Archive for December, 2007

Orienteering, the modern way

Friday, December 14th, 2007

The orienteering club I am a member of has made the decision to purchase electronic punching equipment. Since I grew up with the classic punching equipment that has probably not changed in the last 40 years, this is pretty exciting.

I was kind of puzzled originally that an e-punch card could cost  about $35, while being able to store information about you and up to 33 punches. Assuming a generous 100 bytes per punch, that can’t be more than 4k of storage. 1G USB keys can be bought for a bit over $10 (sometimes even less). I guess there is a lot of research and development invested in making the solution resilient to the elements (the control stations are waterproof and smart enough to save their batteries etc), and there is not a high demand in orienteering equipment either, but I still think there’s a pretty high margin in selling the e-punch cards at $30+.

People are already unhappy that the new cards only work in new control stations  and that forces clubs to ditch their older equipment. There may well be valid reasons to force the incompatibility (after all, the newer e-punch cards are more than twice as fast as the old ones, and it may be that the old stations just can’t handle that).

Since I’m not traveling to A-meets, it doesn’t bother me that I will have a new e-punch card. But it does point out one fact: the good old stapler-like puncher lasted for the past half a century and is still a valid option – can you say the same thing about electronic equipment? I am worried we’ll see the same trend as with modern computers: update every 3 years or you won’t find software to run.

Coincidence? I think not!

Friday, December 14th, 2007

The revised version of RFC2440 is RFC4880.

As someone else pointed out, kind of similar to the revised RFC822 being RFC2822.

The wonderful world of PGP

Friday, December 14th, 2007

In case you’re wondering what I’ve been up to lately, here’s a quick overview.

Aside from the weekly Conary releases and  the regular bug fixes, I’ve been busy trying to make Conary no longer depend on gnupg. As dumb (to re-implement gnupg) as this sounds, it gives us several advantages, one of them being the ability to customize the trust model to our liking. The lack of a good way to tie into gnupg (other than invoking gpg) is another good reason.

Look for the new code in Conary 2.0.