Goals are numbered: is there a reason for it? it suggests a descending level of priority. Would unnumbered bullet listing be better?
You are right. I just numbered it at Wikisym to know how many we had. Changed that now. --Christoph
:-) a minor thought no more. Cheers Christoph! --MarkGaved
What does "extensible by omission" mean? -- RadomirDopieralski, 2006-09-04
It was EugeneKim's idea, but I think he meant that "what is not explicitly specified is okay to develop further." I.e. You can add new things to Creole, unless there is already something written on that.
The other choice would be to have an explicit extension mechanism (like XML namespaces), but they are a bit too messy for this kind of work.
Eugene, feel free to correct me :)
-- JanneJalkanen, 04-Sep-2006
After the discussion on the Talk.BoldAndItalics i think Radomir is right about a missing terminology here, a missing goal. The question is: Should Creole be visual or semantic?
citing him on Talk.BoldAndItalics "visual" markup is much easier to introduce, while the idea of "semantic" markup might be even impossible to explain.
If i look at the two goals of the WikiSym
- Readable, 2 votes
- Easy to learn/teach, 2 votes
and maybe even this here
- Clear separation of markup & content, 3 votes
then i think this adresses this.
They could all be combined to form one goal: Visual Markup -> Therefore Creole should be more visual than semantic. It's an usability issue.
Combining the votes it would be a top goal. So the top five goals would be:
- Collision Free
- Cover the common things people need
- Extensible by omission
- Not new (Principle of least innovation
- Visual over Semantic
Pleas tell me your thoughts on this, vote for it.
I'd count "clear separation of markup & content" as a vote for "semantic", but that's just me. The term "content" is blurry at best, and the whole sentence can mean many things to different people.
About semantic/visual, I don't really have any preference as long as we agree on one of them. The current trend in development of Internet seems to be towards "semantic" things, making it easier for indexing and searching, and even for automatic processing of data. The visual tags are discouraged in modern dialects of HTML, and even obsolete. This is how professionals want to see the future of the Internet.
On the other hand, wikis are just tools "to have the job done" and on top of that "in the simplest way that could possibly work", which usually means delegating as much as possible to the humans using them -- including interpretation of semantics. I can bet that people *will* use the markup in many creative ways we haven't even thought about -- it happened before. Remember using single quotes for escaping wiki markup, or that weird rule for indenting text? Their "semantic" doesn't make sense, but they are rendered the right way. In the end, visual markup leaves more power in the user's hands.
And let's not forget what Creole s supposed to be. If a wiki engine it is used in has "semantic" markup, the text can be refactored to use it by the users who can grasp the meaning and who care. The problem might be clashes between Creole's visual markup, and similar (or identical) semantic markup used in the wiki in question...
It's hard to decide on this, I wish this topic was brought up on the workshop.
-- RadomirDopieralski, 2006-09-15