For those so inclined, this is a rather lengthly yet interesting article. After two days I've only just started chapter 3.

http://www.euronet.nl/users/frankvw/IhateMS.html

Anthony Kerstens P.Eng.

 

There really isn't much new in this rant. However, it's a nice summary of why so many literate technical people shudder whenever someone mentions support for a Microsoft product or platform.

I'm being managed by a bunch of folks who have never known anything but Microsoft products. They think that hacking software until it behaves is normal. They don't see anything wrong with not knowing why some software crashed --as long as you can reset something that will fix it.

And since the true scope of the open source market is so difficult to document, they have a "follow the market" strategy and it blindly takes them wherever Microsoft wants to go.

I'd call them fools and ignormuses, but they're not. They're just a bunch of all too human managers who are unwilling to try anything radically different. They will only move to keep up, not to forge new territory. If the people around them aren't changing, then neither will they.

What I've seen of the Automation industry seems to show that we're overloaded with managers such as these. Open Source Automation software is still in its infancy. Let's show the PHB types how REAL automation is done...

 

I read a few chapters myself. What I see is a guy extremely jealous of a Bill gates for managing to sell a technically inferior product because he has good business and marketing skills.

It should come as no surprise to anyone in the technical world that having the best technology is no guarantee that your product will be a commercial success.

And quite frankly, unitl recently, the MS product just plain was the "best" available single user/home user product available. Linux certainly has the advantage for servers, but for standalone Pcs for the home and office market, Windows wins hands down.

Bob Peterson

 

Hi Anthony

Great link as it provides perspective and history that many folks today aren't at all aware of. Actually a fairly charitable treatment as it doesn't mention the empty buildings, fire sales, home auctions, and the human wreckage wrought by these shining examples of commercial success who negotiate in the same manner as the highwaymen of old. They tended to be highly successful too, as long as they payed the right people.

Regards

cww

 

An interesting article in that rather than being just another diatribe on how unethical Microsoft is (we hear plenty of that already), he actually discusses the fundamental design flaws of Windows, and why they can't be
corrected without tossing the existing Windows out and starting over. Perhaps a better title would have been "Why Windows Cannot be Secure or Reliable". If someone other than you had been the one to post the link, I'm not sure I
would have bothered to read something with a title like that.

Of course knowing just exactly why the problems with Windows can't be fixed (it apparently isn't simply a matter of eliminating the bugs) isn't going to stop anyone here from using it. The problems with security and
reliability (and scalability and performance and etc.) which so exercises the gentleman who wrote the web page are not likely to cause much concern to people on this list. We've come to have rather lower expectations from our PCs than he is used to.

Not many of us trust PCs enough to do anything with them where reliability is critical. We've become used to the idea that computers are unreliable because the ones which we use every day are unreliable. To be quite honest,
solving the "blue screen of death" isn't high on my own list of priorities either.

If you want something to worry about, then worry about this. If you take Microsoft's plans for ".NET" at face value, then today's PC as a stand alone independent computer will disappear and become a client which won't run except as part of a larger integrated IT system. This seems to be the direction the office IT market wants to go for reasons of cost (reducing man hours in managing and maintaining office systems). I don't know whether that
is a good thing or a bad thing from a user's perspective, but I don't make those decisions.

This by the way isn't an issue of whether Microsoft is "good" or "bad". There is certainly nothing wrong with them changing their product line in a way which a lot of large scale IT managers have told them they want to go.
Microsoft has a good idea of where their main market is, and they are going to pursue it. The stock market analysts have also been telling them they need to become more of "services" company, and less dependent upon irregular
revenues from selling software, so I guess a change like this would make the money people happy too.

This may not be a problem if all you want to do is word processing or e-mail, or even CAD. But what if that *doesn't* fit what you are trying to do with a PC? What if you are part of that 0.1% of the market that needs an old
fashioned stand alone computer system with complete control of its own resources to do things like run a piece of test equipment?

If that situation comes about then you might have to switch to a different operating system which is still able to address that 0.1% of the market. That's not a disaster, unless that is the application software you've come to
depend upon can't be economically ported over to something else because of excessive dependence on someone else's proprietary way of doing things.

So here comes the conclusion of this rather long story. If you want to think about the long term viability of systems you are responsible for, you need to think about minimising your dependence on any particular computing platform. You may never need to actually switch platforms, but that option should be there as insurance if you need it.

To do this though, you also need to have a close look at any critical application software which you depend upon (SCADA, MMI, soft logic systems, etc.). It would be nice to be able to leave this whole problem to the application software publishers and count on them to provide the portability when it's required. However, what if their software design has painted them into a corner and their only viable business option is to declare bankruptcy?
It wouldn't be the first time *that* happened in the software business. To some of the clever fellows who create the sales strategies in these businesses, 6 months is long term and a year is an eternity. Anything beyond
that is unimaginable to them.

So I guess you need to look carefully at software companies not only from the perspective of what flashy features they have, but also how vulnerable they are to changes which are outside of their control. If they depend too heavily upon a third party's proprietary features (particularly platform dependent features), then they present a higher risk to your business than if they use open industry standards.

I don't see the problem I have described as a type of "doomsday" issue such as Y2K was presented as being. It is something which is quite forseeable and the consequences of which can be avoided or at least minimised with a bit of forethought and the willingness to make the right decisions.

However, if it does come to pass, then I won't be surprised to see a great deal of wailing and moaning going on as people blame Microsoft for their situation rather than take responsibility for the decisions which got them
into their mess in the first place.

--

************************
Michael Griffin
London, Ont. Canada
************************

 

Hi Bob

Yes, it's obviously envy and jealousy that brings people to contribute to the common good. And all that hogwash about ethics, legality and moral bankruptcy is just sour grapes from people who wish they could be as corrupt and ruthless and unethical in the name of making a buck. I've had an epiphany!

Regards

cww

 

Bob Peterson:
> I read a few chapters myself. What I see is a guy extremely jealous
> of a Bill gates for managing to sell a technically inferior product
> because he has good business and marketing skills.

Hmm, I don't hear much jealousy in it, more like despair at the devastation wrought.

Jiri
--
Jiri Baum <[email protected]> http://www.csse.monash.edu.au/~jirib
MAT LinuxPLC project --- http://mat.sf.net --- Machine Automation Tools

 

> I'm being managed by a bunch of folks who have never known anything but
> Microsoft products. They think that hacking software until it behaves
> is normal. They don't see anything wrong with not knowing why some
> software crashed --as long as you can reset something that will fix it.
>
> And since the true scope of the open source market is so difficult to
> document, they have a "follow the market" strategy and it blindly takes
> them wherever Microsoft wants to go.
>
> I'd call them fools and ignormuses, but they're not. They're just a
> bunch of all too human managers who are unwilling to try anything
> radically different. They will only move to keep up, not to forge new
> territory. If the people around them aren't changing, then neither will
> they.

I agree 100%.

The comments posted above just describe what happens at almost 90% of the companies here in Japan. :-(

Wilson Pardi Junior

 

Hi Michael

I think you've hit on an issue while skirting around it. It's who controls what's available. This shows up in a lot of ways in our world. Not just the .NET stuff.
For example:

Who decides when it's time for ISA cards to go away.
Who decides that serial ports need to go away.
Who is driving to put security in firmware,
to make hardware and software dependant on each other which also conveniently locks out non-serfs.
Who can determine the success or failure of a peripheral simply by supporting it or not.
Who decided that we're all gonna move to a post PC world.

At one time the market decided these sorts of things.
But lately free market control has been preempted by a monopoly that seeks to control every aspect and has the power to do it. There are quite a few who would argue that the market is still in control, but as a practical matter, If MS stops supporting serial ports, they will go away. If I bring out a card or device that MS doesn't choose
to support, it's DIW. And if they succeed with their plans for hdds, CD-R/RW, Platinum, etc, just about everyone is going to have to give up their freedom and control of their computing environment. The market can still exercise control
but it's an all or nothing thing. You have to take the bad with the good with MS or select an alternative. So it's really worse than "the market has decided", because that's a given and has been leveraged to the Nth degree. Nearly to the point where there is no choice for the "market" to make and the market only thinks they have a choice. Just a little further, to widespread .NET deployment and computers that won't work with "unapproved" data and what choice will there be?
You are right that this should be obvious, at least to IT professionals. But as you said "we don't make those decisions".
The PHBs and uninformed are making _non-decisions_ and the default is what has got us here on the brink. My only hope is that _those_ folks begin to think and make decisions.

Regards

cww

 

So anyone who provides software that is not OSS is not contributing to the common good? I realize you have a major issue with corporations making money by creating and selling software, but you can't really believe the hogwash you are spouting. There is nothing unethical, or immoral about doing so. There are some legal issues involved, but most of these legal issues are
secondary.

Bob Peterson

 

I will grant that the author (frank v w) writes as though he has a massive chip on his shoulder (to put it nicely). However, in reading this I have tried to ignore that because he seems to know what he's talking about.

What has captured my interest is the compilation of facts, figures, and anecdotal information (albeit unverified) with the added
"historical" context. It's always nice to have an understanding of where things came from and how they developed.

That said, although I find some of it ringing true and want to believe it, the tone of the article does the information a serious disservice.

If anyone has information to either verify or debunk anything mentioned in this article, I would appreciate it.

Anthony Kerstens P.Eng.

 

Most Microsoft bashing is a form of sour grapes, but at least this article gives some details - like the explanation for what some people call "DLL Hell".

This link http://www.winntmag.com/Issues/Index.cfm?IssueID=97 introduces another point, Is NT Windows? The answer, of course, is yes and no. When I read this article I was impressed that the Microsoft team could kludge together a replacement for VMS is such a short time while producing only a few thousand bugs. Blue Screen of Death? We were lucky!

More to the point, I think, is the struggle inside IBM to cripple the early Windows, to avoid competition for Big Iron. This left us with no security, no internal passwording, and a single-user mentality/philosophy that could not even be safely shifted to multi-tasking! Poor Gary Kildall at least put passwording in CP/M 86. If only he had some marketing sense. Or was it his wife.......

The NT/VMS link starts out:

"December 1998 :Is NT Windows?

Just when NT has begun to gain credibility as an enterprise OS, Microsoft insists on subsuming NT under the Windows brand. But is NT Windows? Not really. Mark Russinovich, a well-respected internals expert, proves that NT is directly descended from VMS, an established enterprise OS. In addition, Russinovich demonstrates some important resemblances between NT and its enterprise rival, UNIX."

 

On December 17, 2002 04:42 pm, Curt Wuollet wrote:
<clip>
> I think you've hit on an issue while skirting around it.
> It's who controls what's available. This shows up in a lot of ways in
> our world. Not just the .NET stuff.
> For example:
> Who decides when it's time for ISA cards to go away.

ISA, VIESA, Micro-channel, STD, Multi-bus, S-100, Nu-bus, etc., all went away or are going. ISA cards are long overdue for their demise in new applications and PCI will in turn also pass away in time. PCI is technically superior, and is widely supported by a lot of different software, including all current PC operating systems.

ISA will disappear regardless of any other considerations. If you really want ISA support now (and many people on our business do), then try one of the passive backplane PCs which are available. I've bought a few, and they were much cheaper than I expected.

> Who decides that serial ports need to go away.

Not many people outside of our business use RS 232 for anything anymore. It was originally built in as a printer interface, and later as a mouse or trackball interface. Can you think of any modern office uses for RS-232 anymore? I can't outside of simple UPS interfaces (any very few people have these). What justification can you think of for the cost of putting the hardware in standard office PCs?

I think this more or less is a hardware example of what I was referring to when I said that the standard office PC environment may be diverging from what we need to use in industrial applications. If you need an RS-232 interface and a desktop PC doesn't have one, then use better hardware. If the operating system doesn't support it, then use a different operating system.

> Who is driving to put security in firmware,
> to make hardware and software dependant on each other
> which also conveniently locks out non-serfs.

This is a different issue than I was trying to discuss. It is however something people have to be very careful of as it may cause problems with the maintainability of their system. This is something the customers will have to deal with in their project specifications, as I don't think most software integrators have much incentive to try to solve it unless the customer makes solving it a requirement of getting the project.

> Who can determine the success or failure of a peripheral
> simply by supporting it or not.
> Who decided that we're all gonna move to a post PC world.

Who decided we are all going to "move to a post PC world"? I believe that I did point out that many company IT managers have been asking for this
(although not necessarily for ".NET"). This can probably be more properly characterised though as "back to a pre-PC world" - back to the mainframe or mini (except now they are called "application servers").

> At one time the market decided these sorts of things.

Don't you mean that at one time IBM decided these things? That is they did with the mainframe market until the American government forced them to open up their proprietary mainframe interfaces and unbundle their hardware and software packages to allow "plug compatable" competition.

Yes the computer industry seems to be prone to monopolisation both now and in the past (probably due to defective laws and regulations concerning patent and copyright). There was a period of competition and intense innovation between the OS360 and Windows eras. A similar period of innovation may occur when Windows joins its predecessors in the dustbin of history.

However, let us assume that there was in fact a competative market today. I think that the point I was raising would still be relevant. As the office environment changes its needs may diverge from what is required in industry to a point at which it is no longer reasonable to attempt to use office derived software for many industrial applications.

People need to be prepared for this, both software vendors and their customers. The customers need to look very closely at what vendors are offering them to ensure they don't lock themselves into a product line which is so rooted in one proprietary way of doing things that it effectively has no future.

I would like to point out that there are two types of business risk involved. One is that a company may go out of business. This is the lesser risk. If its products are unique and have merit, then someone will usually buy them from the bankruptcy trustees and continue on with them. If there are many close competing substitutes, then someone will find it worthwhile to capture those customers by making it easy for them to switch to a competing product.

The other type of business risk is where you are using a product for a purpose for which it was never intended, and for which there are no close substitutes. Since the product is not intended for your market, there is no reason why its future design would continue to make it suitable for your purposes.

This is a much greater risk than the previously mentioned one because there isn't anyone with a close substitute on hand to bail you out. The company which sells the original product doesn't have to go broke to leave you stranded, they just have to continue to follow what they perceive as being their real market. Their efforts (and those of competitors they may have) will be directed towards moving things *away* from where you need to be, not towards solving your problem.

I think the answer to this would be (as I said previously) to look very closely to see what would be required to move an application to a different platform (including different operating system, but also data storage, networking, etc.) You may not ever have to actually do this, but this is a form of insurance in case you do. If the answer looks to be overwhelmingly difficult and expensive, then you definitely have a problem and should be looking for other ways of accomplishing what you want to achieve.

I'm not saying that you should look for an answer which requires the flick of a switch. It should however be an answer however which looks reasonably straight forward, and can be done in a piecemeal fashion.

This is particularly significant as factory systems become more integrated with each other. You don't want to (can't!) have to have to rip out and replace everything at once just because the web server won't talk to the database which won't talk to the network which won't talk to your production test system because you can't use the secure multi-media internet digital rights mumbo jumbo in a real time environment.

I can go on about this subject if anyone is interested in discussing it. My point is not so much a question of whether Microsoft is good or bad as what we see as how the future of automation software (or more accuractely, possible futures), may be affected by events from outside of our own industry. A lot of the discussion here about the direction of the industry tends to be too narrowly focused on the short term.

<clip>
> So it's really
> worse than "the market has decided", because that's a given
> and has been leveraged to the Nth degree. Nearly to the point
> where there is no choice for the "market" to make and the
> market only thinks they have a choice.

Which market? The office market or the industrial market? The two are not the same. For a while the software markets for both have followed each other on closely parallel paths, but what happens when they start to diverge?

<clip>
> The PHBs and uninformed are making _non-decisions_ and the
> default is what has got us here on the brink. My only hope is
> that _those_ folks begin to think and make decisions.
<clip>

Actually, a lot of the bad decisions are made at a fairly low level by the customer's engineering staff. I've probably made a few myself. A lot of bad decisions get made through neglect or laziness or because people don't want to admit they don't understand the technology and won't ask someone who does.

I was going to write something general about how everyone needs to work together etc., etc. However, thinking about it some more, I realised that that isn't how markets work. Markets don't work by the participants having perfect foresight and "knowing" what the future will hold.

What really happens is that the market dinosaurs get wiped out by an asteroid and the survivors are the companies which were lucky enough to have designed their products in a way which allowed them to adapt to the new conditions.

So I guess the real question is, when the upcoming IT technology asteroid strikes, are the companies you rely on to supply you with what you need to run your plant dinosaurs or mammals? Do they tell you that they "know" what the market will be like in 5 years time so they don't have to make any plans, or are they adaptable?

************************
Michael Griffin
London, Ont. Canada
************************

 

I am not all that sure it really matters if the anecdotes he presents are true or not. The fact is that it is very common for marketing and business accumen to win out over "better" technology. A few examples:

- Beta versus VHS. Beta had several technical advantages.
- 3 phase power versus single phase power. Its pretty clear that there are technical advantages to supplying homes with 208/120 service instead of
120/240 service. But there are other reasons why it is not done (cost and need being the big ones).
- Wankel engines. remember them?

There seems to be a few people who just cannot accept that the "best" technology just does not always win out. This is often seen in personnel
decisions as well. Look what happens when corporations downsize. Often the most technically astute ("best") engineers are let go and the lesser lights kept. This is an internal political decision made by marginal managers.
They do not want to have to be compared unfavorably to the few people who are left standing when the dust settles. They would much rather be compared to the less talented because it makes them look relatively better. Its one of
the reasons why you frequently see laid off engineers returning shortly after wards as contractors.

Even if we accept as gospel what is presented, is it really relavent that Bill Gates may have partially developed his BASIC interpreter using computer time he acquired in less than a straightforward way? Much of the innovation
that occurs happens this way. Its just not possible to innovate on schedule and as planned from higher headquarters.

The anecdote about DR refusing to sign IBM's form has been around a long time. It might even be true. Another story that floated around was that he deliberately stiffed them because of his dislike for IBM. Is any of this Bill Gate's fault? Or that he was able to take a piece of software written by a 3rd party, rehash it and repackage it quickly and resell it? Thats just
good business sense, whether you like it or not.

My biggest problem with the MS haters is that there was never anything available that was really substantially better for the purpose. Its like arguing the relative merits of a chevy versus a plymouth. Both got you where you wanted to go, at similar prices, and with similar features. Arguing over the minutia of whther a plymouth engine was better than a dodge is really not important. Whats important is that the market place makes those decisions and companies do things to help the marketplace make decisions that favor their product. To do otherwise is sheer business suicide.

Today there are but three options for personal computers. MS, Mac, and Linux. The fact that a "free" OS like Linux has never been able to grab a substantial part of the PC market ought to tell the MS haters that there is something else at work here. Many people think Mac is inherently "better" than MS as well. So why doesn't Mac take over the world? Its because Mac
and MS are close enough that either will work for most PC users equally well. MS systems tend to be less expensive and have a much greater variety of hardware available. Mac systems are touted as being easier to use and "better" in some applications, but these are not the typical applications the vast majoity of PC users are looking for.

One of the list's Linux pushers has indicated his disdain for Lindows (a Windows emulator running on top of Linux) as (paraphrasing now) not being
technically all that great. So what???? If it gives MS some competition it has not really had before, I think thats good. Nothing has to be perfect to be a commercial success, and often getting into the market as quickly as possible is more important than getting there perfect, which seems to be another gripe of the article writer.

I tend to agree that some of MS activities were probably borderline illegal, only because of their market position and the odd way we handle monopolies in this country. If they had 30% of the market and had done the same thing, no one would care. Everything they did is done by other companies all the time in other markets.

Bob Peterson

 

Anthony Kerstens wrote:
> What has captured my interest is the compilation of facts,
> figures, and anecdotal information (albeit unverified) with the added
> "historical" context. It's always nice to have an understanding
> of where things came from and how they developed.

Some of the stuff he talks about are valid, but like most good demagogues, he takes minor facts and spins them into something more than they actually are.

For example:

DLL Hell and the versioning problems. Yup, that's a problem. However, he makes no mention of the solution for this, namely, COM aka OLE, which indentifies components via a GUID.

The "100% CPU utilization of the Control Panel under NT4 SP3". So in one very obsolete update of NT4, there was a problem which has since been fixed. If I pulled up the Linux kernel updates, I can find more than one instance where a bug made cpu utilization go to 100%.

The "different language version get mixed up problem." This isn't Microsoft's problem in the first place, its the other application vendor.
In fact, I'm trying to figure out which OS has done a BETTER job of internationalization that Microsoft, who has had right-to-left versions of
Windows (Hebrew, Arabic) for many years now.

The "Windows' lack of an adequate repair or maintenance mode". Since Windows 2000, you HAVE been able to boot into a single-user, text mode
repair DOS-like command system which can copy DLL files and restore the registry.

"Unchecked buffers, unverified I/O operations, race conditions, incorrectly implemented protocols, failures to deallocate resources, failures to check environmental parameters, et cetera ad nauseam..." Heck, I can find those
on Linux too. BIND anyone? The DNS server that runs on most UNIX system has had more vulnerabilities and security holes than I can count.

"A good example is Windows' long filename support. In an attempt to allow for long filenames, Microsoft deliberately broke the FAT file system. They stored the extension information into cross-linked directory entries, which is probably one of their dirtiest kludges ever..." But you know what? Its BACKWARDS compatibile. I can probably take programs from the mid-80s and run them on Windows XP as-is (assuming they don't try to access the serial
ports <g>).

"An internal memo among Microsoft developers mentioned 63,000 (yes: sixty-three thousand) known defects in the initial Windows 2000 release." Defects. I'd love to see exactly their definition of a defect, and then compare a Linux distribution with its various bug lists to count how many they have.

....

At this point I gave up.

The guy started at the "Microsoft sux" approach and then cherry picked facts to meet his original hypothesis. He ignores the fact that all this
"technical sub par" stuff is so damn easy to use for simple stuff like file and print sharing.

I use Microsoft stuff on a daily basis. I use Linux stuff on a daily basis. There are advantages to both, and there are things that suck about both. If it does the job -- use it.

Alex Pavloff - [email protected]
Eason Technology -- www.eason.com

 

Hi Bob
Let's be honest.

> cww wrote:
>>Yes, it's obviously envy and jealousy that brings people to contribute
>>to the common good. And all that hogwash about ethics, legality and
>>moral bankruptcy is just sour grapes from people who wish they could
>>be as corrupt and ruthless and unethical in the name of making a buck.
>>I've had an epiphany!

> So anyone who provides software that is not OSS is not contributing to
> the common good?

I didn't say that. But,the author is a contributor and I think it's ridiculous to attribute his dislike of an adjudged illegal monopoly to envy or jealousy. That's simply an attempt to discredit him as if there are no valid reasons for disliking MS.

>I realize you have a major issue with corporations
> making money by creating and selling software, but you can't really
> believe the hogwash you are spouting.

I have absolutely no issue at all with corporations or individuals making money by creating and selling software or whatever thay wish to create and sell. Neither I or the author even hinted at that.

That would be rather hypocritical as I create and sell stuff, as well as contributing what I can. Let's quit changing the subject. My issue is only with unethical, unfair, coeircive, and anticompetitive business practices. And abuse of monopoly power. The telephone monopolies were a good thing too, for example, before the power was
abused. I remember when you had to justify making a long distance call and you could only use "Bell approved" equipment. Where would we be today if that was allowed to continue? Answer, about where we are in computing with a dominant monopoly holding all the cards.

And it seems hard for anyone to stay on track and address what the issues really are. No one is arguing that commerce is bad. Surely you can't be arguing that screwing over everyone you can is good?

> There is nothing unethical, or immoral about doing so.>

If done in an ethical manner, I agree. The line is drawn when the majority of customers feel gouged or are compelled to do things that are more in the company's interest than their own. Business is a fair balance of interests between producer and consumer. And in no civilization is it allowable to do absolutely anything as long as
you make money doing it.

>There are some legal issues involved, but most of these legal issues are secondary.>

They become legal issues after the ethical and moral issues are ignored. Most companies never face indictment. It's a last resort for the incorrigable scofflaws.

Regards
cww

 

Look at the bankruptcy records and suits filed in the state of incorporation. The bodies are there to see. And you don't sway the AG's of states or the USDOJ with groundless charges. And all these incidents were widely reported on in the media. Strangely, as long as the stock price kept going up, only the victims payed attention as to how. Any of the trade rag archives will show the major events that took place.

Regards

cww

 

The article does not make that claim, and neither does Curt here.

The claim the article does make is that the particular methods employed by a particular company are unethical and immoral. The fact that they sell software is secondary here (though of interest to us as computer professionals).

It also makes claims about the quality of that company's products, and that this is a deep, unfixable problem rather than a surface one, but you weren't disputing that, indeed you were agreeing with it. There used to be a time when computers were as reliable as PLCs are today, which is not only not the case today, but actually unimaginable to most.

> There are some legal issues involved, but most of these legal issues
> are secondary.

Most laws on the books are there because society found the behaviour in question to be problematic for some reason. Abusing a monopoly because it distorts the market and gives a (non-democratically-elected) party excessive power, infringing intellectual property (piracy, to use Microsoft's own term) because it deprives artists of livelihoods.[1]


Jiri

[1] Those are the two I can name off the top of my head as actual verdicts againt Microsoft.
--
Jiri Baum <[email protected]>
http://www.csse.monash.edu.au/~jirib
MAT LinuxPLC project --- http://mat.sf.net --- Machine Automation Tools

 

You're right in that the author seems to have a serious mad on about Microsoft and Windows. You have to skim over the condemnations to get to the actually interesting stuff.
The author writes from the perspective of someone working in the large scale IT business. He points out that OS crashes or lock-ups are *not* normal events, and that there is no reason why a reliable operating system running on a server cannot run for long periods of time (years even) without problems. This is reliability not as something unusual like sighting a rare bird, but as what should be expected as the *normal* behaviour of software. He says that this sort of reliability is normal for BSD or Netware, and nearly impossible with any version of Windows (this article isn't a Linux or open source paean, so he doesn't spend much time on that).

If you are looking for the statistics to back up what he says, there is a link to a site which maintains a list of the most reliable servers visible on the internet. When I looked, of the 50 listed for highest average uptime, 48 ran some form of BSD (and the remaining 2 used Linux). For the application software, 49 used Apache web server, and the remaining one was an Oracle database server. Windows (in any form) or IIS didn't even make the list.

It was all rather interesting, although I'm not sure how I would make use of this information in anything that I do. However, I will admit that I am a lot less sceptical of a certain gentleman on this list who has been plugging server based process control systems. He may after all, have a point.


************************
Michael Griffin
London, Ont. Canada
************************

 

I think that most people will be a bit disappointed if they follow your link. The lead article is protected and you apparently must have a subscription to the magazine to read it. Since I don't subscribe to "WinNT Magazine", I'm afraid I wasn't able to find out what it was they had to say.

I did read some of the unprotected articles though, including the introductory editorial which seemed to summarise the lead article. The main thrust of the arguments seemed to be that Windows NT wasn't like Windows 95 at all ("NT is not Windows"), and had a lot of resemblance to VMS or Unix. Whether it resembled VMS or Unix depended upon which article you happened to read.
The general tone of this magazine issue seemed be like those heraldry agencies which will guaranty (for a suitable sum of money) to find that you are indeed somehow related to the kings of England or any other royalty or nobility you happen to fancy. The magazine writers did seem oddly anxious to find resemblences to other operating systems and to dispell rumours of relationship to such peasantly software as Windows 95 or DOS. Quite frankly, I didn't think that DOS or Windows 95 were really so bad that it was necessary to obscure the relationship.

Computer magazines which are oriented towards specific brands are rather like car or stereo magazines in that you will seldom find any serious discussion of the relative merits of the products from the larger companies which buy ads in them. Since advertising is their main source of revenue (not your subscription fee), they can't afford to offend major advertisers. They can really only afford to be objective with companies that don't have much money or who don't advertise there anyway.

> More to the point, I think, is the struggle inside IBM to cripple the
> early Windows, to avoid competition for Big Iron. This left us with
> no security, no internal passwording, and a single-user
> mentality/philosophy that could not even be safely shifted to
> multi-tasking!
<clip>

As I said, I'm not a subscriber to "WinNT Magazine", so I couldn't read thelead article. However, I seriously doubt that IBM was worried about competition for their mainframes from the early versions of Windows. PCs had already eaten deeply into that market without it. The early versions of Windows ran on top of DOS and the 8086, and couldn't provide meaningful security or memory protection anyway.
The "single-user mentality/philosophy" was the whole *idea* of the "Personal Computer" in those days. People *wanted* PCs (of whatever brand) to get away from the restrictions and bureaucracy of the mainframes and minis. That is why they were called "Personal Computers".
Windows NT was also designed for the "single-user mentality/philosophy", and IBM had nothing to do with that, nor did such a "single-user" pose any limits to multi-tasking in any way.

Windows was an evolutionary product, not an innovative one. That was probably its main key to success. DOS provided an easy transition from 8 bit CP/M. The original Windows was built on top of DOS, and allowed existing DOS software to be used. Windows NT originally went nowhere because there were few applications for it until Windows 95 provided a "bridge" between Windows 3.1 and NT. Windows 2000 and XP are just a continuation of NT with incremental improvements.



--

************************
Michael Griffin
London, Ont. Canada
************************