[NCLUG] Looking for user reviews and installation CD-ROMs

J. Paul Reed preed at sigkill.com
Tue Nov 7 00:16:28 MST 2000


Seeing as how this is going to quickly degenerate into a distro war about
the same distro, no less, I will respond to this post, and leave it at
that.

On 07 Nov 2000 at 01:30:43, R P Herrold modified my mailspool to say:

> ... Fair enough; you've changed the terms of your criticism
> from "I've heard ..." to "I've read ..." -- but I still don't
> see a statement that you've actually seriously _tried_ the
> product which you are so venomously tearing down.  

I haven't... because I don't have the time to fix a horked box. I decided
not to install it based upon previous experiences with RedHat x.0 releases,
and the FACTUAL posts on linuxnews.com and others about problems with RHAT
7.0, NOT comments from Slashdot (l)users.

> The original poster asked for user reviews, and I offered a user
> review, rather than repeating hearsay and running a bad news
> clipping service.

And I offered my opinion that they should solicit opinions (or better yet,
test-install) RHAT 6.2, NOT 7.0.

> The GCC / claimed binary incompatiblity criticism again is
> based on the sour grapes of some developers who are unhappy
> that Red Hat was unwilling to wait for them to decide the
> product was ready, and so it finished it to a releasible
> state, differientated it, and released it. -- Tough -- That's
> part of what the GPL is about.  A Freedom to Fork issue is
> present here.  See:
> http://www.gnu.org/philosophy/shouldbefree.html at
> "Obstructing Custom Adaptation of Programs"

No one, including myself, has questioned the Freedom to Fork. You're
arguing a non-issue which no one brought up, and probably no one is
contending.

> A like inability to reach a current featureset stable revision
> caused the GCC/EGCS fork a couple years back -- and the
> issuance of the fork caused the GCC folks to finally get off
> their duff's and relent to committing long needed patches to
> GCC.

egcs 1.1.2 has been good enough for RHAT 6.x, so I don't see why it
wouldn't have been good enough for 7.x; the developers rightly had a bone
to pick with RHAT not particularly because they pulled a beta release and
shipped it, although that is part of it, but moreso because they're putting
a compiler that has no future and could cause untold future problems in the
hands of thousands of people, without fully realizing the consequences of
that decision and attempting to account for them.

Quality software takes time... RHAT should've waited.

BUT, they didn't, so fine... whatever.

I'm sure by RHAT 7.2, they'll have gcc 3.0, which will make me happy...
I'll upgrade then... but not now.

> (I would note that the fix was publicly announced, rather
> than slip-streamed in, as has been a squable over on the
> OpenBSD tech list within the last month.  This is to Red
> Hat's credit.)

Well, the fact of the matter is that RHAT 7.0 still shipped with a sloppy
coding bug that should've been fixed before bits were pressed.

When Microsoft has done this, the Linux community has crucified them...
yet, when RHAT does this, it's somehow OK because it's Linux... uhhh... no. 

> Don't like the RH update network, or related update daemon?
> -0- Okay, don't use them.  

Exactly... I don't.

> That they are installed is not hidden (they're on the manifest, they're
> in the online help at install time, they're in the install log, they show
> as one configures services for a given runlevel), and they can be
> de-selected during install if one wishes not to have them on
> your hard drive.

*shrug*.

Ok. I know certain packages, such as kudzu and linuxconf, get installed
whether or not I want them to, and whether or not they're selected... I
purposefully didn't install RHAT 7.0, because I was afraid of this being
the case with this auto-updater dealy... maybe not.

> Because you've seemingly NOT installed and use-tested, I guess
> that you are unaware that free, anonymous, non-priority update services
> are presently provided by Red Hat -- I don't understand the criticism of
> giving away anonymous free updates of a freely available GPL'd product.

Uhh... I didn't criticize that.

> As to the prior install consent issue, an autorpm driven FTP
> update does not even install those components.  Personally I
> think it is pretty neat that I can bump a 3.0.3 Red Hat box to
> 7.0.  That is impressive support of their line to my thinking.

Have you actually done this, or is that something from their webpage? If
so, I'd agree that it is, but considering I couldn't even install RHAT 5.2
on a 486, much less get it updated to 6.2... I dunno...

> Are you also 'disgusted' with the KRUD CD's, or our (
> http://www.owlriver.com/projects/ ) projects page where GPL'd
> packages are made freely available too? -- 

Uhh... I think you totally missed a point here, and are again arguing
something that is a complete non-issue, and not even something I brought up.

I am "disgusted" with this: http://www.redhat.com/products/network/,
which is not what you're talking about at all. Now, obviously, RHAT has to
make money, and in open source, you do that with services/support, and
that's really a whole different post in and of itself, but again, Microsoft
does the same thing, and we're all skeptical and we think it sucks, and yet
when RHAT releases the same thing, WHICH CRASHED COMPUTERS AT FIRST, it's
suddenly all ok.

Ultimately, I think you missed the point of what I was saying.

I use RedHat. I like RedHat. I defend RedHat against those crazy Debian
folk at CPLUG... except for RedHat x.0 releases.

They have a history of releasing total crap for their x.0 releases, and it
DOES GET FIXED for the x.1/2 releases... which is good. And it is those
releases which I recommend and use whole-heartedly

I find it extremely disconcerting to see someone so forcefully and blindly
promote RHAT without stepping back to look at the problems and issues that
decisions RHAT makes can cause. It sounds to me like you'd bless anything
RHAT released, even if it were a copy of IE for Linux that had to run as
root to be used.

That mentality is dangerous, whether it's pro-Microsoft, or pro-Redhat.

That being said, my original recommendation to the person who was asking
about distros was to use RHAT 6.2 if you care about stability and Linux
shining as a product, instead of frilly "auto updaters" and the "latest
desktop" which will crash as often as your Windows box does (well... every
two weeks).

Later,
Paul
  -----------------------------------------------------------------------
  J. Paul Reed                 preed at sigkill.com || web.sigkill.com/preed
  We're living in a world that's blowing itself to hell as fast as every-
  one can arrange it.       -- First Sgt. Edward Welsh, The Thin Red Line



More information about the NCLUG mailing list