PRM font bounty

There is a problem to be solved and not much in personnel or resources to do it. Looking for an ideal answer feels a bit like hunting for a snark, and we may end up with a boojum. Does KISS apply?

I would like to return to “After considering several possible release formats it was decided that HTML was the best candidate”, because the problem has been faced once before, over two decades ago.

HTML is clearly not the best option now, when there are more modern methods available, but it does work. The export from FrameMaker has been made and the considerable tidying up by David Thomas has be done. It is workable as it stands.

The HTML source is clear and could be edited as it is, but it is even easier if imported to TechWriter to make changes and then re-exported. TW is good, because it retains the structure, but I am sure there are other ways to achieve this. TW can also export it as a PDF, but is nowhere near as good looking as the PRM we love. ;-)

I would also recommend Gerph’s ramble on this topic.

The HTML source is clear and could be edited as it is, but it is even easier if imported to TechWriter to make changes and then re-exported.

The problem is, and let’s be really clear on this:
The HTML copy on the CD is copyright ROL. The content is copyright PACE and will have passed to Castle and ultimately to ROD, but that does not make the HTML preparation a resource for ROD or ROOL unless the current rights holders (4QD??) allow it.
Pain in the proverbial rear, but real. Of course if Aaron were to open source it…

Re-exporting from Framemaker is the other option. With the downside of needing to repeat a whole load of editing.

a whole load of editing.

Probably unavoidable. There are three (and a half) sources which are all out of step with each other (printed/PDF PRMs, the wiki here, StrongHelp manuals, and the documentation on the website of the other branch (that pretty much ignores our existence)).

Plus there’s about a third of book one that can be thrown away (all the stuff about Archi era hardware). All of the examples are 26 bit (and far too few of them), some stuff is barely explained (anybody not called nemo fancy writing a keyboard handler?), a lot has changed in the three decades since the PRMs were originally published.

Plus, there’s the question about what to do about “the other branch”. Personally I’m of the opinion that we should be like the ROLtd site and simply not mention it at all. Because if we do (other than a brief overview), there are numerous API changes and the whole issue of 26 bit to deal with.

So, yeah, a metric *#&+load of editing is pretty much expected.

A question worth asking, I suppose, is:

• Is there some sort of software package (doesn’t have to run on RISC OS) that is easily available (preferably free) that is capable of generating a decent looking PDF, HTML/Wiki pages, and ideally some sort of other format that could be bashed into StrongHelp format with some BASIC code, all from the same source material, so everything can stay in sync.

Indeed, the copyright status is why I assume is it’s not been released thus far.

There’s one other resource, which is The Tekkie Disc that was released on CD by Emerald Publishing in 1993. Looking in there, the text is clear; it seems to be in a simple markup for fonts etc. The text says it is copyright Acorn but the files are copyright Emerald. Being 1993 there’s no PRM5 or 5a.

I wonder who was behind Emerald and whether they might allow release? There’s an ‘Emerald Publishing Group’ owned by Elsevier but they were only renamed in 2002, so I don’t think it’s them. Any ideas who that might have been?

Is there some sort of software package (doesn’t have to run on RISC OS) that is easily available (preferably free) that is capable of generating a decent looking PDF, HTML/Wiki pages, and ideally some sort of other format that could be bashed into StrongHelp format with some BASIC code, all from the same source material, so everything can stay in sync.

Yes, at least five flows have been mentioned on this thread (Docbook, AsciiBook, Sphinx, Latex, PRMinXML). All the tools to process those are free.

The problem is, and let’s be really clear on this:

Are you “really clear” on this? Just asking. ;-)

The only copyright claimed is by Pace. ROL is only named as the publisher. Perhaps that gives some other right, but it it limiting?

AIUI the words would be copyright Acorn and thus Pace/Castle/ROD. But the formatting (ie HTML) is work done by ROL and hence would be copyright ROL. I’m not party to the legal agreements but I would assume that, like other ROL changes, Pace did not assume copyright even if they might have been supplied a copy. The problem being that to allow the documentation to be open sourced it would need consent of the copyright holders to relicense it. Which, in the case of the HTML version, would need ROL to agree.

By starting with something before the ROL era, the question of their copyright doesn’t come into it, and it just needs ROD to agree.

(This is similar to the works of 19th century authors: the words are out of copyright but it doesn’t give you the right to sell photocopies of Penguin Classics. If you do your own typesetting of an out of copyright source text then that’s fine)

But the formatting (ie HTML) is work done by ROL and hence would be copyright ROL.

If I understand these things correctly, formatting (layout) is a design right, not copyright. Design rights lapse quite quickly, unless they have been registered, certainly since 2000.

If I understand these things correctly, formatting (layout) is a design right, not copyright. Design rights lapse quite quickly, unless they have been registered, certainly since 2000.

To save typing, just read Theo’s comments (Penguin Classic etc)

The thing here is that it should be seen as an opportunity to do a proper job.
Back then a quite laudable effort was made to produce a document with pictures/images that represented the then state of the desktop. Surely any new documents should, like others that have been updated, use the current state of the desktop. Thus the content is the important part and the graphical content from old versions is just an indicator of where it might be useful to have a graphic item in the new

Are you “really clear” on this? Just asking. ;-)

I’m really, really clear that the “unpleasantness”¹ in the RISC OS world during the early portion of this century is best buried down deep, perhaps where the earths mantle starts?

Presenting another cause for irritation isn’t a good idea.

¹ There may be more accurate phrases.

If I understand these things correctly, formatting (layout) is a design right, not copyright. Design rights lapse quite quickly, unless they have been registered, certainly since 2000.

No, design rights cover the form of objects. For example I can’t sell something that looks like a Coke bottle even if I make it myself and fill it with another liquid. Even if I did the CAD and manufacturing myself with no reference to Coke’s blueprints, everyone would associate my bottle with Coke. Hence Coke register their design to protect from clones.

Formatting of a book is just the copyright of the page in image form. If the words are out of copyright, the typesetting may not, and reproducing that typesetting may get you sued by the publisher. The main selling point of a Penguin Classic over another edition is the typesetting, since the words are the same.

Now it has been argued that if you take a copyright document and strip out the formatting all you are left with is the bare text which has no typesetting copyright, but there have been court cases in this area. At the end all you have are words and you then need to rebuild the formatting yourself, which is not a small job.

new documents should, like others that have been updated, use the current state of the desktop

That was what took the most time when updating the Impression manual.

Now it has been argued that if you take a copyright document and strip out the formatting …

That is what makes holding it in FrameMaker the ideal solution. Just revise it and output it in PDF or HTML and a completely fresh version is produced. No copyright problem with layout.

First, making the extant PDFs releasable with suitable fonts. I suggest this is done as Druck suggests by extracting the font from the current PDFs, pulling out the ‘metrics’ (ie the width and height of each character), and modifying an open source font to match the metrics. The font can then be patched back into the PDF. The position of each character on the page stays the same, but the look of each character changes a little. This is the way that Homerton, Trinity, Corpus etc were originally designed – they have the same metrics as Helvetica, Times and Courier but slighty different glyph shapes to avoid Adobe copyrights.

Anyone with font editing skills who might be up for this? It’s probably a few hours work in a font editor – the awkward part will be pulling apart and repatching the PDF (I’m sure there are Linux tools to do this, as well as commercial programs like Acrobat.)

I might be up for this.

Exactly how does one ‘as Druck suggests by extracting the font from the current PDFs’ and then ‘patch the new font name back into the PDF’? I can use my ‘skills’ with font editing to pull out the ‘metrics’ (ie the width and height of each character), and modify an open source font (for example from the EFF PD CD) to match the metrics.

To replace a font name with another throughout a PDF (I’m sure there are Linux tools to do this, as well as commercial programs like Acrobat.) Not even sure Acrobat Professional lets you do this.

Has ROOL mentioned which font or fonts are the problem ones?

As most of the Postscript level 1, 2 and 3 fonts have freeware metric compatible replacements already available which would be much preferable to making your own.

https://wiki.archlinux.org/title/Metric-compatible_fonts

On a slightly related note, I spent sometime to convert the risc os fonts to otf/sfd format. So there’s another set of metric compatible fonts for PS level 1 with more familiar names:

https://github.com/marutan/riscos-free-fonts

On a slightly related note, I spent sometime to convert the risc os fonts to otf/sfd format.

Many thanks for that. A small nitpick: Trinity’s equivalent is surely “Times”, as it predates Microsoft’s version “Times New Roman”? With Adobe/PostScript “Times Roman” being the equivalent to “Trinity Medium”.