In the W3C land of XML and RDF (not to mention HTML, XSLT, URIs, etc, etc), we have a lot of acronyms. They don't work well in MoinMoin WikiName's. You can't say `RDFIsCool` as a plain WikiName. You have to write the term like `RDFIsCool`, paying explicit attention to the fact that you're making a link.

Thinking AboutThisService, should we change `word_rule` in `parser/` to "fix" this?

Some things like WordIndex will break, since there's no way for software to know where to split `IsRDFAGoodLanguage` (i.e. IsRDFAGoodLanguage). TWiki does allow such names. What do they do?

  1. How about wikifying everything that matches an extant page name? How heavy on the server would that be if the list of page names were kept in memory and shared globally (if that isn't done already)?
  2. Or how about just a list of acronyms (similar to the moin spelling page, LocalSpellingWords)? Are processed pages cached? Then, even if RDFA is a valid acronym, WordIndex would just list them both. I think you could disambiguate by always following known acronyms with a lowercase letter. e.g. IsRDFaGoodLanguage? MaybeRDFis. IsRDFAaLanguage? (Normally written "IsRDFAGoodLanguage? MaybeRDFIs. IsRDFAALanguage?" None of which work as one would like as I write this.)
  3. Thirdly, we could just write acronyms "Xyz...", perhaps capitalising them on display as they are recognised (this could cause problems if an acronym is a common word).
Man, that was nice capuccino. I think the second was made with espresso by mistake. There was a suspiciously large amount of frothed milk.

This is a possible project for MoinMoinToDo. Other options are discussed in [[[WikiNames|vs. Normal Titles]]] and UseInterWikiNamesEverywhere.

Last modified on 16 August 2005, at 04:03