Feeder Rant / Woe is Me
January 9, 2006
This would go on Reinvented Blog were it not for the bitching. It’s long and boring, but I don’t care. This is my life and my bitchy blog.
I’m trying to work on Feeder 1.3, and I have been trying to work on it since the middle of October. Considering I wrote the 1.0 version in 6 weeks, this is a long time and now it’s starting to piss me off. I’d planned to release it after Macworld next week, but that isn’t going to happen.
Since mid-October, I’ve probably spent only a few weeks on it in little chunks and estimated it would take around 8 weeks of development. That means I probably only get to spend about 50% of my time actually developing. I think the problem is exaggerated at the moment because of Christmas and New Year’s and before that, Podcast Expo and the whole moving house aftermath.
Oh yeah, and the fact that I did everything in the wrong order.
You see, back in early August, when I came up with almost all of the stuff for this release, it was as part of a cunning plan to preview during someone else’s session at Podcast Expo. I would start working on it in September and have enough ready to show by mid-November. With everything arranged, I booked the tickets and flights to the Expo.
It didn’t work out that way. I had really underestimated the effect iTunes would have on my workload. I released Feeder 1.2 with the iTunes stuff at the end of July and ended up working on support emails all day, every day for almost the whole of August. For a while, Feeder didn’t really have any competitors in this department. By September I managed to get a couple more releases out there to fix some of the problems and add the features causing me to get so many emails, but we moved house and before I knew it the month was over and I still had issues to deal with.
So, by the middle of October I was looking at trying to patch enough together on this 1.3 release to be able to show at Podcast Expo three weeks later. It didn’t have to be perfect, but I did want to show quite a few things, so I put the UI together and developed enough of the functionality to be able to show four or five major new features that looked like they were working even if they weren’t.
Come the Expo and while I’d pulled off everything I wanted to show I really wasn’t happy with it. Along the way I’d found my UI design was lacking and started finding fault with everything and that was even before the guy with whom I was meant to be doing the demo completely blanked me two times in a row, shattering my already edgy self-confidence. It’s a long story, but that turned out the be a total misunderstanding, and while we worked it out I’d pretty much set my mind on not doing the demo by that point. All my panicky rush had been in vain and my trip to California’s only purpose had been to schmooze with podcasters and see my friend Hans in San Francisco for a night (which was excellent, if far too brief).
I get back from the Expo and I’m looking at this version of Feeder 1.3, and I’m really not happy with it, and I’m exhausted and, thanks to my own intervention (via Madge Weinstein via Adam Curry) in uncovering a new version of the iTunes spec a couple of days before I flew out to California, was getting yet more panicky support emails, because people’s feeds were no longer validating, because FeedValidator had been changed to reflect the updates. So, I get that fixed and by the time I work the changes into Feeder 1.3 November was almost over.
By December I’d been saying to myself and friends for weeks that I really needed a week off to sort out all the other things piling up in the meantime. I caught up on the previous three months of bookwork, which I have to keep reasonably current for Inland Revenue reasons, sorted out the tons of real mail that had mounted up, got my hair cut, did stuff around the new house, etc. I still dealt with support emails, but they seemed to be calming down.
And then, just as I was about to restart development I get inundated with more support emails, and by the time they’re despatched it’s the middle of December. I squeeze in another week’s work and give up for the rest of the year.
So here I am. It’s January, I still have probably a few more weeks of work to do on the release. I could drop some stuff but since it is in the UI, I have already done at least half the work and told people it is coming, I might as well finish it.
There are lessons to be learned here. The first, which is totally for me, is to know your limits. I feel like I’ve pulled off some amazing stuff over the last year by the skin of my teeth, but I should have seen the futility of the Expo experience coming by October and just not bothered trying to do some superficial demo version. Trouble was I didn’t want to let anyone down, not least myself.
Also in future I won’t have so much stuff in what is, after all, a point release. I never really wanted to do this, but some of these things have been piling up over the course of the last few months and even back to the start of the year, because the whole iTunes podcasting thing disrupted all my plans. Also, I’m now going with the UI design I came up with in August; I tried to cut this down, but it doesn’t work and I don’t know what I was thinking.
Sometimes being the first to get something out there screws things up. It was obviously amazingly cool to have been the first proper desktop Mac app to support the iTunes stuff and not have any direct competitors for a month or so, but it also meant weeks and weeks of support for problems that were, by and large, absolutely nothing to do with me and everything to do with iTunes. I now feel like I’m months behind, and that I should have been more disciplined and done things the way I know to be right.
This is the problem with being a one-man-band. You can only do one thing at a time, and if you’ve got a long and arduous task on your hards, such as reverse-engineering the innards of QuickTime files, it’s that much harder to be productive when you have to drop what you’re doing all the time to do other things. Plus, when you screw up, you alone suffer the consequences and have to deal with the mess.
The most important thing, I think, is not to release something that isn’t ready. So this month I will be as dedicated as possible to getting it all done, without killing myself in the process. I’ve had enough of that for a little while.