Why should any "standards" apply to WikiEngines?
- make it easier to use different wiki sites (not have to learn different SmartAscii rules, Wiki Name rules to create new nodes, etc.)
- make it easier to link from one wiki to another (WikiWeb) (e.g. Remote Wiki Links can't work well to link to a wiki which uses ID numbers as URLs instead of Wiki Name-s, or uses punctuation within Wiki Names, etc.) (see Wiki NameExamples)
- allow for creation of Rich Client apps (maybe to support Off-Line use/Data Synch, maybe for visual interface like Touch Graph/Idea Graph) that could work (InterOp) with any engine.
See also Picking A Wiki For Your Private Notebook.
Should this be structured as a Wiki Pattern Language? See the bold items below.
Specific potential standards
InterWiki - WikiWikiWeb:InterWiki was the original idea to have a single web-wide wiki NameSpace, transparently distributed across many servers. That idea was mostly dead/inactive, until new Federated Wiki projects started.
- update: in building WikiFlux I decided to break this rule, for the sake of normal-reader-friendliness, going with DoubleBrackets
- MeatballWiki:InterMapTxt is (?) the definitive list of possible URLs to look up; surprising-to-me, the 'Hostname' doesn't have to be Camel Case, just initial-cap (so 'Hostname:PageName' works).
- not surprisingly, this assumes that all servers use the same Wiki Name and page URI conventions (Page Name As URL); so if your wiki assigns some other ID as a URI (some use a number, others allow spaces in Wiki Names and replace those with underscores in the associated URI), those pages are less likely to get linked to by other wikis.
- UseMod:InterWiki also uses the Meatball Wiki approach.
- TWiki:InterSiteLinkRules seems to use at least the same naming methods, though the actual list management looks different.
- Zwiki, unfortunately, neither uses this list of hostnames, nor even this approach of having the list contained in a single page. Instead, it scatters them through single-wiki pages like Zwiki.
- the former queries those wikis in Real Time, while the latter stores a directory of PageNames.
- Zwiki doesn't even do title-only searches, I think
- does this belong in the wiki/server, or is in a Rich Client/Universal Inbox?
Sister Sites: a given wiki defines some "sister sites", and caches a list of valid Wiki Name-s for each. Then, for any page it delivers, it checks that list, and if finding matches, shows those "TwinPages" links (typically via icon) at the bottom of the page.
Unified Recent Changes: for a defined set of wikis (the same list used for Sister Sites?), provide a merged RecentChanges view. But note that RecentChanges lists often imply a higher currency (hourly?) than Sister Sites (daily?). Maybe RSS/Universal Inbox is a better approach to this? But note that most people use web-based RssAggregator-s like Google Reader, which can't even deal with BasicAuth for a closed wiki, much less other authentication schemes.
Zwiki, being based on Zope, gets Zwiki:WikiAcquisition. This can mean that when you nest one WikiSpace inside another, inner references to an existing outer Wiki Name will be recognized/activated. Sometimes this isn't what you want.
A different context/perspective
What's the smallest amount of specification you might demand?
To get some Knowledge Management benefits of people learning from each other, would want to make it easy for people to read multiple sites and link from one to the other. So Wiki Name and InterWikiMap features would help.
I could also see insisting on every wiki supporting an RSS format of RecentChanges. But you could probably implement that with some simple scraping from a central spider. (Not unlike an Intranet Search Engine.)
One would hope that people were working together across dept lines. Which means a given person could end up writing in multiple wiki sites. This would seem to push toward a single SmartAscii choice.
You'd probably want Hard Security (so authorized people could use spaces from outside the FireWall). You might not worry about which subset of the company is allowed to use a given space... which might allow Single Sign-On validation through something like a central POP/IMAP server.
How do I personally intend to increase adoption of Wiki Standards?
write up something establishing the position. That way people can find it.
have some running code which support those positions - key for credibility
seek input from authors of key WikiEngines, try to get congruence toward some of the potential standards
Start pushing it around in Social Software Alliance
- re their engine: http://www.socialtext.net/ssa/index.cgi?feedback
Idea for a better use of human intelligence --2003/09/28 09:18 GMT
No doubt the InterWiki under a common administration is the basic-anarchic platform for Collective Intelligence to unfold and Open Society and a World Government to be built upon. See http://www.terrahome.net.tc (link discussions for further Info in German and English on different wikis) --MattisManzel
2003/12/08 09:49 GMT
There is a wiki way to develop wiki standards. The standards you mention above were developed on Meatball Wiki by the collective efforts of all the major wiki developers. Inter Map Txt, InterWiki, Touch Graph link standard, Mod Wiki, and soon perhaps a Wiki Syntax standard were all done there, and I'm proud of that. Not to disrespect the Jsp Wiki guys who made the Xml Rpc For Wiki interface. --SunirShah