Andrew’s posterous

Last night a British newspaper (The Guardian) was prevented from reporting upon a question asked in Parliament. Yup, you're not allowed to read about what the question is, or who asked it or to which minister it was addressed. The BBC, Times, Telegraph, Independent won't even report that they're not allowed to report on the story. It's all gone a bit Kafka.

It's a sad indictment of our free press that you need to look to new media (twitter, digg, wikileaks etc) if you want to find out about what your MPs are doing. Ironically, the question you're not allowed to read about is a one implying that there might be certain judges who will rubber-stamp 'gagging orders' for high-profile clients. Here's a link to the questions (61,62,63).

So, if you want to read about dead pop stars, fat/thin models, 'The Best 50 watches', the latest X-factor/strictly news or the '10 worst car parking spaces', then head over to the homepage of your favourite serious broadsheet. 'Old Media' has never been so irrelevant - no wonder it's dying.

Loading mentions Retweet

Easy. Call in the lawyers...

http://ernestmarples.com/blog/2009/10/ernest-marples-postcodes-has-been-threatened-by-the-royal-mail/

Especially sad given that only a couple of days ago data.hmg.gov.uk was launched, with the stated aims of

"This site seeks to give a way into the wealth of government data. As highlighted by the Power of Information Taskforce, this means it needs to be:
easy to find;
easy to licence; and
easy to re-use.
We are drawing on the expertise and wisdom of Sir Tim Berners-Lee and Professor Nigel Shadbolt to publish government data as RDF – enabling data to be linked together. "

[from http://data.hmg.gov.uk/about]

Loading mentions Retweet

So, at Yahoo's OpenHackLondon earlier this year my Purple Pedal Power hack won "Best Flickr Prize". Basically it's a visualisation tool that displays the photos taken by the Purple Pedals bikes over a moving map that follows the routes the bikes take and was written in a couple of hours as an exercise in learning how YQL worked.

Anyhow, two things have given me a bit of impetus to give it some polish and push it live - firstly at last week's AGI Geocommunity the BBC demoed an application with a vaguely similar concept, but coded in Silverlight, and secondly HackDay comes to New York next weekend. Hope you NYC hackers have fun - can't wait to see what comes out of it.

You can visit Purple Pedal Power at http://labs.andrewl.net/pppp - it's still a hack and likely to be buggy, so be nice to it. :)

Loading mentions Retweet

The dust is slowly settling after the AGI Geocommunity event with lots of neo/paleo love going on, and cries of 'why can't we all just get along?'

Nonsense.

I'm of the opinion that although it's true to say we're all 'doing geography' there are some profound differences in the cultures of the two groups and rather than bury them we should be exploring and investigating them. Incidentally, I speak as someone who has got a well thumbed hardcopy of Snyder's 'Map Projections: A Working Manual' and has served my time writing Avenue and MapBasic applications.  Anyhow, back to Geocommunity...   

'Yes, but how do you make money out of this?' was the first question put to Andrew Turner after his keynote 'How Neogeography Killed GIS'? The general view of the paleo community is that the neos are just a bunch of freeloading cheapskates. In one way it was a banal question (the answer of course being 'we sell consultancy and development services - same as everyone else'). But as the conference unfolded it became a touchstone for what I was actually seeing.

The various paleo-presentations I sat in on were given by the public records office, local councils, university and government departments. Mostly they were very interesting. None of them though were 'making money out of it'. Even Arup's presentation on measuring perceptions of space was funded, in part at least, by the National Lottery. The question of profit was never even mentioned.

So who is 'making money out of it'? Yup, it's the neos. The entrepreneurial spirit is alive and well in companies such as like Cloudmate, Itoworld, GeoCommons, Dopplr. These are the people innovating, building the future, 'making money out of it'.

Yes, there is a distinct divide between the paleo- and neo-geographers. It was plainly visible at the AGI Geocommunity conference, but it's not just about t-shirts vs suits, whether you used a notepad or tweeted, or which tools you use. The divide is rooted in those who are sitting in safe jobs and wait for others to deliver the big ideas and those who are pushing boundaries, exploring new ways of using geography, who are innovating for fun and profit.

Loading mentions Retweet

Trying to reset my password at www.bt.com resulted in this email. Doesn't anyone test this stuff?

Loading mentions Retweet

Better late than never, but here's my very brief DrupalCon Paris roundup...

Drupal 7

I'm always highly sceptical when people start rolling out the Gartner Hype Cycle (which, peculiarly, is not cyclical), but nevertheless it was great to hear about the projected timescales for D7 - code freeze now, followed by full release in early 2010. I'm not a #smallcore or #largecore fan, but #appropriatecore. It's great that blog etc are being moved out of core, but more important that rdf and especially cck are heading the other way in to core. If Drupal is to maintain its trend towards become a web application framework, rather than a CMS+, systems like cck and the new theme engine are vital.

Geo

Jer Davis, Allie Micka, Alan Palazzolo and Bec White gave a great overview to a packed room of the current state of the Drupal Geo stack. There's something at every level from geodatabase to visualisation using OpenLayers. In the coming years Drupal is going to become a really sweet platform for developing geo-centered web applications. :)

RDF

Despite all the whizzy geo stuff going on this was the highlight for me. RDFa as an enabler of the semantic web is one of those topics that's been pushed around for a few years now. With RDFa being pushed into Drupal core this means that by default all nodes created using core modules on any Drupal site will emit RDFa tags. All those 'authored by:' tags will contain dc metadata. Suddenly there's going to be whole load more RDFa tags floating around the web, which with Google and Yahoo! using them to enhance their search results, is no bad thing.

As ever it was great to catch up with other Drupallers from the UK and beyond. See you at the next one!

Loading mentions Retweet

Om nom nom

Loading mentions Retweet

overview
significant changes to themeing/rendering subsystem
easier to change content and positioning dynamically
hook_page_alter == single, powerful hook for site builders

$page
$page array = structure, unthemed representation of page content
enables modification of elements using hook_page_alter, then passed to drupal_render
renderable elements prefixed with #
sidebar_first == old left. semantically correct
new addition to devel module that will display page array
ideal for custom site modules, rather than contribs
duplicating a block simple as copying an array element eg $page['myregion']['myblock'] = $page['oldregion']['oldblock'] - sim to eg moving part of a node (eg links) to another region (eg sidebar)
blurs the line between devs and themers. imo increases significace of The Themer.

new cache
any renderable element is cacheable
arbitrary cache keys (eg OG id, role etc)
arbitrary cache expiry time (eg force element cache reload each hour irrespective of anything else)
pre_renderer function name is assigned in #pre_render, called when cached version is not present

renderer
add css js to render elements #attached_js, #attached_css. Do not use drupal_add_js/css - required for good cacheing

theming
jqueryui included
granular themeing. takes a while to appreciate :)
    hide($content['comments']);
    print render($content);
    print "some stuff";
    print render($content['comments']);
hide function - modifies renderable array so they (and presumably children?) not printed by render
render function - renders individual elements (and presumably children?)

tips for developers
menu callbacks should return array, not strings
use hook_page_alter early as possible -> means that altered stuff is available down the line to other modules as early as possible
do not use drupal_render, drupal_add_css, drupal_add_js

Loading mentions Retweet

State Of The Map 2009 was (bike crash aside) a fantastic event. 250 open source mappers with lots of interesting people doing Good Stuff with open data. Many highlights, but here's my top 5:

• Stamen's Open Paper Maps
• Tim and Frankie's historical double act:  Historical Maps and Mapping History
• PlaneMad's maps of Chennai - beating google and being useful.
• Is Volunteered GI sustainable?
• talking geo-philosophy

Thanks too to Frans and Frieda at Logement De Craecke in Veere for arranging ice, assistance and train times after my crash.

See you at SoTM10 :)

Loading mentions Retweet