Talk:RISC OS/Archive 1 Source: en.wikipedia.org/wiki/Talk:RISC_OS/Archive_1

This is an archive of past discussions about RISC OS. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

"...since a minor 24-bit ARM CPU..." - should that be 26-bit? Not fixed as I don't know. -- S

Yes, it should be 26-bit. But should it say the function was removed in the XScale or simply changed/made to a 32-bit function? I don't know enough about ARM internals myself to give an answer. Crusadeonilliteracy 13:51, 4 Nov 2003 (UTC)

Originally there was only 26bit addressing (24 bits for address in PC, but ARM instructions are always aligned to 4 bytes), then with ARM6 they introduced a 2nd mode, 32bit addressing. Later they invented Thumb mode, which is mutually exclusive to 26bit mode. StrongARM had 26bit, XScale doesn't. mavhc

'certain countries'? Can we specify them?2toise 13:16, 8 Nov 2003 (UTC)

Does this refer to something that's been removed? mavhc

Two similar projects, Impulse and Eidos's Phoenix, both stalled.

What is this Eidos Phoenix? The only mention I can find of it is of a game by Eidos (what has this to do with RISC OS?) Crusadeonilliteracy 04:32, 22 Mar 2004 (UTC)

Eidos was originally a company specialising in Acorn-based video editing solutions. Only later did they become famous as a non-Acorn games publisher! Yes, it is the same Eidos that published Tomb Raider.

The only RISC OS thing called Impulse I know of is Computer Concepts' pre RISC OS OS, which then became part of their applications mavhc

The article says RISC OS had no memory protection. It didn't have `full' memory protection, but AFAICR there was a limited degree of protection - all 'normal' applications appeared to be running at the same addresses in memory, so it took special actions to read or write to another application's memory. This didn't protect against a malicious application, but I think it did mean a buggy application couldn't corrupt the memory of another one under normal circumstances. I could well be wrong, hence putting this in talk rather than the article. --Zorn 08:38, 23 Jul 2004 (UTC)

I've added some info, what do people think? mavhc

Anyone got a screenshot? I think it'd add to the article. Lupin 08:56, 23 Jul 2004 (UTC)

Can the user who added the clean up tag, or anyone else please specify what they see as needing to be done. Andrewduffell 18:30, 30 November 2005 (UTC)

I realised that it was closed-source and that the opening process was not an opening process but a person developing the same system api...but it was corrected before I corrected it...lol 213.189.165.28 16:20, 22 March 2006 (UTC)

I have moved the Talk into chronological order - please add new talk sequentially at the bottom of this page - bottom posting in true RISC OS tradition :-)TerriersFan 07:29, 16 June 2006 (UTC)

{{Talk header}} now added. --Trevj (talk) 13:19, 11 May 2011 (UTC)

Is this worth to add to article? --Yonkie 10:49, 2 October 2007 (UTC)

Image:AcornArthur110desktopsmall.png is being used on this article. I notice the image page specifies that the image is being used under fair use but there is no explanation or rationale as to why its use in this Wikipedia article constitutes fair use. In addition to the boilerplate fair use template, you must also write out on the image description page a specific explanation or rationale for why using this image in each article is consistent with fair use.

Please go to the image description page and edit it to include a fair use rationale. Using one of the templates at Wikipedia:Fair use rationale guideline is an easy way to insure that your image is in compliance with Wikipedia policy, but remember that you must complete the template. Do not simply insert a blank template on an image page.

If there is other fair use media, consider checking that you have specified the fair use rationale on the other images used on this page. Note that any fair use images lacking such an explanation can be deleted one week after being tagged, as described on criteria for speedy deletion. If you have any questions please ask them at the Media copyright questions page. Thank you.

— Save_Us_229 01:15, 31 December 2007 (UTC)

PROJECT IS NOW LIVE

Previous WikiProject talk has now been moved to Wikipedia_talk:WikiProject_RISC_OS

Your source doesn't say anything about the icon bar predating anything. You are taking a few separate facts, and from them drawing a conclusion that isn't mentioned in the source. That's original research.

Also. Mentioning that the icon bar predates Windows' taskbar and Nextstep/Mac OS X's dock is implying that they were influenced in some way by the icon bar. Again, that is original research unless you can provide a reliable source that states same. AlistairMcMillan (talk) 02:29, 2 January 2011 (UTC)

No. The iconbar is part of RiscOS (And Arthur previous to that). Arthur/RiscOS was released in 1987. Ergo iconbar predates Mac OS X & Win 95, which were released in 1995 & 1999. That's hardly original research or even synthesis which would have been be a better accusation.

Furthermore, the only person who is implying that the iconbar influenced either Win95 or Mac OS X is you. The term "predate"[1] offers no definition or suggestion about such an implication, only that it came before another given date - in this case the Win95 & Mac OS releases. Mentioning that it predates them is perfectly valid, given the functional similarities of the respective iconbars. Something had to come first - why not mention it, given that there's a source to show that RiscOS was released prior to both others?

Despite still thinking you're wrong, I'll not revert for a while to give other editors a chance to comment as well. a_man_alone (talk) 08:30, 2 January 2011 (UTC)

Mac OS X is an evolution of Nextstep. When was Nextstep developed? Was it released in the 1990s? AlistairMcMillan (talk) 09:00, 2 January 2011 (UTC)

see: Dock (computing), the first versions of Nextstep were in the same 1986-1988 period as the first Arthur/RiscOS versions. --Egel _Reaction? 11:10, 2 January 2011 (UTC)

While I've no doubt that it's correct, I agree that it's original research, unfortunately. It seems to me, though, that it ought to be comparatively simple to source, since it seems likely that something similar would have been published. Sadly I've long ago lost my Acorn magazine collection, but I wonder whether news items in these about W95 or OS X might make such a comparison with RISC OS? Jakew (talk) 11:30, 2 January 2011 (UTC)

Be careful using RISC OS magazines (or PC ones for that matter) as they may well be not considered the most unbiased of references.--Flibble (talk) 15:40, 3 January 2011 (UTC)

Technically, the Windows taskbar predates Arthur/RISC OS, as it's in Windows 1.0 from 1985. The Arthur taskbar (two years later) looks very similar... —Preceding unsigned comment added by 82.9.85.2 (talk) 12:19, 3 January 2011 (UTC)

I agree the similarity of the Windows 1.0 taskbar to the arthur taskbar seems telling. Perhaps Arthur's is a logical development of it (thus ends my original research ;-) )--Flibble (talk) 15:40, 3 January 2011 (UTC)

Okay I'm confused now. Which Windows 1.0 taskbar are we talking about? I didn't think Windows had a taskbar until the one developed for Cairo that evolved into the Windows 95 taskbar. AlistairMcMillan (talk) 21:39, 4 January 2011 (UTC)

http://www.google.co.uk/images?q=windows+1.0 The green bar along the bottom, dropped by Windows 2.x.--Flibble (talk) 22:00, 4 January 2011 (UTC)

Do you know what this bar was called? Was it an actual bar across the bar or just showing minimised icons like Windows 3 does (or even Windows XP where it will show an icon in the bottom-left of the screen if you kill Explorer.exe and then minimise an application)? AlistairMcMillan (talk) 01:11, 5 January 2011 (UTC)

Interesting... there's no specific mention in Windows 1.0 (or its talk page). --trevj (talk) 10:07, 5 January 2011 (UTC)

Having had a good read of the article and the reference, I'm in agreement with AlistairMcMillan having made his changes. The reference linked, didn't mention Iconbars/taskbars in comparison to win95 or mac os x.--Flibble (talk) 22:24, 5 January 2011 (UTC)

                    GBLS  SystemName
SystemName          SETS  "RISC OS" ; ", p.k.a. Arthur a.k.a. Richard III"

I don't know if this PCW_Jan89 article is referring to "within the OS" or not. There was a taster in PCW_Aug87. --trevj (talk) 16:17, 12 January 2011 (UTC) Shame there's no history in the relevant section of this ART note from Jul96 --trevj (talk) 16:23, 12 January 2011 (UTC)

"RISC OS has featured font anti-aliasing since 1990" (I can't access the link from where I am ATM.) --trevj (talk) 10:09, 13 January 2011 (UTC)

The Jan 89 one does mention Anti-aliased fonts inside ArcDraw (later !Draw). I think that might be enough, but I'm not sure. The desktop itself didn't move from System font to anti-aliased until RISC OS 3.50 (1994) was released and some of the not-public developer modules from 1993. http://www.marutan.net/db/modules.php?kModule=12 --Flibble (talk) 16:23, 13 January 2011 (UTC)

Wasn't the German RiscOS 3.19 was the first RiscOS to have an anti-aliased system font? It was touted in Acorn User, but I very much doubt I still have the relevant magazine. a_man_alone (talk) 16:38, 13 January 2011 (UTC)

No, and I know the exact picture you're referring too as it's caused this confusion ever since. If you're looking for the magazine, it's the Acorn User that covered the launch of the Risc PC on the cover. What that was a picture of an intermediate German build (labelled as 3.12 in the picture) running one of the developer modules I mentioned above. If you'd like I can send you a rom image of 3.19 so you can test it under Arculator or RedSquirrel yourself to test it.--Flibble (talk) 17:03, 13 January 2011 (UTC)

https://groups.google.com/group/comp.sys.acorn.tech/msg/9a314c6c5746eb9a?hl=en --Flibble (talk) 17:04, 13 January 2011 (UTC)

Despite what I said above, I'm pretty sure that I do have the RiscPC launch magazine... Anyway - I also found this [2] which shows an A5000 running RO3.19, and nary an anti-aliased font to be seen. a_man_alone (talk) 17:12, 13 January 2011 (UTC)

Well, it's the Jan 1989 reference for the moment then (I'm assuming the magazine was published in Decemeber 1988). Technically, I suppose ArcDraw is a separate application (can it be confirmed that it was included as part of the OS?) Also, was the DrawRender module around in those days? --trevj (talk) 15:07, 14 January 2011 (UTC)

The "Archimedes user guide", published 1988, writes about 'anti-aliased' fonts in the chapter "Fonts". quote:

The desktop and most simple applications use the system font,........
In addition, there is a second kind of text style, which makes use of so-called 'anti-aliased' fonts.

Arthur included a program for anti-aliasing fonts [3] page 70

--Egel _Reaction? 18:57, 13 January 2011 (UTC)

I've read that but it sounds a lot more involved than ArcDraw. But add it as a reference and bring the date forward if you feel it's justified in the context of the article. --trevj (talk) 15:07, 14 January 2011 (UTC)

These are items I've tagged as ^{[dubious – discuss]} on the article. That means that unlike ^{[citation needed]} I beleive them to be incorrect, or not correct enough to remain as the definitive statements they appear to be.--Flibble (talk) 16:55, 13 January 2011 (UTC)

"RISC OS remains with a co-operative multitasking system. Although this is preferential for RISC OS' many embedded applications"

I've never seen a reference, quote or opinion from any party that could be considered non-biased (e.g. not Acorn trying to sell it too you) that CMT is preferable to PMT for embedded apps. I'm not even sure you can say "RISC OS' many embedded applications" without some further examples and references.--Flibble (talk) 16:55, 13 January 2011 (UTC)

"Consistent look and feel across all applications"

With regard to the style guide, it's only a guide. There are many apps that didn't conform too it, as the OS itself placed no constraints that forced apps to conform. Examples of 'famous' RO apps that didn't conform include Computer Concepts 'Impression 2', some versions of Oregan Developments 'Digital Symphony' in addition to hundreds of PD apps. If more examples are needed, I'll start listing them.--Flibble (talk) 16:55, 13 January 2011 (UTC)

As there appears to be no real contention about those "dubious"s, I've removed one and rewritten the other. If you want to add them back, please find a reference.--Flibble (talk) 14:27, 19 January 2011 (UTC)

In adding (pronounced /rɪskˈoʊˈɛs/), that's my understanding of the intended "correct" pronunciation. Not being a phonetician, what I'm unsure of is the stress. Perhaps (pronounced /ˈrɪskoʊˈɛs/) would be more accurate. Sources RISC, RISC, O, S. --trevj (talk) 13:03, 18 January 2011 (UTC)

I added a reference for this from the RISC OS Open site. In future when adding info, even stuff that seems trivial, please add the reference where you got that information from.--Flibble (talk) 14:23, 18 January 2011 (UTC)

Thanks. Will try to remember that for all future stuff. --trevj (talk) 16:00, 18 January 2011 (UTC)

If we're fussed about this (and I know it's unimportant compared to referencing sources), the template at riscos.info could be imported here. There's already the related Template:Spaces here - I guess a new template can simply be created, unless there's some policy I've missed. In the longer term, perhaps there could be a bot to automatically replace occurrences of a standard space. --trevj (talk) 10:50, 25 January 2011 (UTC)

If automated by a bot, use of the existing Template:Nowrap would do it. --trevj (talk) 13:53, 26 January 2011 (UTC)

There has been previous talk covering style bots. --trevj (talk) 14:42, 26 January 2011 (UTC)

Incidentally, Mac OS X doesn't bother about it. --trevj (talk) 14:42, 26 January 2011 (UTC)

Unless there's a suitable bot within this list or elsewhere [1], [2], I suggest dubmitting a bot request. Such a request could be to augment the specified text (assuming bots can have parameters passed to them) with {{nowrap}} or {{nowraplinks}}. If such a bot is created (or already exists or can be adapted) then other WikiProjects may also find it useful. --trevj (talk) 13:28, 24 February 2011 (UTC)

Manual of Style, for reference. --trevj (talk) 13:53, 26 January 2011 (UTC)

At the moment the ordering of the post acorn sections is a little haphazard with it being difficult to follow who's doing what and how it relates to each other. So I suggest moving from this

* RISC OS 4
* RISC OS 5
* Shared Source Initiative
* RISC OS Six

to something like this

* Work Post-Acorn by RISCOS Ltd
 * RISC OS 4
 * RISC OS Select
 * A9Home
 * RISC OS Six
* Work Post-Acorn by Castle Technology and RISC OS Open Ltd (and potentially mention Pace if I can find enough references)
 * RISC OS 5
  * Iyonix PC
 * Shared Source Initiative
  * Iyonix PC
  * Beagleboard
  * IOMD
* Licensing Controversy

Note I've added a licensing controversy section, this is so all talk in the two sections above that's not releases/features added/development etc can be seperated and not get in the way of the less contentious details. In the Controversy section we would have to be very very neutral and only state publically available info, perhaps finishing on the note that despite all the controversy both strands of development continue to ship without issue.--Flibble (talk) 14:16, 31 January 2011 (UTC)

In the RISC OS, Arthur and NCOS Module Database you can find 3 RISC OS / NCOS versions made by Pace: NCOS 5.13, RISC OS 4.00 (Pace 20000808 aUUP00-00) RISC OS 4.00 (Pace 20010402 Ursula0020). article on Drobe --Egel _Reaction? 10:45, 1 February 2011 (UTC)

I know, but that's my webpage and I wrote the article on Drobe ... and I'm not allowed to cite myself :-) more importantly I know that those Pace builds of RISC OS (not NCOS) were internal and as such are all rather hush hush and not meant to be public knowledge, so finding reliable confirmation of what they were and what they represent is very hard. So despite the fact that I think they're an evolutionary step on the road to RISC OS 5 I can't say that (original research) as I have no verifiable reference for it. The NCOS one however is probably best referenced by Chris's Acorns. Incidentally I think there should be an additional section on NCOS and RISC OS's use in embedded applications.--Flibble (talk) 17:10, 1 February 2011 (UTC)

As of today I did a large rewrite of the post acorn section. I think the detail for ROL work is now pretty good, though still has several ^{[citation needed]} in it. The Castle/ROOL section still needs that level of detail adding, but it seemed a large improvement so I decided to merge now. I've not yet written a section on the licensing controversy, that requires me to build another reference timeline.--Flibble (talk) 16:45, 23 February 2011 (UTC)

Regarding the proposed licensing controversy section, have you considered referring to Fork (software development)? Or perhaps it could go at the end of RISC OS#Demise of Acorn Computers Ltd, by way of introduction to 'Work post-Acorn by RISCOS Ltd' and 'Work post-Acorn by Pace, Castle and RISC OS Open Ltd'. --trevj (talk) 09:38, 22 March 2011 (UTC)

I recently largely rewrote the Phoebe (computer) page. In that there's now a fairly good description of the improvements that went into the Phoebe version of RISC OS that ended up being released as RISCOS Ltd's RISC OS 4, do you think it's worth duplicating in the RISC OS article?--Flibble (talk) 13:54, 3 February 2011 (UTC)

Nice one. For use of the fairly good description within RISC OS, it might be best to leave it a week or two and see what editing tips we can borrow from members of WP:COMP. They may have documented similar occurrences elsewhere. It may be preferable to include the detail within RISC OS, with only a summary/cross-reference within Phoebe (computer). --trevj (talk) 15:04, 3 February 2011 (UTC)

Could the RISC OS cog be used within the pages? Permission/licensing would probably need to be obtained, e.g. [4]. (Did Richard Hallas also create the Iyonix logo?) --trevj (talk) 12:54, 4 February 2011 (UTC)

Not sure if he did the Iyonix logo, but yes Richard Hallas did the Iyonix switcher icon, with the cog shadow, and the rest of the sprites for RO5 [5]. I believe there's support of uploading non-free (e.g. work still under copyright) under certain circumstances, check out the rationale on File:AcornComputersLtdLogo.jpg--Flibble (talk) 14:11, 4 February 2011 (UTC)

Incidental I shall make no comments about whether or not the cog logo is any good nor it being entirely derivative of the KDE logoFile:KDE_logo.svg, oh--Flibble (talk) 14:22, 4 February 2011 (UTC)

Yes, I meant the switcher icon. How long has the KDE logo been in a recognisable form with that cog appearance? There are certainly similarities, but it could be a coincidence I guess. Have you read any discussions of this elsewhere? I'll probably contact Richard for clarification. ----trevj (talk) 08:33, 6 February 2011 (UTC)

It is a coincidence from what I've understood before, but the KDE logo has been a K on a cog since version 1 (if not before), so about 1998.--Flibble (talk) 19:00, 6 February 2011 (UTC)

Much to your expected disappointment, you'll see that the logo is now uploaded. I'm now waiting for comments back from Richard Hallas and Paul Middleton regarding the copyright and its usage here. --trevj (talk) 09:57, 17 February 2011 (UTC)

The logo sucks, but that doesn't make it inappropriate to the article (NPOV and all that). However note that the cog logo has been taken off the RISC OS project page, fair use images can only be used in article space (and then only with one fair use rationale per article).--Flibble (talk) 20:57, 17 February 2011 (UTC)

Yes, I'm now aware of that mainspace exclusivity. Anyway, I think there will be some way to get a recognisable representation (perhaps own work) onto the project page again. Incidentally, I've recently exchanged a couple of interesting emails with Richard Hallas, which I'll summarise here some time. --trevj (talk) 16:19, 18 February 2011 (UTC)

NOTE: No new material can be used in a Wikipedia article, as it would count as original research. The many aspects discussed below (aside from the important image licensing conditions) would have to be published elsewhere first.

Richard

With reference to File:Riscos logo generic cog riscosltd richard hallas 110x110.png, are you please able to clarify the usage requirements?

I can set the copyright position out in very simple terms.

Strictly speaking, RISCOS Ltd has nothing to do with the ownership of the cogwheel device. This is because the device was always intended to be a generic symbol to represent RISC OS, without ties to a particular company.

It's true that the symbol was first used by and issued through RISCOS Ltd; it appeared in Foundation RISC User, which was a magazine that I produced and RISCOS Ltd published, and was used in icons within an early issue of RISC OS Select that appeared at a similar time. This created a widespread but erroneous impression that the cogwheel was in some way tied to RISCOS Ltd and/or RISC OS Select (i.e. that particular version of the OS).

That was not the case. The cogwheel was created to satisfy a need for a generic, unifying OS logo to represent all versions of RISC OS. The Acorn nut had previously served that purpose, but the fact that its ownership was unclear (and it ended up being used by a maker of PC laptops) - and also that it existed in many versions, without a definitive design being widely available - meant that a new symbol to represent RISC OS was desirable.

I think that the logo has in fact achieved its intended purpose - i.e. it's now recognised as representing RISC OS generally rather than RISC OS Select or any other specific version - but I'm keen for it to be kept unlinked from any particular company for the reasons I've just stated. Please note also that it has been widely used by Castle in RISC OS 5-related matters, which diluted its association with RISCOS Ltd and helped to strengthen its independent nature.

Anyway, copyright.

The copyright of the RISC OS cogwheel designs belongs to me; I have never signed it over to anyone. I was not paid for my work in designing it; it was one of various things I did on an unpaid basis for the good of the RISC OS world at the time. I'm fairly keen to retain the copyright largely because I don't think it should belong to any single party with commercial interests, when no single company has sole responsibility for the development of the platform. To tie it to a company would be to compromise its impartiality and its RISC OS version-agnostic nature. I suppose I could give it some sort of copyright-free status, but (a) I have neither expertise nor interest in copyright-related matters, and hence know little about the various potential licenses, and (b) I'd be slightly concerned that, if it left my control, it could be used inappropriately (as happened to the Acorn logo, which ended up on Windows PC laptops and allowed a new company to trade on an old company's reputation without justification).

So, the fundamental position is that I have essentially released the logo for the free use of the RISC OS market, and I've always been keen to see it used as widely as possible in that context. I would never deny permission for it to be used on any genuine RISC OS-related product. I would merely like to reserve the right to deny its use in inappropriate situations, however far-fetched that scenario may be in reality.

It would of course be nice to be asked about its use, or the use of symbols derived from it, especially in platform-related situations. However, it's certainly not essential to approach me about the use of my symbol (or a derivative) on any RISC OS product, and I would never expect to deny any legitimate use of it.

Additionally:
Is the cog logo available in SVG format?

I've never done an SVG version personally, but there's no reason why one shouldn't be created, since the original vector version is available online.

What's the status of the RO4/RO5 logos?

The RISC OS 4 cube logo was designed by Paul Middleton and is the copyright of RISCOS Ltd. There's a 3D raytraced version of it that I created, and I suppose that version is my copyright, but the basic cube logo is RISCOS Ltd's and (obviously) relates specifically to RISC OS 4.

The Iyonix 'blue jellybean' logo is Castle's property, but relates specifically to the Iyonix PC rather than to RISC OS 5. I don't know who created the original design; Castle could tell you. I actually recreated it myself in ArtWorks when working on the RISC OS 5 icons, and it's my own redrawing of it that formed the basis of the 'switcher' icon (as used by the Task Manager) in RISC OS 5, but the basic logo is Castle's. As noted, I designed all the standard icons for RISC OS 5 (plus window furniture and other items), and since it was a paid job, the copyright is presumably theirs (though I don't actually recall signing anything). But that doesn't apply to the cogwheel because I had a conversation with Jack Lillingston that made it explicitly clear that the RISC OS cogwheel was an independent entity whose copyright belonged to me.

The rainbow-cogwheel logo used by RISC OS Open, despite its obvious similarity to my RISC OS cogwheel design, is nothing to do with me.

Incidentally, if possible, I'd quite like the file to be preserved as it exists now - i.e. the page full of alternative versions and logos based on the cogwheel, rather than the single green cogwheel icon. Does that sound reasonable/possible? It's true that the single green cogwheel is the most important of the graphics, and the 'primary logo' - and I support your use of it on the RISC OS page etc., presenting it singly as you have done. Nevertheless, the other versions with four cogwheels (one red, two green, one blue) have both been used in different situations and are equally 'official'. Indeed, the single green logo with "RISC OS" written after it has also been used widely. Ideally, the best way to preserve the logos would be to make that page full of official alternative designs available, rather than only preserving the single green cogwheel.

In due course, I intend to upload to Wikipedia the SVG version of the single cogwheel for use in the RISC OS article. I'd like to discuss the suggested wording.

Right at the start, under Description, it says: "This is a logo owned by Richard G. Hallas [...]". I'd quite like this to say "designed and owned by".

Good point. That had crossed my mind but there seemed to be nowhere to fit it logically. However, I've now done so. Please amend as you see fit.

Your solution seems to work very well, thank you.

Regarding hosting all of the files, I think it would be most appropriate to host them at riscos.info. As you say, ArtWorks images can't be hosted [here] at Wikipedia - although SVG versions of everything could in theory be hosted [here], provided they were legitimately referenced to from the article. The problem with that is that an editor could claim that such inclusion is not strictly necessary to the understanding of the topic and could therefore remove them as a contravention of the 'Neutral Point of View' policy.

I'm not entirely sure I see the neutral point of view argument, given that the other logo variants are just as official as the basic single green logo. However, you're clearly more familiar with these matters than I am, and I accept your judgement.

Well, it's not appropriate to promote things for the sake of doing so. If the article warrants their inclusion, then hopefully it's fair enough. If not, then they shouldn't be hosted [here] solely out of convenience.

I suppose my point of view would be that the set of logos was originally supplied (i.e. the ArtWorks and Draw versions) *as* a set. Certainly there were preferred versions within the set: the green cogwheel alone as the primary OS logo, and the single green cogwheel with the Gill Sans text "RISC OS" following it as the primary /descriptive/ OS logo. But the other variants are also there for equally free use, and some of them have indeed been used widely. Notably, the four cogwheels in a square arrangement as the Task Manager icon in RISC OS Select, and also within the Module filetype icon in RISC OS 5.

What's happening here with Wikipedia is that the primary logo has been selected from the available set and the rest discarded. In a sense I don't mind, because it *is* the primary logo, and if you're going to have only one then it's definitely the right one to use. But it seems a bit arbitrary to throw the other variants away when they can be (and have been) used similarly to represent the OS.

If there has to be a direct one-to-one relationship between the graphic used on Wikipedia pages and the source version supplied as SVG, then I suppose there's little alternative to supplying just the one green cogwheel. However, I would prefer it if the other variants could be made available as well, just to provide potential users with a bit of variety. The other variants are just as 'official', and if a Wiki page were to refer to the logo, I see no reason not to mention that some other official variants of the basic design are also available for use if desired, because that's always been the case. I don't see this as promoting things for the sake of doing so; the other designs are just there as part of a standard set of alternatives, and that's always been true.

Anyway, that's my point of view for what it's worth. As I say, I'm not especially bothered either way (particularly if the full set is to be hosted on riscos.info), because the single green cogwheel is indeed the fundamentally important one.

Certainly, I agree that hosting the files at riscos.info would be an excellent idea.

With the files hosted there, they could be linked to from the Wikipedia article, e.g. perhaps from a new section under logo history/development.

Quite. Sounds ideal.

I'd also previously linked your uploaded logo to the WikiProject Compting template, which meant it could be included on pages within WikiProject RISC OS. Unfortunately, the logo was removed because non-free images cannot be used outside the "Article" Main namespace. It will remain that way for the moment. In the longer term, I can see three possible ways to resolve this situation:

1. We adopt completely different symbol altogether for WikiProject RISC OS.
2. Someone creates a new (low resolution, bitmap) logo. This could be influenced by your work but perhaps would not, strictly speaking, be yours (but IANAL). This could be licensed accordingly, hopefully being able to find something with some suitable restrictions (although it would actually be of too poor a quality for anyone to want to use in any meaningful way).
3. Advice is sought regarding licensing of your logo under free terms (but with restrictions on use) - I know you're reluctant to pursue this and that's perfectly understandable.

Well, creating a new logo is possible, but seems less than desirable unless there's a real need for it. On the one hand, creating a successful logo would require significant effort, and on the other, having a different logo from RISC OS would merely distance the pages from their subject.

I suppose a possible answer would be simply to derive a new logo from the existing cogwheel in a way that makes the link very obvious, and that's a possibility, but it's still much simpler to use the existing logo if that can be achieved. (This sounds similar to your point 2, except that I'm talking about using the existing cogwheel explicitly as an element.)

As far as I'm concerned, use of the logo in the way in which it was being used (until it was removed) was not only acceptable but actively to be recommended. Its use on the page was ideal, because it helps to promote the platform visually and give relevance to the pages that use it, and this kind of usage was one of the reasons why the logo was created in the first place.

If it were possible for me to grant specific permission for the logo to be used in this way on Wikipedia, then I would be only too happy to do so. It's all very well having a general principle about the non-use of non-free images, but it ought to be possible to override the generic case if a copyright holder gives explicit permission, as I would be more than willing to do.

As to the matter of releasing the designs under free terms as you suggest... I'm not totally opposed to the idea, especially if there were some way of preventing potential abuse. I'm just wary because of the theoretical potential for either abuse or loss of the logo's impartiality, and because I'm simply not au fait with legal matters and licensing (and frankly have no interest in the subject). If someone were to come to me with a proposal about licensing terms, though, I'd certainly be prepared to consider it. Overall, though, I see no compelling reason to do anything about it - other than that I don't want my ownership of the logo to hamper its adoption in places where it would be useful, such as on Wikipedia. I wonder if I could actually upload it to the template page myself and state that I'm allowing its use there because it's my copyright. Would that work?

As you'll see, the RISC OS talk page discusses this a little.

Thanks for that. I've now read what's on the Wikipedia talk page, in terms of the brief discussion. Just a few points in response, which you may or may not find interesting. (And should you happen to wish to quote any of this, feel free.)

1. I've always referred to it as the 'cogwheel' logo, not the 'cog' logo. Not significant (they both mean the same thing), but I prefer the longer name, as I think of 'cog' as being sometimes synonymous with the teeth on a cogwheel.
2. Whether one likes the design or not is a matter of personal taste, but it was kept very bold, simple and flat-looking for some very good reasons.

   For what it's worth, the logo is not especially typical of my work generally, and I was far from sure that I liked it myself initially, after I'd first created it. But it grew on me as I got used to it, and several other people have told me the same thing. I believe that Paul Middleton wasn't keen on it originally, but he's continued to feature it on his products and Web site, so maybe it's grown on him too.

   Justin Fletcher was hugely positive about it right from the start, and it was with Justin that I originally had discussions about designing something to replace the Acorn nut. The cogwheel design is mine, but it was designed according to criteria that we had discussed and agreed between us.

   So the key question is: does the logo successfully satisfy the criteria and requirements surrounding it? Before I even started thinking about what a generic symbol for RISC OS should be, Justin and I had come up with a long list of requirements that a RISC OS logo should ideally fulfil.

   Those criteria are set out in the 'Rebranding RISC OS' article that I wrote at the time I created the design. I go into a lot of detail there about what the logo needed to convey, and why, and also its potential needs in terms of simplicity (it's actually no more than a shape; even the colour is optional, though green is recommended if colour is available). The bottom line is that the logo satisfies *all* the requirements that were set out for it when I designed it, and it is also drawn with care and balance (e.g. one thing I don't think I mention in the article is that the central hole is at the Golden Ratio of the radius, and the thickness of the design is also determined by similar principles). Even ten years on, I can't think of another logo that would better satisfy all the requirements that were defined originally. Probably my only regret about the design is that I would have liked it to be a less generic shape, even though that would probably have meant it couldn't have been symmetrical.

   Whether you actually like it is merely a matter of personal taste. I'm sorry if some people don't care for it, but others do like it, and most are at least used to it. Personally, I consider it a pleasing symmetrical shape. But coming up with something *simple* rather than graphically exciting was one of the design goals, not something to apologise for.

   There's a big difference between designing a logo and drawing a pretty picture. The Linux penguin, for example, is a great picture and has lots of potential for cuddly toy spin-offs etc., but it really isn't a successful logo because of the limitations in the ways in which you can use it. It's far too complex to be a successful logo. In my view, one of the world's best logos is actually the Apple logo, but its strength lies in its simplicity: again, at heart it's nothing more than a simple shape, with which you can do interesting things. The Acorn nut logo was actually in exactly the same class as the Apple logo, and really just as good; it's a great shame that it couldn't survive. I'm not saying that my cogwheel logo should be elevated to that level: clearly it shouldn't; it's not as distinctive a shape, and maybe its symmetry - a benefit in some ways - works against it in this respect. Perhaps its shape is too generic. Nevertheless, it does try to satisfy the same basic set of needs and requirements as those other 'simple' logos, as well as having RISC OS- related symbolism. I genuinely believe that it fulfils all of its requirements and is hence a successful logo.

3. If I'd known about the KDE logo I might well have done something different, but I'd never knowingly seen it until some bright spark decided to accuse me of plagiarism. Any similarities are purely coincidental, and the rationale behind the design, which is all written up in the source article, should make the reasons for my choices abundantly clear. The claims that I copied the KDE logo have become quite tiresome, and are easily countered by the detailed explanation for the design that I gave in article form at the time of its release.

As a final point, I'd say that it's impossible to satisfy everyone regardless of what you do. I was personally extremely proud of the work I did on the RISC OS 5 icons, for example. I consider them pretty much wholly successful (notwithstanding a few compromises that I had to make along the way, such as with regard to the hard drive icon and Task Manager icons, which I've never really liked). I'm personally very proud of them, and seeing them in use on my Iyonix still gives me pleasure. Other people have also been extremely enthusiastic about them, which was very gratifying. At the same time, though, I've had what amounted to hate-mail from some people about them, and enough of it to make me extremely upset at one point. It was amazing how diametrically opposed and irreconcilably incompatible some of the views expressed to me were, about how I had (or ought to have) gone about the approach to the designs.

But once again, there's a *very* long article explaining the design process, written just after I'd finished the work, explaining the underlying requirements and the reasons for some of the design decisions. The bottom line is that I'm very happy with what I did here and feel that I achieved what I was aiming for. Moreover, I'm convinced that I took the right approach, and would do the same even if it were possible to go back and change my mind. But the one thing that the experience proved conclusively to me is that it's impossible to please everyone, especially with something as apparently emotive as graphical designs.

By way of conclusion, I'm very happy to engage with anyone who is still involved in furthering the interests of the platform and wants to discuss my past contributions to it, whatever they may be. Even if they don't actually like them!

— Richard Hallas in private emails to/from User:Trevj.

I've just noticed the existence of Template:Infobox OS version. Splitting the OS details up into separate articles may seem a bit of a pain (and would be a backward step in terms of Arthur (OS)) but perhaps it would facilitate improvement in clarity of the main article. Screenshots from each version could easily be incorporated for historical purposes without detracting from the main content. It would also mean that specific hardware articles could link to appropriate OS articles. A version number table could be included in RISC OS as a quick summary. Sorry I didn't suggest this more recently! What do people think? (Note: perhaps now isn't the right time to do this, in which case it could be considered for the future.) --trevj (talk) 12:05, 6 February 2011 (UTC)

I say edit everything in this article, then lets decide if it's too big and worth splitting up when we have more content.--Flibble (talk) 16:31, 6 February 2011 (UTC)

OK, makes sense to me. --trevj (talk) 18:30, 6 February 2011 (UTC)

This is an archive of past discussions about RISC OS. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.