A Call to Arms: developing the Web Browser

Right, I've kicked off some discussions with a couple of open source projects, specifically Dillo and Minimo (a small footprint Mozilla implementation), and we need to know what sort of man (and woman, thank you Reg) we can call on.

If you can offer anything to the project, particularly in the core development skills area, then please please please let me know.

This will be an open source project, and will involve a hell of a lot of work. For Minimo, we will have to write stub routines that are missing from WinCE, write an interface layer (XPC) and then write the User Interface.

XPC may have to be implemented in ASM, so any ARM assembler wizards out there would be most welcome too - failing that, I can dig in there... Although the last assembler I wrote was for 68000s and relied on agnus, paula denise and the copper being in the memory map too!

So, in short, the community now needs to dig in, roll their sleeves up, and get to work. The web browser that we have is insufficient, so let's do something about it.

With enough man (and woman) power, we can make this succeed. However, it is not a quick win, and it will be hard work.

Any takers?
Nick.

Raises hand



"tsubasa at your service
but wait a minute, I don't know assembly.. "

Will be glad to do anything I'm capable of.

UI design, graphics, publicity, someone to organise and keep cracking the whip. I'm your man.

Code wise. About all I can manage is:
10 print "I can't code"
20 goto 10
end

Unfortunately my computer died a few days ago, so I might have to put off learning c++ a bit longer, but as far asthe fundamentals of coding, I have them - commands aren't hard for me to learn. (Dude, I'm using Linux!!!) Yeah, but I can post anything helpful on my site because I now have one.

-Even though it will be open source, will it be released under the gnu public license? Because if it is we have a better chance of people making their own versions or sending code.
(even if we don't do it, the whole goal is just to get a good browser)

Unfortunately my computer died a few days ago, so I might have to put off learning c++ a bit longer, but as far asthe fundamentals of coding, I have them - commands aren't hard for me to learn. (Dude, I'm using Linux!!!) Yeah, but I can post anything helpful on my site because I now have one.

-Even though it will be open source, will it be released under the gnu public license? Because if it is we have a better chance of people making their own versions or sending code.
(even if we don't do it, the whole goal is just to get a good browser)

Thanks Folks, this is appreciated

Does anyone have experience of porting large products that were designed to be multi-platform? We may well be able to steal a march by using a pre-existing library. I know of Rainer Keuchel's CELib.dll, and that might well be a useful starting point...

Cheers,
Nick.

I agree with Chris on this... we will do everything we can to promote and help the developers from a site perspective. We'll even beta test for you...

Code though... uh.... well....

Hi group,

It's great to see determination to develope new Web browser to our lovely handhelds/palmtops and I want to do my best to help with development but I don't have any experience with it.

There are some useful links to PPC programming forum and ARM assembler docs.



http://www.pocketmatrix.com/forums/viewforum.php?f=5&sid=b67151d1b97a41ff27acd8206e0200d8
http://www.heyrick.co.uk/assembler/index.html
http://www.geocities.com/SiliconValley/Hub/6461/arm/
http://www.rainer-keuchel.de/wince/armasm.html

I should emphasise that the assembler coding will be a tiny, tiny part of the project and will only be to provide the interface between the cross platform COM library (XPCOM) and the native implementation. Or at least, that's what I understand it to be at the moment.

I'm glad to see what i started getting somewhere, i once again can say i have'nt a clue about programming, but if one thing im good at is finding software, if it exsist...i'll find it...So if anyone finds a need in me to look for some devlopment tools that might be helpful, i'd be glad to scrounge around...i have the time to do it where i know others do'nt...i've ran across some in the past that i know could be useful and i'm sure that has'nt been seen before by many or forgotten from disintrest and lost.....Steve

Steve, if you know of any libraries such as Rainer's CELib that implement currently missing parts of an API - either Win32 or glib - then shout . I think we're trying to stay on the standard toolset at the moment, rather than use glib and the like - which sorta rules out Gnuwince, or PocketGCC, or maimich's port of the GNU toolset. Although those yet might be a tempting option...

Cheers,
Nick.

I'd really stay away from assembler, if at all possible. There are plenty of people on MIPS and SH3/SH4 machines that could benefit from such a project as well - so ARM assembler would be an issue.

I'm starting to wish I'd never mentioned it

The lead developer of Minimo tells me that a small amount of assembler is necessary for the interfacing and I'm going to believe him. I do realise that this means that we've got "issues" when it comes to SH3, SH4 and MIPS, but they'll not be insurmountable.

Let's forget the assembler issue for the moment, and concentrate on the C/C++ side of things. Doug (Minimo lead) has wandered off to have a look at Rainer's CELib.dll and is wondering whether that might make things a bit (emphasis here, bit) easier.

It should be said that if we don't get a bit more interest from experienced developers, we are really going to be facing an uphill battle.

Once we've got a bit more of an idea of what's involved, I will start casting the net wider and farther, including the PocketPC dev groups as everyone on the WinCE platform could benefit from this: we just need to ensure that as we're moving along, we're being as inclusive device-wise as we can possibly be.

Cheers,
Nick.

I want to throw in a boat achor there.
Projects that wind up involving the PPC community usually wind up staying there. PPC community members wont have access to the Core OS devices that we do. The API's are completly differenet in the PPC world.
Where's the incentive for them to be backwards compatible?

If you lay the foundations it'll get hijacked and I can easly see us loosing out to the more robust API, where the more developers lurk.

IMHO, we have to get an open browser core going and then, once established let them in to do the PPC2000, 2002, 2003, 2003SE and 2005 ports.
We could probably do PS-PC ports internally.

The UI API/look and feel is different, for sure, but the majority of work in a port will be involved "behind the scenes", well away from the UI, where there are little to no differences in the API.

The front end components will, out of necessity, have to be different but if we can share resources to get the core ported to WinCE then we have broken a massive amount of work.