Rent To Buy
Posted December 1, 2006on:
Hello. Whee, nearly a month since my last post. I’m good at this bloggin’ thing aint I?
So how did that Rent A Coder stuff go? Well, it’s kind of a mixed bag. I started off with a bunch of small projects that took a day or two each and paid $10–20. That was alright but I was really hoping to land the bigger jobs of the hundreds of dollars variety. I got one for $65 which was porting an open source program to run on Windows. I downloaded the code and gave it quick test compile before making my bid and it seemed like a very simple job ($65 sounded too much, if anything). However, unfortunately I didn’t realise that in this particular program you have to explicitly switch on nearly all the functionality, so when I tested it I was actually only compiling a very small fraction of the complete project… Whoops. So it turned out to be a bigger task than I first thought, and there was a bugger of a bug that the customer was having but I couldn’t reproduce for ages (it only happened on files bigger than a gigabyte, so it took ages to try each experiment to trigger the bug). By the deadline I’d basically finished the port, but the bug was still outstanding. The buyer kindly paid me anyway, though, and I figured out and fixed the bug a few days later. Still, it was an interesting task, and it was quite satisfying to finally get it working. I was hoping that the buyer would let me contribute the code back to the original project, but he didn’t want to. It’s a shame, because everyone wants to have their work widely used, but the buyer was within his rights to refuse since he paid to have the work done.
After that my “coder rating” (like feedback on eBay) was pretty solid so I got two $200 jobs in a row. The first project was to put together the software for a server which is to be run on Amazon’s EC2 server hosting platform. I was really eager to get this project because EC2 is a very intriguing idea to me, and so it was great to get some hands-on experience of it with someone else paying the bills (I clocked up 153 hours of server time doing the project). Unfortunately this project took longer than I expected so it was a huge rush in the end, but it was only a few hours past the deadline (which my Dad tells me is pretty much early by I.T. project standards).
When I was just starting the EC2 project, the guy from the porting project offered me another one. I’d actually seen the task on the Rent A Coder listings and put it down as a “maybe”, so I took another look at it and decided I could do it. The guy has a team of programmers that write software for Windows, and they’ve decided to get into making Linux software. So what they wanted was to have someone prepare installations of lots of different versions of Linux for them, and pre-install a development environment on each so that they could get up and running as quickly as possible. I also had to write a tutorial for the development environment (it turned out about 3000 words) and some quick instructions about configuring networking and changing passwords on each different version of Linux. Now, I like Linux as much as the next guy, but even before I started I had an inkling that installing it 13 separate times would get tiresome. Well, I was right, it was tiresome, and there were loads of teeny little issues that took two hours each to fix. Add to that the fact that my aging PC is really not cut out for this sort of virtualisation work (I certainly could have used a dual core and a couple of gigs of RAM) and you end up with another project that – have you guessed yet – took way longer than I expected.
The initial deadline was on Thursday 23rd (the day before the EC2 project deadline), and on the Wednesday I realised that I wouldn’t possibly be able to finish both in time, so I asked for an an extension until the Monday which the buyer thankfully accepted. I worked my arse off getting the EC2 project done by Friday evening so I was pretty incapable of doing anything for the next day or so after finishing that. To cut a long story short, I was working as hard as I could but Monday came and went, and so did another new deadline on Tuesday. At this point I was feeling pretty shitty having missed three deadlines, and some other stuff came along and plunged me into a bit of a depressed state. Thankfully though, those clouds cleared after about 24 hours. Again, I’ll omit some details, but I finally finished the bloody thing on Thursday (yesterday) evening and I’m now ready to send off the end product (11 DVDs worth of data).
So, to recap, I’ve been working very hard (like 13 or 14 hour days), messed up my sleeping patterns something rotten (for most of this week I was waking up at 1 a.m. and going to bed at four in the afternoon), and all for not a huge amount of money (just about enough for a Wii). I’m clearly rubbish at estimating the time required for jobs, so perhaps this isn’t the best work for me. It has occured to me that if I had done a $20 job every day then I’d have earnt just as much without the burden of being committed to a large piece of work. However, the $20 jobs are difficult to get because there are so many people who can do them, and it can take me the best part of an hour just to make a really good bid for a project. I’d also have to trawl through all the dregs to find the good jobs, and the dregs can really be quite depressing (reminds me of the depressingness of visiting a Jobcentre).
So, pretty dire, then? I’m pleased that I’ll probably be able to afford a Wii (although I may need some Christmassy assistance from Ma & Pa) but I’m not relishing the prospect of diving back in to more freelance work. The EC2 project people have already offered me another job, but I think I’m going to turn it down because a few days ago I had an idea for a business project of my own which I think will be quite fun. I really don’t know whether it will make any money or not, to be honest, but I think it’s worth the £50 or so that it will take to find out. Watch this space to find out what that project is. Or just ask me; that’d work too.