This blog has moved

Well actually, all this stuff is still here. It's just that I'm going to be posting new things at my new Wordpress powered blog, which is at http://chrs.me.uk ...

My new methodology

I read a fair few blog articles out there on the subject of software development. It isn't often that I read one that makes me mutter to myself "what are you talking about?". This is one of those articles. In it, Glen Alleman states

the net effort to deliver the project is the same no matter what the methodology used.

This is patently not true. But let's suppose it is, just for fun. Given that, let me introduce my own methodology, the aeronautical software development system. This methodology is very much like a traditional software development methodology, but developers are required to spend an hour folding paper planes after each line of code written. These paper planes are to be folded from large A3 sheets of paper and are required to closely model the Boeing 747. Once folded the aircraft must be painted to closely match the livery of a randomly chosen international airline. The developer must then write out a passenger list and for each passenger on this list, must write a short biography of at least 100 words. Then...

Hold on a second, this is just ridiculous isn't it? Apparently not, because remember, whether you are talking to your customer about requirements, typing in code or folding paper planes it doesn't matter. All methodologies are equal.

I can believe it is true that the effort spent doing valuable work is pretty much the same regardless of methodology. So in that sense software development is a zero sum game. The problem is that a lot of effort is spent doing work that isn't valuable. The skill is in reducing this effort as much as possible.

Oxtremists - Oxford Agile Book Group

I've started a book group for anyone in the Oxford area who'd like to meet up and talk about agile software development. There's a blog-based website and a Yahoo group to help spread the word.

The first meeting is at "Far from the Madding Crowd" on 3 May 2006 at 8pm. We'll be discussing "Refactoring" by Martin Fowler. This is a book I've been meaning to re-read for some time, so now I've got a real incentive to do so! Anyone with an interest in Agile is welcome, even if you've never read the book...

Name change

No, I haven't changed my name to a SQL injection attack yet. I'm referring to the blog title. I thought I'd capitalize on my geeky initials.

Incidentally, I've always considered my surname to be of Scandinavian origin. My Dad's family come from Formby just north of Liverpool, which is traditionally considered to be a Viking village. The -by at the end is a bit of clue for starters. My favourite explanation is that "Rimmer" means "Poet" and comes from the same root as "Rhyme". So I was very interested to learn recently that there is a traditional form of Icelandic poetry called "Rimur". Draw your own conclusions...

Now I live at a URL

Yes, my house is at http://107.the.avenue.org.uk...

All I need to do is register road.org.uk, crescent.org.uk, street.org.uk and I can start my own business mapping houses to URLs.

How to break the National Identity Register

I've just been at a public meeting of the Oxford branch of NO2ID, where a couple of MPs spoke about why they are opposed to the legislation. Before the event began a bunch of us in the office were having a database related discussion on the National Identity Register (NIR). We came up with ways to break the Register:

• Change your name to add some funny characters like a single quote or question mark,

or best of all:

• Turn your name into a SQL Injection Attack. I'll let you know when I've changed my name to '; DROP TABLE REGISTER;

Thoughts from JAOO

So, I've finally got round to blogging about my trip to JAOO in Denmark. I've already posted something on my official work blog about the surprising lack of non-Danes at the conference. Looking back on it, I think the most interesting talk was by Jaron Lanier on Phenotropic Computing (see this interview for more). His basic thesis is that protocols are bad, because they don't degrade nicely. As a program gets larger and larger these problems build up until the whole thing falls apart. The solution? Use something like pattern recognition for components to talk to one another. It wasn't something you could take back to the office and use straight away, but it certainly was something to get you thinking.

On a more practical level, the introductory talk by Martin Fowler on Domain Specific Languages was very good. He managed to get a good deal more interesting material into the short, half hour slot he was given than certain other speakers delivered in an hour. The point that I found particularly interesting was that he pointed out that all programming is using domain specific languages (DSLs) to one degree or other. As soon as you start using an API, that is in effect a DSL. The real difference is the programmers attitude rather than anything else.

I think the biggest disappointment was Ivar Jacobson's keynote speech. I didn't really 'get' it and I wasn't the only one. Another of the speakers characterised it as 'basically nonsense'. He seems to be reacting to the fact that people have difficulty using the Unified Process not by thinking 'Hmmm, perhaps this is too complicated', but by trying to create some kind of intelligent agent to guide your development. This sounds like 'Clippy' from Microsoft Word to me - I'd rather pair program with a person than an animated paperclip, thanks!

(By the way, if you're looking for somewhere to eat in Aarhus close to the Scandinavian Center and the Concert Hall (Musikhuset), I'd recommend the Raadhuus Kafeen (ugly website: www.raadhuus-kafeen.dk), right opposite the City Hall. I'm no expert, but this seems to me to be an old fashioned Danish restaurant. It wasn't too busy even on a Friday night, the food was good and the portions were enormous. I just hope they're still in business if I get to go to JAOO 2006...)