It was back in February when I noticed a call from secret london for developers to help build a website and so looking for a small project at the time to sink my teeth into, I thought I'll come down. I head to where they had a small cozy office off oxford circus to meet the small gang of 4. After chatting a bit with a slightly shy girl who I thought was called Georgia (for some bizare reason) go through the website concept. This girl turned out to be Tiffany, the creator of the facebook group! At the time with my tired mind I was unawares of this! (sorry Tiffany if you are reading this)

Anyways…walking through the site idea, I had commented how I could possibly offer a few hours of my time a week over the next few months. Obviously Tim (the main dev guy) had thought about scalability and having a fast backend from the start. It was a shame he had decided to do it in Django and some weird data engine called Solr but I was enthusiastic working with people who knew what were doing. The real kicker or as Greg (who I considered the energetic and charismatic "ideas man") came out with the punch line… they were planning to create the whole thing in a weeks time! I was a bit stunned but said I will try get back to them (but still enthusiastic to try help out). I left thinking decisively "These people were slightly mad.", my favorite type of people

Moving onto the following weekend with some great planning and scheduling done through gdocs, the "guerrilla style" site build was underway starting on the Friday near Farringdon. Unfortunately I could not attend from the start and joined them on the Saturday. My php skills would be slightly rubbish for this project but decided to help out as best can with the javascript side of things.

I met some fantastic peeps on the day, chatting mainly to Ruth and Andy The setup was was very fun with a large area to accommodate the gang of designers, developers, editors and content people with booze and food around. I found it amazing to see so many people spending their weekend pushing forward something they believed in and credit is due to Demetrios for keeping the atmosphere positive, making sure everyone was happy! The content people had a very arduous task of transferring as much of the information from the facebook website to the site database. There was then the 4-5 developers  nd roughly same amount of designers do the front end work.

It was a bit of a shame on my part as I had a couple of bday do's to attend on that particular weekend, I wanted to hang around a bit longer to polish off my section! After spending a few hours here and there, trying to play around with the google maps local search api, I managed to offer a small contribution in the form of a google map showing the secrets for the first version of the website. I finally left late on the sunday after the design team had did the initial load of the site on a beta domain, at that point it truly struck me how an amazing feat they had achieved with the number on the posters indicating how much data they had loaded and the functionality they managed to squeeze in!

The experience was definitely a positive one and it made me realise how doing coding for projects like this is really good fun. I should do more of these! I had a chance to learn some new stuff that I wouldn't otherwise expose myself to and got a tasty curry and beer for my troubles! Would I do this again? Definitely.

Further links

• Secret london blog
  
• The secret london site idea on tech crunch
  
• Secret London guest posting on Tech Crunch - I'm famous! Though my head looks like it's growing out of Tim's

Made it to the PHP UK conference today and this morning got to hear Josh Holmes do the keynote talk about "The lost art of simplicity", of course was much more abstract chat but so many elements ring true such as the instinct of developers to over build for something simple. Then went on to hear Stefan Priebsch chat about Anti-php patterns, a talk on bad habits or mi-interpretations of patterns. Then followed to Rowan Merewood on his chat about "Living with legacy code", a talk about how to deal with legacy code when trying to update/integrate it with new code. I always like Rowans code with his great injection of his personal style of presenting and humour. Loved the comic book style slides

Just had lunch and I really should mingle a bit more…

ps apologies for any typos and mistakes on this post, the usual amount of checking and editing I would normally do!

Check out the blog entry as well! Glad to see the map working, gonna have to put a few more hours in to make it more amazing!

update – Holy crap! My mug is on tech crunch!

Prototyping generates disposable code. Tracer code is lean but complete, and forms part of the skeleton of the final system. Think of prototyping as the reconnaissance and intelligence gathering that takes place before a single tracer bullet is fired.

I realise more from my reflection that my analytical nature is at times stopping me to "just build it" on projects. Seems like a classical case of 'analysis paralysis' at the first step! With a recent project I then noticing myself just trying to put a whole bunch of code together to get past my first hurdle after reading more about the idea of the duck tape programmer. From these sort of related readings I have these guiding principles (until some other crazy idea skips along) :

• Prototyping is to test whether something will actually work if you build it, that you should be prepared to throw away and therefore DONT SPEND TIME MAKING IT PRETTY(forget comments, coding standards and refactoring)
• Tracer code is where you *might* build something that may end up in the first release
• Refactoring is not as important as making something work and breaks sometimes, then something that doesnt work at all
• Tell the user what should work! And what will work at a later …at some point

Maybe this time next year, I'll understand at a better level the approach to development!