victor in verbose mode

Chrys lent me his Canon 100mm macro lens a few weeks ago, and I haven't really played with it until now. Rebecca and I came back from Obi with a ton of new plants and flowers for our balcony, after we planted them I had a good excuse to use the lens. Wow, what a sharp lens. I'm impressed.

If you like the Beatles, you'll find this video clip amazing. I keep expecting him to drop something, but he keeps rocking out.

Brian Fioca of Jobby recently talked to Owen Bryne, senior software engineer and one of the co-founders of digg.com, about the scalability of PHP on big sites.  It turns out at digg.com, PHP scales really well:

To get a better idea of what was in store for a heavily loaded PHP application, I set up an interview with Owen Byrne, cofounder and Senior Software Engineer at digg.com. From talking with Owen I learned digg.com gets on the order of 200 million page views per month, and they’re able to handle it with only 3 web servers and 8 small database servers (I’ll discuss the reason for so many database servers in the next section). Even better news was that they were able to handle their first year’s worth of growth on a single hosted server like the one I was using. My hardware worries were relieved. The hardware requirements to run high-traffic PHP applications didn’t seem to be more costly than for Java.

Paul Graham writes some good stuff. I've stumbled across some of this articles in the past and, unlike most of these types of essays from other dot-com millionaires, I've always been very impressed.

His latest essay takes on the subject of software patents, a subject that nearly every hacker has an uninformed opinion about. I've always had a very libertarian view about patents in general, and was very glad to read that by Mr. Graham's assessment, I'm at least uniform in my opinions on the validity or worth of patents in our society.
Mr. Graham compares software patents to traditional patents, insightfully and accurately explaining what he thinks are some reasons for the sometimes silly patents issued by the USPTO. He compares the rules to a game of hockey, and lawsuits to the rules of body-checking:

Hockey allows checking. It's part of the game. If your team refuses to do it, you simply lose. So it is in business. Under the present rules, patents are part of the game.

What does that mean in practice? We tell the startups we fund not to worry about infringing patents, because startups rarely get sued for patent infringement. There are only two reasons someone might sue you: for money, or to prevent you from competing with them. Startups are too poor to be worth suing for money. And in practice they don't seem to get sued much by competitors, either. They don't get sued by other startups because (a) patent suits are an expensive distraction, and (b) since the other startups are as young as they are, their patents probably haven't issued yet. [3] Nor do startups, at least in the software business, seem to get sued much by established competitors. Despite all the patents Microsoft holds, I don't know of an instance where they sued a startup for patent infringement. Companies like Microsoft and Oracle don't win by winning lawsuits. That's too uncertain. They win by locking competitors out of their sales channels. If you do manage to threaten them, they're more likely to buy you than sue you.

If you've got an interest in intellectual property law and the business of technology and computing, it's certainly a good read. Like always, you'll learn something from one of Paul Graham's essays.

I've always struggled with the Cisco VPN client for linux. It has seemingly never installed smoothly for me, but now I finally have figured out what's going on. Last week I built a Debian Sarge (stable) box, and installed Cisco's VPN client v4.7. I used a patched installer I had made a long time ago (and forgot why I needed to patch it), but the module compile was failing. After lots of frustration, I reversed the patch to the original sources, and then the installer compiled the kernel module without any problems. I was using the Sarge 686-SMP (officially linux-image-2.6-686-smp) kernel on a dual Xeon SMP 3.0Ghz box, which for Stable means kernel 2.6.8-3-686-smp.

Just now I built a Debian testing machine, this time using kernel 2.6.15-1-686-smp. I rsync'd the Cisco VPN client from the previus box (Sarge 686-SMP), and after running 'make clean', ran the installer. This time I got this:

Arg, not this again, I thought. Then I remembered that I had reversed the patch for the last machine, so I re-applied the patch and then re-ran the installer. This time it built the kernel module without a problem, and then all was gravy. What this means is that Cisco's installer does not compile out of the box for kernel 2.6.15, but it does work for kernel 2.6.8. I'm guessing that somewhere in between those releases, code was changed that breaks the Cisco installer.

So, the point of this post is to help anyone out there who has struggled with the Cisco VPN client. First, make sure you've got your kernel's headers installed - apt-get install linux-headers-`uname -r`. If you're still getting a compiler error, try using this patch (I don't even know where I found it anymore):

Copy and paste that as is into your favorite text editor (or click download on the file at the bottom of this post), and then place it into the Cisco VPN installer root directory, then run:

Then re-run the installer. Good luck, and feel free to leave a comment if it worked for you (or if you have any questions).

UPDATE AUG 30 2006 - I recently installed version 4.8 of the Cisco client, and it appears as though Cisco fixed the problem. It works with new >2.6.15 kernels without requiring this patch.