| 1 |
On 8/29/05, Ciaran McCreesh <ciaranm@g.o> wrote: |
| 2 |
> |
| 3 |
> |
| 4 |
> Note that such a number wouldn't be useful for comparing with, say, |
| 5 |
> other distributions or ports, because a) we can SLOT things, so we |
| 6 |
> don't need separate foo-1, foo-2 and foo-3 packages, and b) we don't |
| 7 |
> need to do a zillion foo, foo-python, foo-perl etc packages because of |
| 8 |
> USE flags. |
| 9 |
> |
| 10 |
|
| 11 |
Some wise person once said that "Its not the size of the tool its how well |
| 12 |
it works." I think that perhaps all this fuss over how many packages are in |
| 13 |
the repository is kind of pointless. The issue is really does it work for |
| 14 |
your use. Sure Fedora may have 6 billion packages but if it lacks the ONE |
| 15 |
packages that you need to do that job and its not easy to create a package |
| 16 |
for it then by golly Fedora isn't very useful. From personal experience I |
| 17 |
find it much easier to create ebuilds for non standard packages than it is |
| 18 |
to create RPMs plus the Portage tree has most of the packages I need by |
| 19 |
default (and as an added bonus I don't have to hunt down a thousand |
| 20 |
dependencies to install a single package). |
| 21 |
|
| 22 |
So instead of comparing raw numbers think about this... Does it work for |
| 23 |
you? In non-standard situations is it easy to get it to do what you want it |
| 24 |
to do? If the answer to both is yes I think that says far more about the |
| 25 |
distro than the size of its repository. Just my $0.02 |
| 26 |
|
| 27 |
-Mike |
| 28 |
|
| 29 |
-- |
| 30 |
________________________________ |
| 31 |
Michael E. Crute |
| 32 |
Software Developer |
| 33 |
SoftGroup Development Corporation |
| 34 |
|
| 35 |
Linux, because reboots are for installing hardware. |
| 36 |
"In a world without walls and fences, who needs windows and gates?" |
Archives