On 19.02.24 23:12, Jonathan Corbet wrote: > Thorsten Leemhuis <linux@xxxxxxxxxxxxx> writes: > >> On 16.02.24 20:41, Petr Tesařík wrote: >>> Is this because you want to keep it readable if the target audience >>> reads the source text of the documentation? Otherwise, the .. include >>> directive does not make a difference after rendering to HTML. AFAIK. >> >> It less that I want that, it's more that I got the impression that both >> Jonathan and most of the kernel development community wants the source >> text to be readable; not totally sure, but I think that's the right >> thing to do, too. > > As a general rule, yes. To harp on this one more time, I do think we > could create sections of the manual (a "tutorials" book, say) with a > different set of priorities. Partly answered to that elsewhere in the thread: https://lore.kernel.org/linux-doc/2c5a82e1-31f0-4908-80b7-00b3b0257d59@xxxxxxxxxxxxx/ > In the documentation session at the last kernel summit, I got some > pretty clear feedback that plain-text readability could be made > secondary to getting the best rendered output, at least in some cases. For the record: I didn't feel any constrains while writing and would not know how to improve the "rendered output", except maybe by adding a image or two. But even then I'd say that's not worth abandoning plain-text readability. > Tutorials seems like a good example of such a case, where we could focus > on good web output without, as you say, creating potential maintenance > troubles going forward. It's just a gut feeling, but to me "split the text into smaller parts so those can be included in different documents" sounds like a much bigger maintenance nightmare than "keeping some sections in sync that two or three files (which most likely will be rarely changed!) use in parallel". But I fear our docs translators might have a different opinion. Ciao, Thorsten
