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

Re: An alarming trend (no it's not flaimbait.)

* Pierfrancesco Caci (ik5pvx@penny.ik5pvx.ampr.org) wrote:
> Nice bait.... I'll bite, but if you want to read it you'll have to
> subscribe... It's not fair to throw the rock and hide the hand 
> 1) learn how to properly format a mail message (i.e. fold at 75th
>    column)

Quit pickin at the measly stuff and pay attention to the content of
his words. Laying the bear trap here only gets you laughs from the
other hunters.

> 2) learn how to package a deb and adopt whichever package you think
>    you're better at maintaining than the original maintainer

Pointing out a failure in a system doesn't mean one has the ability
to do what you are asking. It simply means he found a failure. In
this instance, his becoming a maintainer does nothing to solve the
problem he's point at other than for that single package. Pushing
someone off into this section only further proves the point that
debian's starting to potentially fall apart since you completely
prove that you either failed to hear or desired not to hear what
his content.

> 3) if the package is dead upstream, fork it and maintain it
>    yourself. Most free software licences allow it.

Please explain how him maintaining it helps the other packages in
trouble as well. Or are you trying to suggest that the package he
spoke of is the only one of it's kind in trouble or that no other
packages in debian suffer from the fate he describes?

> Have a nice (redhat|mandrake|windowsXP) day

Not even worth more than this snicker.

OK, I'm getting in on this one, regardless of opinions. Sorry, but I
have to agree with Brian here. 

OK, my little history to show what grounds I make my stand on.

I've used linux for quite a few years now. I started with SLS Linux
kern ver 1.0.8), was there at slackware's inception, ran Peanut,
Stampede, Red Hat, Mandrake, and of course Debian.

I've worked in the linux industry for Red Hat, Ensim Corp, and a few
others. I've developed LPIC-II certification tests as a core member of LPI. 

"Why are you listing all that crap bub?" Probably what a few of you
are asking. Only to show my experience with different distros,
linux, and where I feel I gain my credence for my vote for Brian's

Debian is a solid distro to me. It's got heart, strength of
charactor both in it's member software, and it's member users and
developers. It's withstood 99% of the "Let's add every feature we
can lay our hands on cause that'll show we know what we're doing!"
crowd. It's solidly built, loved, and protected over by a loyal
group of users. This is more than I can say for the majority of the
distributions out there. 

yeah I give other distribs a hard time just like everyone else. it's
fun and part of the game. BUT, there are some real things that
happen to real distros when it's members don't speak of what they
see wrong and *developers* __listen__. 

Folks, our user base (non official developers and general users
alike) deserve to be listened to when they say something.

At least Brian took the time to list things out with well thought
out and deliberately worded and experience backed points. Many of
the "complaints" that a distrib gets are from those that aren't
really for the distro, it's just something they use, they want some
help, they find it's not easy without using the docs, and just
start bad mouthing it to cover their own lack of capability.
The real users have something substantial to their words.

Brian does. I've got to admit up front that I've not been a user of
debian as a primary distribution. And for whatever the comments
made, yes, I've been using Red Hat or Slackware as my primary. I do
run multiple distributions at any given time. There are many that do
because it makes it easier to see where users in general are. 

I have been running Debian since Slink. I've left it, come back,
left it, ran it solid for a bit, then left it again, now I'm back.

Why? Simply because of issues with Debian, be it the installer of
old, the lack of certain support, all different reasons. But one
thing is for sure. I've been able to follow quite a bit of the
lifeline of Debian. I too have watched as packages that Debian used
to keep up to date are now getting moldy. I've watched as developers
have gotten the attitude that the user is here to serve their
programming careers or their kudos meter, rather than the developers
realizing that's who they develop for besides themselves. (Face it
you wouldn't be developing for "Debian" as an official developer if
you didn't believe that it's a community thing and community is made
up of more than just the developers.)

Monitoring all the packages that belong in the Debian distribution
is a mighty tough thing. I'm sure even Brian will grant that. The
point of this excersise is the realization that the reason we have
maintainers in the first place is to make sure that stuff like this
*doesn't* happen to debian. If each maintainer is watching his or
her upstream, updates with their source when it's released, and if
the upstream is *not* providing the updates like they should, either
announce to the BTS that the source is cold, or attempt to
request to the upstream that they be named as the new maintainer if
the upstream wants to back out.

Admittedly this is in a perfect world that this works flawlessly,
but the amount of cruft building in Debian shows that we must be
living in a dying world because so many complaints about the bugs
and the moldiness level of the distribution are out there.

yes, developers whether you want to hear it or not, debian is
getting many complaints. And there are a lot fo silent ones as well.
I can tell you of at least 60 or 70 people that I know on a much
deeper level than mere acquantance level that refuse to use debian
in their work or home because of this.

There is a major college in North Carolina whose admin is a good
friend of mine. He refuses to use it in his labs and the 2 reasons
he named were "Debian's stuff is getting too old." and "Debian's
developers act all high and mighty." Not my words, his.

The age of many packages in potato has also caused me to see a major
research lab in Massachusettes decide against debian, as well as
several hosting firms. Some of the training I've had has allowed me
to work as consultant both freelance and company based. Most of the
times it's been Red Hat that has been chosen in those companies.
Definitely not because it's the better distro, but once again a
perceived lack of growth or energy in the maintence of Debian.
I can't remember how many times I've been told that debian is too
slow to keep up with "linux" (this being the community from the way
they expressed it. Community being the software development one :))

This is going on far longer than I planned so I'm going to end with
this. I too see the edges of Debian crumbling as a distribution.
Not heavily to the point of possible extinction, but to the point
that it's definitely noticable that debian has a problem. We need to
fix that. Even the general user can help with that. Look at the
upstream of a package that you use a lot of. Hell, make it 20
packages. Surf and check dates, spend a weekend surfing on them and
email the owners about the age of the product. Give them 4 or 5 days
to respond. Set up a cron job that will mail the BTS with an
important bug if they don't respond. Tell them you're doing this.
(Reason for the cron job is that then you too become a responsibly
party to uphold your end of the bargain. if they *do* respond then
you can remove them from the list. Now you know how many are still
active of the 20 you picked to watch. If they *don't* respond,
you've got the tool in place to make sure that everyone using debian
and that monitors the bug system knows what's going on upstream.)

Get 1000 people that use debian to do this and that's 20,000 packages
with updated information. Do we even *have* that many packages?
There are a hell of a lot more than 1000 people using debian out
there in the world. If you want I can start doing this. I'll anty

In a nutshell folks, we need to take responsibility as a group and
do this, and not just give wisecrack comments in return to good

Cheerfully singing my name to this,

David D.W. Downey - Creator of KiXO Linux - sourceforge.net/projects/kixolinux
Red Hat Certified Engineer - www.redhat.com/training/rhce/courses/rh300.html
LPIC-II Core Test Developer - Linux Professional Institute - www.lpi.org

Attachment: pgpRDFglFaHfJ.pgp
Description: PGP signature

Reply to: