I should be.  But I like doing this, so 
here I am.    ;-)   Beats doing facebook or tweeting at work.  Fundamentally 
the same thing.  Writing on the internet.

If you wouldn't mind, doing exactly as you say - seeing what obvious low 
hanging fruit exists, and trying to get more info about that, it would 
definitely be time well spent.

Also, I don't know if you'll need permissions either.  So simply getting the 
answer to THAT question will already represent forward momentum.  ;-)  I now 
have obtained nearly all the permissions on this project, so I think I should 
be able to grant that.


> Another unrelated thing... currently source is in svn. Git is the new
> hotness, perhaps it would be worth seeing how hard it would be to
> migrate?

You're right.  Until a couple weeks ago, it  was under CVS.  And I shyly with a 
slight little gesture of embarassment, converted to svn intentionally instead 
of git.  From a technical standpoint, I do personally think git will be better 
than svn for this project, but my motiviation for selecting svn was thus:

I mentioned I do a lot of IT.  I've deployed many svn and git deployments for 
various companies over the last several years.  Most of the time, you can't 
have a balanced conversation about it; Linus started it, and there exists 
nowadays, a culture of git-over-svn elitism, which is sometimes accurate and 
sometimes not.  The real truth is, each is a tool (neither Linus nor the 
Internet at large are always fair or well-balanced), each has different 
characteristics, and each tool is better for some situations.  In this case, as 
an OSS software project, git is the *ideal* solution.  But we don't have any 
development effort taking place, and as I said, I've deployed and supported a 
lot of svn and git.  Even in software groups full of pro software developers, I 
see this trend over and over:  The learning curve for git is much longer.  If 
you can use git, you can definitely use svn (although you might poo-poo it).  
Svn is way simpler to setup and understand.  So the reason I chose svn for this 
project is to not-inhibit uptake of new developers.

I figured, being forced to learn CVS would legitimately be an obstacle to 
acquiring new developers.  I figured svn will not be an obstacle.  I figured 
git is a double-edged sword.  As you said, it's "the new hotness," (or "hot 
mess?")  ;-)  but I don't believe running git will *attract* new developers.  
("Hey, did you hear, rdiff-backup is being developed on git.  Sounds like a 
good reason to join them...")    heheheh  ;-)      #join-rdiff-backup-on-git

There are still a bunch of people out there who know svn because of work and 
haven't surmounted the obstacle of learning git...  I chose svn to be 
conservative and avoid creating any obstacles for new developers.

Anybody who's in love with git is probably already using git2svn for other 
projects ...