[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: [hertzog@debian.org: Re: Woody retrospective and Sarge introspective]

* Raphael Hertzog 

| > it seems like you want to use unstable like experimental is supposed
| > to be used.  if you don't want to propagate to testing, upload to
| > experimental.
| I was sure that someone was going to respond that. :-)
| You're missing several points :
| - unstable will always be the default target for uploads, and things
|   will continue to go in unstable making harder for any package to go
|   from unstable to testing (because of a dependent packages that is
|   not stabilizing)

yes, because we think that most of our packages are suitable for
release.  If you think it is unsuitable, then don't upload to
unstable.  Upload to experimental.

| - experimental is not autobuilded

that can be changed just as easy as adding candidate.

| One of the biggest advantage of "testing-proposed-updates" is that it is
| built against "testing". That means that your package will go in
| testing if the delay is passed and if you had no major bugs. You won't
| be kept away of testing because of a nightmare with the dependencies.

having t-p-u is not an argument against using experimental more.
t-p-u is supposed to be used like proposed-updates -- there is
something very wrong with the package in testing (security hole,
doesn't work for some reason, etc), not as a genereal «let's skip this
unstable thing».

| Also you keep the control of what you send to our users.

uhm, yes.  You send stuff to end users (that is, what we release) if
you upload to unstable.  If you want to say «hey, this works (more or
less) for me, but it might cause anything to happen if you try it, but
feedback appreciated», then use experimental.

Tollef Fog Heen                                                        ,''`.
UNIX is user friendly, it's just picky about who its friends are      : :' :
                                                                      `. `' 

Reply to: