Linux-Advocacy Digest #404, Volume #25 Sat, 26 Feb 00 14:13:07 EST
Contents:
Re: Linux Gets Worldwide Recognition (Roger)
Re: The real cost of Linux (Greg Copeland)
Re: New respect for Microsoft (Greg Copeland)
Re: The latest from IDC ( Was Re: Linux sales. ) (Greg Copeland)
Re: Microsoft's New Motto (was: TPC-C Results for W2k!! (Greg Copeland)
----------------------------------------------------------------------------
From: Roger <roger@.>
Crossposted-To:
comp.os.ms-windows.nt.advocacy,alt.destroy.microsoft,alt.microsoft.sucks
Subject: Re: Linux Gets Worldwide Recognition
Date: Sat, 26 Feb 2000 18:17:30 GMT
On Mon, 21 Feb 2000 16:31:06 -0500, someone claiming to be doc rogers
wrote:
>"Dr Suave" <[EMAIL PROTECTED]> wrote in message
>news:[EMAIL PROTECTED]...
>> Look up the recent Caldera lawsuit. Put briefly, Caldera sued MS over
>> MS's disabling DR-DOS using code specifically inserted in win3.11 to
>> do this.
And correctly put, Caldera sued over a warning message which appeared
in a beta version of Win 3.1, back when beta was beta and not a
preview release, which warning never appeared in a retail product,
which error warned of a fairly major incompatibility which did exist
at that time, but which DR (with assistance from MS) patched by the
time the product shipped.
>> Their evidence involved two versions of win311, in which the
>> only difference was that running on top of DR-DOS caused windows to
>> display a dialog stating that a bug has occurred due to
>> DR-DOS. MS settled for $150 million.
The evidence was a beta compared with the released product, and
Caldera settled for less that a tenth of the money they had requested,
and none of the other penalties they ought.
<snip>
>> And MS enforces upgrades by drastically changing the word
>> document format every time they release office.
No. The last format change was several years ago, the one before that
almost a decade. And an upgrade is not necessary to support the new
format -- converters and filter have been made available shortly after
release (and if that's not soon enough, the formats are publicly
documented prior to release, so one could code zir own.)
>> It must be incredibly
>> annoying to be forced to uprade your software because a company with
>> which you do business has upgraded theirs, and you can no longer read
>> their documents.
It might be, if that were the case. But it's not.
------------------------------
Subject: Re: The real cost of Linux
From: Greg Copeland <[EMAIL PROTECTED]>
Date: Sat, 26 Feb 2000 18:19:57 GMT
Actually, if I recall, the first article that I found, was in Home Theater
mag or something like that(something that I picked up from the mag stand
- was helping a neighbor collect DVD information). I've slept since then.
I mostly remember the conclusions of the study rather than the technical
details of it (meaning, I don't remember what encoder or the hardware
lists, etc). The second one I saw was maybe 6-8 months latter on a news
site that happen to have links to audio stuff (I was looking for sat
system). From there, I found a link that had the story. I was not at
any place that I would normally frequent. Nonetheless, two studies seem
to indicate the same thing. I do remember that the type of music was
also a factor. Generally speaking, classical music was considered
the hardest to decern by all involved in both tests.
I understand you've observations. I too have acute hearing (I've been
known as the guy with "doggie" hearing because I've heard supposedly
"sonic" motion detectors etc...they obviously weren't, but I was the
only one that could hear them). At any rate, I have good hearing.
I understand that things like vision and hearing tend to have a lot
of weight when it comes to these things. However, this is why double
blind tests are REQUIRED to get valid results. Otherwise, historical,
the results ALWAYS wind up biased toward CD.
Greg
Steve <[EMAIL PROTECTED]> writes:
> Yikes!!!!!
>
> Where in the world did you read that?
>
> URL? References? Magazine?
>
> MP3's are easily distinguishable from their CD counterparts on all but the
> worst systems.
> My recording studio is about $30k worth of equipment, very modest by todays
> standards, and I can EASILY tell the difference, even through a
> SoundBlaster Live card. Note my pro card is an Event Layla, the SBLive is
> in my general purpose PC.
>
> My home system is about 50k dollars or so (Bryston, B+W. NAD and so forth)
> but my wifes system in the den (her sanctity from me!) is a basic Yamaha
> surround system and the difference is obvious between an MP3 and a 44.1khz
> CD file.
>
> The MP3 sounds compressed, short of bandwidth and thin compared to the CD.
>
> Anyone can hear the difference when a comparison is made between the 2. As
> for just playing an MP3 file, without a reference to the CD, most people
> wouldn't even know it was an mp3 that was playing, on any system.
>
>
> These audioPhiles must have had hearing problems :)
>
>
> As for radio, the difference wouldn't be heard due to compression,
> limiting, modulation noise and so forth.
>
>
>
> On Sun, 23 Jan 2000 23:39:25 GMT, Greg Copeland <[EMAIL PROTECTED]>
> wrote:
>
> >Steve wrote:
> >>
> >> On Thu, 20 Jan 2000 17:55:49 GMT, [EMAIL PROTECTED] wrote:
> >
> >[snip]
> >
> >> >Then, there are the multitude of MP3 and MP4 encoders and decoders,
> >> >which solves that little quirk about high-end audio, rather nicely.
> >>
> >> MP3 is not high quality audio.
> >>
> >Actually, I beg to differ. I've now seen several double bind studies
> >conducted with audiophiles on $60,000+ entertainment systems. In both
> >bases, less than %10 could accurately tell the difference between a CD
> >and a high quality MP3. That means, that if you have quality MP3's,
> >that the chances of you telling the difference (since CD's exceed human
> >hearing capacity to begin with), even on a nice $20,000 system in next
> >to nil. This is why many radio stations are trying to switch to MP3
> >(especially since you can't tell the difference over radio - legal
> >issues that they are looking to resolve). In short, anyone that says
> >they can reliably tell the difference between a quality MP3 and a CD is
> >almost certain to be delusional or full of it.
> >
> >> >And finally, there's always the UCL videocasting suite, which allows you
> >> >to transmit broadcast-quality sound and video across the Internet, to
> >> >any number of users, with no loss of quality. (Multicasting is fun!)
> >>
> >> I'm not familiar with UCL but it sounds interesting.
> >> >> automate your company with group ware,
> >> >
> >> >Lotus Notes, Lotus Domino, SourceForge, CVS, CODA, Oracle 8i, IBM DB/2,
> >> >Informix 2000, Bugtraq, Squishdot, Netscape LDAP, IBM Websphere
> >> >Application Server, UCL Video/Audio/Whiteboard/Shared Editor...
> >>
> >> And what client are they running?
> >Anything they want. I personally don't have a problem with people using
> >a cheapo client. It's the backend that counts for the enterprise.
> >
> >>
> >> >Nuff said?
> >> >
> >> >> actually have a choice in GUI based browsers
> >> >
> >> >Opera, Netscape, Mozilla, Jazilla, Amaya, Arena, the web browser that
> >> >comes with Sun's JDK, etc, ad nausium.
> >>
> >> Opera is released for Linux?
> >> News to me, but if it is I will try it. I use Opera under Windows and like
> >> it a lot.
> >
> >Grail and two others that are not listed and I can't think of the name.
> >
> >
> >>
> >> Netscape is unstable.
> >And IE is? I personally have equal trouble with both. I must admit
> >that IE is faster than Netscape, but Mozilla will comletely change
> >that. It's wicked fast even with tons of debug code running and little
> >to no optimizations.
> >
> >> Same for Arena.
> >> Mozilla is not shipping.
> >>
> >> >> and so forth.
> >> >
> >> >Oh, Linux has Forth, too. Gnu Forth, and a few other variants, are
> >> >available.
> >>
> >> Programmer geek stuff ?
> >It's important to a lot of people. Nothing wrong with that. The nice
> >thing is, when you have good tools available, that's where the next
> >application will be developed. This has historically held true. I
> >don't remember who's study it was that I read within this last week, but
> >it said that the majority of developer's are not planning their next
> >Win32 (specifically, W2K) release until 2002. And that's assuming that
> >it makes sense to do so. That alone indicates, if nothing else, that
> >people are looking elsewhere. Furthermore, that are a TONS of really
> >nice environments available for Linux. Take your pick. Each day more
> >companies announce Linux support; most plan to follow through (unlike
> >Creative).
> >
> >With Microsoft's recent legal loss, their history of well....it goes
> >without saying, and lack of stability on any of their platforms, Linux
> >for backend is EXTREEMLY attractive and looking more so on the desktop
> >daily.
> >
> >Greg
------------------------------
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: New respect for Microsoft
From: Greg Copeland <[EMAIL PROTECTED]>
Date: Sat, 26 Feb 2000 18:30:00 GMT
"Erik Funkenbusch" <[EMAIL PROTECTED]> writes:
> Greg Copeland <[EMAIL PROTECTED]> wrote in message
> news:[EMAIL PROTECTED]...
> > > The distinction between "micro kernel" and "monolithic" kernel is
> > > getting blurred more and more every day.
>
> > Not in my eyes. A microkernel is a minimalist implementation of kernel
> > services, whereby, everything is handled by user space implementations.
>
> Not everything. The definition of a Microkernel is that "as much as
> possible" is handled in user space.
>
> Indeed, NT has 2 kernels. One, the NT native kernel, and the other the
> Win32 subsystem kernel.
I think we are saying the same thing.
>
> But no, NT is not a pure Microkernel. At best it can be called a modified
> Microkernel. Like many real-world uKernel implementations, compromises had
> to be made due to the inherant overhead of a pure uKernel.
>
> > The big advantage of doing so, by intent and convention, is to ensure
> > that as much of the "kernel" remains in the CPU cache as possible. In
> > short, a microkernel is usually nothing more that a scheduler and a
> > queuing subsystem with a well established interface. Thusly, user space
> > applications provide system services like memory, logical IO, buffering,
> > caching, fs, etc, etc, etc. I've also read about user space scheduling
> > for some microkernels, but this always seemed like a core service to me,
> > so I'm not actually sure how common that is.
>
> There is really no way that memory management can be done in user space on
> the intel architecture. modifying the page translation registers is a
> ring-0 function. Additionally, several types of I/O cannot be achieved
> through anything but ring-0 as well.
Actually, I was not very clear on this. There are ways to do memory management
in user space. Basically, the kernel handles the memory request and passes
to a user space to manage. Obviously, I was not entirely clear here. Nonethe-
less, user space schemes can and have been devised.
>
> > I think you need to look again. I didn't say or imply that loadable
> > modules = microkernel. Furthermore, loadable modules (at least in
> > Linux), do not run in user space. They attach and run at kernel
> > priority. They are, exactly as described, a dynamically loadable kernel
> > component. Loadable module doesn't imply user space service either.
> > There is nothing exclusive or inclusive going on here.
>
> Ahh.. but here is where things get blurred. Many portions of NT run in
> "kernel space", but do not run in the thread context of the kernel. They do
> not run at kernel priority either (unless you mean ring-0, versus a
> scheduled priority).
>
> > If it runs at the same priority as the kernel, it is part of the
> > kernel.
>
> It does not run at the same priority level. The reason for moving the GDI
> and USER into kernel was to prevent multiple address space context switches
> from occuring as each GDI or USER function executed. GDI and USER run on
> their own threads at their own priorities.
I understand all that. I can see why you considered it to be "blurred." To
me, don't know exactly why, but my guys says it's part of the kernel. Since
there isn't a clear separation between kernel and user, I tend to say that
it's kernel. To this point, I would allow NT to be a mono-micro kernel hybrid
weighing in much more heavily on the mono side.
>
> > There is no way to make a distinction other than by simple,
> > logical semantics. In other words, is a device driver part of the
> > kernel? Logically, it may not provide a core service and uses an
> > established system interface, yet, it runs with the kernel. I have no
> > problem following the commonly accepted notion that it is part of the
> > kernel. To me, it's either kernel or user space. If you can
> > recategorize, I'd be interested to hear about it.
>
> It's not easy to categorize. If your sole definition of what's "in the
> kernel" revolves around the fact that it lives in the same address context,
> then OS's like AmigaOS, MacOS, and even Windows 3.x mean that all
> applications run "in the kernel".
>
> Clearly there is some differentiation between the kernel and the application
> in these contexts.
Keep in mind, in many of those contexts, the kernel pretty much is an application,
likewise, an application can become part of the kernel. So, in short, that's
one of the main reasons my many complain that DOS/Win/Win9x barely meet the
minimum requirements of a true kernel/OS because it's such a blurred mess.
Greg
------------------------------
Subject: Re: The latest from IDC ( Was Re: Linux sales. )
From: Greg Copeland <[EMAIL PROTECTED]>
Date: Sat, 26 Feb 2000 19:05:12 GMT
[EMAIL PROTECTED] (Donovan Rebbechi) writes:
> On Sat, 26 Feb 2000 12:21:29 GMT, Greg Copeland wrote:
> >Hehe. This is a wonderful example of what I was commenting on before.
> >Maintaining meaningful context goes a long way. You're whole argument
> >is based on the completely wrong assumption that I've presented my
> >comments as fact.
>
> So it would appear as though I've completely misunderstood your post. It
> may help if you quantify opinions with the well known "IMO" or "IMHO"
> quantifier, to make it clear that you are offering an opinion.
I agree. I went back and read. I however, would like to point out that
phrases like, "This is where support revenue will/should start.", tend to
be a give away that at a minimum some level of speculation is involved.
Having said that, you point is well taken. I will try harder to use such
important clues as to the nature of the comment.
>
> > To address ALL of your points, one need only ignore
> >everything you've said and request that you please learn the difference
> >between someone offering fact and someone offering opinion.
>
> Again, this is not always clear. We've seen in this thread people like Jedi
> who present their personal opinions ( that physical property is "real" and
> intellectual property isn't ) as some sort of universal truth.
>
> >Removing the context and making it look like I was presenting fact and
> >not opinion is an incredible poor practice.
>
> If you feel I've misrepresented your position, I'm sorry. This certainly was
> not my intention.
>
You actually seem pretty reasonable. :P~
> > In case it's not already
> >obvious, asking someone to present facts on events that have not
> >occurred yet, especially on topics whereby, there is little to no trend
> >data available is ridicules!
>
> My point is that you cannot cite something as supporting evidence for your
> argument if it hasn't occurred yet.
If it's offered in such a way that it is supported by a logical manner,
I'd beg to differ. Otherwise, no one would ever be able to make a
forward looking statement. To that end, it would make it difficult to even
make basic trend analysis in many cases.
>
> >The original question was asking about a market model that will support
> >Linux. I offered one based on events that are shaping the market. If
>
> The point I was making is that there's really no hard data to indicate
> that these "Linux companies" will be succesful. You seem to be presenting
> reasons why you believe that they may be succesful.
Yes, I agree. In fact, I will offer that many Linux companies do not seemingly
have a good business model in place. I think if you look at what the likes
of Redhat is doing, they understand and acknowledge this. They are actively
seeking to create new business model. Will it work. I don't know. Nonetheless,
I offered my argument of yet another business model (Corel's) and why I thought
it will work. Needless to say, even without a strong model in place, these
companies have sucessfully generate large sums of capital that will allow them
to try many models before the venture it self can be considered a failure.
This, I hope, you will yield to me.
>
> >P.S. There is one exception to my above comments. I did offer that
> >Linux is having market growth. This fact is well understood and
> >accepted.
>
> Having market growth doesn't help that much unless you're either already making
> some kind of operating profit, or you can service a larger market without
> a proportional increase in costs. In other words, I acknowledge your fact,
> but I have doubts as to how much it helps. Another problem is it's really
Okay, let me put it this way. Very simply put. They are creating a market
to leverage for other business models. You will surely agree that having
a foot in the door is the first start to any successful business. This is
what they are wanting to leverage. If you have Corel Linux, when why not use
other Corel products. This is a common tactic that other companies use in
more mainstream models (Microsoft being one of them; OS and apps).
> hard to accurately measure Linux growth though I'm pretty sure it's growing
> ( on a side note, we just started up a user group at our strongly
> windows-centric school, and I'm seeing Corel box sets in the campus
> bookshop ... )
>
> To clarify my position further, I don't really care whether or not Linux
> makes money. To me, OpenSource is not just about money ( or even not at
> all about money ).
>
> --
> Donovan
Agreed.
Greg
------------------------------
Crossposted-To: comp.os.ms-windows.nt.advocacy,comp.sys.mac.advocacy,comp.unix.advocacy
Subject: Re: Microsoft's New Motto (was: TPC-C Results for W2k!!
From: Greg Copeland <[EMAIL PROTECTED]>
Date: Sat, 26 Feb 2000 19:07:15 GMT
Bob Germer <[EMAIL PROTECTED]> writes:
> On 02/26/2000 at 12:01 PM,
> Greg Copeland <[EMAIL PROTECTED]> said:
>
> > > > "Microsoft -- We're the company you can trust, because,
> > > > although we used to lie all the time, now we've stopped."
> > >
> > > ROFL.
> > >
> > > You know, in the rare instance that you abstain from daft conspiracy
> > > theories, you can be quite entertaining.
> > >
> > > --
> > > Donovan
>
> > While it certainly is funny, it does happen to correctly depict
> > Microsoft.
>
> A better motto is:
>
> Never blame on incompetence what can be blamed on malevolence.
Mmmmm....don't know that I really buy into that one. Incompetence
is too easy to find in the real world.
Greg
------------------------------
** 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
******************************
