git/list[1] front-page[2] threads[3] people[4] search[5] about
 

Re: [ANNOUNCE] GIT 1.5.3-rc4

From
David Kastrup <dak@gnu.org>
Date
Aug 5, 2007, 22:15 UTC
Message-ID
<857io9hblv.fsf@lola.goethe.zz>
In-Reply-To
<46B6446D.4030607@gnu.org>
Bruce Korb <bkorb@gnu.org> writes:
Show 7 quoted lines
> Linus Torvalds wrote:
>
>> AsciiDoc is *also* a source format. But the source format is
>> already readable IN ITSELF. Which is the whole point!
>
> Readable, just not writable.  It's markup language is a bunch of
> special characters that require familiarity to understand.

Well, one problem I find that the documentation of both Asciidoc and the connected Docbook toolchain is horribly subpar.

One part of the problem is that asciidoc.txt is written in Asciidoc, and so you can't pick apart markup from content in the explanations when reading the "readable in itself" documents.

Another problem is that most of the details of conversion to Docbook, and what results in what XML output, are completely glossed over. And then there are further problems in that downstream Docbook processors are documented even worse.

Then there is the problem that the markup can be redefined: the sectioning underlines explained in the asciidoc documentation differ from that _used_ in the documentation and again from that used in the git documentation.

Now I can, in fact, use docbook2x-texi --info --to-stdout user-manual.xml >user-manual.info and get a working info file: not just basically working, but quite fine (missing an index, though).

So there are "minor details" to fill in, like generated file names and info directory entries. Would you think that there is _any_ way of finding out how to represent this in Docbook, or if you find out that, how to get it from Asciidoc into Docbook?

Forget it.

Or things like including the manual pages in an appendix or elsewhere. Any chance for that? Slim, at least for me.

Texinfo source certainly looks less pretty than Asciidoc, but then makeinfo can produce plain text output from it looking quite like Asciidoc. And whatever you may think about Texinfo as a format and the generated info files and their readers: it is damn well documented.

Asciidoc is quite readonly in many respects at the moment for me, and I don't even know where to start in order to fix that.

And it does not help that there are multiple conversions involved with a pretty opaque in-between XML representation. In contrast, Texinfo has a single well-documented source format and direct converters to the target formats.

And its syntax is straightforward enough to write additional converters if one wants to (makeinfo can even produce Docbook output, so perhaps I may have a chance to reverse-engineer some required information from there).

Whatever. It is pretty clear that Texinfo is not going to be interesting enough to maintain for most git developers to continue this particular thread, as it won't progress beyond a simple and ugly advocacy and name-calling thread.

Anyway, the necessary structural information (indexing, directory info etc) presumably can be spliced into Asciidoc documents once somebody finds out how to do this, and then Texinfo can be _generated_ from it for those who need it.

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum
Previous: Bruce KorbNext: J. Bruce Fields
Message 37 of 59 in “[ANNOUNCE] GIT 1.5.3-rc4”
  1. Junio C HamanoAug 4, 2007
  2. Ismail DönmezAug 4, 2007
  3. Junio C HamanoAug 4, 2007
  4. Ismail DönmezAug 4, 2007
  5. Junio C HamanoAug 4, 2007
  6. Ismail DönmezAug 4, 2007
  7. Steven GrimmAug 4, 2007
  8. Junio C HamanoAug 4, 2007
  9. Daniel BarkalowAug 4, 2007
  10. Junio C HamanoAug 4, 2007
  11. Daniel BarkalowAug 4, 2007
  12. Steven GrimmAug 4, 2007
  13. Doug MaxeyAug 4, 2007
  14. Sam RavnborgAug 4, 2007
  15. Steven GrimmAug 4, 2007
  16. David KastrupAug 4, 2007
  17. Steven GrimmAug 4, 2007
  18. Johannes SchindelinAug 4, 2007
  19. David KastrupAug 4, 2007
  20. Jeff KingAug 5, 2007
  21. Linus TorvaldsAug 4, 2007
  22. David KastrupAug 4, 2007
  23. Linus TorvaldsAug 4, 2007
  24. David KastrupAug 4, 2007
  25. J. Bruce FieldsAug 4, 2007
  26. Linus TorvaldsAug 5, 2007
  27. David KastrupAug 5, 2007
  28. Linus TorvaldsAug 5, 2007
  29. David KastrupAug 5, 2007
  30. Linus TorvaldsAug 5, 2007
  31. Johannes SchindelinAug 5, 2007
  32. David KastrupAug 5, 2007
  33. David KastrupAug 5, 2007
  34. David KastrupAug 5, 2007
  35. Linus TorvaldsAug 5, 2007
  36. Bruce KorbAug 5, 2007
  37. David KastrupAug 5, 2007
  38. J. Bruce FieldsAug 5, 2007
  39. Man-pages in user manual (was: [ANNOUNCE] GIT 1.5.3-rc4)David Kastrup, Aug 8, 2007
  40. Junio C HamanoAug 5, 2007
  41. Miles BaderAug 6, 2007
  42. David KastrupAug 6, 2007
  43. Jeff KingAug 5, 2007
  44. David KastrupAug 5, 2007
  45. Jeff KingAug 5, 2007
  46. David KastrupAug 5, 2007
  47. Jeff KingAug 5, 2007
  48. David KastrupAug 5, 2007
  49. David KastrupAug 5, 2007
  50. Johannes SchindelinAug 4, 2007
  51. J. Bruce FieldsAug 4, 2007
  52. David KastrupAug 4, 2007
  53. Timo HirvonenAug 4, 2007
  54. Johannes SchindelinAug 4, 2007
  55. Timo HirvonenAug 4, 2007
  56. MichaelAug 4, 2007
  57. Robin RosenbergAug 4, 2007
  58. Julian PhillipsAug 4, 2007
  59. David KågedalAug 7, 2007

Read the whole thread, see it on lore, or plain text.

$ cat FOOTERMessages come from the public archive at lore.kernel.org/git, fetched every hour. The front page is chosen and written each morning by an AI editor and can be wrong; the threads themselves are the record. About and API. For agents: an MCP server at https://gitlist.dev/mcp, and any thread, story or person page as Markdown by adding .md to its URL (or sending Accept: text/markdown). Details in /llms.txt.