In message <20080905150939.29111
...@trite.i.flarn.net.i.flarn.net>
Rob Kendrick <n
...@rjek.com> wrote:
>> OTOH I have been, for some time - well prior to the current,
>> limitless thread - trying to think of a creature (mythical or
>> otherwise) that I could add to the computer lexicon to describe Rob,
>> Ray Dawson and other perpetual critics of RISC OS who infest these
>> NewsGroups.
> Realists? Pragmatists?
Oh yes. But it's all getting really tiresome for those of us
(the majority?) who do not wear ROse-coloured spectacles; esp. the
sub-set who also are /still/ managing to avoid using Windows machines.
> The amount of fun to be had writing software for RISC OS is very
> limited compared to what can be done elsewhere, if only because
> everything's so much more effort and painful. I'm not the only one to
> say this - lots of other developers have, too.
This is well understood. Which is why so much gratitude goes to people
like Martin Weurthner (where's my umlaut?).
His posts to the NGs, also, are splendidly factual; but seldom dismal
or argumentative.
>> Not being a developer, I accept without question :
>>> Unless you have a blind love for RISC OS, move elsewhere to do your
>>> development. Otherwise put up or be prepared to do five times as
>>> much work as your peers for little or no return.
>> - but note that the doom-laden druck (whom we all know and love) has
>> little problem developing, on a Windows machine, software for RISC OS.
> He proves the point; he had to resort to using Windows development
> tools to debug and develop his RISC OS software.
Yes.
> It was easier to write a load of library call emulation and deal with
> Windows than it was to do it on the real thing. That should tell you
> something!
It does. It has. But, in the direct sense, I don't need to be told, as
I am not a developer.
> Initially, the UNIX version of NetSurf only happened because it meant
> the developers could use good debugging tools to find bugs, because
> they simply don't exist on RISC OS. There are still some annoying
> memory corruption bugs that afflict the RISC OS-specific code in NetSurf
> that's next to impossible to track down.
All power to your elbow(s) then :o)
> By comparison, running the UNIX-specific and OS-agnostic NetSurf code
> under a tool called Valgrind under UNIX gives you a nice listing of
> all your memory access bugs, where they are, and their exact nature.
> It makes finding memory bugs a 20 second joy, rather than a 20 hour
> hog with no guarantee of success.
You may rest assured that if I (developer or not) e.g. went the VRPC
route, it would be to a Linux box, aided and abetted by my daughter.
<snip>
>> !NetSurf, however, claims to be a cross-platform browser; which, I am
>> delighted to say, works very well indeed on my [minority platform]
>> machine. That being so, how can it hope to catch on elsewhere without
>> Javascript?
> It's already caught on. Just being available for UNIX has bought the
> attention of several more people who have contributed code - including
> all the Google Summer of Code work.
Which I have gleefully read about in Archive mag.[1]
> Without a UNIX port, hardly any of that work would have been done or
> to the same quality. NetSurf is packaged in several miniature Linux
> distributions. We even have it ported to boot loaders! All this
> extra exposure helps the RISC OS version, both directly and
> indirectly.
For which, as already mentioned, we NetSurf users are very grateful;
and rejoice in the ongoing improvements (I'm stiil on 22.3.08 !)
>> My reading of the various threads about browsers, and my own
>> day-to-day experiences, leads me to the following tentative
>> conclusions :-
>> a) No RO-only browser is, as we say in Norfolk, much cop.
>> b) The Firefox port is a handy port in a serious storm.
>> I use it about once a month, at the most.
>> c) NetSurf is the best of the bunch for routine use.
>> d) Incorporation of JS is Priority No. 1; despite the lack of
>> Flash, Java, Real Audio, or whatever proprietory plug-in.
> Incorporation of JS is perhaps the users' first priority, but not the
> current development team's.
As often stated; and I was not trying to nag - simply summarise.
[1] BTW - What are the comparisons with Google's "Chrome"??
> There are more incremental things to be
> done. Remember: we don't get paid.
Correct. And very praiseworthy.
> Every evening I get the choice to either read a book, socialise with
> friends, play a computer game, work on NetSurf, or do something else
> completely. We've been very fortunate to get some payed work done on
> it via Google.
Indeed.
> Basically, if it's not going in the direction you want it to, you know
> where the source is.
TeeHee. My programming days never really got started.
I could /read/ a COBOL listing, and that's about it.
(Hence the joy of BBC BASIC).
>> So, as Rob was (and is?) involved in the NetSurf Project :
>> a) Did he agree with the inclusion of the RISC OS platform?
> How could I not? It's the native platform. RISC OS users provide the
> vast majority of our feedback and bug reports, which is extremely
> welcome, as well as being essential for its further development.
That's the positive and encouraging reply I was trolling for.
>> b) Is he as keen as ever on upgrading that version?
> I've done a lot of work both directly and indirectly improving the RISC
> OS version. Almost all the work done on NetSurf in the past year has
> been OS-agnostic - it's benefited all the platforms we support
> regardless. (Be it RISC OS, UNIX, BeOS or AmigaOS.)
Right.
>> c) Does it matter if he can't use an Iyonix to do his work?
> I don't own an Iyonix, but most of my code is developed on a Linux
> box. The RISC OS-specific contributions I make are written on Linux,
> built on Linux, tested on an A9 Home.
Now you are preaching to the almost-converted ;o)))
>> d) When is he going to stop pouring cold water on RISC OS?
> When people stop saying "isn't RISC OS fabulous?" with no comparison or
> actual facts.
Well yes. But are you, tempting as it is, inadvertently feeding the
trolls?.
> If somebody says "What can be made better?" don't
> complain when somebody actually provides a list. If somebody says
> "Let's do this!" don't complain when somebody says "You can't do that,
> here's why, and here's what you should be doing instead that's more
> useful."
Quite. But we usually have Peter Naulls to tell us, in his inimitable
way.
You will agree, I hope, that I am not one of the 'us' (complainants)
in question.
> Do complain when somebody just asks vague questions with no regard for
> the question when they have a long and distinguished history of causing
> long, argumentative threads that they stay out of once they've started.
Back to The Trolls, then.
Thanks very much for a speedy, detailed and up-beat reply.
> B.
Cheers,
Terry
--
__o Terry Mills Norwich - Up The Canaries!!
_`\<,_ te
...@norridge.me.uk Clan 3444; Foundation 0037
(_)/*(_) Who needs Bill Gates? Iyonix 512
Phone: 01603-427900
http://www.norridge.me.uk
