Re: Food for thought - SECURITY (design flaw?)

[Nicolás Lichtmaier <nick@debian.org>]

> My guess from the above, is that 0.4.6-3 was uploaded to unstable, but
> before it could get through to testing, 0.4.6-4 has
> uploaded. (probably 2 days latter).

 But there is another problem with testing.  When an important package
(xlibs,libc6) changes incompatibly (eg. bumping up the number in the shlibs
file), all the packages that depend on that package get paralized. All those
packages freeze in a random snapshot that won't change until the important
package gets included in testing (2 weeks, or more if there are problem
building in other archs). Suddenly all those packages are as unstable as sid
packages, but worst, because they aren't updated.