Extension talk:Semantic Forms/Manual/The forminput parser function
The description of this parser function comes half way down the section cryptically titled 'The two-step process'. I know its good to focus on tasks rather than technicalities, but I think the information here is so complex it needs to be presented technically instead of procedurally.
For example, the section starts with the title 'The two-step process', but then has sub-headings that deal with specific technical aspects of forminput tag usage. ... I guess the emphasis is, again, on procedure (i.e. based on use-cases), which is good, but my problem was I didn't understand what I was doing when I read this the first time... I still didn't have a clue what anything was doing in relation to anything else, so use-case style documentation was no good.
I don't want to come over as too critical (the information IS there, which is massive), I just thought it would be a good idea to motivate this rewrite.