One has to ask what is the commercial benefit to Microsoft of providing IE? We don't buy it as a separate product from the OS, it is a maintenance nightmare for MS consuming significant tracts of development $ without benefit, and is a major factor in establishing Microsoft's reputation as a provider of insecure and unreliable product. Microsoft needs HTML and XML parsers as part of it's technology suite to facilitate net-centric application integration, but plenty of other stand-alone browers and mail clients (mostly better than IE) are available for free, so why bother?

Microsoft has failed in it's attempts to dominate the Internet agenda, most of us develop websites that work within standards supported by all reputable products, so IE as a concept, with it's non-standard extensions and capabilities, is an anachronism. If I were Steve Ballmer I would ditch it ASAP.

Of course the other face of the argument is that the single most important part of the MS product portfolio is the operating system, and if IE remains a key component of MS strategy, if MS still believes it can dominate Internet technology, then it makes eminent sense to ensure that access to MS Internet technology is less convenient from other OS's to retain customer loyalty in the light of the emergence of Linux as a desktop force. Recent developments in China show us that Asia is now adopting Linux as it's standard OS, and unless MS can reverse this by for instance ensuring that non-MS OS's are devalued by not being able to access MS Internet technology, MS's aspirations to develop the Asian market will be seriously curtailed.

However one looks at it, the decline of IE would be a good thing, improving PC security and enabling the industry to consolidate on agreed standards rather than having to support manufacturer-specific extensions - but I wont be holding my breath waiting...

Can Microsoft bring out an improved version of IE? I'm not so sure. They're faced with two parralel problems.

First, their browser code is an antique. It's a direct lineal descendant of Spyglass Mosaic, the first graphical browser. Microsoft has owned it since '95. Old, frequently patched code tends to be brittle, tangled, hard to understand.

Second, they've stirred IE so throroughly into the Windows operating system that the dependencies must be hugely complex. Every change must trigger an unpredictable cascade of other small changes. Thus the security nightmare of "service patches" - plug one hole and another opens, fix a bug and you break something else.

Putting those together, I think the problem MS has with IE is that it's a spawling un-maintainable beast of a program, too hairy to fix, too brittle to imrove.

This pruning of all old versions is probably from necessity. The windows OSes have diverged so much that maintaining one such program for all must be hard. By cutting the old OSes loose they get to junk a lot of legacy code. Maybe then they can simplify enough that it can be cleaned up. Then they can try to pick up momentum again, chasing Firefox on features, rather than mere market share.

Microsoft rightly sees browser-based computing as leading to the end of its business model (see Joel Spolsky as to why). So the company will do what it can to push people towards its Windows Forms for Internet based computing, while simultaneously doing what it can to degrade the browser experience.

Nothing new here.

craggy_steve: Microsoft having given up on owning all the standards, the primary purpose I can see for IE is: to make sure the web doesn't commoditize the underlying OS. Thus: be the weak link preventing a general quality improvement (stopping the web from challenging native apps), and add features that can only work on top of Windows.

Sure - Firefox is a viable, and welcome, alternative, but I and several friends and clients have found it to be a little 'overhyped'. Some nice features, but still prone to the accsional lockup and crash, - about once a day it seems. OK if you are a bit of a geek, but not for the run of computer user.

If IE gets dumped by the majority of users, the virus writers will turn to whatever is the latest browser or OS and proceed to make that unusable, it's a fact of life that the crazies will always be with us.

It's always a shame to lose something familiar, but that's progress I suppose.

A lockup a day? Not on my machine. Admittedly I am using Linux. Still, I get maybe one Firefox lockup a week, and by that I mean, Firefox has been running continuously, in active use or open but idle, for that whole time. Not perfect but, for the hammering I give it (routinely opening tens of tabs in many open windows on several virtual desktops) it's not at all unreasonable.

I use Mozilla (same code as Firefox) all day at work, and it never crashes. Perhpas that's because it's running on a well-written OS (linux) instead of Windows.

Ernest writes: If IE gets dumped by the majority of users, the virus writers will turn to whatever is the latest browser or OS and proceed to make that unusable, it's a fact of life that the crazies will always be with us.

This argument presumes that all browsers are equally insecurely written, which is probably not the case. Microsoft software does seem to have more than its fair share of break-in problems. For example, most web servers run Apache, but the Microsoft IIS web server gets most of the break-ins, worms and other attacks.

I'd concur with the guess that MS is dumping legacy support for IE on the older platforms to help boost potential new OS sales. This is a repeat of a familiar pattern with MS - their legacy support rapidly declines after a product/component has been superceded, and virtually ceases once the second subsequent generation of something hits the streets.

As for an abandonment of IE, I don't see that as the case - as it has become so intertwined with the Windows platforms as a result of their conscious move several years ago to move in the extensible desktop direction.

IE will be an unfortunate fact of cyberlife for some time to come. Evidence of this is the existence of numerous sites and still widely used techniquesthat will only render properly in IE.

MS isn't ditching IE...it's ditching the customers that haven't contributed lately to company coffers. Pay tribute to Bill, or be cast out!

The reason Microsoft should worry about losing the browser market is simple; browsers represent platform independence. Firefox is the same whether I'm running Wintel, *nix, or Mac. All of a sudden MS isn't as indispensable as it used to be because a major app that I spend most of my time in is OS agnostic. Further, how many tertiary apps can I then tie in to that application? Video, audio, word processing, spreadsheets, yadda, yadda. The browser is extendable in what it can do, and it doesn't matter what OS I choose to run it in. And that's why MS should very much fear this development.

Goodbye and good riddance. The only useful purpose IE serves anymore is for downloading windows updates.

I've userd firefox for quite a while now and have never had a problem with it. Before that, I used Netscape, until they stopped updating it.

Oddly enough, Firefox for Linux goes haywire and has to be shut down at least once every 24-48 hours, whereas Firefox on my Windows 2000 machine often runs for weeks before needing a restart. The Linux box is superior in every other aspect as far as stability, but Firefox for Linux has been going downhill since 0.5.

Microsoft software does seem to have more than its fair share of break-in problems. For example, most web servers run Apache, but the Microsoft IIS web server gets most of the break-ins, worms and other attacks.

Not being 'server' orientated, I have often wondered whether IIS was more loosely written than say, Apache, or was it just attacked more often because of it being Microsoft. After all, they are a tempting target, and all those 'anti-trust' cases have not helped their reputation.

The perfect bug-free, error-free, hack-proof code has yet to be written, under whatever platform you may choose.

That the internecine battles between Oracle, Netscape, Sun et al and Microsoft will eventually have a big effect on the rest of us and the way we use IT, may not be such a good thing in the long run.

Will we go back to the days where you had to be tech savvy to use a computer, because every company had it's own way of doing things, or will 'open source' be the answer.

Somehow I don't think so, already there has been much litigation over source code in the Linux field, with companies claiming to own the code, and wanting to be paid for the use of it.

Whatever it's faults, (many), MS has been a steadying and homogenising hand on the helm of IT, and those who are relishing its demise should be careful of what they ask for - they may just get it...

Always felt that the MS Wars had the stink of politics behind it, and judging by the vehemence of the viewpoints whenever the subject is mentioned, it certainly arouses similar passions.

Already there has been much litigation over source code in the Linux field, with companies claiming to own the code, and wanting to be paid for the use of it.

"Much" is a bit of a stretch. There's been one company, which has lost every case that they've put up.

Certainly OS wars have a bit of politics to them, especially since they're about personal preferences. Yet, it is true that Microsoft's browser development has stagnated (pop-up blocking, tabbed browsing, etc.), and probably for the business reasons described by Joel of Joel on Software. Purely open web-based computing is a threat to Microsoft, because it reduces the network effects of people using a common OS.

Viewing this as an economic phenomena rather than an IT matter, Microsoft - a mature monopoly - is ripe for some smaller, faster competitor to attack.

Most monopolies in the past were done in by competitors taking advantage of the monopoly's disinclination and inability to change, not by government attacks. The Standard Trust, for example, was doomed by the discovery of oil in Texas in 1903, not by Teddy Roosevelt's "Trust-Busting" a decade later.

Firefox may or may not be the beginning of the end for Microsoft's market dominance, but that dominance is bound to end eventually. If Gate's intention was a company that would last, the monopoly business model was the wrong choice.

Maybe Microsot just decided that to invest in a product that is free, in a market of free competitors, when they already made money by selling the underlying operating system, is a foolish waste of resources. It's not the browser that they are trying to sell, it's the OS. And people are going to be using that OS for a considerable time even if there are free alternatives, simply because everyone has at least one non-free application which runs on Windows for which there is no peer on Linux, or BSD, or Unix. For some that may be Visual Studio, others it may be AutoCAD, or MasterCAM, or a particular desktop publishing and/or graphics package, or more likely a game. Until the application companies can be persuaded to abandon their investment in Windows specific programming, and make an investment in platform-indepentent APIs, Microsoft has little to fear. This will probably not start to happen until the time Linux gains a market share somewhat above the market share that Apple had when Autodesk stopped developing AutoCAD for it.

Not supporting Windows 2000 and earlier with IE security updates devalues both those operating systems and those certified to service them. I was going to wait for Longhorn to update my certs, skipping the 2k3 cycle of MCSE stuff. Now I'm going to have to update earlier than I'd like because I'd be irresponsible to keep my clients on any platform that isn't maintained.

Thanks Microsoft, you've given me new incentive to recommend Linux and Mac (both of which I also support).

I suspect that Microsoft used to be worried that the Broswer would become the OS, and so funded the clearly uneconomic policy of producing a large piece of software for free. Then it took over the market, and they didn't care anymore. It was all an overreaction to a threat articulated by Andersen. In any case, it is still utopian to believe that the browser will guarantee "Platform independence": it will when it can run the fastest first person shooters, space trading games and complicated architectural programs like Autocad in full screen mode without a recompile, or rewrite, on Linux, OS X and Windows. Which is to say, never.

"it will when it can run the fastest first person shooters, space trading games and complicated architectural programs like Autocad in full screen mode without a recompile, or rewrite, on Linux, OS X and Windows. Which is to say, never."

It doesn't have to. The OS isn't that nifty GUI you run around in, it's mainly the kernel that decides how that processing power is going to allocated. Hell, I could configure linux to just load the kernel, an X server (basically just a graphics subsystem) and a browser. There's no reason you couldn't launch other apps from within that browser; it's not handling the processing, the server is. All the browser does is provide a graphical framework for the user, in reality you could do everything you usually do from the command line.

Now do I think that this is likely to come about? No, but it's not far fetched at all, hell one could do it now were one so inclined.

Sorry Eoin, I didn't address the issue of recompiles, and platform independence overall. Yes, you're correct. Developers still have to work around the various architectures out there, and browsers wouldn't change that fact. I'd argue, though, that no one uses Windows for the NT kernel. They use it for the GUI and out of simple inertia at this point. When they've ceded a large portion of the GUI pie to an unaffiliated 3rd party I still think that it's a big deal.

ernest young: "Not being 'server' orientated, I have often wondered whether IIS was more loosely written than say, Apache, or was it just attacked more often because of it being Microsoft. After all, they are a tempting target, and all those 'anti-trust' cases have not helped their reputation."

Possibly. It might also be (regardless of any comparative holeyness in code) that crackers and co specialise in MSFT products because they are so common and standardised, thus the payoff of finding an exploit is vastly greater than against its competitors.

http://transitive.com/

It works. It's coming Q4 2004 and Q1 2005.

I'm running Win XP Home. I'm switching to Safari.

If you think that 'transitive' will let you run OSX/Safari on an X86 system then you are severely mistaken. It's just an API translator that lets you run UNIX binaries compiled on different architectures on other architectures without having to recompile. The systems have to support the same basic API however - i.e. you have to be running a UNIX OS and the binary you are trying to run has to be compiled for a similar UNIX OS.

It's nothing new. Linux binaries run on FreeBSD using a very similar method.

Guy,

Thank you for replying.

I feel it safe to assume that the next 'King of the Software Jungle', will experience the same sort of attacks as those on MS.

As I mentioned above, the crazies will always be with us...

Two pickups -

1) I run Firefox on my three PCs and have done since v0.3, it doesn't seem to crash, but maybe that's because my OS's are stable and my web use tends to be corporate sites. Certainly it is more reliable than IE on my Windows PC.

2) I run a major business on a predominantly Linux back-end, my business has migrated away from MS server products in order to achieve levels of stability. reliability and interoperability which MS seems unable to provide. For me Microsoft has not 'stabilised and homogenised' IT, it has corrupted it and inhibited the development of interoperable systems by eschewing established standards in favour of its' own re-invention and by reserving some functionality for it's own exclusive use to gain (illegally as judged in both the US and EU) competitive advantage in the applications development business.

Ernest Young asks "Will we go back to the days where you had to be tech savvy to use a computer, because every company had it's own way of doing things, or will 'open source' be the answer.". My not very humble opinion is that by attempting to stifle competition rather than meeting and rising above it, Microsoft has significantly held back the development of computing over the past two decades, and the emergence of open source products which have been developed without extensive funding and yet exceed Microsofts products in quality, usablilty and reliability provide a graphic demonstration of how Microsoft has failed us all. If the leading open-source products had been developed with the benefit of the financial resources expended by Microsoft in developing its products we would all be using faster, cheaper, easier to use and more reliable systems today.

craggy_steve,

Interesting viewpoint, as I remember, it was IBM that initially handed a monopoly in OS' to MS, and MS appeared to take it's business plan from IBM, almost by default. To paint MS as an unprincipled business ogre is a bit unfair, they were only protecting their business as they saw fit, and in the same manner as IBM, and others, had done, and still do.

It was only later that MS went around buying up the competition, and bruising a few egos in the process. The anti-trust actions by both the EU and the US, were motivated more by politics than anything else, as several other companies had, and still have, similar business practices as the ones that MS was indicted for.

I always felt that MS did a good job in the late 70's and early 80's in setting some standards at a time when every manufacturer did things differently. I think they just got overtaken by events and their own success, and a belief that BG was some sort of genius, which he is not...

There has always been choice for users, Apple has always been with us, and should have faired a lot better than it has, after all, everyone says that the MacOS is a far superior system, but it remains the domain for only a small percentage of users. and they had to be Win compatible to survive.

Why should this be? maybe software developers did not fancy placing their bets on that horse. Unix has also been around for a long time, and it was always a hotbed of disagreement as to who owns this code or that code. Linux is the first real attempt at standardising and marketing unix for general use, and long may it reign...

Software and applcations are what drives computer use, and if developers opted for Windows, then surely they are as much to blame as MS for any stultification of the IT industry.

By coincidence I have today, received my first 'critical update' from Mozilla, over security concerns. Deja vu?
As more 'off the shelf' software becomes available for Linux, it will become even more competitive, and not just the domain of geeks.

ernest young: the Linux kernel is actually not a direct descendant of unix. It's a work-alike, written from scratch but copying the unix standard (POSIX) so it looks similar in use.

Julian,

So, it's a 'knock-off'.

I am sure you are smart enough to get the drift of my comments, that the partisanship re the various operating systems has acquired an almost political character, which extends way beyond mere discussion over 'closed' or 'open source' code.

I too regret the passing of the days when you could call an illustrious Uni. or Institute, and discuss some esoteric point with someone who was working on a similar project. In those days computing was a very small world. Nowadays it is all very commercial, hence all the animosity.

I've just started using Firefox. Much better than IE. But I still have to boot up IE every now and then for some non-standard websites...

Death to Microsoft!