Why are Databases so &#%-ing Difficult!?


Comments are closed.

This was a very nice plunge into the depths into the issue of why databasing is difficult and how simple, careful thought and a few steps at the beginning before you begin building your database can save you a lot of problems and heartache.

This talk didn't really have a thesis. If it did, I guess I missed it. Obviously Dave is highly knowledgeable, and he presented plenty of avenues for non-expert SQL coders to deepen their understanding and improve code performance. BUT...the talk didn't really answer the question in the title beyond just brushing over what the synopsis already says. In other words, lots of great information but almost no explanation of why I should care or how it fits into the big picture.

Also, please please please don't write an entire paragraph (or three) and throw it into a slide. You want people to be able to glance at a slide, digest what it's saying, and then think "OK, convince me."

Dave gave a great talk on MySQL specific things, and I definitely picked up a few nuggets that I'll keep in mind as I build queries going forward, but while the problem put forward was "why are databases so $(#*%$%ing difficult", I didn't feel like we got much of an answer to that. Quick touches on some higher level things. Maybe needs to be a bit broken up to fill the 50m time slot with focus on more things (the mysql visual explainer could probably fill a talk on its own.)

Overall, I thought this was an enjoyable talk, with lots of audience participation (if prompted by the promise of receiving a plush MySQL dolphin), and as someone less on the DBA side of things, I learned some new tricks about MySQL that I can look into and apply next time I'm writing queries.

I will agree with others that the question posed by the talk's title never directly really got answered, though I think it's inferred - databases are difficult because of the myriad little pieces of knowledge that are needed to properly construct your tables and write optimal queries. Perhaps a bit more explicit focus on answering that question (and, I agree - lessening the amount of text on individual slides) will help make this a talk a home run.