Linux-Advocacy Digest #175, Volume #31 Mon, 1 Jan 01 15:13:03 EST
Contents:
Re: Uptimes (T. Max Devlin)
Re: Uptimes (T. Max Devlin)
Re: Uptimes (T. Max Devlin)
Re: Uptimes (T. Max Devlin)
Re: Conclusion (T. Max Devlin)
Re: Conclusion (T. Max Devlin)
Re: Conclusion (T. Max Devlin)
Re: Operating Systems? Where would you go next? (Marty)
Re: Why Hatred? (J Sloan)
Re: Red hat becoming illegal? (T. Max Devlin)
----------------------------------------------------------------------------
From: T. Max Devlin <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: Uptimes
Reply-To: [EMAIL PROTECTED]
Date: Mon, 01 Jan 2001 19:49:53 GMT
Said Tom Wilson in comp.os.linux.advocacy on Mon, 01 Jan 2001 04:57:52
>"T. Max Devlin" <[EMAIL PROTECTED]> wrote in message
>news:[EMAIL PROTECTED]...
>> Said Erik Funkenbusch in comp.os.linux.advocacy on Sun, 31 Dec 2000
>> >"Tom Wilson" <[EMAIL PROTECTED]> wrote in message
>> >news:ZVA36.799$[EMAIL PROTECTED]...
>> >> Questioning statistics isn't exclusively the right of Microsoft fanatics.
>> >> I base my opinions not on Netcraft numbers, but with prior experience.
>> >>
>> >> Non-MS OS's have better uptimes.
>> >>
>> >> Period.
>> >>
>> >> End of story.
>> >
>> >Well, I have dissimilar experiences. I've had Linux boxes crash dialy, and
>> >NT systems stay up months. I've also seen it the other way around as well.
>>
>> I'm afraid it must be pointed out that, given your historical
>> credibility, your personal experience doesn't have the same value, to
>> you or to anyone else, that Tom's does.
>
>I've heard too many other similar woes about NT to conclude that mine was an
>exceptional case. I'm not about to call Erik a liar as I'm not familiar with
>him, but I meet any pro-NT testimonial with a healthy dose of skepticism. He
>may, indeed, be one of the lucky few who have had consistent success with
>it. I don't know.
I feel compelled to point out that it is more often that selection bias
and hindsight bias are the more likely causes of Erik's "experience"
with NT systems which make him a fan.
>> >All this proves is that your own experience is not the reality of everyone.
>>
>> But it damn well better be compatible with the reality of everyone, or
>> else you're just being deluded. Having personal experience with NT
>> staying up in some arbitrary implementation is not incompatible with
>> reality. Believing that it means NT is not crapware, however, requires
>> denying too many other people's knowledge and experience.
>
>Literal hoards of admins and developers wouldn't be pronouncing it as such
>if the basis for the claim were not present. I'm not above admitting that
>some complaints are fictitious and meant only to tweek selected noses. I
>assure you that my posts are not in that vein.
And I concur with you that, in the present environment, a good portion
of user's complaints are only the "fault" of the OS because the OS is
the single component in the system which cannot be 'swapped out' with
ease. Getting a whole new piece of hardware is often far less trouble
and expense, even if its just a driver problem with Windows software.
That's why I call it monopoly crapware.
>> Regardless, I think everyone here has probably already learned to be
>> wary when Erik Funkenbusch uses the word "proves".
>
>The very word, in most cases, is incorrectly applied.
Thus the "scare quotes". ;-)
--
T. Max Devlin
*** The best way to convince another is
to state your case moderately and
accurately. - Benjamin Franklin ***
Sign the petition and keep Deja's archive alive!
http://www2.PetitionOnline.com/dejanews/petition.html
------------------------------
From: T. Max Devlin <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: Uptimes
Reply-To: [EMAIL PROTECTED]
Date: Mon, 01 Jan 2001 19:49:55 GMT
Said Tom Wilson in comp.os.linux.advocacy on Mon, 01 Jan 2001 04:31:43
> [T. Max:]
>> To rely on "prior
>> experience" and ignore statistical evidence is no more superior an
>> approach than the opposite.
>
>From an analytical standpoint, true.
>From a practical standpoint, I find the experience approach , more often
>than not, produces accurate results. If you've bad experiences with a
>product line over an entire decade, the smart money is on continued
>headaches when you use them again.
>
>Non-scientific, but, consistently accurate none-the-less. Especially when MS
>is part of the equation.
On the contrary, I think if you actually have had bad experiences,
rather than engaging in routine selection and hindsight bias, it
qualifies as statistical evidence. It isn't strictly empirical, so I
guess you're right that it would be non-scientific. But it really just
comes down to how careful you are, and whether others have similar
enough experiences to make it a convincing statistical fact. MS passed
this mark years before anyone even noticed it (if they were ever over
the threshold to begin with, which is debatable.)
>> You are correct that questioning statistics
>> isn't exclusively the right (or habit) Microsoft defenders, but I would
>> further add that confusing the ability to question the statistics with
>> having invalidated the statistics is, indeed, something I expect to see
>> more often from those who aren't intelligent enough to recognize
>> crapware for what it is.
>
>It seems no-one can be impartial about statistics. Either you believe them
>(or how they are represented) or you discount them altogether regardless of
>the source or presentation. So much lying and spinning with numbers occurs
>that most tend to take the latter approach.
There are lies, damn lies, and statistics, yes. But that doesn't make
statistics showing a clear fact, such as that MS software is the most
unreliable, typically, the same as Erik's lies.
--
T. Max Devlin
*** The best way to convince another is
to state your case moderately and
accurately. - Benjamin Franklin ***
Sign the petition and keep Deja's archive alive!
http://www2.PetitionOnline.com/dejanews/petition.html
------------------------------
From: T. Max Devlin <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: Uptimes
Reply-To: [EMAIL PROTECTED]
Date: Mon, 01 Jan 2001 19:49:56 GMT
Said JSPL in comp.os.linux.advocacy on Mon, 1 Jan 2001 10:32:21 -0500;
>"T. Max Devlin" <[EMAIL PROTECTED]> wrote in message
>> My apologies. I obviously did get hot-headed, and misinterpreted your
>> remark. When you mentioned http headers, I presumed you were
>> misrepresenting the issue, since http headers aren't related, it
>> appears, to Netcraft's numbers (or Uptime's numbers).
>
>They are related and the sole source of Netcrafts information.
You are, as always, incorrect.
>The http
>response to a HEAD request to the server at port 80 is the only data in
>which Netcraft uses to come up with their uptime assumptions.
But it is not the http response, but the TCP packets containing this
response, which provides both the OS and the uptime indication.
>Now, if you
>look through the rfc's on the standard, industry accepted header fields
>you'll find that there are none which pertain directly to "time since last
>reboot" or anything even close. Therefore netcraft is using some kind of
>unsupported method which is apparently impossible to duplicate.
Not at all. I'm pretty sure I know precisely how they are doing it.
The TCP packet contains a sequence number for packet ordering. The
number simply has to be unique, what is called pseudo-random. As often
occurs in computer engineering, most OSes use the system timeticks
value, and counter used as an uptime continuity indicator, as the value.
Firewalls, under the general belief that any information is too much
information to let someone else have, provide true randomized (but not
necessarily random) numbers.
None of this has anything to do with HTTP, other than the fact that it
uses TCP and the intent is to measure uptime for various web sites.
>I suspect
>that their method wouldn't stand up to independent analysis, which is why
>they do not release their method.
I'm afraid this is an argument from ignorance. Looking at independent
correlation of their results, we find them to be accurate, consistent,
and practical.
>Uptimes.net purposely puts a header field
>in which indicates time since reboot on participating machines which makes
>it much more accurate if it weren't for the fact that the field can be
>forged by the client at will.
Uptime.net uses a client. The uptime value cannot be forged; the number
of previous counter wraps can be initialized. But that only invalidates
the very first indication provided by a system. On the very first
reboot, the use of a counter as a continuity indicator makes continued
fudging impossible. Sure, after rebooting you could say the system was
just "wrapping" for the 35th time. But the server watching the
continuity indicator can tell it wasn't a wrap, so your baseline starts
at the current counter value, not the number of previous wraps
indicated.
--
T. Max Devlin
*** The best way to convince another is
to state your case moderately and
accurately. - Benjamin Franklin ***
Sign the petition and keep Deja's archive alive!
http://www2.PetitionOnline.com/dejanews/petition.html
------------------------------
From: T. Max Devlin <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: Uptimes
Reply-To: [EMAIL PROTECTED]
Date: Mon, 01 Jan 2001 19:49:58 GMT
Said JSPL in comp.os.linux.advocacy on Mon, 1 Jan 2001 11:25:11 -0500;
[...]
>If netcraft in fact gets their "time since reboot" info from the TCP packet,
>why does it take several pollings?
Because they use the delta of a counter as a continuity indicator, not a
clock.
>Wouldn't that information be available at
>first contact with a given server?
Available, yes, but meaningless.
>The whole problem with Netcraft, and the
>main reason they have no credibility is they are perfoming a feat which is
>impossible for anyone to duplicate.
Their credibility, given that the keep their actual method proprietary
(but it can be easily reverse engineered, which means there is no trade
secret protection, nor copyright or patent protection, so anyone can
duplicate it) is based on how well their results are corroborated by
other methods. And the less those methods have in common, the better.
Such correlations are easily found, as well, and it can be positively
stated, at this point, that it is a fact that Microsoft software is
monopoly crapware. Not so good as to be acceptable in a free market,
but not any better than it has to be to be unavoidable in an illegally
monopolized market.
--
T. Max Devlin
*** The best way to convince another is
to state your case moderately and
accurately. - Benjamin Franklin ***
Sign the petition and keep Deja's archive alive!
http://www2.PetitionOnline.com/dejanews/petition.html
------------------------------
From: T. Max Devlin <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: Conclusion
Reply-To: [EMAIL PROTECTED]
Date: Mon, 01 Jan 2001 19:49:59 GMT
Said Erik Funkenbusch in comp.os.linux.advocacy on Mon, 1 Jan 2001
>"Adam Ruth" <[EMAIL PROTECTED]> wrote in message
[...]
>So in other words, it's reporting the OS as being the proxy servers OS and
>reporting the web server of the real machine. Chances are, if the proxy
>were an OS that reported uptime it would give the uptime of the proxy server
>as well.
"Chances are?" No, sorry, that is a complete fabrication. Worse even
than a gedanken test, not even qualifying as a though experiment, that
is simply a bogus and unsupported assertion, contrary to all known
facts.
[...]
>And this shows that the uptime is accurate, how?
By providing corroboration that the Netcraft statistics work the way
they are reputed to work, and therefore are not inaccurate in the way
you wish that they were. You will find, Erik, that when you argue from
conclusions, expecting that Netcraft's numbers are suspect because you
don't like the conclusions their results support (that NT/W2K is
unreliable.) It turns out that all of your suppositions about how their
statistics are invalid are false. It would seem that this was a useful
exercise in providing several different hypothesis about how their
values might be erroneous, if it weren't for the fact that you have been
so resolutely adamant about ignoring the facts.
[...]
>You never could prove this one. Netcraft has only been reporting uptimes
>for 2 months, your claim that you verified this site with Netcraft is an
>obvious fabrication since you claim to have checked it 6 months ago.
[...]
>> So, anyway, what are those many sites with complex firewall schemes you
>> worked on?
>
>I never claimed I worked on many sites with complex firewalls.
IIRC, you claimed that you had worked on sites which could fool
Netcraft. You have yet again simply failed to provide any information
how, or indicate which sites they might be, or how they could fool
Netcraft's method.
--
T. Max Devlin
*** The best way to convince another is
to state your case moderately and
accurately. - Benjamin Franklin ***
Sign the petition and keep Deja's archive alive!
http://www2.PetitionOnline.com/dejanews/petition.html
------------------------------
From: T. Max Devlin <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: Conclusion
Reply-To: [EMAIL PROTECTED]
Date: Mon, 01 Jan 2001 19:50:00 GMT
Said Erik Funkenbusch in comp.os.linux.advocacy on Sun, 31 Dec 2000
>"T. Max Devlin" <[EMAIL PROTECTED]> wrote in message
>news:[EMAIL PROTECTED]...
>> Said Ayende Rahien in comp.os.linux.advocacy on Sun, 31 Dec 2000
>> >"Adam Ruth" <[EMAIL PROTECTED]> wrote in message
>> >> [...]Since no one
>> >> has yet to come forward with a site that they know is incorrect (and some
>> >> people have claimed to work on sites that they 'know' will fool Necraft),
>> >> then it's reasonable to assume that the numbers are somewhat accurate.
>> >
>> >No one has come forward and said that the numbers *are* correct, either.
>>
>> You are, once again, incorrect. Even if we discount (for no reason but
>> that we don't want their results to be true) Netcraft's own assertion
>> that their numbers are correct, others have corroborated them, and a
>> list of "verified" sites was posted in this thread a couple days ago.
>
>Obviously you didn't bother to actually check those "verified sites"
>yourself, did you?
>
>You would have seen that some of them list no uptimes at all at netcraft,
>thus the list is bogus.
Your conclusion is logically flawed, in the extreme. You have not
provided any indication or reason to believe that any Netcraft numbers
are false, inaccurate, or "bogus". And it hardly needs to be repeated,
yet again, (though I will anyway, in case any reader isn't aware that
you are a bogus Microsoft apologist, repeating fabrications in the
mistaken belief that you are supported in saying that NT is a reliable
platform) that having certain sites list no uptime does not shed any
question at all, in and of itself, that those sites that do list uptime
are incorrect.
--
T. Max Devlin
*** The best way to convince another is
to state your case moderately and
accurately. - Benjamin Franklin ***
Sign the petition and keep Deja's archive alive!
http://www2.PetitionOnline.com/dejanews/petition.html
------------------------------
From: T. Max Devlin <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: Conclusion
Reply-To: [EMAIL PROTECTED]
Date: Mon, 01 Jan 2001 19:50:01 GMT
Said Adam Ruth in comp.os.linux.advocacy on Sun, 31 Dec 2000 18:37:37
[...]
>> An opinion which is not a reasoned opinion is not an opinion; it is a
>> fabrication. It is a fabrication to say that Netcraft's numbers are not
>> accurate.
>
>I disagree, it can still be an opinion. From Websters: "A belief or
>conclusion held with confidence but not substantiated by positive knowledge
>or proof." I'm certain that Chad has confidence that the numbers are false,
>and I'm also sure that he has no positive knowledge or proof.
>
>But anyway, your point is well taken.
As is yours.
Thanks for your time. Hope it helps.
--
T. Max Devlin
*** The best way to convince another is
to state your case moderately and
accurately. - Benjamin Franklin ***
Sign the petition and keep Deja's archive alive!
http://www2.PetitionOnline.com/dejanews/petition.html
------------------------------
From: Marty <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux,comp.os.os2.advocacy
Subject: Re: Operating Systems? Where would you go next?
Date: Mon, 01 Jan 2001 19:46:43 GMT
The Ghost In The Machine wrote:
>
> os2.advocacy removed from followups.
[re-added, in case you have any questions on the response below]
> (I wonder if anyone ported X to OS2? Interesting thought, that.)
They sure did. We have a full implementation of XFree86.
http://set.gmd.de/~veit/os2/xf86os2.html
------------------------------
From: J Sloan <[EMAIL PROTECTED]>
Subject: Re: Why Hatred?
Date: Mon, 01 Jan 2001 19:53:42 GMT
Erik Funkenbusch wrote:
> I doubt that. Don't you think she'd be on the phone to you after about 3
> steps? "What's a mount point? What do I set these partition thingies to?
Why would she ask me about these things?
All she did was click where it said "install kde workstation"
like I told her, and sat back to watch the pretty flashing screens.
>
> What's a server? What's a workstation?" There are tons of questions in
> there that most Windows users, (including your mother unless she's some kind
> of computer wizard) would not know how to answer, even in simple distros.
>
> > > 2) Overwhelming amount of work needed to acomplish even minor tasks.
> >
> > Please explain -
> >
> > To browse the web, you click on the icon
>
> After you've configured PPP.
I have cable, but whatever.
I remember years ago when I had ppp, it was a
3 minute task to go to the red hat control panel
and click on the relevant buttons and type things
like the dialup number, account and password in
the dialog boxes.
> > To get mail, you click on the icon.
>
> After you figure out how to configure the mail program.
Not sure what you mean, please explain. Are you talking
about the MTA? that comes configured "out of the box".
Or are you taling about having to type in the name of
your mail server etc, just as you would in any windows
mail program?
> > To start the word processor you click on the icon
>
> After you figure out how to install it.
Not sure what you mean, it comes installed with the distro.
> > To play quake 3 arena you click on the icon.
>
> I've never seen a distro with quake 3 arena installed.
I have - but even so, installing it is a cinch:
execute the install file, then indicate your acceptance
of choices by clicking on icons in the install dialog box.
> Install something from a tarball? Unless you already know the syntax to
> tar, you'll spend an hour figuring out how to do it. Then you'll spend an
> hour trying to figure out what you're supposed to do afterwards.
Why would mom install anything from a tarball?
She would click on the kpackage or gnorpm icon,
then select the package in the file menu, and
click "install".
> Ok, contact management software. Where is it? CAD software equivelant to
> AutoCAD or other high end CAD tools? What about MIDI tools?
First of all, that's not OS functionality, it's apps.
Yes, windows has had a huge head start, but these things
all exist in some form, and are quickly becoming more and
available under Linux.
What was it you people were saying a couple years ago?
"Linux has no 3D hardware video support, so it will die" -
but now we have support in place and rapidly maturing.
and then what were you saying? "There's no commercial
databases for Linux, so Linux will never make it" - but as
you know, Linux now runs all major commercial databases.
What was your other mantra? "Linux has no support, no
company you can call" OK, is IBM global services good
enough for you? As you can see, your arguments are
all being swept away by the invisible hand of competition.
> > I was a windows user, and I tried Linux.
>
> Oh, so that means all Windows users that try Linux have the same opinion. I
> understand now.
huh?
>
>
> > Guess what? I now prefer Linux, and really don't like using
> > windows anymorel. windows now seems much too limiting.
>
> What do you do with it?
- web browsing with rich multimedia content
- email communications (several hundred messages per day)
- participating in newsgroups like this one (check my headers)
- document processing (word, excel files)
- programming assignments for classes
- chat via irc, aol, yahoo etc
- playing internet games like unreal tournament or quake 3 arena
- playing CDs
- ripping CDs to make mps
- playing mp3s
- downloading/uploading mp3s on napster
- all my work as a system administrator and webmaster.
> Lots of places.
mainly ex windows users, with some long time Unix users
> Where did the 200 million Windows users come from?
kmart, target etc - any place where joe 6 pack buys a
pc and has no choice about what OS comes with it.
jjs
------------------------------
From: T. Max Devlin <[EMAIL PROTECTED]>
Crossposted-To:
alt.destroy.microsoft,comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: Red hat becoming illegal?
Reply-To: [EMAIL PROTECTED]
Date: Mon, 01 Jan 2001 19:54:06 GMT
Said JSPL in alt.destroy.microsoft on Mon, 1 Jan 2001 13:37:16 -0500;
>"Bob Hauck" <[EMAIL PROTECTED]> wrote in message
>news:[EMAIL PROTECTED]...
>> On Mon, 01 Jan 2001 01:28:58 GMT, Chad Myers
>> <[EMAIL PROTECTED]> wrote:
>>
>> >"T. Max Devlin" <[EMAIL PROTECTED]> wrote in message
>> >news:[EMAIL PROTECTED]...
>>
>> >> No, you claimed that the Democrats were trying to "subvert the rule of
>> >> law."
>> >
>> >Which they were. You didn't even answer the claims, instead stooping to
>> >name calling.
>>
>> How is going to court "subverting" the rule of law? I've always been
>> under the impression that you went to court to get an interpretation of
>> the law, not to subvert it. Subverting would involve something
>> extra-legal, such as paying off the people counting the ballots or
>> voting on behalf of dead people.
>>
>> Maybe I missed a civics lesson someplace where they defined lawsuits as
>> "subversion".
>
>Well lets see, a law on the books in Florida stated that the the Secretary
>of State SHALL certify the results by a certain date. The Florida Supreme
>court issued an order stating that the Secretary of State SHALL NOT certify
>by that date. I'd call that "subverting" the existing law, not
>"interpreting" it.
Thus supporting the theory that Republicans are simply wrapping
themselves in the flag, and claiming that any Democratic action
whatsoever was "subverting" the law. A relatively worrisome indication
of the kind of civil environment we can expect under Republican rule.
--
T. Max Devlin
*** The best way to convince another is
to state your case moderately and
accurately. - Benjamin Franklin ***
Sign the petition and keep Deja's archive alive!
http://www2.PetitionOnline.com/dejanews/petition.html
------------------------------
** 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
******************************
