RISC OS source code release news
diodesign
For those who may have missed it, there's views and opinions from
developers on the news of the RISC OS 5 source code release here:
http://www.drobe.co.uk/riscos/artifact1706.html
Original announcement: http://www.riscosopen.com/press_20060929.html
Chris.
Ams
Good comprehensive well written piece on Drobe Chris, well done !
And good news as well. This will allow people, for the first time, to
be come direct stakeholders in RISC OSes future development. It should,
in my humble opinion, promote the development and expansion of RO 5
into new and exciting areas (porting to other ARM devices, developing
support for new hardware, enhancing the UI (an area Select currently
has an advantage in) and perhaps even taking advantage of the dual core
ARM's that are on the way).
More importantly it *may* draw back in developers who left the platform
- who perhaps have honed their skills elsewhere (e.g., Linux) and may
now be willing to use those skills to enhance and advance our platform.
Seems like Christmas has come early this year ;-)
Regards
Annraoi
Malcolm Ripley
"Ams" <a...@globalcafe.ie> wrote:
My worry is that developers are going to play catchy-up with those
aspects of the RO5 code that are behind the RO4 select code. This
would be a dreadful waste of time and effort. It would better to get
all those potential developers to kidnap Jack Lillingston and Paul
Middleton. Then lock them in a room and only release them when they
have agreed in blood to do what we all know should be done to get
RISCOS back on a single OS.
Re-joining the OS'es will have a much greater impact on future
development than the open sourcing of some out-of-date modules and
apps.
Now if we had access to the most up-to date modules and apps from a
re-joined OS......wowee things would definitely look rosy.
regards,
Malcolm
--
Malcolm Ripley uses a RiscOS StrongArm computer.
Current new sender keyword is : terrier (use this word in the
subject line if you are contacting me for the first time)
Email address is : "mripley" not forgetting: "@ukonline.co.uk"
--
Posted via a free Usenet account from http://www.teranews.com
Steve Fryatt
<d3cdf56e4...@mripley.ukonline.co.uk>:
> My worry is that developers are going to play catchy-up with those
> aspects of the RO5 code that are behind the RO4 select code. This would
> be a dreadful waste of time and effort. It would better to get all
> those potential developers to kidnap Jack Lillingston and Paul
> Middleton. Then lock them in a room and only release them when they
> have agreed in blood to do what we all know should be done to get
> RISCOS back on a single OS.
>
> Re-joining the OS'es will have a much greater impact on future
> development than the open sourcing of some out-of-date modules and
> apps.
In what way are they out of date?
> Now if we had access to the most up-to date modules and apps from a
> re-joined OS......wowee things would definitely look rosy.
It has been explained many times why this is never going to happen, and
also why it isn't so much of a problem as some people (looking for an
excuse to leave RISC OS) want to make it into.
--
Steve Fryatt - Leeds, England
Peter Naulls
Malcolm Ripley <mri...@ukonline.co.uk> wrote:
> My worry is that developers are going to play catchy-up with those
> aspects of the RO5 code that are behind the RO4 select code. This
> would be a dreadful waste of time and effort. It would better to get
> all those potential developers to kidnap Jack Lillingston and Paul
> Middleton. Then lock them in a room and only release them when they
> have agreed in blood to do what we all know should be done to get
> RISCOS back on a single OS.
It wouldn't help. I think enough time has passed that it's safe to say
the following. I sat in on a number of talks between those two parties.
Admittedly those sessions were very much informal, and there were many
many more occasions on which I wasn't present and which were much more
structured. Nevetheless, I'll let you guess which party had difficulty
even focussing on the issues at hand, but rather preferred to talk about
some mythical future convergance media/home storage centre, and just
ignored any questions to do with RISC OS. (Yes, similar to a
presentation this party gave at a RISC OS Show)
If you want to blame someone else too, I take some minor responsibility.
Around about the same time, I was contacted by a respected member of the
RISC OS community (whose name I won't sully by mentioning it in this
context, but wouldn't be surprising) in order to suggest someone who
might be appropriate to try and help negotiate between these two
parties. In my naivete, I suggested John Cartmell, not knowing any
better. You can surmise from the subsequent break down shortly after
that he totally failed, and of course this vaulted position explains
much of his currently over-inflated sense of self-worth.
--
Peter Naulls - pe...@chocky.org | http://www.chocky.org/
----------------------------------------------------------------------------
RISC OS Community Wiki - add your own content | http://www.riscos.info/
druck
> My worry is that developers are going to play catchy-up with those
> aspects of the RO5 code that are behind the RO4 select code. This
> would be a dreadful waste of time and effort. It would better to get
> all those potential developers to kidnap Jack Lillingston and Paul
> Middleton. Then lock them in a room and only release them when they
> have agreed in blood to do what we all know should be done to get
> RISCOS back on a single OS.
Well kidnapping asside, if you really think an OS with unified features from
both Select and RO5 is important to the future of RISC OS, then you have to
consider the ways it can be realistically acheived.
On one hand there is ROL with their dismal record of providing releases to
paying customers in the last 2.5 years, their refusal to commit to supporting
a large chunk of the market in the form of Iyonix users, the the erratic
personality of its director and its resourcing problems.
On the other hand, if there are developers willing to give their time to
providing a level playing field of APIs, backed by the ongoing security of
making the source available under an open licence, then while duplication of
effort is never ideal, it certainly would not be a waste.
---druck
--
The ARM Club Free Software - http://www.armclub.org.uk/free/
The 32bit Conversions Page - http://www.quantumsoft.co.uk/druck/
Ams
Malcolm Ripley wrote:
<snip>
> My worry is that developers are going to play catchy-up with those
> aspects of the RO5 code that are behind the RO4 select code.
To be fair many of these are UI (User Interface) issues. Updating
!Paint, !Draw etc., so that they look more appealing and easier to use
is a *good thing* even if it means re-inventing a wheel or two. From
other commentary it's pretty obvious that ROL either are unwilling or
unable to put any Select features into RO5.
They had their chance - now it's out of their hands.
> This would be a dreadful waste of time and effort.
If ROL can't/won't put select UI features into RO 5 then what's the
alternative? We'll eventually wind up with an OS that is modern and has
some Select UI features as well - not exactly a waste of time or effort
- an unavoidable one.
> It would better to get all those potential developers to kidnap Jack Lillingston and Paul
> Middleton. Then lock them in a room and only release them when they
> have agreed in blood to do what we all know should be done to get
> RISCOS back on a single OS.
>
Well that joint announcement someways back by Castle and ROL might have
suggested that they'd agreed in principle on remerging the OSes - but
given recent pronouncements from ROL that would suggest they [ROL] are
unlikely to do anything towards forwarding Select on Iyonix.
> Re-joining the OS'es will have a much greater impact on future
> development than the open sourcing of some out-of-date modules and
> apps.
This is just the *first* trench of OSS releases. Further will come
along in due course (e.g., Kernel, various drivers etc.,). I can forsee
a RISC OS 5 variant that shares some UI features with Select - but is
able to run on Iyonix (and possibly even earlier platforms - or on
alternative hardware).
Yes the OS will be unified - but it'll be a derivative of RISC OS 5, in
due course with the continual non-release of Select - Select will die
off. Let's face it Select was never really a "developer friendly" set
up was it? The API of Select was never *publically* documented by ROL -
you had to buy select to obtain it. You *can* get RO5's API from
Castle's website - and now you can get source as well.
>
> Now if we had access to the most up-to date modules and apps from a
> re-joined OS......wowee things would definitely look rosy.
Without starting to foam at the mouth in anticipation (not a pleasant
sight I assure you!) one can forsee a future RO5 based OS that will, by
being open to external development, benefit from being that little bit
better/faster at adapting to new challenges and hardware. Overall a
good thing.
Regards
Annraoi
Steve Potts
druck <ne...@druck.freeuk.com> wrote:
[snip]
>
> On one hand there is ROL with their dismal record of providing releases to
> paying customers in the last 2.5 years, their refusal to commit to
> supporting a large chunk of the market in the form of Iyonix users, the the
> erratic personality of its director and its resourcing problems.
Dave, do you have the number of Iyonix users to hand? I didn't realise this
was publicly available information. I didn't think the Iyonix userbase was
all that large when compared with RISC OS 4 (Risc PC era and A9) users, but
would be interested to know for sure.
[snip]
> ---druck
Cheers
Steve.
--
StevePotts at blastzone DOT demon STOP co DOT uk (www.blastzone.demon.co.uk/)
Written on RISC OS.
http://www.riscos.com/
Steve Potts
"Ams" <a...@globalcafe.ie> wrote:
[snip]
>
> This is just the *first* trench of OSS releases. Further will come
> along in due course (e.g., Kernel, various drivers etc.,).
Do you have those examples on authority, or are they assumptions? Just
checking out of interest.
[snip]
> Annraoi
Thanks.
Steve.
--
StevePotts at blastzone DOT demon STOP co DOT uk (www.blastzone.demon.co.uk/)
Written on RISC OS.
http://www.riscos.com/
--
Steve Fryatt
<d3b40d6f...@blastzone.demon.co.uk>:
> In message <1159716178.7...@m7g2000cwm.googlegroups.com>
> "Ams" <a...@globalcafe.ie> wrote:
>
> > This is just the *first* trench of OSS releases. Further will come
> > along in due course (e.g., Kernel, various drivers etc.,).
>
> Do you have those examples on authority, or are they assumptions? Just
> checking out of interest.
It's been stated on Drobe, by one of the ROOL team, that it will be
possible to build a full, saleable system using the released components.
Steve Fryatt
> In message <0b4cfe6e...@druck.freeuk.net>
> druck <ne...@druck.freeuk.com> wrote:
>
> > On one hand there is ROL with their dismal record of providing
> > releases to paying customers in the last 2.5 years, their refusal to
> > commit to supporting a large chunk of the market in the form of Iyonix
> > users, the the erratic personality of its director and its resourcing
> > problems.
>
> Dave, do you have the number of Iyonix users to hand? I didn't realise
> this was publicly available information.
I don't think it is. However, the RISC OS (developer) community being
fairly small, it's possible to gauge what others are using.
> I didn't think the Iyonix userbase was all that large when compared with
> RISC OS 4 (Risc PC era and A9) users, but would be interested to know
> for sure.
It may not be, but as I've pointed out before, denying Iyonix-owning
developers access to Select and -- more crucially -- access to the Select
documentation, makes it that much harder for those developers to support
users who have Select or Adjust. ROL seem unable to grasp this.
I've no idea if any of my stuff still runs on the latest version of
Select, let alone Adjust 32. I should probably care, but I don't want to
waste nearly 150 quid a year to get the documentation to find out; if
Select actually ran on my main machine, I would be running it.
If any A9home users want to do my testing for me... :-(
Ray Dawson
>
> If ROL can't/won't put select UI features into RO 5 then what's the
> alternative?
Do ROL have the RO 5 source to work on and add Select features to?
Ray D
Ray Dawson
>
> I've no idea if any of my stuff still runs on the latest version of
> Select, let alone Adjust 32.
But as long as any of your apps don't have RISC OS 5 specific code, why
shouldn't they work on Select or Adjust?
Ray D
david.b...@alligatagroup.co.uk
I'd say that if the stated aim of this project is completed then yes,
they will end up with the full RO5 source code to add features to. But
then so will many others.
Ams
Steve Potts wrote:
> In message <1159716178.7...@m7g2000cwm.googlegroups.com>
> "Ams" <a...@globalcafe.ie> wrote:
>
> [snip]
> >
> > This is just the *first* trench of OSS releases. Further will come
> > along in due course (e.g., Kernel, various drivers etc.,).
>
> Do you have those examples on authority, or are they assumptions? Just
> checking out of interest.
The ROOL FAQ states that they hope to release more. A lead ROOL
developer stated the same on Drobe.
No I don't have *specific* indications of what additional material will
be released (if that's what your asking) - but the intent seems pretty
clear that users will be able to build and develope RISC OS systems.
It makes for a refreshing and more inclusive alternative to Select
where even the API documentation is off-limits to non-subscribers
doesn't it ?
Regards
Annraoi
Steve Potts
Steve Fryatt <ne...@stevefryatt.org.uk> wrote:
> On 1 Oct, Steve Potts wrote in message
> <f7490d6f...@blastzone.demon.co.uk>:
>
> > In message <0b4cfe6e...@druck.freeuk.net>
> > druck <ne...@druck.freeuk.com> wrote:
> >
[snip]
> > Dave, do you have the number of Iyonix users to hand? I didn't realise
> > this was publicly available information.
>
> I don't think it is. However, the RISC OS (developer) community being
> fairly small, it's possible to gauge what others are using.
>
> > I didn't think the Iyonix userbase was all that large when compared with
> > RISC OS 4 (Risc PC era and A9) users, but would be interested to know
> > for sure.
>
> It may not be, but as I've pointed out before, denying Iyonix-owning
> developers access to Select and -- more crucially -- access to the Select
> documentation, makes it that much harder for those developers to support
> users who have Select or Adjust. ROL seem unable to grasp this.
>
> I've no idea if any of my stuff still runs on the latest version of
> Select, let alone Adjust 32. I should probably care, but I don't want to
> waste nearly 150 quid a year to get the documentation to find out; if
> Select actually ran on my main machine, I would be running it.
>
> If any A9home users want to do my testing for me... :-(
I'd be more than happy to give your stuff a try on my A9 as and when I have
time, but that's a problem, as you know, these days :-(. Will talk to you on
Wednesday evening.
Steve Potts
"Ams" <a...@globalcafe.ie> wrote:
>
> Steve Potts wrote:
> > In message <1159716178.7...@m7g2000cwm.googlegroups.com>
> > "Ams" <a...@globalcafe.ie> wrote:
> >
> > [snip]
> > >
> > > This is just the *first* trench of OSS releases. Further will come
> > > along in due course (e.g., Kernel, various drivers etc.,).
> >
> > Do you have those examples on authority, or are they assumptions? Just
> > checking out of interest.
>
> The ROOL FAQ states that they hope to release more. A lead ROOL
> developer stated the same on Drobe.
>
> No I don't have *specific* indications of what additional material will
> be released (if that's what your asking) - but the intent seems pretty
> clear that users will be able to build and develope RISC OS systems.
Thanks for clarifying.
> It makes for a refreshing and more inclusive alternative to Select
> where even the API documentation is off-limits to non-subscribers
> doesn't it ?
Sounds like it it may be (but I've always thought that the people who've
complained loudest about the documentation from ROL are those whom did
subscribe at one time or other and had therefore had the opportunity to have
access to at least early versions), but I think we need to see exactly what
happens. I'm being very cautious about the whole ROOL at the moment. I
think ROOL needs time to do what they say and others need to wait and see.
>
> Regards
>
> Annraoi
>
Cheers.
Ams
Steve Potts wrote:
<snip>
>
> Thanks for clarifying.
>
No problem.
> > It makes for a refreshing and more inclusive alternative to Select
> > where even the API documentation is off-limits to non-subscribers
> > doesn't it ?
>
> Sounds like it it may be (but I've always thought that the people who've
> complained loudest about the documentation from ROL are those whom did
> subscribe at one time or other and had therefore had the opportunity to have
> access to at least early versions), but I think we need to see exactly what
> happens. I'm being very cautious about the whole ROOL at the moment. I
> think ROOL needs time to do what they say and others need to wait and see.
Never subscribed to Select myself (on principle as I wouldn't want
something if I couldn't get documentation up front). By contrast I have
an Iyonix, purchased the C/C++ tools in advance and the Iyonix
Technical Reference *afterwards*. The difference being I could see the
Iyonix RO5 API upfront. In my case ROL shot themselves in the foot and
lost a sale.
How *can* people develope for (or at least allow for) differences
inherent to Select if they don't have documentation? Even Castle
release API info for RO5 and that's available to *everyone* not just
Iyonix purchasers.
I'd agree with you that it's necessary to see how things pan out, but
if what is being suggested happens then that *would* be a good thing
IMHO, and have no specific reason for doubting that source will be
released. If that does occur and developers take up the oppertunities
that represents then enhancements to RO 5 will follow.
It in a stroke will remove the accusation that RO 5 is not being
actively developed (an untruth anyway - but at least this will finally
kill off that notion). It will also mean that potentially (at least) UI
improvements to RO5 in the short term may happen - with more "deep"
improvements in future.
But as you suggest we will have to wait and see. My advice would be
that of cautious optimism, that usually is the best tactic !
Regards
Annraoi
Tim Marson
They may have just saved RISC OS :)
I'm happy with that.
Tim
--
VinceH
Because where an API is being used that is, at face value, the
same in different OS versions (which should be the case for
non-version specific APIs) it doesn't necessarily mean the
underlying code is the same. There could be very subtle
differences to the way values are returned from a given SWI, for
example, which could result in a program behaving differently on
different OS versions.
--
VinceH - http://www.softrock.co.uk/info/vinceh.html
Please sponsor me to abseil forwards down the Avon Gorge:
http://www.justgiving.com/vinceh/
Andrew Hodgkinson
> It's been stated on Drobe, by one of the ROOL team, that it will be
> possible to build a full, saleable system using the released components.
I'm not sure if that will be the case in the initial release. You can see
what ROOL is planning to initially offer here:
A full saleable system is certainly a position we would very much like to
end up in. However, as I've said in a comment on the ever-lengthening
Drobe article, we can't release everything in one go. We've started off
with something that can be achieved with reasonable time and resources
and hope to move from there as quickly as possible. No promises for
future components, though.
As you've hopefully gathered from how quiet we were about ROOL when it
was first noticed by the RISC OS community, we're trying very hard not to
mislead people about any aspects of our activities. We don't see any
point in making promises for things that we later fail to deliver.
BTW, ROOL .sig coming RSN ;-)
--
TTFN, Andrew Hodgkinson
Find some electronic music at: Photos, wallpaper, software and more:
http://pond.org.uk/music.html http://pond.org.uk/
Steven Pampling
I have scripts at work that should work on all the PC's in our domain,
until I'd tested them on all the OS revisions in use I wasn't aware of the
shortcomings of some of them. Of course with more knowledge of the setup of
all of the target machines it would have been easier to get things right
first time.
Steve Fryatt
<gemini.j6gtls0...@magray.freeserve.co.uk>:
Because they're written in C and linked to the Castle stubs: I use GCC,
and so the use of StubsG is at best tricky. ROL were going to be dealing
with this issue "soon" when I asked PM about it earlier this year
(probably March or April, as ISTR it was snowing heavily in the Cardiff
area the evening he came up to Yorkshire to see us). Guess what's
happened so far...
Since I've not seen /any/ documentation on CLib version numbers or
compatibility, I have no idea what happens if you try and run CashBook or
PrintPDF on an A9. They may fail to run because the CLib version numbers
are too low; they may start but then crash or behave unpredictably because
the version numbers are OK but the native library isn't fully compatible;
they may be fine. Who knows? Can the Castle SCL be softloaded onto an A9
yet?
And that's not even considering the API issues.
Ste (news)
Ray Dawson <r...@magray.freeserve.co.uk> wrote:
Over arching all of the responses you've had so far to this question is this
simple fact: ROL have changed a number of APIs and bits of system behaviour
in ways which are not backward-compatible (e.g. path variables and transient
utilities, to name but two).
These changes break existing code which has been carefully constructed to
adhere to the published and/or "well known" specifications. Unfortunately,
the ROL changes are not in the public domain so developers who aren't ROL
donors (oops, I mean Select subscribers) have no idea about these changes
until the fault reports start to roll in.
Steve
--
Steve Revill @ Home
Note: All opinions expressed herein are my own.
David Pitt
Steve Fryatt <ne...@stevefryatt.org.uk> wrote:
[snip]
>I've no idea if any of my stuff still runs on the latest version of
>Select, let alone Adjust 32. I should probably care, but I don't want
>to waste nearly 150 quid a year to get the documentation to find out;
>if Select actually ran on my main machine, I would be running it.
It would be unreasonable to expect developers to have all the variants
of our platform to hand, the hardware that is. It would be very
reasonable for the documentation to be available. (Or perhaps not if the
Select/Adjust info had been published there might be more application
software divergence than there actually is. (Must learn smileys.))
>If any A9home users want to do my testing for me... :-(
I am certainly willing to do any testing or diagnostics, and I make the
offer to all. Just ask.
The A9home Adjust32 is beta which does come up with the occasional
surprise. ROL is still working towards a final release (I hope), there
may be a bit of a wait.
--
David Pitt.
Computing with RISC OS.
Ray Dawson
> In article <gemini.j6gtls0...@magray.freeserve.co.uk>,
> Ray Dawson <r...@magray.freeserve.co.uk> wrote:
> > Steve Fryatt <ne...@stevefryatt.org.uk> wrote:
> >
> > > I've no idea if any of my stuff still runs on the latest version of
> > > Select, let alone Adjust 32.
> >
> > But as long as any of your apps don't have RISC OS 5 specific code,
> > why shouldn't they work on Select or Adjust?
>
> Over arching all of the responses you've had so far to this question is
> this simple fact: ROL have changed a number of APIs and bits of system
> behaviour in ways which are not backward-compatible (e.g. path variables
> and transient utilities, to name but two).
>
> These changes break existing code which has been carefully constructed
> to adhere to the published and/or "well known" specifications.
I have to say that none of the existing applications I use 'broke' when I
went from RISC OS 4.02 to Select to Adjust. I do vaguely remember
something with Choices: on one app, but that was fixed by changing a !Run
file.
Cheers,
Ray D
Alan P Dawes
<david.b...@alligatagroup.co.uk> wrote:
Which sidesteps the question as to whether Castle provided the source code
of ROS 5 to ROS Ltd to allow them to produce a proper ROS 5 version of
select. Considering that at the time ROS5 select was first mooted, ROS Ltd
was developing a 32 bit version of ROS which would be in competition with
Castle/tematic's version, I do wonder how much of the cooperation talked
about by Jack and Paul at their joint show appearance actually
materialised.
Alan
--
--. --. --. --. : : --- --- ----------------------------
|_| |_| | _ | | | | |_ | alan....@argonet.co.uk
| | |\ | | | | |\| | | alan....@riscos.org
| | | \ |_| |_| | | |__ | Using an Acorn RiscPC
Kevin Wells
Steve Fryatt <ne...@stevefryatt.org.uk> wrote:
>On 1 Oct, Steve Potts wrote in message
> <f7490d6f...@blastzone.demon.co.uk>:
[snip]
>If any A9home users want to do my testing for me... :-(
>
I will be pleased to.
--
Kev Wells http://kevsoft.topcities.com
http://kevsoft.co.uk/
ICQ 238580561
Live to ride Ride to live.
Steven Pampling
Alan P Dawes <alan....@argonet.co.uk> wrote:
> Considering that at the time ROS5 select was first mooted, ROS Ltd
> was developing a 32 bit version of ROS which would be in competition with
> Castle/tematic's version, I do wonder how much of the cooperation talked
> about by Jack and Paul at their joint show appearance actually
> materialised.
Now that couldn't possibly be true.
You see when RO5 Select was first mooted (outside ROL) ROL were busy
saying they had no intention of making Select 32 bit.
From all outward indications only when the need for a 32 bit version of
RISC OS for another machine (A9) came up was any work undertaken. This in
one respect is understandable in that the 32 bit work had a commercial
target in that the A9 was a non-starter without that work or a deal with
Castle for a modified version of RO5.
John Cartmell
Steven Pampling <steve.p...@dsl.pipex.com> wrote:
> In article <4e6f79f2fe...@argonet.co.uk>,
> Alan P Dawes <alan....@argonet.co.uk> wrote:
> > Considering that at the time ROS5 select was first mooted, ROS Ltd
> > was developing a 32 bit version of ROS which would be in competition with
> > Castle/tematic's version, I do wonder how much of the cooperation talked
> > about by Jack and Paul at their joint show appearance actually
> > materialised.
> Now that couldn't possibly be true.
> You see when RO5 Select was first mooted (outside ROL) ROL were busy
> saying they had no intention of making Select 32 bit.
Some confusion here?
ROL always had 32-bitting on the agenda. Work on it starting years ago when
Select first started (actually before Select was announced). What was needed
was the money to do the bulk of the work that wasn't being updated along with
the Select releases. Had someone paid for that work at the start it would have
been done at the start.
> From all outward indications only when the need for a 32 bit version of
> RISC OS for another machine (A9) came up was any work undertaken.
I'm not sure what you're meaning but think you may have this back to front.
The bulk of the 32-bitting was done because Ad6 were able to offer a deal that
would get the 32-bitting done and still allow the completed code to be
available to other hardware users. ROL wouldn't have been able to accept a
deal that allowed 32-bitting to be made available just to one hardware
developer - and presumably no-one before Ad6 were able to come up with a deal
that produced what the customer wanted *and* the 32-bitted code for everyone
else.
[and apologies to pedants for the loose use of the phrase '32-bitted']
--
John Cartmell john@ followed by finnybank.com 0845 006 8822
Qercus magazine FAX +44 (0)8700-519-527 www.finnybank.com
Qercus - the best guide to RISC OS computing
Ste (news)
diodesign <diod...@gmail.com> wrote:
> Hi,
>
> For those who may have missed it, there's views and opinions from
> developers on the news of the RISC OS 5 source code release here:
>
> http://www.drobe.co.uk/riscos/artifact1706.html
A few more entries have been added to the shared source FAQ on the RISC OS
Open web site:
http://www.riscosopen.com/ss_faq.html
Specifically:
* Why haven't you used the GPL?
* Why have you seleted these components to release first?
* What tools do I need to build these components?
Steven Pampling
<jo...@cartmell.demon.co.uk> wrote:
> In article <4e6fa21796st...@dsl.pipex.com>, Steven Pampling
> <steve.p...@dsl.pipex.com> wrote:
> > Now that couldn't possibly be true. You see when RO5 Select was first
> > mooted (outside ROL) ROL were busy saying they had no intention of
> > making Select 32 bit.
> Some confusion here? ROL always had 32-bitting on the agenda. Work on it
> starting years ago when Select first started (actually before Select was
> announced).
Since the source of the information about not having any intention of 32
bitting the source was a statement from PM you might possibly understand
any confusion (not that I think any confusion is at this end)
> What was needed was the money to do the bulk of the work
> that wasn't being updated along with the Select releases. Had someone
> paid for that work at the start it would have been done at the start.
As various people who are programmers have pointed out updating the code
for any reason would be done on a bit neutral basis and the final build 26
bit or otherwise is a matter of compiler options.
> > From all outward indications only when the need for a 32 bit version
> > of RISC OS for another machine (A9) came up was any work undertaken.
> I'm not sure what you're meaning but think you may have this back to
> front. The bulk of the 32-bitting was done because Ad6 were able to
> offer a deal that would get the 32-bitting done and still allow the
> completed code to be available to other hardware users.
In which case I don't have it back to front. ROL did the work with a
specific target which would provide money rather than a more speculative
set of work - which would work on existing machines but as people have
pointed out provide no benefit to those and break some existing
software.(of the time)
> ROL wouldn't
> have been able to accept a deal that allowed 32-bitting to be made
> available just to one hardware developer - and presumably no-one before
> Ad6 were able to come up with a deal that produced what the customer
> wanted *and* the 32-bitted code for everyone else.
As I recall the request from PM was for funding for the project but that
anyone providing the funding would be doing so on a philanthropic basis for
the rest of the community.
and as this conversation stands you seem to be trying to start an us and
them argument.
John Cartmell
Steven Pampling <steve.p...@dsl.pipex.com> wrote:
> In article <4e6fb5f...@cartmell.demon.co.uk>, John Cartmell
> <jo...@cartmell.demon.co.uk> wrote:
> > In article <4e6fa21796st...@dsl.pipex.com>, Steven Pampling
> > <steve.p...@dsl.pipex.com> wrote:
> > > Now that couldn't possibly be true. You see when RO5 Select was first
> > > mooted (outside ROL) ROL were busy saying they had no intention of
> > > making Select 32 bit.
> > Some confusion here? ROL always had 32-bitting on the agenda. Work on it
> > starting years ago when Select first started (actually before Select was
> > announced).
> Since the source of the information about not having any intention of 32
> bitting the source was a statement from PM you might possibly understand
> any confusion (not that I think any confusion is at this end)
I think there may have been a misunderstanding there. They were 32-bitting!
(but slowly).
> > What was needed was the money to do the bulk of the work
> > that wasn't being updated along with the Select releases. Had someone
> > paid for that work at the start it would have been done at the start.
> As various people who are programmers have pointed out updating the code
> for any reason would be done on a bit neutral basis and the final build 26
> bit or otherwise is a matter of compiler options.
It was done on a 'bit neutral basis' (if I understand what you are saying) and
the version for the RiscPC or 32-bit machines is produced from the same source
at the 'flick of a switch'. That means that everyone can use what is presented
to the software as the same version of the OS.
> > > From all outward indications only when the need for a 32 bit version
> > > of RISC OS for another machine (A9) came up was any work undertaken.
> > I'm not sure what you're meaning but think you may have this back to
> > front. The bulk of the 32-bitting was done because Ad6 were able to
> > offer a deal that would get the 32-bitting done and still allow the
> > completed code to be available to other hardware users.
> In which case I don't have it back to front. ROL did the work with a
> specific target which would provide money rather than a more speculative
> set of work - which would work on existing machines but as people have
> pointed out provide no benefit to those and break some existing
> software.(of the time)
Not true. The 32-bit Select works on all machines (new and
old-that-can-run-RO4) with the exception of the Iyonix. Like the A9 the Iyonix
needs work doing to bridge the gap between the hardware and the OS. For the A9
that was done by Advantage 6 (not ROL). When it's ready for the rest ROL could
complete work on the Iyonix links.
> > ROL wouldn't have been able to accept a deal that allowed 32-bitting to
> > be made available just to one hardware developer - and presumably no-one
> > before Ad6 were able to come up with a deal that produced what the
> > customer wanted *and* the 32-bitted code for everyone else.
> As I recall the request from PM was for funding for the project but that
> anyone providing the funding would be doing so on a philanthropic basis for
> the rest of the community.
Yes. That's what Ad6 has done.
> and as this conversation stands you seem to be trying to start an us and
> them argument.
No argument. The work has been done. There are criticisms that should be made
to ROL - but not the one you made above. ROL always had the intention to
32-bit (given the wherewithal) but the blocking point is that the work needed
a hardware developer who was happy to see the resultant work made available to
all. I don't know if Castle ever went to ROL with a proposal; I do know about
others and, having heard their story, can see both sides. I'm grateful that it
has been financed and the work done.
Ams
> In article <1159720265....@h48g2000cwc.googlegroups.com>,
<snip>
>
> Which sidesteps the question as to whether Castle provided the source code
> of ROS 5 to ROS Ltd to allow them to produce a proper ROS 5 version of
> select.
For anything other than *low level* hardware access they *should not
need to*. In fact most of RO5 API is well documented and publically
available - that should be *all* that is required to provide the UI
enhancements in Select under Iyonix's native RISC OS 5.
<snip>
I note that ROOL will be providing source for items like !Paint and
!Draw so it's not beyond the realm of possibility that users (without
the RO5 kernel source) may yet complete UI enhancements to !Paint and
!Draw that ROL seem unwilling or unable to provide. And if they do that
will neatly underscore the point.
Regards
Annraoi
Peter Naulls
"Ams" <a...@globalcafe.ie> wrote:
> Alan P Dawes wrote:
> > In article <1159720265....@h48g2000cwc.googlegroups.com>,
>
> <snip>
>
> >
> > Which sidesteps the question as to whether Castle provided the source code
> > of ROS 5 to ROS Ltd to allow them to produce a proper ROS 5 version of
> > select.
>
> For anything other than *low level* hardware access they *should not
> need to*. In fact most of RO5 API is well documented and publically
> available - that should be *all* that is required to provide the UI
> enhancements in Select under Iyonix's native RISC OS 5.
I wouldn't be too quick to make such judgements. RISC OS is a greatly
complex beast, and certain changes can have unexpected results.
Equally, modifications might need to go deeper than you expect.
> I note that ROOL will be providing source for items like !Paint and
> !Draw so it's not beyond the realm of possibility that users (without
> the RO5 kernel source) may yet complete UI enhancements to !Paint and
> !Draw that ROL seem unwilling or unable to provide. And if they do that
> will neatly underscore the point.
Be careful with your phrasing. ROL have already considerably enhanced
Pain and Draw, and have indeed provided them - to Adjust/Select users.
It's a shame that we need to see repetition here.
--
Peter Naulls - pe...@chocky.org | http://www.chocky.org/
----------------------------------------------------------------------------
RISC OS Community Wiki - add your own content | http://www.riscos.info/
Ams
John Cartmell wrote:
> In article <4e6fe48086st...@dsl.pipex.com>,
> Steven Pampling <steve.p...@dsl.pipex.com> wrote:
> > In article <4e6fb5f...@cartmell.demon.co.uk>, John Cartmell
> > <jo...@cartmell.demon.co.uk> wrote:
> > > In article <4e6fa21796st...@dsl.pipex.com>, Steven Pampling
> > > <steve.p...@dsl.pipex.com> wrote:
<snip>
>
> > Since the source of the information about not having any intention of 32
> > bitting the source was a statement from PM you might possibly understand
> > any confusion (not that I think any confusion is at this end)
>
> I think there may have been a misunderstanding there. They were 32-bitting!
> (but slowly).
>
Publically all that was known was that ROL had (in effect) stopped
planning to convert RISC OS to 32bit. In fact I think they initially
said there'd only be releases of an enhanced 26bit RISC OS variant to
be called RISC OS 4.5
<snip>
> When it's [32bit select] ready for the rest ROL could complete work on the Iyonix links.
I am now completely confused, what about the comment about a minimum
number of new Iyonix Select subscribers and all that - and the then the
subsequent rolling back on that ?
> ROL always had the intention to 32-bit (given the wherewithal) but the blocking point is
> that the work needed a hardware developer who was happy to see the resultant work
> made available to all.
With the ROOL announcement this make all of that somewhat moot. Large
chunks of RO5 source (all 32bit compliant) will be made available to
*all*. This means any "nice" features (usually UI) can be added by
independant developers to RO5 - ROL need do nothing.
ROL have missed the boat with respect to Iyonix (IMHO), as to ROOL the
opening of RISC OS key components to the developer community means
incremental improvements for all - surely a good thing ? Let's enjoy
this - I believe it represents another oppertunity for RISC OS to
expand and develope... and not before time too !!!!
Regards
Annraoi
Ams
Peter Naulls wrote:
> In message <1159898610.6...@k70g2000cwa.googlegroups.com>
> "Ams" <a...@globalcafe.ie> wrote:
>
> > Alan P Dawes wrote:
> > > In article <1159720265....@h48g2000cwc.googlegroups.com>,
<snip>
> > For anything other than *low level* hardware access they *should not
> > need to*. In fact most of RO5 API is well documented and publically
> > available - that should be *all* that is required to provide the UI
> > enhancements in Select under Iyonix's native RISC OS 5.
>
> I wouldn't be too quick to make such judgements. RISC OS is a greatly
> complex beast, and certain changes can have unexpected results.
> Equally, modifications might need to go deeper than you expect.
>
Given your knowledge and experience Peter I'll accept that. But surely
if some UI features *do* require such "depth" then surely that's going
to be a recurrent problem that *may* even restrict what people can do
even when ROOL source to Paint and Draw and the like become available ?
Well at least cause a bit of delay while developers "get to grip" with
the nuances and complexities of it all.
<snip>
> > I note that ROOL will be providing source for items like !Paint and
> > !Draw so it's not beyond the realm of possibility that users (without
> > the RO5 kernel source) may yet complete UI enhancements to !Paint and
> > !Draw that ROL seem unwilling or unable to provide. And if they do that
> > will neatly underscore the point.
>
> Be careful with your phrasing. ROL have already considerably enhanced
> Pain and Draw, and have indeed provided them - to Adjust/Select users.
> It's a shame that we need to see repetition here.
Sorry about that, I did not mean to infer that ROL *only* did UI
changes to those apps - although that could be read into what I said,
sorry about the repetition and lack of clarity.
Regards
Annraoi
Steven Pampling
<jo...@cartmell.demon.co.uk> wrote:
> In article <4e6fe48086st...@dsl.pipex.com>, Steven Pampling
> <steve.p...@dsl.pipex.com> wrote:
> > Since the source of the information about not having any intention of
> > 32 bitting the source was a statement from PM you might possibly
> > understand any confusion (not that I think any confusion is at this
> > end)
> I think there may have been a misunderstanding there. They were
> 32-bitting! (but slowly).
Like I said the source of the confusion was PM. Making misleading or untrue
statements hasn't anyone.
Ste (news)
Ams <a...@globalcafe.ie> wrote:
> Peter Naulls wrote:
> > In message <1159898610.6...@k70g2000cwa.googlegroups.com>
> > "Ams" <a...@globalcafe.ie> wrote:
> > > For anything other than *low level* hardware access they *should not
> > > need to*. In fact most of RO5 API is well documented and publically
> > > available - that should be *all* that is required to provide the UI
> > > enhancements in Select under Iyonix's native RISC OS 5.
> >
> > I wouldn't be too quick to make such judgements. RISC OS is a greatly
> > complex beast, and certain changes can have unexpected results.
> > Equally, modifications might need to go deeper than you expect.
>
> Given your knowledge and experience Peter I'll accept that. But surely
> if some UI features *do* require such "depth" then surely that's going
> to be a recurrent problem that *may* even restrict what people can do
> even when ROOL source to Paint and Draw and the like become available ?
> Well at least cause a bit of delay while developers "get to grip" with
> the nuances and complexities of it all.
I think Peter is talking in the general case, where every now and again you
stumble across a feature which has sneaky (often unknown to the public)
connections with some lower level of the OS.
That sort of thing is not going to be (or is highly unlikely to be) a
blocker though for things like Paint and Edit.
> > > I note that ROOL will be providing source for items like !Paint and
> > > !Draw so it's not beyond the realm of possibility that users (without
> > > the RO5 kernel source) may yet complete UI enhancements to !Paint and
> > > !Draw that ROL seem unwilling or unable to provide. And if they do that
> > > will neatly underscore the point.
> >
> > Be careful with your phrasing. ROL have already considerably enhanced
> > Pain and Draw, and have indeed provided them - to Adjust/Select users.
> > It's a shame that we need to see repetition here.
>
> Sorry about that, I did not mean to infer that ROL *only* did UI
> changes to those apps - although that could be read into what I said,
> sorry about the repetition and lack of clarity.
In some respects I agree that we should avoid putting development time into
performing work which has already been done but on the other hand, the
improvements in the ROL versions of tools like Draw and Paint /haven't/ been
done for the Iyonix and I don't see them moving over any time soon.
If some newish developer should decide to cut their teeth by adding new and
interesting features to these apps, which will then be available to all, I
wouldn't complain.
Alan Wrigley
Peter Naulls <pe...@chocky.org> wrote:
> ROL have already considerably enhanced
> Pain
So a lot of people have said... ;)
Alan
--
RISC OS - you know it makes cents
Ams
Ste (news) wrote:
> In article <1159899907.5...@k70g2000cwa.googlegroups.com>,
> Ams <a...@globalcafe.ie> wrote:
>
<snip>
>
> I think Peter is talking in the general case, where every now and again you
> stumble across a feature which has sneaky (often unknown to the public)
> connections with some lower level of the OS.
>
Ok, that makes sense, thanks for the clarification Steve.
> That sort of thing is not going to be (or is highly unlikely to be) a
> blocker though for things like Paint and Edit.
>
Ok.
<snip>
>
> In some respects I agree that we should avoid putting development time into
> performing work which has already been done but on the other hand, the
> improvements in the ROL versions of tools like Draw and Paint /haven't/ been
> done for the Iyonix and I don't see them moving over any time soon.
>
A pity.
> If some newish developer should decide to cut their teeth by adding new and
> interesting features to these apps, which will then be available to all, I
> wouldn't complain.
Nor I ;-)
Thanks and regards
Annraoi
druck
> Publically all that was known was that ROL had (in effect) stopped
> planning to convert RISC OS to 32bit. In fact I think they initially
> said there'd only be releases of an enhanced 26bit RISC OS variant to
> be called RISC OS 4.5
At the 2000 ROL developers conference they were still planning to 32bit the
whole OS, but in the two years following they obviously decided that the
resources available as the result of the Select scheme would not allow this,
stating that a full 32bit conversion would take many hundreds of man hours,
and quoting the number of developers working on the 32bit Pace version. At
that point there was then a lot of speculation as to whether the Pace version
could be licensed, but sources both inside and outside Pace stated that the
management would not entertain the idea at that time, as the revenue would be
insignificant compared to the volume of STB shipments.
This is why it came as a complete shock to me when I meet up with Jack and
John in 2002 and they revealed the plans to make the Iyonix based on a 32bit
only processor. "What about the OS?" was my first question, and it was a
surprise that they had licensed the Pace version, rather commission ROL to
produce it (who at the time were still delivering regular Select releases and
everyone was happy with them). Luckily for us a down turn in STB business
made Pace look to capitalise on the assets they held, and Castle managed to
get access to a completed 32bit OS, at least 18 months earlier than ROL could
have delivered it.
> With the ROOL announcement this make all of that somewhat moot. Large
> chunks of RO5 source (all 32bit compliant) will be made available to
> *all*. This means any "nice" features (usually UI) can be added by
> independant developers to RO5 - ROL need do nothing.
ROL can go one of three ways here:-
1) They can ignore it and carry on as present releasing nothing and shedding
select subscribers like autumn leaves, as PM's statement on drobe implies.
But I hope this was rash statement, and he reconsiders this position, or
its just matter of time before they quietly cease to exist.
2) They just take additional RO5 features such as as unicode font manager
and add it to Select. If they don't do it, someone else will port it
anyway, and at least then it the unicode API becomes universally
available, and more programs can take advantage of it. But it doesn't
really move things further forward.
3) ROL fully embrace the RO5 source release and integrate Select features in
to it, producing variants which will run on all machines including the
Iyonix. They should also become the official distribution for enhancements
made by 3rd party open source programmers, packaging the results in to a
tested and easy to an easy to install single ROM image, for people who
would be wary of installing numerous separate patches.
> ROL have missed the boat with respect to Iyonix (IMHO), as to ROOL the
> opening of RISC OS key components to the developer community means
> incremental improvements for all - surely a good thing ? Let's enjoy
> this - I believe it represents another oppertunity for RISC OS to
> expand and develope... and not before time too !!!!
Its does, if only PM has the vision to see the way forward. I've been very
critical of ROL in the past 2 1/2 years, and I will continue to be so while
they treat the few remaining supporters of this platform so badly and give
them even more reason to jump ship, than there is with all other things
consider. However, all ROL has do is demonstrate the commitment to the
platform that option 3 above would entail, and I will again back them to the
hilt as I did during the early good years of Select. But its releases that
count and not words if they want my support.
---druck
--
The ARM Club Free Software - http://www.armclub.org.uk/free/
The 32bit Conversions Page - http://www.quantumsoft.co.uk/druck/
John Cartmell
<a...@globalcafe.ie> wrote:
> John Cartmell wrote:
> > In article <4e6fe48086st...@dsl.pipex.com>, Steven Pampling
> > <steve.p...@dsl.pipex.com> wrote:
> > > In article <4e6fb5f...@cartmell.demon.co.uk>, John Cartmell
> > > <jo...@cartmell.demon.co.uk> wrote:
> > > > In article <4e6fa21796st...@dsl.pipex.com>, Steven Pampling
> > > > <steve.p...@dsl.pipex.com> wrote:
> <snip>
> > > Since the source of the information about not having any intention of
> > > 32 bitting the source was a statement from PM you might possibly
> > > understand any confusion (not that I think any confusion is at this end)
> >
> > I think there may have been a misunderstanding there. They were
> > 32-bitting! (but slowly).
> Publically all that was known was ...
[Snip]
We're at cross-purposes. I'm talking about 6 years ago or more when Select was
first announced.
And I've no intention on commenting on the rest that is just wishful thinking.
John Cartmell
Steven Pampling <steve.p...@dsl.pipex.com> wrote:
> In article <4e6fed3...@cartmell.demon.co.uk>, John Cartmell
> <jo...@cartmell.demon.co.uk> wrote:
> > In article <4e6fe48086st...@dsl.pipex.com>, Steven Pampling
> > <steve.p...@dsl.pipex.com> wrote:
> > > Since the source of the information about not having any intention of
> > > 32 bitting the source was a statement from PM you might possibly
> > > understand any confusion (not that I think any confusion is at this
> > > end)
> > I think there may have been a misunderstanding there. They were
> > 32-bitting! (but slowly).
> Like I said the source of the confusion was PM. Making misleading or untrue
> statements hasn't anyone.
From recent experience I have to treat that opinion with caution. In any case
I don't think there is any serious doubt about what was happening as it was
publicly announced right at the start of the Select scheme.
Doug Webb
druck <ne...@druck.freeuk.com> wrote:
[snip]
As a long time select subscriber I to wish to see ROL grasp the
initiative.
I have backed them with my cash since the start and despite what some
people have stated I think they have done a great job with the
enhancements given the constraints they have been under. However, I
was disappointed with ROL's attributed comments on Drobe and feel that
if this is the case then with no Select for Iyonix forthcoming I may
have to look again at my subscription which is a shame but keeping us
in the dark about what they were doing re-32bit/A9 didn't help their
cause. However as we don't know what NDA they may have been under I
will reserve judgement on that.
I agree that Option 3 seems the way to go and would make sense for a
revenue stream as I think a lot of people who didn't join the Select
scheme but purchased Adjust would testify to.
If ROL did go for option 3 and did commit to Iyonix then I'd be
certain to retain my subscription but I would want some more
assurances than the seemingly moving goal posts of the past.
If ROL go for Option 2 then this may be the final nail in the Select
for Iyonix saga as that way may only see RISCPC and A9 users getting
the features in RISCOS5 that are lacking in RISCOS4/Adjust and ROL
would further marginalise the Iyonix base.
Perhaps the SE Show will see some positive words, who knows.
Welcome to another RISCOS will they/won't they saga
--
Using a Iyonix PC 512Mb, RISC OS5.11, !Dialup & !MessengerPro.
All based on superior UK technology
Steven Pampling
<steve.p...@dsl.pipex.com> wrote:
> In article <4e6fed3...@cartmell.demon.co.uk>, John Cartmell
> <jo...@cartmell.demon.co.uk> wrote:
> > In article <4e6fe48086st...@dsl.pipex.com>, Steven Pampling
> > <steve.p...@dsl.pipex.com> wrote:
> > > Since the source of the information about not having any intention
> > > of 32 bitting the source was a statement from PM you might possibly
> > > understand any confusion (not that I think any confusion is at this
> > > end)
> > I think there may have been a misunderstanding there. They were
> > 32-bitting! (but slowly).
> Like I said the source of the confusion was PM. Making misleading or
> untrue statements hasn't anyone.
Or in English "hasn't helped anyone"...
Steven Pampling
<jo...@cartmell.demon.co.uk> wrote:
> In article <4e70221ef1st...@dsl.pipex.com>, Steven Pampling
> <steve.p...@dsl.pipex.com> wrote:
> > In article <4e6fed3...@cartmell.demon.co.uk>, John Cartmell
> > <jo...@cartmell.demon.co.uk> wrote:
> > > In article <4e6fe48086st...@dsl.pipex.com>, Steven Pampling
> > > <steve.p...@dsl.pipex.com> wrote:
> > > > Since the source of the information about not having any intention
> > > > of 32 bitting the source was a statement from PM you might
> > > > possibly understand any confusion (not that I think any confusion
> > > > is at this end)
> > > I think there may have been a misunderstanding there. They were
> > > 32-bitting! (but slowly).
> > Like I said the source of the confusion was PM. Making misleading or
> > untrue statements hasn't anyone.
> From recent experience I have to treat that opinion with caution.
Ah, I take it you're sulking because I won't answer e-mails.
> In any
> case I don't think there is any serious doubt about what was happening
> as it was publicly announced right at the start of the Select scheme.
The "no 32 bit" statement? You could be right about the timing of that, I
don't actually recall precisely.
John Cartmell
Steven Pampling <steve.p...@dsl.pipex.com> wrote:
> > From recent experience I have to treat that opinion with caution.
> Ah, I take it you're sulking because I won't answer e-mails.
Sulk?
I'm responding reasonably to reasonable comments from you. I'm still waiting
for an apology for a disgusting tirade that you unleashed after being
'affronted' by an entirely innocuous comment of mine. I still think that Paul
B. presents his Christian message in such a way that it isn't forced upon his
readers - and cannot understand why saying so causes your blood pressure to
reach boiling point.
Steven Pampling
<jo...@cartmell.demon.co.uk> wrote:
> In article <4e703a9cabst...@dsl.pipex.com>, Steven Pampling
> <steve.p...@dsl.pipex.com> wrote:
> > > From recent experience I have to treat that opinion with caution.
> > Ah, I take it you're sulking because I won't answer e-mails.
> Sulk? I'm responding reasonably to reasonable comments from you. I'm
> still waiting for an apology for a disgusting tirade that you unleashed
> after being 'affronted' by an entirely innocuous comment of mine. I
> still think that Paul B. presents his Christian message in such a way
> that it isn't forced upon his readers - and cannot understand why saying
> so causes your blood pressure to reach boiling point.
Tirade it wasn't.
I still think that in the context of the thread your phraseology was at
best lacking in tact.
Ams
> On 3 Oct 2006 "Ams" <a...@globalcafe.ie> wrote:
> > Publically all that was known was that ROL had (in effect) stopped
> > planning to convert RISC OS to 32bit. In fact I think they initially
> > said there'd only be releases of an enhanced 26bit RISC OS variant to
> > be called RISC OS 4.5
>
> At the 2000 ROL developers conference they were still planning to 32bit the
> whole OS, but in the two years following they obviously decided that the
> resources available as the result of the Select scheme would not allow this,
> stating that a full 32bit conversion would take many hundreds of man hours,
> and quoting the number of developers working on the 32bit Pace version.
Yep I remember that too, I know when I attended the show in Epson in
2000 ROL were upbeat about 32bit - the next time I visited (2002) the
mutterings were about 26bit only and a "RISC OS 4.5" whatever that
was.... your explanation tallies with that.
<snip>
> This is why it came as a complete shock to me when I meet up with Jack and
> John in 2002 and they revealed the plans to make the Iyonix based on a 32bit
> only processor. "What about the OS?" was my first question, and it was a
> surprise that they had licensed the Pace version, rather commission ROL to
> produce it (who at the time were still delivering regular Select releases and
> everyone was happy with them). Luckily for us a down turn in STB business
> made Pace look to capitalise on the assets they held, and Castle managed to
> get access to a completed 32bit OS, at least 18 months earlier than ROL could
> have delivered it.
It might have been *nice* if Castle and ROL had managed to come to some
agreement to use RISC OS jointly at that point as it may have saved
much later hassle (and yes I know that the chance of that was probably
the same as a member of the porcine family being presented a pilot's
license ;-).
<snip>
>
> ROL can go one of three ways here:-
>
<snip first two which I agree with>
> 3) ROL fully embrace the RO5 source release and integrate Select features in
> to it, producing variants which will run on all machines including the
> Iyonix. They should also become the official distribution for enhancements
> made by 3rd party open source programmers, packaging the results in to a
> tested and easy to an easy to install single ROM image, for people who
> would be wary of installing numerous separate patches.
>
I'd agree on that one too - ROL being a sort of "Red Hat" of the RISC
OS world might well be an option they should consider - it would give
them a larger number of customers, could allow them *still* to build on
their base developments (while getting an income for the broadest
market possible) and overtime perhaps even feed some RISC OS Select
features into RO5 and get in turn some improvements backported from RO
5 and open source developed based on RO 5.
I am not sure how the RISC OS world will look after all these changes
what with ROL, ROOL and Castle all sharing duties. To say times might
be interesting would be an understatement IMHO.
<snip>
Thanks for your considered reply Dave.
Regards
Annraoi
Malcolm Ripley
"diodesign" <diod...@gmail.com> wrote:
> Hi,
>
> For those who may have missed it, there's views and opinions from
> developers on the news of the RISC OS 5 source code release here:
>
I found the cascade of replies to my comments earlier very telling. I
remember distinctly who said what, why, predictions of the future etc
way back when the Iyonix first appeared.....
Bottom line : so frigging what. The RISCOS market is very small and
fragmented, split down the middle. The majority on the Select/Adjust
side.
All this talk about open sourcing one side of the fork fixing the
problems is pie in the sky. As likely as the predictions for the
RISCOS market when Iyonix was released!
Developers need ONE OS to develop against. Some developers see the sun
shining out of Jack Lillignstons rear end, others see it appearing out
of Paul Middletons rear end. Both of these groups need a dog and a
white stick. NO EXCUSES (you know who I am targetting that comment
at!).
Get a grip everyone, step1 : ONE OS, FULL STOP.
regards,
Malcolm
--
Malcolm Ripley uses a RiscOS StrongArm computer.
Current new sender keyword is : terrier (use this word in the
subject line if you are contacting me for the first time)
Email address is : "mripley" not forgetting: "@ukonline.co.uk"
--
Posted via a free Usenet account from http://www.teranews.com
druck
> Developers need ONE OS to develop against.
Developers might like a single OS to develop against, but they've never had
it. We've always had to support a range of operating systems with varying
features sets, running on the current and previous couple of generations of
machines. Users wouldn't like it if we said we are only supporting RISC OS
X.YZ, and you have to buy the particular machine it runs on just to make our
life easier.
John Cartmell
Steven Pampling <steve.p...@dsl.pipex.com> wrote:
> I still think that in the context of the thread your phraseology was at
> best lacking in tact.
That says plenty about what you were thinking. Nothing about what I said or
meant.
Steven Pampling
<jo...@cartmell.demon.co.uk> wrote:
> In article <4e709fe038st...@dsl.pipex.com>, Steven Pampling
> <steve.p...@dsl.pipex.com> wrote:
> > I still think that in the context of the thread your phraseology was
> > at best lacking in tact.
> That says plenty about what you were thinking. Nothing about what I said
> or meant.
Please do explain what you were saying and meant.
Ned Abell
druck <ne...@druck.freeuk.com> wrote:
> On 4 Oct 2006 Malcolm Ripley <mri...@ukonline.co.uk> wrote:
> > Developers need ONE OS to develop against.
> Developers might like a single OS to develop against, but
> they've never had it. We've always had to support a range
> of operating systems with varying features sets, running on
> the current and previous couple of generations of machines.
> Users wouldn't like it if we said we are only supporting
> RISC OS X.YZ, and you have to buy the particular machine it
> runs on just to make our life easier.
I think most users will be a bit more pragmatic than that.
I see the single OS as the way forward and from memory we took
a leap forward when Authur was dropped. Now is the time to do
this with non 32bit OS's.
There will be those who decide to not upgrade but I suspect
these people will be content to carry on with software that
does the job for them.
For the others of us and for future new users we need to
minimise the development time bringing new products to market
to make the OS capable of competing with newer trends and
fulfilling future computing requirements.
So the announcement about open sourcing creates a pivital
point and one which I for one am very happy about if it brings
movement towards a single OS allowing hardware manufacturers
to develop around that and to expand the market.
So roll on OS6 and leave the baggage behind!
--
besters
Ned
(this email address is unused)
John Cartmell
Steven Pampling <steve.p...@dsl.pipex.com> wrote:
> In article <4e70b8d...@cartmell.demon.co.uk>, John Cartmell
> <jo...@cartmell.demon.co.uk> wrote:
> > In article <4e709fe038st...@dsl.pipex.com>, Steven Pampling
> > <steve.p...@dsl.pipex.com> wrote:
> > > I still think that in the context of the thread your phraseology was
> > > at best lacking in tact.
> > That says plenty about what you were thinking. Nothing about what I said
> > or meant.
> Please do explain what you were saying and meant.
Sorry Steve. Too many assumptions.
What you perceive is conditioned by a great many things - not just what
impinges on your senses. In this case your perception included your own
thoughts and feelings prior to what you read of mine. For some reason you read
an innocuous comment from me as (presumably) anti Paul B. and/or
anti-Christian when an objective observer would have (rightly) seen no such
thing. It was whatever was going around your head at the time was responsible
for your response - not what I wrote. People would normally shrug this off as
a 'misunderstanding' and lightly accept it as such. That's OK by me - but I'm
still waiting for the key "Sorry, I read that wrong" comment from you.
Responding nastily and then refusing to comment further is OK for people whose
comments one considers to be of no worth - but that 's not you.
John Cartmell
Ned Abell <ne...@weatherpost.org.uk> wrote:
> So the announcement about open sourcing creates a pivital
> point and one which I for one am very happy about if it brings
> movement towards a single OS allowing hardware manufacturers
> to develop around that and to expand the market.
I find myself agreeing with that statement - but do note your 'if' clause. No
one has shown how open source could "bring about a single OS", could "allow
hardware manufacturers to develop" or could "expand the market".
There is already a move towards a single OS, that would allow manufacturers to
develop around that - and perhaps expand the market. That move is about to
release its version of the OS: Select 4. Like the 'open' version it will be
available for hardware developers at a cost. Unlike the 'open' version it will
be available from a company that doesn't develop its own hardware - but has
all the RISC OS hardware developers - and retailers &c - as its shareholders.
Why is it a good thing to begin work on a second OS that will, by definition,
move *away* from a "single OS" and leave that OS in the hands of a hardware
manufacturer who might reasonably not want to licence it to competitors -
hence not "allowing hardware manufacturers to develop ..."?
[NB I realise there are many problems my comments ignore - far too many to
cover in one thread]
--
John Cartmell
Martin Wuerthner
Ned Abell <ne...@weatherpost.org.uk> wrote:
> In article <5997b770...@druck.freeuk.net>,
> druck <ne...@druck.freeuk.com> wrote:
>> On 4 Oct 2006 Malcolm Ripley <mri...@ukonline.co.uk> wrote:
>> > Developers need ONE OS to develop against.
>
>> Developers might like a single OS to develop against, but
>> they've never had it. We've always had to support a range
>> of operating systems with varying features sets, running on
>> the current and previous couple of generations of machines.
>> Users wouldn't like it if we said we are only supporting
>> RISC OS X.YZ, and you have to buy the particular machine it
>> runs on just to make our life easier.
>
> I think most users will be a bit more pragmatic than that.
>
> I see the single OS as the way forward and from memory we took
> a leap forward when Authur was dropped. Now is the time to do
> this with non 32bit OS's.
Sorry, but that does not make the slightest bit of sense. We have got
many problems but the distinction 32-bit OS vs. 26-bit OS is not one
of them. The split between OS versions that we witness is NOT a split
between 26-bit and 32-bit, which is nicely exemplified by the fact
that Adjust comes in 26-bit and 32-bit variants. A 32-bit clean
program does not notice any difference between these variants.
Martin
--
---------------------------------------------------------------------
Martin Wuerthner MW Software http://www.mw-software.com/
ArtWorks 2 -- Designing stunning graphics has never been easier
spam...@mw-software.com [replace "spamtrap" by "info" to reply]
Ned Abell
Im sorry I cant let you get away with this. Im looking at the
potted CVs of the members of ROOL, a wide range of experience
across the industry with a good mix of technical and marketing
abillity. If anyone can make a purse out of the sorry sows ear
we have been left with then I think they can and I fully
support their efforts. The fact that they have that abillity
allows them to act as honest brokers.
And you seem to suggest they have ulterior motives.
The bit of my post you chose to snip was about "leaving the
baggage behind".
Lets please do that.
Martin Wuerthner
Malcolm Ripley <mri...@ukonline.co.uk> wrote:
> Developers need ONE OS to develop against.
Ah, so you must have been really annoyed when ROL released RISC OS 4
(oh, no, a new version of the OS) or the various iterations of Select
(oh, no, more different new versions of the OS). Not to mention the
time when Acorn still developed the OS and released various of
different versions of the OS.
We have never had ONE version of the OS to develop against and while
the OS is being developed at all there will never be such a situation,
nor do we actually need it. This is completely independent of whether
there is only one party developing the OS or many and whether there is
only strand of development or many. Of course, supporting different
versions of the OS causes effort, but this is what application
software development is like on all platforms.
All we need is a common base API, which we have. For instance, any
developer can develop against the RISC OS 3.6 API, which seems to be
the least common denominator at the moment and which is supported by
all later variants of the OS. Alternatively, the RISC OS 4 API.
Even if only a single company (say, ROL) developed the OS programmers
would still have to decide whether to develop against RO 3.5/ 3.6/
3.7/ 4.0/ Select1/ Select2/ Select3/ Adjust/ etc., so does it matter
whether there is one version more or less on the list?
I am really tired of this ONE OS discussion. It is completely
pointless. A lot of fuss about nothing.
Ray Dawson
> In article <4e70ea23e5st...@dsl.pipex.com>,
> Steven Pampling <steve.p...@dsl.pipex.com> wrote:
> > In article <4e70b8d...@cartmell.demon.co.uk>, John Cartmell
> > <jo...@cartmell.demon.co.uk> wrote:
> > > In article <4e709fe038st...@dsl.pipex.com>, Steven Pampling
> > > <steve.p...@dsl.pipex.com> wrote:
> > > > I still think that in the context of the thread your phraseology
> > > > was at best lacking in tact.
>
> > > That says plenty about what you were thinking. Nothing about what I
> > > said or meant.
>
> > Please do explain what you were saying and meant.
>
> Sorry Steve. Too many assumptions.
>
[snip]
You really should have breakfast before you take your medication.
Cheers,
Ray D
Ray Dawson
>
> With the launch of ROOL I think there is renewed optimism in the market
> and again, I'd like to echo the sentiments above - that the people
> behind ROOL are certainly the ones who can help spearhead RISC OS into
> the 21st century if anyone can.
With what user base? Existing, expanded niche, or general public?
> I'm supporting them!
I'm sure we all are.
Cheers,
Ray D
mark.NOSPA...@ukonline.co.uk
Paul Vigay <invalid-em...@invalid-domain.co.uk>
wrote:
> There is nothing from Select/Adjust that I miss, and a couple of things I
> did like, I've implemented myself under RISC OS 5 - so I don't have any
> need for Select now - and speaking to others, quite a few people feel that
> way.
>
The Image filing system was good and the improvements to !Draw and
!Paint were worth having.
Select features would still be welcome under RISC OS 5 along with the
'outward' appearance of a single OS from the users perspective. At
the moment its irritating having 2 different versions with mutually
exclusive features.
I still hope ROL will come to the party...
> With the launch of ROOL I think there is renewed optimism in the market and
> again, I'd like to echo the sentiments above - that the people behind ROOL
> are certainly the ones who can help spearhead RISC OS into the 21st century
> if anyone can.
Movement and optimism - great to have them back!
RISC OS is still a really pleasurable OS to use for many tasks.
MArkee
John Cartmell
Ray Dawson <r...@magray.freeserve.co.uk> wrote:
> You really should have breakfast before you take your medication.
If I had taken time over a late breakfast I might have noticed the article in
today's New Scientist earlier. Confabulation is a normal part of cognitive
processes but can be a neurological deficiency. In your case I'll take the
older diagnosis.
I eagerly anticipate your reply* once you have read the article. ;-)
*but best make that somewhere other than csa misc
george
Paul Vigay <invalid-em...@invalid-domain.co.uk>
wrote:
> In article <4e7104c...@weatherpost.org.uk>,
> Ned Abell <ne...@weatherpost.org.uk> wrote:
>
>> Im sorry I cant let you get away with this. Im looking at the potted CVs
>> of the members of ROOL, a wide range of experience across the industry
>> with a good mix of technical and marketing abillity. If anyone can make a
>> purse out of the sorry sows ear we have been left with then I think they
>> can and I fully support their efforts. The fact that they have that
>> abillity allows them to act as honest brokers.
>
> Hear hear!
Seconded and thirded!
>
> The people behind ROOL are some of the very best RISC OS programmers in the
> world, having worked on RISC OS since the beginning. I would trust their
> code over RISCOS Ltd's code any day.
I find it encouraging that people of such experience and ability are
still committed to the development of RISC OS.
>
> Plus, it looks like they're actually doing something to help the users.
>
> I'm sorry but RISCOS Ltd have missed the boat now. I don't believe they
> ever wanted to produce Select for the Iyonix.
[Snip]
I don't think so either. Paul Middleton has always maintained that
Iyonix users were a minority of the market, but the point that seems
to have escaped his notice is that, even if that's true, Iyonix users
have /ipso facto/ invested serious cash in the platform compared to
legacy machine users, and therefore comprise a marketing opportunity
out of proportion to their numbers. But you're right, things have
moved on now.
>
> There is nothing from Select/Adjust that I miss, and a couple of things I
> did like, I've implemented myself under RISC OS 5 - so I don't have any
> need for Select now - and speaking to others, quite a few people feel that
> way.
I got as far as Select 4.37, then I sold my Kinetic RPC and that was
that. TBH it always felt a bit like tuning up an old car: fun, but
essentially pointless, when far better performance could be got from
newer hardware.
George
--
Steven Pampling
<jo...@cartmell.demon.co.uk> wrote:
> In article <gemini.j6nw9i0...@magray.freeserve.co.uk>, Ray
> Dawson <r...@magray.freeserve.co.uk> wrote:
> > You really should have breakfast before you take your medication.
> If I had taken time over a late breakfast I might have noticed the
> article in today's New Scientist earlier. Confabulation is a normal part
> of cognitive processes but can be a neurological deficiency. In your
> case I'll take the older diagnosis.
It's OK Ray, that wasn't an insult either.
If you take it as such then you were mistaken.
VinceH
[...]
> There is already a move towards a single OS, that would allow
> manufacturers to develop around that - and perhaps expand the
> market. That move is about to release its version of the OS:
> Select 4.
So you have inside knowledge that there's definitely going to be a
version for Iyonix, then? No? Thought not.
--
VinceH - http://www.softrock.co.uk/info/vinceh.html
Please sponsor me to abseil forwards down the Avon Gorge:
http://www.justgiving.com/vinceh/
Eddie Lord
george <george.g...@tiscali.co.uk> wrote thus:
[snip]
>>
>> There is nothing from Select/Adjust that I miss, and a couple of things I
>> did like, I've implemented myself under RISC OS 5 - so I don't have any
>> need for Select now - and speaking to others, quite a few people feel that
>> way.
>
> I got as far as Select 4.37, then I sold my Kinetic RPC and that was
> that. TBH it always felt a bit like tuning up an old car: fun, but
> essentially pointless, when far better performance could be got from
> newer hardware.
Very nicely put, if I may say so.
Anyone want to buy an old Kinetic RPc - I have one under the stairs.
All my money has gone to Castle, not to mention Martin (Artworks and
TechWriter:-)
--
Best Regards
Eddie
Mike Clark
george <george.g...@tiscali.co.uk> wrote:
> In message <4e71097042inval...@invalid-domain.co.uk>
> Paul Vigay <invalid-em...@invalid-domain.co.uk>
> wrote:
[snip]
> >
> > The people behind ROOL are some of the very best RISC OS programmers
> > in the world, having worked on RISC OS since the beginning. I would
> > trust their code over RISCOS Ltd's code any day.
>
> I find it encouraging that people of such experience and ability are
> still committed to the development of RISC OS.
>
What are the motivating factors though? Is it based on altruism? If
there are commercial interests whose interests are best served? Where is
the (anticipated?) cash flow coming from, and where is it going too?
Don't forget that a lot of ex-acorn senior staff, executives and
shareholders have already made a lot of money out of the assetts of
Acorn, and perhaps some think that there is still potential to make a
bit more.
> >
> > Plus, it looks like they're actually doing something to help the
> > users.
> >
> > I'm sorry but RISCOS Ltd have missed the boat now. I don't believe they
> > ever wanted to produce Select for the Iyonix.
>
> [Snip]
>
> I don't think so either. Paul Middleton has always maintained that
> Iyonix users were a minority of the market, but the point that seems
> to have escaped his notice is that, even if that's true, Iyonix users
> have /ipso facto/ invested serious cash in the platform compared to
> legacy machine users, and therefore comprise a marketing opportunity
> out of proportion to their numbers. But you're right, things have
> moved on now.
>
There's cash expenditure on hardware, cash expenditure on software and
cash expenditure on peripherals and ancillaries.
Over the years I've spent far more money on software, peripherals and
ancillaries, than I have on the basic hardware. I've got business
accounts that detail my expenditure. It runs into the 10s of thousands
of pounds spent with companies and businesses associated with the RISC
OS community.
I don't have an Iyonix and I still haven't seen an overwhelming case
that would encourage me to buy one. It could never (at least in the
foreseeable future) replace all my requirements for software so I'd
still definitely need to have space for a PC. I can save physical space
and yet have access to all the RISC OS applications that I use by either
using VirtualRiscPC and/or having an A9.
Any developer who stops supporting their software running on Virtual
RiscPC or the A9 is going to be cutting off a significant source of
revenue. So that means that the majority of developers are likely to
continue to produce applications that people like me will continue to
pay money for.
Because the Iyonix is a full sized desktop machine you've got to
convince people that they want to dedicate the desktop space (shelf or
floor) to it. It not only needs to provide a few killer apps, but also
not lack ANY necessities. A PC/A9 combination, or a PC/VirtualRiscPC
combination fills a big gap that the Iyonix alone just doesn't, yet
occupies about the same amount of space.
I've yet to see any argument that convinces me that Pseudo Open Sourcing
RISC OS 5 addresses and counters the reasons that large numbers of
people are voting with their wallets and running VirtualRiscPC on a PC
or buying an A9 to sit alongside a PC.
Whilst the OS is a nice feature of RISC OS, it is after all the
applications that run on the OS that make it useful for most users and I
don't see that this latest move is going to change or improve the
immediate availability of applications. Indeed it might even distract
some developers away from time spent on applications into time spent on
the OS (including the basic bundled applications). How does that make
things better?
Of course the motive behind Pseudo Open Sourcing of RISC OS 5 may have
nothing (or little) to do with the marketing potential in the desktop
market but may be driven by initiatives in other potential markets (such
as set top boxes?).
Mike
--
o/ \\ // |\ ,_ o Mike Clark
<\__,\\ // __o | \ / /\, "A mountain climbing, cycling, skiing,
"> || _`\<,_ |__\ \> | immunology lecturer, antibody engineer and
` || (_)/ (_) | \corn computer user"
News poster
Mike Clark <mr...@cam.ac.uk> wrote:
> In message <b5102371...@tiscali.co.uk>
> george <george.g...@tiscali.co.uk> wrote:
>
> > In message <4e71097042inval...@invalid-domain.co.uk>
> > Paul Vigay <invalid-em...@invalid-domain.co.uk>
> > wrote:
> [snip]
> > >
> > > The people behind ROOL are some of the very best RISC OS programmers
> > > in the world, having worked on RISC OS since the beginning. I would
> > > trust their code over RISCOS Ltd's code any day.
> >
> > I find it encouraging that people of such experience and ability are
> > still committed to the development of RISC OS.
> >
>
> What are the motivating factors though? Is it based on altruism? If
> there are commercial interests whose interests are best served? Where is
> the (anticipated?) cash flow coming from, and where is it going too?
Not quite sure what the relevance of this is. The same questions could
easily be applied to ROL.
>
> Don't forget that a lot of ex-acorn senior staff, executives and
> shareholders have already made a lot of money out of the assetts of
> Acorn, and perhaps some think that there is still potential to make a
> bit more.
...and? ROL has no interest in extracting money from their RISC OS IP?
<snip>
Regards
Stan
--
Ams
Mike Clark wrote:
> In message <b5102371...@tiscali.co.uk>
<snip>
>
> What are the motivating factors though? Is it based on altruism? If
> there are commercial interests whose interests are best served? Where is
> the (anticipated?) cash flow coming from, and where is it going too?
>
Well the open sourcing is, theoretically *free*. Code can be shared
freely, save for an agreement of the user not to commercially exploit
it without paying a license fee. And sure what's wrong with that.
Maybe the cash flow will be used to fund a dual core 1.2GHz version of
the Iyonix - who knows - if so *cool*.
> Don't forget that a lot of ex-acorn senior staff, executives and
> shareholders have already made a lot of money out of the assetts of
> Acorn, and perhaps some think that there is still potential to make a
> bit more.
>
And now the headlines "Company attempts to make a profit. Prime
Minister is shocked and dismayed - more at 11".
And we all know that Select is free and you get free A9's with Coco
Pops ;-).
<snip>
>
> There's cash expenditure on hardware, cash expenditure on software and
> cash expenditure on peripherals and ancillaries.
>
> Over the years I've spent far more money on software, peripherals and
> ancillaries, than I have on the basic hardware.
Well as software on the RISC OS platform is usually somewhat cheaper
than the PC world that wouldn't be so much then ;). Peripherals - well
very few are *actually* made by *any* vendors in RISC OS land - so
again *very little* of that money stays within the RISC OS vendor
community. If you buy (say) a Canon printer the bulk of the money goes
to Canon - a small slice goes to CJE (for example).
In comparison if you buy an A9 or Iyonix the hardware has a larger
proportion of its value remain *with* the RISC OS vendor.
> I've got business accounts that detail my expenditure. It runs into the 10s of thousands
> of pounds spent with companies and businesses associated with the RISC
> OS community.
Again how much of that *stays* depends on the added value those
companies provide. Anything Windows based you can take as meaning the
*bulk* of the value goes to the OEM (someone outside the RISC OS world
who actually built the box you're using), Microsoft (for their OS that
runs VRPC), some small amount to VA and a slightly bigger chunk to
ROL.
>
> I don't have an Iyonix and I still haven't seen an overwhelming case
> that would encourage me to buy one.
Depends on your criteria I suppose. Can't be cost (you've spent in your
own words 10's of thousands), so I imagine it's for other reasons.
> It [Iyonix] could never (at least in the foreseeable future) replace all my requirements for
> software so I'd still definitely need to have space for a PC.
So the volumetric capacity of Iyonix is a "problem" ? You *do* know
they have a small footprint version (a cube box). Ok not as small as an
A9 but smaller than a PC.
> I can save physical space and yet have access to all the RISC OS applications that I use > by either using VirtualRiscPC and/or having an A9.
>
You could do the same with an Iyonix - can't think of any apps of the
top of my head (other than Select - but then it's not an App) that can
run *only* on A9/VRPC but not on Iyonix. In fact if you need to run old
26 bit code the Iyonix would probably be better than A9 as it's version
of Aemulor uses special instructions (only found in the Iyonix 80321)
to speed things up somewhat.
> Any developer who stops supporting their software running on Virtual
> RiscPC or the A9 is going to be cutting off a significant source of
> revenue. So that means that the majority of developers are likely to
> continue to produce applications that people like me will continue to
> pay money for.
It's a two way street. Developers are more inclined on the whole to
support OSes and vendors who (at least) provide them with access to API
information (documentation if you will) and where possible source. This
open sourcing and the existing API information for RO 5 do that -
Select fails on both counts. To support *you* (as a Select user having
an A9) means that the developer has to *pay* for an A9 and a yearly
Select subscription. For supporting RO 5 from now on there's
(theoretically) no cost other than their time.
I don't believe anyone would purposefully write code that would *not*
work on the A9 or VRPC it's just that to support *you* will *cost them*
generally users are most likely to get an immediate response and the
most help if they choose hardware the developers use. Thing is *many*
developers chose to support the RISC OS platform to the largest
possible extent and bought Iyonixes early on.
>
> Because the Iyonix is a full sized desktop machine you've got to
> convince people that they want to dedicate the desktop space (shelf or
> floor) to it.
Again what's with the volumetric argument? Technically Iyonix is *NOT*
a full sized desktop computer - the standard model is a "mini-tower",
the smallest variant is a small "cube". Yes you can also get the
enormous Panther (have a look at www.iyonic.com for some pictures).
As the Iyonix is relatively cool you could mount a cube Iyonix *below*
a cube PC that runs VRPC. That way for floor mounted set ups an Iyonix
(cube) and PC (cube) would probably occupy less space than a normal
desktop PC.
>It not only needs to provide a few killer apps, but also not lack ANY necessities.
Rubbish - it runs the same apps as an A9 ! So any deficiencies *it* has
are shared with the A9. It does other things the A9 can't and never
will do (display off air TV, burn CD's/DVDs, handle more than 1 hard
disk, optical drive; access optical drive at faster than USB1.1, handle
expansion cards and so on).
>A PC/A9 combination, or a PC/VirtualRiscPC combination fills a big gap that the Iyonix >alone just doesn't, yet occupies about the same amount of space.
Its not an either or. In reality *no* computer of any make or type does
*everything*. All things are compromises. I could make a more
convincing case for a PC/VRPC and Iyonix than A9 - the PC/VRPC option
would allow you to run Select (and network with the Iyonix at gigabit
speed) yet run RO 5 or speed dependant RO apps on the Iyo - the A9/PC
combo would offer you none of the RO5 features and not as much speed or
expandability.
>
> I've yet to see any argument that convinces me that Pseudo Open Sourcing
> RISC OS 5 addresses and counters the reasons that large numbers of
> people are voting with their wallets and running VirtualRiscPC on a PC
> or buying an A9 to sit alongside a PC.
What reasons would those be, in your opinion? Sure you were advocating
Windows a few paragraphs back. With that sort of "support" RISC OS has
*serious* problems I would have thought.
>
> Whilst the OS is a nice feature of RISC OS, it is after all the
> applications that run on the OS that make it useful for most users and I
> don't see that this latest move is going to change or improve the
> immediate availability of applications. Indeed it might even distract
> some developers away from time spent on applications into time spent on
> the OS (including the basic bundled applications). How does that make
> things better?
This argument has also been advanced by Peter Naulls and I must admit
it does have possibly some validity. Thing is that if you fix/improve
the underlying OS, improve the development tools and provide updated OS
documentation (and source) it makes it *easier* to develop
applications. You may find that apps will have a little bit of a hiatus
followed by increased developments when RISC OS 5 gets some more bells
and whistles.
OS improvements are available to *all* applications. RISC OS'
impressive Anti-aliased scalable fonts (an OS facility) helped launch a
raft of DTP apps. If the DTP apps had *not* had this support in the OS
each one would have had to implement their own font system.... (in
short a mess). In such circumstances I suspect you'd have seen *fewer*
apps of that type because the task of implementing them would have been
more difficult (i.e., take longer, cost more).
I expect the same improvements to apps to happen after the current Open
Source initiative.
>
> Of course the motive behind Pseudo Open Sourcing of RISC OS 5 may have
> nothing (or little) to do with the marketing potential in the desktop
> market but may be driven by initiatives in other potential markets (such
> as set top boxes?).
>
And where do you suppose the bulk of A9's are sold ? Desktop use or
somewhere else? The purpose of the small box is because the A9 is a
derivative of am embedded controller (a small compact device used to
control displays, devices or processes where a full scale computer
could not be justified). That's not to denegrate it or anything - but
that what it's genesis is. So Iyonix was used in STB development - so
what?
Regards
Annraoi
John Cartmell
It wasn't an insult. It was an explanation of how Ray is able to maintain
belief in contradictory information. It has been discussed elsewhere.
Martin Wuerthner
> I've yet to see any argument that convinces me that Pseudo Open Sourcing
> RISC OS 5 addresses and counters the reasons that large numbers of
> people are voting with their wallets and running VirtualRiscPC on a PC
> or buying an A9 to sit alongside a PC.
If you think that this was the point of open sourcing RISC OS, then
you are mistaken. And the fact that this idea does not make much sense
probably explains why nobody has managed (or even tried) to convince
you of it. ;-)
> Whilst the OS is a nice feature of RISC OS, it is after all the
> applications that run on the OS that make it useful for most users and I
> don't see that this latest move is going to change or improve the
> immediate availability of applications. Indeed it might even distract
> some developers away from time spent on applications into time spent
> on the OS (including the basic bundled applications). How does that
> make things better?
It makes things much better because we, the developers, can finally
fix things that neither party has tackled in the past 10 years. For
instance, being able to create PDF files is a crucial requirement for
many users. However, many PDF files created using the RISC OS
PostScript driver (i.e., created using any RISC OS application with
the notable exception of ArtWorks 2) are of poor quality. An enhanced
PostScript driver would improve things a lot, but to make that
possible its source code needs to be available, which will soon be the
case. Spending some development time on the PS driver would improve
what you can achieve with *many* applications in one go and would thus
achieve much more than spending development time on applications.
> Of course the motive behind Pseudo Open Sourcing of RISC OS 5 may have
> nothing (or little) to do with the marketing potential in the desktop
> market but may be driven by initiatives in other potential markets (such
> as set top boxes?).
Aha, what a surprising thought. You have probably not had a look at
the RISC OS Open Limited web site yet...
Keith Hopper
Mike Clark <mr...@cam.ac.uk> wrote:
[snip]
> There's cash expenditure on hardware, cash expenditure on software and
> cash expenditure on peripherals and ancillaries.
> Over the years I've spent far more money on software, peripherals and
> ancillaries, than I have on the basic hardware. I've got business
> accounts that detail my expenditure. It runs into the 10s of thousands
> of pounds spent with companies and businesses associated with the RISC
> OS community.
I haven't spent that much but quite a bit over the best part of twenty
years since I saw the first prototype in the Acorn lab in Cambridge.
[snip]
> Whilst the OS is a nice feature of RISC OS, it is after all the
> applications that run on the OS that make it useful for most users and I
> don't see that this latest move is going to change or improve the
> immediate availability of applications. Indeed it might even distract
> some developers away from time spent on applications into time spent on
> the OS (including the basic bundled applications). How does that make
> things better?
Indeed! However!!! ... and its a big however too - the OS API is what
matters to the developer - and the availability of modern development tools
(and I do not mean just an XYZ compiler here. Until there is a stable API
with no sneaky use of extra 'bits' in arguments or responses then life will
inevitably be difficult for developers. This is really the crucial aspect
of the game. It was the same with Unix - to POSIX - and very slightly
changed over many years - but the original API is still valid so far as I
am aware. I think Martin has it right when he commented that the 3.6
version (I think he said) or at best 4.0 would be something which should be
frozen.
I must admit to using the POSIX API - via Unixlib as something which
is portable - albeit with some extra wrapping code.
While I deprecate on theoretical grounds most of the programming
languages which seem to be used for appplication development of RISCOS
applications, this is really only a prod to others to look at using
high-level tools (one of the nicest amodal single file editors I know is
written in Python!). I am looking at porting YAFL using gccsdk over the
next year or so - as an intermediate tool for my own use. YAFL -is- a
high-level language development out of the Modula-2/Oberon-2 stable
although it is still only half way to modern thinking on language design.
I, too, am suggesting therefore that a straw poll or something like
that should be taken about what actual API will be used for further work on
applications. Let the open software lobby go ahead -knowing- that
developers are going to use some particular interface. Doing things
underneath the interface is fine - but please let us have one single fully
defined API. With that in place there is a single fixed target - OS
development can take place to offer improved performance - even, if you
like, pre-emptive multi-tasking -without- changing the API. Developers can
work happily away knowing that the ground will not shift under their feet -
or over their heads.
Just a thought!
Keith Hopper
--
Sky Development
David Holden
On 5-Oct-2006, Mike Clark <mr...@cam.ac.uk> wrote:
> There's cash expenditure on hardware, cash expenditure on software and
> cash expenditure on peripherals and ancillaries.
>
> Over the years I've spent far more money on software, peripherals and
> ancillaries, than I have on the basic hardware. I've got business
> accounts that detail my expenditure. It runs into the 10s of thousands
> of pounds spent with companies and businesses associated with the RISC
> OS community.
I think this is something that a lot of people don't understand.
Back in the days when the majority of users had pre RiscPC machines a *huge*
proportion of the money spent on Acorn computers went to 3rd party hardware
suppliers. If you needed a hard drive for your A3000 or A3010 or RAM
expansion for any model you couldn't go to your local PC dealer. With the
arrival of the RiscPC CDs, hard drives, RAM, etc. were standard(ish)
components so a big chunk of turnover was lost by most dealers.
Currently the biggest amount of money spent by RISC OS users doesn't go on
new hardware from RISC OS dealers, it's spent on software. It's often said
that the people who spend most *must* be the people who've bought an Iyonix
or A9 because they're obviously the ones still committed to RISC OS. I don't
know about the second part of that statement, but the first part certainly
isn't true. Individually it might be the case (after all, a lot of people
need to spend money on upgrades for their software if they buy a 32 bit
machine), but the fact is that numerically there are far more people still
using 26 bit OSs so collectively they spend more.
I can't see this situation changing in the foreseeable future.
--
David Holden - APDL - <http://www.apdl.co.uk>
Mike Clark
Martin Wuerthner <spam...@mw-software.com> wrote:
> In message <a27831714e....@mrc7acorn1.path.cam.ac.uk>
> Mike Clark <mr...@cam.ac.uk> wrote:
>
> > I've yet to see any argument that convinces me that Pseudo Open Sourcing
> > RISC OS 5 addresses and counters the reasons that large numbers of
> > people are voting with their wallets and running VirtualRiscPC on a PC
> > or buying an A9 to sit alongside a PC.
>
> If you think that this was the point of open sourcing RISC OS, then
> you are mistaken.
No I didn't think it was the point, but at least some contributors to
the newsgroups were desperately trying to fit that aspiration into their
utopian RISC OS world vision.
> And the fact that this idea does not make much sense probably
> explains why nobody has managed (or even tried) to convince you of
> it. ;-)
>
> > Whilst the OS is a nice feature of RISC OS, it is after all the
> > applications that run on the OS that make it useful for most users and I
> > don't see that this latest move is going to change or improve the
> > immediate availability of applications. Indeed it might even distract
> > some developers away from time spent on applications into time spent
> > on the OS (including the basic bundled applications). How does that
> > make things better?
>
> It makes things much better because we, the developers, can finally
> fix things that neither party has tackled in the past 10 years. For
> instance, being able to create PDF files is a crucial requirement for
> many users. However, many PDF files created using the RISC OS
> PostScript driver (i.e., created using any RISC OS application with
> the notable exception of ArtWorks 2) are of poor quality.
OK that is an example of something I can buy into since it would be of
direct benefit to me as a user.
My daily general use of RISC OS applications is now mainly restricted to
Techwriter, Zap, Messenger, Fireworkz, RiScript, HTMLEdit Studio and
Impression Publisher. Less frequently I use Fresco, WebsterXL,
DataPower, Artworks, Draw and PhotoDesk. I also use a collection of
utilities for processing data for and between these above applications
and a few legacy applications written by my colleagues and I for
personal specialist requirements.
For things like web browsing I now almost exclusively use Firefox on the
PC (backed up with IE when a site proves to be particularly badly
written). Also for all music, video, digital mapping and digital imaging
I use the PC. Recently I've started using voice recognition software on
the PC.
Regarding printing and handling of difficult PDF files I have found
RComps UniPrint and UniLaunch to provide the best solution. Click on the
URL, or print, in RISC OS and then see the data handled seamlessly in
Windows.
> An enhanced PostScript driver would improve things a lot, but to make
> that possible its source code needs to be available, which will soon
> be the case. Spending some development time on the PS driver would
> improve what you can achieve with *many* applications in one go and
> would thus achieve much more than spending development time on
> applications.
>
> > Of course the motive behind Pseudo Open Sourcing of RISC OS 5 may
> > have nothing (or little) to do with the marketing potential in the
> > desktop market but may be driven by initiatives in other potential
> > markets (such as set top boxes?).
>
> Aha, what a surprising thought. You have probably not had a look at
> the RISC OS Open Limited web site yet...
>
> Martin
You're correct I was going by comments I'd read in these newsgroups and
on Drobe, I have now taken a more detailed look at their website.
The licensing conditions are as I said above, what I would regard as
Pseudo Open Source. Working in an academic environment I'm much more
familiar with the GPL type of licence.
Clearly your comments above about the owners of the OS (and their
licences) not bothering to fix major long term issues such as the
PostScript drivers, combined with the description of the probable
licence terms of ROOL, serve to reinforce the view I've had for a long
time, which is that they (the owners of the OS) haven't got a sufficient
cash flow to enable them to properly finance their own in house
development of the OS at the required level.
The licence terms mean that they might get bugs fixed for them, without
having the expense of paying for the salary of a developer, yet at the
same time the terms mean that they get to keep most of the resulting IP,
including the bug fixes and improvements, for themselves. Clearly if
such fixes found their way into commercial applications of RISC OS in
large numbers of embedded devices, then the royalty payments resulting
would swell the coffers of Castle/ROOL and not the contributing
developers to the fixes.
I'm getting a clearer picture of where the aspirations of cash flow are
coming from, and where they are going too.
george
> In message <b5102371...@tiscali.co.uk>
> george <george.g...@tiscali.co.uk> wrote:
>
>> In message <4e71097042inval...@invalid-domain.co.uk>
>> Paul Vigay <invalid-em...@invalid-domain.co.uk>
>> wrote:
> [snip]
[snip]
>
> I don't have an Iyonix and I still haven't seen an overwhelming case
> that would encourage me to buy one.
Everyone's got to decide for themselves, and I wouldn't presume to
challenge your particular reasons. For myself, I wanted a RO machine
that would deal more easily with substantial graphics files (8Mpix
JPEGS amounting to approx. 25MB as sprites) and do 16 million colours
in a 1280 x 1024 pix display. The RiscPC's hard disk and native VRAM
performance ruled it out*, so I had the option of going VRPC or
Iyonix. As one of the main attractions of email and browsing on a RO
machine was immunity from viruses, I chose the latter. AemulorPro
enabled me to continue to use Rhapsody4 for my music setting, and
indeed any other 26-bit app I wished to use, so that wasn't an issue.
(*of course a Viewfinder would have given me the display performance,
and I had one in my Kinetic for some time, but the use of the podule
bus is a definite handicap despite the clever s/ware).
> It could never (at least in the
> foreseeable future) replace all my requirements for software so I'd
> still definitely need to have space for a PC. I can save physical space
> and yet have access to all the RISC OS applications that I use by either
> using VirtualRiscPC and/or having an A9.
>
> Any developer who stops supporting their software running on Virtual
> RiscPC or the A9 is going to be cutting off a significant source of
> revenue. So that means that the majority of developers are likely to
> continue to produce applications that people like me will continue to
> pay money for.
Yes, but that's a different issue from continuing to expend effort, as
ROL did for several years, refining the OS for what is technologically
a dead end, i.e., 26-bit hardware.
>
> Because the Iyonix is a full sized desktop machine you've got to
> convince people that they want to dedicate the desktop space (shelf or
> floor) to it. It not only needs to provide a few killer apps, but also
> not lack ANY necessities. A PC/A9 combination, or a PC/VirtualRiscPC
> combination fills a big gap that the Iyonix alone just doesn't, yet
> occupies about the same amount of space.
On the desk here I have one CRT monitor, keyboard and mouse, the
Iyonix, an ADSL router, a KVM switch box and a PC laptop (which I own
but hardly use). Another (work) PC laptop sits on top of the Iyonix.
On the floor is a scanner, and on the other side of the desk is one of
those old wheeled metal workstation tables with two printers (a colour
inkjet and a b/w laser). This (plus 2 x 3-drawer filing cabinets) all
fits in a home office measuring 10ft x 8ft. Compared with all this
gear, the space occupied by the Iyonix is trivial.
>
> I've yet to see any argument that convinces me that Pseudo Open Sourcing
> RISC OS 5 addresses and counters the reasons that large numbers of
> people are voting with their wallets and running VirtualRiscPC on a PC
> or buying an A9 to sit alongside a PC.
>
> Whilst the OS is a nice feature of RISC OS, it is after all the
> applications that run on the OS that make it useful for most users and I
> don't see that this latest move is going to change or improve the
> immediate availability of applications. Indeed it might even distract
> some developers away from time spent on applications into time spent on
> the OS (including the basic bundled applications). How does that make
> things better?
It may be a distraction, but AFAICS there are a number of aspects of
the OS (for example, the Postscript driver in !Printers) that directly
affect the ability of applications to work well. These need to be
fixed. If work can be done on issues like memory protection, stability
etc that will also benefit application users. Hopefully the
ROOL/Castle initiative will galvanise development of these and other
areas.
>
> Of course the motive behind Pseudo Open Sourcing of RISC OS 5 may have
> nothing (or little) to do with the marketing potential in the desktop
> market but may be driven by initiatives in other potential markets (such
> as set top boxes?).
Why should desktop users complain about that? For a start,
developments of, say, video streaming support for the embedded market
may directly benefit desktop users; even if not, I'd be delighted if a
wall of embedded application cash descended on ROOL and Castle, since
that would help to guarantee faster development in the desktop area
and hopefully, new native hardware as well.
George
--
Robert Seago
> > I find it encouraging that people of such experience and ability are
> > still committed to the development of RISC OS.
> >
So do I
> What are the motivating factors though? Is it based on altruism? If
> there are commercial interests whose interests are best served? Where is
> the (anticipated?) cash flow coming from, and where is it going too?
I can only think that it is pretty well altruistic. I used to enjoy
building little apps for the education side and other utils , and think I
got a donation of a few quiid about once.
> Don't forget that a lot of ex-acorn senior staff, executives and
> shareholders have already made a lot of money out of the assetts of
> Acorn, and perhaps some think that there is still potential to make a
> bit more.
I hope so
--
Regards from Robert Seago : http://www.users.zetnet.co.uk/rjseago
John Cartmell
Robert Seago <rjs...@zetnet.co.uk> wrote:
> In article <a27831714e....@mrc7acorn1.path.cam.ac.uk>,
> Mike Clark <mr...@cam.ac.uk> wrote:
> > > I find it encouraging that people of such experience and ability are
> > > still committed to the development of RISC OS.
> > >
> So do I
> > What are the motivating factors though? Is it based on altruism? If
> > there are commercial interests whose interests are best served? Where is
> > the (anticipated?) cash flow coming from, and where is it going too?
> I can only think that it is pretty well altruistic. I used to enjoy
> building little apps for the education side and other utils , and think I
> got a donation of a few quiid about once.
Castle and RISC OS Open are companies that need cash or promise of cash. Pure
speculation may assume that Castle have licensed to RISC OS Open bits of the
OS that is not already held exclusively by RISCOS Ltd in return for RISC OS
Open administering and adding to the work done free by volunteers which Castle
can then sell.
News poster
John Cartmell <jo...@cartmell.demon.co.uk> wrote:
> In article <4e71b098...@zetnet.co.uk>,
> Robert Seago <rjs...@zetnet.co.uk> wrote:
> > In article <a27831714e....@mrc7acorn1.path.cam.ac.uk>,
> > Mike Clark <mr...@cam.ac.uk> wrote:
>
> > > > I find it encouraging that people of such experience and ability are
> > > > still committed to the development of RISC OS.
> > > >
> > So do I
> > > What are the motivating factors though? Is it based on altruism? If
> > > there are commercial interests whose interests are best served? Where is
> > > the (anticipated?) cash flow coming from, and where is it going too?
> > I can only think that it is pretty well altruistic. I used to enjoy
> > building little apps for the education side and other utils , and think I
> > got a donation of a few quiid about once.
>
> Castle and RISC OS Open are companies that need cash or promise of cash.
Uhu and RISCOS Ltd, Advantage6 are not?
> Pure speculation...
<snip in mid sentence>
Indeed pure speculation.
ROL are clearly NEVER going do anything to 'enhance' RISC OS 5 and
Castle quite rightly have their priorities aimed at maintaining the
profitability of their business* and securing a future for RISC OS. So
why not take advantage of an initiative to enthuse the enthusiasts?
RISC OS 5 is an extremely stable base to build on, and adding a measure
of quality control to changes made by outside parties seems to me to be
very sensible.
Seems to me like a win-win situation, for Castle and those of us lucky
enough to be running RISC OS 5 who are looking for a more dynamic user
driven process of OS improvements.
Regards
Stan
*You should know all about the importance of profitability of business
and timely product releases for market confidence. Or at least
understand the principles if not the practice.
--
--
Sandy Morton
<jo...@cartmell.demon.co.uk> wrote:
> Castle and RISC OS Open are companies that need cash or promise of
> cash.
And presumably subscribers to Quercus need a similar type of promise?
I don't understand why you are going to attend the imminent shows
when you appear to have nothing to sell - other than empty promises.
--
T h e - e x t e n d e r ! ! ! !
druck
> Pure speculation may assume that Castle have licensed to RISC OS Open bits
> of the OS that is not already held exclusively by RISCOS Ltd
Pure speculation, and complete nonsense. A classic JC post in the making.
---druck
--
The ARM Club Free Software - http://www.armclub.org.uk/free/
The 32bit Conversions Page - http://www.quantumsoft.co.uk/druck/
John Cartmell
druck <ne...@druck.freeuk.com> wrote:
> On 6 Oct 2006 John Cartmell <jo...@cartmell.demon.co.uk> wrote:
> > Pure speculation may assume that Castle have licensed to RISC OS Open bits
> > of the OS that is not already held exclusively by RISCOS Ltd
> Pure speculation, and complete nonsense. A classic JC post in the making.
Whilst I clearly label it pure speculation druck dismisses it without
revealing that he has no idea what he's talking about. Of course if he knows
what RISC OS Open are getting out of the deal then maybe he could tell us. If
not we'd perhaps treat his comments as 'pure speculation' even though he wants
us to believe otherwise.
Speculation is sometimes OK but only when clearly labelled. druck is always so
certain even when he's wrong - like when he told the world the A9 still had no
sound three weeks after mine was putting out quite decent harmonies.
I'm quite happy to see my pure speculation corrected. Flat denials and insults
from druck do *not* count! ;-)
druck
> In article <fb27cc71...@druck.freeuk.net>,
> druck <ne...@druck.freeuk.com> wrote:
> > On 6 Oct 2006 John Cartmell <jo...@cartmell.demon.co.uk> wrote:
>>> Pure speculation may assume that Castle have licensed to RISC OS Open
>>> bits of the OS that is not already held exclusively by RISCOS Ltd
>
> > Pure speculation, and complete nonsense. A classic JC post in the making.
>
> Whilst I clearly label it pure speculation druck dismisses it without
> revealing that he has no idea what he's talking about. Of course if he
> knows what RISC OS Open are getting out of the deal then maybe he could
> tell us. If not we'd perhaps treat his comments as 'pure speculation' even
> though he wants us to believe otherwise. Speculation is sometimes OK but
> only when clearly labelled. druck is always so certain even when he's wrong
> - like when he told the world the A9 still had no sound three weeks after
> mine was putting out quite decent harmonies.
Well I bow to your world authority of being wrong, and in such a wonderfully
verbose manner too, you could fill an entire magazine with such stuff -
why aren't you?
> I'm quite happy to see my pure speculation corrected. Flat denials and
> insults from druck do *not* count! ;-)
I was referring, should you not realise from the precise quoting, your
preposterous bluster about "bits of RISC OS not already held exclusively by
RISCOS Ltd" as if it were some diminutive part. Castle as the owner the head
licence have rights over the whole of the RISC OS code base up to and
including Select 2. It is only at that point did new code written by ROL
become their sole property. Whether ROL submitted back to the holder of the
head license (then Pace) all the code they were obliqued to is another
matter.
Chris Joseph
>So the volumetric capacity of Iyonix is a "problem" ? You *do* know
>they have a small footprint version (a cube box). Ok not as small as an
>A9 but smaller than a PC.
But you can get PCs in cases that size too, and if space is *really*
tight.... So I'm afraid that's something of a null argument.
Chris.
News poster
John Cartmell <jo...@cartmell.demon.co.uk> wrote:
> In article <fb27cc71...@druck.freeuk.net>,
> druck <ne...@druck.freeuk.com> wrote:
> > On 6 Oct 2006 John Cartmell <jo...@cartmell.demon.co.uk> wrote:
> > > Pure speculation may assume that Castle have licensed to RISC OS
> > > Open bits of the OS that is not already held exclusively by RISCOS
> > > Ltd
>
> > Pure speculation, and complete nonsense. A classic JC post in the making.
>
> Whilst I clearly label it pure speculation druck dismisses it without
> revealing that he has no idea what he's talking about. Of course if he
> knows what RISC OS Open are getting out of the deal then maybe he
> could tell us. If not we'd perhaps treat his comments as 'pure
> speculation' even though he wants us to believe otherwise. Speculation
> is sometimes OK but only when clearly labelled. druck is always so
> certain even when he's wrong - like when he told the world the A9
> still had no sound three weeks after mine was putting out quite decent
> harmonies.
>
> I'm quite happy to see my pure speculation corrected. Flat denials
> and insults from druck do *not* count! ;-)
Now I wonder who was it who said this around the 13th of September?
"I know you think you're being funny but your ill-informed and
destructive speculation gets a touch wearing after a while. The next
issue of Qercus will be up to date on release which will be soon. As
I've already clearly said I will not qualify that 'soon' until I have
the printed copies in front of me ready for posting."
As regards speculation I suggest you think about what people in glass
houses should not do, before you start speculation on the motives of
people running their companies with rather more success than you seem
to run yours.
Perhaps with the aim of reducing speculation about Qercus and all of
this unnecessary sniping that contributors to this forum apparently
don't like, you could answer some questions yourself?
Lets start off with you telling us;
a) what anyone in the RISC OS market (let alone subscribers) is getting
out of Qercus?
b) how much money you have taken in subscriptions in the last year is
actually left over for paying contributors and other production costs?
c) Are you financially able to pay back subscriptions to subscribers of
Qercus if compelled to do so by court orders?
Regards
Stan
--
--
John Cartmell
druck <ne...@druck.freeuk.com> wrote:
> I was referring, should you not realise from the precise quoting, your
> preposterous bluster about "bits of RISC OS not already held exclusively by
> RISCOS Ltd" as if it were some diminutive part. Castle as the owner the head
> licence have rights over the whole of the RISC OS code base up to and
> including Select 2. It is only at that point did new code written by ROL
> become their sole property. Whether ROL submitted back to the holder of the
> head license (then Pace) all the code they were obliqued to is another
> matter.
As someone who has read the contract you obviously know all that. As someone
who has read the contract you are bound by NDA not to discuss the matter.
Lying or breaking contract?
Or much more likely: haven't got the slightest idea what you're talking about
or are intending to simply confuse matters.
Whichever it's as irrelevant as all your comments on the matter - and quite
off-topic. I think we'd best agree to end the matter here.
Ray Dawson
> In article <29971072...@druck.freeuk.net>,
> druck <ne...@druck.freeuk.com> wrote:
> > I was referring, should you not realise from the precise quoting, your
> > preposterous bluster about "bits of RISC OS not already held
> > exclusively by RISCOS Ltd" as if it were some diminutive part. Castle
> > as the owner the head licence have rights over the whole of the RISC
> > OS code base up to and including Select 2. It is only at that point
> > did new code written by ROL become their sole property. Whether ROL
> > submitted back to the holder of the head license (then Pace) all the
> > code they were obliqued to is another matter.
>
> As someone who has read the contract you obviously know all that. As
> someone who has read the contract you are bound by NDA not to discuss
> the matter.
It is obvious that are not privy to the contents of the contract as you
would also be subject to the same NDA. The fact that you are giving
opinions all the time on the contents of the contract means that you
either breaking the NDA, or weren't told in the first place.
> Lying or breaking contract?
Lying or speculating?
> Or much more likely: haven't got the slightest idea what you're talking
> about or are intending to simply confuse matters.
Ah - talking about yourself again.
> Whichever it's as irrelevant as all your comments on the matter - and
> quite off-topic.
Although I often don't agree with druck, I do respect the fact that he
does have inside knowledge, but cannot always disclose it for obvious
reasons. You have no such knowledge.
> I think we'd best agree to end the matter here.
So, you are losing the argument.
And there's me promised myself I wouldn't respond to any of your stupid
posts again :-(
Ray D
John Cartmell
Ray Dawson <r...@magray.freeserve.co.uk> wrote:
> Lying or speculating?
The phrase I used was "Pure speculation may assume that ..."
When I've specifically marked it down as 'pure speculation' can you please
explain why you want to make a (in a very nasty way) complaint that I might
have been speculating?
Off-topic discussion about your mental state taken to appropriate forum.
VinceH
> Whichever it's as irrelevant as all your comments on the matter
> - and quite off-topic.
Huh?
Discussion/speculation of the legal ownership of RISC OS, which is
pertinant to the release of some or all of RISC OS under the
shared source plans of Castle/ROOL, is "off-topic" in a thread
entitled "RISC OS source code release news" ?
Okaaaaaaay then.
And there are people out there telling me *I'm* mad!
--
VinceH - http://www.softrock.co.uk/info/vinceh.html
Please sponsor me to abseil forwards down the Avon Gorge:
http://www.justgiving.com/vinceh/ - Okay, so I've done it now -
why not stump up some money anyway, tight-arse?
druck
> druck <ne...@druck.freeuk.com> wrote:
>> I was referring, should you not realise from the precise quoting, your
>> preposterous bluster about "bits of RISC OS not already held exclusively
>> by RISCOS Ltd" as if it were some diminutive part. Castle as the owner the
>> head licence have rights over the whole of the RISC OS code base up to and
>> including Select 2. It is only at that point did new code written by ROL
>> become their sole property. Whether ROL submitted back to the holder of
>> the head license (then Pace) all the code they were obliqued to is another
>> matter.
>
> As someone who has read the contract you obviously know all that. As
> someone who has read the contract you are bound by NDA not to discuss the
> matter.
>
> Lying or breaking contract?
I am only repeating publicly made statements made by Paul Middleton and
other ROL employees between 1999 and 2003.
> Or much more likely: haven't got the slightest idea what you're talking
> about or are intending to simply confuse matters.
You could of course pretend you were still involved in the press and look up
the notes of meetings with ROL that Acorn User attended, such as on 17th
August 2001 at the Pace offices in Cambridge. I'm sure Steve Turbull can
fill you in if you don't have records from back then.
> Whichever it's as irrelevant as all your comments on the matter - and quite
> off-topic. I think we'd best agree to end the matter here.
I hate to have to remind you in every post, but as someone who is currently
masquerading as a magazine producer, repeatedly lying to and effectively
defrauding a large number of RISC OS users, I don't think you can really go
questioning the honesty and integrity of other people.
Ams
If you'd read my post *in full* you'd have spotted that I also wrote:
"As the Iyonix is relatively cool you could mount a cube Iyonix *below*
a cube PC that runs VRPC. That way for floor mounted set ups an Iyonix
(cube) and PC (cube) would probably occupy less space than a normal
desktop PC."
Emphasis on the word "Cube PC", the case the Iyonix Cube uses is (in
actual fact) a small cube PC case. My point (which still stands) is
that if the Mike Clark's objection to an Iyonix was it's size - it's
not a valid one in that he could use a PC (cube) and an Iyonix (cube)
and be able to run Windows/VRPC (RISC OS Select) on the PC and RO5 for
speed on the Iyonix in a total space that his normal PC occupies.
He'd also have the best of both worlds having the option of running
RISC OS 5 *and* Select.
Regards
Annraoi
David Pitt
druck <ne...@druck.freeuk.com> wrote:
[snip Quercosities]
>Well I bow to your world authority of being wrong, and in such a
>wonderfully verbose manner too, you could fill an entire magazine with
>such stuff - why aren't you?
He is filling up my hard drive! This is a splendid test of MPro kill
file arrangements. 32 posts in 3 days.
--
David Pitt.
Computing with RISC OS.
Ams
<snip>
>
> Castle and RISC OS Open are companies that need cash or promise of cash. Pure
> speculation may assume that Castle have licensed to RISC OS Open bits of the
> OS that is not already held exclusively by RISCOS Ltd in return for RISC OS
> Open administering and adding to the work done free by volunteers which Castle
> can then sell.
As Castle own the head license and the bulk of the OS what ROL own
doesn't really matter does it? Developers it seems likely will
(eventually) be able to build a RISC OS release *without* ROL's
assistance.
Developers who sign up with ROOL know *exactly* what they're doing, the
FAQ clearly states what will happen, code can be shared so long as it's
not exploited commercially. If a developer want's to exploit it
commercially they can still do so - but have to pay a royalty. As much
of ROS existed before the developer gets their mitts on it - it follows
that had they not obtained the source under those very license terms
they would not have been in a position to exploit (and therefore)
profit from it. If they feed improvements back to Castle/ROOL then
Castle is free to use them - that's what the developer signs up to. And
I bet you many will.
Regards
Annraoi
John Cartmell
> I am only repeating publicly made statements made by Paul Middleton and
> other ROL employees between 1999 and 2003.
In that case you're seriously mistaken and seriously off the current topic.
John Cartmell
John Cartmell
John Cartmell
Ams <a...@globalcafe.ie> wrote:
> As Castle own the head license and the bulk of the OS what ROL own
> doesn't really matter does it?
Yes. Try asking anyone who has negotiated an exclusive contract to use/develop
anything. Example: The person who owns the land my house is on has very few
rights other than the collection of a small annual fee. I have the exclusive
right to build on the land and can exclude trespassers including the land
'owner'.
How much it matters depends on the detail of the contract that you and druck
pretend to understand and of which you don't even have the faintest knowledge.