[Rule-list] The state of the installer message

[Martin Stricker]

Devon wrote:

> OK, a few random thoughts on progress and direction.

> To solve this, perhaps we could add a little code to the memory test.
> If memory < 10000, sorry, can't be done, exit. If memory < 17000,
> display disclamer, offer only the 'lowmem' install classes, and a
> slimmed down comps file. Offer to quit or continue.
> Just something to consider.

The disclaimer would perhaps stop Red Hat from doing this. See below...

> The reason I bring this up is that I believe we have hit the low end
> of the memory limit, while continuing to use anaconda and a stock Red
> Hat distribution. I believe that limit is going to be ~12M. Yes, with
> more changes, we could possibly slim it down more, but I doubt Red Hat
> would be interested in supporting that ability. On top of that, how
> many end users would be willing to endure a basic install that takes
> 12 -16 or more hours, resulting in a machine that uses 98% of the cpu
> time just managing swap? Swap is a wonderful thing, but it isn't a
> substitute for RAM. Can you imagine trying to compile a kernel or
> rebuild a large RPM under those circumstances?
> 
> I'm not trying to be negative, just realistic here. Red Hat has to
> support what they sell, and isn't likely to want to deal with the
> support calls a low memory install might generate.
> 
> Thoughts, opinions, counter claims, flames?

I agree with your thoughts. We indeed have reached the limit of what
only slight modifications to anaconda can accomplish, and I don't think
Red Hat will accept major changes to it in their distro.

Someone at Red Hat does the final decision on what to add into the
distribution and what not. We should talk to that person ASAP! Brock,
can you tell us whom to ask, or, if you aren't allowed to do that, help
us in other ways to establish contact? Or do we need to file a RFE
(request for enhancement) on anaconda in bugzilla and wait what might
happen? Or is anaconda-devel-list the place to ask?

=====================================================================

Let's think about what *Red Hat* wants for a moment.

Red Hat's main target market is business  PCs and business servers. They
make most of their money on selling support contracts and Red Hat
Network (up2date), not the boxed sets. (Please correct me if I'm wrong!)
So what could Red Hat gain with adding low-memory support, and what
could they loose?

Individuals who install Red Hat Linux on their low-memory machines might
be poor (some of our targets are schools and underdeveloped counties),
so it is not very likely that they will buy a boxed set => no gain for
Red Hat in this market!

Another target could be governments, authorities and universities in
underdeveloped countries (as Romualdo pointed out in the initial thread
that started RULE). These most probably will buy a boxed set for their
IT staff, maybe in some cases even the server boxed set. Also some will
buy Red Hat Network subscriptions and support contracts. This target
market could be of interest for Red Hat. Maybe even some small- to
medium-sized companies could be interested in using Red Hat Linux - try
to run Win2k on less than 32 MB of RAM!

Red Hat is *very* cautious in adding any experimental or just not well
enough proven software into their distribution. Just think abaout all
these discussions about adding installer support for ReiserFS... But why
is Red Hat that careful? In comanies, especially the large ones,
decisions are done by the Evil Suits (TM), the managers. They lack any
technical background and thus distrust technology. While a techie thinks
"so there they did the *interesting* stuff" when seeing something
experimental or "be careful", a manager just gets frightened and
wouldn't buy that product under any circumstances. What managers want to
hear is that the product is safe, proven, QAed. And the two magic words:
It has to be "standard" and "supported". It's amazing what one can
accomplish with using these words. I have to cope with the Evil Suits
(TM) every day, and that's my main strategy. It works! :-)

What does that mean for the RULE installer? (Marco, where's the name you
promised? ;-) ) It has to be tested *very* thoroughly, and the memory
minimum must be high enough to assure the install *will* work. Maybe
that can be 12 MB, maybe it must be higher, only testing more will show.
Any experimental things or anything that might not work always like
using swap early (especially if it throws a nasty error message or
warning!) cannot make it into the official Red Hat Linux. We can provide
information on how to do this, but I'm sure Red Hat wouldn't include it
into the distro.

Another thing is reliability: Both the install and the installed system
have to work as flawlessly as possible. Selling support contracts only
pays off if there are not too much support calls. This again means heavy
testing. Sadly old hardware has two downsides in this regard (which I
suppose were two of the reasons for Red Hat to drop support for old
hardware in the first place): It's not very much available for testing
purposes, and hardware problems will occure more often, causing the
customers to call Red Hat support more often.

Sorry for strengthen your patience so long. I'm really curious what Red
Hat staff/managers do think about this!

Best egards,
Martin Stricker
-- 
Homepage: http://www.martin-stricker.de/
Red Hat Linux for low memory: http://www.freesoftware.fsf.org/rule/
Registered Linux user #210635: http://counter.li.org/

_______________________________________________
Rule-list mailing list
Rule-list at mail.freesoftware.fsf.org
http://mail.freesoftware.fsf.org/mailman/listinfo/rule-list