Linux-Advocacy Digest #371, Volume #35 Mon, 18 Jun 01 18:13:07 EDT
Contents:
Re: Linux penetration MUCH lower than previously claimed (.)
Re: Linux inheriting "DLL Hell" (Craig Kelley)
Re: Linux inheriting "DLL Hell" (Craig Kelley)
Re: PC power switch wont shut down Windows (Peter Hayes)
Re: More micro$oft "customer service" (The Ghost In The Machine)
Re: The beginning of the end for microsoft (Bernd Paysan)
Re: PC power switch wont shut down Windows (Peter Hayes)
Re: Why homosexuals are no threat to heterosexuals ("Aaron R. Kulkis")
Re: Linux inheriting "DLL Hell" (Craig Kelley)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (.)
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: Linux penetration MUCH lower than previously claimed
Date: 18 Jun 2001 21:39:47 GMT
In comp.os.linux.advocacy Jon Johansan <[EMAIL PROTECTED]> wrote:
> "." <[EMAIL PROTECTED]> wrote in message
> news:9gj5b5$c0v$[EMAIL PROTECTED]...
>> In comp.os.linux.advocacy Jon Johansan <[EMAIL PROTECTED]> wrote:
>>
>> > "." <[EMAIL PROTECTED]> wrote in message
>> > news:9gdtug$kt1$[EMAIL PROTECTED]...
>> >> In comp.os.linux.advocacy Jon Johansan <[EMAIL PROTECTED]> wrote:
>> >>
>> >> > "." <[EMAIL PROTECTED]> wrote in message
>> >> > news:9gdd68$e1$[EMAIL PROTECTED]...
>> >> >> In comp.os.linux.advocacy Jon Johansan <[EMAIL PROTECTED]> wrote:
>> >> >>
>> >> >> > "JS \ PL" <[EMAIL PROTECTED]> wrote in message
>> >> >> > news:[EMAIL PROTECTED]...
>> >> >> >> I wasn't aware that netcraft was counting physical servers. When
> did
>> >> > that
>> >> >> >> start happening?
>> >> >> >> The way they count has nothing to do with server market share.
>> >> >> >> No more than counting houses shows the amount of cities.
>> >> >>
>> >> >> > Netcraft has never claimed nor is it even capable (or anyone for
> that
>> >> >> > matter) of counting physical servers.
>> >> >>
>> >> >> > So, a mom&pop ISP running a single BSD box using Apache with 2000
>> >> > virtual
>> >> >> > hosts (those little 5 meg sites that joes diner and franks car
> repair
>> >> > puts
>> >> >> > up their one or two pages created in dreamweaver or frontpage)
>> >> >>
>> >> >> Apparently youve never had to deal with such a box. Heres an
> example:
>> >> >>
>> >> >> I can run 3500 of those websites *easily* on a dual PIII 600 w/1 gig
> of
>> >> > ram
>> >> >> and freebsd 4.3.
>> >>
>> >> > Um, thank you. You've even further proven my point. 3500 of _those_
>> > sites
>> >> > easily - I believe that because I've seen it. Some are name only
> virtual
>> >> > hosts. Only 3500?? I've seen more.
>> >>
>> >> So have I, I was talking about *small* machines, dipshit.
>>
>> > wow - what a come back, of course it's the small ones - but who cares if
>> > your box can run 3500 or 5000 single page sites?
>>
>> Apparantly youve never purchased for a web company. I have, and those
> numbers
>> mean quite alot.
> You'r right, I never purchased for a "web company" because at my company we
> have people to do that. We don't have the system administrators doing the
> buying, only the recommending.
Then you doubltessly make lots of mistakes. At my company, my word was final
ultimately on what hardware was purchased, case closed. It was done that way
on purpose; because the people who USED to do the buying there (much like
your own, apparantly) couldnt tell their asshole from a well.
> Must be a pretty small operation, do you also
> make the coffee? When I said "who cares" it wasn't in terms of money it was
> in context with capacity/managability alone. But I'm sure you knew that and
> changed the topic/direction.
The more you type, the less expertise you display actually.
>>
>> >>
>> >> >>
>> >> >> IIS cannot handle 3500 websites, no matter what kind of hardware its
>> >> > running
>> >> >> on, and never has been able to. Microsoft likes you to pile em on
>> >> > lightly,
>> >> >> or buy one box per site.
>> >>
>> >> > "That is completely untrue. Here is a little tool from MS that will
>> > assist
>> >> > you in creating and managing up to about 5000 virtual hosts on a
> single
>> >> > server (Scalable Hosting Solutions):
>> >>
>> >> > http://www.microsoft.com/TechNet/IIS/shsover.asp
>> >>
>> >> Which doesnt, and has NEVER worked. Never actually tried it, have you?
>>
>> > Personally no.
>>
>> Didnt think so.
> Because our clients don't like to share a single box with 5000 others.
> because our clients typically have a much higher load than you would put on
> a box with 5000 others. If you didn't work for a mom&pop isp with a few
> boxes spuing $14.95/month 5 megs free sites you'd know how this worked.
Sweetheart, the last company I worked for had more than 100,000 *access*
customer alone. Count on another 80K web customers. Count on half of those
being business. And thats not counting leased lines, satellite uplinking,
telephony, etc, etc, etc.
>>
>> > Someone I've asked about it, yes. Over 5000? no, about 2700
>> > is what he is running on a single box but they sure are tiny dipship
>> > accounts that's for sure...
>>
>> Hardly comparable. I'm talking about thousands of LARGE, BUSY SITES.
> Oh really. So in your mind it's a nice safe thing to put "thousands" of
> "LARGE, BUSY" sites on a single box eh? They can't be very large sites if
> you can put thousands on a single box and not have any performance issues
> not related to the os or HTTPD choice. Then again, when I think "LARGE,
> BUSY" I think in terms of saturating T3s ... each - doing e-commerce.
And right there, youve just absolutely killed your point with the buzzword
"e-commerce". Having built dozens of machines for 'e-commerce' sites, I can
say that beyond the shadow of a doubt, the reason why they are consistently
such heavy loads on hardware is because the people who write them have
NO IDEA WHAT THEYRE DOING.
Heres a handy example: why do with java what you can do with frames?
>>
>> > just like what the majority of tiny ISPs host on
>> > their apache general web account boxes.
>>
>> There are hardly any tiny ISP's left, first of all. Most of them were
> purchased
>> by bigger fish.
> um... yea...
Its true. Ive been in this end of the industry for 15 years, and I know
exactly what im talking about.
>>
>> And of the bigger fish, *all* of them do *most* of their webhosting with
> some
>> kind of unix, period. Every once in a while youll see a big company that
>> keeps an NT webserving cluster around for neat things like Frontpage
> extension,
>> but thats becoming more and more a rarity.
> That's funny. You are funny you know that? Oh yea, no NT servers anywhere...
> we must be imagining them. Then again, who cares what little ISPs do with
> their white-box servers. I'm much more interested in branded servers in the
> enterprise myself.
Earthlink is a little isp. Neat.
Enterprise? You havent a CLUE as to what that is all about, sweetie. When I
talk about enterprise, I'm talking about hundreds of machines with thousands
of processors tied into drive arrays that you havent even HEARD of.
>>
>> > These are not figures or specs to be
>> > proud of on EITHER platform - you do realize that? I'm sure you don't -
> you
>> > seem to think that bigger is always better.
>>
>> Having experience in the field (which you clearly do not have), I know
> exactly
>> when bigger is better: when it saves money, period.
> ya ya, yours is bigger than his. Sure... but is his penetrated like yours?
> sheesh... how empty. "I'm experienced and you are not, nah nah nah" ...
> silly. instead of waving your dick around, how about proving some points?
> Never mind, facts would slow you down.
Ive been proving points this entire time, youve simply been ignoring them---
or not understanding them. I begin to suspect the latter, and that you
simply arent very intelligent.
>>
>> >>
>> >> Try running 500 high traffic coldfusion sites on one W2K box of ANY
> size.
>>
>> > Coldfusion sites, nope, never did run that many on a single box. Got me
>> > there. Don't have any idea how they'd do.
>>
>> Terribly.
> Which is very likely why we don't - DOH!
Which is also likely why I ween large companies off of W2K as much as possible
and as soon as I can in every single case. Coldfusion works much better
under solaris.
>>
>> > But then again, who'd want to run
>> > 500 high traffic sites on a single box anyway?
>>
>> A company that wants to spend less money on hosting so that it can impress
>> investors with its efficiency and profitability. That would be all of
> them,
>> these days.
> "Like to" and "should do" are differences reconciled between bean counters
> and system administrators. Sure, the accountant would love for us to have
> one single server that runs absolutely everything over a "unlimited monthly
> access" 56K modem account - but... eh? DOH! You really missed the point
> didn't you...
Sweetie, you are not a system administrator.
>>
>> > Ever heard of load balancing
>> > and not putting all your eggs in one basket?
>>
>> Yes. Thats why its a handy idea to build out linux clusters (virtually
>> on big iron, or physically on small iron).
> So why don't you - why put 500 "high traffic" (still thinking you've got a
> different defination for this than me) sites on a single box?
Because it costs much, much less---without any loss of stability, speed or
functionality. In short, it just plain *makes sense*.
>>
>> > Sounds like someone is trying
>> > desperately to improve a slim profit margin and cheating his
> customers... If
>> > my "high traffic" cold fusion site was on a server shared by 500
> others - I
>> > would be VERY pissed.
>>
>> You would have no idea, actually. None at all.
> again, I think you don't know what you are talking about.
I think its just the reverse actually, whoever is paying you to take care
of their computers (if anyone is at all) has made a terrible mistake.
>>
>> >>
>> >> Never done that before either, have you?
>>
>> > Nope but I'm sure I could name something you've never done before and it
>> > would prove... that you've never done it before. so?
>>
>> What you never doing it before proves is that you have no experience with
>> the subject. I on the other hand, have mounds of experience with the
> subject;
>> and I can say unquestionably that you are absolutely incorrect.
> Hmm... so this is how debate works, ok, lemme try:
> I, on the other hand, have mounds of experience with the subject and I can
> say without reservation that you are in fact the clueless one.
You have experience with what, exactly?
>>
>> >>
>> >> > Depending on the application, a single IIS 5.0 server can host up to
>> > 5,000
>> >> > sites due to the amount of storage required in the Metabase for each
>> >> > additional site.
>> >>
>> >> Which is generally seen (even by microsoft engineers, ask paul salada)
> to
>> >> be the biggest braindeath of IIS. Next to its allowance out of the box
> of
>> >> random writings to the registry of course.
>>
>> > Why not have Paul post his comment here then?
>>
>> Ah, you do not know any microsoft engineers. Tell me, do you know any
>> MCSEs at least? Any one of them with half a brain in their head will
>> gladly back me up on this. Dozens and dozens already have.
> Dozens and dozens - woo hoo, care to add some testimonials too? "I used
> linux and gained 30 lbs overnight on the amazing mandrake diet" And these
> mean what?
I would, but that would mean that id have to admit something that I am
very embarrassed about.
> The fact that I do not know Paul Salada well enough to speak on his behalf
> is meaningless. I'm doubtful he would like you to speak on his behalf
> either.
Yes, which is why I havent.
> MCSEs are human and some are smart and some are not.
Agreed. Except that *most* of them are stupid, and only a *few* are smart.
> Some actually
> have field experience and some do not. But I can't speak for others or what
> they might or might not say. I can only speak from my own experience in a
> very large datacenter in Florida where we've had excellent success with MS
> products. So, I talk about what I've seen and done myself. We don't run
> 1000s of websites on a single box because we don't have to nor need to but
> not because we can't. And we HAVE tested running quite a large number of
> virtual hosts and haven't encounter any problems.
I'm running 41 thousand virtual hosts on a large freebsd box at the moment.
Are you running that many?
> You can live in denial of
> these facts but they are facts nonetheless. I see no reason why you have to
> behave as if you are the only person in the world that has ever done
> anything right and only your way... weird.
I'm simply pointing out that you have no position at all in which to talk about
linux or any other flavor of unix, because you have clearly never used any of
them and do not know what youre talking about in their concern.
=====.
--
"George Dubya Bush---the best presidency money can buy"
---obviously some Godless commie heathen faggot bastard
------------------------------
From: Craig Kelley <[EMAIL PROTECTED]>
Subject: Re: Linux inheriting "DLL Hell"
Date: 18 Jun 2001 15:43:05 -0600
drsquare <[EMAIL PROTECTED]> writes:
> On 18 Jun 2001 08:56:14 -0600, in comp.os.linux.advocacy,
> (Craig Kelley <[EMAIL PROTECTED]>) wrote:
>
> >drsquare <[EMAIL PROTECTED]> writes:
>
> >> >Name one.
> >>
> >> xcruise.
> >
> >In stable.
>
> Not there.
>
> >> parted.
> >
> >In stable.
>
> Not there.
>
> >> sc.
> >
> >In stable.
>
> Not there.
>
> >> gnotepad.
> >
> >In stable
>
> Not there.
Look again; perhaps you need to `apt-get update` or you screwed up
your sources.list.
http://packages.debian.org/stable/allpackages.html
--
It won't be long before the CPU is a card in a slot on your ATX videoboard
Craig Kelley -- [EMAIL PROTECTED]
http://www.isu.edu/~kellcrai finger [EMAIL PROTECTED] for PGP block
------------------------------
From: Craig Kelley <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: Linux inheriting "DLL Hell"
Date: 18 Jun 2001 15:46:22 -0600
"Seán Ó Donnchadha" <[EMAIL PROTECTED]> writes:
> "Craig Kelley" <[EMAIL PROTECTED]>:
> > >
> > > Actually, Microsoft's OS provides versioning support for all
> > > executables, including shared libraries. See the VS_VERSIONINFO
> > > resource. But again, that doesn't solve the problem. See below.
> >
> > It does little good if visual studio supports versioning (it's a good
> > thing it does, though) -- the convention is to overwrite the filename
> > with the same filename and to link against filenames.
>
> First of all, it isn't DevStudio that supports versioning; it's the
> OS. The OS provides file installation APIs that check the
> aforementioned version resources when installing files.
Translation: If a newer library is present it throws a flag. It
doesn't let you install both.
> The OS also provides an extremely comprehensive standard
> installation system (Windows Installer) that also works with these
> resources. Second, the convention is to overwrite only when the
> revision is minor, and to create a new filename when the revision is
> major. This is almost exactly the same as the Unix convention,
> except that the Unix symbolic link allows the older minor revision
> to remain on the disk, unused.
Explain that to Microsoft then; I've seen MFC42.DLL and the file info
says it's 'version 5.xxx'.
> > > Not only do you not know anything about Windows (see above), but
> > > you're also sadly mistaken to think that versioning solves the
> > > problem. It doesn't. It doesn't even come close. The simple example
> > > is as follows. You install libfoo-1.1. You install appfoo, which
> > > requires libfoo-1.1 and inadvertently relies on a bug therein. You
> > > then get appbar, which requires libfoo-1.2. You install libfoo-1.2,
> > > which fixes the aforementioned bug. You install appbar and
> > > everything seems fine. Then you run appfoo, and it breaks. Don't
> > > tell me it's appfoo's author's fault, because blame is
> > > irrelevant. This kind of shit happens in the real world. The point
> > > is that versioning is not a silver bullet by any stretch.
> >
> > No, but it does solve the *more common* case that goes like this:
> >
> > o Install appfoo which uses libbar-1.1
> > o Install appmoo which uses libbar-2.0
> >
> > They both co-exist just fine. I never said UNIX version was perfect,
> > I'm just saying that its *better than what Windows offers*.
> >
>
> Then, again, you don't know what you're talking about. Windows works
> exactly the same way. Wordpad requires MFC 3.x. Some other app
> requires MFC 4.x. They both co-exist just fine. This isn't due to
> some all-powerful versioning scheme. Just like Unix, it works simply
> because major revisions get different filenames (MFC30.DLL vs
> MFC40.DLL).
I must be imagining it then. Sorry for the misunderstanding.
--
It won't be long before the CPU is a card in a slot on your ATX videoboard
Craig Kelley -- [EMAIL PROTECTED]
http://www.isu.edu/~kellcrai finger [EMAIL PROTECTED] for PGP block
------------------------------
From: Peter Hayes <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy,alt.destroy.microsoft
Subject: Re: PC power switch wont shut down Windows
Date: Mon, 18 Jun 2001 22:45:48 +0100
Reply-To: [EMAIL PROTECTED]
On Sun, 17 Jun 2001 17:29:16 -0700, "Stephen S. Edwards II"
<[EMAIL PROTECTED]> wrote:
> If you could give some hardware specs, I'm sure we
> might be able to give you some assistance of some
> sorts.
AMD Thunderbird 1.33 GHz with a big fat cooler
Asus A7V133 motherboard *not* overclocked and with all Asus's updates
256 Mb Pc133 ram
20Gb UDMA100 hdd
Nvidia TNT2 video card
100mbps network card
Peter
------------------------------
From: [EMAIL PROTECTED] (The Ghost In The Machine)
Crossposted-To: comp.sys.mac.advocacy
Subject: Re: More micro$oft "customer service"
Date: Mon, 18 Jun 2001 21:37:14 GMT
In comp.os.linux.advocacy, Dan
<[EMAIL PROTECTED]>
wrote
on 17 Jun 2001 09:39:04 -0500
<[EMAIL PROTECTED]>:
>In article <[EMAIL PROTECTED]>,
> Woofbert <[EMAIL PROTECTED]> wrote:
>
>
>> >It's actually very useful.
>> > Perhaps you should withold your opinion until you've actually used it.
>> >
>> > Dan
>>
>> Who controls the content of these added links?
>
>Who cares? Who "controls the content" of *any* link on *any* page?
The HTML server gives a suggestion for the link, but it's ultimately
up to the browser to present said link in a usable form -- both IE
and Netscape have options for downloading a file, as opposed to displaying
it, using the right mouse button instead of the left, and a menu pick,
for example. Or one can just click on it and view it.
Suppose, for example, Somebody'sBrowser(tm) decided to implement
a URL => URL mapping, so that any displayed <A HREF=> tag would point,
instead to its intended target, a constructed target based on the mapping?
For example, http://www.microsoft.com/products/index.asp would be
mapped to http://somewhere.nasty.com/stupid/products/index.asp, if
the mapping specifies
http:///www.microsoft.com/ => http://somewhere.nasty.com/stupid/
as one of its elements. (Some issues ensue regarding names vs.
IP addresses, rotating DNS, forwards, and such; these are secondary.)
Suppose further that this mapping is controlled, not by the customer,
but by the browser company. Would that be illegal, unethical, or
just silly?
Suppose further that this mapping is not keyed on <A HREF> tags,
but on certain keywords (e.g., "Microsoft" might get a squiggly line
under it, pointing to the company's main website; "Oracle Database"
might get a squiggly line pointing to something relating to, well,
Oracle databases -- such as http://technet.oracle.com).
I'm not saying that this is what Smart Tags are doing (I have no idea,
but from what I've heard, they are "somewhat like hyperlinks except
not really, keying off keywords in the text as opposed to tags in the
HTML, and having a different popup menu"). But I see some
possible issues here.
Microsoft most likely wants to enhance the experience of web browsing
for the end user. That's fine. However, it's not clear whether Microsoft
also wants to *control* the experience. That's not.
>
>Dan
--
[EMAIL PROTECTED] -- insert random misquote here
EAC code #191 3d:00h:08m actually running Linux.
This is my other .sig.
------------------------------
From: Bernd Paysan <[EMAIL PROTECTED]>
Crossposted-To: comp.arch
Subject: Re: The beginning of the end for microsoft
Date: Mon, 18 Jun 2001 23:51:22 +0200
Michael Marion wrote:
> I've experienced this myself. Try doing something like upgrading the mobo and
> CPU but leaving everything else the same. Windows will go completely wacko
> upon reboot, loading drivers for things it already had, asking for drivers in
> an impossible manner (like asking for the IDE driver from the win CD when your
> CD-ROM is attached to that very same IDE channel).
Some years ago, I changed everything in my computer, excluding hard disk
and case (both mistakes: The case died first, the HD a bit later ;-). At
that time, I had three OSes, OS/2, Windows 95, and Linux. Before
changing, I installed the necessary drivers for OS/2, compiled a new
Linux kernel with drivers for old and new system, and did not find
anything useful to do in Windows (obvious, it's supposed to be P&P,
isn't it ;-). You guess. Linux did only boot faster. OS/2 also worked
out of the box. Windows did not recover this change, it required a
complete reinstall. The difference is: With OS/2 and Linux, you had to
fiddle a bit, but you could at least do it (and for Linux even so far
that the same system booted on both machines). With Windows, no chance.
--
Bernd Paysan
"If you want it done right, you have to do it yourself"
http://www.jwdt.com/~paysan/
------------------------------
From: Peter Hayes <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy,alt.destroy.microsoft
Subject: Re: PC power switch wont shut down Windows
Date: Mon, 18 Jun 2001 22:51:25 +0100
Reply-To: [EMAIL PROTECTED]
On 18 Jun 2001 21:28:07 GMT, [EMAIL PROTECTED] (.) wrote:
> In comp.os.linux.advocacy Jon Johansan <[EMAIL PROTECTED]> wrote:
>
> > "Peter Hayes" <[EMAIL PROTECTED]> wrote in message
> > news:[EMAIL PROTECTED]...
> >> On Sun, 17 Jun 2001 10:25:39 -0700, "Stephen S. Edwards II"
> >> <[EMAIL PROTECTED]> wrote:
> >>
> >> > "LShaping" <[EMAIL PROTECTED]> wrote in message
> >> > news:[EMAIL PROTECTED]...
> >> >
> >> > > >> Shut Down... path is for? Thanks to Microsoft for extending
> >> > > Windows slimey tenticles to my power supply. I can't wait to find out
> >> > > what "PCHealth" is going to do to my other hard disk partitions.
> >> >
> >> > *sigh*
> >> >
> >> > I find it quite amazing that we live in an age where
> >> > people actually complain about convenience. Go fig.
> >>
> >> If I use Start -> Shutdown -> Shutdown my machine shuts down and powers
> > off
> >> so fast that at the next switch on Scandisk goes through its thing and
> > gives
> >> me a row. (Win98).
> >>
> >> On windows 2000 it shagged it good and proper.
> > "winnt\system32\config\system
> >> is missing or corrupted".
> >>
> >> Reiserfs seems to have survived so far...
> >>
> >> I now select restart and switch the UPS off when the reboot starts.
>
> > I love stories likes these from the (your pick) a) incompetent or b) liars
>
> I actually had the same problem with winME on my asus ATA/266/athlon thunderbird
> 1.33ghz board running 2 ATA/100 60 gig IBM drives. According to microsoft, this
> happens because the motherboard I have responds to the signaling to spin down its
> drives WAY too fast, and spins them down before the OS is finished writing to
> them. There is a fix on the software updates page for winME concerning this
> issue.
That seems similar to my setup and my problem, thanks.
> Now then jan, what exactly were you saying about liars or incompentants?
Just ignore him...
Peter
------------------------------
From: "Aaron R. Kulkis" <[EMAIL PROTECTED]>
Crossposted-To: soc.men,soc.singles,alt.fan.rush-limbaugh
Subject: Re: Why homosexuals are no threat to heterosexuals
Date: Mon, 18 Jun 2001 17:56:35 -0400
SSunbird wrote:
>
> Rich Soyack wrote:
>
> > When one asks sensible questions about AIDS all one gets is dogma.
>
> it ain't dogma, it's science. there really are studies that show
> that protection works.
The best protection is not associating with fags.
>
> Pt
--
Aaron R. Kulkis
Unix Systems Engineer
DNRC Minister of all I survey
ICQ # 3056642
L: This seems to have reduced my spam. Maybe if everyone does it we
can defeat the email search bots. [EMAIL PROTECTED] [EMAIL PROTECTED]
[EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED]
[EMAIL PROTECTED]
K: Truth in advertising:
Left Wing Extremists Charles Schumer and Donna Shalala,
Black Seperatist Anti-Semite Louis Farrakhan,
Special Interest Sierra Club,
Anarchist Members of the ACLU
Left Wing Corporate Extremist Ted Turner
The Drunken Woman Killer Ted Kennedy
Grass Roots Pro-Gun movement,
J: Other knee_jerk reactionaries: billh, david casey, redc1c4,
The retarded sisters: Raunchy (rauni) and Anencephielle (Enielle),
also known as old hags who've hit the wall....
I: Loren Petrich's 2-week stubborn refusal to respond to the
challenge to describe even one philosophical difference
between himself and the communists demonstrates that, in fact,
Loren Petrich is a COMMUNIST ***hole
H: "Having found not one single carbon monoxide leak on the entire
premises, it is my belief, and Willard concurs, that the reason
you folks feel listless and disoriented is simply because
you are lazy, stupid people"
G: Knackos...you're a retard.
F: Unit_4's "Kook hunt" reminds me of "Jimmy Baker's" harangues against
adultery while concurrently committing adultery with Tammy Hahn.
E: Jet is not worthy of the time to compose a response until
her behavior improves.
D: Jet Silverman now follows me from newgroup to newsgroup
...despite (C) above.
C: Jet Silverman claims to have killfiled me.
B: Jet Silverman plays the fool and spews out nonsense as a
method of sidetracking discussions which are headed in a
direction that she doesn't like.
A: The wise man is mocked by fools.
------------------------------
From: Craig Kelley <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: Linux inheriting "DLL Hell"
Date: 18 Jun 2001 15:56:17 -0600
"Seán Ó Donnchadha" <[EMAIL PROTECTED]> writes:
> "Craig Kelley" <[EMAIL PROTECTED]> wrote in message
> news:[EMAIL PROTECTED]...
> > > >
> > > > So? An application can link against minor revisions.
> > >
> > > Well yeah, except they never do, because the whole point of using
> > > shared libraries is to allow your app to inherit library bug fixes
> > > in the field. If you're going to link against a minor revision, you
> > > might as well link statically.
> >
> > ... unless there are a whole suite of programs using said minor
> > version (ahem, like ORBit with GNOME perhaps?).
> >
> > [...]
> >
> > It happens all the time; especiall with pre-1.0 libraries.
> >
>
> There is absolutely no compelling reason to link against a minor
> revision unless you assume that the library is full of bugs and
> won't be back-compatible when revised. Only then (and only if static
> linking is impractical because you'd be pulling in too much bloat)
> does linking against a minor revision make sense. That would explain
> your examples with GNOME and pre-1.0 libraries. But these are freak
> cases, not the accepted Unix norm, which is to use shared libraries
> the way they were intended to be used.
You said "they never do"; I'm providing proof to the contrary. Just
because you think they ought not do so in order to bolster your
misguided defense of Windows' DLL hell doesn't make it so.
Another example (I guess I'm using "freak" software, hmm?): Oracle 8i
doesn't like libc-2.2.x. No problem, I install libc-2.1.3 alongside
2.2.2 and happily use Oracle from there. I re-linked it *to an older
version of libc* after installation.
In short: Yes, programs can and do link against various minor version
of a library. Yes, this is a good thing. No, Windows can't do it
without playing around with paths and having an installer that knows
about such things.
> > > > I must be crazy then, because I did this exact same thing with libc
> > > > just last week to install Oracle 8.1.7 on our new ten-thousand dollar
> > > > box.
> > >
> > > Did what?
> >
> > I Installed an earlier minor revision of glibc to make Oracle happy
> > under RedHat 7.0 (which has a known buggy glibc).
>
> Did you have to clobber the symbolic link, or is Oracle linked against a
> minor glibc revision?
Neither. I chose to link it against 2.1.3. You know, dynamic linking
being dynamic and all that. (to be fair, Oracle facilitates this by
allowing one to specify the version to use -- but it is completely
doable to relink by hand if absolutely necessary)
--
It won't be long before the CPU is a card in a slot on your ATX videoboard
Craig Kelley -- [EMAIL PROTECTED]
http://www.isu.edu/~kellcrai finger [EMAIL PROTECTED] for PGP block
------------------------------
** 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 by posting to comp.os.linux.advocacy.
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
******************************
