Mar 5: Hotsos - Day 1.07.0
(I have to avoid clashing blog posting titles with previous years, so introducing a day.year.fraction convention should solve that problem)
This entry is brought to you courtesy of J. Lewis. More on that later.
Here it is so far ...
8:00 - Downstairs for the usual nice breakfast. I say 'usual' but the food at this conference is unusually good in my opinion, and that's coming from someone who doesn't eat any of the damn stuff, but knows a good spread when he sees one.
8:30 - The opening rock-style video showing snippets of previous conferences (I think) and snippets of preparations for this conference (definitely). Then Gary Goodman said a few words before introducing Ken 'Dr. DBA' Jacobs. Ken's presentation was a brief walk through the history of Oracle versions and the different features that were introduced. I made a ton of notes during this presentation, so let me pick a couple of things that stood out.
First he was very specific in acknowledging all of our debt to people like Ted Codd and Jim Gray and hats off to him for that. Oracle may be a powerful acquisition-hungry giant, but the product was originally based on a foundation of the dedication and focus of great men (and I don't use the term lightly). It's a shame my mate Peter Robson wasn't here, because I think he would have appreciated credit given where it's due.
On to some more trivial snippets you may or may not have heard already.
- The founders of Oracle had been working on a project for the CIA called 'Project Oracle'. Nothing to do with relational databases, but they liked the name so asked if they could use it for their fledgling company.
- Oracle's Initial Public Offering, when they became a company listed on the stock markets was 12th March 1986. Microsoft's was 13th March 1986. In March 1986, I'm pretty sure I was writing a version of Konami's Ping Pong for the ZX Spectrum
- Why was there only a version 2 of Oracle? Because those smart people knew that no-one would be comfortable buying a version 1
Oh, one thing that I didn't think rang true in Ken's presentation was his description of ASM as a feature to give you more performance and more reliability. It was one of the few times when I felt like it was an Oracle person presenting, rather than just a database guy, but maybe it's just me.
Overall, though I enjoyed the presentation.
9:45 - On to Cary Millsap next and I was entertained as usual. I particularly enjoyed his opening discussion of sequence diagrams, which I think people struggle with a little and the concept seems easier to digest via a verbal description than in 'Optimising Oracle Performance', at least for me.
There were quite a few points in Cary's presentation where I thought - 'you know, that relates quite directly to something I'm trying to get across in my presentation, too. How can I fit that in?' I haven't solved the problem yet, but I hope to before Wednesday. Oh, that's a bit of a tease, isn't it? He was talking about performance data skew and how easy it is to be misled by initial impressions of aggregated data. Actually, I'm para-phrasing from memory.
11:00 - I was particularly looking forward to Christian Antognini talking about Oracle RAC, Parallel Execution and Data Warehouses because it's a subject he and I have talked about once or twice in the past about other conferences and I wasn't disappointed. Christian didn't seem completely satisfied when I bumped in to him later, but I was very happy with the presentation, I learnt new things, he had played with things I'd been meaning to play with myself and this presentation alone made the conference worth it for me, seriously. At one point during the presentation, Alex G turned and whispered, 'Hey, you guys could have worked on your presentations together!' because it was so close to the subjects I'm interested in. Top marks, Christian!
12:00 - It always happens at every conference I go to, but is probably difficult for people to understand, particularly if you're not presenting. I decided to skip lunch to finish my paper but then also skipped the Stephan Haisley presentation I wanted to attend. It's just difficult to concentrate when you have your own responsibilities at the back of your mind. Sure enough, when I eventually made it back downstairs, Alex confirmed that it was the best presentation he'd seen so far and he was clearly very impressed. Sigh .... What I will say in my defence is that I skipped food too, so I'm hardly slacking!
14:15 - I was very excited about hearing Neil Gunther speak because I've read parts of one of his books after reading Cary Millsap and Jeff Holt singing his praises in 'Optimising Oracle Performance'. I say 'parts of' because I really struggled to get on with the style of the book, despite being able to see that Neil is a man who knows what he's talking about. I have to say that hearing him present was a very good move and I found it much easier to digest because he's an entertaining, funny speaker who, although obviously working within limited time, didn't completely avoid any depth. Another presentation I was happy with and, judging by the volume of laughter coming from the other room, it sounded like Julian Dyke's presentation went well too.
15:30 - Regular readers know you'd have to shoot me to stop me from attending a Jonathan Lewis presentation. Today that extended to me doing some minor furniture re-arrangement so I could sit next to Shrek, right at the front. Then, just before he started, Jonathan shuffled up (well, if a big English gent can be said to 'shuffle') and informed me that, not only is that particular presentation available online in audio (I'm sure he'll blog about it later) but he will be giving it at the Miracle Scotland DBF too. He suggested I could go next door instead, but I also recognised a perfect opportunity to catch up on some other things. (Hey, you're reading it!)
A Hotsos audience obviously enjoys the real live thing best, so I'll leave it there.If I dash, I might make the panel session ...
Nice post Doug, that way I've an idea what's happening... It's a pitty I couldn't join, if I read all that, it's even more painful.
But I can't do everything
Look forward to the next post. Say also hi to Carry and Marco!
Don't get me started on this CIA thing
Did he tell they were into writing drivers for tape streamers before doing databases ?
Why would the CIA need a company who writes tape streamer software ...
... because in 5 seconds the tape will self-destruct.
Good luck, Jim.