> written for plain TeX (acctually CONTeXT), but supports LaTeX (i.e. sty
> files, ifx when loading PiCTeX and using the \f@size [latex2e] and
> \@point [latex209]). This package would be latex4.
CONTeXT is a rather special case, as its effectively a rewrite of
the ideas of LaTeX with some modules that can be used with LaTeX.
i wouldn't classify it here at all, excellent tho it is. its
"context1" on its own
> 'not must have' and 'must have' depends also from the point of view:
> if I don't need e.g. chemical formulars, than these packages are not a
> `non must have'(latex3-4). For chemists those are real 'must haves'
in my book, one or two good chemical packages will be in 2b; one can
argue about which ones, but do you want us to keep chemstruct there
the LaTeX Companion effectively classified packages in the way we are
describing, by including a set of `known good and useful' material. it
descended from the CERN local guide. we all do, why pretend otherwise?
any of us who maintains a TeX setup for others keeps a set of packages
available, and recommends them to people. when people come to me and
say `how do i do Harvard references', i don't say `well here are 6
packages which have worked once, try for yourself', i say `use
natbib'. i am sure most of us say to people, `use the graphics package,
stop using epsf.sty/boxedart/psfig/psbox etc etc', dont we?? of
course, in some cases we say `you choose between ppchtex and xymtex,
it depends on your needs and working methods', but often not.
Bias? Aggressive to package writers? Sure. When I recommend a novel to
you, I say `I think Antonia Byatt is a very good writer, I think you
should read Possesssion', i don't say `i have 500 novels in my house,
pick one, any one, i won't advise you'. Some friends who know more
than me read the reviews, go to the bookshop etc, others are pleased
to have a recommendation. Antonia Byatt won the Booker Prize for
Possession, that makes it a `book1' package, does that mean we
shouldnt give prizes?
sebastian `fascist' rahtz