WebSeitz/wikilog
Future Wiki Weblog Features
is seeking a full-time developer.

(backlinks off) (map off)
(search off)
last edited by BillSeitz on Feb 2, 2008 10:12 pm

Future things to consider tweaking in this code. Meta-question: should I prioritize on (a) making writing easier for me, (b) getting others to use same tools, (c) nudge (or wiki-specific variety), (d) other -related, or (e) non- () ( - actually, that's relevant to blogging too, but with lots more work)?

Upgrade code. Refactor my pages to fit more compatibly/optionally into official distribution. Get my changes adopted by official distribution.

Track down bugs in rendering [URLs] and [HREFs] (may get this for free if I just update the codebase).

Change create-page label (the "standard" is a question-mark) to something less distracting (the question-mark is especially confusing when printing a page, and even moreso when it occurs at the end of a question!). Maybe an asterisk? Caret? Plus-sign?

Explode [Wiki Words] for public search engine hits (and greater normal-human readability)? hmmm. (Related idea: put all exploded contained [Wiki Names] into meta tag.) (note some ugly cases like ) (note: doing this already for title tag, but want to do it in all body text words)

When rendering external [HREFs], also render a link that runs a search on the [URL]. And maybe a link search (except that the google search includes [URLs] on the same host, which often ends up including piles of junk).

Find on -s (not body content) (cache list for super performance!). Find "like" pages (same first word or same last word). ()

- provide feed of this

Cache and local [Forward Links] for each page, render with the page (in right column) to encourage their use. (This data could probably also feed /.)

Find "candidate" [Wiki Names]. (-s that don't have nodes)

Admin functions: , delete page

Better validation, support.

Cache of pages I blog.

When [URL] asks for a page that doesn't exist (maybe an artifically generated from an ), provide a create-new-page option.

Render diffs (or first lines, or line containing match) for pages in main listings. Return first lines with [Wiki Names] as search/ results (at least for z items)

Improve hack of views that skip page types. e.g. if I import a bunch of history, then the page could deliver an empty page as it cycles through 50 overall pages but finds that all of them are and not .

Take current page and generate long page combining all the pages directly referenced within (e.g. you could make a table of contents page for a "master document", and be able to pull together all the referenced pages into one big monster, maybe with aname links also).

Generate map for ? (maybe just a small graph of all nodes within 2 jumps of current page) (or ?)

link to

true server-based : Easier process for adding new blogbit. ? Just simpler web process (auto-assign id)? (And auto-ping weblogs.com?)

system of rendering static files and upstreaming to vanilla webserver, so people can run Zope on their desktop and not have to pay premium for zope-based hosting.

Grab Referers of each individual page (from outside). Add them to the view. (Maybe order them by most-recent-clickthrough?)

Features inspired by : http://hypertext.rmit.edu.au/vog/vlog/blog/32002/113tinde.html

automatically disregard singular/plural (and other weird variations in spelling of )

see MeatballWiki:MeatballWikiSuggestions

Other list-navigation views: MeatballWiki:IndexingScheme

For what it's worth, I have an channel of you over here via [Rss Distiller] in [Radio User Land]: http://www.decafbad.com/deus_x/radio/RssDistillerChannels/webseitz.xml --

A page diff feature would be really useful catching up on changes. --

See : | | | |


 



Bill Seitz, fluxent at gmail dot com, Weblog