Log Buffer #30: A Carnival of the Vanities for DBAs

Welcome to the thirtieth edition of Log Buffer, the weekly review of database blogs.

We start with an item on O’Reilly Radar by Tim O’Reilly: Data is the Intel Inside. Tim shows some evidence in favour of the the notion that, in the post-Web-2.0 world (“Web 2.0” being a coinage of his own), where data and database-driven applications are as prevalent and popular, and as cool, as rock music once was, SQL is the new HTML. Buzzphrases aside, I think he has a point. A lot of people know a little SQL (enough, say, to either manage or damage their or their group’s blog or wiki), much as they know some HTML, or some P-language programming.

What is it they know? Don Seiler of die Seilerwerks points out a webpage with an informative tabularization of various major SQL dialects. In the O’Reilly scenario, I suppose this would be like the phrasebook for visitors to the land where SQL is common argot. (I wonder what is SQL for, “Which way are the bathrooms,” or “How much does it cost?”)

I think we can agree that the big news this week came from MySQL AB, mostly spinning off of an interview with CEO MÃ¥rten Mickos in Computer Business Review’s Open Source Blog. First came Mickos’s revelation that Oracle has itself been hinting that they will support MySQL. A surprising turn perhaps. Or perhaps not, given their recent Unbreakable Linux move.

DB bloggers covered this news, of course. Oracle + Open Source noticed, and saw it favourably: “…(Moves) like this not only help Oracle, but they can…embolden the technology.”

There were some posts on MySQL’s disruption of the database business. On TheOpenForce.com, Zack Urlocker makes some observation about patterns in disruption in the software industry, and shows that the Oracle/MySQL thing would conform to them. “…Oracle and MySQL are both in the transportation business. But Oracle is a 747 and MySQL is a Toyota. Unless you’re very rich, a 747 is not a great commuter vehicle. But admittedly, I would not want to drive cross-country for a meeting in New York.”

Zack also mentioned MySQL AB’s introduction of MySQL Enterprise Unlimited: “For $40K (the price of a single CPU of Oracle) you can get an enterprise wide use of MySQL Enterprise, with production 24×7 support and unlimited use of the MySQL Network Monitoring & Advisory Service for a year.”

On rand($thoughts), Savio Rodrigues reponded: “It is likely that the $40k offer will cause folks using Oracle, SQL Server or DB2 for ‘Toyota’ projects to think twice.”

On Infoworld’s Opens Sources, Matt Asay responds: “…I suspect people will find ways to use the ‘Toyota’ a lot more often – including for long-range travel…and this will end up hurting Oracle in a big way.”

There was at least one skeptic, Kevin Closson’s Oracle Blog, where Kevin answered with a post about a non-trivial MySQL arithmetic bug: “It appears as though MySQL is exhibiting a nasty arithmetic bug. It seems when they ran the pricing numbers through the database, free or $0.00 dollars became $40,000 dollars.”

From the second part of Computer Business Review’s interview with MÃ¥rten Mickos came the news that MySQL AB are preparing to make an IPO.

On DBMS2, Curt Monash opined, “Frankly, I think they shouldn’t come public until they can prove solid acceptance of Version 5, because Version 4 remains in too many ways an embarrassment.”

rand($thoughts) again: “Unbreakable Support for MySQL is likely the first step towards an acquisition: There’s really little long-term benefit from Oracle providing support to a competitors database.”

Oh… not to be outdone big-news-wise, Oracle has a rumour of their own: the Oracle bid for SAP and Business Objects, courtesy the OracleAppsBlog. “…I don’t see how this would get (past) the monopolies and (mergers) commissions in the various countries affected and I think its more likely that Microsoft will take over SAP…”

There has been plenty of talk about the DST (daylight saving time) thing and Oracle. And here’s a little more from the Dizwell Blog. From one of the commenters: “March 11, 2007 at 2.00am is the absolute deadline, just like year 2000 issue for mainframes back in the days when people didn’t think of the future. Of course, this affects countries like USA, CANADA. Not sure about others…And the clock ticks away, tick, tick, tick…”

Chris Foot’s Oracle10g Blog published a summary of info on the DST change as it may affect Oracle, operating systems, and Java virtual machines.

On Perspectives on Database Management, Craig Mullins has a few notes on how the DST change affects, yes, DB2: “Basically, IBM has posted that DB2 for z/OS is not expected to be impacted by the changes … Of course, that being said, there are follow-up actions recommended in the previously referenced post.” Better read them.

There was a little conference news this week. On The GoundBlog, Duncan Mills has an item about the European Oracle Usergroups User Conference taking place in Amsterdam in early May.

J.Pipes let us know about the nerdliciousness of the 2007 MySQL Conference and Expo, where “nerdliciousness” means that registration is now open, and the event has lots of tasty stuff, which he is previewing in other blog entries. I leave those to your own curiosity.

Sheeri Kritzer, the MySQL She-BA, has posted the seventh episode of her podcast, OurSQL: Whats it Like to be Normal?. It’s about database normalization in general, and the 1st Normal Form (1NF) in particular. She also has a link to a centralized List of MySQL 5.x binaries.

On Informix technology, Fernando Nunes summarizes and gives examples of new features in IDS (Informix) 10.00.*C6.

On the Discombobulator.net blog, Scott Gruber has a piece on the ancient art of interviewing — a response to items of Howard Rogers’s on getting a job as a junior DBA, covered here last week. He writes, “Howard didn’t articulate what he was looking for particularly well. … Secondly perhaps Howard didn’t know quite as much as he thought he did about some of the finer points of the inner workings of Oracle. But most importantly thirdly most people had missed the point of the series.” Scott continues with his thoughts on the less-tangible qualities of a successful DBA job candidate: book-learning and certification may be of limited value, in his opinion.

An interviewer might admire Padgraid O’Sullivan’s broadening perspectives. Here’s an entry of his about feeling hemmed-in as a DBA. He says, “Being a DBA is cool and all but…I defintely miss that extreme sense of achievement you feel when you develop something that has not been around before and someone actually uses it.” Okay, so maybe that’s not the most accurate picture of how developer time is spent, but the item does point out that DBA/developer gulf again.

zillablog’s Robert Treat found himself in a tricky situation involving PostgreSQL and killing invincible processes. His solution involved Solaris’s dbx debugger.

Pythian’s own Julien Lamarche has an item on an Oracle bug/non-bug (feature??), with comments comparing Oracle to a certain fictional pirate.

If it could all possibly go wrong on an Oracle 9-10 update, it probably will. And it did, for OracleDoc of the So What Co-operative. “For those of you who are still in the 9i world and are going to be upgrading here in the future please for God’s sake bookmark this blog entry or the Metalink notes provided. Maybe it will save you, maybe it wont

It can indeed all go wrong, and Arjen Lentz’s Journal has a picture that says so much more about this than words can. It also illustrates the SQL-everywhere notion from the top of this article.

Until next week, cheers!