The Icon Bar: RISCOS_Ltd._survey_comments: The Icon Bar: RISCOS Ltd. survey comments

RISCOS Ltd. survey comments

RISCOS Ltd. have put up a page on their website listing some of the responses they had to the last question of their recent survey: "What are the main things currently missing from either RISC OS itself or from current RISC OS computers that you think are essential for the future?"

There are several hundred comments listed, with an updated web browser being highest on most people's wish list (people seem especially concerned about support for JavaScript and CSS). Updated printer drivers, support for USB devices, DVD playback and an updated version of Java also appear quite frequently.

Which of these things do you think is most important? Are we likely to ever see any of them? Why not share your views on our forums.

Source: Drobe

RISCOS Ltd. survey comments

(16:16 7/4/2001)
  Anon (00:11 8/4/2001)
    Francis (11:32 8/4/2001)
      Tim Fountain (13:26 8/4/2001)
        Maurice (20:08 8/4/2001)
          Lee Johnston (08:04 9/4/2001)
            Andrew Flegg (08:13 9/4/2001)
              Guy Inchbald (08:18 9/4/2001)
                Rob Kendrick (08:53 9/4/2001)
                  Nick Wright (17:27 9/4/2001)
                    Annraoi (17:45 9/4/2001)
                      Steffen Huber (08:35 10/4/2001)

David Bryan

Message #88457, posted at 16:16, 7/4/2001

Unregistered user

595 users, who'd have thought it would still be
that many ;-) While it is commendably open to
put up all those comments, it hardly represents a
compelling endorsement. I didn't think the site's
remit was to run down RISC OS. It would be better
to have a page saying "Many users thought RISCOS
Ltd should concentrate on getting the following
areas improved." Then give a short list of key
points.

[ Log in to reply ]

Anon

Message #88458, posted at 00:11, 8/4/2001, in reply to message #88457

Unregistered user

It's interesting tht so many people have commented on a need for a particular type of app - even though this has nothig whatsoever to do with ROL.

Quite a lot of respondants seem to be still using 3.7 or earlier.

Lots of advocacy for adobting Windows traits. This can be good (eg spring-loaded directories) but it would be bad if RISC OS lost it' individuality. The main chanes needed are under the hood.

[ Log in to reply ]

Francis

Message #88459, posted at 11:32, 8/4/2001, in reply to message #88458

Unregistered user

Spring-loaded directories are a Mac feature not a Windows trait - and having not actually used it I'm not convinced of its usefulness.

I do agree with the suggestions about proper native support for the global clipboard (i.e. Ctrl-Z Ctrl-X Ctrl-C Ctrl-V) in all text icons. From using implementations on other OS's I am beginning to find it a real pain not having this feature in RISC OS - although !Swipe use a useful kludge.

A few other points that seem to be quite important:

A good C/C++ development IDE
----------------------------
(IDE = Integrated Development Environment). This would address another common moan of "more software".

USB support
-----------
This is in line with RISC OS's user-friendly approach. It would be lovely to be able to use USB scanners, Zip drives, CompactFlash readers and the like! Firewire support, anyone?

Support for future non-26 bit ARMs
----------------------------------
In my opinion the XScale's speed is needed now as one of the major uses of RISC OS - image creation and manipulation - is held back by lack of hardware "grunt". And yes it needs to go hand in hand with fast memory and disk subsystems.

[ Log in to reply ]

Tim Fountain

Message #88460, posted at 13:26, 8/4/2001, in reply to message #88459

Unregistered user

The question does ask what RISC OS or RISC OS computers need so I think an app is a perfectly acceptable response.

[ Log in to reply ]

Maurice

Message #88461, posted at 20:08, 8/4/2001, in reply to message #88460

Unregistered user

I agree, RISCOS Ltd. should be concentrating on getting the OS working properly, and leave the applications stuff to tho
se that know.Unless they can come up with anything radical they would just be repeating existing applications anyway.
We already have a few browsers, and internet connection suites, why bother?

Sorry to be so cynical, but it just seems to be a touchy-feely smoke screen.
Can't they just get on with it??

[ Log in to reply ]

Lee Johnston

Message #88462, posted at 08:04, 9/4/2001, in reply to message #88461

Unregistered user

I think listing the apps that people want is a good thing. I agree that RISC OS Ltd shouldn't be the ones writing them but at least it allows developers to take a look at what people want.

It's interesting how often the same stuff comes up again and again. What stood out to me are the comments about development tools and 3D support. While I agree with 3D support I hope people don't think it'll magically make 3D games appear because it won't.

As for development tools, I do feel that a fully featured C++ compiler and Java would be nice. I suspect that both of these might be beyond RISC OS Ltds (and even Pace's judging by comments on the newsgroups) resources. OTOH a decent Java VM could really help on the apps front as you can get complete editors, web browsers etc for it now.

I'm not sure about the comments about "Windows" compliance - the MFC is big, slow, bloated and not very nice and yes I have developed with it. With libraries like Desk, the toolbox and hopefully GuiLib I don't think libraries are a huge problem - it's more a case of complete lack of documentation (although the toolbox manual isn't bad). Also the comment about CORBA is interesting - I wonder just how RISC OS would benefit from an implementation of a distributed object platform *scratches head in puzzlement*.

There are other comments about Rapid Application Development (RAD) tools along the lines of Delphi. Given the availability of WimpBASIC, WimpWorks and Dr Wimp it seems that these products would need highlighting (I would offer to write an article for TIB but I've never used them being a C / ARM coder).
Perhaps if the TIB people have time they could approach Clares, Jaffa Soft and whoever looks after WimpBASIC (sorry, memory fails me) and see if each could write an article on their product, perhaps with a simple tutorial on how to build a small app (note, riscos.org are running a Dr Wimp tutorial).

[ Log in to reply ]

Andrew Flegg

Message #88463, posted at 08:13, 9/4/2001, in reply to message #88462

Unregistered user

I noticed the number of comments about RAD/IDE tools and I was going to point out WimpWorks, WimpBASIC and Dr. Wimp.

I also think Lee's on to a good idea with simple tutorials - certainly we (Jaffasoft) would be glad to assist in the production of a tutorial series or something...

[ Log in to reply ]

Guy Inchbald

Message #88464, posted at 08:18, 9/4/2001, in reply to message #88463

Unregistered user

Frighteningly long list, really. Dev tools, OS functionality, utilities, webby stuff.

Hypothetically, if ROL were to open up the source code, maybe even open up the copyright a little, would this encourage part-timers to cotnribute more than it would discourage salaried developers?
ie would it boost or stifle development?

Meanwhile, anybody know what happened to the port of the GNU C++ compiler?

[ Log in to reply ]

Rob Kendrick

Message #88465, posted at 08:53, 9/4/2001, in reply to message #88464

Unregistered user

Nick Burrett got caught up in something much more interesting :-/

http://www.freevsd.org/

[ Log in to reply ]

Nick Wright

Message #88466, posted at 17:27, 9/4/2001, in reply to message #88465

Unregistered user

Unfortunately, as far as Im concerned, the current Rapid Application Development style environments dont really cut the mustard - not enough anyway. Cant we have a delphi-esque tool that uses BASIC instead of Pascal?

[ Log in to reply ]

Annraoi

Message #88467, posted at 17:45, 9/4/2001, in reply to message #88466

Unregistered user

I was thinking that myself.

I've used Delphi quite extensively on the PC and it is an excellent language. It is highly object oriented and has (thankfully) improved upon the Pascal language. The fact it can make developing for Windows tolerable (considering how ucky that is) is no mean feat.

I can, however, appreciate that trying to encourage RISC OS users to adopt something as "foreign" as Delphi might be problematic, and a BBC BASIC compliant language (coupled with some of Delphi's cooler features) might be a viable compromise.

I am no dyed in the wool Pascal fan (vanilla Pascal is boring) but Delphi ISN'T vanilla Pascal, it has a sane object oriented model, and the VCL (Visual Components Library) means that programmers need not delve into the murky depths of windows. The debugger and IDE work seemlessly together and make coding and debugging relatively painless. A similar language for RISC OS would be a real boon for developers reducing development time (time=money) and making code more reusable (another big plus).

Bear in mind Delphi 6 and Kylix (Delphi for Linux) do allow cross platform development, this must surely show the strength of the underlying OO principles of that language. Such a language on an RPC may well allow rapid cross platform development (allowing additional income from sales on platforms a RISC OS only developer would not normally consider). Just something to think about folks !

[ Log in to reply ]

Steffen Huber

Message #88468, posted at 08:35, 10/4/2001, in reply to message #88467

Unregistered user

If someone wants a "Delphi-like" language, I would suggest to take a look at Ada 95. There is a port of GNAT for RISC OS available (unfortunately a quite old version based on GCC 2.7.2, but it is much more mature than the C++ part of GCC ever was).

Of course there is no such thing as a rapid development environment available. In fact, I am probably the only one who has ever developed a WIMP application with Ada. My RISC OS library is very rudimentary (things only get added when CDBurn needs them badly) and very procedural in style, but a replacement is already planned (see http://www.huber-net.de/ada_e.htm for more information about that whole stuff).

Ada is syntactically a lot closer to BBC Basic than C or C++, and it is a *very* powerful language. It supports a lot of low level stuff ideally suited for bit fiddling and hardware access. And interfacing to already existing C stuff is very easy, too.

Hopefully, the GCC project will be able to provide a new GNAT port once GCC 3.0x is out.

Steffen

[ Log in to reply ]