Fog Creek Software
Discussion Board

Microsoft's Future

I apologise if this has been raised before, but what do you think will happen to Microsoft when everybody has 100Mbit broadband straight into their house and office?

There would be no need for Windows or Office for the majority of users as the web would be quick enough to run whatever Web App is required. And they wouldn't have to use Microsoft software either.

For example: I already use Hotmail for my email and if I had a fast enough link with remote storage I might use an online word processor or spreadsheet too. That way my data would be available wherever I happen to be. Just like Hotmail.

Tuesday, April 20, 2004

Whoa there, cowboy! You're extrapolating like crazy from a small trend.

First of all, I highly doubt you would want to spend very much time using a word processor delivered to you in the form of a web page, especially given the state of HTML. User interfaces designed for IE 6.x are klunky and slow. This is not entirely an accident -- I don't think Microsoft *wants* it to be possible to develop great user experiences in HTML form.

Second, lots of people do have very high bandwidth connections, and I have yet to meet the person that does their word processing and spreadsheeting in a web browser. The Microsoft Office business is doing fine, thank you very much.

Even if it were to be threatened by hypothetical web-browser-delivered productivity tools, you can bet Microsoft would get into that business too.

In the early 90's everybody thought IBM was going to disappear because they just didn't understand PCs and could not get over their mainframe mentality. What people forgot is that a large successful company has a LONG way to fall and during that fall they have YEARS to regain their footing. Microsoft is the same way now. Whatever fanciful theory you come up with about how Microsoft is going to be wiped out (web apps, broadband, Linux, antitrust law, or black helicopters) can't change the fact that Microsoft has enough cash to keep operating for something like 5 years even if their revenue suddenly plummetted to zero without laying off one programmer. And their revenue is not going to zero tomorrow: even in their worst nightmare scenario you might expect their revenue to drop, say, 20% a year, so now they can probably hang around for ten years before they have to start contemplating the first round of layoffs.

So it's not enough to come up with some reason why Microsoft will be threatened by a change in technology: you also have to assume that they would fail to respond to such a change for years and years.

Joel Spolsky
Fog Creek Software
Wednesday, April 21, 2004

Joel's right, but it still is possible for really big companies to fall all the way to the ground.  Anybody remember RCA & DEC?

J. Peterson
Wednesday, April 21, 2004

Which RCA fell all the way to the ground?,,CI257,00.html

Maybe this is a revival of RCA, but they claim a contiguous history. (Note that I started the web search based on the assumption that JVC was actually a continuation of RCA under a different name, with different owners, and in a different country. That may be at least partially true, but I stopped digging once I found the RCA site).

DEC seems a more likely case for 'fallen and can't get up':

Ron Porter
Wednesday, April 21, 2004

Is it me or Joel really sounds like a Microsoft apologist. He has valid arguments, but the way he passionately talks about Microsoft makes you wonder.

I have started observating.
Wednesday, April 21, 2004

If he has valid points, why should I care if he's an apologist?  Aren't valid points, um, valid, no matter who makes them?

Boofus McGoofus
Wednesday, April 21, 2004

Anyway, so you are saying that Microsoft does not *wants* it to be possible to develop great user experiences in HTML form?

There are other players out there OK,
if they show signs of success, Microsoft will all over and Player 2 ... Game Over

I think Macromedia has come up with some ideas using XML and Vector stuff, but they still have some to go to establish this I think. But what is threating them is ... Longhorn which use similar technologies for UI, and since Longhorn is developed by MS, I guess the guys at Macromedia fear this like ****

Web applications will be the future.
OK technology like HTML, Flash, HTTP, server side stuff and so on is maybe childish today, but what does we know about tomorrow?

Maybe I am wrong, but after all I've read and seen by Microsoft and other major players lately they are showing signs of the future that is web apps (not necessary through browser). I don't think anyone can stop this, Microsoft or not.


Wednesday, April 21, 2004

I think Hotmail is owned by Microsoft.

Wednesday, April 21, 2004

You're already accessing your apps from a computer on the net.  It just happens to be the same one you're typing on.

Tayssir John Gabbour
Wednesday, April 21, 2004

I think I see the future.  We will all be using computers called "thin clients" that access the web for all of our functionality.  "Fat clients" like Microsoft Office will go away because they cause everyone to need a powerful computer at the client end. Since high-power client computers are soooo expensive (like hundreds of dollars), we need to push that expense to a server on the net.

While we are at it, we will subscribe to web services that provide "lite" word processors, because nobody really needs all of that word processor functionality that Microsoft shoves down our throat.

I can't believe nobody has ever thought of this before.  Bill Gates is doomed when word of this gets out!!!

Thursday, April 22, 2004

Kyle, I sincerely hope that was sarcasm.  If so, very nice.  If not, I think you are pretty naive.  Personal opinion, mind you.

Aaron F Stanton
Thursday, April 22, 2004

I'm reading a book about Xerox PARC at the moment and one of the factors that motivated them to develop the Alto was the desire to get away from large time sharing systems where you felt guilty if you weren't using the thing. They wanted to give computing power to individuals.

John Topley (
Thursday, April 22, 2004

Good point about time-sharing, old mainframes used to have time allocations for the users so you would have to get ready for your time slot and then use it and then stop. This was because of limited resources. Any general thin client solution would have to be prepared to have limitless resources available.

I have seen spreadsheets that will bring *any* PC to it's knees - how will some remote server cope with that kind of thing without limiting the user in some way? Fat clients are not going away any time soon IMHO and if you have fat clients why wait for any kind of download?

Thursday, April 22, 2004

More likely MSFT will rot from the inside, sorta like the way IBM did.

They're a big comapny now, and if there's one thing I know about big comanies is that they get slow and stupid. And political. If it hasn't happened to MSFT yet, it will.

The most dangerous game MSFT is playing, IMHO, is trying to move everyone to a $30/mo/person subscription model for Windows and Office. Companies out there just don't have the money anymore, and if MSFT squeezes too hard, everyone will phase change, and look for another option.

Unlike some people, I don't knee-jerk hate MSFT - youi may despise their business practices, but when it comes to accomodating programmers who want to write stuff for their platforms, they're about as nice and friendly as you can get. Everyone likes to laugh at that Ballmer video, but there is a kernel of truth to it.

anonymous strikes again
Thursday, April 22, 2004

All these notions of using applications running on somebody else's computers are obscenely stupid and pointless, initiated by people with shabby insight and perpetuated by frantic lemmings.

Taking account of Joel's absolutely correct argument, the only grave danger Microsoft faces is not a technical but a market issue: the alienation of their customers and potential customers.  If most people decide they can't trust Microsoft with something important like the software that rules their increasingly data-centric lives, Microsoft will tank no matter how good their technology portfolio and labor force might be.  This too will never happen.

As Ballmer -- likewise Carly Fiorina -- gets deeper into bed with Hollywood seeking new markets to corner, he skirts forgetting who pays for his product, but even that will never go so far as to wound the company's reputation deeply enough for catastrophe.  They'll also resolve the security nightmare of carelessly dropping a swiss-cheese desktop platform into a networked world.  Since that hasn't hurt their credibility already, it never will.

Microsoft is a fixed star.

But on the bright side for those who fear them, you can already do pretty well entirely without Microsoft.  GNU/Linux if you're of the fiddling variety, Mac if you're into easy productivity and beauty, and a few others too.  I think the options will only get broader.  For every software vendor obsessed with market share, there is another one-person shop willing to craft nice solutions for a niche market such as the 3-million-new-computers-a-year Mac niche.  Microsoft's dominance in the corporate world has utterly failed to reach absolute monopoly over software platforms and office suites. 

Nature favors variety.

Thursday, April 22, 2004

On the other hand, I'd have no problem using a word processor on my local machine that was well integrated into a private web storage mechanism, allowing my key documents to be saved onto a central, well-managed server with a local cached version available.

Ben Combee
Thursday, April 22, 2004

The real problem for Microsoft is that everyone realizes just about every conceivable feature has been already shoved into the word processor, and, hey, everything I want to do I can do in Star/OpenOffice anyway, and we're running out of ways to break file compatibility.

Windows and Office can no longer grow significantly faster than the general economy, as they are already ubiquitous.  Linux and OpenOffice will just continue to slowly eat away at their profit margins.

As for web apps, something always pops up that requires dedicated resources.  Now people upgrade their computers to edit their photos, music and videos.  Big honkin' hard drives, tons of RAM, and processors with lots of cache (interestingly, clock speed is becoming a little less of a selling point).

An example of the convergence of all this is iTunes.  It's the front end for a web store, but it's a desktop app (although using a fair bit of HTML).  It also rips MP3s, requiring a decent processor on your desk.  And you need a lot of HD space to save all your songs.

Then you need a high speed FireWire or USB 2 interface to sync everything with your iPod (this was the key initial advantage iPod had over its competitors).

Jim Rankin
Thursday, April 22, 2004

"As Ballmer -- likewise Carly Fiorina -- gets deeper into bed with Hollywood seeking new markets to corner"

Actually Steve Jobs, I think, is a bit more of an insider with the Hollywood/entertainment crowd.

Jim Rankin
Thursday, April 22, 2004

"Actually Steve Jobs, I think, is a bit more of an insider with the Hollywood/entertainment crowd. "

WITH?  How about he IS an insider. Can anyone say P I X A R.

Mr. Analogy
Thursday, April 22, 2004

"As Ballmer -- likewise Carly Fiorina -- gets deeper into bed with Hollywood seeking new markets to corner"

I'm fairly sure that Carly Fiorina has never been in bed with anybody.

Caroline from The Apprentice though... that's a whole different ballgame. Hubba-hubba.

A. Portly Guy
Friday, April 23, 2004

Broadband is irrelevant.

You can do most routine tasks on a computer without any Windows software whatsoever. You just install Linux on your PC.

The fact that this is not happening shows that MS is safe for some time yet.

Stephen Jones
Friday, April 23, 2004

The shift from desktop apps to web apps has already started.  It's slow so far, but happening.  Here are some examples:

- Webmail, so far only your mum uses it, but gmail may change that.

- CD-ROMs of information and encyclopedias - the web did away with those.  Does anyone buy those phonebook CD-ROMs anymore?

- Newsgroup readers - I haven't touched a newsgroup reader really since Google Groups started.

- CRMS like

It will take a long time until you don't need any desktop applications, but web applications will slowly errode most desktop applications one day - kind of like how PCs slowly erroded mainframes.

Matthew Lock
Friday, April 23, 2004

Actually Paul Graham has a nice take on this:

"Web pages weren't designed to be a UI for applications, but they're just good enough. And for a significant number of users, software that you can use from any browser will be enough of a win in itself to outweigh any awkwardness in the UI. Maybe you can't write the best-looking spreadsheet using HTML, but you can write a spreadsheet that several people can use simultaneously from different locations without special client software, or that can incorporate live data feeds, or that can page you when certain conditions are triggered. More importantly, you can write new kinds of applications that don't even have names yet. VisiCalc was not merely a microcomputer version of a mainframe application, after all-- it was a new type of application.

Of course, server-based applications don't have to be Web-based. You could have some other kind of client. But I'm pretty sure that's a bad idea. It would be very convenient if you could assume that everyone would install your client-- so convenient that you could easily convince yourself that they all would-- but if they don't, you're hosed. Because Web-based software assumes nothing about the client, it will work anywhere the Web works. That's a big advantage already, and the advantage will grow as new Web devices proliferate. Users will like you because your software just works, and your life will be easier because you won't have to tweak it for every new client. I would not even use Javascript, if I were you. Viaweb didn't. [16]"

Matthew Lock
Friday, April 23, 2004

"The shift from desktop apps to web apps has already started.  It's slow so far, but happening"

Dude, don't be so rigid. Web apps are crapola compared to what you can do on a PC. If you said that connected applications were the wave of the future, I might almost agree with you.

The thing about computers in general is they are able to accomplish so much, why would you niche them into such a specific role as web apps? I don't want to run my media center off the web. I don't want to run my games off the web. I do want to connect my computer to the web and use the huge network available.

Does this mean web apps are useless, no. Collaborative applications are good for a web solution for the most part.

Technology moves fast, but it never stops doing what it did before.

If you think that everything will be web apps in the future, talk to the CEO of a bank and see where they will want to store their data from now until the end of time. I am sure they will not say the web.

Humans want to interact and communicate with minimal effort. The Internet is part of the solution to this. Applications that help people communicate and share information, in my mind, is what the web is all about. But that doesn't mean that I will run Doom III on a browser.

Friday, April 23, 2004

Do we still have coal powered trains? the pony express? Technology often "stops doing" what it previously did. CPU speed is doubling every 18 months, which works out to be about a hundred-fold increase every decade. Bandwidth, on the other hand, is doubleing every 9 months i.e. there is a ten-thousand-fold increase every decade. So why are web apps 'crapola'? What is the fundamental thing that prevents web apps from being 'good'? I hope your answer isn't bandwidth or latency :-) I've already covered bandwidth, and as for latency with games e.g. Doom III, all of the important data is already kept on the server, so you don't *really* get a reduction in latency from running on the client.

"Humans want to interact and communicate with minimal effort. The Internet is part of the solution to this. Applications that help people communicate and share information, in my mind, is what the web is all about. But that doesn't mean that I will run Doom III on a browser."

Just because A can be used for B, and has been in the past, doesn't mean that A can only be used for B in the future.

This is a bit of an aside, but I'm curious to know if bank branches actually store all of their data on site. That would seem odd to me, but I'm not in IT, so I don't really know what's standard.

Sunday, April 25, 2004

Webapps are crapola because HTML+Javascript is crapola for user interfaces, and the industry can't agree on a decent rich UI which is based on having lightweight modules downloaded on the fly.

T. Norman
Sunday, April 25, 2004

I happen to like the web UI. I like the way I can move forwards and backwards with the back button. I like the way I can bookmark or email a link to a particular point in an application. I like the way I can just decide to open 2 new windows when I choose. I don't think I'm the loan ranger here either - products like Hotmail are amazingly popular.

By all accounts gmail has a great UI with keyboard shortcuts fast loading etc. I think it would be interesting to revisit this debate when gmail becomes widely available.

I admit that the DHTML landscape is a mess, but there are ways to draw arbitary shapes and lines if you are patient enough:
Therefore there's nothting really stopping anyone making a web UI as rich as they like.

Matthew Lock
Sunday, April 25, 2004

Sorry not enough coffee, replace the "loan ranger" with "lone ranger"

Matthew Lock
Sunday, April 25, 2004

Maybe you can make a web UI as rich as you want, but that doesn't mean the majority of your users will be willing and able to get past the numerous plug-in or browser dependencies in order to make use of it.

Monday, April 26, 2004

The majority of users I know are infected with spyware from allowing everything to install ;-)

Monday, April 26, 2004

It is perfectly possible to create "rich" web applications. Check out the web version of Outlook that comes with Exchange. That's pretty much a perfect clone of the Windows version of Outlook.

I have to admit though, it does take more effort to get something like this working, compared to knocking up a similar Windows app.

Steve Jones (UK)
Tuesday, April 27, 2004

Outlook web access is anything but a perfect clone of Outlook I can assure you. When I access it over trhe dail up connection I find I am taking at least three times as long to do my work as when I am using normal Outlook in the Office.

Stephen Jones
Tuesday, April 27, 2004

While there is a trend these days towards web apps, that does not mean that the trend will continue until ALL applications become web apps.  Keep in mind that our threshold of where a thin client becomes "fat" is also in motion.  The 1gig processor on my computer is far from cutting edge, but it does everything that I ask of it, including office work, browsing and light Photoshop retouches. 

I see no reason for productivity apps like Office to migrate to the web.  Storage of documents may go there (and that also presents challenges), but what would a server-side spreadsheet get me?

As we move into the third decade of spreadsheets and personal word processors the solutions required to construct those applications are becoming generally known.  I do believe these products will cease to be proprietary, but I don't believe they will fully migrate to the web unless desktop computers themselves become obsolete.

Ran Whittle
Tuesday, April 27, 2004

Who did tie original posting on this thread? Was it George Gilder in disguise? 

Throughout the nineties similar arguments were presented. Gilder phrased it well, I think he said something like "assume you had infinite and free bandwidth, what would computing look like?" (search Google for Gilder and Telecosm). He drew all sorts of conclusions and made many predictions based on this assumption. Web Applications, Network apps using Java, Network thin client computing, and the death of Microsoft were often predicted using this argument as a basis.

Few (if any) of these predictions have held up. As a matter of fact, after the Dot-Com and telecom crash, many of these older predictions just seem silly. Very simple things were often overlooked, like what happens when I take my laptop off the high speed connection and want to continue my work offline? Also, as bandwidth increased during the decade, so did hard disk space and CPU speed. So, many of the limitations that people saw in the early 90s that could have been addressed by network computing ended up being addressed by better software running faster and stored locally.

The ironic fact of Gilder's predictions is that people used the argument "assume bandwidth is free and infinite" to justify the huge Telecom capital budgets and debts burdens. When they telecoms and network equipment manufacturers crashed I think they realized that Gilder's assumption was based on low-cost (freely available) bandwidth. The market wouldn't pay the rates to support the build out and now we have bankrupt companies and over capacity.

Changes in the industry are coming, but it is slow, and Microsoft will certainly survive. Here is an example of a new product I use that is bandwidth enabled. I receive video on demand now from my cable company. I can watch movies on my own schedule now over the network.  I used to have to rent them on video or DVD. So, in a way I'm a user of the network model and this has displaced the incumbent video store for me. However, many others have found the Tivo/Replay solution and they rely on local storage rather than bandwidth to watch the shows on their own schedule.  Which is better? Neither. Which will survive? Both. 

So, back to the Microsoft question: Will all applications move to a network model? Of course not. Can some applications thrive in a network model? Yes (on line commerce is a perfect example). Here is the boundary condition test to use: Would you want to use video editing software with storage over a network?

Wednesday, April 28, 2004

*  Recent Topics

*  Fog Creek Home