Linux-Advocacy Digest #626, Volume #31 Sat, 20 Jan 01 23:13:03 EST
Contents:
Re: A salutary lesson about open source (The Ghost In The Machine)
Re: Windows curses fast computers (spicerun)
Re: A salutary lesson about open source (Charlie Ebert)
Re: NT is Most Vulnerable Server Software (T. Max Devlin)
Re: NTFS Limitations (Was: RE: Red hat becoming illegal?) (The Ghost In The Machine)
Re: NT is Most Vulnerable Server Software (T. Max Devlin)
Re: NTFS Limitations (Was: RE: Red hat becoming illegal?) (T. Max Devlin)
Re: Is Bill Gates MAD?!?!? (Marten Kemp)
Re: NTFS Limitations (Was: RE: Red hat becoming illegal?) (T. Max Devlin)
Re: Windows curses fast computers (Peter =?ISO-8859-1?Q?K=F6hlmann?=)
Re: "The Linux Desktop", by T. Max Devlin (T. Max Devlin)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (The Ghost In The Machine)
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: A salutary lesson about open source
Date: Sun, 21 Jan 2001 03:28:49 GMT
In comp.os.linux.advocacy, J Sloan
<[EMAIL PROTECTED]>
wrote
on Sun, 21 Jan 2001 01:15:11 GMT
<[EMAIL PROTECTED]>:
>Chad Myers wrote:
>
>> "J Sloan" <[EMAIL PROTECTED]> wrote in message
>> news:[EMAIL PROTECTED]...
>>
>> > And you carefully snipped my references to amazon, google, deja,
>> > which, like yahoo and other large sites, use apache.
>>
>> So a handful of large sites use Apache, so what? Many more use IIS.
>
>Sorry kid, you're just not believable anymore -
>
>> Also, Yahoo is "unknown on FreeBSD".
>
>It's apache -
They hide it well, and it appears slightly misconfigured, then.
A HEAD request generated *lots* of glop.
$ telnet www.yahoo.com
Trying 204.71.200.74...
Connected to www.yahoo.akadns.net.
Escape character is '^]'.
At this point, I typed in
HEAD / HTTP/1.0
and it responded with
HTTP/1.0 200 OK
Content-Length: 16273
Content-Type: text/html
<html><head><title>Yahoo!</title><base href=http://www.yahoo.com/>...
[rest of response snipped]
Now, I'd have to look, but doesn't RFC1945 say not to do this?
As far as I can tell, it's a proper request (although a number of
sites running IIS don't like it much; they tend to throw back
'400 Bad request' errors).
HTTP/1.1 isn't much better, and I doubt RFC2068 allowed them that
much leeway, anyway.
To be fair, I do note that the response is coming not from
Yahoo itself, but from someplace called 'www.yahoo.akadns.net',
which is probably an Akamai cacher, to help performance.
Maybe it's their fault. :-)
>
>jjs
>
--
[EMAIL PROTECTED] -- insert random misquote here
EAC code #191 1d:13h:52m actually running Linux.
Linux. The choice of a GNU generation.
------------------------------
From: spicerun <[EMAIL PROTECTED]>
Subject: Re: Windows curses fast computers
Date: Sat, 20 Jan 2001 21:39:11 -0600
[EMAIL PROTECTED] wrote:
> In article <J4ma6.185472$[EMAIL PROTECTED]>,
> Pete Goodwin <[EMAIL PROTECTED]> wrote:
>
>> [EMAIL PROTECTED] wrote:
>>
>>>> It's supposed to WAIT on the fucking interrupt you god damn idiot!
>>>
>>> This is a general problem these days. Most software engineers just don't
>>> know how the hardware works. In the old days we learnt what the hardware
>>> could do because we were coding in assembler (even toggling in machine on
>>> occasions:-). Now software engineers are so isolated from the hardware
>>> they resort to adding delay loops in their software instead of finding out
>>> how the hardware works. It is clear that Erik is of this sad generation.
>>> He might know how windows works but hasn't a clue how it interacts with
>>> the hardware. It appears that most Microsoft developers don't know how it
>>> interacts with hardware.
>>
>> I know how it works, and I'm a Microsoft developer.
>>
>> Ah, but then I started as a Electronics Engineer at EMI, then switched to
>> software. That's one reason in my favour when I took my current job writing
>> audio device drivers.
>
>
> I must say that some of the best software engineers that I have worked with
> were electronic engineers who made the transition to softwae. What went wrong
> in your case?
I suspect, in his case, he wasn't an Electronics Engineer long enough to
really understand devices (and got very frustrated)
....in fact, that's probably the reason he switched to Software, so that
he wouldn't have to understand Hardware.
------------------------------
From: [EMAIL PROTECTED] (Charlie Ebert)
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: A salutary lesson about open source
Date: Sun, 21 Jan 2001 03:38:24 GMT
In article <gfsa6.4989$[EMAIL PROTECTED]>, Chad Myers wrote:
>
>"Cliff Wagner" <[EMAIL PROTECTED]> wrote in message
>news:[EMAIL PROTECTED]...
>> On Sun, 21 Jan 2001 00:35:20 GMT, Chad Myers typed something like:
>> >
>> >"Cliff Wagner" <[EMAIL PROTECTED]> wrote in message
>> >news:[EMAIL PROTECTED]...
>>
>> [SNIP]
>>
>> >> Probably because you make up your own analysis as opposed
>> >> to actually addressing what is on the page you are
>> >> so happy to quote:
>> >> -------- BEGIN QUOTE FROM URL CHAD LOVES SO MUCH -----
>> >> The 1999 Fortune 500 list of companies ranks the top corporations
>> >> in the United States. We expected the results to be dramatically
>> >> different than the Netcraft results because upper management in
>> >> big business generally don't understand open source software (OSS).
>> >> They often forbid the use of OSS because they confuse it with the
>> >> FreeWare and ShareWare from the 1980s. They're not aware that
>> >> the quality of Apache rivals the commercial products and surpasses
>> >> the commercial products in terms of flexibility and functionality.
>> >> ------- END QUOTE ----
>> >> That basically say that the Fortune500 is skewed due to business
>> >> politics (in case you have problems comprehanding it).
>> >> And are you ever going to get me a number of "all these unimportant
>> >> low-traffic sites" that run on apache? I know of a lot that
>> >> run on IIS as well. Hell, i've designed some for both platforms
>> >> (gotta love companies that believe, if you build it they will come)
>> >
>> >
>> >So now what biznix says in opin is the bible? The facts speak for
>> >themselves. You can make up all sorts of fairy-tales about business
>> >politics to make up for Apache's poor showing, but it's just
>> >that, fairy tales.
>> >
>> >Please show conclusive proof that the reason Fortune 500 companies
>> >don't use Apache is because they think it's shareware. That's an
>> >assinine statement. That URL shows numbers, the rest is just all
>> >idle conjecture.
>> >
>> >-Chad
>>
>> In other words, your opinion is more valid then
>> those who actually did the study.
>> How nice.
>
>Please show me proof that Fortune 500 execs were making these
>kinds of decisions based on an OSS/shareware confusion.
>
>I'm just going on the plain facts. The numbers speak for themselves.
>
>It's pretty sad for you guys now. You're no longer trying to
>argue for Apache, you're trying to make excuses why its failing,
>and you're eager to use anyone else's conjecture (like the Biznix
>article) to support your excuses.
>
>-Chad
>
>
To even remotely suggest that websites such as Yahoo under FreeBSD
or Google.com under Linux clustering are NOT running apache and
are possibly running a Microsoft product running on any *nix machine
is absolutely insane.
Where are you going with this dribble Chad?
Charlie
------------------------------
From: T. Max Devlin <[EMAIL PROTECTED]>
Crossposted-To: alt.destroy.microsoft
Subject: Re: NT is Most Vulnerable Server Software
Reply-To: [EMAIL PROTECTED]
Date: Sun, 21 Jan 2001 03:40:54 GMT
Said [EMAIL PROTECTED] () in alt.destroy.microsoft on Sat, 20 Jan
[...]
>>Again, your knowledge seems sound, but your understanding is flawed.
>>There are no subnets 'reserved' for anything, unless you're dealing with
>>a full-blown firewall (which by nature breaks all those rules about how
>
> Or routers.
>
> There are certain addresses that aren't meant to be routed.
> To do so will cause name collisions.
Routers don't have anything to do with names; I'm not sure what a 'name
collision' is, or what mechanism you're trying to explain. The only
addresses that "aren't meant to be routed" are 127.0.0.1, and 0.0.0.0.
Perhaps you got something about DHCP a bit mangled. There was much ado
about how 0.0.0.0 needs to be handled concerning Cisco's "DHCP
forwarding".
>[deletia]
>
> No matter how you are connected to the net, there WILL be a
> local router somewhere that's not supposed to forward traffic
> with certain destination or source addresses.
>
> These are defined in RFC 1597.
Regardless of what is in RFC 1597, your previous statement was again
nonsensical. Routers that "are not supposed to forward traffic with
certain destination or source address" are typically called *filtering
routers*, and are actually indistinguishable from simple firewalls.
More complex firewalls (and routers which have more complex filtering or
firewall features themselves) also pay attention to the source or
destination ports, typically. "Real" firewalls are generally protocol
(application) specific, and can filter traffic related to certain
transactions to prevent logins and such.
So what's 1597 about?
--
T. Max Devlin
*** The best way to convince another is
to state your case moderately and
accurately. - Benjamin Franklin ***
------------------------------
From: [EMAIL PROTECTED] (The Ghost In The Machine)
Crossposted-To:
alt.destroy.microsoft,comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: NTFS Limitations (Was: RE: Red hat becoming illegal?)
Date: Sun, 21 Jan 2001 03:49:25 GMT
In comp.os.linux.advocacy, Chad Myers
<[EMAIL PROTECTED]>
wrote
on Sun, 21 Jan 2001 00:52:55 GMT
<Hpqa6.6224$[EMAIL PROTECTED]>:
>
>"The Ghost In The Machine" <[EMAIL PROTECTED]> wrote in message
>news:[EMAIL PROTECTED]...
>> In comp.os.linux.advocacy, Chad Myers
>> <[EMAIL PROTECTED]>
>> wrote
>> on Sat, 20 Jan 2001 16:14:48 GMT
>> <YPia6.2990$[EMAIL PROTECTED]>:
>> >
>> ><[EMAIL PROTECTED]> wrote in message
>> >news:[EMAIL PROTECTED]...
>> >> On Sat, 20 Jan 2001 06:38:55 GMT, Chad Myers
>> >> <[EMAIL PROTECTED]> wrote:
>> >> >
>> >> ><[EMAIL PROTECTED]> wrote in message
>> >> >news:[EMAIL PROTECTED]...
>> >> >> On Sat, 20 Jan 2001 04:25:52 GMT, T. Max Devlin
>> >> >> <[EMAIL PROTECTED]> wrote:
>> >> >> >Said [EMAIL PROTECTED] () in alt.destroy.microsoft on
>> >> >> >Sat, 20 Jan
>> >> >> >>On Fri, 19 Jan 2001 15:25:22 GMT, T. Max Devlin
>> >> >> >><[EMAIL PROTECTED]> wrote:
>> >> >> >>>Said Ayende Rahien in alt.destroy.microsoft on
>> >> >> >>>Fri, 19 Jan 2001 06:58:01
>> >> >> > [...]
>> >> >> >>>>I'm not sure exactly *what* you can put into a file to get
>> >> >> >>>>into that size.
>> >> >> >>>
>> >> >> >>>Precisely what they said about the 2 Gigabyte limit. ;-)
>> >> >> >>
>> >> >> >> Databases.
>> >> >> >
>> >> >> >A 'database' is not, by definition or even by convention,
>> >> >> >a single file.
>> >> >>
>> >> >> There's that magic word: "convention".
>> >> >>
>> >> >> That's all that separates a table spread across 10 files
>> >> >> and 5 physical disks from the video that for some
>> >> >> strange reason can't be similarly divided.
>> >> >
>> >> >Man, you must really have you head up your ass.
>> >>
>> >> No, I just don't see what the big deal is.
>> >>
>> >> I still don't, especially after you've broken down
>> >> the "process". The "need" to have a certain duration
>> >> of video in a single file is entirely arbitrary.
>> >
>> >Man, you really are dense. I've explained the obvious 4 or
>> >5 times now. When you digitize, it's easier to digitize
>> >the entire clip and work with it as a whole.
>>
>> It depends on the clip. Were I a moviemaker (I'm not), I would
>> not want to have a single gigantic file until the very end, during
>> final assembly; prior to that, I'd want scenes.
>
>Then you don't understand video processing.
Well, OK, I was characterizing the wrong problem. :-)
>Note: these are not
>feature-length movies, these are video lectures. When digitizing
>them to the computer (through firewire) stopping every 15 minutes
>to account for Linux's poor design is not an option.
Then don't use Linux. I'd say that's a no-brainer. :-)
Either that, or get an Alpha or PPC box which can handle
huge files, or rewrite the apps so that they can handle
long seeks -- of course, that option is only available if
one actually has the source code, but the recompile is simple
enough in that case:
g++ -D_FILE_OFFSET_BITS=64 -D__USE_FILE_OFFSET64 ...
assuming the application uses off_t for a seek key. Presumably
there's a FAQ on this somewhere, and it may involve quite a bit
of work depending on the app.
The limitation is not really in Linux, or even the file system.
However, most apps don't use off_t, they just use int, which
causes problems.
(Yes, I've tested this.)
>
>We have a farm of machines that just take in digital video and
>run for hours as they digitize every second of the video. Managing
>those every 15 minutes is not an option.
That would be a tricky problem, admittedly. I'm not sure how I'd
manage it, either (my contribution to video is a 1.7 gigabyte
hard disk on an Amiga Personal Animation Recorder -- and it's
*ancient*).
>
>-Chad
>
>
--
[EMAIL PROTECTED] -- insert random misquote here
EAC code #191 1d:14h:12m actually running Linux.
You were expecting something relevant down here?
------------------------------
From: T. Max Devlin <[EMAIL PROTECTED]>
Crossposted-To: alt.destroy.microsoft
Subject: Re: NT is Most Vulnerable Server Software
Reply-To: [EMAIL PROTECTED]
Date: Sun, 21 Jan 2001 03:52:28 GMT
Said [EMAIL PROTECTED] in alt.destroy.microsoft on Sun, 21 Jan
>In article <[EMAIL PROTECTED]>,
> T. Max Devlin <[EMAIL PROTECTED]> wrote:
>>
>> Said [EMAIL PROTECTED] () in alt.destroy.microsoft on Sat, 20 Jan
>>>On Sat, 20 Jan 2001 04:14:17 GMT, T. Max Devlin <[EMAIL PROTECTED]> wrote:
>>>
>>> Actually, there are several subnets that are reserved for
>>> local use. Assuming your local router is not malconfigured,
>>> the traffic you see on those subnets should be rather
>>> limited.
>>
>> Again, your knowledge seems sound, but your understanding is flawed.
>> There are no subnets 'reserved' for anything, unless you're dealing with
>> a full-blown firewall (which by nature breaks all those rules about how
>> IP networking works). There are such things as local subnets; those
>> that are directly connected on the transmission system(s) the computer
>> is connected to. But that doesn't inform anything but the routing
>> decision; which MAC address to put on the frame bearing the packet. It
>> doesn't have squat to do with the packet itself, nor certainly the
>> datagram (potentially fragments of datagrams, in fact) in the packet.
>
>You are completely wrong here. There are class A, B, C (to use the old
>naming convention) subnets that are reserved for private use only and
>must never be seen on the Internet (10.x.x.x and 192.168.x.x for
>example (plus the 172.x.x.x ones)).
This is a convention, not a rule of routing or the IP protocol.
>You clearly have no understanding
>of Internet routing and firewalls.
To be modest, my knowledge and understanding literally dwarfs your own.
>How is a firewall breaking all the
>rules? Do you know what a firewall is? Most firewalls allow address
>translation.
Indeed; first you ask a question, then you answer it. ;-)
BTW, the function of Network Address Translation (NAT) is potentially
related, but entirely independent and separate from firewalling. NAT is
also very similar to proxying, which is what "most firewalls" do.
>They use PAT to translate many internal addresses to a single
>external routable address.
Indeed; Proxy Address Translation.
>The good
>firewalls also allow 1 to 1 NAT.
No, a "good" firewall will leave NAT up to a NAT system, just as a NAT
leaves firewalling up to the firewall. Sure, in theory we could munge
all these things together. Have fun; its too braindead an idea for me
to waste my time on, personally.
>MAC addresses are used to send the packet
>to the next hop on its route.
Yes, MAC addresses are used to *transmit* the *frame*, containing the
packet, to the receiver. Note that whether the receiver is "the next
hop on its route" is one of those rules that you should fuck around with
lightly. The rule is, "you don't know, you can't know, and it doesn't
matter." Because routing doesn't have anything to do with
transmissions. So MAC addresses don't have anything to do with packets.
>When the packet arrives at a router connected
>to the destination host the MAC address is that of the destination host.
No, when the packet is *transmitted* *from* a router connected to a
transmission channel shared by the destination host, the destination
address of the *frame* bearing that packet is the MAC address of the
transceiver related to that host.
>May
>I suggst you purchase a copy of the late Mr. Stevens' books on TCP/IP. If
>you can understand and digest those then you may talk about Internet subnets
>and routing.
Been there, done that. Any further 'errors' which you find in my
explanations can be assumed to be things *you* don't understand; forgive
me, but my theories and understanding are rather advanced, and often
less knowledgeable people find them confusingly distinct from the
standardized explanations they've been taught previously. Feel free to
ask questions, I'll be happy to explain them; I love explaining this
stuff.
>>> Likewise, if you see such traffic coming to you inbound
>>> parts unknown it's and obvious danger signal.
>>
>> This is a mechanism for firewall software. IP (routing) software can't
>> be *allowed*, in fact, to pay attention to it, or you make IP nothing
>> but a virtualization, not a routing method, and that removes all
>> benefits from using it.
>
>Please give up talking about something that you are totally ignorant of.
>You obviously know nothing of how IP routing works and even less about
>firewalls.
Guffaw. And here I was just about to submit my resume to a local
company that needed a router and firewall person. What a shame I'll
have to forget all about my ten years of experience in networking and
the dozens of professional courses I've taught.
--
T. Max Devlin
*** The best way to convince another is
to state your case moderately and
accurately. - Benjamin Franklin ***
------------------------------
From: T. Max Devlin <[EMAIL PROTECTED]>
Crossposted-To:
alt.destroy.microsoft,comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: NTFS Limitations (Was: RE: Red hat becoming illegal?)
Reply-To: [EMAIL PROTECTED]
Date: Sun, 21 Jan 2001 03:54:02 GMT
Said Chad Myers in alt.destroy.microsoft on Sun, 21 Jan 2001 00:52:55
>"The Ghost In The Machine" <[EMAIL PROTECTED]> wrote in message
>news:[EMAIL PROTECTED]...
>> In comp.os.linux.advocacy, Chad Myers
>> <[EMAIL PROTECTED]>
>> wrote
>> on Sat, 20 Jan 2001 16:14:48 GMT
>> <YPia6.2990$[EMAIL PROTECTED]>:
>> >
>> ><[EMAIL PROTECTED]> wrote in message
>> >news:[EMAIL PROTECTED]...
>> >> On Sat, 20 Jan 2001 06:38:55 GMT, Chad Myers <[EMAIL PROTECTED]>
>> >wrote:
>> >> >
>> >> ><[EMAIL PROTECTED]> wrote in message
>> >> >news:[EMAIL PROTECTED]...
>> >> >> On Sat, 20 Jan 2001 04:25:52 GMT, T. Max Devlin <[EMAIL PROTECTED]>
>wrote:
>> >> >> >Said [EMAIL PROTECTED] () in alt.destroy.microsoft on Sat, 20 Jan
>> >> >> >>On Fri, 19 Jan 2001 15:25:22 GMT, T. Max Devlin <[EMAIL PROTECTED]>
>> >wrote:
>> >> >> >>>Said Ayende Rahien in alt.destroy.microsoft on Fri, 19 Jan 2001
>06:58:01
>> >> >> > [...]
>> >> >> >>>>I'm not sure exactly *what* you can put into a file to get into that
>> >size.
>> >> >> >>>
>> >> >> >>>Precisely what they said about the 2 Gigabyte limit. ;-)
>> >> >> >>
>> >> >> >> Databases.
>> >> >> >
>> >> >> >A 'database' is not, by definition or even by convention, a single
>file.
>> >> >>
>> >> >> There's that magic word: "convention".
>> >> >>
>> >> >> That's all that separates a table spread across 10 files
>> >> >> and 5 physical disks from the video that for some
>> >> >> strange reason can't be similarly divided.
>> >> >
>> >> >Man, you must really have you head up your ass.
>> >>
>> >> No, I just don't see what the big deal is.
>> >>
>> >> I still don't, especially after you've broken down
>> >> the "process". The "need" to have a certain duration
>> >> of video in a single file is entirely arbitrary.
>> >
>> >Man, you really are dense. I've explained the obvious 4 or
>> >5 times now. When you digitize, it's easier to digitize
>> >the entire clip and work with it as a whole.
>>
>> It depends on the clip. Were I a moviemaker (I'm not), I would
>> not want to have a single gigantic file until the very end, during
>> final assembly; prior to that, I'd want scenes.
>
>Then you don't understand video processing. Note: these are not
>feature-length movies, these are video lectures. When digitizing
>them to the computer (through firewire) stopping every 15 minutes
>to account for Linux's poor design is not an option.
>
>We have a farm of machines that just take in digital video and
>run for hours as they digitize every second of the video. Managing
>those every 15 minutes is not an option.
And I guess we're all supposed to take it on faith that these big
impressive machines cannot be told to do this automatically.
--
T. Max Devlin
*** The best way to convince another is
to state your case moderately and
accurately. - Benjamin Franklin ***
------------------------------
From: Marten Kemp <[EMAIL PROTECTED]>
Subject: Re: Is Bill Gates MAD?!?!?
Date: Sun, 21 Jan 2001 03:58:51 GMT
"Aaron R. Kulkis" wrote:
>
> Steve Mading wrote:
> >
> > Aaron R. Kulkis <[EMAIL PROTECTED]> wrote:
> > : Steve Mading wrote:
> > :>
> > :> Aaron R. Kulkis <[EMAIL PROTECTED]> wrote:
> > :> : Steve Mading wrote:
> > :> :> Specify, what is it that is illegal?
> > :>
<<snip>>
> >
> > : New Jersey and Colorado (?) have laws prohibiting "self serve" gas sales.
> >
<<snip>>
Maybe New Jersey but definitely not Colorado (at least as of a month
ago).
-- Marten Kemp
------------------------------
From: T. Max Devlin <[EMAIL PROTECTED]>
Crossposted-To:
alt.destroy.microsoft,comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: NTFS Limitations (Was: RE: Red hat becoming illegal?)
Reply-To: [EMAIL PROTECTED]
Date: Sun, 21 Jan 2001 04:00:34 GMT
Said [EMAIL PROTECTED] () in alt.destroy.microsoft on Sat, 20 Jan
>On Sat, 20 Jan 2001 19:10:17 GMT, T. Max Devlin <[EMAIL PROTECTED]> wrote:
[...]
>>Actually, Jedi, in this case, that's not true. A video stream is not at
>>all similar to a discrete data store. The very concept of 'structure'
>>is different, in fact, and there is, despite your claims, some
>
> For compressed digital video there is some truth to this.
>
> HOWEVER, compressed digital video is quite successfully
> spread across multiple files on a regular basis.
Yes, well, "successfully" might or might not be related to
"conveniently", but you're getting dangerously close to forcing me to
suggest that Chad is right about something, so do you mind....?
[...]
>>If they were supposedly professional video processors, you'd expect that
>>they might consider that idea. But Chad said they wanted 'cheap', so
>
> I can't imagine skimping on the FPU performance (or even
> the disk subsystem) would get them anywhere in the long run.
He already said, this box was used for "big dumb storage", not active
processing. Would you get a grip?
[...]
> The real question is HOW cheap?
>
> Are we talking about the bargain bin stuff that I would
> give my mother-in-law or serious workstations that would
> NOT be cheaper than faster Alpha hardware?
Basement. Cheap, as in "bog standard and whatever happens to be handy".
> As long as we're indulging in conjecture, why not wonder
> why they didn't consider an Altivec system? Digital video
> is probably one of the few areas where the inflated
> performance claims of MacOS cheerleaders actually hold up.
>
> Those too (Macs) should not be uncompetitive when compared
> to a serious video editing workstation.
>
> I am a nitpick and suspension of disbelief was not achieved.
Obviously.
--
T. Max Devlin
*** The best way to convince another is
to state your case moderately and
accurately. - Benjamin Franklin ***
------------------------------
From: Peter =?ISO-8859-1?Q?K=F6hlmann?= <[EMAIL PROTECTED]>
Subject: Re: Windows curses fast computers
Date: Sun, 21 Jan 2001 04:57:03 +0100
Erik Funkenbusch wrote:
>
> Actually, FreeBSD's halt command calls Sync, which is documented to have a
> bug which allows it to return before buffers are completely flushed, and
> causing a shutdown. Further, FreeBSD's "reboot" function in it's kernel
> calls a function called poweroff_wait() which merely waits 5 seconds
> before powering off the computer.
>
> You can see it for yourself here:
>
>
http://www.freebsd.org/cgi/cvsweb.cgi/~checkout~/src/sys/kern/kern_shutdown.
> c?rev=1.91&content-type=text/plain
>
> So, do you then think that FreeBSD is also "shit" engineering?
>
No, not if it works properly. could you just make a small computation in
mind HOW MUCH the drive can write in those 5 seconds? It is at least 30
times as much as the LARGEST caches you get on drives. So they wait enough
to make sure the drive has time enough.
Wintendo, on the other hand, shuts power off almost at once. This pisses me
off because my daughters computer (a 450MHz) manages to shut off too fast 2
out of 3 times. So, scandisk runs the next time. Then 1 out of 3 times that
blasted thing manages to find a new Ethernet card and installs it a second
time, rendering Network inoperable. You have to remove those bogus entries
AND install that damn card again. Well, this has no direct connection with
shutdown, but would not occur if MicroShit had done their job a little
better. But then again, why do they have to do that FUCKING hardware
detection every power on?? It is plain dumb to do it that way. I know a lot
of people who call me and ask what in hell that damned computer is up to
because it finds "new" hardware where none was installed.�
But, back to the drives.
Your statement that it was the drives fault is even dumber than that, and
by now you have probably realised it. You should at least say so, there are
(even here on advocacy) people who manage to say "here I was wrong".
Even Flatfish did it, he escaped the killfile that way (I had enough of his
rantings, but he was at least once grown up enough to admit error).
You have not done so, at least I have never read a statement from you to
that effect (and no, it's nearly impossible to be always right, and you
know that. It's quite easy to recall something wrong)
------------------------------
From: T. Max Devlin <[EMAIL PROTECTED]>
Crossposted-To: alt.destroy.microsoft
Subject: Re: "The Linux Desktop", by T. Max Devlin
Reply-To: [EMAIL PROTECTED]
Date: Sun, 21 Jan 2001 04:02:29 GMT
Said Ayende Rahien in alt.destroy.microsoft on Sat, 20 Jan 2001 21:24:02
>"T. Max Devlin" <[EMAIL PROTECTED]> wrote in message
>news:[EMAIL PROTECTED]...
>> Said SomeoneElse in alt.destroy.microsoft on Sat, 20 Jan 2001 13:47:06
>> >On Sat, 20 Jan 2001 04:29:04 GMT, T. Max Devlin <[EMAIL PROTECTED]>
>> [...]
>> >>Thanks, Gary. I still hope to avoid having to use gcc or any other
>> >>compiler, but I appreciate the conversation.
>> >You may not want to write any programs, but...
>> >
>> >sometimes the cutting edge stuff only comes in source and you have to
>> >build it yourself. Typically that is very simple:
>> >./configure
>> >make
>> >make install.
>> >
>> >But you will need gcc.
>>
>> Thanks, 'someoneelse'. I still hope to avoid having to use gcc or any
>> other compiler, but I appreciate the chance to repeat myself. ;-)
>
>The simplest workaround is to avoid using gcc, and using kgcc instead.
>Reason is that kgcc is the previous version of gcc, which will work
>correctly, unlike the gcc that you'll have with RH.
Maybe I've been fooling myself all these years, by continuing to think
of Usenet as an intellectual exercise. Why does nobody seem to be
getting this?
What am I getting myself in to?
--
T. Max Devlin
*** The best way to convince another is
to state your case moderately and
accurately. - Benjamin Franklin ***
------------------------------
** 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
******************************
