It's T-SQL Tuesday again today! I'm feeling a little distracted because I'm in Poland for SQLDay, suffering from jet lag, and with my mind turning to the presentations I have to give this afternoon and tomorrow. Yesterday I gave a full workshop on SQL Internals and Performance without turning on my computer even once, but I can't let the day go past without writing a post.
The topic for the month is hosted by Matt McGiffen (@mattmcgiffen) and is on puzzles. It takes me back to the second ever T-SQL Tuesday, almost a decade ago in January 2010. Back then I wrote about an interesting puzzle I had come up with to explain how GROUP BY and HAVING work (the answer is here – and as my blog has moved in recent years as has Adam's post, I can't guarantee any of the links there).
This month I was reminded of something from years back (2007!), when Itzik Ben-Gan challenged people to solve a palindrome puzzle – how to find palindromes made up of words in a list. Even that post has moved – originally being at SQLMag, and now being at ITProToday. I took on the challenge, and was one of the solutions listed by Itzik in his follow-up post. I've never tried to create code to solve sudoku puzzles or other things, although when I was at university I did a subject that included using Prolog and needed to solve number-letter substitution puzzles, which I enjoyed in a strangely geeky way.
The reason why I mention the palindrome puzzle is that the approach I took was very similar to the approach that I'd take if I were trying to do it by hand – looking for words that start with the letters that I'm missing from the end. I've learned (and taught many times) over the years that if I can try to persuade the SQL engine to run a query the same way that I would in real life on paper, then when the computer does it, I've got a strong chance of it performing well.
These days the puzzles that I spend time on are the ones that my clients need me to solve, although I'm often tempted to pick up puzzle books and solve some things. We should never fail to find puzzles that exercise our minds – for what else will keep us sharp?
Today I’m sitting in the historic city of Wrocław in Poland. I’m not sure I ever would have thought to visit, except that over a year ago I was asked to one of the special guest speakers at SQL Day – one of the largest SQL Server-centric events in the world.
The city is a charming place, and despite the rain that’s falling today and my jet lag from travelling for over 27 hours, this place has fascinated me, from the architectural mix of plain v elegant and old v new to its friendly people. I knew of Silesia from a board game, but didn’t know of Wrocław, and certainly didn’t know how to say it (which is roughly “Vrotz-waff”). I’ve walked around, enjoyed some coffee in a Czekoladziarnia, looked around the museum in the Old Town Hall, and will enjoy seeing the attendees at the event which is in the Centennial Hall – which I’m told is the second most significant building here (I even saw photos of it in the museum) and is on the UNESCO World Heritage List. The image below is from VisitWroclaw.eu.
Along with the PASS Summit (USA), SQLBits (UK), and the Data Platform Summit (India), I would consider this to complete the other Grand Slam of SQL events around the world. I’ve done precon seminars at the other three, and tomorrow rounds it off. It’s a great honour to be invited to be one of the special guests – when I look at the list of others they’ve had over the years I know I’m in excellent company.
I’m probably the only speaker here who won’t be using a computer. I’ve discovered over the past couple of years that if I can explain concepts without using one, it exercises a different part of the brain and people understand in a deeper way. And so I’m going to spend a day talking about SQL Internals and Performance, using a flip chart, and using audience members to help demonstrate concepts. Hopefully they’ll enjoy it and be able to approach their data in a different way. I have two other presentations to give later in the week which will also be different to the rest – one will hopefully involve audience discussion, and the other is a session that I’ve done as a keynote before, about story-telling and seeing yourself in movies. I hope to be thought-provoking and entertaining. If I can inspire the data community in any way, then I’ll be happy.