Richard,

> I'm not convinced I'd go for a daily build, or the testing distro, either, 
> but that's another issue.

I used daily 'testing' installers more often and there never was a problem with 
the installer or base system. If there would be a minor bug with a desktop app, 
later. then you can update or change app.

It is strategic recommendation. I was using stable installers in the past and 
always ended up with upgrade to testing, then unstable, because i wanted more 
recent features. For example, latest video codecs, or new website features 
(css3, html5, flash).  Especially big internet players never wait for Debian 
with their upgrades ... Then, upgrading from stable to testing can be quite 
cumbersome, since the distance can be huge, but from testing to unstable is not 
such a big deal.

Agreed, Ubuntu always is an option and the most popular. But over the last 
years, whenever i installed the latest Ubuntu (up to 11) for someone else (or 
did remote support), i found so many deficits and bugs, that i always ended up 
with installing better or less annoying apps, or just newer versions, and then 
the whole thing just sucks. And i finally switched back to plain Debian. It was 
rather surprising to me too.

But even inexperienced users have specific requirements (even when they got 
used to Windows or Ubuntu since long) and then you can adjust a plain Debian 
much easier. Like, resolving dependencies when upgrading certain packages (or 
all) to unstable.

Of course YMMV and there are good reasons for staying stable too. It's also 
true that even trivial problems can be mountain ranges for someone who no idea 
what's up.

> ie replace ia64 with amd64. ia64 is for Itanium machines, which is mostly 
> expensive servers.
oops...so 'amd' is for intel too ? I always forget it and confuse it....sorry!


-- 
To UNSUBSCRIBE, email to debian-laptop-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20120322021638.1c61d...@mirrors.kernel.org

Reply via email to