Michael Nolan

I spent a good proportion of last weekend doing web development - what an exciting life I lead! First up on Friday night I finally managed to get a few moments to knock up a new Advent Calendar. Basically the same symfony-powered site as last year (which is also available for them’s that missed it) but with a new colour scheme and now using the commentable behavior plugin instead of my homebrew effort.

On Saturday I headed up to Waddecar to work on the new West Lancashire Scout County website. Mark Charlton, Pierre and myself did most of the work in a weekend before the summer but life caught up with us and it never got finished. So in 7 hours it was polished off, and that included accidentally running:

$ symfony propel-build-all

Learnt my lesson now

I’m pretty happy with the site for a first release. The top navigation bar is quite neat - a bit of JavaScript created using sfPJSPlugin which can be included on other sites in the West Lancs “network”. Should hopefully build a bit more cohesion between the various sub sites. Options in the back end allow it to be customised with different colour schemes and functionality. Already Network are interested in having some extra buttons to use across Local Scout Network sites.

My already-infrequent blog entries might be a bit less frequent while I’m updating the Advent Calendar but I’m going to try to post occasionally to the Web Services blog.

I’ve already blogged about the event for work but since people are more likely to find my own site I thought I’d mention it here as well.

My presentation “symfony: PHP doesn’t have to be crap… and how we used it to bring Web 2.0 to Edge Hill University” seemed to go okay from my point of view - I’ll leave the glowing reviews to members of the audience!

Met a bunch of cool people and heard about lots of interesting things that they’re doing. Etap Hotel gave me two rooms that hadn’t been cleaned - I must be unlucky as the same thing happened at IWMW.

We’ve just advertised a web applications developer post at Edge Hill University. Alison has given a good insight into what we’re doing, where we’re going and why you might want to work here on the Web Services blog so go check that out.

I mentioned this on the Edge Hill Web Services blog last week but it’s worth mentioning again because the the GoPHP5 website has launched and the symfony project have publicly announced their support. It really is vital that hosts and projects move to PHP5 as earlier versions are frankly dire. If you’re with a crap host, upgrade now!!

It seems that Google/FeedBurner have seen the light and made some of their premium features available for free. In particular MyBrand is now available to everyone meaning I can remove the hack I installed last month.

Via Techcrunch.

Check out the iPhone guided tour.

He’s smug because he’s got an iPhone and you haven’t.

Update: I love how in the stock quotes application Apple are up and Adobe are down

For the last two weeks I’ve been hunting down obscure bugs since upgrading PHP, Apache etc. First there was an issue with Gallery, which I solved with a very hacky fix. But the one that’s been bugging me since was with WordPress - it would display the wrong feed for some pages. Finally after many failed attempts to find anything, I noticed mod_cache had been enabled as part of the upgrade to Apache 2.2. Searching the internet is all about knowing what keywords to find - mod_cache wordpress picked out this from Julien Valroff as the top result. Hopefully that should be the problem solved!

I noticed earlier today that Google Maps seems to have improved its satellite coverage for the UK. I haven’t noticed any announcements about it but you can now see pretty cool images of all sorts of places:

• Edge Hill University
• Great Tower
• Hylands Park

It’s not complete, but it’s a lot better and more recent imagery that before, and it doesn’t have a join running right through my house!

Update: Seems they updated on 2nd June.

The Guardian Technology blog linked to a list on Mashable.com titled “My Soul, and 10 Other Things that Google Owns“. It’s a fairly obvious list of things that Google do to rule your life. Many of them apply to me - I use Google Calendar and Google Reader all the time. Google Maps is pretty good for mapping although I wish they’d open up a UK geocoder in their API. Jack Schofield thinks the most interesting one is “My paycheck”, maybe, but not for me - but please click the adverts and prove me wrong!

For me, the interesting one is the recent announcement that Google is to buy Feedburner.

With its acquisition of Feedburner, Google now controls the leading company for managing RSS feeds. Thus, Google knows everything about my readers – how many of them there are, where they come from, and how they access my content. How might Google use this information? Targeting ads in my feeds based on context or geography sounds like a start, but using cookies the company could also theoretically collect data on my readers and better tailor ads to them throughout Google’s product line.

Feedburner is a really cool service - it allows you to track statistics on your feeds and see how many people are reading your site and in what types of RSS reader. But I’ve always had a major problem with Feedburner - the way they suggest you redirect your feed to their servers. This means that as soon as a user subscribes to your feed you lose control over how they access your site. I’m constantly amazed that people are willing to sign over their feeds to Feedburner just to get some stats or a few bucks from advertising revenue.

I am interested in better stats on my blog though, and as much as I wish there was an open source system that was able to give me good information about who’s reading the site, I can’t find one. So I set out looking for a way to use Feedburner without signing my soul away and with a bit of help from Google knocked up this bit of mod_rewrite magic:

RewriteCond %{HTTP_USER_AGENT} !^FeedBurner.*
RewriteRule ^feed/ http://feeds.feedburner.com/MichaelNolan?format=xml [P,L]


All it does is check to see if it’s the Feedburner feed fetcher requesting my feed, and if it is return the RSS feed, otherwise we proxy the content of the Feedburner feed which adds to our stats. The only downside I’ve noticed so far is that all stats appear to come from the UK, where my server is hosted. I can live with that.

Strangely, this failure could be corrected by Google. If Feedburner was brought into the Google Apps feature-set rather than just Analytics/Adsense then they could allow you to host stats under your own domain, thus protecting the security of your feed.

There’s much talk on the superinterwebs about the future of Facebook. What I hadn’t thought about before was how big they are for photo sharing. Facebook say 60 million photos are added each week which makes them bigger than Yahoo/Flickr and MySpace/Photobucket. My money’s on Google picking them up to get into the social network market properly (Orkut? Saywhat?) which should really give MySpace a much needed kicking.