Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

[Home]GoodLinkStyle

MeatballWiki | RecentChanges | Random Page | Indices | Categories

See AvoidClutterLinks

Should we try to make links and pages self-sufficient? I feel we should follow a link if we are interested in its subject, but we shouldn't have to follow a link to discover what that subject is.

I think it is almost never good to have a page consisting entirely of a link; there should always be some explanation. Often that would encourage debate local to this Wiki.

On the other hand, maybe that would just add redundancy, and it's only old fuddy-duddys like me that resent link-chasing.


I absolutely detest link chasing. If all you have is a link, then include it in the page where you reference it. Don't make me bounce to another (shallow) wiki page and then to the link. This practice reminds me of the early commercial sites that were (still are for all I know) afraid to use offsite links for fear of losing traffic. It's user-hostile. Don't do it.
It's good style to expand WikiQuickies. If you only have a URL, at least go to the site in question and extract a paragraph that effectively summarizes the content. This would include foreign Wiki links. The OtherHypermedia pages are failing badly at this, however.
Another question: I created this link directly; no page links to it. This is contrary to custom, and arguably defeats a benefit of the Wiki approach that everything be connected, exist in context and have a need. Have I sinned?

No sin. Some people remember orphaned pages and link them as needed. The search, AllPages, and RandomPage will also retrieve this page. If we link to an existing page, the BackLink will bring us here.

I think it may be affected by whether you see WikiAsConceptDictionary?. MeatballWiki started off like that, but I think it may have drifted away from that model, which I think may be a good thing. Some pages make sense as standalone articles.

I created MeatballWikiFaq? as a kind of manual RecentChanges, as a place where pages like this could be linked in. I don't think it's worked; probably I made the name too specific. Suggestions for a better one? How about, PleaseReadMe?, or BurningIssues?? -- DaveHarris

I can see some need for a manual list of interesting pages. (A simple list of all pages can be provided by the index action.) Even though the Meatball RecentChanges keeps all changes (at least for now), most people won't look past the default number of days. A link to the manual list could be placed at the top of RecentChanges (which is just another wiki page) for easy access..)

One good way to manually list pages would be by high-level categories, like people, sites, Meatball-specific, and wiki ideas. Each category could have a list of pages, with a brief description. (If a page isn't interesting enough to describe, it's probably not interesting enough to list.) --CliffordAdams

By BurningIssues? I meant not just "interesting" but somehow "current", as in, "this is a new thing which I've written which I offer for peer review". Pages would be moved off as they get old and/or better places are found for them. RecentChanges does some of that, and is presumably how people found this page.

A categorised index should also be useful. WardsWiki has 2, one just a hierarchy of normal pages (which is not updated very often) and one generated by searching for CategoryXXX? (ditto).

Here's a more-or-less concrete suggestion in an attempt to refine the idea a bit more. Have a page called, eg, NewPages?. The system automatically adds links to new pages to it as they are created. This is a "major edit", refreshing NewPages? in RecentChanges. Pages are not automatically removed from NewPages? unless they contain a CategoryXXX? link. There may be other criteria, eg based on page age or the size of NewPages? itself. (Removing a page is a minor edit.)

Such a page could be generated automatically (and indeed retrofitted to WardsWiki with a script, if someone wants to try it). It might be better if pages were only removed manually, though. -- DaveHarris

As for the name, completely general names like "PleaseReadMe?" are probably not a good idea. I had a similar problem trying to find a name for Wiki:ChangeSummary. I found the best way to choose a wiki name was to list all the main words I wanted to describe, then select two or three of the most important words. What are the important ideas of the page?

Reasonably soon I hope to add page deletion and renaming so that choosing names doesn't have to be so difficult. (If you get it wrong, you'll be able to have the name changed.) In the meantime, perhaps a page like PageNameSuggestions would be a good idea? People who want suggestions for names could describe their idea and ask for community suggestions. (If someone likes this idea, please create the page, and I'll fill it in.) --CliffordAdams

ZWiki uses the "HotTopic?" to tag interesting pages (cf ZWiki:HotTopic). Interested readers can search for the backlinks of "HotTopic?" to find interesting discussions. I think they use it a little too liberally there, personally. Also, since ZWiki hosts a WebLog-like FrontPage (semi-templated as well), it is possible to put the list of HotTopics? there. Or, even better, the list of the ten most recently changed hot topics. -- SunirShah (The ironies: today, as I post this, the HotTopic? list has been removed from the ZWiki:FrontPage.)

We can already RenamePages by creating a new page, copying the content across, and removing it from the old page. Admittedly this leaves the old page as a stub, but I don't think that's a big problem and it can be used to store a pointer (or REDIRECT) to the new page. The problem area is updating other pages to use the new name, and that really needs to be decided by a human on a case by case basis. -- DaveHarris


I'm considering adding more tools for categories. (These tools could be part of the 0.9 or 1.0 version of UseModWiki.) Categories will probably become metadata, rather than part of the page content. Indexing tools could use the category data. I might even make categorization as easy as selecting from a drop-down list, or a text field similar to the current search field. (The category tools might even be displayed on every page). Suggestions and comments are welcome... --CliffordAdams

How would they work? If they look anything like Yahoo!, forget it. The Yahoo! categorization system is pretty pathetic. Anyway, the more I think about it, the less I'm enamoured with categories. It's too difficult to assign a moniker/motif to a page in order to relate it to a set of other pages. After all, that's what linking is for and it's far more fluid. The big deal is that all links are egalitarian. So rank them. But a "problem" with ranking systems is that they are self-influential. Ok, more questions than answers, but I did offer one idea with InterestMap which could be tied with PersonalCategories into a mini-ViewPoint. -- SunirShah

Don't worry. Categories will be optional, and those that don't like categories should be able to completely ignore them. One reason for the metadata approach is so that category edits do not pollute RecentChanges. Users should also feel extremely free to categorize, since they won't be distracting the RC-watchers. Tools like CategoryFilteredRecentChanges and category-sorted search results should be possible.

The first implementation will probably add a small text area (2-3 lines) to the edit page for the categories. Initially, PersonalCategories would be run on an honor system. (If a secured version is developed, it could use a separate metadata area.) Tools will be the main users of categories, although I may add a user option to display categories on each page. Soon afterwards I hope to add (optional) tools to make categorization easier. --CliffordAdams

Putting the categories into meta-data would also allow the distinction between talking about a category and assigning a category. Do a backlinks search from CategoryHomePage and you'll find a lot of non-homepages (including this one, now). The distinction can be somewhat useful for human readers (redundant except on some edge cases), but particularly useful for non-discretionary automata ... isn't there some script or other that deliberately ignores pages with CategoryHomePage on them? --EricScheid


CategoryMeatballWikiSuggestion CategoryLink

Discussion

MeatballWiki | RecentChanges | Random Page | Indices | Categories
Edit text of this page | View other revisions
Search: