On Wed, Feb 05, 2003 at 11:28:18AM -0800, Seth Cohn wrote:
> > > Every user of testing knows that he must read
> > > debian-security-announce and if needed install fixes from unstable
> > > since it can take an arbitrary amount of time until security fixes
> > > from unstable enter testing (most likely none of fixes from the
> > last
> > > 70 security advisories is in testing).
> 
> Joseph and I were just talking about this yesterday.
> Stuff only makes testing once's it's hit a stability of a few weeks.
> So testing is the least secure because of that.  Unstable is second,
> because it needs to be compiled for many architechtures.
> Stable is usually good, and adding security.debian.org's stable updates
> often is the answer, though it won't always work for all packages if
> you run testing.

Backwards..

Stable requires 13 machines recompiling the security fix (including an old
Amiga, Atari, or Mac using a Motorola 680x0 chip - hope the security fix
isn't in XFree in that case..)

Testing requires the 13 machine recompile and to be stable for a couple
weeks.

Unstable gets fixes as soon as the mirrors are synched, which is about
once a day.


> I used to subscribe to all of the mailing lists, but I tried to get a
> life.  Not that it helped.

It can be confirmed that Seth has no life.  ;)


> Then install stable.  If you choose cutting edge (or near cutting edge)
> over security & stability, of course it takes a bit more 'bother'.

You get more security at the cost of potential stability.

-- 
Joseph Carter <[EMAIL PROTECTED]>             You're entitled to my opinion
 
<Teknix> our local telco has admitted that someone "backed into a
         button on a switch" and took the entire ATM network down
<netgod> hopefully now routers are designed better, so the "network
         off" swtich is on the back

Attachment: msg12838/pgp00000.pgp
Description: PGP signature

Reply via email to