Re: Sun Drops Its Linux Distribution

From: Rev. Don Kool (oldno7@comcast.net)
Date: 04/05/03 

Date: Sat, 05 Apr 2003 08:34:13 -0500
From: "Rev. Don Kool" <oldno7@comcast.net>

Chris Cox wrote:
> Rev. Don Kool wrote:
>> Chris Cox wrote:

>>> IMHO, there's no need for Linux to degrade to Unix.

>> "Degrade"... nice troll bait there, my penguin entranced young
>> friend. LOL!!

> No troll bait.

        Yes, troll bait.

> In fact, it's mainly because of your intelligent
> remarks that I stated that. Linux isn't Unix.

        That's not my remarks, that's simply a fact. LINUX isn't even an
operating system. It's just a UNIX wannabe kernel for PCs.

> Why should it be limited by that?

        It shouldn't. It should be a PC as it can be. What it will never
attain is UNIX status. LINUX remains a kernel in search of third party
utilities.

> You always say "penguin entranced",

        Not always. Only when the shoe fits, my defensive young friend.

> I'll pit my 19+ years of Unix against yours anyday.

        That wouldn't exactly be a good bet for you, Chris.

        [...typical LINTEL kiddie ad hominem snipped...]

>>> One could
>>> just as easily argue for the features in the Linux kernel to be
>>> placed into Solaris (in fact, the list is much longer going that
>>> way).

>> Or features of Windows, OpenVMS, DOS, etc. since you're touting
>> adding non UNIX features to the SOLARIS UNIX kernel.

> Yes... at least you understand this point.

        It is only the penguin heads who lack understanding, my young friend.

> Certainly, there are
> features to be emulated from everywhere, and new ideas to
> make things even better.... once you get outside of the
> box (which Unix apparently isn't allowed to do if I understand
> your prior posts on the subject).

        You don't understand but given your limited UNIX only experience, that
is not surprising. You need to experience other enterprise class
operating systems so that you get a feel for all that is available.
Take a Windows, OpenVMS or MAC course. Spread your wings.

>>> I'm not convinced of the superiority of the Solaris
>>> "multi-threaded kernel" implementation... but it certainly
>>> has the maturity that Linux lacks.

>>> Don't get me wrong, Solaris is solid and somewhat well supported
>>> (somewhat since the support seems to be degrading over time).
>>> But as the Rev will gladly say... GNU Linux is not Unix... and
>>> that's a good thing IMHO.

>> <...immature editorializing snipped...>

> That's sad.

        That's why I snipped it.

        [...typical penguin head ad hominem snipped...]
        
>>> Are there some
>>> very important elements in Unix that appear to be working
>>> better than Linux, perhaps so... as the Rev would say, Linux
>>> is a "hobbiest" OS... that is to say, it's in a constant state
>>> of evolution. And moving forward is not a bad thing, though
>>> it moves people away from their comfort zone of dealing
>>> with the "tried and true" older and perceived to be more
>>> stable technologies. Ultimately, it's about making money.
>>> People like old and reliable.

>> "Old and reliable" is generally a good thing to base multimillion
>> dollar production systems on. Job security generally suffers when
>> production systems pile up downtime waiting for "sURferDuDE" to come
>> up with a patch in his free time.

> I realize you really aren't trying to make intelligent conversation
> here, but you ARE agreeing with the point I made about "old and
> reliable" and "tried and true" being very comforable for enterprise
> businesses.

        It's not about "comfortable", it's about performance. Only an idiot
would put production on a Best Buys PC running LINTEL. HA, redundancy,
survivability, etc. That's what matters. Not the OS. For the most
critical apps, OpenVMS is the only way to go. For the desktop, Windows
is the only way to go. If LINUX is to have a place, a tenuous assertion
at best, it will be embedded in cheap, mass produced appliances like
dishwashers or cable TV boxes. While trusting LINUX to start the wash
in 4 hours might be OK, trusting it in a multimillion dollar production
system is not. Professional Systems Administrators know that basing a
business on when SURferDuDe decides to come back from the beach isn't a
good idea.

> The "sURferDuDE" comment however, again, makes me believe that
> you've only recently graduated from grade school.

        Why; it's a fact.

> I'm still waiting for Sun to release a patch for the latest
> sendmail issue.

        SENDMAIL is a third party app. Go to "sendmail.org" and get the patch.
  Where is the patch from non UNIX systems like BSD, Windows and LINUX?

> I guess some would say, compile it from scratch,
> but of course that defeats the whole Solaris mantra that
> you're preaching.

        It is extremely insulting for you to fantasize that I preach a "Solaris
mantra". I, like most people experienced in the field, consider SOLARIS
to be the worst UNIX system ever foisted upon the public.

>>> My GNU Linux host uptimes are usually twice as long as Solaris just
>>
> ... But
>
>>> I have replaced $75K-150K Sun boxes with <$10K Linux platforms
>>> without any change in performance or reliability. Not sure
>>> about replacing the $500K-$xM types... there really is no
>>> Linux equivalent today (apart from some not quite ready for
>>> prime time stuff... or the z-Series efforts).

>> Exactly. You lack the experience to make an intelligent choice.

> Huh? I do this for a living. I don't (as apparently you do) engage
> clients and say "Solaris is great, Solaris is perfect, death to
> all else"... I keep an open mind instead of a closed one. I deploy
> solutions not ego trips.

        ROTFLOL!!! The last thing I would ever recommend to a UNIX client is
SOLARIS. The last thing I would ever recommend to anyone is LINTEL.
You need to try to gain some security in your abilities, Chris. Then
perhaps you could engage in adult discussion without all the defensiveness.

        [...typical penguin head ad hominem snipped...]

>>> I can't say if Sun is working on a next generation Solaris or not.

>> Of course they are, SOLARIS 10.

> My point is

        If only you had a point, my young penguin obsessed friend. If only you
had a point.... :-(

> that Solaris 8 was baby steps from Solaris 7 and
> 9 was baby steps away from 8. I'm not sensing anything
> grand inside of Solaris 10. However, Sun does do some interesting
> things in HW (though it's been a bit lacking lately probably
> due to the high cost of those things). As I mentioned, the
> fact that Sun (unlike Linux and Windows) can focus on a controlled
> platform, this should really give them a much greater
> advantage....

        Like Apple. We've all seen how well they're doing. LOL!

> and perhaps through hardware, Solaris 10,
> working with that hardware, might do something truly
> new and amazing. But I'm not counting on it right now.
> My guess is that Solaris 10 will continue in 9's path of
> supporting more GNU and other free software products that
> have helped Linux/BSD/Mac to make inroads on Windows (Windows
> being a target still in McNealy's sites).

        Making "inroads on Windows" is a quixotic dream of penguin heads and
Gen-Xers. Windows won the desktop war. Time to move on. SOLARIS and
OpenVMS own the data center. LINTEL is for students, hobbyists and
tinkerers. FreeBSD/OpenBSD/etc are not even a blip on the radar except
for zealots. The OS field is mature. Outside of the major players
there are only rabble rousers who are easily ignored. As you only have
19 years in the field, it is little wonder that you are fascinated by
the latest fad. Fads pass. Performance matters. If you remain in the
field, you will learn and mature.

>>> But I think it would be a great idea for them. But I doubt
>>> seriously they have the in-house expertise to even fully
>>> comprehend the menagerie that is Solaris today (well, the have
>>> expertise, but probably not the time to unravel the mysteries).
>>> Certainly they understand parts of it, but they didn't write 50-75%
>>> of the code base. Microsoft (though to a much lesser extent)
>>> faces some similar issues due to attrition over time. So realistically,
>>> I think we can expect further small steps of evolution in
>>> future releases of Solaris (just like we've seen going from
>>> 2.0 to 9) mostly centered on augementing and fixing of
>>> the existing code base. This of course is great for
>>> corporate enterprises because they don't like radical
>>> change (even when the cost benefits are high... "if it
>>> ain't broke, don't fix it").

>> Your are of course, in way over your head here, Chris. "Corporate
>> Enterprises" are not in the business of LINTEL advocacy. They can't
>> tinker while Rome burns.

> Huh? I work for Fortune 100 companies. How many Fortune 100 companies
> have help to found, Don? Are you on any boards? I'm not talking
> about something I've made up. I'm talking about direct
> experience, something that you obviously lack in this case (though
> I certainly recognize your experience with Solaris).
>
> You obviously must go by another well know industry name
> Don? What is it?

        Please repost in English.

> Maybe we can do lunch sometime?

        No problem. I applaud you LINTEL kiddies reaching out for mentors in
the unfamiliar UNIX field. When and where?

        [...typical truth hurting the LINTEL
        kiddie's feelings snipped...]

        Yours in the glory that is our Lord Jesus Christ,
                                Don 

-- 
***************************   Q: How many frenchmen does it take
* Rev. Don McDonald, SCNA *      to defend Paris?
* Baltimore, MD           *   A: No one knows, it's never been tried.
*************************** 
http://mywebpages.comcast.net/oldno7/TheFrench.mpg
Reluctant Yamaha YZF-R1 Owner