[gccsdk] Supporting riscos.info

Peter Naulls peter at chocky.org
Fri Nov 23 09:29:36 PST 2007


In message <15715c414f.druck at druck.freeuk.net>
          "David J. Ruck" <druck at druck.org.uk> wrote:

> On 13 Nov 2007 Peter Naulls <peter at chocky.org> wrote:
> 
> > In message <471C16EA.8040509 at chocky.org>
> >           Peter Naulls <peter at chocky.org> wrote:
> 
> >> It's that time of year when we need to pay the bills for the hosting of
> >> riscos.info.
> 
> [snip]
> 
> > Many thanks to those who've contributed.  I'm over half way there. I'll
> > add an item to the front page of the wiki to see if I can collect the
> > rest.
> 
> I know riscos.info isn't just about Firefox, but I'll happily donate 
> another £50 if there is any work forthcoming on Firefox 2. Its is so =20
> frustrating to see it seemingly abandoned just short of being usuable.

n.b.: the amount has now been covered.  Thanks to all.   Any excess
I'll put towards supporting RISC OS in some manner.  

disclaimer: what follows is not an attack on druck; on the contrary,
it is an extension of things he said many years ago regarding the future
of RISC OS.  Some of what I say in the following presumes that the
suggestion is that I personally do the work, even though that isn't the
only option.

Essay follows:

Adding all the final RISC OS nicities to Firefox 2 wouldn't just be a
matter of a few comfortable evenings hacking in some Wimp stuff.  More
likely it would several weeks work digging through layers of XUL and
obscure C++.   It'd still be relatively slow (and never mind the RiscPC
users complaining that moderm applications need to run on their
hardware), and it'd still have people complaining that it "didn't
support flash", even though that's a completely separate project.

I know it's real tempting to suggest quick fix ups for something which
seems so obviously trivial to fix, and let's assume for a moment that
might be the case, the problem is that this is the attitude which has
led to so many of the bad designs, duplication and hackery that is now
present in RISC OS.  c.f. my complaints about !HForm.    For that, I
could blame Acorn in part, but it wouldn't be contructive to take issue
with a company that has been gone so long.  In any case, there are a
number of parties still actively ensuring that RISC OS development
remains like this. I'll merely point to some of Castle's activities in
the past at this point in time that I was vocal about.

My real problem is the complete lack of direction that is apparent in
RISC OS.  There's no one going in any obvious direction.  No plans to
fill in the glaring holes in the application space (with some exceptions
regarding the work taking place on this list), and a great deal of
apathy.  Again, we can point fingers at various parties who've been
instrumental in misinformation and general BS, but it doesn't get anyone
anywhere.

Perhaps worse still, depending upon your POV, Linux does almost
everything I require these days, and a great deal more than RISC OS,
even if there are still many problems (and I know better than most). A
lot of my browsing involves media sites using flash.  Now, I'm not sure
it's realistic to demand that even an Iyonix can play streaming video
from YouTube, but it seems insane in 2007 to point out that it can't
stream music from the likes of deezer.com (French free music site).

Instead people are still going on about porting RISC OS to x86 (never
mind the fallacies involved in that, and even though the _two_ emulation
solutions run exceptionally well and are a far better solution), asking
for faster hardware, and asking that we resurrect the corpse of !Browse.

Anyway, the point is I wouldn't want to embark on any RISC OS work
unless I felt there was some kind of direction or emphasis on working
though the issues that it has.  We've spent much of the 10 years since
Acorn pissing in the wind, for what?  As an experienced software
engineer I must insist upon reusability and doing work properly.   At
this point, for Firefox, even though I'm sure it will seem absurd to
many of you, that would mean working on Firefox version _3_ (about to be
released) or perhaps even the version after that.  Only this way could
we ensure that furture work is minimised (the RISC OS support for the
nspr is in fact present in the upstream Firefox 2 sources). And again,
it wouldn't be work on Firefox per se, but work on all the technologies
surrounding it so as to make other applications easier to convert to
RISC OS.

Essay ends.


-- 
Peter Naulls - peter at chocky.org        | http://www.chocky.org/
----------------------------------------------------------------------------
RISC OS Community Wiki - add your own content   | http://www.riscos.info/



More information about the gcc mailing list