My article on DB caching and scalability went up on TheServerSide.

Introduction

How many projects have you worked on where the bottleneck in testing was the database, for both performance and scalability? In my experience, getting to the data, creating new data, and updating data takes a fair amount of effort.

If you are developing a small application, then you may not be worrying about scalability as much. However even small applications like to run fast! When you get to larger enterprise applications, then the concern grows significantly. As you add more and more concurrent users your poor database gets more and more bogged down.

This article discusses how caching data in front of the database can give our DB a break, and allow for faster running, and more available applications.

In particular, We will look at:

• What is clustering?
• Using a distributed cache
• Caching strategies
• Read-through / Write-behind caching
• Technologies that integrate nicely into this architecture

Related link: http://www.forbes.com/home/2003/10/14/cz_dl_1014linksys.html?partner=newscom

We can agree, or disagree, on whether the GPL is a good idea, and whether viral software licenses are a good idea, and whether open source software, in general, is a good idea. There are legitimate points of disagreement with respect to all those questions.

I’m always surprised, however, to find people upset at the idea that the Free Software Foundation (FSF) is actually attempting to enforce the GNU Public License (GPL). Of course the FSF is going to try and enforce the GPL– there’d be no point to the license otherwise. And it’s perfectly obvious and clear from the history of the FSF and the GPL that they expect you to live up to it.

Anyone who builds on top of GPL’d code and is surprised when the FSF comes knocking was, like Rick in Casablanca, misinformed.

In the story in question, the Free Software Foundation claims that Broadcom used GPL’d software in its devices, without following the provisions of the license (Broadcom didn’t share changes, and didn’t perform source-code level publication of derivative works). I don’t know if the claim is true or not; it should be easy enough to determine. Forbes doesn’t dispute it though; Forbes seems to grant that Broadcom is in violation of the GPL.

And hence, according to the story,

For several months, officials from the Free Software Foundation have been quietly pushing Cisco and Broadcom for a resolution. According to Free Software Foundation Executive Director Bradley Kuhn, the foundation is demanding that Cisco and Broadcom either a) rip out all the Linux code in the router and use some other operating system, or b) make their code available to the entire world.

In other words (e.g. when you remove biased terms like “rip out”), the FSF is asking that Broadcom either live up to the license or stop using GPL’d software. Which seems perfectly ordinary to me. In almost any other licensing dispute, it seems clear that Forbes would be on the side of the licensers. Indeed, it seems clear that a market economy (and Forbes likes market economies) relies on things like contracts and licenses being enforced (or, at least, being enforceable).

In this case, however, Forbes clearly sympathizes with the companies that are violating the GPL. Attempting to enforce a license is seen as an evil act, and open-source developers are viewed with scorn, as in the following paragraph:

The dispute, which was leaked to an Internet message board, offers a rare peek into the dark side of the free software movement–a view that contrasts with the movement’s usual public image of happy software proles linking arms and singing the “Internationale” while freely sharing the fruits of their code-writing labor.

Leaked? Dark side? Happy proles singing the “Internationale” ? Can you count the ways in which that paragraph is an offense to journalism?

Is Forbes going to go around making excuses for other companies that violate the provisions of their licenses (more rabidly: Has Forbes made a conscious decision to approve of theft)? Why does Forbes refer to fairly routine attempts to enforce software licenses as “the dark side” of the open source movement? Isn’t it more reasonable to refer to the people who violate licenses they’ve agreed to as “the dark side of commercial software”?

This article raises many questions, none of them complimentary to Forbes. Somehow, I expected better of them.

Related link: http://howtoons.net/

(photo from Doc Searl’s web site, as modified by Andrew Watt).

I spent the weekend in Sebastapol, at a retreat organized by O’Reilly and Associates (and named “FOO Camp”; FOO standing for “Friends of O’Reilly”). It was your basic two-day-long, two-hundred-people-big, brainstorming-style event, centered around self-organized hour-long talks (after the introductions on Friday, people who wanted to speak, or facilitate a session, stepped up and scheduled a time. This was a great way to take advantage of the group’s expertise and enthusiasm).

I missed Tim O’Reilly’s breakdown of the book market, but I did participate in the discussion about search (same link, scroll down) and I got to hear Scott McCloud give a talk that was almost as good as the book.

I learned about subversion, I spent some time learning about the truly astonishing growth of wi-fi networks, I heard some cheese jokes I didn’t quite understand (apparently programming in Python leads to watching Monty Python, which leads to the fact that Cheddar is the world’s most popular cheese which leads to a table filled with great cheese at the Friday night reception. I’m not comfortable with the logic, but I can’t argue with the outcome), and ….and I got to meet a lot of great people. I’m not going to list them all, but I will say it was especially nice to finally meet so many of the the O’Reilly staff I’ve corresponded with over the years.

For me, the highlight of the conference was the presentation by the folks from HowToons (check out the references at the bottom of the page; some of those books look interesting). The basic motivation behind HowToons is simple: children today aren’t taught the joys of fiddling with things. Science education has swung too far towards book-learning and observation. The solution? Create a library of simple projects that:

• Appeal to kids.
• Are easy to complete.
• Rely on very easy to find and highly available (to children) materials.
• Illustrate a scientific or engineering principle.
• Engender a “tinker with it” attitude.

In addition, the project descriptions are illustrated cartoons that fit on a single sheet of paper (so they can easily be photocopied and passed around).

If that doesn’t make sense, check out the draft HowToon for the ice-butt skateboard. That’s something I would have loved to build when I was 10.

Note: You can also see some movies of people on real-world ice-butt skateboards at ZeroPrestige. Be warned, however: the movie files are very big.

FOO Camp was a very good way to spend the weekend.

Did you attend? If so, what sticks out in your mind as the highlight of the weekend?