Linux-Advocacy Digest #960, Volume #29 Tue, 31 Oct 00 11:13:04 EST
Contents:
Re: Astroturfing (Jason Bowen)
Re: 2.4 Kernel Delays. ("Les Mikesell")
Re: Linux growth rate explosion! (Roger Lindsj|)
Re: Ms employees begging for food ("Les Mikesell")
Re: MS Hacked? (chrisv)
Re: Windoze 2000 - just as shitty as ever (T. Max Devlin)
Linux 2.4 needed now (R.E.Ballard ( Rex Ballard ))
Re: Ms employees begging for food (T. Max Devlin)
Re: Ms employees begging for food (T. Max Devlin)
Re: IBM to BUY MICROSOFT!!!! (2:1)
Ethernet saturation and throughput (was: Ms employees begging for food) (hack)
Re: Once agian: Obscurity != security (Was: Tuff Competition for LINUX!
([EMAIL PROTECTED])
Re: 2.4 Kernel Delays. (R.E.Ballard ( Rex Ballard ))
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (Jason Bowen)
Crossposted-To: comp.os.os2.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: Astroturfing
Date: 31 Oct 2000 14:22:47 GMT
In article <39fe2e88$1$yrgbherq$[EMAIL PROTECTED]>,
<[EMAIL PROTECTED]> wrote:
>In <8tl8lo$ajm$[EMAIL PROTECTED]>, on 10/31/00
> at 01:52 AM, [EMAIL PROTECTED] (Jason Bowen) said:
>
>>In article <39fe1d3d$3$yrgbherq$[EMAIL PROTECTED]>,
>> <[EMAIL PROTECTED]> wrote:
>>>In <8tk10o$ib4$[EMAIL PROTECTED]>, on 10/30/00
>>> at 02:36 PM, [EMAIL PROTECTED] (Jason Bowen) said:
>>>
>>>>Who's the one that is obsessed and angry? You are frothing :-). I only
>>>>mentioned this in passing when bringing up the kind of person you are in
>>>>response to the getting paid comment. You never did provide a reason why
>>>>your cheerleading is different from anybody others. You make baseless
>>>>attacks and don't like it when you are called on it. Tell me Ed, why should
>>>>anybody not believe you are petty, immature etc.. when your best lines are
>>>>grade school insults and claims of being paid because you don't like the
>>>>message?
>>>
>>>
>>>No I'm not frothing at all. I just say what I mean and I mean what I say --
>>>You're a complete asshole. I could use other words, but there is nothing else
>>>in the English language that quite means the same thing when one is trying to
>>>break through to a moron who refuses to listen to reason. The other problem
>>>is that you're just like the other asshole who chimed in. You think direct,
>>>accurate and concise words that point to your personality defects show anger
>>>-- when in fact it only means I can express the reality here. Even though you
>>>are too dense to understand it.
>
>>So I'm an asshole for being right when sticking to the conversation topic
>>then? I'm understanding now. And Ed, you use words like asshole constantly
>>because your vocabularly isn't exactly ripe with much else. You seemed to
>>have chimed in to a few conversations with not much else but irrelevant
>>off-topic information or claims that you can't provide proof for.
>
>No. You an asshole -- for carrying a grudge for nearly a year now . The fact
>that you are now complaining without end that I have called you an asshole for
>this behavior -- is proof that you are what I call you, e.g. "Asshole." I
>could use bigger words, but there is no indication that you would understand
>them.
>
I would consider somebody to be an asshole that was wrong about a subject
and never admitted to it. Remember when you said earlier in this thread
that you never claimed that OS/2 overcame hardware problems?
http://x60.deja.com/[ST_rn=ap]/getdoc.xp?AN=596562408&search=thread&CONTEXT=972958734.160628801&HIT_CONTEXT=972958420.159383593&HIT_NUM=529&hitnum=0
http://x60.deja.com/[ST_rn=ap]/getdoc.xp?AN=597133173&search=thread&CONTEXT=972958734.160628801&HIT_CONTEXT=972958420.159383593&HIT_NUM=529&hitnum=39
The second one, which is between you and somebody else is particularly
funny. That is where you claim that OS/2, "can overcome" what is being
talked about. The person replying, Dave, points out that OS/2 may have an
advantage over 95 if it loads bottom up rather than top down but that it
isn't overcoming the hardware.
>Furthermore I have no intention of trying to prove anything. After all you
>just whined the other day that you are great EE engineer. Even though in the
>year that has gone by you could have designed and verified your own test to
>prove me wrong -- but all you have done is bring up the same old crap numerous
>times about how I injected myself into your little war with someone else.
>
>Go away little boy. Your time in life has not come yet -- and if you don't
>smarten up it never will.
I'm not an EE and I just provided proof for you lies above.
>
>>>
>>>I didn't expound on your asinine cheerleading whine, because its the only
>>>thing you can point to that even resembles cheerleading (and then its only in
>>>your mind). It was close to year ago, and not a cheer. It was a salient point
>>>that ou just happen to not like, because it made the little game you were
>>>playing essentially moot. I also mentioned this before -- but you are too hot
>>>headed to read or perhaps your pea brain just didn't get it.
>
>>Your lost. Your cheerleading is your pro-OS/2 stance. I have as much
>>evidence as you have for anybody else being paid to support a product so you
>>must be paid using your logic. See how that works Ed?
>
>You are using a one-time comment from a year ago to point to me as a
>cheerleader! And you whine that I must have been paid to say it? What else
>should we think of the college boy with the big EE who uses a one-time year
>old remark to prove his point -- that you are young and dashing and brilliant?
>The fact is you are an idiot with a grudge, e.g., an asshole! You ought to be
>spending your time with one of th 15 thousands chicks there, yet you're here
>with your whining and pissing like that is going to prove you are right. If
>you were a complete person you would be doing something different.
>Something's not right with you kid and its time you went looking for it.
>
You make no sense. You just attack, attack, attack with no logic to your
arguments other than a grudge. I just proved a lie and have gone after
you for not providing proof for a claim. If you think the time it takes
for me to compose this message somehow takes away from my social life it
must be a reflection of your own life. I do hang out with many "chicks",
in fact two "chicks" bought me a ticket for my birthday to go to a concert
this Thursday night with them.
>
>>>The rest of your stuff is simply bullshit and you know it -- or if not, then
>>>you have a genetic defect. Now, I think the Boys section is in another
>>>newsgroup. You and the other fool should go there and play. When you get a
>>>full set of working brains and stop carrying grudges around, comeback.
>
>>What grudge? Being right about the topic at hand and then questioning your
>>logic for being sure that others are being paid for having viewpoints not
>>favorable to yours? You still haven't pointed to evidence that proves that
>>people are being paid to post.
>
>You are whining about this like M$ is Lilly White and never does anything
>wrong. If you don't think that M$ pays people to spread the word according to
>Gates, then you are a bigger asshole then I had thought.
>
>Go away little boy. Your time hasn't come yet -- get a girl and stop whining.
>
You really have a problem. I haven't mentioned Microsoft at all, I told
you to provide proof. I'm aware of Microsoft's tactics in the past and
present but I think that they would have more compentent people than the
ones you feel are being paid. You really think I'm rooting for Microsoft
since I'm against you? That sums you up all nice and neat.
>
>-
>-----------------------------------------------------------
>[EMAIL PROTECTED]
>-----------------------------------------------------------
>
------------------------------
From: "Les Mikesell" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy,comp.os.ms-windows.advocacy
Subject: Re: 2.4 Kernel Delays.
Date: Tue, 31 Oct 2000 14:33:01 GMT
"Chad Myers" <[EMAIL PROTECTED]> wrote in message
news:H3AL5.26592$[EMAIL PROTECTED]...
>
> > > Meanwhile, Look at sites like microsoft.com, barnesandnoble.com, ebay,
> > > NASDAQ, and hundreds of other major ebusiness sites that all run
mutiple
> > IIS
> > > hosts on a single site.
> >
> > That just suggests to me that IIS is not robust enough to count on one
> > machine staying up, but I don't see much of a point here.
>
> So all these sites would run on one single Linux/Apache box? BS, you're
> full of it. Yours is a rediculous statement, let alone retarded.
>
> *PL0NK*
>
> -Chad
At least they could count on it to run long intervals without crashing.
Here's
one of mine:
$ uptime
8:26am up 348 days, 13:47, 1 user, load average: 0.15, 0.24, 0.26
(It handles financial data from commodity exchanges so the load hasn't
ramped up yet - it will peak at about 1.5 at 10 AM). I have a backup
machine of course, but haven't seen any particular need to deal with
automatic failovers.
Les Mikesell
[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (Roger Lindsj|)
Crossposted-To:
comp.lang.java.advocacy,comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: Linux growth rate explosion!
Date: 31 Oct 2000 14:37:32 GMT
In article <UEpL5.25589$[EMAIL PROTECTED]>,
Chad Myers <[EMAIL PROTECTED]> wrote:
>
>"Roger Lindsj|" <[EMAIL PROTECTED]> wrote in message
>news:8tk19l$42h$[EMAIL PROTECTED]...
>>
>> What applications do you need?
>
>Real ones, not cheap knock-offs.
What a great answer. Looked at several sites selling software, and
none of them sells "Real ones". They do however sell editors, word
processors, layout programs, CAD programs, compilers etc.
So I aska again, what applications do you need?
Roger Lindsjö
------------------------------
From: "Les Mikesell" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy,comp.arch,comp.os.netware.misc
Subject: Re: Ms employees begging for food
Date: Tue, 31 Oct 2000 14:51:42 GMT
"Andy Newman" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Les Mikesell wrote:
> >A socket is a file once you get past the magic of opening it.
> ^^^^^^^^^^^^^^^^^^^^^^^
> That's the problem.
Oh, I agree. Much of the beauty of unix is the general lack of magic
needed at the application layer, and the way this has allowed
existing programs to work with new kinds of media with no
changes.
> Sockets are a (very useful) wart. With no file system representation
> you can't manipulate them without special programs. And there's
> an extra I/o mechanism to worry about and its properties. Consider the
> mess that is SysV IPC. Peter said it previously but Plan 9 got it right,
> everything in one name space with a few common "methods" applied to the
> "objects" in that name space (with location transparency thrown in for
> free, on a per-process basis). When followed to the letter the true value
of
> the abstraction is evident.
Named pipes are the most sensible SysV IPC mechanism in this respect.
You can throw a perfectly usable client/server mechanism together
with nothing but shell scripts, using a shell that has not been modified
since the invention of the FIFO. URL-named sockets could give the
same ability across machines, although it isn't too bad to run helper
programs (rsh, etc.) on pipes so you don't have to deal with opening
the socket directly. And of course perl has changed what we expect
script interpreters to be able to do. Once upon a time the shell had
a command that just about matched every system call in unix, but
this was never updated when the SysV IPC and BSD sockets were
added to the kernel. Perl on the other hand, knows all the magic and
has an extension mechanism to add more. If we didn't have perl
someone would have had to add those to the shells or come up with
a new shell.
Les Mikesell
[EMAIL PROTECTED]
------------------------------
From: chrisv <[EMAIL PROTECTED]>
Subject: Re: MS Hacked?
Date: Tue, 31 Oct 2000 15:07:16 GMT
"Erik Funkenbusch" <[EMAIL PROTECTED]> wrote:
>Correct. I should have said "but seldom why".
I still disagree. This is extremely basic information.
------------------------------
From: T. Max Devlin <[EMAIL PROTECTED]>
Crossposted-To: alt.destroy.microsoft,comp.os.ms-windows.advocacy
Subject: Re: Windoze 2000 - just as shitty as ever
Date: Tue, 31 Oct 2000 10:10:19 -0500
Reply-To: [EMAIL PROTECTED]
Said Ayende Rahien in alt.destroy.microsoft;
>> You could have a small linux partition. to do the real work
>>
>> Right now, I'm burning a CD-Rom with some MP3 files. I was surpassed,
>because I
>> did a "cdrecord -scanbus" and then "cdrecord -v speed=2 dev=0,5,0 -eject
>> /mnt/gnapster/*.cdr" and the darn thing found my CD-Rom drive (a rather
>old,
>> obscure, inexpensive drive) and created a perfect CD-ROM the first time I
>tried
>> it. Of course, all the CD-Rom burning software is absolutely free. And I
>can
>> create the CD-Rom image as a disk file and "mount" it and actually see the
>disk
>> image as a filesystem to make sure everything is OK.
>>
>> On the other hand, when I boot the same computer into Win95 (game machine
>for
>> kids), and forget and leave ANY CD-Rom in the drive when I boot, I get a
>message
>> saying windows has seg faulted, and I have to remove the CD-Rom and
>reboot. And
>> forget about any free CD-Rom burning software!
>
>Obviously you haven't looked very hard:
>
>http://download.cnet.com/downloads/0-1896417-100-1757739.html?tag=st.dl.1000
>1_103_1.lst.td_1757739
>(not sure, but it might be just for cd-audio)
>http://download.cnet.com/downloads/0-14480-100-909951.html?tag=st.dl.10000_1
>03_1.lst.td_909951
>This would let you write your own cd-burner for windows machines.
>
>Both are free, (I've tried neither) I found them by going to downloads.com
>and typing burning cd and choosing free licensing.
>1 minute search.
>You could dig up something better if you would try.
So you haven't ever used either, but you're just sure they're going to
be fine. Thanks for your intensive research. No, the request was not
for cd-audio-only burning or for "writing your own cd-burner". If this
is all that came up on a cnet search, it appears that the situation has
remained unchanged from when I checked last year. Forget about any free
cd-rom burning software for Windows; years ago there were a couple, but
they had all been yanked two months or so before I started looking. Use
the crap that's bundled with the unit or use Linux.
--
T. Max Devlin
*** The best way to convince another is
to state your case moderately and
accurately. - Benjamin Franklin ***
======USENET VIRUS=======COPY THE URL BELOW TO YOUR SIG==============
Sign the petition and keep Deja's archive alive!
http://www2.PetitionOnline.com/dejanews/petition.html
====== Posted via Newsfeeds.Com, Uncensored Usenet News ======
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
======= Over 80,000 Newsgroups = 16 Different Servers! ======
------------------------------
From: R.E.Ballard ( Rex Ballard ) <[EMAIL PROTECTED]>
Subject: Linux 2.4 needed now
Date: Tue, 31 Oct 2000 15:02:19 GMT
Here are some links about some of the actual and potential candidates
for commercial adoptions of Linux.
http://www-4.ibm.com/software/is/mp/linux/
http://www.informationweek.com/shared/printArticle?article=infoweek/799/
ibm.htm&pub=iwk
http://www.netaddress.com/tpl/Info/Popup?hidden___url=http%3A%2F%2Fwww.n
ytimes.com%2Fads%2Femail%2Fcompaq
http://www.dell.com/us/en/bsd/topics/linux_linuxhome.htm
This machine was originally intended to be a Linux machine but the
Linux version had to be delayed due to unavailability of Linux 2.4
http://www.netaddress.com/tpl/Info/Popup?hidden___url=http%3A%2F%2Fwww.n
ytimes.com%2Fads%2Femail%2Fcompaq
--
Rex Ballard - VP I/T Architecture
Linux Advocate, Internet Pioneer
http://www.open4success.com
Linux - 60 million satisfied users worldwide
and growing at over 9%/month! (recalibrated 10/23/00)
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: T. Max Devlin <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy,comp.arch,comp.os.netware.misc
Subject: Re: Ms employees begging for food
Date: Tue, 31 Oct 2000 10:14:05 -0500
Reply-To: [EMAIL PROTECTED]
Said Terje Mathisen in comp.os.linux.advocacy;
>Bengt Larsson wrote:
[...]
>Re. the current "discussion" with Mr. Max; since that time, I have
>_known_ that the 'classic' 30% max utilization number is plain wrong.
It isn't "wrong" in the slightest bit. You just don't seem to
understand how to apply it, why it exists, or what to use it for.
--
T. Max Devlin
*** The best way to convince another is
to state your case moderately and
accurately. - Benjamin Franklin ***
======USENET VIRUS=======COPY THE URL BELOW TO YOUR SIG==============
Sign the petition and keep Deja's archive alive!
http://www2.PetitionOnline.com/dejanews/petition.html
====== Posted via Newsfeeds.Com, Uncensored Usenet News ======
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
======= Over 80,000 Newsgroups = 16 Different Servers! ======
------------------------------
From: T. Max Devlin <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy,comp.arch,comp.os.netware.misc
Subject: Re: Ms employees begging for food
Date: Tue, 31 Oct 2000 10:23:53 -0500
Reply-To: [EMAIL PROTECTED]
Said [EMAIL PROTECTED] in comp.os.linux.advocacy;
[...]
> You're completely missing my point. First of all, there are loads
>of cheap systems that implement shared 100BaseT -- they're called hubs,
>and there are probably 100 brands of them. I think I *personally* own
>at least three different brands. Of course, there might only be five
>(or fewer) chipsets in wide use in these products, which may be what
>you meant but -- nah, you're just a stupid troll.
Shared media 100BaseT hubs? It doesn't matter how many you own, they're
outnumbered by shared 10BasteT hubs by factors of tens of thousands, and
100BasetT switches by factors of hundreds. 100BaseT hubs (shared media
repeaters) are very rare in professional networking.
As far as being a troll is concerned, I'm not insisting my personal
experience is global. I'm just pointing out my global experience is
personal.
> Second, while a hub-based system works fine, the total bandwidth
>the network can carry is 100Mbps, regardless of which hosts are doing what.
>In practical terms, it's less than 100Mbps, as you have noted. It's more
>than 10Mbps, though. Quite a lot more.
Blah blah blah. You're still wasting *way* too much time worrying about
bit rates, and ignoring the entirety of the network outside one single
(and generally non-limiting) component system. Good job.
> The reason for going switched is that you can now get 100Mbps between
>hosts A and B while at the same time getting 100Mbps between C and D, and
>so on. Thus, the total carrying capacity of the network is measured
>(typically) in Gbps, though the total capacity available to any given host
>is 100Mbps (well, 200 if you can persuade it to go full duplex).
The reason for putting switched in is that you don't really understand
how your network works, so you throw some bandwidth at it. Sometimes it
even works.
> So, no, going to switched is not because shared 100BaseT networks
>can't comfortably sustain 50% utilization or more, it's because switched
>buys you a LOT more capacity without changing your endpoints at all.
Only if you conveniently redefine "end-point". Which is trivial to do
for the channel user; it just makes network management a nightmare. So
being able to sustain 50% utilization or more (or whatever you meant)
isn't buying a lot more capacity? Hmmm....
> Of course, you know all this, since you're just a stupid troll
>amusing yourself (oddly enough) by making people explain really simple
>things to you. I would be happy to walk you through simple arithmetic
>some day too.
Actually, I happen to be a real-life expert in real-world (as opposed to
'vendor theory') networking and network management. Your problem,
obviously, is that you think this is an issue solved by simple
arithmetic. Believe me, I can hold my own with any specialist in these
particular boxes, (its amazing what you can learn just by reading the
802.3 spec directly), the trick is to not get hung up on a particular
box as if it is the whole network.
--
T. Max Devlin
*** The best way to convince another is
to state your case moderately and
accurately. - Benjamin Franklin ***
======USENET VIRUS=======COPY THE URL BELOW TO YOUR SIG==============
Sign the petition and keep Deja's archive alive!
http://www2.PetitionOnline.com/dejanews/petition.html
====== Posted via Newsfeeds.Com, Uncensored Usenet News ======
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
======= Over 80,000 Newsgroups = 16 Different Servers! ======
------------------------------
From: 2:1 <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: IBM to BUY MICROSOFT!!!!
Date: Tue, 31 Oct 2000 17:23:53 +0000
Chad Myers wrote:
>
> "2:1" <[EMAIL PROTECTED]> wrote in message news:[EMAIL PROTECTED]...
>
> <SNIP>
>
> > I saw 3.5.1 on floppies. I think someone at the company i worked for
> > actually installed it. The mind boggles.
>
> I had to install NetWare 3.11 from floppies once... I think it had
> somewhere in the neighborhood of 50 floppies? I could be wrong, but
> it was some ungodly amount. Inevitably, Disk 24 would be bad and would
> screw the whole installation.
>
> -Chad
Win95 on about 20 floppies or so was bad enough. And did it take a long
time.
Fortunately, the days of doing huge installs from floppies is over.
So are the days of backing up your hard disk on to 20 floppies.
-Ed
--
Konrad Zuse should recognised. He built the first | Edward Rosten
binary digital computer (Z1, with floating point) the | Engineer
first general purpose computer (the Z3) and the first | u98ejr@
commercial one (Z4). | eng.ox.ac.uk
------------------------------
From: [EMAIL PROTECTED] (hack)
Crossposted-To: comp.os.ms-windows.nt.advocacy,comp.arch,comp.os.netware.misc
Subject: Ethernet saturation and throughput (was: Ms employees begging for food)
Date: 31 Oct 2000 15:37:38 GMT
I think Ethernet saturation in the face of rising traffic depends a lot on the
transmission protocol. Window mechanisms (like TCP) tend to generate more
retry traffic as packet loss increases, leading to a steep saturation curve
(where effective throughput drops precipitously when traffic (tries to)
increase. In the mid-80s our research group used a protocol where retry
traffic behaved well when load went up (by increasing interpacket delay when
the proportion of lost packets went up, and decreasing it when it went down).
This led to a much smoother saturation curve, and it was my impression that
total effective bandwidth was pretty high without severe impact on individual
traffic bandwidth. (Of course, the more concurrent transmissions there were,
the lower the individual bandwidth: it's a shared medium after all.) We never
did any serious measurement, unfortunately, since the network effort was only
a side show in our project.
Michel.
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: comp.os.ms-windows.nt.advocacy,comp.os.ms-windows.advocacy
Subject: Re: Once agian: Obscurity != security (Was: Tuff Competition for LINUX!
Date: 31 Oct 2000 10:40:08 -0500
"Chad Myers" <[EMAIL PROTECTED]> writes:
> "Perry Pip" <[EMAIL PROTECTED]> wrote in message
> news:[EMAIL PROTECTED]...
>
> <SNIP>
>
> > 200,000 open source programmers looking at the code. Only one needs to
> > be honest enough to reveal a trojan and then the whole world
> > knows. Because of this, no open source programmer even thinks of
> > putting a trojan in the code, becuase he knows it would be there for
> > the public to see, he'd get caught, and no one would ever let him
> > contribute to an OSS project ever again, not to mention he could go to
> > jail. Funny how human behavior changes when it's open for the public
> > to see.
>
> Of course, this is false, because there are several documented
> occurances, the most notable was the cc compiler that would muck with
> the login process to create a backdoor. It was compiled into the
> original cc, but the source distributed didn't have it, so no one knew.
Several documented occurrences of open-source backdoors? Cite please.
You need to control the binary form of the compiler that the end user
uses to hide such a back door.
The one case you do cite was only an internal experiment; it was never
distributed.
In any case, it would have been a case of a closed-source back door
rather than open-source, in that the source distributed would not match
the binary distributed.
--
Bruce R. Lewis http://brl.sourceforge.net/
------------------------------
From: R.E.Ballard ( Rex Ballard ) <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: 2.4 Kernel Delays.
Date: Tue, 31 Oct 2000 15:34:37 GMT
In article <8tmke7$9er$[EMAIL PROTECTED]>,
[EMAIL PROTECTED] wrote:
> R.E.Ballard ( Rex Ballard ) <[EMAIL PROTECTED]> blessed us
> with this writ:
>
> > On the other hand, corporate decision makers look to the
> > publishers for guidence as to what is appropriate for general
> > use.
> >
> > Compaq **could** put 2.4 on their appliance, but the investment
> > community would consider it an excessive risk. Dell **could**
> > put 2.4 on their laptops, but the stock would dip and corporate
> > customers wouldn't go for it.
> >
> > The "test" or "beta" designation is a red light. Until that is
> > removed and stamped "production" (the green light), the corporate
> > market won't go for it.
> >
>
> Rex, you apppear to be (more) histerical (than usual).
> This is*your* job-to talk to those 8000 publishers you
> have in your thrall and explain how this works.
Very true. I've been looking at this as a two-way street, a dialogue.
I listen to my corporate contacts, listen to their concerns and
their fears.
> I.e. that if there is a slight
> chance that the new kernel will 'destroy the harddrive' (or
> however you yourself put it) then simply renaming it 'final' will
> solve nothing- the risk will remain.
I've been reviewing the Linux kernel status page:
http://linux24.sourceforge.net/status-changes.html
There are some things that should be delayed. Perhaps they could
be put into a later 2.4.x release. Others, such as USB Audio and
USB with APM could be released as beta.
Linux is still a modular kernel, and the biggest problems seem to be
in modules, not in the core kernel.
Either that or a formal 2.2 release that supported the USB devices
and other drivers that have been successfully back-ported.
Unfortunately Linux 2.2 is being orphaned while Linux 2.4 is going
feature-happy as if the ONLY way new features could be added is if
they were installed in 2.4.1 "Gold".
Remember, when 2.2.1 was released, USB barely supported mice, and some
keyboards. Linmodems such as the Lucent drivers didn't exist, and
hundreds of other features have been added from 2.2.1 to 2.2.17,
including enhancements.
Many of the GNU products used to be released very often with new
revisions published every week. This gave packagers the chance to
get bug-fixes and accept/reject new features that might be less than
perfectly reliable.
> The learning curve is
> steepest right now on the side of the 'industry' that is caught
> up in this unseemly craving for Linus-product. Explain to them
> that as the kernel assumes a larger footprint geometry takes hold
> and they may have to keep waiting for their new, free, toy, er,
> keystone.
The issue here is whether the 2.4 release should be treated as a
monolithic product. In fact, most of the system is designed
in a modular structure. There are several reliable file-systems
that do work. But we're waiting for the buggy one. There are
a number of new USB drivers and features that do work, but
we're waiting for the buggy ones. There are hundreds of new
hardware configurations that now work under Linux 2.4, but we're
holding for very specific hardware that acts wierd.
> Or else perhaps the world is nothing but a net of vast
> conspiracies and once again I have no clue whatsoever- my usual
> (implied) disclaimer.
I don't think it's a function of conspiracies. I think it is an issue
of many manufacturers trying to get every possible feature and
configuration supported before the first release is announced.
In some cases, the delays are specific to extremely high-end servers
such as 32 processor SMP clusters. In some cases the bugs and race
conditions are specific to these massively parallel clusters.
Meanwhile, back at the ranch, there are dozens of workstations
and appliances that will miss their market window.
It may be necessary for Linus and team to prioritize the features,
and get a reliable version out as quickly as possible.
Again, if one is to speak of the "Open Source Way", that would be
to release "early and often". Furthermore, there was no such thing
as a "beta" release. There were "attic" projects or "alpha" projects,
but very often the transition from what the industry would call a
"beta" release to "production" would have been 2-3 revisions focused
on bug fixes, followed by incremental releases to add features and
fix bugs. In many cases, the patches are even released separately.
> > --
> > Rex Ballard -
>
> Vacuo
> --
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>
>
--
Rex Ballard - VP I/T Architecture
Linux Advocate, Internet Pioneer
http://www.open4success.com
Linux - 60 million satisfied users worldwide
and growing at over 9%/month! (recalibrated 10/23/00)
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
** FOR YOUR REFERENCE **
The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:
Internet: [EMAIL PROTECTED]
You can send mail to the entire list (and comp.os.linux.advocacy) via:
Internet: [EMAIL PROTECTED]
Linux may be obtained via one of these FTP sites:
ftp.funet.fi pub/Linux
tsx-11.mit.edu pub/linux
sunsite.unc.edu pub/Linux
End of Linux-Advocacy Digest
******************************
