Daily Abstraction

StreamBase is looking to hire a programming language expert, partly to fill the gap I'll be leaving behind. The opportunity to work on a programming language full-time doesn't come along all that often, so if you're into languages you may want to check it out.

There are several interesting parts to the job:

StreamBase finally decided to release the complete set of product documentation online, for free. So now, if you're interested in seeing what the StreamSQL language looks like, you can find out without having to download and install a hundred megabytes of binaries.

Via reddit, I came across news that Opera supports server-sent-events (aka SSE). The SSE model for streaming events is very similar to the model used by StreamBase, so last night I whipped up a simple little StreamBase to SSE adapter. More...

blogs.streamsql.org is a new multi-author blog about interesting topics in stream processing. It's hosted by StreamBase, and the main authors are probably going to be university professors and/or developers associated with StreamBase. I think it would be great if we could get posts by people from Coral8 and Apama too. So far there are posts by Mitch Cherniack (from Brandeis) and Ugur Cetintemel (from Brown).

I'll probably write a few posts for that site over the next few months. Topics I'm considering writing about:

• The difference between events (i.e. change notification messages) versus behaviors (variables whose value changes continually). Why I argued against the "inserts" stream.
• Syntactic and semantic drawbacks to basing a language on SQL. What StreamSQL might look like if it had combinators, and what the consequences would be for optimization.
• Static schemas versus dynamic schemas (self-describing data).
• Why StreamSQL needs compound data types, even though relational SQL doesn't.

There's an easter egg in StreamBase. It was added way back in 2004, and it's still there as of 3.5, both in the free version and the enterprise version. Here's how to trigger it:

1. Make sure the components view is open, and isn't obscured by any other view.
2. Add a new input stream to the canvas (using a graphical sbapp, not textual ssql)
3. Give the input stream fields with the following names, in exactly this order: all, your, stream, are, belong, to, us. The field types don't matter.
4. Click on the canvas, so that your schema changes take effect.

Now you should see a picture of all the members of the company as of summer 2004. I'm the one squatting in front on the left.

When I worked at Endeca, I was in the "infrastructure" group, which had about eight people. I just saw a press release saying that Adam Ferrari and Daniel Tunkelang, two of the other people from that group, are going to be at SIGIR 2006 in Seattle. Apparently Daniel will be giving a presentation called "Dynamic Category Sets: An Approach for Faceted Search".

The company I work for, StreamBase, has released our product, also called StreamBase, for free download. If you're interested in seeing what I've been working on for the past 2.5 years, now's your chance to find out first hand. You're required to register to get to the download, but don't let that stop you -- you can use a bogus email address since there's no verification.

I'll be writing more about StreamBase over the coming months, especially once we've released the first public implementation of our textual query language, StreamSQL. The streamsql.org site is still pretty empty, but you can see a "preview" of StreamSQL if you download the product.

If you do try it out, please let me know what you think.