cute post evaluating HG

[Giorgos Keramidas]

On 2007-04-15 10:14, TK Soh <teekaysoh at gmail.com> wrote:
>On 4/15/07, Matthieu Moy <Matthieu.Moy at imag.fr> wrote:
>>From what I've read, they did a lot of experiments, but the conclusion
>>boils down to :
>>
>> * Windows portability => only bzr and hg match.
>
> Although hg (can't speak for bzr) run on Windows, it's still not
> appearing to the Windozers largely due to the lack of GUI.  These
> days, we can't expect most [average] developers on win32 to be
> command-line savvy.

Right on the spot again, as usual, Thomas :)

I don't know enough about the internals of Windows, but I have been
tempted several times to start with TortoiseSVN as a basis, and make
something like TortoiseHG.  Integration with the Windows GUI shell,
similar to that of TortoiseSVN is going to make Mercurial extremely
appealing to Windows users.

Right now, with the rich feature set of Mercurial, and the beautiful
opportunities it provides for scaling much better than Perforce for a
distributed environment (which we are using in the place I work), there
is really only *one* major obstacle which blocks me from filing an
official request to switch from Perforce to Hg: the lack of a pretty
GUI interface for Windows using people.

For a UNIX user, it may sound silly, but having a nice UI which works
seamlessly with the Windows GUI shell, is *very* important for the
people who like working with Windows.  A typical example of this are
the people at my workplace who keep a shared Perforce workspace between
Windows and Solaris systems, because they like the GUI of Araxis Merge
a lot more than Emacs ediff-mode.

Having something like TortoiseHG, with configurable GUI options for the
stuff that usually goes into .hg/hgrc and the global .hgrc would be a
killer feature for Mercurial on Win32 :)

- Giorgos