Yesterday, an idea crossed my mind. I enjoy using CWEB when writing programs. I also like the documentation produced by Doxygen. Is it possible to write one set of source files and get the benefits of CWEB and a pretty source code narrative while also getting API reference documentation out of Doxygen? A quick search didn't get me anywhere, so I sat down and figured it out.

On the face of it, these two systems seem incompatible with each other. Doxygen won't read the source files used by CWEB. It will read the output of ctangle, but that program strips code comments which Doxygen needs. There is, however, a command available that forces text to be included in the generated source file verbatim. If the source file contains text such as:

@=//! Doxygen can see this@>

the comment will end up in the output from ctangle. Unfortunately, cweave also understands that command and will display the comment with an ugly box drawn around it. This is particularly bad with the multiple line comments common for use with Doxygen. Doxygen markup is usually not interesting to someone reading the output from cweave, so it would be better if this could just go away. If we restrict these comments to the form above, this is easy enough to do with sed:

sed 's/@=\/.*@>//g' input.w > output.w

With this, input.w can be run through ctangle to generate a source file for use with gcc and Doxygen while output.w can be run through cweave.

A barley shortage caused by government promotion of corn based ethanol production is currently causing the price of malt to rise. The price of hops has also exploded after years of oversupply. An email from Weyerbacher Brewing Co. provides the perspective of an industry insider.

I've just written a little blurb about my first few hours with Leopard. Here's the link.

This isn't really news (which is why the story has not been submitted for front page consideration), but I've noticed that Google indexes my posts here nicely so I'm throwing up a link to a new site that's coming soon. Random Field. Don't bother following the link unless you're a bot. There's nothing there yet. If you're a human, sorry for wasting your time.

The Ethiopian Herald (link via allafrica) reports that a new coffee museum will be opened in Jimma with the goal of commemorating the region's history of being where Coffea arabica was first discovered and turning Jimma into a tourist hub by the millenium (is this a mis-reporting? Looking out to 2999-3000AD seems like an excessively long term project).

The museum is expected to cost 2.5 million Ethiopian birr. When I was in Ethiopia early in 2006, the exchange rate was about 8 birr to the U. S. dollar, so this comes to about $312,500, fairly cheap for a museum.

I visited Jimma briefly last year. I had hoped to visit a coffee research station while I was there, but my timing coincided with a holiday celebrating a battle in which the Ethiopians kicked out the Italians during World War 2.

Today Apple released a new security update for Mac OS X. This is an update to the last security update previously discussed on Slashdot. Many users of older Macs with the original AirPort card discovered that the previous version of this update caused wireless networking to fail when waking from sleep, an issue specifically addressed in the new version of this update.

Isn't this the sort of problem that should not happen when one company handles both the hardware and the operating system?

There's been a lot of talk about Pluto lately and how it's no longer a planet. It has been reclassified as something else. I've heard dwarf planet and irregular planet as the new category. So, no, not a planet.

But that's absurd. Now, I'm not going to trot out some argument that I learned that Pluto is a planet and I'm not going to unlearn that. Rather, it seems that Pluto is still classified as a planet.

Huh?

Yes, that's right. Pluto is still classified as a planet. It seems to me that dwarf and irregular are adjectives. In this case they modify the noun planet. I haven't consulted a dictionary, so I may be completely off on this, but it seems that neither of these adjectives mean "not" and under such an interpretation, Pluto would still be a planet. It would be a special sort of planet, but hasn't Pluto always been a little special?

Perhaps this argument is a little technical, so here's a more down to Earth example. Suppose I have a box. The box is red. Now I can call that a red box. Now suppose somebody were to say that my red box weren't really a box at all. It's something completely different. After all, it's a red box.

Absurd, right? And that's exactly what's going on with Pluto.

So, Pluto is still a planet. Nothing to see here. Please move along.

Trusted Computing (TC) cannot be allowed to become a reality unless it can be easily bypassed with no loss of functionality (in other words, TC must fail due to infeasibility or poor design). I say this not because I want a choice in what I do with my computer. I say this not because I believe that I should be able to do anything I want with bits stored on my hard drive and in my RAM. I say this because if TC emerges in the way its backers envision, it will become a powerful and unstopable tool in the terrorist arsenal.

A major selling point for TC is its potential use in Digital Rights Management (DRM). The idea of DRM is that the distributor of digital content such as movies, music, and books, can provide rules for how that information can and (more importantly) cannot be used. It is a technological solution to the social problem of copyright infringement. That DRM can also trample on fair use provisions of copyright and is potentially inflexible to changes in legislation is beside the point. A TC DRM solution gives media distributors the ability to decide which programs are trusted to restrict what a user is able to do with that media.

Furthermore, TC can be used to enforce software that is used on a subscription model. It allows software to determine if its subscription is current and if it is not, the software no longer runs. Again, this can be used to prevent people from infringing on the copyright of software companies.

This is a potentially powerful and enabling tool for companies of all sizes and the potential for new markets it brings can only be imagined at this stage, but it comes at a prohibitively high cost.

A terrorist network would be able to distribute its plans and correspondence much more effectively and with no chance of interception by law enforcement and intelligence agencies. Terrorists would be able to ensure that any material used in planning an attack could only be viewed with a registered copy of Terrorscape. Copies of Terrorscape that might fall into the hands of law enforcement could be remotely disabled within the terrorist network. Worse, this software could, rather than simply fail to work, provide vast quantities of detailed misinformation to strain the ability of nation states to provide security against these false threats, weakening their enemy without need for an attack at all.

Governments, including the government of the United States of America, are actively pursuing this technology because they see the potential benefits without being made aware of serious danger it poses to free and democratic societies. The advantages this technology provides to terrorist and organized criminal organizations cannot be overlooked.

If you value a free and democratic society, I urge you to write your representatives in government and make them aware of the extreme risk in allowing the development of trusted computing. TC must be banned and continued development halted because if TC becomes a reality, the terrorists win.

Well, today I got my first fan who wasn't first my friend (on Slashdot, I have fans elsewhere or at least I'd like to think so. I've gotten fan mail). I suspect this has something to do with my anti-cell phone stance as that's the pager poll seems to be the only recent thing I've posted on in common with lildogie (54998).

It should come as no surprise to anybody who's been here a while that the moderation and metamoderation features of Slashdot can result in some peculiar scoring. Recorded here are some oddities from my own posts:

This comment asserts that if a command line utility is going to try to keep me from shooting myself in the foot, there should be a switch for releasing the safety. Perhaps the moderators think that I would do well by shooting myself in the foot.

Friended someone for thinking about CoreWars at the same time I was. Surprised there weren't more posts suggesting the same thing.

(Update June 10) Did same with someone who quoted the same bit of an article and ended up with the post being under mine (I replied to something else rather than starting a new topic). Finally, a friend with something in the journal.

The FAQ says that Karma is not a score in a video game. Fair enough, but what if you do want to play Slashdot like a video game? Here I propose rules for calculating such a score, a slashdot score.

Okay, so how do you earn points? You earn points by logging in and posting and by having others reply to your posts. The rules should be such as to encourage reasonable Slashdot behavior. Posts moderated to -1 are worth 0 points. Posts with higher scores are 1 point for 0, 2 for 1, 4 for 2, 8 for 3, 16 for 4, and 32 for 5. So, you don't get any points for trying to be incendiary (unless the mods are crazy (and mustn't they be?)) and you get lots of points for those score 5 posts.

Replies act something like a bonus modifier. To calculate this bonus, take the post score as outlined above and multiply that by the number of replies you get. So, if you get 0 points for a -1 post and get any number of replies, the whole thing is still worth 0 points. On the other hand, if you get a +5 post worth 32 points, 1 reply makes the whole thing worth 64 points, 10 replies makes it worth 352 points.

So, to maximize your score, you'd want to get moderated up to +5 and say something that gets lots of replies.

I don't intend to post paying any attention to the above, but I might periodically check in on that. Scores, as with most video games, don't go down. As of today, that score is 52. Not bad for three days. Years of posting as AC does nothing for the score.

On a side note, somebody understood my sig today. I'd befriend that person, but it was an AC, so I couldn't.

(2 December, 2005 update)

478 through Re: V-chips (July 22), might not count things that have gotten bumped into oblivion.

Was thinking moderation should count in the score, but decided against it. No handy record to check against.

A long time ago, I sent messages to a Usenet group called alt.games.final-fantasy. To provide some perspective on the time, DejaNews and other Web based Usenet archives fail to record my first posts. Some of the earliest recorded posts have quotes from me in the sig (specifically, from a thread called 'Final Fantasy and Science' which over the course of a few years of collecting new material from other AGFFers and keeping it edited as a reasonable whole came to be known as 'The Laws of Final Fantasy'. If you haven't encountered this yet, I haven't touched it in years, but various versions can be found on the Web and Google Groups must have the most recent version. Go read it, it's funny.

Toward the end of my stay on AGFF, I remember that someone was writing a bit of fiction there that featured me as one of the characters. For some reason, I decided today that I was curious to know what the fictional I had been up to.

A quick search revealed that I was looking for the chapters of one of Eyeglazer's fics. Seems it hasn't been collected together and put on a Web site yet, but it's recent enough (Y2K) that it's not hard to find.