Will monodevelop case impact to attracting new user?

[Troodon]

I think this is an extremely interesting option that we have, on all platforms.

Also, I assume you've seen this:

http://wiki.lazarus.freepascal.org/Using_Pascal_Libraries_with_.NET_and_Mono

Thanks.

-Phil

[Phil]

I think this is an extremely interesting option that we have, on all platforms.

Also, I assume you've seen this:

http://wiki.lazarus.freepascal.org/Using_Pascal_Libraries_with_.NET_and_Mono

One limitation, though, is that you do have to "flatten" your code to call the native dll/so/dylib from .NET/Mono. That is, you need to pass simple "C" types and can't pass objects the way you can, say, with a VB.NET assembly calling a Delphi Prism assembly (and vice versa). However, once you've done that you can use the native library from any language.

With Delphi, I was even able to use Delphi forms in my DLL. They look just slightly different from a WinForms form, but good enough for my use in a .NET app. Don't know if the issue of LCL forms in a native library were ever solved with Lazarus.

Thanks.

-Phil

[Phil]

IMO Oxygene (the Delphi Prism language) is only remotely related to Object Pascal, mostly because the .Net framework is a different paradigm when compared to the Windows API. More exactly: Oxygene is not Pascal. But the main hurdle to Delphi Prism adoption, just like for Delphi itself, is the lack of a free version and documentation. The Embarcadero folks argued that, while M$ can afford to make VS Express available for free, they can not. Try finding documentation on the Web for Delphi 2007 for .Net -- there isn't any. Delphi Prism will always be one step behind Visual Studio for C#.

Just so it's clear: Delphi.NET is unrelated to Delphi Prism.

There is a free Delphi Prism command line compiler - I've used it with both .NET on Windows and Mono on OS X.

https://downloads.embarcadero.com/free/delphi_prism

Thanks.

-Phil

[Troodon]

Just so it's clear: Delphi.NET is unrelated to Delphi Prism.

Thanks, I was aware of that

There is a free Delphi Prism command line compiler - I've used it with both .NET on Windows and Mono on OS X.

Yep, and I hope some bright people will integrate it into Eclipse or something similar.

[Phil]

For all the 3 phones using it =) Seriously, the largest seller of Windows Mobile phones, HTC, is moving to Android, just like all other companies that use Windows Mobile too. People aren't much interrested in rewriting their Windows Mobile code to .NET. Restricting development on the Phones to .NET will just remove Microsoft from this market. And they will get what they deserve for such an unethical move.

Microsoft doesn't have to dominate to be successful. I don't see anything unethical about requiring developers to use certain tools for a particular phone platform. Don't make the mistake of assuming this decision is about us developers. It's really about the users (buyers) of the phones and about how to provide them with the best experience, best apps, and quickest turnaround when new OS and device releases are made.

Thanks.

-Phil

[Troodon]

With Delphi, I was even able to use Delphi forms in my DLL. They look just slightly different from a WinForms form, but good enough for my use in a .NET app.

Thanks for the info.

[marcov]

For all the 3 phones using it =) Seriously, the largest seller of Windows Mobile phones, HTC, is moving to Android, just like all other companies that use Windows Mobile too. People aren't much interrested in rewriting their Windows Mobile code to .NET. Restricting development on the Phones to .NET will just remove Microsoft from this market. And they will get what they deserve for such an unethical move.

Microsoft doesn't have to dominate to be successful. I don't see anything unethical about requiring developers to use certain tools for a particular phone platform

True, but it doesn't make it quite attractive either. And not just for that 1% of developers that use something else. Usually such behaviour indicates a certain iron grip that the other 99% will face sooner or later too. E.g. because they don't care about backwards compatibility, because they can change their tarifs, because they can easily impose other requirements that wreck your business.

Don't make the mistake of assuming this decision is about us developers. It's really about the users (buyers) of the phones and about how to provide them with the best experience, best apps, and quickest turnaround when new OS and device releases are made.

A platform is an equilibrium of vendor, users and developers. Piss off one group too much (or, leave them without a profit/something usuable), and it will fail.
 
Personally I think Windows mobile 7 is an attempt to differentiate the Windows CE market into a cheap market where the Telco's have full control over what is run (and thus can improve yield/customer), and whatever real CE comes next that will remain reserved for the more highend and enterprise devices and other, non phone embedded devices.

[felipemdc]

Microsoft doesn't have to dominate to be successful.

Microsoft is too large to define it's success only by making a good % out of a small market share. And also from the developer point of view, if they have a very small market share, they aren't relevant, and there is no use rewriting your application to support a platform which is incompatible with everything else and holds a small market share.

So quite the contrary, they do need at least more then 10% of the market share to avoid being a failure considering the huge amount of resources they have.

I don't see anything unethical about requiring developers to use certain tools for a particular phone platform. Don't make the mistake of assuming this decision is about us developers.

Sure it is about making their .NET platform more valuable by making it the only way to develop Windows Phone apps. Nothing monopolistic about that =) Just that this time it's backfiring and we already see an exodus of manufacturers.

It's really about the users (buyers) of the phones and about how to provide them with the best experience, best apps, and quickest turnaround when new OS and device releases are made.

Offering a native API doesn't change anything for users in any of these terms. And quickest OS development is a terrible excuse, like if they don't have enough money to pay a couple dozen developers to work on the native API.  

[Phil]

A platform is an equilibrium of vendor, users and developers. Piss off one group too much (or, leave them without a profit/something usuable), and it will fail.

Not sure I would include us developers in this love triangle. With consumer phones (almost all, that is), the relationship is between vendor and user. We're in there somewhere but we're a tiny group, fragmented, and don't influence either group. With desktop computers the most important relationship has been between vendor and IT groups within organizations and companies (note that Steve Jobs has commented on this, referring to IT people as "confused"). It remains to be seen whether IT will have the same role for mobile devices, which in some cases are already locked down and auto-updating, meaning these guys don't have as much to do.

Personally I think Windows mobile 7 is an attempt to differentiate the Windows CE market into a cheap market where the Telco's have full control over what is run (and thus can improve yield/customer), and whatever real CE comes next that will remain reserved for the more highend and enterprise devices and other, non phone embedded devices.

I would expect them to kill off CE, if they haven't done so internally already after the Kin failure (http://daringfireball.net/linked/2010/07/07/kin).

We still don't really know how well Windows 7 and .NET scale down to mobile devices. OS X and Cocoa apparently scale down pretty well on iOS, as do Linux and Java on Android. I think it's funny to see how after years of steadily increasing processor clock speeds (which we hardly notice while plugged into A/C), that the mobile device is constrained by battery life (too fast and the thing will burn your hand). This goes a long way in explaining at least in part many of Apple's restrictions on developers (re multi-tasking, VMs, Xcode, etc.).

Thanks.

-Phil

[Troodon]

By analogy with another high tech field, drug development (in which I work), I think that we will see more attempts from large IT companies at controlling niche markets. The time of the blockbuster products may be gone. Battles will be waged over smaller fields, IT companies will address the needs of smaller groups of customers provided they have large pockets.

[JD]

But the main hurdle to Delphi Prism adoption, just like for Delphi itself, is the lack of a free version and documentation. The Embarcadero folks argued that, while M$ can afford to make VS Express available for free, they can not.

This is the main reason why new developers & beginning Computer Science courses systematically turn to Microsoft's Visual Basic/C++/C# Express offerings. I really regret the demise of the free Turbo Explorer offerings. Last thing I want is for Pascal to turn to a hobbyist language.

Unless compelled to do otherwise, I'm still more inclined to develop in Pascal (Delphi/Lazarus) than in C# because the tools are cheaper or free & deployment is easier!

In my opinion, Embarcadero dosen't seem to know what to do with Delphi Prism/Oxygene. Just visit their website & you'll see that the product seems to be an "extra" offering, a poor cousin of their mainline products & not a worthy strategy to give Pascal a voice in the .Net/Java world. It may end up going the way of Delphi.NET 

[Phil]

I don't see anything unethical about requiring developers to use certain tools for a particular phone platform. Don't make the mistake of assuming this decision is about us developers.

Sure it is about making their .NET platform more valuable by making it the only way to develop Windows Phone apps. Nothing monopolistic about that =) Just that this time it's backfiring and we already see an exodus of manufacturers.

.NET is what MS is betting on, just as Apple is betting on ObjC and LLVM. It would be very odd if MS did not try to push .NET on mobile devices. The exodus of manufacturers is unrelated to that.

It's really about the users (buyers) of the phones and about how to provide them with the best experience, best apps, and quickest turnaround when new OS and device releases are made.

Offering a native API doesn't change anything for users in any of these terms. And quickest OS development is a terrible excuse, like if they don't have enough money to pay a couple dozen developers to work on the native API.  

Reviewing the timeline for the release of iOS 4 is instructive. This is a major update and the 4th major release in as many years. That alone is just staggering compared to desktop OSs.

Since I'm an Apple Developer I have all the e-mail announcements from the last several months that went out to all of their developers.

April 9, 2010: First beta of iOS 4 (still called iPhone OS) and SDK available for download.
April 20: Beta 2
May 5: Beta 3
May 19: Beta 4
June 10: GM released during WWDC conference
June 24: iPhone 4 released
July 1: Apple reminds developers that new and updated iOS apps need to be compiled against SDK 4.
July 7: Almost all of the apps I have on my iPod touch have been updated to iOS 4.

Compared to desktop OS's, where tool developers typically have months and even years to update tools for a new release, this all happened in less than 3 months. How did Apple accomplish this? Well, for one thing, they control the tools and have invested heavily in open source projects like LLVM, WebKit, etc.

The result is that user expectations are met: many apps are immediately available in updated versions almost simultaneously with the new OS/device launch. Developer expectations are also met: Just recompile in many cases.

Where do 3rd party tools fit into this new world? Well, maybe they don't. Or at least not in the way that we traditionally think of them.

A similar timeline could probably be constructed for Android, which is also moving incredibly fast, maybe even faster than iOS. This is really unprecedented, I think. But I think it represents a tipping point of sorts, a culmination of all the long hard years of development of Unix, Linux, OS X, GCC, Java, Internet standards, etc. It appears as though finally the fruit can be harvested by the public.

Thanks.

-Phil

[felipemdc]

.NET is what MS is betting on, just as Apple is betting on ObjC and LLVM. It would be very odd if MS did not try to push .NET on mobile devices. The exodus of manufacturers is unrelated to that.

Both are unethical monopolists IMHO, and the fact that both are targeted by regulation agencies just proves that. And I don't think it unrelated. Indeed Windows Mobile was already falling, but now if you have to rewrite all your code to support it, it just doesn't make sense. It's a good opportunity to just migrate to something more stable, of which you have a higher degree of control, like Android.

Where do 3rd party tools fit into this new world? Well, maybe they don't. Or at least not in the way that we traditionally think of them.

There are solutions to support a 3rd party environment in very rapidly changing platforms:

1 - Run as a web page instead of a compiled program. HTML + Javascript works like that. But then Pascal gets limited to the server side.

2 - Emulation, which usually means supporting a bytecode. Just like Java and Flash works. In that sense I think that the future of FPC/Lazarus in such unstable targets would be generating Java Bytecode. I've already done some experiments with that, and I'm pretty sure 99% of our code will keep running. It's just a metter of doing a good mapping between the capabilities of the Java assembler and Pascal and you can emulate almost any Pascal code.

Further, there are international specifications for generic phone/PDA APIs (Such as accelerometer, SMS, Phone, etc) being developed, so you can just implement them for your language and you have cross-platform PDA support.

Obviously the venders can then push even harder to block cross-platform development, like Apple is doing, which goes back to my point of them being unethical monopolists.

[marcov]

A platform is an equilibrium of vendor, users and developers. Piss off one group too much (or, leave them without a profit/something usuable), and it will fail.

Not sure I would include us developers in this love triangle. With consumer phones (almost all, that is), the relationship is between vendor and user. We're in there somewhere but we're a tiny group, fragmented, and don't influence either group.

If so, why do they bother to have free development tools, api's etc ? They need us, so we are part of the triangle.
Sure, they rather have us mute, and believe we can change anything.

I'm not some visionary that now comes with great freetard theory, and that steers the discussion in the direction of Linux.

But the reality IS simply that developers vote with their feet if they don't like the target. They might be meek sheep in the beginning when margins are high (like e.g. with IPhone), and it can take a while. But when a significant other contender appears, it usually comoditizes the platform, and has an edge.

With desktop computers the most important relationship has been between vendor and IT groups within organizations and companies (note that Steve Jobs has commented on this, referring to IT people as "confused"). It remains to be seen whether IT will have the same role for mobile devices, which in some cases are already locked down and auto-updating, meaning these guys don't have as much to do.

I usually think Jobs is full of it.

Personally I think Windows mobile 7 is an attempt to differentiate the Windows CE market into a cheap market where the Telco's have full control over what is run (and thus can improve yield/customer), and whatever real CE comes next that will remain reserved for the more highend and enterprise devices and other, non phone embedded devices.

I would expect them to kill off CE, if they haven't done so internally already after the Kin failure (http://daringfireball.net/linked/2010/07/07/kin).

I think they remove it from the end-user market, except maybe the high end.

We still don't really know how well Windows 7 and .NET scale down to mobile devices.

What does Windows 7 have to do with it?   CF.NET scales down fine, it has done this longer than Apple has with COCOA.

OS X and Cocoa apparently scale down pretty well on iOS, as do Linux and Java on Android. I think it's funny to see how after years of steadily increasing processor clock speeds (which we hardly notice while plugged into A/C), that the mobile device is constrained by battery life (too fast and the thing will burn your hand). This
goes a long way in explaining at least in part many of Apple's restrictions on developers (re multi-tasking, VMs, Xcode, etc.).

I don't think there really are technical reasons behind that.

[Phil]

I'm not some visionary that now comes with great freetard theory, and that steers the discussion in the direction of Linux.

But the reality IS simply that developers vote with their feet if they don't like the target. They might be meek sheep in the beginning when margins are high (like e.g. with IPhone), and it can take a while. But when a significant other contender appears, it usually comoditizes the platform, and has an edge.

Is that even true? How could we test that assertion? Well, what about this:

2001: Windows 92% of desktop market, Mac 4%, Linux 1%
2009: Windows 91%

http://www.windowsitpro.com/article/desktop-management/study-shows-that-windows-owns-the-desktop-market.aspx

http://en.wikipedia.org/wiki/Microsoft_Windows

That's why the new markets that Apple has created with the iOS devices is interesting. Suddenly the mix is not static (ie, stagnant) as it has been for the last decade on the desktop. Historically very few people have had any direct experience with OS X or Linux. Now suddenly a great many people appear to be enthusiastically embracing a non-Windows OS on a computer. It's not a phone, it's a tiny computer connected to a tiny touchscreen and to a mobile phone radio modem.

Another measure of sorts:

http://www.prnewswire.com/news-releases/freelancercom-announces-the-fastest-growing-online-jobs-for-q2-with-the-release-of-the-freelancer-fast-50-97839529.html

This is a pretty oddball list. I'm not sure where this stuff comes from or how the categories are determined, but one thing it does do is scream, "it's not the desktop anymore". It confirms what I've thought for a while, that desktop software development is on the verge of irrelevancy. Well, let me restate that: Desktop development, even new projects,will continue; most desktop software development will continue to target Windows; much of it will continue to be Windows-only. However, Web development and mobile development has sucked most of the oxygen out of the desktop air.

Thanks.

-Phil