RE: Future CVS Development

[Greg A. Woods]

[ On Tuesday, June 19, 2001 at 11:48:46 (-0400), Noel L Yap wrote: ]
> Subject: RE: Future CVS Development
>
> CVS only requires files to be mergable.  This has a different meaning from
> requiring files to be non-binary.
> 
> One thing that may be done is to add a hook for pluggable diffing/merging
> engines.

Maybe.  This trick is actually necessary at the RCS level....

It's also not obvious how to manage file content type identification (I
mean initial identification, not the tracking of such identification --
that's pretty easy if you steal ideas from MIME), and of course the
whole concept of content typing has many complex implications for any
tool that tries to track and merge changes (which might include type
changes).

I'm not entirely sure what the benefit is either.  It seems much more
productive to group files by type instead of treating them all the
same.  Even when it's beneficial to mix file types (eg. in e-mail), it's
not always easy or universally done.

For example when you're building a program with a graphical user
interface, do you really need to branch and merge the images
representing graphical components?  Do the changes in the images even
relate in any way at all to changes in the code?  Often they do not.
Why not manage the changes to images in a manner more suitable to the
development processes for images?  It's very trivial to build a higher
level integration layer that can put together the appropriate versions
of source code (or the products created from the sources) with the
appropriate versions of images.

-- 
                                                        Greg A. Woods

+1 416 218-0098      VE3TCP      <address@hidden>     <address@hidden>
Planix, Inc. <address@hidden>;   Secrets of the Weird <address@hidden>