[OS X TeX] gtamacfonts ligatures: PDF searchability

[Gerben Wierda]

I will largely stay out of this discussion, but I would like to make one
remark/clarification.

> On Mar 29, 2006, at 3:35 AM, Thomas A. Schmitz wrote:
>
>> 1. "At a guess the gtamacfonts package converts the fonts in some way,
>> munging the encoding which causes the difficulty you see"
>> This is patent nonsense. You can try this yourself: take a tool
>> like fontforge. Convert a font (say Baskerville) to  otf. Adjust
>> the mapfile accordingly so it will now embed the otf font instead
>> of the ttf (and run updmap if you're using LaTeX). Now copying from
>> a pdf will work flawlessly. The reason this difficulty crops up is
>> precisely because we did not want to convert the fonts and did not
>> want to violate Apple's license.
>
> Okay, how's this.
>
> The gtamacfonts package modifes a .dfont, reducing it to that portion
> which should constitute a valid .ttf --- but, because Apple has
> removed tables from them, the font doesn't work right, which
> apparently is how Apple wants it.
>
>> 2. "So you're converting from a Mac format .dfont to a set of .ttf
>> files
>> by extracting them."
>> Yes, that's the same kind of "conversion" that's taking place when
>> you unwrap your new computer and put it on your desk.
>
> But the .dfont doesn't stand in for the .ttf, nor vice versa --- it
> is a modification of how the font is packaged and Apple's license
> prohibits this.

As some people may know, when I first attempted this a few years ago, I
quit my attempt because Apple's license did IMO prohibit font conversion
(e.g. to type1). I wanted type1 (or another dvips/pdfTeX compatible
format, like maybe type42) mainly because I did at that time not make a
difference between TeX+Dvips+distiller and pdfTeX in usage.

What did make me reconsider and accept unpacking a .dfont into separate
.ttf files (and thus pdfTeX only) is that I have been told that this
unpacking process is also performed by the OpenOffice installer and I have
not seen Apple starting to wage a licensing war on OpenOffice (and how
does OpenOffice work with Baskerville ligatures, btw?). A second point is
that I do not redistribute the unpacked fonts, what I do distribute is a
method to convert them (as others do as well). With SIL Charis, I do
redistribute the fonts, though, as I am allowed to do that. Note also that
using fonts and redistributing the documents also sometimes redistributes
the fonts, so some font licenses may in fact formally forbid you to
redistribute documents you have made, which is plainly absurd. E.g. would
it in principle be possible to extract a Baskerville font from a
XeTeX-produced PDF document? If it is possible, would not distributing the
document be forbidden by th eliteral wording of the font license? I do not
have the specifics here, but something like that maight apply.)

My position in all of these legal situations is largely pragmatical. If
the de facto situation is that something is acceptable, I will possibly
support it/make use of it until I am told de jure to desist. Hence, gwTeX
has for years contained some packages (like prosper) that formally forbade
redistribution but allowed personal download (which in fact also means
that someone is redistributing, albeit passively) and free personal use.
In practice, this was meant as a measure against commercial exploitation,
but it is difficult to get these legalese texts changed. Hence, I would
take the stance that my tools only automate what a user could do if he/she
would act by himself (in the case of prosper, download and install). That
way, I was less strict than teTeX and TeX Live and could be more
supporting for the user (which in my pragmatic world take precedence over
legal millimeter-work that has in the real world no real effect, e.g.
prosper license and OpenOffice .dfont conversions)

As soon as a rights owner tells me that my actions are against the
(intention of) a license, I will desist. Given that no such actions are
taken (against OpenOffice or myself) I conclude that these actions are
acceptable.

I know that this is not de jure perfect, but let me stress once more: I
have chosen a pragmatic approach for my TeX redistribution works.

I do not have our discussions (Thomas, Adam and myself) available, but I
know we discussed this and concluded that this would be acceptable. If
not, Apple can easily point it out and the support for automation of user
actions (like automatically running fondu to unpack the .dfonts) will
immediately be removed by me from the TeX i-Package as it will have to be
removed from OpenOffice if that conversion there is indeed the case.

G

------------------------- Info --------------------------
Mac-TeX Website: http://www.esm.psu.edu/mac-tex/
          & FAQ: http://latex.yauh.de/faq/
TeX FAQ: http://www.tex.ac.uk/faq
List Archive: http://tug.org/pipermail/macostex-archives/