Recently I’ve been making more modifications to the MySQL source. Part of making modifications is testing them. For years I’ve been curious why there are both tests and mysql-test sub directories in the source tarballs. Arjen Lentz tells me that tests is old and that mysql-test is the current testing framework. This makes sense since the manual testing pages are all about mysql-test.

Are there plans to clean up the remnants of the old testing framework? What else is there in the tarball that’s outdated and needs to be removed?

I worked out these rules from speaking at MySQLcon, and attending OScon this year.

• Speaking at geek conferences is fun.
• Don't put your talk on your slides!
• Don't put everything on the slides, don't put the outline on your slides, don't put a lot of stuff on the slides.
• Put questions on the slides, instead. Each question is the one you answer while that slide is up.
• You can occasionally have an answer, a good quote, or a graphic.
• If someone can't QUICKLY write down what is on the slide, it's too complex.
• Do write speaker notes. Make them as complete as possible. Try to memorize them, but don't just recite them. You can refer to them if you have to during your talk.
• Practice your talk at least once. Check time. You may feel silly talking to yourself in front of the wall, but you will feel dumb instead if you run way …

I just unpacked and started setting up the Kysoh Tux toy that I got at OScon. The lucky folks with a press badge were given a free one. I had to pay for mine.

The good:

Physically, it feels to be well and solidly made. It didn't feel like they were being cheap with the fabrication. It weighs just over half a kilogram.

Physical setup is straightforward. Unstrap the parts from the box. Click in the US power module into the internationalized power cord. Plug in the penguin unit, to charge it up. It can operate both plugged in, and on internal batteries, which it charges itself. Plug the fish unit into your computer with the included USB cord. It uses a mini-usb plug.

The firmware running in the "Penguin" main unit, and the firmware running in the "Fish" USB/wireless transmitter unit, and the interface library and control software on the host are …

If you are running a Linux laptop that uses NetworkManager, you can get a list of every wireless network that NetworkManager knows about by doing

gconftool-2 --all-dirs /system/networking/wireless/networks

If you want to delete one, do

gconftool-2 --recursive-unset /system/networking/wireless/networks/BADNETWORKNAME

I find this useful when I'm in a venue, such as a tech conference, where there is both a free conference wireless network, and a paid wireless network run by the convention center. If you ever accidentally associate with the paid network, then whenever the signal strength of the paid network is higher, NetworkManager will prefer it to the free one. The fix is to delete the paid network data from the gconf database.

Currently there isn’t a good method to control sending individual queries to the slow query log. Typically this isn’t an issue. It becomes an issue when using very short query times and importing sql files. The slow query log doesn’t have a limit on the length of queries it will write to a log. If you’re importing a sql file with inserts that get sent to the log mysql will dump the entire insert query. This can cause the slow query log to grow to gigs in size in a very short time.

I’ve patched mysql 5.0.41 to add a session variable called sql_log_slow (think sql_log_bin) that when set to 0 will prevent queries from that session from being sent to the slow query log. The patch also updates mysqlbinlog to set that variable in it’s output. One issue with this patch is that the executable comment version in mysqldump is set to 5.0.41 but it will error when imported on any 5.0.41 server that doesn’t have this patch. If it’s merged …

A few months ago, I had a meeting with a small local startup. Their tech and their pitch is pretty neat. They had an angel kick them a megabuck of seed to get started. It's the classic geek startup: two main guys, one wearing the CEO/sales/biz/money hat, and the other wearing the CTO/it/tech/arch/geek hat. And they've hired a couple of coders.

But they've also rented some office space: two cubes, a meeting room, a front desk, and a lockable office for the locking file cabinet. Now, it was cheap office space, but still, why?

It's a waste of the angel's money, is increasing their burn. For what it's costing them, they could hire another coder. Dev speed is their current bottleneck, and going from 2 to 3 causes only minimal invocation of Brook's Law, especially if it's early. If they start to grow, they'll have to hire #3 and #4 soon enough anyway, at which point they've outgrown their space, and now also have the cost, …

In a decade, on-demand virtualized utility computing will be an invisible utility, part of the vital infrastructure of the technological economy.

People will mostly have forgotten what an enormous pain in the ass provisioning computation was today. Today, we don't truly feel that pain, because it seems "normal", everyone has to suffer it together.

The situation right now is, if you have a delivery van, you have to make your own gasoline. And you have to hire and pay for your own mechanics. Seems stupid, doesn't it? It's amazing that there are any delivery vans at all …

Think of the internet itself, what it did to telecoms.

Twenty-five years ago, if you wanted a high speed data connection to a computer in San Francisco, it was a pain. You'd have to come up with a pile of money, and wait a couple of months, at best. Hardware would be dedicated and provisioned, and then finally you would have your connection. To …

I’ve talked to several people that have questions about how alter table works under the hood. They want to know how it handles locking tables why they can sometimes use a table during alter table and other times they can’t. Also why it’s so slow

First let’s look at the basic process alter table typically goes through.

1. If a transaction is open on this thread, commit it.
2. Acquire a read lock for the table.
3. Make a temporary table with new structure
4. Copy the old table to the temporary table row by row changing the structure of the rows on the fly.
5. Rename the original table out of the way
6. Rename the temporary table to the original table name.
7. Drop the original table.
8. Release the read lock.

The slowest part of the process is copying rows from the original table to the temporary table. For large tables this can take minutes …

While I'm being insane...

CREATE TABLE housing ...
 ENGINE=FEED
 CONNECTION="http://seattle.craigslist.org/hhh/index.rss";

SELECT title FROM housing WHERE ...;


I think I can get it working with a week or less work.

I thought of doing this at the Yahoo Pipes talk at the Vancouver PHP conference a few months ago, but dimissed it as a joke. I thought of it again at Pasha Sadri of Yahoo's final keynote yesterday. This time, I know that I can do it, but I don't know if I can dismiss it as a joke.

Is this a great idea, or a sick one?

The MySQL Con has done something amazingly intellegent. Taped down to the floors in all the session rooms and meetings rooms, there are power strips. No matter where you sit, there is a powerstrip within a couple of meters. So instead of spending time looking for power outlets and having clumps of people gathered around the few outlets around the walls, you can sit where you want, and take notes and do research while listening and speaking.

This is something that all other tech conferences would do well to emulate.