Re: [cobalt-developers] nice knowing you all!

[corliss@xxxxxxxxxxxxxxx]

On Fri, 22 Sep 2000, Jeff Lasman wrote:

> Not really DOS-based, though it's still got lots of DOS-code in it. 
> This has been discussed over and over ad-nauseum in both pro- and con-MS
> groups, so I won't go into it in detail here.  But Win98 isn't
> DOS-based.
> 
> Just like WinNT isn't DEC VMS-based, though it's got a lot of VMS legacy
> behind the windows interface.

Justify it any way you want, but to deny that the 16 bit DOS kernel is *still*
bootstrapping the rest of the services Win98+ provides in in complete denial
of the facts.  Furthermore, MS has admitted that WinME is an evolutionary step
from Win98, which was a step from Win95, and so on.  So what we have here is
complete failure on MS' part to migrate to the NT kernel as promised four
revisions ago.

> My networks are NOT large, but I currently administer for myself a
> Cobalt RaQ2 and RaQ3, generic RedHat Linux 6.x, (with Red Hat's secure
> server), BSDi4.1, FreeBSD4.0, WinNT4.0SP4, and Win2K Advanced Server
> (and a Win98 desktop).  For others a lot more RaQs and generic Linux
> boxes.

Then you're blowing my mind.  You don't see *any* problems for the less
technically inclined trying to work around the various differences of these
Unices?  That's the market Cobalt is catering to.  I can't wait to see what
happens when, as another poster mentioned, these people try to install an x86
binary RPM on Solaris.  That should be fun.

> That's not what I was thinking of.  I was thinking of using the RaQ
> strictly as an appliance, using the gui interface, which I expect would
> be evolutionary rather than revolutionary, as it was in the switch from
> the Qubes to the RaQs, from the RaQ2 to the RaQ3.  That's why I talked
> about washing machines.  They're appliances after all, and most of us
> don't know or care that different models come from different OEMs.

I don't know how realistic that is.  Many of these "appliances" are so simple
that it whets the appettite for fancier functions, given that these people are
already under the illusion that it can't be that much more difficult.  I'm
willing to bet that there's a fairly high percentage of Cobalt owners (say
20%-40%) that are doing more with the units than what was possible out of the
box.

> Look again at what I've posted.  I said that, basically, for those of us
> on the developers list. But most of Cobalt's market is not developers,
> and is people who are happy with the web interface.  And again, it
> should have been clear from my context that that was what I was
> referring to.

Look at what I've written above, I don't think your observation is relevant.
Even normal users are going to see crap on other web sites that they want to
implement on their Raq, and Pandora's box is opened.

> Personally, I find myself constrained enough by Cobalt RaQ software that
> I use my RaQs in very limited application.
> 
> I do continue to support them, and their interface, for my cleints,
> though.  That's my market niche.

If you can keep your users contrained to that little niche, then more power to
you.  But I don't that would fly for long up here.

	--Arthur Corliss
	  Programmer/Administrator
	  Gallant Technologies (http://www.gallanttech.com/)