1 |
Hello, konsolebox. |
2 |
|
3 |
On Sat, Nov 29, 2014 at 11:18:49PM +0800, konsolebox wrote: |
4 |
> On Sat, Nov 29, 2014 at 10:28 PM, Alan Mackenzie <acm@×××.de> wrote: |
5 |
> > Hello, everybody. |
6 |
|
7 |
> Good day. |
8 |
|
9 |
> > instead of conceptualising a "branch" (as you would do with Mercurial, |
10 |
> > Bazaar, Subversion, or even CVS), you need to think about "commits |
11 |
> > reachable from a certain head (excluding commits reachable from some |
12 |
> > other head)". |
13 |
|
14 |
> I actually see that as a more flexible approach. git is designed to be |
15 |
> distributed and that's what everyone loves about it. |
16 |
|
17 |
We're in violent agreement, it seems. git is very flexible, just like |
18 |
programming in assembler is. git is certainly a distributed system, |
19 |
just like Mercurial, Bazaar, etc., but seems to be the only one of its |
20 |
kind that imposes this degree of flexibility on its users. Hence the |
21 |
multi-hundred line man pages for each of git's 155 sub-commands. |
22 |
Mercurial has a mere 50 sub-commands (plus, to be fair, a few one's |
23 |
going to need which are classified as extensions), and a single, very |
24 |
readable, man page ~8000 lines long. |
25 |
|
26 |
> For everything: |
27 |
|
28 |
> http://stackoverflow.com/questions/802573/difference-between-git-and-cvs |
29 |
> http://eclipsesource.com/blogs/2011/06/09/git-lessons-learned/ |
30 |
|
31 |
> Cheers, |
32 |
> konsolebox |
33 |
|
34 |
-- |
35 |
Alan Mackenzie (Nuremberg, Germany). |