Windows for ARM ... PB for ARM ?

[smishra]

I guess the issue can be divided into two parts

- porting the language over
if the language translates to something close to ANSI C, or another portable intermediate language, porting it is not as big of an issue
if it translates to assembler, the effort is harder

- porting the libraries and commands over
this is really the fun part
maybe all commands need not be ported over

We do not need to port the development environment over, since most phone development is done as a cross build anyway.

[freak]

I guess the issue can be divided into two parts

- porting the language over
if the language translates to something close to ANSI C, or another portable intermediate language, porting it is not as big of an issue
if it translates to assembler, the effort is harder

- porting the libraries and commands over
this is really the fun part
maybe all commands need not be ported over

We do not need to port the development environment over, since most phone development is done as a cross build anyway.

You are forgetting the largest part:
- maintaining the separate port in all future updates

This is by far the most work intensive part of it all. We already have 6 different OS/CPU configurations to maintain, so i know what i am talking about. I don't think we have the resources to maintain another one.

[IdeasVacuum]

...ah, no problem, just drop support for the Mac

[skywalk]

Well, this industry is too fluid for permanent declarations.
Amiga started PureBasic and it is no longer supported.

Computing is now like a stick of dynamite with a fuse lit at both ends.
The resulting explosion forming an ubiquitous, yet blurred product that is both mobile and tethered.
When my cell phone has the same Mips or screen projection as my laptop, then it would be wise to have applications available.

[USCode]

...ah, no problem, just drop support for the Mac

Don't you mean Linux?

Operating System / Total Market Share
Windows 89.70%
Mac 5.25%
iOS 2.05%
Linux 0.95%
Java ME 0.81%
Android 0.49%
Symbian 0.28%
BlackBerry 0.12%
Windows Mobile 0.03%
Playstation 0.03%
FreeBSD 0.01%
BREW 0.01%
...

http://marketshare.hitslink.com/operati ... px?qprid=8

[smishra]

Freak, I agree, the maintenance issue is huge.

On the other hand, there really is nothing simple and clean to use for phones. There are all kinds of custom development environments, but nothing to get things done in a casual manner which still generates small, efficient code. If I am wrong, I will be happy to learn otherwise.

The reason PB appeals to me is that I can write casual programs quickly. I don't have to deal with semi religious mumbo jumbo about object oriented this and overloading that. The libraries are great and the command set is intuitive. The compiler is fast and the code generated is reasonably efficient and small.

PB is probably a labor of love. I would love to see additional versions, but understand that unfortunately there are only 24 hours in a day.

[rsts]

I'd like a PB like solution for mobile platforms too. I've considered java, but would very much prefer something PB like, if possible.

Supposedly, or maybe hopefully, all the additional platform support has yielded new customers and additional revenue. I would pay additional for a PB license for a "new" platform and I imagine others would also, not to mention the possibility of "new" customers.

Might there be a point when increased demand/sales justifies additional resources, so the entire burden does not fall on the existing staff?

Or are we just better off moving to a different platform for all other development?

Probably a question only fred can answer.

cheers

[smishra]

Let us start a bounty to see if we can convince the development team.. I would gladly pay $400 for a PB port to an Android smartphone.

If the development team does the initial language port, I might be able to add some library functions.

If PB port does not happen, I will eventually have to move to another platform for phones. Something which I am not looking forward to, because to me it means additional maintenance of keeping my knowledge of another platform current.

[blueznl]

I already own two PB licenses, but would gladly buy a third if it would include Android (tablet) support

[H.Brill]

look here :

http://www.basic4ppc.com/

[rsts]

look here :

http://www.basic4ppc.com/

have seen it - it looks interesting but I would prefer a PB offering it it was to be.

cheers

[smishra]

Thanks for the link. will check it out.

However I agree with blueznl, I would prefer PB

It is a question of having the similar platform, similar set of commands and the same level of comfort.

[Seymour Clufley]

Perhaps it is pointless commenting on this, because Fred will do whatever he thinks is best. And there's only so much you can do with a small team and limited resources. Perhaps it's better for PB to be a great language for the major desktop platforms, than a crappy language for many platforms.

But anyway, here are some thoughts...

It seems to me that programming is going through a huge change just now. In the past it was a choice between Windows, Mac or Linux. And those platforms will be important for a long time yet. However, nowadays people expect things to be available no matter what platform they're on. This is because of websites, really. Imagine if you could only look at Facebook on a Mac, or MySpace on Linux. Also because of websites, people have got accustomed to using the latest version of software without having to update it manually. This is how websites behave, and it's how programs should behave.

This is where the modern web browser comes in. It makes the underlying OS irrelevant. The browser is becoming capable of 2D drawing, 3D graphics, sound synthesis, file management, multithreading, databases, drag&drop, etc. The web browser is becoming THE platform. And this is how things should be. Who wants to worry about Mac/Windows/Linux?! Who wants to worry about not having the latest version of a program?

Part of PureBasic's charm has been that, to a large extent, you don't need to worry about Mac/Windows/Linux because it compiles to all three. But the browser is now, effectively, becoming the 4th platform. It's a few years away yet, but it's coming. And if you want your program to get to consumers, the web browser will be the platform to choose - not Windows or Mac. Google is working on Native Client to allow binary "modules" (executables) to run inside Chrome, and other browsers will follow suit.

Separate from that is the mobile phone platform. Writing mobile phone apps is already a bankable skill, and it will only become more so as people migrate away from Mac/Windows/Linux.

People will expect programs to be delivered either as webpages (you use it in your browser), or as mobile apps (you use it on your phone).

So there are two new kinds of "program" emerging: the webpage and the mobile phone app.

You can't use PureBasic to make either of these. For now, that isn't a big problem. All of us on this forum use PB to make programs for "the old platforms", and we'll be doing it for many years to come.

But in five or ten years from now, "the old platforms" will be much less relevant than they are today.

I'd say that it would be wise for PB v5.0 to do these things:

• compile NativeClient modules
• compile mobile phone apps

In the first case, a lot of PB's own libraries would be redundant. We don't need 2D drawing, 3D, drag&drop, database etc... because the browser provides APIs for those things. PB would communicate with the APIs. Essentially PB would just be the top layer, with the "nuts and bolts" handled by the browser. We'd simply be programming webpages in PureBasic rather than JavaScript (I'd LOVE that!) but because the browser is a modern one, such as Chrome, we'd have access to 3D, 2D drawing, databases etc. just like we have now with PB's own libraries.

In the second case, compiling mobile phone apps would be different, because it wouldn't be for a single target platform (Native Client) but at least 3 different operating systems (iOS, Android, Symbian, etc.). I don't know what would be involved in this. Would PB's libraries need to be re-written? PB v5.0 could have basic support for this, with new features added as and when the libraries were updated, if at all.

So essentially there are four new operating systems for PureBasic to support: NativeClient, iOS, Android, Symbian.

LLVM may reduce the workload, but I don't understand how it all works. It does compile for NativeClient, though.

I hate saying all this, because it sounds like a hell of a lot of work. But I believe it is the future, and it'll be a great future. I really want PureBasic to be a part of it. I don't want it to disappear or become a language for "the old platforms". I want to be able to use it for the new era of mobile apps and web apps!

[USCode]

Given their limited resources, the PB guys have to wait for things to really settle out before they can commit to another platform.
They'll be able to drop OS X PPC support I imagine one of these days but that won't free up the tremendous amount of time required to support an entirely new OS.
However, using PB to develop Windows tablet apps on ARM seems like a logical potential next step ... but it's too early for PB to commit that ... yet ...

[skywalk]

Is the PureBasic licence valid for all OS ?
Yes. The PureBasic licence is an user-based licence, which means you can install your full version on all of your computers without the need to buy another licence for each. PureBasic currently runs on four different operating systems (Windows, Linux, MacOS X and AmigaOS) and a single licence cover them all. Even if PureBasic is ported to other OSes (OpenBSD, Mobiles..) you will have free access to them.

Even if...