Fog Creek Software
Discussion Board




How Microsoft Lost the API War

Very depressing article. What else can I say.
Don't get me wrong. I really really like your articles, but this one is just very dark.

Cheers,
- Madsen -

Mads Boyd-Madsen
Wednesday, June 16, 2004

Yeah, the article is very interesting. Makes me think twice about what type of application to develop.

So, let's say I do develop in .NET, is there anyway to have the winforms exe run from a URL in a nice way? Even with such requirements as I know that the client host has .NET framework and IE6.

Interesting
Wednesday, June 16, 2004

A very good piece.  I liked the quote from Raymond Chen "I get particularly furious when people accuse Microsoft of maliciously breaking applications".

Microsoft would NEVER do such a thing.  (wink wink)

My Cousin Vinniwashtharam
Wednesday, June 16, 2004

DR DOS

christopher baus (www.baus.net)
Wednesday, June 16, 2004

Sorry for my over quoting, I would like to be just as precise as possible.

You wrote:

Two new web applications, Gmail and Oddpost, both email apps, do a really decent job of working around or completely solving ... issues.
....
Which means, suddenly, Microsoft's API doesn't matter so much. Web applications don't require Windows.

Oddpost FAQ:

Hey, that reminds me: what are Oddpost’s system requirements?
Oddpost requires Internet Explorer 5.5 or higher for Windows XP, 2000, NT, Me, or 98. Please note that we provide POP and IMAP access to Oddpost, which means that you can still get at your mail when visiting Oddpost from a Mac....

Indeed sir, Just as you say ( Jeeves ? )

Edouard Chevtchouk
Wednesday, June 16, 2004

Another great piece from Joel. Thanks.

The problem, then, that Microsoft is going to run into is that LAMP is destroying Microsoft on the web server. Whether or not ASP.NET is more productive is not as important as LAMP's dramatic edge in accessibility (both in cost and deployment at hosting providers).

I will put in a small defense of Apple. First, if you look just at the home market, Apple share is double digits and may even be north of 20%. And certainly the bulk of the interesting home usage is on Macs (iPhoto, iTunes, iMovie, etc.) Second, having a small base does help Apple avoid a lot of the problems that Microsoft faces.

pb
Wednesday, June 16, 2004


Very penetrating article, with just one slight stone left unturned.  How Microsoft didn't just lose the API war, they shot themselves with friendly fire a couple of times.  Several very loud cases can be made for Symantec Norton Disk Doctor, Stacker, etc... Where how not only Microsoft was interested in making life hard for 3rd party developers, but worked a little too hard in competing against them.

Now, on the other hand, the DLL Hell situation could have been resolved, had Microsoft really wanted to, they have the solution to it, WindowsUpdate, but it cannot work if Microsoft milks its control of the desktop as an asset against its own developers(which is something they have done in the past, as you clearly demonstrated with your fine article).  Microsoft COULD have expanded the MSI format to include better dependancy handling and tracked libraries like DirectX straight in WU(like freebsd ports, debian apt and yum do, to name a few).  Of course, that would have meant that developers would have trusted Microsoft with their dependancy trees and updates, and Microsoft would have footed the bandwidth bill for the download of updates.  Instead, many critical applications have their own update system(including Office, which does NOT use windows update).
Here F/OSS users are particularly blessed: one command to update them, and dependancy track them(with apologies to J.R.R. Tolkien).  Debian's apt-get is even used for commercial, licensed software, so I know it can be done.  Of course, that means you trust your OS maker, something we tend to do less and less with each iteration of windows.

I congradulate you again on highlighting so dramatically how the rapid-reinvention cycle of windows, combined with the sheer size of their product lines means they are competing with their own developers while depending on them for their very survival.

Eric Robibaro
Wednesday, June 16, 2004

What has not been mentioned so far is that .Net web applications modify the generated HTML code for some of the .net web controls.  Now the HTML code that .Net itself generates - for non-IE browsers - in some cases is invalid HTML.  Look at what happens if you use an asp:panel control.

I could be cynical and say Microsoft is trying to make .Net applications NOT work on browsers other than IE.

And don't get me started on how VS.Net deliberately corrupts valid HTML markup.

Ken Ray
Wednesday, June 16, 2004

Ironic that that the brakes are on and the theme is: "We're investing on the desktop, ..." but everything is .Net

Mr E Lurker
Wednesday, June 16, 2004

"I will put in a small defense of Apple. First, if you look just at the home market, Apple share is double digits and may even be north of 20%."

Bullcrap.  Care to cite sources?

Mr. Fancypants
Wednesday, June 16, 2004

How can different web applications run by different providers for one user share this user's data? (assuming the user wants them to do that).

a) The web applications must be extended; the providers must be willing to do that and they want to be paid for that.

b) The integration happens on the client side. A simple browser can't do that. Some client-side programming required.

Heribert Slama
Wednesday, June 16, 2004

Dot Net is semi-useless for software that you intend to distribute to the masses, especially if they have to download the runtime. But I have no problem deploying a Dot Net app to a client who is only going to use it internally.

Catatonic
Wednesday, June 16, 2004

Here comes the .NET linking argument again...

But Joel was right about that one too.  Microsoft royally fucked up by designing .NET in such a way that it is all or nothing.  Bloating software downloads by 25 megs and/or having a potentially troublesome and lengthy extra install step has certainly slowed .NET's adoption for client-side applications quite a bit.  I've yet to see much use of .NET for projects that weren't internal-use-only or web based with ASP.NET...

Mr. Fancypants
Wednesday, June 16, 2004

A great article and certainly thought provoke. Personally, I hate web apps so my view is probably biased anyway. But surely certain apps just can't be reduced to a web based format?

I'd like to think that things like ITunes (the music store bits) and blogging clients like ecto (and of course CityDesk although its a different sort of beast), are all just going to herald a brave new world of rich client apps.

Kenshi
Wednesday, June 16, 2004

People are thinking that PC hardware will follow the same path it has been on the past few years, i.e. things will get keep getting faster, but will basically remain the same. I think Microsoft is planning on major changes in hardware, and is preparing to exploit that opportunity.

From what I recall, Avalon is designed to take advantage of very high DPI flat displays. You can't buy this yet, but when they are available MS will have some really cool Longhorn screen savers. If they convince game companies to use the new API's, by showing them how easy it is to develop for the new hardware, there is a big incentive for people to upgrade.

Bill K
Wednesday, June 16, 2004

"If they convince game companies to use the new API's,.."


This is going to be a chicken/egg problem.

Game companies won't develop games that will ONLY work on the new platform.


In terms of new high resolution displays:

One interesting problem is Multiplayer games.  I play Starcraft, which will display only at 640x480 screen resolution. This limitation is to keep things fair between those with big displays and those with smaller displays. I.e., if I could set my screen res higher, I'd see more of the map.

So, it'll be difficult to exploit higher resolution screens for games that have a multiplayer component (and I'm sure they ALL will. It's the future of games).

Mr. Analogy
Wednesday, June 16, 2004


" is that LAMP is destroying Microsoft on the web server. ...important as LAMP's dramatic edge in accessibility (both in cost and deployment at hosting providers)."

Destroying? I think that's a bit optimistic. LAMP is more common that IIS/ASP, but I haven't seen any figures that suggest that MS is losing ground fast enough to refer to as "destroying".

Also, I can't spit without hitting a hosting company offering .NET.  They're not any less common than LAMP hosting companies. Of course, most corporate shops host their own sites so it's really a moot point.

"First, if you look just at the home market, Apple share is double digits and may even be north of 20%."

You're kidding, right? Seriously. North of 20%? Even Steve Jobs doesn't dream that big.

"And certainly the bulk of the interesting home usage is on Macs (iPhoto, iTunes, iMovie, etc.) "

I suppose it depends on what you call "interesting". None of those apps you mention really appeal to me. I'm a gamer, so Mac doesn't offer much that is "interesting" to me.

Mark Hoffman
Wednesday, June 16, 2004

> Microsoft was interested in making life hard for 3rd party developers, but worked a little too hard in competing against them.

Exactly!

Look at history.  That's why I have a hard time trusting them. 

Developers, Developers, Developers, Developers.

Yea unless you go into a space deemed too juicy for them to resist.  Then watch out!  Feel the wrath of 1000 developers come down on you.  Watch your top guys leave with million dollar signing bonuses.  Watch your bread and butter  become an "integral part" of their overall platform.

I have a lot of respect for Eric Sink, but I just don't see how Microsoft can be a good partner in that space.  All it is going to take is a decent source control system free with MSDN, and the game's over.  I'm sure he's got this in mind, but it has to be frustrating. 

That's exactly why people start to resent Microsoft.  They'll use their advantage and bowl you over.  This strategy can only work for so long, before developers get smart and start looking for alternatives. 

Microsoft might have won "fair and square," but they played a pretty physical game.

christopher baus (www.baus.net)
Wednesday, June 16, 2004

>> I'd like to think that things like ITunes (the music store bits) and blogging clients like ecto (and of course CityDesk although its a different sort of beast), are all just going to herald a brave new world of rich client apps.

Agreed.  Everywhere  I look, I see successful web enabled desktop apps (I'm listening to MusicMatch).  Other than bank access apps, account mgt, and OWA, can't think of many successful web apps (and SalesForce.com) The thing is that the customer doesn't care what technology is used.  Other than deployment, which is a problem that is solved, there is absolutely NO advantage to a pure web application. Some folks are too developer & technology centric to see that the hype of the 90s is mostly bull.

Distributed processing is the trend. Centralized processing is stupid.  1000 people using a web enabled desktop application have 1000 pentiums of processing power.  It would be stupid to try to compete with that processing power with one computer.

Gunnar Skogsholm
Wednesday, June 16, 2004

Just a couple of thoughts: the thing about the lack of C++ programmers on Windows and the salaries they currently make: is Joel saying he's having a hard time finding a someone to develop who knows what pointers are?

And the thing about Microsoft: does Joel feel he's about to be smashed just like Eric Sink was(no matter what he says)?

RP
Wednesday, June 16, 2004

"And the thing about Microsoft: does Joel feel he's about to be smashed just like Eric Sink was(no matter what he says)?"

There is good reason for both Eric and Joel to be feeling a little animosity towards Microsoft.

I forget the name dujour, something like Visual Studio Team, but MS will be rolling out a new source control system and a bug tracking tool around the same time it ships VS 2005.

Much has been made that it won't be free, but let's face it..Most MS shops have MSDN subscriptions so it might as well be free because it will be available with MSDN.

So here I am in a MS development shop with my MSDN Universal subscription. I'm looking for a bug tracking system and a source control system. Hmmmm...Where do I go? Do I try to get money to buy Fogbugz or SourceGear? Or do I just download Microsoft's tools?

Not a tough choice.

Granted, I'm assuming that Microsoft's product aren't crap and they are just as good as what Joel and Eric offer. But assuming they are decent quality, Eric and Joel are gonna have a tough time selling their apps to new customers.

I've got a great deal of respect for Eric Sink and my heart sank a little when I heard the news that MS was going to enter the source control market because I knew this would be bad for Eric. (Regardless of what Eric says, I think it will be far worse for him than what he is saying.)

So I can understand MS developers beginning to feel a little resentment, anger and distrust of Microsoft. If you are any type of market that deals with developers, there are good odds that MS will eventually come and squash you.

And that sucks.

JT
Wednesday, June 16, 2004

See  http://radio.weblogs.com/0001011/2004/04/03.html#a7150  for some earlier discussion on this.

It's almost funny to see how Microsoft is now trying in some ways to portray earlier versions of Windows as fundamentally broken somehow, as some kind of twisted excuse of why they have to deploy stuff like Avalon for Longhorn only.

Like Scoble gives this excuse:
"Almost all of Avalon requires the GPU. Even fonts now are being rendered in the GPU. Avalon simply WILL NOT be portable to older OS's. Not possible even if we wanted to do it."

As if somehow it is not possible to also do font rendering without the GPU. Well, guess what, I'm running on XP, and what is this that I see on my screen, oh my god, it's a bunch of fonts being rendered. It's too bad that they weren't aware that XP had some font rendering technology in it when they were designing Avalon.

There is still some time for them to regroup and fix this, but their desire from their upper management to have some kind of Longhorn technology gap similar to DOS/Windows is really strong. It's sort of a case of "careful what you wish for, you just might get it", because they are on track for creating a damaging artifical technology gap instead of a real one.

It was sure depressing to see Microsoft's response to the earlier "please may I have a linker" thing too, which I think was also right on the mark. It seems like the people over there have been kind of blinded by the success of the CLR on server-side applications, and are just ignoring the massive bloat download problem that they have created for client-side apps. They wouldn't have to look any farther than Microsoft's own product development groups to find out why it is not being deployed for their own client-side applications (Visual Studio doesn't count).

Michael
Wednesday, June 16, 2004

Ok we've been kicking your article around at work now for a while and there's one thing we can't answer... Joel, What about the next generation smart clients? You get a lot of both worlds from that.

You get easy deployment just like a website
You can take advantage of the new api on the server side
You get richer ui
You get rapid development
We're supposing Mono will support smart clients eventually too, so they'll run on most machines.

Is the only drawback the requirement of a .net runtime? Assuming we're all going to be using some memory-managed language soon anyway, does that matter?

Have I had too much Kool-Aid today? Probably.

Bucho
Wednesday, June 16, 2004

One thing:  you say this:

"The real significant productivity advance we've had in programming has been from languages which manage memory for you automatically. It can be with reference counting or garbage collection; it can be Java, Lisp, Visual Basic (even 1.0), Smalltalk, or any of a number of scripting languages.

Whenever you hear someone bragging about how productive their language is, they're probably getting most of that productivity from the automated memory management, even if they misattribute it."

I think you miss a lot of the productivity gap in thinking this.  You might try looking here:

http://www.cincomsmalltalk.com/userblogs/avi/blogView

and have a look at Continuation based web development.  And take a look here:

http://www.cincomsmalltalk.com/userblogs/travis/blogView

just keep reading down - now, try to figure out how you would do anything like that in any of the "mainstream" languages and runtimes....

James Robertson
Wednesday, June 16, 2004

Great article, Joel. Thanks for sharing it.

ian ashley
Wednesday, June 16, 2004

Oh yeah, there's another thing I wanted to mention. It's not just making APIs backwards compatible to existing Windows versions that has been thrown out recently at Microsoft, there is also a new wave of ignoring what used to be a high priority of presenting a uniform platform on a relatively low performing machine.

This priority seems to be gone with Longhorn, now they are planning on relying on faster hardware (always a risky gamble, IMHO, but even more so nowadays as we near limits in power consumption and heat dissipation), and also relying on more specific hardware such as programmable GPUs.

I predict that this will be a big problem, because there will be a significant wave of devices for a long, long time, that will choose stuff like lower power consumption over CPU power (think like a TabletPC or a laptop with a 24 hour battery life). Also, the really low priced end of the spectrum will also be lacking in great CPU and GPU power for a long time to come.

Longhorn will not run well on these types of devices, right now their plan is to scale back to a totally non-themed Win2k like shell.

This is going to suck, because someone who gets a brand new machine of one of these types (either low power, or just the absolute cheapest), is actually going to end up with a crappier experience using Longhorn than they would with XP.

Furthermore, if they try to run applications on this device that were developed using Avalon targeting the more powerful computers, the apps are probably going to behave like crap. Stuff like animations will either stutter or not show up, it is just not likely to work smoothly for a number of reasons like that.

This is another pretty big area that is being screwed up right now, it's another big change from all previous releases of Windows.

Michael
Wednesday, June 16, 2004

>Re: mixed message on Longhorn

This is what is going on with C++ now.  Personally I don't think C++ needs garbage collection, but Herb Sutter is hell bent to make sure MS ships a C++ with .NET garbage collection amoung other .NET specific support.  Which pretty much deprecates "Managed C++."  If the C++ memory management isn't confusing enough, wait until this thing ships.  C++ is becoming the PL of programming languages.  It is like they want to destroy C++ on Windows, but I digress.

In the mean time the only practical way to integrate C++ with .NET is either via "extern C" or COM.  The message certaintly isn't clear. 

christopher baus (www.baus.net)
Wednesday, June 16, 2004

I have no idea how the quoting on this forum works.  Forgive.

>> I'd like to think that things like ITunes (the music store bits) and blogging clients like ecto (and of course CityDesk although its a different sort of beast), are all just going to herald a brave new world of rich client apps.

What's driving this little surge in the popularity of fat/smart/thick/real clients is surveillance and control.  Need to track someone's habits more easily than you can on the web?  Need to make sure users don't copy that floppy?  Well, then you have to bother to write and convince people to download real code.  It helps to have the Apple brand (and the excuse of "good UI") behind you.

Of course, if you stick with Microsoft standards, all the DRM is built into the clients ("components?") shipped with the OS anyway.  But then only Microsoft gets to do the spyi^H^H^H^H^H^H^H^H^H^H^Hextract the 'performance statistics.'

"Agreed.  Everywhere  I look, I see successful web enabled desktop apps (I'm listening to MusicMatch).  Other than bank access apps, account mgt, and OWA, can't think of many successful web apps (and SalesForce.com) The thing is that the customer doesn't care what technology is used.  Other than deployment, which is a problem that is solved, there is absolutely NO advantage to a pure web application. Some folks are too developer & technology centric to see that the hype of the 90s is mostly bull."

That's a pretty verbose way of saying "people will use what works."  And they do.  Browser (or HTML) bugs and DLL/deployment hells are equal poisons.  "Web-based" does mean you can fix it for most of the people most of the time... or break it for most of the people in one fell swoop (but it's easier for a user to sit and wait without Hotmail and with a good excuse, than to have to 'do' something and potentially not accomplish the task).

"Distributed processing is the trend. Centralized processing is stupid.  1000 people using a web enabled desktop application have 1000 pentiums of processing power.  It would be stupid to try to compete with that processing power with one computer."

So is this an argument for smart clients (because they're "smart"), or web clients, because they're "networked?"

Distributed processing is great if you can find a problem(/solution) without security or latency concerns.  It's more "hype of the 90s" if you can't.

As to what will "win?"  Whatever works best within the bounds of attainability (where said bounds can, of course, be annoyingly limited) for the most people.

Also note that DR-DOS (and OS/2) weren't applications, they were OSes; Microsoft don't want you to have a bad XPerience with 'Windows software,' but they have no institutional qualms against denying you the XPerience (Hello, Mono?) if you do something naughty, like try to compete.  After all, users blamed and threw out DR-DOS, not the Windows that ran so well on their copy of MS-.

Anonymous Peon
Wednesday, June 16, 2004

"I predict that this will be a big problem, because there will be a significant wave of devices for a long, long time, that will choose stuff like lower power consumption over CPU power (think like a TabletPC or a laptop with a 24 hour battery life). Also, the really low priced end of the spectrum will also be lacking in great CPU and GPU power for a long time to come."

B-b-but hardware will be free!!11oneone

The only boring thing is that low-power GPUs actually are catching up, so I can't see things being too much worse than Quartz on an iBook.  Even if I'd like to.  Hell, you can get a cellphone today that'll hardware-accellerate GLQuake.

Anonymous Peon
Wednesday, June 16, 2004

Nice insights.  However, the appeal of upgrading/changing to a new OS is not all about applications.  The security and stability of a platform is also very important, as is manageability/TCO, and performance (getting the most bang for your buck).  I agree that it is very difficult to keep up with all the dev tool/platform changes lately from MS, so they need to take a hard look at making stuff like WinForms work under Avalon and co.

There are some really neat things coming up in Longhorn, but if MS is having trouble getting developers to upgrade to .NET right now, they will face an even bigger hurdle in the next couple of years. 

The good news for MS is that I have seen a lot of developers and corporations start moving to the .NET platform, especially those who are sick and tired of the shortcoming of Java and its really horrible JNI.  As more and more people realize they don't have to throw out all their old stuff to start using .NET because of its stellar interop features, more and more people will adopt the technology.  The last big hurdle for .NET is that it does not come with a linker, although there are several good 3rd party products in the works.  See the following:

http://www.remotesoft.com/linker/

Those who have never tried .NET are missing out on faster development time and lower maintainence costs.  These two factors taken together free up resources that can be refocused toward creating new features for the end user.  Therefore, while ramp-up time may seem to be a waste at first, as Joel believes, the long-term benefits are certainly worth the short-term investment.

Kurt
Wednesday, June 16, 2004

On the LAMP vs. APS.NET war, here's a personal account:

I'm not a programmer by trade.  It's a hobby I dabble in.  I work in electronics.  A couple of years ago I bought myself a cheap computer to learn Linux on, and set up a server.  I taught myself LAMP because the peices were all there in the Linux distrubution and it looked interesting.  When the company I work for needed to provide up to the minute information to their customers from their database, a web app was the obvious way to go.  I wrote the software to export the data from their database to a hosting service and the search engine to allow the customers to look up the information on the webhost's servers.  This took me, a self-taught amature, 3 months to develop (in my spare time), earned me a trip to Hawaii with my wife, and saved my employer thousands.

My total monetary investment: $200.  Try that with ASP.NET!

This experience taught me the power of the browser as a UI.  It's by far the easiest GUI programming environment I've learned.  Like Joel mentioned, it doesn't do everything, but if it will do what your application needs, it takes alot of the grunt work out of your UI coding.  It also taught me the power of open-source software.  Open source means open documentation, and I had all the documentation I needed instantly in my browser window.  Easily searchable - right to the answer I needed - complete with examples and comments from other developers.  I remember thinking to myself 'WOW! this is so easy to learn, easy to program with, and powerful!  When it really catches on it's going to turn the software industry on it's head!'.

scotty d
Wednesday, June 16, 2004

"Distributed processing is the trend. Centralized processing is stupid.  1000 people using a web enabled desktop application have 1000 pentiums of processing power.  It would be stupid to try to compete with that processing power with one computer."

Dude, those pc's are just idling.  You don't need a server as big as the number of client pc's connecting.  You really should get out more.  It's called Unix, iSeries, OS/390, etc.  Work that matters in business is done on the server.  The client be it pc or terminal is a window to it.

Logical
Wednesday, June 16, 2004

Microsoft used to be primarily a desktop software company with a good desktop os.  Now there server side is scaling up and they are trying to break into the enterprise where Unix and Java are stalwarts.  That is why as Joel says MSDN is winning.  The reason is that the server side is not only important for enterprise, but is becoming much more important for apps that were traditionally the realm of the desktop.  No touch deployment, and non-existant client side maintenance make web apps a huge win, even with factoring the UI degradation. 

So if we take this story further along.  What can Microsoft sell?  If the desktop is less important you sell less.  It has very few server components that aren't matched and or exceeded by other vendors with respect to reliablity or cost.  So that means you don't make the best, you don't make the cheapest.  Maybe you can make the prettiest.  Microsoft is gotta be thinking along the lines of reinventing itself.  I think Cringley was probably onto something when he talked about Microsoft trying to be a bank (THE bank).  It will take something like that for them to maintain the level of profitablity they have had.  Right now I'd rather be buying SUNW than MSFT.  SUNW has a good shot at going up, MSFT has only a slim chance.

You might think Microsoft could make a .net especially for Unix, the problem is anything they ever make for Unix or Linux, they won't be able to charge anything for, so that idea is out.

To summarize Microsoft ran into not being able to fool all of the people all of the time.

Logical
Wednesday, June 16, 2004

>> Dude, those pc's are just idling.  You don't need a server as big as the number of client pc's connecting.  You really should get out more.  It's called Unix, iSeries, OS/390, etc.  Work that matters in business is done on the server.  The client be it pc or terminal is a window to it. <<

Depends on the software, doesn't it?  You should get out at least once.  Some are stuck in the 90s, others are stuck in the 70s or 80s. 

Gunnar Skogsholm
Thursday, June 17, 2004

Re: DR DOS and breakage.

Microsoft didn't break DR DOS, they changed Win386 so that it would not run on DR DOS, and if I remember rightly did so during our beta cycle and just before release.

It got fixed though, and overnight if I recall, though I had nothing to do with that.

Simon Lucy
Thursday, June 17, 2004

It was just a warning in one beta. There was nothing to "fix", if I recall.

Just me (Sir to you)
Thursday, June 17, 2004

"Depends on the software, doesn't it?  You should get out at least once.  Some are stuck in the 90s, others are stuck in the 70s or 80s. " 

Then I guess almost all of the Fortune 100 are too.

Quad
Thursday, June 17, 2004

There is a big difference between 1000 people running financial and ERP style apps (one big ol' database) on a server and having 1000 copies of Excel running at the same time on the same server.

Bear in mind you can write pretty huge financial and ERP systems USING Excel.

DomF
Thursday, June 17, 2004

"From what I recall, Avalon is designed to take advantage of very high DPI flat displays. You can't buy this yet, but when they are available MS will have some really cool Longhorn screen savers."

For a preview of this technology, check out Mac OS X.

Jim Rankin
Thursday, June 17, 2004

""And certainly the bulk of the interesting home usage is on Macs (iPhoto, iTunes, iMovie, etc.) "

I suppose it depends on what you call "interesting". None of those apps you mention really appeal to me. I'm a gamer, so Mac doesn't offer much that is "interesting" to me."

Evidently, you don't take pictures, listen to music, or use a digital camcorder, then.

Jim Rankin
Thursday, June 17, 2004

"Other than bank access apps, account mgt, and OWA, can't think of many successful web apps (and SalesForce.com)"

Amazon.  Ebay.  Google.  Travelocity.  Netflix.  MapQuest.  Bunch of stuff from Yahoo.

Etc. etc.

Jim Rankin
Thursday, June 17, 2004


"Evidently, you don't take pictures, listen to music, or use a digital camcorder, then. "

I use Photoshop for my pics. Did you hear? They have a version for the PC, too.

I don't listen to music much and I don't use a digital camcorder. But if these were more appealing to me, there are plenty of PC software.

The OP suggest that these type of applications are more widely available on the Mac, or that somehow the PC is second-rate when dealing with these types of app. I'm not going to get into a religious war over Mac and PC, but don't disconnect your brain when your heart swells up with admiration for that colorful Apple sitting on your desk.

Mark Hoffman
Thursday, June 17, 2004

*  Recent Topics

*  Fog Creek Home