I'm finding this thread to be quite distressing.
On 26/04/2014, at 6:45, Philip Taylor <p.tay...@rhul.ac.uk> wrote:
> Wagner wrote:
>
>> XeTeX can do via xdvipdfmx specials almost everything explained in the
>> PDF reference and pdfmark reference. Will you insist to include these
>> 1000+ pages in the XeTeX manual?
>
> No, Zdeněk; I am asking for /important/ facts to be documented,
> not the entire known universe.
Phil, who is supposed to write the documentation that you desire?
You know that all TeX development is done voluntarily.
Most documentation is written by whomever wrote the code.
So you cannot expect any XeTeX developer to write beyond \special and it's
initial keywords.
Anything about the arguments to \special will necessarily be written by the
authors of the driver applications, or someone else who has kindly donated
their time to contribute their less than complete knowledge, based upon their
own specific experience.
Asking for anything else is unreasonable, and insisting upon it is arrogance.
Yes, you do need to understand that there is a driver application, even if only
one is currently supported with XeTeX, and that it currently — it wasn't always
this way — is called automatically. The TeX world has always been this way, in
that tasks are devolved to different application programs, and their
documentation is typically written independently.
>
> And how about the code below:
>
>> \bf\special{pdf: code q 2 Tr 0.4 w 0 .5 .5 .05 k 1 1 0 .1 K}Hello
>> world!\special{pdf:code}
>> \bye
>
> The number of such productions is infinite; no documentation system,
> no no matter how complex and complete, can fully document an infinite
> universe of discourse, and therefore unless you can show that whatever
> your write-only code accomplishes is something that B L User is likely
> to want to accomplish, then I can see little point in documenting it.
Of course.
So please take the obvious hints, face reality, and desist on pursuing this
thread.
Off list I have given you the advice of:
1. employing the miniltx.tex input file, to enable you to load important
LaTeX internals, without having to submit to LaTeX's model of what is a
document and how it might be structured; and
2. use \tracingall with LaTeX examples to see what is really happening.
And being prepared to open the package files themselves, to see what other
branches are possible with the package's internal coding.
Method 2. has always worked for me, as it reveals far more accurate information
than any documentation can ever do. Yes, it can be difficult and daunting, but
it is accurate.
I mean, if a computer can understand it, then surely so can I.
>
> ** Phil.
> --
> All duplicate recycled material deleted on principle.
A fine principle.
Please apply such empathetic principles also to developers, who supply their
efforts entirely voluntarily, and respect the fact that they may have a
different perception to you of what is important, and what is not.
Best regards,
Ross
--------------------------------------------------
Subscriptions, Archive, and List information, etc.:
http://tug.org/mailman/listinfo/xetex
