[kennedy]

Apple Insider reported rumors that the new Tiger Xcode would include an implementation of the .NET CLR.

What's the latest word on that?


Oracle and DB2 have both announced that they are implementing CLR's for their stored procedures! Mono is trying to implement CLR for Linux. It would be a HUGE win to have a CLR on Mac OS X!!

My business, to meet my customer needs most effectively, will be delivering on .NET and MS Office... but I hate that I won't be able to run my own software on my beloved PowerBook. I'm sure someday Mono will get ported to Mac OS X... but that'll be like the decade of waiting for Mozilla... then finally Apple stepped up and delivered Safari! If Apple would step up and deliver a CLR that was even better at compiling to Cocoa than MS's CLR is compiling to Windows API... that would be SWEET!!!


Anyway, any word on the rumors above?

[larkost]

Personally I would doubt that the .NET frameworks will ever come to any non-Microsoft operating system. That is what Microsoft views as being their next set of crown jewels, and having it run anywhere else would undermine their business plans (read: monopoly). They are perfectly willing to have C# runtimes and compilers anywhere else (they have an implementation you can download for MacOS X already), but not the frameworks.

Mono on the other hand already runs on MacOS X. Take a look at the screenshot from their site. I would be very interested if Apple decided to add the ability to create Mono applications through XCode, but I don't think that it will be on the menu anytime soon.

Apple has been doing a lot of work expanding the tschnologys avalible to Java and Obj-C Cocoa programmers (bindings and key-value-codeing being the big ones), and adding a whole other language with it's own compiler just does not seem like the right step to take at this point.

Executive summary: interesting, but highly unlikely. .NET is virtually guaranteed not to happen, and Mono is slightly less likely for 10.4.

Oh... and minor note... there are a number of CLR's out there, Parrot would be another major example. You should be more specific and refer to the ".NET CLR" or "Mono CLR". This is another one of Microsoft's attempts at making the naming of things confusing so that people think they are the only game in town (Windows, SQL Server, Word, etc...). Don't play their game.

[MacGorilla]

It is a port in progress:
http://homepage.mac.com/griffincapri...oOnMacOSX.html

[waffffffle]

If would be awesome if Apple took the initiative to add mono to the OS and give it their official support. It's one big step towards future compatibility.

Right now mono works great for CLI apps. Ideally Apple will create some sort of compatibility API to provide a smooth transition from .net to a native OS X framework, or even better, a complete virtual machine to provide java-like support for .net apps. That would rock, and give us access to so many more new software titles.

[kennedy]

Originally posted by larkost:
Personally I would doubt that the .NET frameworks will ever come to any non-Microsoft operating system. That is what Microsoft views as being their next set of crown jewels, and having it run anywhere else would undermine their business plans (read: monopoly). They are perfectly willing to have C# runtimes and compilers anywhere else (they have an implementation you can download for MacOS X already), but not the frameworks.

Huh? The .NET Framework is the CLR plus the .NET Framework class libraries.
Mono is an implementation of CLR and the .NET Framework class libraries.
Thus, your own statement on Mono says the .NET Framework is already running on Mac OS X.

Mono on the other hand already runs on MacOS X. Take a look at the screenshot from their site. I would be very interested if Apple decided to add the ability to create Mono applications through XCode, but I don't think that it will be on the menu anytime soon.

Apple has been doing a lot of work expanding the tschnologys avalible to Java and Obj-C Cocoa programmers (bindings and key-value-codeing being the big ones), and adding a whole other language with it's own compiler just does not seem like the right step to take at this point.

But consider how many people are coding software on .NET right now!!

Consider if Apple created an implementation of the WinForms library that was built on Cocoa rather than on Mono's Gtk+ and GNOME libraries.

QUESTION for you Cocoa programmers: what is the degree of mismatch between WinForms and the Cocoa interfaces? How hard would it be to take Mono, but for Mac OS X re-implement the WinForms library using Cocoa (rather than use Mono's generic Unix implementation of WinForms)?

[kennedy]

Just to clarify:

Mono 1.0 is currently only a PARTIAL implementation of the .NET Framework.
Their plan for Mono 1.2 is to be a near-complete implementation of the .NET Framework.

In particular, WinForms is NOT implemented in Mono 1.0.
Hence, waffffle's comment on CLI programs.
They do plan to have WinForms implemented in Mono 1.2.
However, that implementation will be built on Gtk+ and GNOME libs;
great for general Unix, perhaps; not so sure for Mac OS X.

Without a good implementation of WinForms, the vast majority of .NET apps will not run on Mac OS X.

[DeathMan]

Apple should not port windows.forms to native os x. The last thing we need are more Limewires floating around out there. I'm all for writing cocoa wrappers for native .NET apps, where the programmer has to rething the UI with a mac user in mind. If the .NET programmer does not want to bother with a unique Mac UI, they shouldn't port the app.

[Angus_D]

Didn't I hear that Quark wants to write the next version of Xpress in .NET, with Mono/Cocoa# for the interface on OS X? Or was that just lies?

[kennedy]

Originally posted by DeathMan:
Apple should not port windows.forms to native os x. The last thing we need are more Limewires floating around out there. I'm all for writing cocoa wrappers for native .NET apps, where the programmer has to rething the UI with a mac user in mind. If the .NET programmer does not want to bother with a unique Mac UI, they shouldn't port the app.

I understand your concern... however, don't cut off your nose to spite your face!

As a Mac lover:

1) My Windows-targeted apps will be as Mac-like as I can make them in Windoze; thus, a port of my apps won't be so horrible... though admittedly not ideal.

2) My Windows-targeted apps will get almost no Mac usage as they are enterprise apps... yet, I'd still like to have them able to run there so that I can use my Mac for demo'ing AND so that I'm not contributing one more reason that companies should not buy Macs.

3) Some people are just fine with the Windoze interface... but like the Mac product offering better except for one overriding fact: there's some key software that they can't run on their Mac. If that stuff could run on the Mac, even with an ugly Windoze interface, then Apple would sell more Macs, and...

ONLY when Apple starts selling more Macs will more software companies be able to afford to do specialized ports leveraging the Mac interface properly... so, first we need to get the ugly apps running well... THEN we can sell more Macs... THEN we'll start to see more apps leverage the Mac... THEN the Mac becomes even more superior and visibly so to more people and we sell even more Macs... and so on.


I think the .NET Framework could be a huge win for Mac. If they can go through and figure out how to best convert each construct into Cocoa and make it Mac-like automatically, that would be HUGE. Then, they have the ability to write a "Mac .NET Coding Guidelines" that recommend certain ways of doing your .NET code such that it ports well, certain things to avoid (stuff that doesn't convert well), and perhaps some auxiliary constructs that are only used on the Mac but allow the Mac .NET implementation to result in more Mac-like stuff...

Many software companies could afford that investment for portability... you may not get "ideal" Mac apps... but you'd get a lot more "really good" Mac apps.

Finally, consider this: if Apple comes out with a great .NET implementation that is fast and does a really nice job of implementing WinForms using Cocoa, then the Unix shops may find the Mac OS X platform far more appealing than the Linux or older Unix platforms stuck with Mono mapping WinForms to Gtk+.


I think .NET represents a HUGE opportunity for Apple... I hope they capitalize on it.

[goMac]

Having done some .Net, its not bad, but I couldn't see doing full applications in it.

[wadesworld]

I think .NET represents a HUGE opportunity for Apple... I hope they capitalize on it.

I disagree. I think it represents a ton of work for very little return.

Companies that are completely unfamiliar with the Mac marketplace are still going to be loathe to enter the Mac market, even if they could compile their .NET apps for the Mac.

Why? Because it would still cost them money to market on the Mac and it would still cost them money to support the Mac. So, with a small marketshare, additional costs to support the platform and a .NET implmentation the produces an ugly interface, developers are going to be lining up? I don't think so.

This is the same reason Apple killed YellowBox - the expected returns just weren't there. Sure, it's nice, but it would have taken a lot of resources to maintain as Windows changed and Apple just couldn't see a huge influx of Windows developers changing over to the Mac platform (or any for that matter).

Wade

[kennedy]

Originally posted by wadesworld:
Why? Because it would still cost them money to market on the Mac and it would still cost them money to support the Mac. So, with a small marketshare, additional costs to support the platform and a .NET implmentation the produces an ugly interface, developers are going to be lining up? I don't think so.

Hmmm... hope you're wrong... because without .NET, those costs that you seem to find so high a barrier to entry are just a tiny fraction of the costs... meaning Mac will never be able to expand its market. That would be a shame... I'll always need a damned Windows machine on my desk next to my Mac.

As a business owner, I could afford the minimal costs of supporting the Mac platform... the marketing costs would be trivial (the Windows-side marketing will hit plenty of Mac users) and having my support people know how to operate a Mac won't be that big a deal... I'm sure I'll have at least one Mac nut in my support group.

But having to invest 50% more into development to support the Mac... not even a remote option for me, or the vast majority of the software developers.

[Chuckit]

Originally posted by kennedy:
Hmmm... hope you're wrong... because without .NET, those costs that you seem to find so high a barrier to entry are just a tiny fraction of the costs...

It doesn't matter. I doubt most Windows devs would be interested in supporting the Mac even if it were free. Any cost is too much.

In all likelihood, nobody's making your company write to .NET. You'd have Apple's support (as well as full support from the Linux community) if you were using, say, Java or Python. Quite frankly, I'm glad Apple isn't encouraging .NET. Every bit of support Microsoft doesn't get is one more smile to my day.

[TETENAL]

What's wrong with Java?

[bewebste]

I'm still waiting for Xcode to remember vertical splits when reopening editor windows in a saved project...

[theolein]

AFAIK Microsoft's CLR with the C# compiler for BSD already runs on OSX due to OSX' BSD system. I downloaded it more that two years ago. The rest of the frameworks including WinForms will probably ont make it to OSX for the simple reason that there isn't much of a market for it right now. The large majority of enterprise stuff is done in Java at the moment and Apple covers that for those who want to use OSX in the enterprise. Those who use Windows servers and clients are those who will use .Net applications. Those places that use Linux on the server side will probably not switch to .Net anytime soon for the plain reason that despite Miguel's optimism, mono might very well suffer under Microsoft's patents. Until it is sure that mono won't be take to court for infringing on Microsoft's patents I don't think it's going to be big news on other platforms.

[Millennium]

Originally posted by TETENAL:
What's wrong with Java?

Nothing, but try telling that to people wbo've bought into Microsoft's marketing.

[kennedy]

Originally posted by TETENAL:
What's wrong with Java?

Nothing's wrong with Java as a *language*. You can use Java/J# to program .NET if you wish. And C# is very comparable, itself, to Java the *language*.

But "Java" can also be used to refer to a larger framework including the JVM and all the Java libraries. It is the latter two elements that map to .NET's CLR and .NET Framework libraries.

As a software developer, given almost all of your target users have Windows machines at their disposal AND almost all are heavy MS Office users, you are best to target that platform. And the most productive environment for delivering solutions that leverage MS Office and work well in Windows is .NET, not Java. It can be done in Java, but you take a productivity hit.

Now, the Mac world can thumb their noses at those developers not willing to take that productivity hit to port to the Mac. Or the Mac world could make it easy for all those .NET apps to be ported, minimizing the productivity hit.

From my searching Monster.com, there is a TON of C# and .NET activity... based on the resumes, there's a staggering percentage of the code the last two years done in C# on .NET. I was not expecting that going in... I was looking for Java... but I kept finding C#... made me take a new look at the state of the world.

[Chuckit]

Originally posted by kennedy:
Now, the Mac world can thumb their noses at those developers not willing to take that productivity hit to port to the Mac. Or the Mac world could make it easy for all those .NET apps to be ported, minimizing the productivity hit.

I'll buy my software from companies that don't consider me a "productivity hit," thanks. I didn't become a Mac user because I wanted to be a second banana platform for devs that don't care about us.

[kennedy]

Originally posted by Chuckit:
I'll buy my software from companies that don't consider me a "productivity hit," thanks. I didn't become a Mac user because I wanted to be a second banana platform for devs that don't care about us.

Oh, please! Porting is ALWAYS a productivity hit. So, in other words, you'll only buy software from companies that ONLY target the Mac. So, no MS Office for you! No Palm Desktop for you! No QuickBooks for you! No Quicken for you!

Again, you're cutting off your nose to spite your face. Yes, I buy the Mac for the great apps targeted for it... iLife, iWork, Final Cut, Apple Mail, iChat, Sherlock, Safari, and the OS itself... but that doesn't mean I don't want Intuit to build QuickBooks for the Mac, even if it is an inferior version to the Windows QuickBooks (which it is). If you take away MS Office and QuickBooks and Palm Desktop from the Mac platform, it would become real hard for me to use the Mac as my only machine. I'd rather have an inferior version of those programs than none at all and have to deal with a damned Windoze machine for my normal work!

The easier Apple makes it to port Windoze apps to Mac OS X, then the more apps we'll have available to us, and the more of us can ditch our Windoze machines, and the more Macs that will sell, and the better our beloved Mac platform will become!!

[Chuckit]

Originally posted by kennedy:
Oh, please! Porting is ALWAYS a productivity hit. So, in other words, you'll only buy software from companies that ONLY target the Mac. So, no MS Office for you! No Palm Desktop for you! No QuickBooks for you! No Quicken for you!

Do you really think these companies all view it as a "productivity hit"? If they really thought porting to the Mac was counterproductive, they wouldn't have done it, would they? Clearly they feel making a Mac version is a productive use of their time.

Originally posted by kennedy:
The easier Apple makes it to port Windoze apps to Mac OS X, then the more apps we'll have available to us, and the more of us can ditch our Windoze machines, and the more Macs that will sell, and the better our beloved Mac platform will become!!

Better by having a bunch of half-assed Windows ports? I don't think so. As somebody else said earlier, I don't want the Son of Limewire.

[itai195]

Originally posted by kennedy:
As a software developer, given almost all of your target users have Windows machines at their disposal AND almost all are heavy MS Office users, you are best to target that platform. And the most productive environment for delivering solutions that leverage MS Office and work well in Windows is .NET, not Java. It can be done in Java, but you take a productivity hit.

A productivity hit? I disagree. No clear leader for enterprise apps has emerged between .NET and Java. A lot of companies use both. .NET is better at some things; Java, others.

The main problem with the dream of .NET apps on the Mac is, however, the fact that Mac users will never accept them. The differences in application design between the two platforms are not skin deep -- Mac users expect their applications to look and function a certain way. Making .NET successful on the Mac will require more than merely porting the runtime. (Edit: as I see others have pointed out above)

[leperkuhn]

Originally posted by Chuckit:
Do you really think these companies all view it as a "productivity hit"? If they really thought porting to the Mac was counterproductive, they wouldn't have done it, would they? Clearly they feel making a Mac version is a productive use of their time.


Better by having a bunch of half-assed Windows ports? I don't think so. As somebody else said earlier, I don't want the Son of Limewire.

Unfortunately the rest of the world isn't as elitist. The ability to run software on the Mac that was written for Windows would be great for people that are switching, and for businesses that are intersested in getting macs but don't want to find Mac versions of the software.

It's a catch 22- people don't develop for the mac because it's a niche, and people don't buy macs because people don't develop for it. And yes, I know there's a lot of software. But some people are stuck in their ways and want to use what they've been using all their lives.

[kennedy]

Originally posted by Chuckit:
Do you really think these companies all view it as a "productivity hit"? If they really thought porting to the Mac was counterproductive, they wouldn't have done it, would they? Clearly they feel making a Mac version is a productive use of their time.

If you're not spending resources porting or "increasing portability", then those resources could be spent adding functionality, reliability, or polish. So, as long as the number of Windows users dwarfs the number of Mac users, then the effort invested in the Mac platform by the mainstream products will be dwarfed by the effort put into the Windows versions. Sooo, if you want better Mac versions, you need to make it as easy as possible to port to the Mac, so that they can use that time leveraging the Mac platform rather than just "making it work".

If there was a good implementation of the .NET framework on the Mac, mapping the WinForms to the Cocoa equivalents, such that the porting effort was trivial... then the limited investment they can justify on the Mac could be entirely spent on leveraging the things the Mac does better than Windows... such that the Mac version might even be superior to the Windows version.

For example, Apple could offer some improved derived versions of the WinForms classes that leverage Cocoa features. Then, on the Windows-side, those classes could just provide the behavior from the base class provided by MS. Then that bit of Mac investment by software developers could be used to introduce those CocoaForms into their apps... CocoaForms that won't affect the behavior on Windoze, but will dramatically improve the Mac behavior. Just one example.

[Chuckit]

Originally posted by leperkuhn:
Unfortunately the rest of the world isn't as elitist.

Fortunately, the rest of the world can keep using cheapo products with bad UI while the elitists like me get the good stuff. That works out fine in my eyes.

Originally posted by kennedy:
If there was a good implementation of the .NET framework on the Mac, mapping the WinForms to the Cocoa equivalents, such that the porting effort was trivial... then the limited investment they can justify on the Mac could be entirely spent on leveraging the things the Mac does better than Windows... such that the Mac version might even be superior to the Windows version.

Such a thing isn't unheard of even without a .NET port � MS Office for the Mac is widely considered to be leaps and bounds better than Windows Office. As I see it, the situation would be mostly unchanged: Developers who do not see it worth their time to focus on the Mac now would not give it much thought then, and the only conceivable upshot (if any) would be a flood of horrible crap that makes AppleWorks look like an awesome port.

Originally posted by kennedy:
For example, Apple could offer some improved derived versions of the WinForms classes that leverage Cocoa features. Then, on the Windows-side, those classes could just provide the behavior from the base class provided by MS. Then that bit of Mac investment by software developers could be used to introduce those CocoaForms into their apps... CocoaForms that won't affect the behavior on Windoze, but will dramatically improve the Mac behavior.

Cocoa is not magic. Layering WinForms on top of it won't somehow make a UI good if it wasn't good before. Basically, this would give you automatic spell-checking for WinForms elements � big whoop. I really can't see how this would make a big difference. It's not as though I've been thinking, "Gee, Windows is generally really great except it doesn't have spell-checking or emacs shortcuts in text views."

[Thinine]

There will be no major apps written in .NET until MS can figure out how to prevent .NET executable from being decompiled.

[Millennium]

Originally posted by Thinine:
There will be no major apps written in .NET until MS can figure out how to prevent .NET executable from being decompiled.

If that's the case, then there will never be any major apps written in .NET. Java apps, too, have this issue. Even native code can be decompiled. This has not stopped major apps from being written in the past.

[leperkuhn]

Originally posted by Chuckit:
Fortunately, the rest of the world can keep using cheapo products with bad UI while the elitists like me get the good stuff. That works out fine in my eyes.

as long as it's worthwhile for developers to keep writing for the mac it does. but I'm not sure if you've noticed, but there's a lot of stuff that will never come to the mac.

thinking long term, why would anyone invest hundreds of hours in a software project to sell to a niche market when they can make 20x the profit selling for windows?

I am continually amazed at how people can't understand such a simple concept.

[Chuckit]

Originally posted by leperkuhn:
as long as it's worthwhile for developers to keep writing for the mac it does. but I'm not sure if you've noticed, but there's a lot of stuff that will never come to the mac.

Yes, I have noticed that. Have you noticed me caring?

Quite frankly, most of the really important software has come to the Mac because the smart companies understand that it isn't a fruitless effort. If some little Windows company that makes .NET enterprise solutions doesn't port its app, I'm not going to cry about it.

Originally posted by leperkuhn:
thinking long term, why would anyone invest hundreds of hours in a software project to sell to a niche market when they can make 20x the profit selling for windows?

I don't know. Ask Microsoft or Adobe, or any of the many Mac-only software companies. Clearly the answer isn't "There's no reason," as you seem to be suggesting.

One possible reason: Just because Windows has 20 times the users does not mean you can make 20 times the profit. In fact, many of those users don't buy jack. I'm sure it depends on the situation, but I've heard of cases where the Mac is actually a more profitable platform.

Originally posted by leperkuhn:
I am continually amazed at how people can't understand such a simple concept.

Uh, yeah, that's it. I'm just stupid and your superior intelligence has baffled me into making meaningless noises that just happen to sound like responses.

[gatekeeper]

Originally posted by theolein:
Those places that use Linux on the server side will probably not switch to .Net anytime soon for the plain reason that despite Miguel's optimism, mono might very well suffer under Microsoft's patents. Until it is sure that mono won't be take to court for infringing on Microsoft's patents I don't think it's going to be big news on other platforms.

Software patents is indeed a big problem, but despite what some people would like others to think, it's not a problem specific to Mono.

Sun has already lost one Java patent lawsuit. Are they going to pay up if they lose another one so that Java users don't have to?

Linux potentially infringes on 283 patents, at least one of them owned by Microsoft. Should people stop using Linux on the server side?

Microsoft has patents on the CIFS/SMB protocol. Samba developers claim that the patents don't cover anything in Samba, but even if they're right that doesn't mean that Microsoft couldn't sue.

And from the schadenfreude department: Microsoft will probably soon have to pay up if they want to continue distributing their own codec

[leperkuhn]

Originally posted by Chuckit:

I don't know. Ask Microsoft or Adobe, or any of the many Mac-only software companies. Clearly the answer isn't "There's no reason," as you seem to be suggesting.

Let's ask Microsoft.

Products for the mac: Office. Virtual PC. MSN Messenger. that's about it.
Products for windows.. SQL Server, Exchange Server, and at least 50 more.

How about Adobe?
Funny, I don't see premeir.. audition, encore, actually, when you look at their lineup.. most of their stuff isn't mac compatible.

I'm not saying YOU need these apps. But some people do, and the fact that they aren't on the mac is a reason not to buy one. And the fact that people aren't buying macs for this very reason enough to not develop for it.

However, if a company could sell the app to mac users, how do we not benefit? What do you care if you have that ability

Uh, yeah, that's it. I'm just stupid and your superior intelligence has baffled me into making meaningless noises that just happen to sound like responses.

Sarcasm is anger's ugly cousin.

[theolein]

This whole thread is really just pissing in the wind, though. The chances of .Net coming to the Mac anytime sson are minimal. Microsoft is definitely not going to write that port and since Apple has put its eggs in the Java and Cocoa basket, I don't see Apple doing that either.

Not only this but .Net is and will remain mostly stuck in the enterprise for the near future. (Notice I said mostly). Recently, on slashdot, there was a story about some free paint programme done in C# on .Net. The people that tried it out pointed out that it was so slow, even on high end hardware (3GHz), that it was for all intents and purposes unusable.

What that means is that .Net, like Java (and Java runs very well on x86 these days, ever used NetBeans?) is fine for applications that use standard forms, like most business apps do, but is singularly useless, at the moment at least, for stuff like games or video and audio production.

I always wondered why Microsoft forbids anyone to publish benchmark of .Net in the EULA, and after seeing that paint programme, I know why.

[Chuckit]

EDIT: Never mind. Like Theolein said, it doesn't matter. Microsoft's frameworks ain't becoming part of the Mac OS anytime in the near future no matter what any of us say.

[gatekeeper]

Originally posted by theolein:
The chances of .Net coming to the Mac anytime sson are minimal.

Parts of .NET are already here for MacOS X. Windows.Forms is on its way.

Mono is already useful for some MacOS X users. A .NET app is the only way to stream audio to an AirPort Express without using iTunes.

Recently, on slashdot, there was a story about some free paint programme done in C# on .Net. The people that tried it out pointed out that it was so slow, even on high end hardware (3GHz), that it was for all intents and purposes unusable.

Did someone analyze it and determine that the primary reason it was slow was that it was written in C#, and not e.g. that it used GDI+?

If I come across a Cocoa app that is slow, would it be OK for me to just assume that it's a result of using Obj-C and Cocoa?

Anyway, I just tested Paint.NET on a P4 2.6GHz and it's only a tad slower than the regular mspaint. But then again, it's all relative. I'm sure there are Slashdot posters who would claim that mspaint or even notepad is slow.

useless, at the moment at least, for stuff like games or video and audio production.

Yes, but then again, has anyone claimed that .NET (or Java) is a suitable platform for those industries?

I always wondered why Microsoft forbids anyone to publish benchmark of .Net in the EULA, and after seeing that paint programme, I know why.

There's a standalone installer for .NET which doesn't have a EULA at all.

Or you could just get it trough Windows Update with the EULA and ignore that clause.

[Brass]

Originally posted by leperkuhn:
thinking long term, why would anyone invest hundreds of hours in a software project to sell to a niche market when they can make 20x the profit selling for windows?

The answer to that is simple, as I discovered by accident when I started developing freeware stuff for Mac OS X... and having and unexpected income that I didn't plan for due to the generosity of many of the users...

Because there is less competition for developers on the Mac. Simple. There may be 20 times the number of Windows users, but there's also 20 times the number of developers. Therefore profitibility works out to be much the same, in some cases.

[leperkuhn]

Originally posted by Brass:
The answer to that is simple, as I discovered by accident when I started developing freeware stuff for Mac OS X... and having and unexpected income that I didn't plan for due to the generosity of many of the users...

Because there is less competition for developers on the Mac. Simple. There may be 20 times the number of Windows users, but there's also 20 times the number of developers. Therefore profitibility works out to be much the same, in some cases.

That is a good answer.

[leperkuhn]

Originally posted by Brass:
The answer to that is simple, as I discovered by accident when I started developing freeware stuff for Mac OS X... and having and unexpected income that I didn't plan for due to the generosity of many of the users...

Because there is less competition for developers on the Mac. Simple. There may be 20 times the number of Windows users, but there's also 20 times the number of developers. Therefore profitibility works out to be much the same, in some cases.

not to beat a dead horse but I was thinking about this conversation today...

the real issue is when there's only 1 app out there for windows, and not 20. Or, something is client-server dependent (like exchange). I used to work at a small theater in VT as a sys admin, and it was windows only. They used a program called Tessitura. It was based on a MSSQL server, with the front end written with *shudder* Infomaker.

Needless to say, there was no mac version. It forced every non-profit that used this particular piece of software to use windows xp.

It also didn't help that they also used exchange for everything. (in the case of exchange, you can use entourage, I know). I'm not sure what the story is for volume licensing of Office 2004 Mac, anyone know? this particular non-profit was part of the "pay a bunch of money get every update" program... I forgot what it was called.

[theolein]

Originally posted by gatekeeper:
Parts of .NET are already here for MacOS X. Windows.Forms is on its way.

The OSX port of Windows.Forms uses GTK. Think about that for a moment with respect to OSX native GUI toolkit Cocoa/Aqua. If someone writes a Cocoa Framework that binds the Forms and and perhaps an ObjC .Net compiler (which doesn't exist even on the Windows implementation, even though there are compilers for exotic languages such as OCaml AFAIK). I have nothing against GTK but most Mac users will appreciate applications that are more integrated visually into the OS, not to mention things like drag and drop and copy/paste etc.

Mono is already useful for some MacOS X users. A .NET app is the only way to stream audio to an AirPort Express without using iTunes.

That's not a very good example. The fact that it was written using .Net doesn't mean it can't be done with Cocoa or Java or even plain C for that matter.

Did someone analyze it and determine that the primary reason it was slow was that it was written in C#, and not e.g. that it used GDI+?

I don't know, but since other paint applications use GDI+, I don't think that's the case.

If I come across a Cocoa app that is slow, would it be OK for me to just assume that it's a result of using Obj-C and Cocoa?

You're right about that one, but read the next point.

Anyway, I just tested Paint.NET on a P4 2.6GHz and it's only a tad slower than the regular mspaint. But then again, it's all relative. I'm sure there are Slashdot posters who would claim that mspaint or even notepad is slow.

Try using the app with multiple layers and large images and watch its performance then.

Yes, but then again, has anyone claimed that .NET (or Java) is a suitable platform for those industries?

No, and that's the point I made. Those industries are important in the consumer world.

There's a standalone installer for .NET which doesn't have a EULA at all.

Or you could just get it trough Windows Update with the EULA and ignore that clause.

Just because you don't read the EULA, which is included, even if it's not click through, doesn't mean it doesn't apply.

[gatekeeper]

Originally posted by theolein:
The OSX port of Windows.Forms uses GTK.

No, it doesn't. It uses Carbon.

And finally over the winter break Geoff Norton added a native MacOS X driver.

That's not a very good example. The fact that it was written using .Net doesn't mean it can't be done with Cocoa or Java or even plain C for that matter.

I doubt there is anything that one can implement using C# that can't be implemented using Java/C or vica versa. My example is neither good nor bad. It simply shows that there is one task that currently requires Mono to be performed under MacOS X.

Just because you don't read the EULA, which is included, even if it's not click through

The standalone installer doesn't have a EULA at all. No EULA presented during install. No EULA in the menu after install.

doesn't mean it doesn't apply.

Actually, that's exactly what it means. A contract that the user has never seen, let alone agreed to, does not apply.

If the user had to agree to a EULA before being able to download a installer which doesn't include a EULA, that would obviously be different. But that's not the case here.

[waffffffle]

I think full .Net support (via a Java-like virtual machine, or even better a pure Cocoa sort of integration) would be a double-edged sword for Apple. The upside is the increase of applications available to Mac users, the down side is the poor quality of those applications and the possibility that many vendors will drop Mac ports in favor of .Net apps.

However, there is a real advantage for Apple here in the corporate world. I'm the IT manager for a small company. I would love to move us away from Windows for many machines since most applications we have use mostly MS Office. But we're tied to Windows because of our internal application, which basically runs our business. The program is a Visual Basic app, and is tied to Access and MS SQL Server (this was all done long before I got to this company, I would have never let them do this today - and our enw developer creates web-based systems with PHP and MySQL). I have been working with our developer to try and move us away from the Access and Office dependencies, but Visual Basic programs can only run on Windows.

Many internal applications are built this way, and .Net seems to be most popular for internal apps at businesses. If companies could be sure that their .Net apps ran on any platform then it would help Apple in entering the workplace. I really want to bring more Macs into our company. I already have management convinced on the Mac mini (we bought one for a creative designer recently) but since it can't run our internal software we're sort of stuck. If I am not mistaken, .Net is not a difficult transition from Visual Basic studio, so it seems like its possible to move our app to .Net and then eventually cross platform, thanks to mono.

[TETENAL]

Originally posted by waffffffle:
Many internal applications are built this way, and .Net seems to be most popular for internal apps at businesses. If companies could be sure that their .Net apps ran on any platform then it would help Apple in entering the workplace.

What's wrong with Java?

[kennedy]

Originally posted by TETENAL:
What's wrong with Java?

Java is not appealing to the HUGE army of VB programmers.
Java, the framework, does not help you write programs on top of MS Office.

Furthermore, the next version of Visual Studio, due out later this year, will sooo eclipse Java IDEs that .NET productivity will be double Java Framework productivity. (That's a conservative estimate.) And Visual Studio supports VB, C#, C++, and J# (aka Java the language)... and development on top of MS Office. Plus there are a dozen other languages for which compilers are being or have been built compiling to .NET/CLR.

For business apps, .NET will dominate... not Java. Thus, if Apple wants its G5's to sell into other than graphic designers and such niches, then it had better figure out how to get .NET running on it... and running well!

[TETENAL]

Originally posted by kennedy:
Java, the framework, does not help you write programs on top of MS Office.

All right, but if you program on top of MS Office you bound yourself to the Windows platform anyway. It doesn't matter whether Apple or a third party provides .NET for Mac because Office is Windows only.

[Mr Scruff]

Originally posted by kennedy:
Furthermore, the next version of Visual Studio, due out later this year, will sooo eclipse Java IDEs that .NET productivity will be double Java Framework productivity. (That's a conservative estimate.) And Visual Studio supports VB, C#, C++, and J# (aka Java the language)... and development on top of MS Office. Plus there are a dozen other languages for which compilers are being or have been built compiling to .NET/CLR.

For business apps, .NET will dominate... not Java. Thus, if Apple wants its G5's to sell into other than graphic designers and such niches, then it had better figure out how to get .NET running on it... and running well!

This is FUD. All that .NET will provide is a drop in replacement for VB for those companies that write their in-house stuff in VB. But companies like that wouldn't touch Apple with a 1000ft bargepole anyway.

Who says VS will 'eclipse' (pun intended?) Java IDEs? Surely that's entirely a matter of personal preference anyway. Can you list the features that the next version of VS will provide that will make .NET developers 'conservatively' double their productivity over Java developers using Eclipse 3.

Java isn't going to shrivel up and die just because MS provides an alternative to VB that doesn't suck. As linux gets more traction in business (as it inevitably will), Java developement will become more, not less relevant.

[Chuckit]

Originally posted by kennedy:
For business apps, .NET will dominate... not Java. Thus, if Apple wants its G5's to sell into other than graphic designers and such niches, then it had better figure out how to get .NET running on it... and running well!

What you mean is, if Apple wants G5s to sell into the much tinier niche that your company apparently fills. Seriously, OS X is viable for most purposes. Gushing over everything Microsoft does (or, in the case of the next version of VS, even things Microsoft hasn't yet done) isn't going to change that fact. Yes, .NET is pretty popular, but there are plenty of good developers who aren't surgically attached to the MS teat. Until Windows is universal, .NET will not be either.

[CharlesS]

Originally posted by TETENAL:
All right, but if you program on top of MS Office you bound yourself to the Windows platform anyway. It doesn't matter whether Apple or a third party provides .NET for Mac because Office is Windows only.

[TETENAL]

Originally posted by CharlesS:

He is using .NET to program on top of Windows Office. That locks him to Windows. Period. Microsoft has an Office-suite available for Mac, but that's a totally different set of applications. Mac Office can not be automated with .NET.

That is this guys actual problem. There is no equivalent to Windows Office on Mac. Windows Office is the reason why he uses .NET, so he asks for .NET on Mac, overlooking that this doesn't magically make Office available.

For other companies whose in-house applications are not tied to Office there is Java available if they care about platform independence.

[kennedy]

Originally posted by TETENAL:
He is using .NET to program on top of Windows Office. That locks him to Windows. Period. Microsoft has an Office-suite available for Mac, but that's a totally different set of applications. Mac Office can not be automated with .NET.

That is this guys actual problem. There is no equivalent to Windows Office on Mac. Windows Office is the reason why he uses .NET, so he asks for .NET on Mac, overlooking that this doesn't magically make Office available.

For other companies whose in-house applications are not tied to Office there is Java available if they care about platform independence.

Valid point. But the vast majority of the users don't need the advanced authoring tools (built on Office)... the rest just need more basic interaction (built only on .NET). The minority doing the authoring are already locked to Windows via their other CAD tools. Its the rest of the company that I don't want to lock into Windows.

And yes, I could program the authoring tools in .NET and the rest in Java... but considering that means a lot of code duplication (both need to visualize the same stuff) and dealing with the quirks of two wholly different environments and learning those two very different environments... I will take a huge productivity hit (where productivity is functionality delivered times the # of users that can use it).


Finally, note that Office for Windows is NOT yet built on .NET... nor will it be during this decade. Rather, they've moved to a new XML-based alternative to the binary file formats. Both Office for Windows and Office for Mac support the BINARY formats. The new XML based format is there for greater interoperability. Given that, I bet Office for Mac will be supporting the XML formats in the next release. That leaves just the XML functionality and exposure of the XML event model to other programs... if Apple offers a good .NET implementation, MS will WANT it to work on MS Office. And they won't bother rewriting MS Office for Mac in .NET!!! After all, Office for Windows is not written in .NET. In fact, given how Office for Mac has often led Office for Windows in functionality, I'll bet the .NET interoperability for Mac would be very well done.

Maybe Mono will get us there... but Apple could do it so much better & faster if they chose to.

[waffffffle]

I don't understand this "what about java:" nonsense. Microsoft created .Net as a Java killer and in many respects it is succeeding. java of course won't go away completely but on the Windows platform .Net and C# are the new buzzwords, just like Java was years ago. If a company could add Macs without even needing to recompile their in house applications I think we'd see a lot more companies looking at Macs.

I know I'd certainly have an easier time convincing management to buy Macs. With the Mac mini, Macs really aren't more expensive today. We can buy PCs for $300 without Windows, add an OEM copy of Windows XP Pro for about $100and yearly antivirus for $50 a year, and that PC costs the same price as a Mac after two years. Let's not get into the cost of cleaning up the spyware every few months. Although to be fair, the Mac will require about $40 more for a USB mouse and keyboard (I'm partial to the Apple keyboards for better integration into the Mac OS). Still, Macs aren't expensive anymore for the corporate world. I know I'm not the only IT manager taking a long serious look at Apple here. The compatibility is almost there.

Also, Apple may actually be able to work with the Microsoft MacBU to give compatibility to the VB for Applications which comes with Office for Mac and Apple's mono/.Net implementation. However that may be wishful thinking.

[CharlesS]

Originally posted by TETENAL:
He is using .NET to program on top of Windows Office. That locks him to Windows. Period. Microsoft has an Office-suite available for Mac, but that's a totally different set of applications. Mac Office can not be automated with .NET.

That is this guys actual problem. There is no equivalent to Windows Office on Mac. Windows Office is the reason why he uses .NET, so he asks for .NET on Mac, overlooking that this doesn't magically make Office available.

For other companies whose in-house applications are not tied to Office there is Java available if they care about platform independence.

Yes, but that's quite a bit different from saying that Office is Windows-only.