How does one publish/promote a new package?

I just took all the wonderful information supplied here, and wrote a small package that makes it possible to use Unicode characters for section, subsection, subsubsection, paragraph, subparagraph and item. I think it makes the text look neat, and it might be very useful for RTL users.

\NeedsTeXFormat{LaTeX2e}
\ProvidesPakcage{usection}
{2011/02/07 Unicode characters instead of section, subsection, etc.}
\newcommand*{\toEolnCommand}[2]{%
\newcommand*{#1}{%
\begingroup%
\escapechar=\\%
\catcode\endlinechar=\active%
\csname\string#1\endcsname%
}%
\begingroup%
\escapechar=\\%
\lccode\~=\endlinechar%
\lowercase{%
\expandafter\endgroup
\expandafter\newcommand\csname\string#1\endcsname##1~%
}{\endgroup#2\space}%
}

\toEolnCommand\toEolnSection{\section{#1}}
\toEolnCommand\toEolnSubsection{\subsection{#1}}
\toEolnCommand\toEolnSubsubsection{\subsubsection{#1}}
\toEolnCommand\toEolnParagraph{\paragraph{#1}}
\toEolnCommand\toEolnSubparagraph{\subparagraph{#1}}

\catcode"2022=\active \newcommand•{\item}         % Bullet
\catcode"B6=\active  \newcommandf¶{\@xparagraph} % ¶: U+00B6 PILCROW SIGN
\catcode"A7=\active  \newcommand§{\@xsection}       % §:  U+00A7 SECTION SIGN

\newcommand\@xparagraph{%
\@ifnextchar ¶ {%
\expandafter
\toEolnSubparagraph
}{%
\toEolnParagraph
}%
}

\newcommand\@xsection{%
\@ifnextchar§{% Two or more § were found.
\expandafter
\@sub@or@subsub@section
}{% Single § found.
\toEolnSection
}
}

\newcommand\@sub@or@subsub@section{
\@ifnextchar§{%
\expandafter
\toEolnSubsubsection
}{%
\toEolnSubsection
}%
}



So, (a) do you think I should publish this, and if yes, (b) how should I do so?

EDIT I correct my question to make it useful to the general public: (a) What are the cirteria to be applied in considering a package worthy of publications and (b) which steps should be taken to realizing such a decision?

-
You are trying to ask two questions at the same time (I think). One is about how to publish/promote a package, and the other is perhaps about getting feedback on the particular package you wrote. I think that both questions are quite good questions! But the answers will be probably most useful to you (and the community in general), if you ask these two questions separately. –  Juan A. Navarro Feb 8 '11 at 14:11
I think the first thing you should to if you decide to publish it is write some documentation, with examples. Look at indian.cse.msu.edu/pub/mirrors/CTAN/info/dtxtut/dtxtut.pdf for a tutorial on creating documented TeX source files. –  Jan Hlavacek Feb 8 '11 at 14:25
BTW: You are missing a % behind { inside the \@sub@or@subsub@section and behind a } in \@xsection (ok \@ifnextchar will gobble it, but anyway). –  Martin Scharrer Feb 8 '11 at 15:07
You should also make sure that all your macro names, espacially the internal ones, are unique. This is most commonly done by using the (maybe shortened) package name followed by @ as prefix. Example: \usection@xsection instead of \@xsection which might be already defined by another package. –  Martin Scharrer Feb 8 '11 at 15:09
@Martin: Although it's common to prefix macros with the package name and @, I realized recently that it's much better to suffix them. That way, if you index your documentation (which you should do), the alphabetization is meaningful. –  Ryan Reich Feb 8 '11 at 16:00

Martin has already explained quite well how to publish a package.

I think there is scope for a package that can make characters active and assign meanings to them. (I've thought about writing something similar before, actually.) But there are some tricky things to consider, and I think it is confusing the issue to start changing LaTeX markup at the same time — if your package also changes •, ¶, § in text then this, IMO, should be customisable. You mention it's useful for RTL users, but as I recall you're editing your TeX files in OpenOffice, which I suspect is somewhat unusual.

I'm also not entirely sure how robust these "grab until end-of-line" commands are. (Consider \foo{¶ something}.) But if you're only dealing with \section and so on I guess it's not too much of a problem.

If you only intend to keep the code as-is in its fairly specific form, I'd suggest not publishing it for the time being while you use it for your own work. If it does turn out to be quite useful to you over time then, yes, I would consider polishing it up and submitting it to CTAN.

-
I think this is a very important point that you make: do not publish until you found it useful for you, and experimented with it for a little bit. –  Yossi Gil Feb 8 '11 at 17:07
¶ could be made outer, since one can still use \paragraph and the like if a non-outer macro is needed. –  Bruno Le Floch Feb 9 '11 at 3:55
@Bruno: there's one for tex.stackexchange.com/questions/2012/…! –  Ryan Reich Feb 9 '11 at 14:22

To (a): why not, as long it is useful for a hand of people its worth the HD space on the CTAN servers

To (b): You should do the following steps

1. Convert it to a .dtx file, either manually (see dtxtut) or using sty2dtx.
4. ZIP the .dtx, .ins and the generated .sty and PDF as well as a short README together. The ctanify tool is also very useful for this.
True it is not mandatory for acceptance on CTAN, but I would recommend it. It should be pretty easy with sty2dtx anyway and having a PDF on CTAN simplifies things a lot for people which looking for the documentation online. @Ulrike –  Martin Scharrer Feb 8 '11 at 14:53
@Ulrike — I don't think this is true any more regarding the lack of PDF documentation. The CTAN (or perhaps TL) maintainers have even gone back through the old packages without PDF documentation and created it manually. (E.g., see texdoc printlen, which was not created by me.) –  Will Robertson Feb 8 '11 at 14:54
@Martin: This sty2dtx tool of yours seems pretty handy, thanks :). –  Bruno Le Floch Feb 8 '11 at 15:03
I think that as part of point 1, when you convert it to .dtx you should document the code. Well-documented packages are fun to read, sparsely-documented ones are frustrating. Undocumented ones are worthless for learning, and that's a shame since I've learned a lot from reading, say, the verbatim` package. –  Ryan Reich Feb 8 '11 at 17:44