Linux-Advocacy Digest #500, Volume #30 Tue, 28 Nov 00 13:13:02 EST
Contents:
Re: Whistler review. ("Kevin M. Taggart")
Re: Linux for nitwits ([EMAIL PROTECTED])
Re: Whistler review. ("Ayende Rahien")
Re: Whistler review. ("Ayende Rahien")
Re: Windoze 2000 - just as shitty as ever ("Ayende Rahien")
Re: Of course, there is a down side... (Donovan Rebbechi)
Re: Whistler review. (J.C.)
Re: The Sixth Sense (T. Max Devlin)
Re: Uptime -- where is NT? (T. Max Devlin)
----------------------------------------------------------------------------
From: "Kevin M. Taggart" <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy,comp.sys.mac.advocacy
Subject: Re: Whistler review.
Date: Tue, 28 Nov 2000 11:37:21 -0600
Josiah Fizer wrote:
>
> Ayende Rahien wrote:
>
> > "Edward Rosten" <[EMAIL PROTECTED]> wrote in message
> > news:[EMAIL PROTECTED]...
> > > Matthew Soltysiak wrote:
> > > >
> > > > Aaron, get a life. The guy likes Whislter, so be it. Leave him alone.
> > In fact, i like
> > > > it too. My eng. buddies and I love it. Beats the hell outta Linux for
> > usability. Take a
> > > > look a KDE2, ultimate bug shit edition. GNOME, ultimate bloat-ware and
> > bug shit
> > > > edition. Both segfault quite frequently. So, what does this say about
> > Whislter?
> > >
> > > KDE and GNOME say nothing about Whistler.
> >
> > Correct, but one has to wonder, due to Aaron's constant nagging about
> > Windows not getting features that Unix had ages ago (his words, or at least
> > his meaning), why did Unix & Linux, so far, have been unable to produce a
> > good GUI system.
> > Apple had it for how long, 17 years?
>
> While it may just be me. I find the GUI in IRIX (OpenMagic) to be far better
> then the Mac or Windows UI and light years ahead of anything I've found on
> Linux.
The OpenMagic environment seems very derivative (or at least, very
reminiscent) of the Mac UI to me. The alignment of icons along the right
side of the desktop, removeable media mounting on the desktop, Get Info,
the dumpster, etc. There are a lot of nice touches like the shelf,
minimized windows, the NeXT-like Toolchest menu, etc., but with the
inclusion of AppleTalk/AFP, and Quicktime being the default compressed
video standard, it is obvious that the implementation of the useablility
features of IRIX are meant to be relatively seamless with the MacOS.
When you consider that many effects houses use both at different stages
of production, this makes a great deal of sense.
When I switch between MacOS and IRIX, the environment is so familiar,
that I don't even skip a beat. I appreciate this greatly.
Just my opinion.
--KT
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: Linux for nitwits
Date: Tue, 28 Nov 2000 16:59:09 GMT
In article <[EMAIL PROTECTED]>,
Glitch <[EMAIL PROTECTED]> wrote:
> mark wrote:
>
> > In article
> > <mCxT5.10916$[EMAIL PROTECTED]>,
> > Jose Mirles wrote:
[snip!]
> > > I have seen tons of people in Best Buy and numerous
> > > PC shops with simple Windows problems. Frankly, I am
> > > surprised by some of the things I have seen people
> > > bring back their PC for. I have come to the conclusion
> > > that some people just want a simple looking, simple
> > > running OS to use. They don't care about the problems,
> > > securities, etc. The average Joe and Jane will not use
> > > MS's Windows update or install any patches. I can't
> > > even count the times I have heard users complain of a
> > > virus and they are running a AV program with a year old
> >
> > Err, year old what?
>
> .dat file. Your virus definitions. Each virus has a
> signature that an antivirus looks for. Each signature
> for every virus is stored in the .dat file. It obviously
> gets bigger over time as more viruses are identified.
> This is the file that needs updated as it is what really
> powers the AV program so that you can find and eradicate
> the newest viruses. If you have an old .dat file from a
> year ago you might as well not have an AV program running
> as it just about defeats the purpose of having it.
I would disagree that running an AV program with an old
.DAT file "defeats the purpose of having an AV program."
While you're right that all the *new* virii will not be
detected and defeated by the old .DAT file, there are
numerous *older* virii which would be caught and defeated
by the old .DAT file.
Ergo, not having an AV program at all will always be more
dangerous than having an AV program with a year old .DAT
file. Like the current resurgence of the leprosy virus
in the human population (armadillos in America are
currently carrying it), older viruses can and do reappear
to wreak havoc on the computer community.
I myself recently saw a new executable application
(shareware, of course) pop up the AV message "The file
is infected with the STONED virus. Disinfect?" Anyone
on here who remembers what the STONED virus was is
definitely dating himself or herself. The last time
I saw a file with the STONED virus was back in 1990.
The other thing to remember is that AV programs with
older .DAT files still often have the feature of
monitoring suspicious activity. Again, if you don't
have an AV program, even one with an older .DAT file,
all sorts of "suspicious activity" will go on and you
will be unaware that it is happening.
So no, I wouldn't say that running an AV program with
an old .DAT file "defeats the purpose of having an AV
program."
It mostly defeats a purpose of "preventing new viruses
from being a problem", but if you think about it, new
viruses are appearing at a rate which dwarfs any
anti-virus software company's ability to create new
.DAT files.
Ergo, *all* AV programs are using "older .DAT files",
older than the newest viruses, and thus your reasoning
declares that one should get rid of all AV programs
because they can NEVER have new .DAT files.
Absolutism never really helps anyone. It just makes
it possible for someone somewhere to gain quite a lot
of power and/or money--and/or votes--at the expense of
quite a lot of other people.
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: "Ayende Rahien" <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy,comp.sys.mac.advocacy
Subject: Re: Whistler review.
Date: Tue, 28 Nov 2000 18:57:05 +0200
"Josiah Fizer" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Ayende Rahien wrote:
>
> > "Edward Rosten" <[EMAIL PROTECTED]> wrote in message
> > news:[EMAIL PROTECTED]...
> > > Matthew Soltysiak wrote:
> > > >
> > > > Aaron, get a life. The guy likes Whislter, so be it. Leave him
alone.
> > In fact, i like
> > > > it too. My eng. buddies and I love it. Beats the hell outta Linux
for
> > usability. Take a
> > > > look a KDE2, ultimate bug shit edition. GNOME, ultimate bloat-ware
and
> > bug shit
> > > > edition. Both segfault quite frequently. So, what does this say
about
> > Whislter?
> > >
> > > KDE and GNOME say nothing about Whistler.
> >
> > Correct, but one has to wonder, due to Aaron's constant nagging about
> > Windows not getting features that Unix had ages ago (his words, or at
least
> > his meaning), why did Unix & Linux, so far, have been unable to produce
a
> > good GUI system.
> > Apple had it for how long, 17 years?
>
> While it may just be me. I find the GUI in IRIX (OpenMagic) to be far
better
> then the Mac or Windows UI and light years ahead of anything I've found on
> Linux.
Can't say I ever saw an IRIX, do you mind telling why you think it's better?
------------------------------
From: "Ayende Rahien" <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy,comp.sys.mac.advocacy
Subject: Re: Whistler review.
Date: Tue, 28 Nov 2000 19:04:50 +0200
"J.C." <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> On Tue, 28 Nov 2000 17:26:51 +0200, Ayende Rahien <[EMAIL PROTECTED]>
wrote:
>
>
> [snip]
>
>
> >Alpha - not working.
> >Beta - still not working
>
> Nope. Far too simplistic.
No, it's a joke.
I thought it was a well known one, but apperantely it isn't.
> >> And I've read
> >> >documents created in FP from variety of browsers (the most horrible
part
> >of
> >> >web designing. I'm trying to stick to HTML 3.2 for the most part, it's
> >> >widely supported.) There have rarely been problems with it.
> >>
> >> Thank God...
> >>
> >>
> >> >ActiveX is not a standard. It's a de facto standard, which is
different.
> >>
> >> I'll let that speak as a testament to your mental capacity.
> >>
> >> ("...which is different." How? You haven't even defined what makes
> >something
> >> "a standard". In any case, the only thing that matters is market
> >penetration,
> >> and when you integrate this shit with the OS, MS gets a lot of
> >penetration...
> >> get it?)
> >
> >There is a different between a standard and a de facto standard.
> >
> >A standard is something that is published openly.
> >De facto standard is something that almost everybody use.
>
> ... and in the computer industry, market penetration is everything. So
whether or not
> something is "published openly", if it's used by a large chunk of the
relevant population,
> it can be referred to in the capacity of a standard, because the meaning
has changed (to
> everyone but yourself, as it appears). Whether or not the standard is
"published
> openly" makes no difference in the computer industry, particularly to end
users.
So does this make windows the standard OS, and IE the standard browser, and
OE the standard Email Client, and MS the standard software-producer company?
> >> >IIRC, the Java they wanted to develop would've supported windows
spesific
> >> >commands or libraries.
> >>
> >> ... so that way, it wouldn't run on OSs other than Windows... (duh...)
> >
> >If you used the windows spesific commands/libraries, yes.
>
> ... and that's exactly my point. MS did this to screw up Java. Have you
tried wasting a brain
> cell on it? MS encouraged developers to do this, precisely to remove
Java's primary benefit:
> write once, run anywhere. MS realized this as a threat and worked to kill
off Java.
I don't program in Java.
> >> >If you wouldn't use the windows spesific commands/libraries, then you
> >should
> >> >be able to move it around.
> >>
> >> ... but if you do, then your portability goes to shit, which is exactly
> >what
> >> MS wanted...
> >
> >Whatever they may have wanted.You could develop for multi platforms, or
just
> >for windows.
>
> MS's implementations of Java were fucked, to say the least. So, why did MS
release these
> munted versions of Java? To run apps written for this munted Java. Why did
MS screw up Java
> in the first place? See above.
My knowledge of this is third party at best, as I don't program or interest
very much in Java.
> >> >They can do whatever they want with the office documents, there isn't
a
> >> >standard for office documents, therefor, you can't claim a propreity
> >> >standard here.
> >>
> >> Um, you need a good English-language dictionary. It might help your
> >spelling as well.
> >> MS "can do whatever they want with the office documents" precisely
> >_because_ the various
> >> Office file formats are _proprietary_.
> >
> >But not _standards_ . Which is what you said they were.
>
> Did I say that the Office file formats are standards? Where?
Sorry, you said formats, I didn't notice it.
> (Well, given MS's market penetration in the office-app market, it makes no
difference; you're
> playing semantic games with the meaning of `standard' and `de facto
standard'...)
No, there is a big difference between those things.
<blink> used to be a de facto standard, frex.
<table> is standard.
[snip]
> >> > which MS hasn't done.
> >>
> >> You are very naive. Are you not old enough to remember, say, DR-DOS?
Try
> >> http://usvms.gpo.gov/ for infos.
> >
> >There has been two incidents with DR-DOS, the 3.X one, with the warning
in
> >the beta, and the one with 95, which one are you talking about?
>
> There are two? I refer you to the 3.1x incident, with the dialog box that
appeared, erm...
Oh, you mean the dialog that said:
"Error number: <can't recall>
Please contact Microsoft Beta Support"
Which only appeared on the beta release of win3.1X?
Nothing to suggest what the problem was.
Nothing to suggest that DR-DOS and its like are invovled.
------------------------------
From: "Ayende Rahien" <[EMAIL PROTECTED]>
Crossposted-To: alt.destroy.microsoft,comp.os.ms-windows.advocacy
Subject: Re: Windoze 2000 - just as shitty as ever
Date: Tue, 28 Nov 2000 19:11:09 +0200
"T. Max Devlin" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Said Ayende Rahien in alt.destroy.microsoft on Tue, 28 Nov 2000 04:56:37
> >"T. Max Devlin" <[EMAIL PROTECTED]> wrote in message
> >news:[EMAIL PROTECTED]...
> >> Said Ayende Rahien in alt.destroy.microsoft on Mon, 27 Nov 2000
01:40:18
> >
> >> >Here is another one:
> >> >Create a shortcut to explorer.exe, check the "run as another user"
> >> >Double click it, explorer will open, you are now, within this windows
and
> >> >within any windows or applications that you launch from this widnows,
an
> >> >admin.
> >>
> >> Now that I would have bet hard money wasn't possible. Seems like MS
had
> >> to work pretty hard to make W2K mimic Unix enough to be acceptable
> >> enough to extend the monopoly.
> >
> >You've a very strange way of thinking.
>
> Indeed I do. Others have called it a "unique perspective". It works
> quite adequately.
>
> >If MS makes bad software, it's because they are monopoly and evil.
> >If MS makes good software, it's because they are monopoly and evil.
> >
> >But why make the seperation?
> >If MS, bad & evil.
>
> I have never said that MS is either bad or evil, merely that they are a
> monopoly. When it is bad software, I explain why: its monopoly
> crapware, and doesn't actually have to be competitive (i.e. good enough
> to compete) in order to be sold. When it is "good software", I explain
> why: Microsoft monopolizes, which means they design their product to
> remove any reasons to avoid it. While this does coincidentally happen
> to satisfy some certain users, supposedly, it can be distinguished from
> competitive behavior by the fact that it is designed not to maximize
> Microsoft's profits on Windows, but to maximize the market share of
> Windows (or, more properly, to defend the almost 100% market share of
> Windows).
Where did you pull *that* out?
Almost 100%?
In their wildest dream, maybe.
Right now, MS has about 90% of the desktop market, and about 20% on the
server market.
> Now, you can refute my reasoning, provide evidence to the contrary, or
> attempt to ridicule my position by building strawmen about "evil
> empires" and such. Your choice.
So, what you want MS to do is to go for vacation for a year or two, in which
their market share will shrink.
Then, when they're back on bussiness, you will judge their software by its
merits instead of its maker?
There is more to a book than its cover.
------------------------------
From: [EMAIL PROTECTED] (Donovan Rebbechi)
Subject: Re: Of course, there is a down side...
Date: 28 Nov 2000 17:26:23 GMT
On Tue, 28 Nov 2000 08:18:36 -0600, Tim wrote:
>Donovan Rebbechi wrote:
>> rpmfind has been around for more than a year now, so this
>> debian-is-better-because-of-apt-get doesn't really cut it any more.
>>
>
>Actually I think it still is. When I do "apt-get install packagename" all
>of it's dependencies are checked and downloaded too and installed in the
>right order. rpm doesn't do that, and when you're installing a large
>complicated series of packages with a lot of dependencies rpm is a massive
>pain in the ass ... I don't know what new features rpm v4 has, hopefully
>it's a better system.
I'm not sure if dependencies can cascade. However, it's certainly possible
to do something like remotely apply all the updates, for example.
However, most of the dependency problems people experience arise when they
try to install RPMs that were built for a different distribution. The
only other place you get large dependency cascades is if you try to
upgrade say KDE or GNOME (in which case just download the whole distribution
+ Qt or GTK) So the main advantage Debian has here is the somewhat
questionable advantage of obscurity -- noone else is using that format.
--
Donovan Rebbechi * http://pegasus.rutgers.edu/~elflord/ *
elflord at panix dot com
------------------------------
From: [EMAIL PROTECTED] (J.C.)
Crossposted-To:
comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy,comp.sys.mac.advocacy
Subject: Re: Whistler review.
Reply-To: [EMAIL PROTECTED]
Date: 29 Nov 2000 04:29:20 +1100
On Tue, 28 Nov 2000 18:47:59 +0200, Ayende Rahien <[EMAIL PROTECTED]> wrote:
[snip]
>I'm very happy for you.
>However, your initial attidue is that 2K isn't even
>"somewhere-remotely-distantly-approaching" unix.
>I've proven to you that it's more does more than that.
No, it doesn't, and it's very arrogant of you to presume that some study proves what
my empirical observation, my experience, my _job_, most clearly disputes.
When I mentioned "somewhere-remotely-distantly-approaching" unix, I was being
_sarcastic_. So
maybe NT/2k _is_ "somewhere-remotely-distantly-approaching" unix. I don't care. It's
not up to
the standard of the unices I admin. It's still not good enough.
>Your needs are met with Solaris. Fine.
>Your attidue is that of 2K is totally useless. And therefor no one's needs
>can be met on it.
For me, it _is_ totally useless. I strongly believe, however, that there
are a lot of people for which 2k may be a worthwhile upgrade. But for _me_,
speaking in the capacity of a typical sysadmin, 2k _is_ useless. It is no
better than NT in handling load, and just as insecure...
[snip]
>Business Research Group studies.
*shrug*
50% less on labor? Hahah, laughing already. The amount of work I had to put into NT
and 2k...
thank fuck that's all over now...
Value-added software? Most software I use as a sysadmin is free... (what software
did they use for the survey?)
46% per user? How does that work?
>> >> Where Solaris gets beaten? "GUI management tools" Hahaha, I'm laughing
>> >already. "Application
>> >> availability"? I use mostly free and open-source, and that which isn't,
>is
>> >readily available.
>> >> Unix has no lack of server apps...
>> >
>> >Development Considerations, Platform Choices, Distributed Systems
>> >Management, clustering, tco.
>>
>> Wow. But none of this matters to _me_. Maybe to other people it does. But
>I'm a typical Unix
>> sysadmin -- I'm in good company. "Platform choices"? I don't give a shit
>about platform
>> choices. I run the OS I want on the hardware I want. The only reason why
>2k wins that category
>> is the number of "vendors that sell and support Windows 2000 on their
>products." OTOH, only
>> Sun markets Solaris. Doesn't change the quality of either product.
>
>
>You run Solaris on Sun hardware. Hardly a choice here.
Gets the job done with superb reliability.
>Oh, you can run it on x86, would it be of the same quality?
In my experience, yes. Solaris on x86 is a good product. [On x86 boxes,
I tend to run BSD variants, keeps my multi-platform skills up ;) ]
>Then why did you say this "Try to get 2k boxes up to
>(okay, sorry, somewhere-remotely-distantly-approaching)..."
>Thus implying that Win2K isn't a good system.
... and for me, it isn't. It just isn't up to Solaris' or BSD's standards. There
_are_ people out there who've got 2k down as a good server for something or the
other, or as a good desktop. I don't care -- and there are millions of sysadmins
who agree with me in my saying that NT/2k just isn't up to Unix's standard.
>> "Development Considerations"? 2k has "a slight advantage" due to "its high
>degree of
>> integration". Phooey. I don't want shit integrated with more shit. I
>_like_ the relatively
>> compartmentalized nature of Unix/clones.
>>
>> Clustering? In all the passages containing `clustering', it was either a
>tie, or Solaris came
>> up on top...
>
>TPC disagree with you.
Should I care? Is Mr. TPC a Unix admin? (*sigh*, that's sarcasm...)
Why should I trust TPC over my own judgement?
>> >> And security wasn't mentioned as a separate category... hrmm... don't
>get
>> >me started...
>> >
>> >Does Solaris has ACL? (I'm asking, not tounting)
>>
>> No. Do I care? No. Does Solaris occupy less than 21% of the webserving
>market
>> but count for over 60% of the breakins? No. Does NT/2k? Yes!
>(netcraft.com).
>
>Netcraft isn't highly accurate.
Would have to disagree with you there -- "Netcraft isn't highly accurate" sounds like a
copout. I knew you'd say that. Can't you come up with anything better than that? Why
wouldn't
it be highly accurate?
If you actually put a braincell towards it, you'd realize that I didn't even mention
where the
breakin statistics were, geez. (How can you criticize statistics you haven't seen?)
(they're at www.attrition.org/mirror/attrition/os-graphs.html, in case you're
interested...
netcraft is for the general, web-wide OS marketshare stats...)
>How long Solaris exist?
>On comparable time scales, what was the Solaris share compare to NT/2K one?
>(IE, it's 8 years since NT was released, what was Solaris like 8 years from
>the commercial release?)
The history of Unix is quite murky -- sometimes, you don't know who's got code
from whom... ;) In any case, it's irrelevant; I'm criticizing NT/2k on an absolute
scale -- if it doesn't work for my needs, but it's targeted there, then I'm going
to criticize for lack of quality.
>> (can I secure a Solaris box, even though ACLs aren't implemented? Yes...)
>
>Who claimed that you can't?
It wasn't an issue. I was just pointing it out. Did I mention anywhere that anyone
claimed
that I couldn't?
>ACL are better then the grouping permissions because they allow much more
>control on the permissions you are delegating.
Fair enough. NT/2k would be even worse off without it, I suppose...
[snip]
>Please don't compare NT & 2K on terms or reliablity, security, and... well,
>practically anything.
>In doing so, you reveal your ignorance about it.
Don't reduce this to personal attacks. In any case, I find NT and 2k to be on
a par as far as stability and security are concerned. So maybe your experience
is different... should I care? I only care about what's relevant to _my_ job,
and my empirical observations of NT _AND_ 2k is that they both suck.
[snip]
--
J.C.
"The free flow of information along data highways being piped into our
homes and offices will permit unimaginable control by a small elite..."
-- 'The Thunder of Justice', pg. 264
------------------------------
From: T. Max Devlin <[EMAIL PROTECTED]>
Crossposted-To:
alt.destroy.microsoft,comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: The Sixth Sense
Date: Tue, 28 Nov 2000 12:39:34 -0500
Reply-To: [EMAIL PROTECTED]
Said Ayende Rahien in alt.destroy.microsoft on Tue, 28 Nov 2000 03:39:23
>"T. Max Devlin" <[EMAIL PROTECTED]> wrote in message
[...]
>An application that is handling files has a good reason to either create its
>own file type or add itself as alteranative to the file that it handle.
Applications don't handle files; people do, using applications.
>By handling file I mean that the program's is mainly open, changing, and
>using in some way in file with common extention(s).
File extensions are a convenience for the user. Or not, as the case may
be.
>For example, a program that rip CD-Audio has every reason in the world to
>add itself as an alternative to the CD-Audio menu.
Add itself as an alternative? I thought you said they had "every
right"; why on earth would anyone want their app to be an alternative?
>An HTML editor should put himself as an alternative on the HTML file type,
>and maybe ask whatever I want it to be the default.
And maybe it will just make itself the "owner" of *all* .html and .htm
files, and a dozen other extensions, because this enables them to
'integrate' themselves on the users desktop. After all, you bought the
app didn't you (or accepted it after it was shoved down your throat,
maybe); why wouldn't you want it to hijack as many extensions as
possible?
>A program that doesn't handle files directly has no need of it, though.
>I agrees with you that not every applications need it, but most of the
>programs do.
Yes, the problem of programs registering file types for stuff the user
will never see is pretty stupid. A hold over from back when they were
"extensions", and every app would use a certain extension for its file
type. But what do you do about file extensions like .cfg?
>> 90% of the file types listed in your Open With... dialog have no
>> practical, productive reason for being there. To hell with creating new
>> ones; how about getting rid of the useless ones?
>
>I disagree with you on the principal, and agree on practice here.
Then why disagree on principal? Is it just too much of a leap for you
to recognize what an abominable job overall MS has done building a
usable desktop OS? I would think allowing flexible and robust control
over document launching should be high on the list, yet the "File Types"
interfaces that MS provides are disfunctional, at best, and usually
prove to be entirely counter-productive, unless your needs are the most
trivial imaginable (no doubt all you've ever imagined to begin with).
>When I use the Open With dialog (I assume you mean the one for unknown
>files), I want to be able to associate the file with every program that I
>like.
There is only one Open With... dialog. You are correct; it hasn't bee
brought up to date, and still presumes that the default action is the
only one you'd ever want to modify. You know that isn't true,
apparently.
>On practice, I hardly ever use the Open With dialog, and when I do, I choose
>one application to open it with, and ignore all the rest.
That's all you *can* do with the Open With... dialog, unless they
changed that in W2K, as well. Its a pretty pathetic system, overall.
--
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
====== 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.os2.advocacy,alt.destroy.microsoft
Subject: Re: Uptime -- where is NT?
Date: Tue, 28 Nov 2000 12:39:36 -0500
Reply-To: [EMAIL PROTECTED]
Said Giuliano Colla in alt.destroy.microsoft on Tue, 28 Nov 2000
[...]
>> There are multiple timers in Windows. GetTickCount() is simply one the
>> lowest resolution timers.
>
>I didn't pick any particular tick counter. You did. You described how it
>works. It is stupid. If another one is smarter, why didn't you pick the
>smart one?
>Sort of pity for the poor little stupid counter nobody takes seriously?
LOL. :-D I'm still giggling, as I write this. "Poor little stupid
counter..." :-D
>> > Now you have decided to provide a function whose specs
>> > require to expose, as a continuity indicator, a 32 bit
>> > counter which increments in units of 10 ms. Remember that
>> > you may not provide it. IBM's AIX doesn't, just to make an
>> > example. It's just a function you may have or not. But
>> > you've decided to have it.
>>
>> Gee, what spec is that? The only spec I know of that requires a 10ms
>> resolution is the SNMP spec, and we know for a fact that the method that
>> netcraft is using is not based on SNMP since this is filtered by most
>> firewalls and filtering routers. I don't know if the SNMP function returns
>> the correct time or not, I've not tried it.
>
>We don't know for a fact the method used by Nectcraft. But I have much
>less pain thinking that netcraft has devised a way to obtain an
>information from a function which is exposed to the network, than
>thinking that netcraft has devised a way to perform remotely, through
>firewalls and routers, a call to the GetTickCount() API!
Obviously, the web servers do provide some sort of 'uptime' information
via HTTP. I guess nobody knows much about it, though. So other than
saying "they use a 1ms counter" with no further explanation, I guess we
have no idea at all why MS would do such a stupid thing.
Mark's comment did seem to resurrect the possibility it may be my point,
though, given all that. After all, it *is* common, obviously, for
people to presume that a continuity indicator is a clock, and try to
read the uptime directly. In such a scenario, where it is assumed that
a rollover makes the value unusable, it would make sense to "short" the
counter so that it never got noticed the system is never up for more
than a few weeks at a time. Is it too late for me to take back what I
said about giving you the point? ;-)
--
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
====== 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! ======
------------------------------
** 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
******************************
