phil wilson :: a geek commodity

Apparently it’s all the rage on the interweb but I’ve never really used IRC all that much, and since Firefox is my main browser I haven’t got easy access to ChatZilla, or so I thought. Of course, it’s available as an extension and works just great with Firefox 0.8 and is even executable from the command line with a quick

firefox -chrome chrome://chatzilla/content/

Finally I can keep up with what's going on over in #foaf!

AOL: New Mozilla-based Netscape 7.x coming shocker

AOL are going to be releasing a new version of the Netscape 7 browser based on a recent Mozilla milestone. This is really awesome. Despite AOL’s financial, uh, downturn millions of people use it around the world, and know it as a brand (let alone as a coaster manufacturer).

Hopefully bundling the latest Mozilla code will serve to raise people’s expectations of what a browser should be able to do, and then, when they move on from AOL replace their flaky IE with the browser they’re used to. From tiny acorns. :)

Sometimes spellcheckers can spread a lot of light.

Appalled by my recent low level of productivity, I’ve set myself a number of small achievable tasks to perform. The first took a few hours of coding and is presented here:

The UK Blogger Map

What surprised me the most about doing this was that I ended up with so few entries. Admittedly there are two layers of selection - first you must have registered your blog with weblogs.co.uk, and secondly you have to provide the geo data the map needs, but there are ~250 sites listed, only 87 of which have geo data.

Of course I could have used the data from GeoURL, except then I'd have had to do much more processing - first to limit the locations to the UK and secondly to filter out all the crappy deviantART sites. Neither of which I wanted (or could be bothered) to do.

Finally! Google’s true technology exposed! Pigeonrank

By collecting flocks of pigeons in dense clusters, Google is able to process search queries at speeds superior to traditional search engines, which typically rely on birds of prey, brooding hens or slow-moving waterfowl to do their relevance rankings.

Everyone else is talking about it, so why not me?

Jesus set to play the new Doctor Who.

Almost a year old now, but the PyCon Introduction to the Semantic Web and RDF is still very good, and most usefully provides sample Python code.

In the same vein, Building Metadata Applications with RDF very ably demonstrates how to handle RDF documents in Python and is an excellent primer in getting started with Rdflib (both of these articles use RDFlib).

Question answering

This is now my favourite site ever.

Despite earlier opinion, Audioscrobbler wasn’t up as fully as it could have been. But it is now.

I just stumbled across htmldog.com – A Good Practice Guide to XHTML and CSS which covers a lot of ground and breaks up its tutorials into sections (beginner, intermediate and advanced) and has other good articles such as Bad Tags for those which give problems or are best avoided altogether.

Dave Winer wants to merge RSS and Atom

Well, OK, that’s not strictly the gist of his post, but it’s a good attention-grabber isnn’t it? :)

The interesting thing to note is that it went up almost two full days ago and only has 46 comments (that are still there – there may have been a load of posts deleted of course).

Has it just not got the blogging press that this kind of thing would have garnered 18 months ago, or do people just not care any more? There are so many tools and libraries that for producers it’s become almost entirely trivial to provide multiple versions of RSS as well as an Atom feed.

Consumers just don’t care about the format as any aggregator worth its salt will be able to read any kind of feed. Will Atom just forever be the scorned younger brother of web syndication? Will Atom rise up by virtue of its syndication capabilities and overthrow the RSS incumbent?

Hands up if you care any more.

Old-fashioned sources only work the best when you use old-fashioned search queries.

Not true – in her solution the librarian used an encyclopedia. Taking the same approach to the web would have led me to use Wikipedia, which would have given the librarian the answer she was looking for.

And now for something, err, completely normal for bloggers: talking about aggregators.

Because of Gareth I’ve been spending some time looking at different desktop aggregators. I haven’t used one as my main aggregator for about a year now, choosing instead to use the pretty-good JabRSS so it’s been an interesting journey to see what’s been happening in my absence. By all accounts – not much.

This is a great opportunity to give lots of constructive feedback to all the aggregator authors, but I’m too lazy, so here are my five minute reviews:

Jyte

I started on the Weblogs compendium aggregator list and my first download was Jyte. Small, fast, and rubbish; mainly because each time I tried to add a feed, or import an OPML feed, or create a search it crashed. Only beta though, so I got fair warning. Can’t parse Atom. Had to register to be able to use, which was a very simple and quick process.

Tristana

Looks good, quick and easy to use. Failed to import my OPML file and didn’t tell me why. Doesn’t render HTML properly so that my big post about mobile devices is impossible to read. Has a “search web for feeds” function which searches syndic8. Managed to crash it within ten minutes and then couldn’t close it down (and it spawned about five identical processes). Minimises to tray which is nice. Can parse Atom. Website has a nice “share your OPML” section. special mention: Adds a new item to the IE context menu when right-clicking a link “Add content to Tristana” useful for all those orange RSS, RDF and XML icons.

Hot Zoup

Looks good, but doesn’t do OPML import. Supports Atom and recognised that my already added RSS feed was the same thing and so didn’t add it (I would have preferred to be able to add it anyway, but nice to see some smarts going on). The website made it look like I could search my feeds for stuff but it only provides a “search google” function. The HTML pane looks really nice but again didn’t parsemy big post about mobile devices well. Has a nice “flag item” function for marking items to read later. Adding an item is harder than it should be and there’s no drag and drop from browser to aggregator. Have to double click the item title to be able to open it in the embedded browser.

Abilon

Screenshots shows that it doesn’t give me an address bar, so I didn’t download it. I must be able to see my current URL at all times.

Active web reader

Ditto.

ActiveRefresh

Woah. Pay-for software and I don’t see an address bar. Not a chance. Although on the plus side, it apparently supports OPML and OCS import and has a built-in blogging tool. Which is nice.

Beaver

I would have downloaded this if the screenshots alone didn’t make me go blind.

Here ends today’s lesson.

Absolutely amazing, decent CSS support for IE6, managed by the client side! Get sibling and descendant selectors just by adding this script to your pages. Really difficult to state just how much this advances the use of CSS
http://dean.edwards.name/IE7/ (via Eric Meyer)

(I'd get a linklog, but then it'd look like even fewer updates than there are already ;) )

It seems like forever since I posted about CSS and web development in general, but I’m sure it’ll pick up again when I move onto rewriting our corporate site. :)

In the meantime, there are a couple of things I picked out of Dunstan’s blogmarks that I think are worth pointing out:

A really nice technique for highlighting the form element which has current focus by ryan brill which he later revists.

Also, a big list of screencaps of blogs with ace design

Of interest to people who want to check the latest TV guide on their PDAs (i.e. me) are:

• Radio Times on your PDA (via The Daily Chump) which costs money
• TellyGuide which can utilise AvantGo and Plucker, so they get points for good use of existing applications as well as for being free
• Zookware which is also free but Palm OS only (which is fine by my Clié) and has too few screenshots for my liking.

I’ve not actually used any of these and they may all suck, but you don’t want me to spoil the surprise for you do you?

Recently I was whining about Ananova stopping their XMLTV feeds but hooray for open source cos bleb.org provide This week’s UK TV in XMLTV format. woo!

Even 2lmc are spooling Ed talking about Making GNOME hot for content creation so in the name of all that’s fair, I’d like to plug KimDaba (again), which a decent image database manager thing with good categorisation and allows you to specify who each picture depicts and also to annotate certain areas of it, but which could do with being able to generate RDF to describe the pictures (as well as the related FOAF).

And no, I still don't support trackback.

Staying on the politics theme, Ben Hammersely had a great post yesterday called Stop asking questions and vote, dammit. addressing some of the issues about just why the connected generation just aren’t voting any more, summed up quite nicely with his closing line:

Why should I bother to vote, says the connected net-savvy demographic, when it’s plain to see that in every single arena the government is either making shit up or trying to appease Paul Dacre.

DowningStreetSays (via 2lmc) and Public Whip

the latter of which is a very decent roundup of Hansard voting. I used to get the info from the Grauniad’s Aristotle but this is much more interesting. :)

This all cropped up yesterday, but I was too lazy to blog it then. ;)