On Thursday, April 10, 2014 8:41:32 PM UTC+2, rusi wrote:
> I am interested in a rather old program -- first version was written more
> than 40 years ago -- in the Algol era!
> Its gone through various IBM/CDC machines until it ran on Irix in the 90s
> and briefly Windows until .Net 2003. After that its not exactly runnable.
> As may be evident the original programmer is elderly and while spunky, not
> exactly upto dealing with modern technology. He has shown interest in
> working with some of us who want his codebase modernized.
> So far thats not really relevant to git-usage.
> The issue that is a bit non-standard from git pov is that there is no
> clear starting tarball because some of the stuff is so old. So while we
> work on cleaning up and getting compiles through he is scrounging around
> for missing files (code and data). So should I handle this with git tools
> like branches, tags etc or just maybe keep a directory called
> original-stuff and put things there as they are found/provided?
It's very hard to advise on this without seeing exactly what kind of stuff
he's outputting in what order. I would just go with folders, which I then
start reverse-disecting into actual commits in a repository, mimicking what
he would have done if he had been using Git at the time. Over time, I would
tweak and improve history using interactive rebase, alternatively
filter-branch for larger operations.
> The other (minor) issue is whether to use bitbucket or github. I myself
> prefer bitbucket but would be interested to know others' findings on pros
> and cons.
I think feature-wise they are very similar. If you're making an open source
project of it, GitHub is definitely the way to go. If you want private
access, you have to compare the pricing models.
You received this message because you are subscribed to the Google Groups "Git
for human beings" group.
To unsubscribe from this group and stop receiving emails from it, send an email
For more options, visit https://groups.google.com/d/optout.