Discussion on:

Gartner analyst thinks that Windows is 'collapsing'

I know that some of us aren't exactly the most ardent of Microsoft fans. But Windows dead? Just how will that affect you?

many fine points already presented here in favor of MS's continued existence, is the fact that Gartner frankly has their heads up the @sses. These are the same people that kept saying over and over that Linux had a higher TCO then windows, ohhh, and was completely not ready for commercial use. I am not saying they are completely wrong, just prone to believing fud, starting more fud, and giving fud to their friends =\
Also, let us not forget MS's vast pool of capitol. It would take several years of NO income to break them, or at least start pay cuts and layoffs.

How about a ban from all public procurement in the EU market?
Look at
http://www.groklaw.net/article.php?story=20080410050845876

European consumers and businesses are no different than those elsewhere with regards to their likelihood of arbitrarily abandoning their present investments.

I agree, but what if the public sector can't by anything from Microsoft for a period of time. According to EU statutes a ban can be imposed for up to five years. I don't think that is very likely though.

However, my gut feeling is that, as long as they can maintain their position in the private sector, their pain would be small.

Maybe, maybe not. There is a substabtial move going on toward open source here in Europe, with Open Source gaining terrain in schools and other public service sectors. Schools have been an important sector for MS. Once you have becomed customed to Windows, it's not so easy to even want to change to something else. I think MS should be worried. I even think they are worried. Why else almost bribe people to vote for OOXML for instance.

See sub-thread beginning at http://techrepublic.com.com/5208-6230-0.html?forumID=102&threadID=260406&messageID=2473699

the horrible incompatibilities of dift versions of windoze. It is almost like all the idft flavors of Linux (or Bennetton?)

The only way windoze could have previously 'stopped the madness' was to have a standard 'execution model' that DIDN'T CHANGE over versions. i.e. you could add new functionality or create new execution models but preserve the ability for older software to run. Dumb 'newbie' consultants and programmers made it so upgrades to .dlls previously were 'supposed to add functionality for free' when instead they BROKE older software versions, perpetuating '.dll hell', separate from not even having an 'execution model' compatibility mode.

Their current OSes promote 'software as use-once-dispose of' where you HAVE TO pretty much upgrade to newest versions or suffer both interchange problems and problems running the software itself.

I have stacks of old software that won't run now becuase of this, and biz have code that won't run also. NO WONDER some biz are still running COBOL. At least if Billy Boy is having a bad day, their software will continue to run!

saying that reports of Window's death are premature.

While Vista may not be gaining the market share that MS projected, users are not going to abandon their present investments in Windows based machines.

Even were users desirious of such, where would they turn to for a quick, easy and inexpensive transition?

... depending on the particular market. Keep in mind that
you said "Quick, easy and inexpensive."

First off the corporate market. They have hundreds and
even thousands of x86-based machines of many vintages
and capabilities. Oddly enough, Linux would be a very
good fit here for most of their needs. It's a very slim OS
that doesn't demand massive resources and supports a
large number of Open Source solutions that should be
capable of replacing almost all existing Windows
applications they might use except for the most esoteric.
Even the custom-written applications should be capable
of being ported to Linux relatively inexpensively when
compared to rewriting for a new version of Windows.

For the consumer market it honestly depends on what the
customer wants. For most, it may be as simple as
switching to Apple's Macintosh. They have models priced
for most consumers that give more capability than the
low-end hardware they usually buy. True, a Mac Mini can
cost $500 or so, but it has more functions and
capabilities than a $300 Dell. Add to this the noted
reliability of Apple products (Yes, I know. Every brand has
its lemons, but Apples do tend to break much less often
than their competitors) and you have a compact machine
that's likely to run 2, 3, maybe even 4 times as long as
that $300 Dell.
Of course, there are other models that fit into every level
of budget by Apple, all the way up to corporate servers;
but you asked for "quick, easy and inexpensive."

For the consumer-level hobbyist it's back to Linux. With
Linux a hobbyist can play to his heart's content; tweaking
and rewriting and making it do almost anything
conceivable, including building their own robots. Linux is
very inexpensive and uses existing hardware so the user
doesn't even have to buy a new machine. Most software is
free and if you can't find what you need, you can learn
how to code your own and join the expanding world of
Open Source software.

In other words, while I don't believe Microsoft is going to
fade away, there are many options available were it to
happen. I personally believe that what's going to happen
is that Microsoft will discover they aren't the only game in
town.

May be they should, I can pretty much guarantee they aren't though. Most businesses that use the in house/bespoke application route have a shed load of it.

The big boys who only do software haven't managed Vista compatibility properly just yet.

'non-standard' software apps are the biggest block to any change. Nothing to do with windows to nix or vice versa, or VMS or ....
Just replacing it with something else can run into major disruptions and man years of effort.

By way of analogy, consider the barriers to a rail company making the transition from narrow gauge to standard gauge; such is neither easy, inexpensive nor quick.

With respect to changing platforms, whether hardware or software, their needs to be a strong enough incentive in order for one to under such a disruptive and costly change.

Businesses most certainly will not simply change merely because the opportunity for such exists; there needs to first be the necessity to change, and, at a realistic cost.

Consumers are the more likely to change, as, for them, the computer is treated almost like any other appliance. However, like businesses, they are not going to invest more than they can afford. And, as with businesses, the time required to migrate their applications and data is included in their measure of the cost entailed.

Therefore, while there are indeed many options for those who are just beginning, there are very few realistic ones for those already committed to a particular platform.

Well, I migrated to Linux last christmas. The only file format that caused me a bigger headache was .mid
Apart from that tools to open, read, play any other files were readily available.
Tell me, what is not quick, easy and inexpensive?

twenty years of in house fortran applications running on sixty dec boxes under various flavours of VMS, with custom links to 22 different models of PLC, approximately a 100 intelligent devices, sixty of so TCP/IP socket links, running nine heavy industrial plants in one large site.

Extreme but real, strangely enough my main task at this place was to start offloading some tasks on to Linux. I was there two years and I aren't cheap, because it wasn't easy.

It's not moving from VMS, or to linux, or from or to anything else, it's just moving.

I do think MS needs to become more agile in its code development; and has missed a trick to streamline its OS code base. Vista seems to be resource hungry and bloated OS.

However, Windows is far from collapsing; MS is too market savvy to allow that to happen. Also it will take a long long time for people to move away from such a prominent vendor whatever their flaws.

Windows ME was also a failed operating system but Microsoft still made money from it, eventually making the hugely sucessful although update hungry XP.
As others have already stated what else is there to run all the code base (that of course is why Vista isn't being used).
The only big mistake Micrsoft could make is stopping selling XP. If this happened they WOULD be in trouble.

Of course Windows is bloated,and their code does need revamping,but MS are too tech-savvy to "fail",as they have been in the business far too long for that.Of course I use linux on my home-desktop,but,and I know this will upset some of my Open Source friends,I do not want to see Microsoft fail.Apart from the obvious (workforce,jobs,etc),if each "offering" is worse than the last,that can only "drive" more folks to the "distro community".In the end MS will have to revamp,and "open-up" accordingly.

Haven't you heard? They're doing that in June. No more box
copies of XP and only preinstalled on 'budget' laptops. They
may continue to support it, but they are clearly planning to
force new machines to be Vista-only, despite its ongoing
problems.

What you say is true to a point. Microsoft will stop selling XP over the counter, but they will sell Vista with an XP downgrade option as part of the CAL. This plan is to run until April '09.

Microsoft is nothing if not market savy. When original sales of Vista were flat and the market was complaining, they extended the life cycle of XP. When XP adoption was slower than expected they extended the life cycle of 2000. When many companies were holding onto their Windows NT Server boxes, waiting for Server 2003, Microsoft extended the life cycle of NT.

have collapsed in 2000, possibly earlier. The selling point of backwards compatibility was always a nightmare from a technical point of view. When you add in monalithic design, it just got worse.

Collapsing no, steadily less marketable and lucrative definitely.

Lets see what happens when Bill and his boys bite the bullet and release a new OS and run windows in a vm under it, which to me is their only practical alternative to going out out of business, eventually....

Absent the ability to continue to use existing applications and data files, any new OS has a very hard row to plow.

Imagine if it were the case that one needed separate appliances in order to receive AM radio, FM mono, FM stereo, etc..

MS built their empire on backwards compatibility.

If everybody had to 'start again' each new version, they would have had a much harder time in the market, no matter how significant the improvements between one version and the next.

As it is they had to constrain what they could offer in the next version to avoid it being effectively a new product.

Now unless they go for a significant loss in backwards compatibility (bigger than Vista), they stay where they are spending more and more to achieve less and less.

This is where their monolithic design, ridiculous levels of integration and moronic blurring of function, really bites them in the ass.

There's no reason why backwards compatibility could not have been retained under a modular architecture. That other OSes have been so able to thrive while being improved and expanded to meet new needs stands as proof that monolithicity is unnecessary for long term viability in the marketplace.

That MS both choose this path, and has clung to it for so long, stands as evidence that it has failed to learn from the historical examples of other industries and companies that attempted the same, only to eventually find it to be their Achilles Heel.

Thinking for the short term never suffices for the long run.

considering windows started out out as a gui for dos. I can see why they went down the monolithic route, I can't figure out why they didn't know it was a dead end though.

There again maybe each guy who made the decision for one more version knew he was up for winning the next round of musical management chairs and figured he could lumber his successor with the pain.

I have one question. The article says, It is ironic that one of
Window???s most touted ???features??? - its backwards
compatibility, is fast proving to be its Achilles heel.
Has Windows ever really been backwards compatible? I played
games in Win95; they broke and became completely
unplayable in Win98. I bought new games for Win98. They
broke and became completely unplayable in WinXP, even when
using their so-called Compatibility mode. In other words, for a
minimum of 12 years, I have never been able to play any
games written for an older version of Windows despite their
"touted backwards compatibility."

What did Apple do right that Microsoft has been unable to do?
Until OS X Leopard, I was able to play games and run any
software I possessed for all versions of the Mac OS all the way
back to OS 7, the version I started with. And the thing is, I can
prove that OS X 10.4.11 can run Photoshop 2.5 for Mac: circa
1993.

I couldn't agree more - MS is losing ground at every turn that doesn't involve cars. What's interesting is that it still may not be the end, if they play it right. Microsoft has been working on a new, from the kernel-up, operating system dubbed Singularity for some time. Putting the recent advances in making virtualization hypervisors invisible to the end-user together with Singularity, and you can have a new operating system with backwards compatibility that doesn't break new ground in cloud-computing (see Parallels Desktop for Mac and its Coherence feature for the first real steps in this direction). Add in cloud-computing operating systems like g.ho.st, or a Microsoft "Live" (choke) version of it, and they may yet pull themselves out of the fire.

All file types must be defined by Government laws---someplace.Microsoft never talks about file attributes at all.What is a bin file and so on.I think that the writing of an OS is something that you would have to teach yourself.I do not think that this would be impossible.It's not code,I have called code geeks to task and have received no response.The only other method is modules like SynthMaker.The first imperative would be to remove all virus from the electronics of the computer.The next would be to work for the revelation.

Ummm... Why?

Files types are not defined or controlled by any one
Government. They are created by corporations and
individuals and from there offered to other corps and
individuals to use.

Yes, there is a Standards body; the ISO (International
Standards Organization.) Here is where a new file type
may be submitted to become a Standard that will work the
same on any computer or software that chooses to adhere
to that Standard. The ISO is not controlled by any one
government and as has already been alleged by many
voting countries the system of confirming those standards
has itself become corrupted.
As for the latter part of your comment, it is clear you have
no concept of the difficulties involved in the creation
process. You can't just say, "Virus, begone!" You have to
work your tail off first to make code work, then try to
figure out how someone could use it in a manner it was
not designed for. It is far too easy to misuse a tool of any
kind. No matter how foolproof you make it, someone will
make a fool of you with it.

...it's basically been bought and run over by Micro$oft with the process to get OOXML approved as standard.

Everyone here has to take a long look back.... Redmond isn't as blind or stupid as we all think. They have taken the old dos labs area and are running and developing linux, yes linux....the word is that thay are using it to make their products more compatable to linux....(dont't get me wrong... all of my machines at home are dual boot boxes, and so are the two servers I run...) but really...take a good look at the default desktop in Vista "Ultimate"... looks and kinda acts a little KDE to me....I do believe that we will all be in for a large surprise comming out of Redmond in the NEAR future... and don't be surprised if it isn't a Redmond flavored linux.

Windows is monolithic, if you don't like KDE, you can use anyone of several front ends, even the command line.

Redmondux is and always has been a possibility, unfortunately, the entire software base currently used on ms platforms would be relegated to WINE. Non MS distros get hammered for that.

It's not that they can't move the OS, or even that they can't move office et al. What they can't move is us.

I consider this poetic justice, as it was MS who nailed our feet to floor, so they could force us to buy their stuff in the first place.

In the end it comes down to if a product gives you value for your money. In the past with minimal competition Micorsoft did not need to take this into consideration, even now the competion is still only small but it is growing at a significant rate. In the end it may not be a quick process, but the often "lack of value fo the money" of many Microsoft offerings is starting to catch up with them. The last boxed OS I purchased was Mandrake 9.2 Power Pack from 2003. I could have gone and purchased winXP (which is the same vintage as Mandrake 9.), it would have cost much more and is an inferior product.

Transitioning to new platforms brings with it far greater costs, such that undertaking such a step is not one to be taken lightly.

See sub-thread beginning at http://techrepublic.com.com/5208-6230-0.html?forumID=102&threadID=260406&messageID=2473699

As I only had to concerm myself with required changes. Software package came with all I imediately required eg Open Office Gimp choice of better web browses etc. The cost of transition always is "put out there" when disscussions of a change to Linux comes up. It probably would be on a case by case cost, but I think often the figure is just pulled out of thin air, coloued by opinion rather than fact. Does anyone have real experience and costings for transition to another OS that are real. Ernie Ball made the change some time back and has gone on record, that even though this was not their reason for change it has also saved them money.

Very few IT folk want to move to Vista. Some CFOs and CIOs are forcing the move, but for the most part it's XP or a total migration to thin clients at this point.

Citrix has REALLY make the whole idea of thin clients smarter, and with MSDTC, ClickOnce, and Silverlight, very few apps need to be installed on the thick client anymore.

MS wrote their own eulogy with Vista.

As noted above, in sub-thread that starts at http://techrepublic.com.com/5208-6230-0.html?forumID=102&threadID=260406&messageID=2473699 , transitioning to new platforms is neither quick, easy, nor cheap for most.

NT

Don't get me wrong, here; I am not a fan of Vista. Nor am I suggesting that anyone voluntarily use it.

Rather, I am pointing out the total costs of migrating to a completely new platfrom far exceed that of simply procuring a new OS.

You need to update your hardware and peripherals. To top off the problems, you also need to make sure your third party apps work as well, and you'll probably need a costly upgrade.

There are too many issues with Vista beyond just migrating to the OS. It almost is like moving to another platform.

That having been said, I do allow that this particular "upgrade" may be more painful than earlier ones for some; however, bear in mind that there are plenty of machines still running Win 9x and W2K owing to their being insufficient for the task of running XP.

And, in this regard, the same problem has arisen with any other number of OSes and applications. Witnesses the grief caused by the endless streams of upgrades to the AOL client, which caused many to ultimately get new machines, simply so as to avoid having to give up their invested efforts in AOL's services.

One finds the same pattern being repeated beginning even in the earliest days of mainframes. And, from experience, I can say that such migrations then, even without adjusting for inflation, were a whole lot more costly than that of migrating to Vista.

The bottom line is that there are costs involved here no matter which path one chooses. The only way to break the cycle is for customers to refuse to jump on MS's latest bandwagon. Were that to happen in sufficient numbers, perhaps MS would finally abandon its abject obeisance to monolithicity and move to a flexible modular OS, such that both a smaller footprint and backwards compatibility could be realized.

Damn you for being so eloquent!!!