Good idea!

Since I'm trying to code with C++ for AOS I've to hassle around with creating own classes for reusable common stuff like GUI elements, message handling and so on!

Hopefully your framework will cope with things like this too!

This is quite a development and one that will unfortunately be overshadowed by the news of the new hardware.

How much of QT framework are you intending to support?

@ne_one
Sounds like NONE of the QT framework will be supported - he only says "inspired by" it.

_________________
Author of the PortablE programming language.
It is pitch black. You are likely to be eaten by a grue...

@jeanluc72

Maybe you could write a little more about what your framework will eventually provide. Right now I see some very low-level fundamental classes, some of which BOOST and STL provide similar functionality for already. There's nothing Amiga specific.

So, what are your goals? What else is it going to do once you get beyond the low-level stuff?

Hans

_________________
Join the Kea Campus - upgrade your skills; support my work; enjoy the Amiga corner.
https://keasigmadelta.com/ - see more of my work

@chrisH

Quote:

Sounds like NONE of the QT framework will be supported - he only says "inspired by" it.

Hmm... well if that's the case it's sort of like 'brought to you by the letter Q.'

A little more information would be helpful.

This is not Qt, and it appears to be so early in development that it could have been "inspired" by many things.

Qt also has its own implementations of lists (QList) and whatnot. I don't use them unless there's a good reason to.

I'd like to have an actual port of Qt. I use Qt at work and know it quite well.

Anyway, good luck with this project.

I think, if something comes out of this, it might be really really giving development a big push.

Having to go through all that hassle with Intuition stuff, when all you want is to simply open a window or display a button or a menu is quite unnerving these days.

So, I for one, am really excited about this project. Seems to be early on, but let's hope the best :)

@all

It seems a lot of people are reading this news, so I will try to give you all a bit more information.

1. No, it will not be a QT port.
We have been thinking about this option, and discussed it a lot. I personally think porting is in many cases a usefull method of getting software from one platform to another one. But it also includes the risk not using the full potential of the platform you are porting towards. There are several reasons why porting QT to the Amiga would need so much changes in the internals of the framework, it would need so many changes that you are not porting QT, but mainly rewriting it. So we dropped the idea of this, and started from "scratch". For us, Qt is only a class template to get some orientation (as mentioned, it was and is our inspiration).

2. Yes, we knew this news would be overshadowed by the announcement of the new hardware, but we all will get back to "usual business" in a couple of days/weeks We did not plan to release this news this early, because we have some big surprises for you all, but they still are in real "cooking" stage, so we keep them down until now, because it is all in a very early stage. On the other hand, we could not resist on giving NEAT a birthday to remember, so, we decided to push our announcement a bit forward. I mean, we would not get another chance to announce on the day of birth of a new Amiga .

3. The framework will provide as much as possible. The only barrier is fantasy and time. We are going to implement easy handling of tasks, processes, inter-process-communications, different socket handling, file handling, memory handling... on the low level side, but we are already planning on how to implement the GUI and multimedia things of the Amiga. I have to be honest we do not know yet on what this GUI implementation is going to be based. A long time we were not active in the Amiga scene (came back in December 2008), and are a bit confused of all the "third party" GUI toolsets which are available. At the moment, only intuition and Gadgets included in the SDK seems to be intresting, because we do not want to be depending on other suppliers. We did a try on Cairo which worked extremely good, but as said, no decissions made yet. I would like comments from all of you on this topic, we need to know what you guys are using now, and plannung to use in the future!

4. We are very excited about the XMos implementation into the new Amiga, and hope we will be able to put some work in Neat also to be able to use this feature. As we do not know any specs yet, I have no idea how this is going to look like.

5. We also think this will give a real push to development, because starting to use QT pushed us a long way on Linux back then. Without QT, doing my projects would be much slower and more complicated as I do them now. For instance, the QtComMon was written in 1 day (yes, you read it well, ONE day!). If I would had to build this on framebuffer technology (only for the displaying issues), you would not be using QtComMon from the distcode.lha package right now, you would probabely be waiting for another 2 months. I think that says enough.

We are really proud of what we are doing, and we hope we get a lot of feedback from our users. Everything you bring to us is important, even negative criticism, because we do not want to sit there in our Ivory tower, but want to work on the base, the things you all need to get work done better and faster. Help us to get it done right!

Thank you all,

Jeanluc.

Since you are using namespaces, why do you prefix each class with N?

Good luck!

I really appreciate your efforts this is someting that is missing in the Amiga land. I'm a C++ developer myself and it's not easy to get into Amiga development when you are "brought up" in a Windows environment and lack the time in the evenings. A framework like Qt for GUI is more than welcom!

But i think that a port of Qt would be most beneficial because developers in the Amiga comunnity would have the power to reach out to other platforms while remaining at "home" :) and have an increased user base and income (if commercial). but if it's a bigger task to adapt Qt to Amiga OS than to write one from scratch than so be it, pity though...

_________________
_________________________________________________
--Regards Muffin
--Amiga a Brave New World!

@jeanluc72

I think if you guys can produce a comprehensive and mature toolkit then this may well be very helpful to many AOS developers.

What would be really nice is an easier method for designing GUIs. I have been having a go at a spot of programming but have hit a sort of brick wall when it comes to Intuition programming... it's just far too easy to lock up the whole system when experimenting and trying to get a feel for it. Plus documentation for Intuition is fragmented at best with no top down conceptual information.

_________________
A600 OS3.1 ACA620 | '030 A1200 OS3.5 | µA1-C 750GX OS4.1 | SAM460 OS4.1 | '040 A3000 OS2.1 | Christian Aid |

I have to say, I like the way Intuition handles messages, very OOP. But the Gadget system (creating Buttons, sliders, etc) is really horrendous.

Well, at least for the standards of 2010. It's more fun to do these things on your own, but who wants to reinvent the wheel everytime?

Guess, that's where NEAT might come in handy..

@Muffin

Quote:

a port of Qt would be most beneficial because developers in the Amiga comunnity would have the power to reach out to other platforms while remaining at "home" :)

Agreed.

Not to descredit the project, but there are a lot of efforts to come up with better alternatives to things that already exist.

What the Amiga is sorely lacking is tools that make it easier to develop and port software from other platforms.

I think they better go this way instead, when you think about it, I believe QT is appealing enough someone else will pick it up sooner or later (and this will end up being something more for Amiga, not something less).

Is a QT 4.5 like IDE planned?

_________________
SamFlex Complete 800Mhz System + AmigaOS 4.1 Update 4
Amiga 2000 DKB 2MB ChipRam GVP G-Force040 Picasso 2 OS3.9 BB2
AmigaCD 32

As I have done somthing very similar already for Amiblitz3, and planning to port this to a new, more C Style language (A/A++), maybe we should get in touch and exchange experience.
I am also about to finish a GUI toolkit (NTUI) that is quite powerful (but not overcomplicated) and abstracts from the AmigaOS API, and it is highly portable.

_________________
--
Author of
HD-Rec, Sweeper, Samplemanager, ArTKanoid, Monkeyscript, Toadies, AsteroidsTR, TuiTED, PosTED, TKPlayer, AudioConverter, ScreenCam, PerlinFX, MapEdit, AB3 Includes and many more...
Homepage: http://www.hd-rec.de

@all

I am overwelmed how many people are reading this announcement Thank you all for your intrest and for your comments!

I wanted to say something again about "qt or not qt". As I mentioned before, we have been discussing this internally quite a long time, and it was no easy decission to make. The reason why we decided not to do it, was not purely technical, it had also something to do with our philosophy, how we see the Amiga in the future. We think of porting software as a very important strategy to get software fast from other platforms to the Amiga, so we Amigans can use it too. It is a quantity decision. Whether it is the best way or not, that's up to anyone to decide on his own, and can also be measured on the quality of the software.

We definitely do not want to take part in the porting business. We both have been doing this in the Linux world, and to be quite honest, we do not want to do it again. Not because it is not good, just because we want to create something new (and hopefully better as a port ).

We want to implement the "Amiga way" in Neat, not the "QT/Linux" way. QT can be ported to the Amiga until a certain level. If you go behind this level, you will see porting does not make sense anymore, because both systems have completely different aspects and ways of doing things. This is the only reason why the "basics" of NEAT look like "QT". Basics are basics, and file IO is file IO. But drawing GUI things on X is somewhat different as doing GUI on the Amiga. And here the paths separate completely. Not to speak of things which are completely new on the Amiga like XMos. What could we possibly take from other platforms to port for it?

I do think it is important that some people will take care of the porting issue, and some people will take care of something completely new. We hare here for the "new" things. That is our goal. It is up to you to decide what path you want to go. Take over habbits from other systems, or create new habbits on a new system.

Just imagine the Amiga 1000 starting in 1985 with "Commodore BASIC V3.0 System ready"... I don't think that would have been a big success... (or even better, starting MSDOS 6.22 lol).

Last but not least, we can only say future will tell. If someone comes along and starts porting QT, we would definitely be interested what the future developers are going to use. If it would be the QT port instead of NEAT, then NEAT would become obsolete. That's the risk of the game isn't it? But it could also kick our ass to get even better and faster in our framework development and to get users on our side.

And about the GUI thing. I can't say anything about that one. Someone would kill me at the moment

Yours,

Jeanluc.

@jeanluc72

Quote:

We definitely do not want to take part in the porting business

I'm not sure that anyone is suggesting that you port Qt - although by mentioning it you obviously knew that it would be the focus of the conversation... right?

What we need to see is a level of abstraction so that developers familiar with other frameworks can immediately begin writing code for the Amiga. This doesn't require a full port. If calling conventions for Qt or GDI or (insert your favourite toolkit) were available, the learning curve for development on the Amiga would be reduced significantly.

Equally important is the fact that already at least 1 other developer has indicated that he is working on another project that could mesh with this one. The Amiga community really needs to do a better job of cooperating to get things done. This should also include dialog with Hyperion to help reduce the possibility of reinventing the wheel.

I am not familiar enough anymore with qt (used it one time ten years ago, i think), but Amiga needs some basic modern OOP-styled GUI Framework...

something along the lines of (simplified):

[code]
window1=new Window();
window1.width=400;
window1.height=300;

button1 = new Button();
button1.x=100;
button1.y=200;

window1->add(button1);

[/code]
and stick some good eventlistening system on top of that...

Stuff like MUI looks so outdated these days, I really don't want to touch that.

If you can accomplish that, all my good wishes to you.. :)

@steril
What you write above it quite outdated, not MUI. Dont mix the syntax of writing it with the concept behind.
BTW, that is why I propose NTUI.

Here, you code would look like (much less headache than your pixelcounting above):

[code]
/* This comes from a file or a string:

<Window id="window1">
<Button id="button1" text="Click me!">
</Window>

/* .. and if you want to create the window:

ntui_CreateFromXML(...)

/* ... and the loop goes (Pseudo Code):

.mainloop
Repeat
*tuiEvent = ntui_GetEvent(...)
Switch(ntui_EventNotify(*tuiEvent) {
case ...
}
Until !*tuiEvent
Goto mainloop

ntui_FreeObject("window1") /* <= this is actually not really needed due to autoexit

[/code]

It is highly prortable too, no amiga specific things are involved in the high level.

Last edited by Wanderer on 07-Jan-2010 at 09:59 PM.

_________________
--
Author of
HD-Rec, Sweeper, Samplemanager, ArTKanoid, Monkeyscript, Toadies, AsteroidsTR, TuiTED, PosTED, TKPlayer, AudioConverter, ScreenCam, PerlinFX, MapEdit, AB3 Includes and many more...
Homepage: http://www.hd-rec.de

@Wanderer

Not to burst your bubble but there is an Aros Public Licensed project that does the same exact thing as you propose at Feelin.FR. It works on AOS 3 and 4 IIRC. It should be easy enough to port to the other Amiga-like OSs also.