29-Mar-2024 09:27 GMT.
UNDER CONSTRUCTION
Anonymous, there are 83 items in your selection (but only 27 shown due to limitation) [1 - 50] [51 - 83]
[News] $1000+ to the first person who ports Mozilla to AmigaOSANN.lu
Posted on 19-May-2003 20:57 GMT by catohagen (Edited on 2003-05-20 08:26:37 GMT by Christian Kemp)83 comments
View flat
View list
"catohagen" quotes Bill Panagouleas, Founder/CEO of DiscreetFX: Looking to make a $1000, port this app and I will pay you. If the dead BeOs can have a port so can Amiga OS, spread the word. More information can be found here. As of right now, there is a $2,150 pot.
$1000+ to the first person who ports Mozilla to AmigaOS : Comment 57 of 83ANN.lu
Posted by corwin on 20-May-2003 12:52 GMT
Thoughts about some of the opinions expressed in previous posts against a mozilla port :

1 it can't be done on the amiga because AmigaOS isn't modern enough

-> an early Mozilla version for an older AmigaOS version was already ported which proves that it can be done

2 it's a linux thing and therefore not easy to port to the amiga platform

-> Mozilla was designed to be portable and already works on systems which are not based on Unix like MacOS9 or Windows9x. According to Mozilla.org, 90% of the code is platform-agnostic. I don't mean it's easy, this is probably a very difficult task, but the fact that Mozilla is available on all linux distros does not make it a unix-only app.

-> Mozilla.org is very supportive and helpful to new ports of Gecko, porting Gecko to AmigaOS would mean that amiga developpers could benefit from the experience and help of the mozilla community and developer tools like bugzilla, dedicated newsgroups or websites like mozdev.org to host development. it would also bring a lot of exposure to the amiga platform.

3 we should give this money to Ibrowse (or other commercial browsers)
-> Mozilla is open-source, the other browsers are commercial products

-> Ibrowse would NEVER be able to approach Gecko capabilities in a near or even distant future. All current modern browsers (Opera/Gecko/Khtml/Tasman) needed *years* of full-time development by dedicated teams, all these teams included the best W3C experts (like Daniel Glazman, Tantek Celik, Hakon Lie, Eric Meyer...). Please, do not reinvent the wheel and do not dive into the NIH (Not Invented Here) syndrom ! Thinking that implementing CSS2 in current amiga browsers would be an easy task when the people who *created* the CSS specs needed years to do it in their own projects is... not realistic. It would make more sense to port Gecko and use it as Ibrowse rendering engine, that's what Omniweb is precisely doing with Khtml.


4 we don't need Mozilla, we just need a bit of CSS and DHTML support

-> Basic DHTML/CSS support is what 1997 browsers did on other platforms. Current minimum requirements for a rendering engine are XML/CSS2/DOM2/ECMAScript... Having a modern browser is now a minimum requirement to hope to get switchers to the amiga platform some day.

-> mozilla is not just a browser but a set of cross-browser technologies. It means for instance that all the XUL software and XPI entensions created by the mozilla community would be available to the Amiga (go and see the Amazon Catalog Browser or MozBlog for instance). It also means that you would have access to the many localization packs available (catalan, breton, nynorsk, hebrew...).

-> A Gecko port would benefit to all amiga users AND developpers since it could be embeddable in other amiga products. On windows, software packages like HTML-kit or TopStyle use Gecko as the internal viewer.

5 Mozilla is bloated, current amioga brrowsers aren't

-> Mozilla runs fine on modern hardware and each new version is faster and better than the previous one. Kmeleon, a light-weight Gecko browser, runs pretty fast on my old Pentium 120Mhz.

-> Ibrowse, Voyager, Aweb are simply not in the same league, it's like comparing multiview with Photoshop. Of course the Mozilla suite is more heavy, but can you with current amiga browsers browse RSS feeds, use it as a blogging tool, display XML, make XSLT transformations, manipulate the DOM tree (via the DOM inspector), use mouse gestures...
$1000+ to the first person who ports Mozilla to AmigaOS : Comment 58 of 83ANN.lu
Posted by Bill Panagouleas on 20-May-2003 13:12 GMT
In reply to Comment 57 (corwin):
This long comment does make a lot of sense.
$1000+ to the first person who ports Mozilla to AmigaOS : Comment 59 of 83ANN.lu
Posted by Henrik Mikael Kristensen on 20-May-2003 14:59 GMT
In reply to Comment 58 (Bill Panagouleas):
I've posted a comment about this effort on our developer list, and await response on this from the developers. We already are on a path of deciding which solution will be the best for the future of AWeb (such as porting Gecko, KHTML or the likes).

Regards,
Henrik Mikael Kristensen
AWeb Development Team
$1000+ to the first person who ports Mozilla to AmigaOS : Comment 60 of 83ANN.lu
Posted by Ferry on 20-May-2003 15:10 GMT
In reply to Comment 39 (Olegil):
By the way, I do not find Opera to be that good. I do not browse many pages, but it has failed badly to access in two of them, one webmail page while browsing messages (javascript), and a password protected page (I haven`t been able to access with the correct login/password). And I can read the webmail even with AWeb without problems...

Ferrán.
$1000+ to the first person who ports Mozilla to AmigaOS : Comment 61 of 83ANN.lu
Posted by 3seas on 20-May-2003 16:29 GMT
Not to take away from this effort but really only to point out that there does seem to be a trend developing of contributing to software development.http://www.chodorowski.com/sponsoring.html - for aros developmentthere is also some teamaros effort.There are different approaches to how such a thing is done.I suppose the spectrum ends might be the following:The Bounty System - payed after the work is done.Sponsoring - buying time for the coder to work regardless of what gets done.There is perhaps the logistics of how these things actually happen.What if: coder does the work but it's not accepted/payed for?What if: what if the code doesn't work correctly?What if: monies are collected but not payed out (someone took the money and ran)?etc.. there are a bunch other possible situations. What it all comes down to is that of having some assurance on both sides of the fence between developer(s) and funder(s).So long as there is trust and it's never broken..... but even then ....in life shit happens.(OH Hell, what am I talking about....one word..... Amiga....) Point MADE!Anyway, I have myself gone thru www.rentacoder.com to get python based code written.Maybe it's a route to go, certainly it has more developers looking for such work, But the reason I mention them is because they have already worked out alot of the logistics, including dealing with different currencies and exchange.Perhaps it'd be worth doing some research on.
$1000+ to the first person who ports Mozilla to AmigaOS : Comment 62 of 83ANN.lu
Posted by Robin Cloutman on 20-May-2003 16:47 GMT
For those who are looking into it (I've had the source for a good while, but at a 50 meg download, and god knows how much un-tgz'd, i never went any further till now) -

http://www.mozilla.org/hacking/coding-introduction.html

Basically there are only a few parts that need to be ported -

1. NSPR - Netscape portable runtime
This is the under-the-covers part of mozilla, it deals with memory, threads, files, and other (non gui) stuff. While ixemul can probably deal with everything other than the thread/pthread code, it would still be a lot better to port every internal method to AmigaOS native functions (as mos/aos4/aros should all be able to use it transparently). This is probably the easiest part to port, but also the most cpu intense part, so needs to be as fast as possible...

2. XPCOM / nsISupports / nsCOMPtr
This shouldn't need any work, *but* it does seem to use threads etc a lot, so might need some #ifdef stuff in to make it amiga compatable (we can start a new task with an existing function, but *nix forks start the new task from the fork() call)

3. Graphical User Interface / XUL
Not sure exactly where this part is in the sourec tree, but this is the part that translates internal gui represenations (from buttons to the web page itself) into on-screen display. If we're lucky then it might just need some simple converting of basic pixel/line drawing functions, but there's also gfx clipping, font display, and more... I can easily see this taking the longest to port as getting it working is easier than getting it working efficiently ;-)

If my miggy was still working (hd died last week, and will take a little while to get it repaired/replaced) I'd be *really* busy right now... though somehow i think ./configure might take an hour or two... Oh, and that needs to be ported to so it knows about AmigaOS ;-)

I reckon the hardest part of porting it for normal amiga coders will be the fact it's written (or the parts needing ported at any rate) in C++, rather than C - though a quick note, it's not using any C++ exceptions, it uses return values to simulate it (sigh of relief)...

Oh, and if ppl are porting, don't forget to make the 256 colour chrome (front end) the default, rather than the 16-bit it uses by default - not everyone has got gfx cards yet... but it'd be *really* slow on those ;-P

Robin
$1000+ to the first person who ports Mozilla to AmigaOS : Comment 63 of 83ANN.lu
Posted by Robin Cloutman on 20-May-2003 17:26 GMT
Just thought to add - the reward *must* add that the person/group getting the port to work must supply their source too (so it can be put back into the cvs source tree) - that way bugfixes etc can be done to the amiga specific stuff, and we can keep with the up-to-date versions...

Robin
$1000+ to the first person who ports Mozilla to AmigaOS : Comment 64 of 83ANN.lu
Posted by MarkTime on 20-May-2003 17:26 GMT
@Robin

nothing is going to make a 50mhz machine fast.
I would assume that the person has a graphics card and 64mb of RAM.
I wouldn't even think of worrying about support for some ancient piece
of garbage..

If they really want mozilla, they can upgrade.
However, I would also target OS 3.1 and 68k, only because how can
one program for a non-existent OS...or a beta, or one under NDA,
or one with ever changing spec's and api's...etc., etc.

so i.e. I would assume, that someone either has a classic 68k amiga,
in which case they are accustomed to, and proud of being slow...or
aros/amithlon/morphos/aos4 in which case they can run the 68k code anyway.

thats my two cents

I agree about breaking it into such parts...now this may or may
not be relevant anymore, but probably the easiest way to port something
based on GTK+ is just to port GTK+, oh sure, thats not the only way to do it,
one could remove the GTK+ dependency from mozilla (I assume in the linux version
it still uses GTK+)....

but another way is to port GTK+, and then also, not only will it make porting
mozilla that much easier, but allow for ports of GIMP and other neat stuff.

Plus its a definable block to do first. Now....of course, its not trivial,
but I think its the first step.

I would be glad to hear comments though....of course.
$1000+ to the first person who ports Mozilla to AmigaOS : Comment 65 of 83ANN.lu
Posted by Robin Cloutman on 20-May-2003 17:55 GMT
In reply to Comment 64 (MarkTime):
I used to run it (about 1.0 release time) on a p166 mandrake setup - knowing how efficient the things are, i'd compare it to *feeling* (on an 060/50) like mui on a vanilla 1200 (with some extra mem) - so useable, and scrolling shouldn't be too bad, but clicking stuff and rendering would probably make it seem slower...

I'd personally want to compile it for 3.1 - though with 3.9 includes, all the other aos alike things (mos, aros, aos4) will hopefully not change any of the basic API stuff that would be needed. winuae (what i'm limited to atm), amithlon, etc, would probably just be emulating 3.1/5/9 anyway.

I'd not assume the gfx card, but i would make 256 colours an absolute minimum - it's unrealistic to do less (maybe something like guigfx.lib could get used to simplify colour reduction transparently to the port). If not using guigfx, then speedwise would probably want to code direct calls for p96/cgfx, so you can guess my choice ;-)

Memory - well, 32mb i'd say is the minimum you'd need, but would probably feel slow - not looked into the NSPR api, but it might deal with virtual memory directly, or the port might need to be able to deal with lower memory well (it *could* run on 32 megs on mandrake with vram - just not very enjoyable). As with anything else though, the more memory you have the better. I'd say port it with 64 megs in mind, but see what it needs once done.

When i was looking into it last year, i did consider something along the lines of porting X/gtk/etc - but in the end decided the overhead of that (already slow) system would make it feel like reaction... i mean mui on a vanilla a600 ;-) It is a possibility, but the XUL framework, once ported, shouldn't need much more than minor bugfixes to keep it current, and would no doubt feel more than twice as fast as anything else... even making a statically linked gtk library would probably add too much to the overhead, but as i've not looked at gtk at all (unlike say sdl and X) i can't really give anything other than my opinion...

Is it just me, or is this the first time i've seen ANN turn into a coders forum yet this year? 8-)

Robin
$1000+ to the first person who ports Mozilla to AmigaOS : Comment 66 of 83ANN.lu
Posted by MarkTime on 20-May-2003 18:09 GMT
In reply to Comment 65 (Robin Cloutman):
@Robin

ahhh, occasionally people start talking code.
I looked at Mozilla's site, and see their reasoning behind XPToolkit
et al....which is fine, they have unlimited resources compared to us,
I think we want the easiest/fastest way to get a modern browser.
And by easiest, I mean easiest to code, not to use, and also, fastest to code,
not to use.

perhaps xpfe is still the answer...but after looking around gtk+ seemed the best to me. as GTK+ could still be used for the front end

now, I don't mean to port gtk and x windows both...sure someone could do that, in fact, that has already been done, I imagine...
but as you mentioned...it has overhead, and worse, the user doesn't think they are running
a native app, they won't install it anyway...not if they have to
configure a whole x windows environment.

I was thinking more along the lines of making a native GTK+ port, and that
will allow, not only to do a kind of early mozilla 'classic' front end, but
also, port gimp, or another gecko based web browser.

as a matter of fact, after porting GTK, could wet people's appetite with a quick port of dillo or something.

but anyway, I just want to help, so someone take over the project and I'll see where I can help.
$1000+ to the first person who ports Mozilla to AmigaOS : Comment 67 of 83ANN.lu
Posted by amorel on 20-May-2003 18:33 GMT
In reply to Comment 21 (Bill Panagouleas):
Such little money is peanuts for years of work, besides the fact that it is close to impossible to port mozilla/firebird. Good luck to whoever tries it, they`ll find out sooner or later :-) and wasted a lot of time and energy.
$1000+ to the first person who ports Mozilla to AmigaOS : Comment 68 of 83ANN.lu
Posted by PMC on 20-May-2003 18:35 GMT
In reply to Comment 6 (hooligan/dcs):
Hahahahahahaha!Yep, tried Voyager and it fell over. A lot. Damn shame as it is a really nice browser. Still, IBrowse used to crash with clockwork regularity until I installed "StackAttack"....
$1000+ to the first person who ports Mozilla to AmigaOS : Comment 69 of 83ANN.lu
Posted by Rafo on 20-May-2003 18:42 GMT
yeah...sure. And M$ is offering a free copy of windows XP and 2 years
in jail to someone porting MSIE to AmigaOS.
$1000+ to the first person who ports Mozilla to AmigaOS : Comment 70 of 83ANN.lu
Posted by MarkTime on 20-May-2003 18:46 GMT
In reply to Comment 67 (amorel):
agreed, you wouldn't do it for the money, its not enough.

You'd do it, cause you'd be legendary...thats the hacker ethic...nearly impossible is exactly what you want to do.

sheez, kids.....go watch some mtv or sumtin
$1000+ to the first person who ports Mozilla to AmigaOS : Comment 71 of 83ANN.lu
Posted by Bill Panagouleas on 20-May-2003 18:51 GMT
Amiga Inc. vs Genesi donation Challenge now up on AmiZilla website, will the real Amiga please stand up.

Pot now @ $2321
$1000+ to the first person who ports Mozilla to AmigaOS : Comment 72 of 83ANN.lu
Posted by Troels Ersking on 20-May-2003 20:17 GMT
Would be nice if the diffrent Amiga companies would donate a little amount. I am well aware that Amiga inc. doesn't have a lot of resources but if ALL companies payed just a little to the fund it would increase it a lot.

Amiga, Eyetech, Hyperion, Point.design, Kdh, Genesi, Vesalia, soft-hut, GGS-data, Datakompagniet etc...
It's in everybody's interest that the Amiga (in all variations, Aros, OS4 etc..) gets a decent browser as it (eventually) will increase the sales.

Best regards
Troels
$1000+ to the first person who ports Mozilla to AmigaOS : Comment 73 of 83ANN.lu
Posted by MarkTime on 20-May-2003 20:45 GMT
In reply to Comment 72 (Troels Ersking):
>>I am well aware that Amiga inc. doesn't have a lot of resources

well now the latest in that drama is Ben suggesting Amiga, Inc. has lots of money, in fact a person suggesting otherwise, or even suggesting that they'll take a 'wait and see' attitude is 'making a fool of themselves' according to Ben Hermans of Hyperion.

:-)

I kid Ben...I kid him cause he's over the top, but....anyway, don't leave out the possibility that Amiga, Inc. can bank roll the whole thing!!!

;-)
$1000+ to the first person who ports Mozilla to AmigaOS : Comment 74 of 83ANN.lu
Posted by Bill Panagouleas on 21-May-2003 00:12 GMT
Thanx to the great Amiga developer individual Computers and a few others the pot is now @ $2666.
$1000+ to the first person who ports Mozilla to AmigaOS : Comment 75 of 83ANN.lu
Posted by brotheris on 21-May-2003 03:08 GMT
In reply to Comment 64 (MarkTime):
For this, just look at Freeciv port for Amiga.
$1000+ to the first person who ports Mozilla to AmigaOS : Comment 76 of 83ANN.lu
Posted by Elwood on 21-May-2003 10:51 GMT
And what about not reinventing the wheel and joining the Ibrowse team ?
$1000+ to the first person who ports Mozilla to AmigaOS : Comment 77 of 83ANN.lu
Posted by Alkis Tsapanidis on 21-May-2003 13:38 GMT
In reply to Comment 76 (Elwood):
It's not reinventing the wheel. We need an open, modern browser, now, portable
to all Amiga compatible systems (IB will not be ported to MOS afaik).
This way, the *AMIGA* browser authors will have to work harder.
$1000+ to the first person who ports Mozilla to AmigaOS : Comment 78 of 83ANN.lu
Posted by Robin Cloutman on 21-May-2003 14:42 GMT
Ok, just set up a mailing list on yahoo so ppl who want to group together to port it can do so - if we do finish first then the prize money will be shared between the *coders* - so if you want to code and don't agree then don't join, and if you want to join, steal any code, and try to win yourself, then p*ss off...

http://groups.yahoo.com/group/amizilla/join

Personally I'm not planning on doing any coding (though I might offer the odd snippet), and if I did qualify for a share, even though I'm broke, I'd put my share towards some other amiga related project that i felt worthwhile (just so you know I'm not in this for personal gain).

If anyone else feels this way then make sure you say on the list or something, so we don't have arguments when (if?) it comes to sharing out the money. ;-)
$1000+ to the first person who ports Mozilla to AmigaOS : Comment 79 of 83ANN.lu
Posted by Bill Panagouleas on 21-May-2003 17:35 GMT
In reply to Comment 78 (Robin Cloutman):
Thank you very much for setting up the list Robin! Everyone that plans to help please join the yahoo mailing list.
$1000+ to the first person who ports Mozilla to AmigaOS : Comment 80 of 83ANN.lu
Posted by Don Cox on 21-May-2003 17:41 GMT
In reply to Comment 77 (Alkis Tsapanidis):
"It's not reinventing the wheel. We need an open, modern browser, now, portable
to all Amiga compatible systems (IB will not be ported to MOS afaik)."

Doesn't IBrowse work on MorphOS now? It should do. It works on Amithlon.
$1000+ to the first person who ports Mozilla to AmigaOS : Comment 81 of 83ANN.lu
Posted by Tof on 21-May-2003 18:36 GMT
Maybe we should do the same thing to buy the Amiga OS source to free it ....

If someone whant to take over .... 10$


Tof
$1000+ to the first person who ports Mozilla to AmigaOS : Comment 82 of 83ANN.lu
Posted by T_Bone on 21-May-2003 21:57 GMT
In reply to Comment 71 (Bill Panagouleas):
> Amiga Inc. vs Genesi donation Challenge now up on AmiZilla website, will the real Amiga please stand up.

Don't hold your breath on one of those two. (I'll leave "which one" up to the imagination of the reader)
$1000+ to the first person who ports Mozilla to AmigaOS : Comment 83 of 83ANN.lu
Posted by Kelly Samel on 21-May-2003 22:34 GMT
I would like to see Mozilla/Netscape rewritten
to have an MUI interface and use other Amiga API
functions as a replacement for the toolkits it
is currently using. Perhaps, this is akin to
porting the rendering engine to AWeb or Voyager etc.
but this would be a better solution to use the
existing GUI and OS instead of writing more new
libraries that would add more complexity in my
opinion. Sounds like the AWeb team are going in
this direction already so perhaps we will see some
progress sooner than it will take to port the
entire Mozilla sources, I hope so. :)
Anonymous, there are 83 items in your selection (but only 27 shown due to limitation) [1 - 50] [51 - 83]
Back to Top