Moving to D

[David Nadlinger]

On 1/6/11 8:19 PM, "Jérôme M. Berger" wrote:
> 	Here's a comparison. Although I am partial to Mercurial, I have
> tried to be fair.

Jérôme, I'm usually not the one arguing ad hominem, but are you sure 
that you really tried to be fair? If you want to make subjective 
statements about Mercurial, that you personally like it better because 
of this and that reason, that's fine, but please don't try to make it 
look like an objective comparison. A fair part of the arguments you made 
in the linked post are objectively wrong, which is understandable if 
you're mainly a Mercurial user – but please don't make it look like you 
had done more in-depth research regarding both to other people…

For example, you dwelt on being able to »hg pull help« being an 
advantage over Git – where the equivalent command reads »git pull 
--help«. Are you serious?! By the way, at least for Mercurial 1.6, you 
need to pass »--help« as a »proper« argument using two dashes as well, 
your command does not work (anymore).

> 	An extra point I did not raise at the time: Git is deliberately
> engineered to be as different from CVS/SVN as possible (quoting
> Wikipedia: "Take CVS as an example of what /not/ to do; if in doubt,
> make the exact opposite decision").

You missed the »… quote Torvalds, speaking somewhat tongue-in-cheek« 
part – in the talk the quote is from, Linus Torvalds was making the 
point that centralized SCMs just can't keep up with distributed 
concepts, and as you probably know, he really likes to polarize. In the 
same talk, he also mentioned Mercurial being very similar to Git – does 
that make it an unfavorable switch as well in your eyes? I hope not…

David