Windows XP, Anyone?

C
Hi Alex.

Alex Pavloff wrote:
> Have you considered the fact that while Windows is a suboptimal control
> platform, its a decent desktop OS well suited for programming packages
> and the like? My new project is using Linux. However, I have no plans
> at this point to spend the large effort needed to port the programming
> enviroment to Linux because, well:

Yes Alex, this is a most thoughtfully considered position. And Linux is an even better environment for programming packages and quite usable for control. especially with extensions. And because it can be improved by people who want it to be an even better control platform and tailor made for the automation market, inside a year, there would be no comparison. In contrast, it is unlikely Windows will ever go in directions that make it more suitable for our particular needs. Indeed, as far as I can tell, all the changes seem to be in the wrong direction.

> 1) I have never run into any of my customers that use Linux on their
> desktops
And you never will if the monopoly gets their way.

> 2) Selling software to the Linux market is a problem because, well, they
> aren't used to paying for it.

You wouldn't be selling to the Linux market, most of them are singularly unconcerned with automation and controls. You would be selling to the automation market where downtime is expensive and MS licensing and policies are a PITA. Decreased cost couldn't hurt either.

> Curt, here's your problem. You sit there, tell everyone that Microsoft
> and all the stuff they use right now sucks, and tell everyone how you
> can solve all your automation problems with a 386 running Linux, a
> tie-wrapped ISA board, and plenty of elbow grease. Good for you and
> your customers, but... err... if you can do that err.... you're aren't
> going to be buying any software or hardware.

I'm obviously not your target candidate. I make my living solving the problems not addressed with OTS stuff. And I haven't said MS sucks nearly as often as the folks here that use it. I have said that they are a monopoly and have successfully wiped out just about any choice in this market segment. I respectfully submit that I am about restoring choices in a market that would benefit greatly from having some. Given a supported, realistic, alternative, competition would soon solve many, if not most, of the problems we see here. Without that alternative, folks will always have to simply accept whatever misery is visited upon them. Such is the case with monopolies. If you don't like the reliability of your local power company, you can generate your own, but it's not much of a _real_ choice.

> The current user base of Linux is highly technically inclined, not
> afraid of programming, and willing to build rather than buy. Automation
> vendors like myself aren't going to jump until we have solid indications
> that people will actually BUY our software if sold for Linux.

This is the "chicken and the egg" problem. I'm pretty sue you won't see this indication as long as there are no comparable Linux products for them to buy.

>>Notice
>>what happened to Wince once there was competition from the
>>embedded Linux folks :^)
>
> Bull. Windows CE (or whatever they're calling it) was giving priority
> because they wanted to take the lucrative handheld market away from
> Palm. Linux has only shipped on, what, one PDA (Sharp Zaurus). People
> flashing their IPAQs with a mostly-working linux distribution
> competition does not make.

And they are giving it away out of kindness? And showing source? That doesn't sound like it was induced by pressure from their proprietary comtemporaries.

I'm not sure why some folks look at my intent as evil or driven by malice. If you can think of a downside for having choices, I'm all ears. I simply want _all_ of us to be able to do automation and controls work with a more reliable, more flexible and adaptable OS that solves a lot of automation world problems. It's better engineering. People will buy better, more reliable, solutions if they become available.

Regards

cww
 
L
On November 8, 2002, Curt Wuollet wrote:
> If everyone reading this simply _asked_ their vendor when they
> will be able to buy Linux based tools and took serious interest
> in the answer, we would have have alternatives eventually.

I might perhaps capriciously suggest that a good start here would be to prohibit Open Source projects from producing Win32 binaries. I would suggest that somewhere in the high 90s in terms of percentages of downloads from any given open source project are Windows based freeloaders, who never even download the open source, let alone look at it, and this is really not helpful in persuading people to shift to Linux on their desktop...

Cheers

Tim
 
Alex Pavloff (in response to Curt):
> Have you considered the fact that while Windows is a suboptimal
> control platform, its a decent desktop OS well suited for programming
> packages and the like? My new project is using Linux. However, I
> have no plans at this point to spend the large effort needed to port
> the programming enviroment to Linux because, well:

> 1) I have never run into any of my customers that use Linux on their
> desktops

That's almost a tautology, though: you're selling a Windows program, so only people who use Windows will become your customers, therefore your customers will be using Windows.

> 2) Selling software to the Linux market is a problem because, well,
> they aren't used to paying for it.

That's not really a fair comment, but there is a true observation behind it; people use Linux for any number of reasons, many of which would be compromised by putting a closed-source program on it.

One of those possible reasons is, as you observe, the price. Others are things like avoiding dependence on a single vendor, access to the source code on reasonable terms, including the right to get someone else to change it, peer review and the resulting reliability, unlimited license `seats' which simplifies license management, and last, and in this case least, philosophical grounds.

You would have to provide truly compelling value to outweigh any of those reasons.

> Curt, here's your problem. You sit there, tell everyone that
> Microsoft and all the stuff they use right now sucks, and tell
> everyone how you can solve all your automation problems with a 386
> running Linux, a tie-wrapped ISA board, and plenty of elbow grease.
> Good for you and your customers, but... err... if you can do that
> err.... you're aren't going to be buying any software or hardware.
I suspect there's enough elbow grease to go around...

(a) Hardware will always be needed. Tie-wrapped ISA boards are OK for once-off things, specials and pilots, but for real applications, real hardware is needed.

(b) For software, it becomes a service. You don't sell software as such, you sell the software doing exactly what the customer wants. The great majority of software development already follows that model anyway, so it's not a big difference.

Jiri -- Jiri Baum <[email protected]> http://www.csse.monash.edu.au/~jirib MAT LinuxPLC project --- http://mat.sf.net --- Machine Automation Tools
 
C
Indeed, that's my point. And Alex will probably be glad to hear this. In this specialized market, What the Linux community does or doesn't do, is not going to make the major impact. Our GPL project will appeal to a certain class of folks, but again, these are not end users. The right people for the job are the Major Vendors. They can make Linux tools accessable and familiar and steal a march on the competition. It's obvious from the interests in Europe and Japan that we are going to get there eventually, the sooner we start, the better off we are in the race.

What I'm saying is that, if AB released a line of products on Linux tomorrow, that would have more impact and move the process faster than anything anyone else could possibly do. You would have the warm fuzzies you need to recommend a Linux solution, I would be able to work with Linux tools and do a great deal more with the installed base, and AB would have both a hi-rel solution set and would certainly have leverage with Microsoft to solve some of their problems. And a valid PR claim on something better and more Open.

Everyone would win, even Windows fans. Which brings up the question of why they would do this. Even if the obvious advantages of having a really bulletproof, adaptable, and customizable OS can't be envisioned, I would expect them to do it if enough of us ask for it. I would be willing to bet that NI's Linux version of LabView exists because someone really important asked for it. If not a major customer, then a large part of the scientific and engineering community.

Regards
cww
 
P
Hi All,
Did a bit of digging myself. It's a bit old but the Windows 95 Web Kiosk How-To at "www.kiasma.fi/~ooland/win95/":http://www.kiasma.fi/~ooland/win95/ does provide some suggestions
which may work.

One possibility is to replace the standard explorer.exe desktop with iexplore.exe or even possibly netscape.exe which forces the screen into a web browser rather than a conventional desktop. This appears to be possible in all versions of Windows. See "www.codeguru.com/mfc/comments/29090.shtml":http://www.codeguru.com/mfc/comments/29090.shtml .

Another alternative was to replace explore.exe with litestep.exe but litestep has been withdrawn pending release of a reworked version so this
may not be an option at present.

It is/was also possible to lock down the screen using Microsoft TweakUI 1.1 utility for Win95. This has been discussed on the list some time ago and from memory TweakUI is either not avialable or requires a different version for later versions of Windows.

SecureDesk! ("www.cursorarts.com/ca_sd.html":http://www.cursorarts.com/ca_sd.html ) seems to be another possiblity.

Anyone used any of these approaches and like to comment.

Regards
Peter Whalley
Magenta Communications Pty Ltd
Melbourne, VIC, Australia
e-mail: peter*no-spam*@magentacomm.com.au
delete *no-spam* before sending
 
Jiri;

If you're user is only supposed to 'run the machine' why on earth is he poking around inside the files with Explorer? FYI, the policy editor allows the admin to make 35 changes to Explorer, most of which will stop the user from causing damage to anything.

In the policy editor can also be used to lock down any part of the GUI, including the ability to resize or move the taskbar.

Have you actually looked at the policy editor in XP? If you had, you probably wouldn't be making so many comments that are obviously incorrect.

Mark Hill
 
Curt Wuollet:
> > If everyone reading this simply _asked_ their vendor when they will
> > be able to buy Linux based tools and took serious interest in the
> > answer, we would have have alternatives eventually.

Tim Linnell:
> I might perhaps capriciously suggest that a good start here would be
> to prohibit Open Source projects from producing Win32 binaries.

That wouldn't help... anyone, really.

> I would suggest that somewhere in the high 90s in terms of percentages
> of downloads from any given open source project are Windows based
> freeloaders,

1) There's no such thing as a freeloader.

Or, rather, the cost is so small as to be negligible. On the other hand, even so-called freeloaders will produce bug reports, informal help for other users, and so on. (What is the value to Rockwell of this mailing list? Yet none of the people on here have seen the RSLinx source...)

> who never even download the open source, let alone look at it,

2) Neither do most linux people. Only those who have a reason to do so download the source - to add a feature, find a bug, learn from it, etc.

Even those who do, generally only download and look at the source to a couple of programs. The rest just comes straight off the Debian CD or website, already compiled and configured to work together.

> and this is really not helpful in persuading people to shift to Linux
> on their desktop...

3) Sure it is - when the time comes to switch over, they're already familiar with a couple of programs. In some ways this is *excellent* help: in the middle of the great upheaval that the shift entails, there's at least one familiar thing that doesn't change.

In other words, share and enjoy!

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

OSS ia not exclusively Linux although Linux is (or at least should be) exclusively OSS. I really hadn't thought much about that as I certainly have no need to cross compile. I guess I wouldn't want to restrict anyone else's choice. Right now, I'm looking at issues seperately. What I am proposing would upset a great many Linux folks, but as a pragmatist, if the major automation vendors were to continue in their proprietary vein, but offer Linux binaries it would still be a great leap forward as at least the OS facilities would be Open. All the comm issues and OS issues could at least be addressed. I just thought I should be clear that I wouldn't expect them to release source amytime soon, we have to have realistic goals. Even if things stayed pretty much the same, with only the Linux platform added. much the way that NI did with LabView, we and our customers would gain great benefit. That would provide the springboard for perhaps working on the Tower of Babel problem and encourage some commonality and interoperability. As an integrator that would solve an unbelieveble number of problems.

Regards

cww
 
B
In a message dated 11/13/2002 3:29:03 PM Central Standard Time, cww writes:

> Indeed, that's my point. And Alex will probably be glad to hear this. In
> this specialized market, What the Linux community does or doesn't do, is
> not going to make the major impact. Our GPL project will appeal to a
> certain class of folks, but again, these are not end users.
> The right people for the job are the Major Vendors. They can make Linux
> tools accessable and familiar and steal a march on the competition. It's
> obvious from the interests in Europe and Japan that we are going to get
> there eventually, the sooner we start, the better off we are in the
> race.

Obvious? I am not all that sure. MS has shown itself to be pretty adaptable. Its not perfect, but it has an advantage that Linux does not, and probably won't anytime soon. The advantage is the MILLIONS of people who are "competant" to manage and run their machines. There are very few people by comparison competent in Linux. I think MS is more worried by things like the Lindows $199 PC than by anything done in the automation field.

If Lindows ever supports windows apps to the extant that virtually any windows app would run on a lindows machine, wouldn't that be an even better idea? Maybe you have been barking up the wrong tree trying to take away MS Office from people and should have been spending your time making Lindows run more windows apps. few end users will care what the OS is as long as it flawlessly runs their favorite apps.

> What I'm saying is that, if AB released a line of products on Linux
> tomorrow, that would have more impact and move the process faster
> than anything anyone else could possibly do. You would have the
> warm fuzzies you need to recommend a Linux solution, I would be able to
> work with Linux tools and do a great deal more with the
> installed base, and AB would have both a hi-rel solution set and
> would certainly have leverage with Microsoft to solve some of their
> problems. And a valid PR claim on something better and more Open.
> Everyone would win, even Windows fans.

This would benefit only the very few integrators conversant with Linux, to the detriment of those not familiar with it. I can see why you might be in favor of AB doing such a thing as it would enhance your tenuous market position in Linux. Why would AB do something that would benefit only a few dozen people? The tiny bit of sales from these people would never cover the expense of porting software to Linux nor maintaining it. Besides, Linux people rarely buy software, at least that is the rub on them. And its probably pretty close to true, so even if they did release Linux versions of their software, the sales might end up being zero because the Linux fans would say its not "good enough" because it is not open source. And doing open source is a financial rathole.

> Which brings up the question of why they would do this. Even if the
> obvious advantages of having a really bulletproof, adaptable, and
> customizable OS can't be envisioned, I would expect them to do it
> if enough of us ask for it. I would be willing to bet that NI's
> Linux version of LabView exists because someone really important
> asked for it. If not a major customer, then a large part of the
> scientific and engineering community.

I'd guess its more like someone paid them to do it, either directly or indirectly OR someone at NI thought it was a neat idea and just did it. I wonder what percent of their sales are the Linux version versus the Windows version. I'd bet the Linux version has only a tiny percent of their sales, maybe only to a few customers.

Bob Peterson
 
A

Anthony Kerstens

I welcome what Curt suggests. One of the things that has me tied to a Windows box is my industrial software.

Anthony Kerstens P.Eng.
 
Bob Peterson:
> If Lindows ever supports windows apps to the extant that virtually any
> windows app would run on a lindows machine, wouldn't that be an even
> better idea? Maybe you have been barking up the wrong tree trying to
> take away MS Office from people and should have been spending your
> time making Lindows run more windows apps. few end users will care
> what the OS is as long as it flawlessly runs their favorite apps.

There are people working on that, of course, but the problem is that in that field MS calls all the shots. They can make changes to Windows and Office, release or don't release any details of them, and so on. As DR DOS found out, they can even do things like (accidentally?) checking for such situations and popping up scary-looking warnings.

Microsoft would also dearly love to make it illegal outright, and they appear to be making some progress on that front.

I have great respect for the Samba team. They do an excellent job under circumstances most would consider unacceptable. Thankfully, theirs is just a small piece of Windows, so it's manageable.

Jiri -- Jiri Baum <[email protected]> http://www.csse.monash.edu.au/~jirib MAT LinuxPLC project --- http://mat.sf.net --- Machine Automation Tools
 
If someone doesn't make the products we'll never have the option of buying them and demonstrating the market. I can guarantee you that they sell more of the Linux version than if they hadn't written it.

I realize the point you're making is that companies can be expected to make decisions based upon expected ROI, and that is right. Even so, I'd be glad to make you a small wager that there are many people like me who, when the software of their choice is available to run on the latest platform from Bill G. and also to run on an open source OS, will choose the open version.

If the vendors stay away from the platform because they underestimate the market potential, we'll never get the chance to find out.
 
C
Hi Bob

List Manager wrote:
> ------------ Forwarded Message ------------
> From: PETERSONRA
>
> In a message dated 11/13/2002 3:29:03 PM Central Standard Time,
> cww writes:
>
>>Indeed, that's my point. And Alex will probably be glad to hear this.
>>In this specialized market, What the Linux community does or doesn't
>>do, is not going to make the major impact. Our GPL project will appeal
>>to a certain class of folks, but again, these are not end users.
>>The right people for the job are the Major Vendors. They can make
>>Linux tools accessable and familiar and steal a march on the
>>competition. It's obvious from the interests in Europe and Japan that
>>we are going to get there eventually, the sooner we start, the better
>>off we are in the race.
>
> Obvious? I am not all that sure. MS has shown itself to be pretty
> adaptable. Its not perfect, but it has an advantage that Linux does
> not, and probably won't anytime soon. The advantage is the MILLIONS of
> people who are "competant" to manage and run their machines.

Yes Bob, but we aren't talking millions of people here. We are talking about the people who craft automation solutions. Very few people reading
this would have any difficulty managing and maintaining a Linux machine. They deal with much larger problems with what they are using now. I have managed and maintained quite a few but, I'm at a loss regarding many of the Windows problems seen here. Linux issues tend to be more "fixable".

There are
> very few people by comparison competent in Linux.

It would depend on your definition of competent. Many more people know how to reboot and reload Windows, but anyone who can actually fix Windows problems should be way ahead on Linux where information is more available and there are no secrets.

I think MS is more
> worried by things like the Lindows $199 PC than by anything done in the
> automation field.

I'm more worried about Lindows than MS is. It's a very poor introduction to Linux in my view. It's much less robust and extremely limited in it's
present form and shares many of the same problems with Windows. It's a remarkable piece of reverse engineering and impressive as an exercise, but nowhere near automation grade. And it's dependent on MS. I wouldn't want to sell life insurance to competitors who depend on MS, based on
historical data.

> If Lindows ever supports windows apps to the extant that virtually any
> windows app would run on a lindows machine, wouldn't that be an even
> better idea?

If it even comes close, MS will just change the API's and send them back to square one. I see it as a last ditch solution where you must run
something unavailable on Linux and for a few special applications

Maybe you have been barking up the wrong tree trying to
> take away MS Office from people and should have been spending your time
> making Lindows run more windows apps. few end users will care what the
> OS is as long as it flawlessly runs their favorite apps.

I just finished explaining that I don't favor taking anything away from anyone. Let those to whom Office is critical, enjoy. I would not care to have someone running Office on a production machine, and for programming and documentation, OpenOffice will integrate just as well with the
tools. It's all about choices. There are a few snags, like AutoCAD, but I expect those to go away soon as well. The EDA market is moving to
Linux as an option and AutoDesk will have to compete or lose market share.

>
>>What I'm saying is that, if AB released a line of products on Linux
>>tomorrow, that would have more impact and move the process faster than
>>anything anyone else could possibly do. You would have the
>>warm fuzzies you need to recommend a Linux solution, I would be able
>>to work with Linux tools and do a great deal more with the
>>installed base, and AB would have both a hi-rel solution set and would
>>certainly have leverage with Microsoft to solve some of their
>>problems. And a valid PR claim on something better and more Open.
>>Everyone would win, even Windows fans.
>
> This would benefit only the very few integrators conversant with Linux,
> to the detriment of those not familiar with it.

How would having a choice be detrimental? I certainly wouldn't force anyone to use it.

I can see why you might
> be in favor of AB doing such a thing as it would enhance your tenuous
> market position in Linux.

I don't have a market position in Linux, it's free. I would greatly prefer working with Linux and the free tools make development much
cheaper. And the system cost would be less, which means more profit. And I really like the lack of phone calls at odd hours.

Why would AB do something that would benefit
> only a few dozen people?

Well there's the whole argument. People jump through a lot of hoops to deal with the present products I'm pretty sure lots of people will try
anything that that has the name on it. And anyone who does, would benefit from it. Low cost, no EULA or license tracking, much higher
functionality out of the box. Nothing else to buy except your automation tools, no activation hassles, no hardware snooping or sending your info
to Redmond. Almost everything that is a PITA with Windows in autmation is fixed with Linux. My guess is that it would be useful to a lot of
people and that number would increase rather rapidly.

The tiny bit of sales from these people would
> never cover the expense of porting software to Linux nor maintaining it.

Everyone who has ported to Linux so far has done OK, except those that compete with MS directly like Corel. Even some of those are doing well
like IBM, SAP and ORACLE. IBM mentioned there are over 4000 applications in precess. Somebody thinks there's a market.

> Besides, Linux people rarely buy software, at least that is the rub on
> them. And its probably pretty close to true, so even if they did
> release Linux versions of their software, the sales might end up being
> zero because the Linux fans would say its not "good enough" because it
> is not open source. And doing open source is a financial rathole.

As I mentioned, they would be selling to automation people, not Linux people. Automation people are at the opposite extreme, used to paying
astonishing prices for their applications. why would this be different? IBM has stated that their OSS effort has been worth about $2 billion and now that the initial investment has been repaid, a substantial chunk of their revenues will be from this "non-existant" market. HP and SUN are chasing their taillights.

>
>>Which brings up the question of why they would do this. Even if the
>>obvious advantages of having a really bulletproof, adaptable, and
>>customizable OS can't be envisioned, I would expect them to do it if
>>enough of us ask for it. I would be willing to bet that NI's
>>Linux version of LabView exists because someone really important asked
>>for it. If not a major customer, then a large part of the
>>scientific and engineering community.
>
> I'd guess its more like someone paid them to do it, either directly or
> indirectly OR someone at NI thought it was a neat idea and just did it.
> I wonder what percent of their sales are the Linux version versus the
> Windows version. I'd bet the Linux version has only a tiny percent of
> their sales, maybe only to a few customers.

It would be interesting to have those figures, wouldn't it. I noticed they upgrade it and it hasn't gone away. If they pushed it, it would
probably do a lot better. But their relationship with MS would suffer and they (like everyone else) can't afford that. Maybe they did it just
for the leverage with MS! Kinda nice to have a backup position if they get too greedy.

Regards

cww
 
M

Michael Griffin

On November 14, 2002 10:45 pm, Mark Hill wrote: <clip>
> If you're user is only supposed to 'run the machine' why on earth is he
> poking around inside the files with Explorer? FYI, the policy editor
> allows the admin to make 35 changes to Explorer, most of which will stop
> the user from causing damage to anything.
>
> In the policy editor can also be used to lock down any part of the GUI,
> including the ability to resize or move the taskbar.
<clip>

There's 35 things we have to know about Windows security to prevent people from damaging anything? Most people would be hard pressed to name 35 things of any kind they know about Windows. A lot of people want to use Windows because they figure they won't have to know anything about computers that way.

A great disservice has been done to the industry by people who have been disseminating the falacies that "Windows is easy", or that "there are lots of people who understand Windows". The fact is that Windows is very complex, and most people know just enough about Windows to be dangerous with it. Very few IT people really know much about Windows other than how to click on menus to set up a typical network, so they're often not much help either.

I think it's reasonable to say that properly configuring any operating system requires more knowledge than can be gained by clicking on icons at your desk all day. In other words, the fact that someone has been using Windows on a daily basis doesn't mean they really know anything significant about it. Any discussion about Windows that begins with the premise that "everyone knows Windows" should be treated with the suspicion it deserves.

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

Ranjan Acharya

<clip>
I have great respect for the Samba team. They do an excellent job under circumstances most would consider unacceptable. Thankfully, theirs is just a small piece of Windows, so it's manageable.

Jiri
</clip>

Yes and from what I have been reading Microsoft is trying to build a patent wall to keep Samba out.

RA
 
B
Curt Wuollet wrote:
> Hi Bob
>
> List Manager wrote:
> > ------------ Forwarded Message ------------
> > From: PETERSONRA
> >
> > In a message dated 11/13/2002 3:29:03 PM Central Standard Time,
> > cww writes:
> >
> >>Indeed, that's my point. And Alex will probably be glad to hear this.
> >> In this specialized market, What the Linux community does or doesn't
> >> do, is not going to make the major impact. Our GPL project will appeal
> >> to a certain class of folks, but again, these are not end users. The
> >> right people for the job are the Major Vendors. They can make Linux
> >> tools accessable and familiar and steal a march on the
> >>competition. It's obvious from the interests in Europe and Japan that
> >> we are going to get there eventually, the sooner we start, the better
> >> off we are in the race.
> >
> > Obvious? I am not all that sure. MS has shown itself to be pretty
> > adaptable. Its not perfect, but it has an advantage that Linux does
> > not, and probably won't anytime soon. The advantage is the MILLIONS
> > of people who are "competant" to manage and run their machines.
>
> Yes Bob, but we aren't talking millions of people here. We are talking
> about the people who craft automation solutions. Very few people reading
> this would have any difficulty managing and maintaining a Linux machine.
> They deal with much larger problems with what they are using now. I have
> managed and maintained quite a few but, I'm at a loss regarding many of
> the Windows problems seen here. Linux issues tend to be more "fixable".
> There are
> > very few people by comparison competent in Linux.
>
> It would depend on your definition of competent. Many more people know
> how to reboot and reload Windows, but anyone who can actually fix
> Windows problems should be way ahead on Linux where information is more
> available and there are no secrets.
>
> I think MS is more
> > worried by things like the Lindows $199 PC than by anything done in
> > the automation field.
>
> I'm more worried about Lindows than MS is. It's a very poor introduction
> to Linux in my view. It's much less robust and extremely limited in it's
> present form and shares many of the same problems with Windows. It's a
> remarkable piece of reverse engineering and impressive as an exercise,
> but nowhere near automation grade. And it's dependent on MS. I wouldn't
> want to sell life insurance to competitors who depend on MS, based on
> historical data.

As best I can tell it is intended as a way to run your Windows software without having to buy Windows. However, as I understand it, a full version of Linux is included that could be used by the end user if so desired. The Lindows part of it just allows some (incomplete) Windows functionality.

> > If Lindows ever supports windows apps to the extant that virtually any
> > windows app would run on a lindows machine, wouldn't that be an even
> > better idea?
>
> If it even comes close, MS will just change the API's and send them back
> to square one. I see it as a last ditch solution where you must run
> something unavailable on Linux and for a few special applications

A few "special" apps? How about the tens or even hundreds of thousands of apps totally unavailable on Linux but readily available to Windows?

> Maybe you have been barking up the wrong tree trying to
> > take away MS Office from people and should have been spending your
> > time making Lindows run more windows apps. few end users will care
> > what the OS is as long as it flawlessly runs their favorite apps.
>
> I just finished explaining that I don't favor taking anything away from
> anyone. Let those to whom Office is critical, enjoy. I would not care to
> have someone running Office on a production machine, and for programming
> and documentation, OpenOffice will integrate just as well with the
> tools. It's all about choices. There are a few snags, like AutoCAD, but
> I expect those to go away soon as well. The EDA market is moving to
> Linux as an option and AutoDesk will have to compete or lose market
> share.

I tend to doubt this. AutoCad is not exactly an EDA product. Its more of a drafting product. Big difference. I have played with OpenOffice. It looks to have a pretty good emulation of Office. So why is it OK to emulate Office in this way but emulating Windows ala Lindows is not? Could it just be that you are really opposed to the profit motive involved in Lindows???

> >>What I'm saying is that, if AB released a line of products on Linux
> >> tomorrow, that would have more impact and move the process faster than
> >> anything anyone else could possibly do. You would have the
> >>warm fuzzies you need to recommend a Linux solution, I would be able
> >> to work with Linux tools and do a great deal more with the
> >>installed base, and AB would have both a hi-rel solution set and would
> >> certainly have leverage with Microsoft to solve some of their
> >>problems. And a valid PR claim on something better and more Open.
> >> Everyone would win, even Windows fans.
> >
> > This would benefit only the very few integrators conversant with
> > Linux, to the detriment of those not familiar with it.
>
> How would having a choice be detrimental? I certainly wouldn't force
> anyone to use it.

Because everyone who did not use it would be forced to pay for it, even though they have no interest in it, since it could not pay its own way.

> I can see why you might
> > be in favor of AB doing such a thing as it would enhance your tenuous
> > market position in Linux.
>
> I don't have a market position in Linux, it's free. I would greatly
> prefer working with Linux and the free tools make development much
> cheaper. And the system cost would be less, which means more profit. And
> I really like the lack of phone calls at odd hours.

I have not gotten a phone call at odd hours in at leats 5 years, maybe longer. It has absolutely nothing to do with Linux versus Windows. It has
to do with competence in setting up the automation in the first place.

> Why would AB do something that would benefit
> > only a few dozen people?
>
> Well there's the whole argument. People jump through a lot of hoops to
> deal with the present products I'm pretty sure lots of people will try
> anything that that has the name on it. And anyone who does, would
> benefit from it. Low cost, no EULA or license tracking, much higher
> functionality out of the box. Nothing else to buy except your automation
> tools, no activation hassles, no hardware snooping or sending your info
> to Redmond. Almost everything that is a PITA with Windows in autmation
> is fixed with Linux. My guess is that it would be useful to a lot of
> people and that number would increase rather rapidly.

If that were the case there would be a lot of interest in the big automation venders in working with Linux. There just isn't. A few software engineers expressing an interest in Linux that just happen to work for an automation vender is not the same thing as interest by the company itself. Until they can find a way to make money at it, it just won't happen.

> The tiny bit of sales from these people would
> > never cover the expense of porting software to Linux nor maintaining
> > it.
>
> Everyone who has ported to Linux so far has done OK, except those that
> compete with MS directly like Corel. Even some of those are doing well
> like IBM, SAP and ORACLE. IBM mentioned there are over 4000 applications
> in precess. Somebody thinks there's a market.

Everyone? The fact is that Corel was near death long before it ported anything to Linux. I am unaware of any widely used products ported to Linux that have been a commercial success in the mass market. They just don't exist because the Linux mindset is "free" software. This is a huge
disincentive to software developers.

> > Besides, Linux people rarely buy software, at least that is the rub
> > on
> > them. And its probably pretty close to true, so even if they did
> > release Linux versions of their software, the sales might end up being
> > zero because the Linux fans would say its not "good enough" because it
> > is not open source. And doing open source is a financial rathole.
>
> As I mentioned, they would be selling to automation people, not Linux
> people. Automation people are at the opposite extreme, used to paying
> astonishing prices for their applications. why would this be different?
> IBM has stated that their OSS effort has been worth about $2 billion and
> now that the initial investment has been repaid, a substantial chunk of
> their revenues will be from this "non-existant" market. HP and SUN are
> chasing their taillights.

I find it humurous that the formerly evil IBM is now being touted as the future of computing. From what I can tell, virtually all of IBM's work in Linux is in developing what is nothing more than a closed system running on Linux. I don't see that as all that much different than what MS does.

> >
> >>Which brings up the question of why they would do this. Even if the
> >> obvious advantages of having a really bulletproof, adaptable, and
> >> customizable OS can't be envisioned, I would expect them to do it if
> >> enough of us ask for it. I would be willing to bet that NI's
> >>Linux version of LabView exists because someone really important asked
> >> for it. If not a major customer, then a large part of the
> >>scientific and engineering community.
> >
> > I'd guess its more like someone paid them to do it, either directly or
> > indirectly OR someone at NI thought it was a neat idea and just did
> > it. I wonder what percent of their sales are the Linux version versus
> > the Windows version. I'd bet the Linux version has only a tiny
> > percent of their sales, maybe only to a few customers.
>
> It would be interesting to have those figures, wouldn't it. I noticed
> they upgrade it and it hasn't gone away. If they pushed it, it would
> probably do a lot better. But their relationship with MS would suffer
> and they (like everyone else) can't afford that. Maybe they did it just
> for the leverage with MS! Kinda nice to have a backup position if they
> get too greedy.

And that backup position would be what? Order their customers to use Linux or else?

Bob Peterson
 
R

Ralph Mackiewicz

> If someone doesn't make the products we'll never have the option of
> buying them and demonstrating the market. I can guarantee you that
> they sell more of the Linux version than if they hadn't written it.

...snip...snip...

> If the vendors stay away from the platform because they underestimate
> the market potential, we'll never get the chance to find out.

It's that "vision" thing. The companies that grow are the ones that spot an opportunity before there is a market for it (aka "vision").
Established companies tend to look at the market numbers and plan accordingly. But these numbers only measure "what is", not "what will be". Don't look to the dominant vendors to find these new
opportunities. They aren't really looking there. They are focused on their existing markets, not the markets that don't yet exist. So, if you think running automation systems on Linux is of benefit to you, don't wait for the big guys to do something (they'll only do it after it has already happened) and don't wait until you can get it for free (you'll have a long wait). Buy something from those companies on the periphery that have a vision of what can be now. That will really make a difference. And, you'll be able to take advantage of those benefits now (instead of having to wait for them).

Regards,
Ralph Mackiewicz
SISCO, Inc.
 
V

Vladimir E. Zyubin

Hello Ralph,

And there is the brain-washing campaign MS doing. The team consists of professional psychologists, and admen, and marketing analysts, and the like. The "second-level" companies are under the hard pressing too (as well as end-users).

And if we precisely look at the antimonopoly suit against MS we can understand that people do not believe in the antimonopoly law and healthy competition.

Independency of the vendors' analysis you try to refer to is incorrect.

--
Best regards.
Vladimir E. Zyubin mailto:[email protected]
 
M

Michael Griffin

On November 20, 2002 04:49 pm, PETERSONRA wrote: <clip>
> I tend to doubt this. AutoCad is not exactly an EDA product. Its more of
> a drafting product. Big difference.
<clip>

AutoCad is starting to look like a dinosaur. The CAD market is driven by the mechanical drafting field, and I don't personally know of *anyone* who is still using AutoCad for mechanical drafting. Everyone seems to be switching to 3-D modelling with SolidWorks or some other competitor. It's an entirely different way of approaching the problem and I don't know of anyone who is impressed with AutoCad's offering in this field. AutoCad was so busy defending their market share in their old product that they seemed to have missed the boat when the market changed.

I think the control system drafting market is ripe for a big change also. I would think it would be logical to allow electrical drafting systems to share a symbol database with PLC programming software. We don't need AutoCad on Linux. We need common data standards to allow us to get out of the electric pencil era and move on to something that integrates the electrical hardware design with the software design.

> And that backup position would be what? Order their customers to use
> Linux or else?
<clip>

Software companies need to make what their customers want to buy. Of course figuring out what somebody wants to buy isn't always that easy, since pretty often the customer doesn't realise they want it until you offer it to them.

However, looking back on the history of the software business, if I were selling software I would be rather nervous about staking my future (or rather, my money) on the continued dominance of a specific product from another company (or even the existence of that company). People change; companies change. A lot of dominant market leaders have disappeared into oblivion. I don't think that any responsible businessman can afford to forget that.

I think the prudent thing is to be careful not to tie yourself too closely to someone else's proprietary software unless there are several close equivalents from their competitors you can choose from. Support open standards rather than proprietary ones whenever possible. Minimise your exposure to proprietary features you can't avoid. There are a lot of basic software design principles you can use such as isolating operating system dependencies to specific parts of your program, rather than scattering them everywhere.

You many not eliminate all your risks, but you can at least reduce them to managable proportions. This isn't social philosophy, it's just good business.

--

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