Bad design

Early on this year, when I first took delivery of my telescope, and after a couple of "test" sessions out in the garden, I decided that I should keep a log of all of my observing sessions. Within a couple of weeks of starting to keep that log I then decided to place a copy on my website. The two main motivations for doing this were that it would mean that I've got a backup copy should anything happen to any of my logbooks and it also meant that I'd have something that I could, at a push, index and search.

The initial version of the online version was done with a bunch of code written in GNU emacs that generated the template HTML and I then wrote the text into it.

It wasn't very long before I got fed up with that approach and decided that I wanted to try and hold each session's details in a more standardised format and then generate the HTML from that.

I decided on an XML layout of my own design (that was probably the first mistake — possibly a bad case of NIH; I should have known better) and then wrote a set of tools, mostly based around Sablotron, that converted the XML files into HTML for use on my site. This approach also meant that it was pretty easy for me to add different views and listings of types of observing sessions (such as this, this, this, this, this or this).

Up until now it's worked a treat.

Until, that is, I got the Solarscope and until Sunspot 798 turned up and made for a great solar observing target.

I won't bore you with the details but the current log processing and presentation system is based on the idea of me only having one observing session per day. Although it's not happened yet there's a good chance that I'll find myself having two sessions in a day at some point.

It's an annoying problem that came out of a silly assumption on my part and the irony is that it's an issue because I'm in danger of doing too much observing in one day. Who'd have thought that too many chances to observe would ever be a problem?

Oh well, at least it's an exscuse to do some more astronomy-related hacking.

No comments: