New Home Page Ideas

``Every good home page has a good quote in its heading bar.'' -- Keith Wansbrough.

I've decided (August 1999) that it's time to redo my home page from scratch, totally redesigning it (thanks to a number of people for this inspiration). But I want to do the job properly, and so I'm going to spend a lot of time thinking and designing first, before I do anything. I'm keeping these thoughts on paper and in notebooks originally, but I decided that putting them onto a web page would serve two purposes: it will help me organise the ideas better, and it will enable others to critique them before they become cast in stone (well, bytes).

Your comments are welcome: please email me on kw217@cl.cam.ac.uk.

Content ideas

• Needs more stuff about me: thoughts, ramblings, ideas etc. [inspiration: Mike Riddell's book God's Home Page, about how the Bible is God's home page, effectively.] Another example of a good home-page-about-someone is my friend Currie's home page.
• Clear all the existing stuff up: old stuff to `archival' section or (shock, horror) deleted (!!), external stuff (like the Lovelace stuff or the Pratchett stuff) removed.
• `Books you must read' section.
• Add comments on books to my books read section.
• Add my CV (at last).
• Add some genealogy info for Wansbrough-hunters.
• A `things I believe in' page, containing yes things like God, Christ, relationships, but also things like Greenbelt, WinterComfort-type stuff, etc. Currie's page has a good explanation in his own words of Christianity and why he's a Christian.
• Want stuff representing (an important part of) me to be clear and distinct from stuff I merely find interesting. Art movies, modern art, techno are part of me (and lots of other things, too!).
• Openness not definitiveness.
• A very good example of a large website involving a fair amount of self-revelation in an extrememly useful way is the Christian Think-Tank. Look at his journal entries and this discussion for examples of what I mean.
• Actually, there seems to be a fair bit of this on the web: here's another interesting approach to journalling: Leonard Sweet.
• Add a mini-photo album/gallery entitled The Adventures of My Adventuring Blanket.
• Add an annotated (image-map) photo of my room, as a way of explaining my weird life (thanks Carrie for this idea!).
• A New Zealand section with my favourite places. Scan in the photos I took in February 1999 when I was back for a month.
• Web Site of the Month list. This is a very bad name; what I mean is a web site that I repeatedly return to, one that has real content. I can only think of two examples: the WikiWikiWeb (design patterns site), and Unravelling Wittgenstein's Net: A Christian Think-Tank. Don't count news sites (eg, Slashdot) or reference (eg, FOLDOC).
• See Heather's comments in inbox/450, 16 Sep 1999, subject "Webpage".
• A brain-dump of potential pages / content (needs to be appropriately structured / categorised):

  Contact details. Work (now and past). CV (PDF, standard version, full version?, other details?, students?). Things that interest me. Things I believe are important / things I care about. Reading list. Top `n' books. Christianity. Current (and past?) research threads and interests. PhD thesis. MSc thesis. Honours project. Publications (short). Abstracts of publications. Code I've written, for distribution. Bookmarks / links. Top `n' links / web resources. My (view of) Cambridge. KeithNet (augmented with links?). Scotland. Glasgow. Linlithgow. New Zealand. Auckland. Carrie. Photo gallery. Bible study group. Humour. Archival (all old stuff goes here; online stuff is regularly pruned). What's New / change-log. Occasional journal / diary. Wedding page. Who am I really? What makes me tick? The Cambridge Adventure. People (could be dangerous...). Credits. Worldview: God, self, people (= community), structures, cosmos. Activities. Interests. Art: film, literature, writing, music, graphic design. Sci/Tech: maths, astronomy, physics, electronics, computer science, .... CompSci: theoretical, functional programming, programming, .... Creative writing. Technical writing and publications. Programming. Theory. Service: FORUM, Christian Graduate Society, ....

Conceptual structure ideas

• My target is not a cool site, but a well-designed site. (thanks Adrian for clarifying this).
• Should be a web not a tree. Look at the original Vannevar Bush article (from 1945!) that predicted what is now the Web.
• My site structure must be easily extensible (including for unexpected additions).
• Divided into different areas / facets of `me'. Broad categories.
• Should the tree (/web) structure have deep or wide branching? This is less of a problem if there is a lot of cross-linking; it's fun to keep browsing without having to go back to the top all the time.
• Semantic net: one mode of navigation of at least part of the site is as a semantic net: each node has a concept or word as title, and a para or so of discourse or discussion, and a list of linked nodes. Normal pages can be linked to as well as pure semantic-net nodes.
• Spatial / text adventure (ADVENT, a.k.a. Colossal Cave) metaphor, like Peter Gutmann's old home page.
• Links as a story rather than a straight menu (see Currie's page again).

Graphic design ideas

• Consistent design (but I can leave the design until I've thought through the conceptual structure, since this is logically prior and most important).
• Theme images (photos? icons?) for each section of the site (/of me); some identifying characteristic in the style of the page (like background colour). Style must still be consistent, however. This idea is tricky for things that overlap multiple categories, especially since I aim to have a web not a tree.
• Page layout: simulated-frames style. Divide page in 4. Icon in TL corner, heading in TR box, navigation in BL box, text in BR box. (see ActiveHaskell stuff, etc., or for another very nice example, Sven Panne's pages).
• A quote on each page (in the header section) (see book of quotes). I've found the perfect quote for the reading page: it's on p.32 of Don Quixote.
• Just read some interesting thoughts on being well-designed rather than cool. Ouch. Goes against some of the cool ideas I had, but I agree with a lot of the philosophy behind the page.

Technical implementation ideas

• Need to replace my current bookmarks system with a simple (Java?) applet that is as easy to use as Netscape's `edit bookmarks' (and perhaps takes stuff off my existing bookmarks list, so I can still use the same technique of `add bookmark' for an interesting site, then every so often spending some time filing). It should file things suitably on a web page (or several). Put a link to it on my bookmarks so it's as easy to invoke as `edit bookmarks', too.
• Currently my HTML is generated by a (Perl) script that processes the same HTML file. Better to have separate source and genHTML directories, and generate HTML from a separate file.
• This file should be in a custom XML format, and the HTML generated by a HaXML filter.
• `What's new' should be automatically generated (from info in the changed file itself). Site plan and/or index should be too. Thus the genHTML stuff should be able to process multiple files into a single one, not just one-to-one.
• Make all links relative. Thus preprocessor must be clever for [HOME] link etc.
• Use CVS on the XML source files, so I can access the history.
• An alternative: Andy Gill has some cool stuff mentioned to generate his pages; he mentions an Html pages generator library. This would be excellent.
• Would be nice to have a web access stats program that runs regularly, and gives info on sites (sorted by domain), which pages were viewed, in what order, etc. Should generate a nice, categorised report. This could be world-visible. [One idea for how to distinguish robots: are the images loaded?]

Back to my home page.

--KW 8-)