Whoah! It’s been a long time since I’ve been around these blog parts. Lets get that French monkey off my back aye.
The interesting thing about being on both sides of the fence, both a developer and a business person (I graduated with a Master of Commerce in Marketing in 2005. *toot* – thats my trumpet done) and shutting off one half of your brain to get things done. First and foremost in this venture, we need to build the product, and but before this we need to be able to build the product.
Growing into Rails and Programming
I have had to learn and extremely large amount about rails, ruby and programming in general. Basically I came into this startup with some extremely rudimentary rails skills (through building a seat-of-your-pants, kick-ass sales tracking system for a small web development firm here in Dunedin, where I was the Sales and Marketing Manager) and some pretty good (I would like to think) CSS and HTML skills. These were gained primarily from building, updating (and unfortunately designing) the Dunedin Drum n Bass website, launched in February 2007 after 6 months in development. Pure, straight HTML, CSS and Javascript, no content management, weekly updates. And still going. Fun.
I had basically learnt rails as a away of enabling me to learn how to do more programming and build what is I believe a very good take on a sales logging and tracking system for small sales teams. It better be a good take; I built it purely out of need. I wasn’t a programmer, however such was the beauty of the company I was working with – I was able to do what ever was required to succeed. Prospector enabled extensive sales-process streamlining, in a very short space of time. Microsoft CRM freakin sucked, and it also sucked our souls. So it died a quiet death on that day of Prospectors launch in September 2007.
However I have not looked at this code in around two months; and although I knew at the time I started building it that the code was ugly and extremely inefficient, and it only got more so through further iterations of the least DRY code ever (or you could say, extremely wet code, haw haw). Now however, I see it as completely grotesque and would certainly bring servers to their knees with extreme database thrashing left right and center.
One mongrel instance was hurting my dual-core workstation pretty bad with a three-person sales team. Bad, bad rails code.
But I couldn’t do a fresh install, start over and build it fresh; that was completely out of the question. Firstly in a SME sized web development firm where team member roles are as loosely defined as they were, your PC stores a huge amount of (seemingly important) data. Secondly, I had no freakin’ idea how I got the application working, it just kind of happened. I knew nothing of gems, servers, databases (especially); nothing. I followed one tutorial religiously for syntax. Rough. So it was a dash of good luck that it was running at all; so I wasn’t touching that. Thirdly: time. No time. Struggling web SME being turned around. Can’t rebuild work machine.
The Next Stage – codenameplannr
Then serious planning discussions started taking place with Jason and Francois right at the end of last year. I started tinkering with rails more then; from reading more rails to coding the first draft of the application with the most butchered AJAX interfaces you ever would see. But you won’t ever see them, rest assured. By March however I had learned a lot and was more than ready to get started.
However the horizon was fixed on 12 – 18 months – ourselves and the entire team we were working with (10 including us) were turning this dog around; it was challenging for every one of us. But fuck it we were up to it and we had already been doing it for nine months. However things did not quite pan out in this way, so we three became full-time codenameplannr on the 23rd of June 2008.
Anyway; I have been heads down in code for eleven weeks now. Pulling crazy hours, not sleeping, all the good stuff. Thinking about very little else, but achieving a helluva lot. The learning about the craft gets much deeper when a feature complete, public application is being developed. The challenge is massive, and I am loving every second of it. I look forward to solving tomorrow’s inevitable problems.
If I can. Which is probably about a 50/50 chance at best. User interface problem, no worries. Anything that makes those numbers mutate into data that is usable beyond a spreadsheet – talk to Jase on that one at this stage. I am getting there; however a wee while to go… yeah.
All this coding has meant however that I have pretty much turned that other half of my brain off.
And Moving to the Other Side
But now as we approach the closing stages of the primary initial development cycle, I will soon be switching my head to the strategies involved in pimping this thing we call PocketSmith. So more things will be popping up on the blog from me in this frame of mind – as you would notice the majority of my posts so far are tending towards my discoveries for tech problems; the preceeding ones are likely to be bit of a different flavour. In parts.
Anyway, through all the learning that I have done, and through knowing all that we have yet to do, I think we three have ended up building something that is pretty awesome and will help a lot of people. I hope that you think so as well when you get to test it out in a few weeks or so. If you haven’t, hit the beta signup!
