tds 0.100 feedback

Ulrik Vieth
Thu, 19 Oct 1995 11:07:16 +0100

It's been a long time since I last read a TDS draft, but after
a fresh reading (without checking what's old or what's new),
here's what I noticed.

  modern systems.  In particular, this Technical Working Group (TWG)
  believes it is usable under Unix, \abbr{ms-dos}, Windows \abbr{nt},
  \abbr{os/2}, MacOS, and \abbr{vms}.  We hope that administrators and
  developers of both free and commercial implementations of {\TeX} will
  adopt this standard.

Why not: Mac\abbr{os}?  (For visual consistency with \abbr{os/2})

  Names and extensions may consist of \emphasis{only} the
  characters \literal{A}--\literal{Z},
  \literal{0}--\literal{9}, and underscore. 

What about hyphen?  It's so common in filenames on some systems 
that I'm really surprised to read here that it's not allowed.  
Could anyone confirm this?

  \item Names of {\TeX} input files must be unique only within each
  first-level subdirectory of \path|texmf/tex| and
  \path|texmf/tex/generic|; i.e., different formats may have files by the
  same name. (\ref{sec:Macros} discusses the macro structure.)
Please make this: Section~\ref{...}  (If that causes bad line breaks
you might say `{\TeX} formats' instead of `formats' to fill the line.

  The \path|dtx| files used for {\LaTeX} distribution contain
  both the program sources and the documentation sources, so they 
  are kept in the \path|source| tree.

I think saying `macro sources' would be clearer.  Speaking of `program 
sources' in the context of LaTeX might be confusing to some people.

  \item \path|config| is reserved for configuration files, whether used
  only by {\iniTeX} or not.  Configuration files are used by several
  formats; for example, {\LaTeX} and Babel.

Is Babel a format?  Anyway, I thing the matter of config files may need 
some clarifications. 

1. Should \path{config} be reserved for config files that affect the base 
   format (and closely related packages) or should config files for 
   individual packages go here as well? (Take the example of tdsguide.cfg 
   if someone wanted to install tdsguidle.cls as a LaTeX package/bundle.)

2. Should \path{config} be reserved to local/user generated config files
   while leaving distributed config files (like Babel's hyphen.cfg) 
   in the corresponding package directory? (Babel's configuration file
   is language.dat, which for some reason doesn't have a .cfg extension.)

3. What about packages that may have arbitrary config files, such as
   myletter.cls, which has a \usename{} feature to load config files
   for individual letterheads.

4. Should \path{config} be restriced to the local tree if there is 
   a distinction?

  Usually, this is usually the name of the {\MF} mode used to build the
  ^^^^^^^          ^^^^^^^ 
  \path|PK| file.

Drop the second `usually'!

  This information is easy to supply: a simple addition to the local modes
  used for building the fonts with {\MF} will automatically provide the
  required information.  If you have been using a local modes file derived
  from (or that is) \path|| (complete reference in
  Section~\ref{sec:Related References}), the required information is
  already in your \path|PK| files.  If not, a simple addition based on the
  code found in \path|| can be made to your local modes file and
  the \path|PK| files rebuilt.

Is it necessary to repeat this everytime when is mentioned?
(It appears at least twice with `complete reference in ...'.)

  The documentation directories may contain {\TeX}
  sources, \path|DVI| files, {\PS} files, text files, or
  whatever form of documentation will best explain the package.

Maybe add that it may contain TeX, LaTeX, MF, MP examples as well?
(Remember David Rhead's mails about whether to separate examples from 
documentation or not?  Personally , I can't remember the outcome.) 

  We recommend all implementations have default search paths that start
  with the current directory (e.g., `\path|.|').

Maybe add the Knuth recommends including \path|..| as well?

  \item{\path|libkpathsea.*| in} \replaceable{prefix}\path|/info|
              ^^^                                           ^^^^ 
You can't mean that.  Obviously a silly typo.

  \section{File Formats and TDS Locations}

  This section lists file types used in the {\TeX} system and their
  placement according to the \abbr{TDS}.  If you are unfamiliar with any
  of these, consult the references (Section~\ref{sec:Related References}).

It might be a nice idae to flesh out this list and add a more verbose 
description of what the individual file types are.  If I had more time, 
I might write up something, but not at the moment.

  This improves the maintainability of the font tree,
  but, unless all the programs that search this tree employ some form of
  caching, there are serious performance concerns.  For example, in order
  to find \path|cmr10.tfm|, {\TeX} would potentially have to search
  through all the directories that contain \path|pk| files
  in all modes and at all resolutions.

I don't understand this. Is this a mix-up or something?

  In the former case, it isn't easy to limit subdirectory searching to only
  the mode required for the particular printer you are using.  And in the
  latter case, leaving \path|dpi|\replaceable{NNN} at the bottom
  allows that level to be omitted on systems which allow long filenames
  (e.g., \path|cmr10.300pk|).

I find this confusing since in the second case |dpi| is not at the bottom.
Or did you mean `... not leaving |dpiNNN| at the bottom doesn't allow ...'?

  All discussions were held by email. The discussions
  are archived on \path|[twg-tds]|.

Maybe add alternative archive addresses, since has problems?

  \section{About the Committee}
  The \abbr{TWG} had no formal meetings (many, but not all, members attended
  \abbr{tug} '94).

And some attended TUG'95 and/or EuroTeX'95 as well.  I don't quite see what 
the reference to TUG'94 should imply here, or maybe it's just outdated.

  Rahtz, Sebastian
  (\literal{}). Production Methods
Obviously a typo.

  Vieth, Ulrik (\literal{}).  Graduate student
  in theoretical physics, maintainer of a local {\TeX} installation for
  a few months, adapted {\MP} to work with \application{Web2c} in early 1995.
This is getting out of date, better drop it.

As a general note: What about mentioning that other people participated 
in the TWG who aren't listed here.  Otherwise, readers might get the 
impression that VMS people are under-represented in this group, even
though we have Phil and Joerg keeping a watchful eye on us Unix people.

That's all for this time.


P.S. Joachim, could you arrange to mirror Karl's new drafts in the
usual location at, or would that cause them to
be mirrored on CTAN as well (which some didn't want as I remember)?  
For some reason, I always have connection problem's to Karl's site.