2000-03-30

(23:25:30)

I had a long chat with an OpenLaw person about some of the DVD cases.

I had several very long chats with various people about my ex-girlfriend (and why she's my ex-girlfriend instead of my girlfriend).

I looked up the "number of ones in the binary representation of n" sequence on Neil Sloane's Encyclopedia of Integer Sequences, and found that the first journal reference to it was before I was born.

In Pirke Avot, there is a line about the prospect of having someone tell you "I knew that before you were born!" -- and indeed I know a lot of people who knew a lot of things before I was born.

Bill Gosper has a nice line about an even more extreme version:

I've often been traumatized by discovering something and then learning that Euler knew about it in his adolescence.

I also came up with a very simple algorithm to convert from Gray code to normal binary representation of a number.

I was thinking about the number of ones in the binary or Gray code representation of n, and then how to convert between the two representations, and in general how many of the digit d appear in the base b representation of the number n... not much progress on most of these questions, although I have nice recursive rules to solve the digit problems for binary Gray codes and normal counting...

Also lots of discussions today I can't really talk about here.

A couple of days ago, I got a Tektronix X terminal at a computer store in Daly City. I've been trying to get it to work -- I found a monitor it will talk to, and a keyboard and a mouse, and I put it on a LAN and managed to send windows to it. Very nice. Unfortunately, it seems to have a hardware problem, because it keeps rebooting randomly. Also, the flash ROM is missing files...

2000-03-31

(18:18:41)

Last night was a warm night in San Francisco, but I went out too late and didn't manage to get any dinner. I did offer to trade a copy of K&R for a Blue Book, though.

I found my copy of Smullyan's book 5000 B.C., which I had misplaced. Umberto Eco quotes from that in the epigraph to Foucault's Pendulum: "Superstition brings bad luck."

Maybe I should extend a standing offer to send a copy of K&R to any lawyer who wants one. I wonder how much demand there would be.

At work, I'm going to do some sendmail stuff -- I'm starting to forget about sendmail, myself, but customers still use it. I still use it! Maybe if I fell in love with some other MTA, I could add it to my list of "cool stuff on modern Linux systems that will get people with a Unix background excited". (I'll probably talk more about that list in the future, and probably put it on-line.)

Company lunch: Pizza. I passed on that. I still eat pizza sometimes, but I didn't really feel like it today; I am still "slouching toward veganism" and making only very slow progress.

Nick's "Ctrl-V, e" trick didn't work for me, even using the copy of lynx which is installed on his server. I am confused; if it worked for me, I too would be able to write very long diary entries.

People keep sending tech support questions to the CalLUG Board, and I keep getting copied on them. Sometime people send other things -- someone is wondering how he can submit an idea to Linux developers so that nobody will "just copyright it". I imagine he means "patent" instead of "copyright", or else some more elaborate thing like "use it to create a proprietary program". OK, I'm trying to set him straight on the legal issue, but his other practical question is pretty solid. Suppose that you are not a programmer and have no contacts with The Free Software Community(tm). But you have a great idea for a project. What do you do to propose this project such that it is likely that somebody will actually be inclined to implement it? What are the odds that the program idea you thought of will actually turn out to be worth someone else's time to implement? :-)

I have a few people expressing their condolences to me on the end of my relationship with my girlfriend, a few people theorizing about the event, and a few people planning to get together with me sometime to talk about it. I'm spending more time on IRC as a result -- mostly private servers as opposed to OpenProjects, I'm afraid.

I didn't have a productive week at work because I was thinking about her a lot and generally pretty messed up. I thought that coming back from Southern California would let me get caught up on things in San Francisco, but I hadn't anticipated being broken up with (never mind the many strange consequences of this). My room is pretty messy and hard to walk across, and I may have missed some assorted deadlines at work. I'll hope that people are willing to excuse me for this. I probably should simply have stayed home -- I didn't manage to concentrate much.

I don't understand why Yosh and Uzi and other people I know are so comfortable down in Southern California. But Public Opinion here in San Francisco is pretty soundly against Southern California as a concept.

We had a really cool demo of a Linux-related thing I can't talk about yet. It's really nice to get a concrete example that people are not only still doing Cool Stuff with Linux, but even doing Hella Cool Stuff. Sorry for the allusion to secrets; I know it's not very nice.

2000-04-01

(13:45:22)

That's a bad situation. One nice thing about having data in a database is that you can ordinarily export it or import it using any number of formats -- from tabbed text or CSV, which you can process with cut, paste, awk, sed, Perl, or Python, to SQL, which you can use (with a few caveats) on any modern industry standard RDBMS, free, proprietary, or whatever.

If you have stored procedures, well, there's typically no standard and portable way to represent them. As a result, if you want to export data from your database, and an essential part of the data or its structure is contained in a stored procedure, that part will be lost when you try export the data. Oops!

Some people certainly have other philosophical objections to databases. They do not seem to be part of the mathematical formalism that database theorists such as E. F. Codd (of "Codd's Ten Commandments" and such -- does anyone have an actual list, or is this just computing folklore?) have set up as the Platonic ideal of the Relational Database. They may make the database more opaque and harder to maintain. And they're not necessarily consistent with popular paradigms like "three-tier client-server computing". Indeed, "database abstraction" may be as much of a religious principle as is the "end-to-end model" (although for much less important reasons).

In practice, DBAs seem to use stored procedures anyway -- because they're efficient and fast and do wonders for database performance, if you've learned the particular stored procedure language and semantics that your DBMS happens to support.

I looked in C. J. Date's Introduction to Database Systems to see what he had to say about stored procedures (and related schemes for putting code into the server rather than the client). Basically, Date more or less ignores stored procedures for his whole book -- he's a Codd fan and he wants to talk math, or else techniques for implementing Codd's ideas -- but he does mention them in passing. He notes that databases with stored procedures can be more efficient, but complains that there is not yet any standard, portable way to define and represent stored procedures. He says that the lack of such a standard has hindered their use.

Presumably each proprietary RDBMS vendor would like to see a world in which there is a very large crew of professional DBAs who know how to write procedures for that vendor's particular database. This is not a situation that free software enthusiasts should greet with enthusiasm. DBAs are clearly serious, dedicated, competant professionals (although the stereotypes of them are perhaps more on the "suit" side than the "hacker" side, because of the association of large databases with corporate IT). It is troubling to think that, because of their skills, they should become wedded to a particular proprietary platform -- but that's what we see happening, to some degree. People say "I'm an Oracle DBA" (or a Foo DBA, except that most professional DBAs do happen to be Oracle DBAs). Why Oracle? Well, mainly because they've become intimately familiar with the particular details of maintaining efficient databases using Oracle.

This specialization does not need to be as rigorous as it is today. If there were portable standards for more aspects of database design and implementation, it would be easier for DBAs to apply their skills regardless of the particular database platform.

A portable method for representing stored procedures would be a big win in any case, though. It doesn't necessarily have to be thought of in terms of making DBAs' skills portable; it also makes the data itself portable.

There is still the question of whether stored procedures are a good idea in the abstract. Many people would argue that you should just use a standard (SQL) database back-end, and then write, say, a Python program to implement the particular functionality you want. There is a lot to be said for that, really. Not using stored procedures in the first place is indeed an ideally portable technique. :-)

Thanks to various people at Linuxcare for talking about this with me.

I got a great letter from Wolfgang. Well, every letter from Wolfgang is great. I really need to write back to her. She's learned a lot this year. Some of you out there in Advogato land may still be in high school and thinking of going to college. If so, do me a favor, if you would. If you don't know exactly where you are going to college and why, or what it means to you to be going to college, please take a year off and do something completely different first. I say this because I neglected to do it, and ended up dropping out; Wolfgang had the foresight to go and do it, and her letters from all over the Pacific Northwest are full of insight and wisdom and sparkle (not that they ever weren't before).

It's anecdotal evidence, here. If taking a year off terrifies you, then don't do it; but consider that, for a lot of people, it seems to mean the difference between starting college with a halfway decent understanding of yourself, and starting college with no firm sense of who you are. For a lot of people, the former is probably going to turn out to be preferable.

Nobody has any obligation to go to college, either. I have only the highest respect for formal education Done Right (which in that connection is still not a clearly defined term for me), but it's likely much more important to understand and be content with why you're doing what you're doing than to fit someone else's expectations of what you ought to be up to.

I envy some grad students, like daw and Matthew (and I would have put daw's sometime officemate Ian here, if he were only a full time grad student at the moment), who, aside from being rather brilliant, have all sorts of interests and passions which graduate school gives them a series of opportunities to develop seriously. I mean, I don't know how much any particular grad student is actually enjoying grad school, but some of them are at the very least making pretty cool contributions to the intellectual world at large. (Some of the law students I've met through OpenLaw and other ways are also rather impressive activists.)

The vast majority of people in any given educational system don't really seem to know why they are there, or give any indication that, given a serious choice, they'd like to be. This is a serious problem, and I'm no longer so sure where to assign blame for it.

One time I was talking to someone about the proposition that "everyone should go to college" (it being assumed for the moment that everyone can afford to, and that everyone is admitted -- assumptions whose complete falsity are possibly hard to realize when you grow up in a middle-class college town). I said that I thought that "everyone should go to college, but not for the reasons most people do now". Then I elaborated and said that "if they can't find better reasons than their present reasons, they should not go to college". I think I still believe in this sentiment, more or less.

Sometimes I've thought of trying to go to grad school with no undergraduate degree. A few schools will consider this, if you can make them see why they should admit you. It is known that MIT has done this; they pride themselves on having "no admission requirements" and considering every case individually.

Wolfgang is still wandering around and still having all sorts of experiences, about which she writes wonderfully. I really need to write back to her, but I'm so caught up at the moment that it's hard to find the time and peace of mind to do a good job.

I found a bug in lynx when I wrote a script to generate links which perform queries on BookFinder.com. I haven't had the time of inclination to track it down and report it or fix it. I'm sure it would be a nice contribution to make. If somebody reads this and wants to remind me sometime, it would be a nice favor to me (and perhaps to the lynx folks).

Much discussion at great length about (and with) my ex-girlfriend. Maybe I should just try to write a script which will automatically add this paragraph to diary entries every day.

These things are very painful and very complex. They are worth mentioning. They are not topical for free software development, but no less topical than messy rooms, movies, food, Canada, air travel, or rats or ferrets as pets. (Hi, Ryan.)

I am looking for several quotations, but the question occurs to me to wonder who is reading this and whether any of you would actually be familiar with the quotations in question.

I did find out recently two things I was wondering about state bar exams: (1) you typically do have to have a J.D. (2) in order to be admitted to practice law before courts in states like, oh, California, you generally do have to swear a loyalty oath (because lawyers are officers of the court and all). I wonder whether my anarchist law student friends are aware of the latter. I had a little chat recently with a friend about what anarchist law students study -- whether they must not be students of anarchist law.

This is kind of tragic, in that it means that random antiauthoritarian types who are good at taking tests (like most of the people I know from CTY) are probably never going to end up becoming civil liberties lawyers on a whim. Perhaps that's good, because it means that civil liberties lawyers will actually really know what they are doing, but by the same token they are stretched pretty thin, and could probably use some relief. (Oh, it also means that I'll probably never be a lawyer myself; I had been wondering whether I could get there by buying a few thousand volumes worth of law books and reading them, without going to law school and without signing a loyalty oath.)

The question of whether it is good to become a lawyer is one I've addressed many times with many people -- sometimes, after I've recited pi to more places than the person listening happens to know, someone will pipe up "Gee, with a memory like that, you ought to be a lawyer!". (This also happened when I first started reading random snippets of USC and Supreme Court cases -- it seems that most people have never read any of either, at all.) This tends to provoke a long discussion about "why I am not in law school" (aside from the ordinary discussion about "why I am not in college").

I'm expecting to go to Berkeley for lunch today and then perhaps to a party in the evening. I hope that I manage to make it to both.

I think the main reason I haven't gone to parties much is the very frequent presence of (ethyl) alcohol. I like visiting with a small number of friends; when people gather in larger quantities, the odds of their all getting drunk seem to increase dramatically. (This is unless they are Freebornians or the equivalent.)

I think I've stayed less friendly toward drinking than a number of people I knew in junior high or high school. When they went to college, a lot of them started to drink, at least a little (or to use other drugs).

Nick once made a clever joke that I was a Pythagorean. I don't eat animals, I don't drink wine, and I am a number mystic. It works pretty well, doesn't it? The only problem is that I do eat beans; I guess I never took that part to heart. Otherwise, I imagine I would be an excellent Pythagorean.

The other joke: I regularly go to the liquor store to buy alcohol. But I have never been carded. How come? Because the alcohol I'm buying is isopropyl alcohol. I find it instructive that heavily diluted and relatively less poisonous ethanol is much more strictly regulated than the extremely hazardous methanol and isopropanol. I can buy all kinds of deadly poisonous concentrated volatile alcohols over the counter in virtually any quantity... but let me try to buy cooking sherry (not that I know how to cook or have any recipes that use it), and I'd be breaking California law. Indeed, all you have to do to ethyl alcohol to free it from most taxes and age restrictions is denature it -- that is, make it poisonous and foul-smelling. So sometimes it's easier to buy more dangerous things than less dangerous things.

Someone will dispute the idea that food-quality ethanol is less dangerous than properly denatured ethanol. Hmmmm...

Thanks to Nick and the lynx.cfg file, I got the "Ctrl-v, e" thing working, hence the long entry. You need to bind some key to EDITTEXTAREA first (which is not default as lynx.cfg is shipped). This is a very nice feature, except it really promotes being enormously verbose in these diary entries.

2000-04-03

(10:09:53)

I bought a VCR, because I thought I wanted one (I have no television, but a Macintosh that can take NTSC input), though don't ask me at the moment what clever applications I can find for it (aside from the obvious advantage of being able to watch "Magic Tricks, Card Shuffling, and Dynamic Computer Memories", a lecture by S. Brent Morris). I was also thinking about the DMCA and the fact that VCRs which don't implement hardware copy protection for analog video signals will become illegal this month. (Search for "automatic gain control" in 17 USC or in the Digital Millennium Copyright Act. VCRs which were legal for years and years are now becoming illegal.) I don't know whether or not the VCR I purchased implements copy protection systems (and in particular automatic gain control) or not; I didn't check beforehand, and the manual is rather ambiguous on the subject. (The clerk at Target definitely would not have known. I myself don't know the difference between AGC and Macrovision, because I'm not enough of a video geek yet.) In the interest of reducing the impact of the DMCA, I'm tempted to urge everyone to stockpile copy-protection-free VCRs this month. Indeed, I'd be tempted to say that everyone in the United States who might ever want a VCR should buy one this month, and never buy one thereafter. But that's not going to happen, realistically, and so lots of people are going to end up with VCRs that do support AGC copy protection.

I should check on the date on which non-AGC VCRs become illegal, as well as good sources and models of non-AGC VCRs, and try to put out a note letting people know. It's a pity to hear after a law takes effect -- "Oh, yeah, it's been illegal to buy those ever since last year!"

Michelle bought an electric fan. It was cool. It was fancy. I never realized how many puns it is possible to make around fans.

I talked about ancient coins and the Cyber Patrol case with Paul, who is a law student and an ancient numismatist. (This is sort of like the "anarchist law student" question. Perhaps there is something inherently difficult to parse about sentences about law students.) Of course I mean that he is interested in ancient numismatics, not that he's a numismatist who is ancient.

I think that the Perl/Python rivalry is really heating up. I mentioned Python to some people and got somewhat strong reactions -- much as though I'd mentioned vi to them or something.

I didn't realize that the transition to DST in the United States was going to happen Sunday morning. I think this can be taken as a sign that I've been very "out of it" this past week; I almost always know about DST transitions well in advance. This week, I simply had no idea.

The new lynx (current dev version as of Saturday) is very nice and very full of interesting features. I've been using somewhat older versions which shipped with various distributions; lynx isn't a package I usually update very much. I also tried w3m recently (because I wanted to see the famed HTML table support in a text-only browser). I was suitably impressed, but w3m seemed difficult to use; perhaps that's only because I'm so accustomed to lynx. I'll give w3m another try, and keep it handy in case I want to look at things which really rely on tables.

w3m did a very elegant job with my DVD copy protection table.

I don't use Netscape much because I don't use X11 much -- not because Netscape is non-free (which also makes me unhappy). When I'm using X, I tend to use Netscape, too -- I haven't taken the Mozilla plunge just yet. I talked to Paul about just how many web browsers there are -- though the general public may know of two, and many geeks know of five or six, I think there are about a hundred floating around somewhere. (I think I can think of ten off the top of my head.) Of course, web pages are really Best Viewed with telnet to port 80.

As I was on my way back to San Francisco, a guy asked me for a dollar (which I gave him) and then told me a story about why he needed it (which I don't know whether to believe). My previous experience is that the great majority of people who ask for money and have a non-trivial story to tell about why they need it are being less than completely candid. Later on, I remembered that my friend Micah once said "Everybody's hustling something, Seth" -- this after only a year in the Village in New York City. I imagine Micah was in some sense right about that.

I poked around with the problem of converting binary to Gray code and vice versa. (This is a slightly imprecise statement of the problem, because there are many ways of writing things "in binary", and there are many different Gray codes. I'm talking about the "standard" binary numeral system, and the "standard" binary Gray code.) After writing a few equations with exclusive OR, and doing a little induction, I got a pair of very simple formulas.

Let b[i] be the ith digit (the rightmost digit is b[0], the previous digit is b[1]) of the binary representation of some number. Let g[i] be the ith digit (according to the same digit-numbering scheme) of the Gray code representation of the same number. Then, for all i,

g[i]=b[i+1] XOR b[i]

and, in the other direction,

b[i]=g[i] XOR g[i+1] XOR g[i+2] XOR g[i+3] XOR g[i+4] XOR ...

The infinite regress on the right hand side of the second equation should not cause any problems in practice, because all finite numbers start with an infinite number of repetitions of the digit "0", and repeated XORing with 0 doesn't cause any change in a value. Thus, in practice, it's only necessary to continue the XORing in the second equation until the leftmost non-zero digit of the Gray code representation is reached.

So, I actually understand the reasons why these conversions work, but they are difficult to explain here. They can be justified based on a pair of recursive generating rules which yield lists of all binary numbers (or lists of all Gray code numbers) of n digits.

A simpler way of saying this would be that

A particular Gray code bit is equal to the XOR of the corresponding binary bit with the binary bit to the left of that.

A particular binary bit is the XOR of the corresponding Gray code bit with all of the Gray code bits to the left of that.

It was a beautiful day yesterday in San Francisco, and my friend Ben came over, and we spent a long time walking around. We went up to Coit Tower, from which there was a very clear view of the whole Bay Area, and we wandered around the Embarcadero and some other places. It was really very, very nice. Ben told me (most topically for Advogato, of the many things he told me) about getting fan mail for a free software project of his -- a fairly simple script which Ben wrote "to scratch an itch", and which happened to fit the needs of somebody else precisely. Ben's correspondent was so pleased to find this particular script that, he said, he wanted to thank Ben -- and Ben's mother! :-)

South of Market is in transition. It's getting bought up amd retrofitted and filled with offices and apartments. In walking around yesterday, I saw a sign on a fence:

The City Planning Commission will meet in City Hall, Room 400, 1 Dr. Carlton B. Goodlett Place, on ______ beginning at ______ to consider case no. _______, an application for THE DEMOLITION OF A 3-STORY BUILDING AND THE CONSTRUCTION OF A 24-STORY, 314-FOOT TALL BUILDING, WITH APPROXIMATELY 252,000 GROSS SQUARE FEET OF OFFICES AND RETAIL USES AS WELL AS APPROXIMATELY 6,000 SQUARE FEET OF OPEN SPACE, 2 LOADING DOCKS, AND 43 PARKING SPACES.

It's amazing!

OK, the hyperlink was added later. This is San Francisco, but public comment notices posted on the street still don't have hyperlinks to the City of San Francisco web site. Maybe next year.

My own inclination would have been to certify jimd as Master, but I'm not sure that he would have liked that. :-)

I'm noticing some discussion about my comments on law and anarchy by Kelly and dhd; I want to comment in return, but I don't have time yet. If you're interested, take a look back.

I definitely remember talking to Nick about school, and I even remember once knowing where "The Idea of a University" comes from. But I can't remember that at the moment.

I seem to have provoked the College Thread (with my diary entry of several days ago on Wolfgang, and the fact that her decision to put off college for a year was far better than my decision to go immediately). I hope all of you are having a good time -- the reason I didn't post it as an article was that I thought it was not topical for an article, but pretty much anything is topical for a diary entry, right?

slashdot had an article a while back on "Should Geeks Skip College?"; I know from on-line the person interviewed for the piece which provoked that discussion, and so I found it quite interesting, but I also thought that the level of discussion in the slashdot thread was not very impressive. So I would be afraid of writing an article for Advogato which might provoke the same phenomenon. (I don't know whether the certification system is any more effective than slashdot moderation and karma, although I hope so -- I like it here!)

What I was talking about, incidentally, is much more general than whether geeks should skip college. I suppose it's more like "who should be in college or a formal educational environment, why, and when?". (Sorry, Raph, for even posing the question in a diary entry.)

2000-04-03

(12:26:36)

I thought these articles were strange, because they took the position that college was about vocational training or at least about becoming qualified to get some kind of job. They didn't discuss what college was good for, or who should be in college, so much as the far narrower issue of whether some particular people (geeks) could get good jobs without going to college.

The obvious answer is that lots of people can get good jobs without going to college at all, and lots of people do. (This is true regardless of what definition of "good job" you use.) This is something that not everyone realizes, but it's also not particularly interesting, because it ignores the other, potentially much more important reasons why someone might want to go to college.

It's very possible that students are being told that the purpose of their college experience is to allow them to get a good job; in that case, what they are being told is, as a general rule, doubly wrong.

2000-04-04

(00:44:21)

Random notes

It's still hot in San Francisco. I think this must be that new-fangled "summer" thing everybody's been talking about.

Isopropyl alcohol is cheaper in Berkeley. I wonder if you're supposed to pay import duties when you bring it back with you to San Francisco, though. I wonder whether someone will start an isopropyl alcohol smuggling ring, sending dozens of students back and forth across the Bay on BART with their backpacks full of Swan's or Shur Fine or one of the pharmacy chain brands...

I seem to be drinking water again (in addition to Arizona green tea and soda).

I'm very pleased that so many people are finding useful things to do with the BBC. It's really fun to see people's enthusiasm for the thing. Thanks for the fan mail and comments and criticisms, everybody.

Interesting Linux software

Two Kernel Monte was mentioned on linux-kernel (I saw it through Kernel Traffic) -- it's like a Linux version of LOADLIN, so you can boot a new kernel on an already-running system without a complete reboot. I like this a lot, and I think it certainly belongs on the BBC -- not to mention in Debian soon, I hope. (Just in case any Debian developers are listening...)

Nick got me to try a browser called "links" (as an alternative to w3m). It's really impressive, and I guess I'm managing to do pretty well with the (very lynx-like or DOS-like) interface. Wow! So, it has default key bindings which will be intuitive to people who've used DOS applications in the past (including heavy optional use of arrow keys), it supports tables very nicely (including attractive presentation of tables with borders), and it is quite fast. And it has a character-based menu interface, which, per ESR, is one of the "things hackers detest and avoid" (but I'm still impressed by links, for what it does). links includes color support, even parsing HTML hex triplets and mapping them into what seems to be ANSI color. Yikes!

Kangaroo

Sorry to hear about mbp and the kangaroo. I will admit that I had never before met anyone who had hit a kangaroo.

I was talking to Michelle over the weekend about the vastly different problems people in different places and at different times have to contend with. She works with a charitable organization which sends a lot of educational materials around the world, and also makes grants to community organizations in various places. One of the organizations, said Michelle, which had received a grant had recounted one of their biggest problems when they had been getting started: "monkeys were stealing their chickens".

So, here in San Francisco, I don't have to worry too much about monkeys stealing my chickens, nor about hitting a kangaroo while riding around. But most people in the world don't worry about the same things I worry about, either. There really are very amazing discrepancies in what people pay attention to and consider a big deal.

Martin, I'm very sorry about the kangaroo, and I hope you feel all right soon.

Discrete mathematics

I finished up the recursive calculation of the number of 1s in certain binary representations. I am frustrated not to have a closed form for these -- only recursive formulas. (My ex-girlfriend had found a very nice iterative solution using logarithms, which is all the more impressive since she has no training in computer science or discrete math, and she didn't really have the vocabulary. She described it anyway, and, although she probably didn't realize it, it was an iterative procedure relying on logarithms. It was very nice -- but, like my solution, it's still not a closed form expression.)

Neil Sloane's page seems to suggest that no closed form solution is known. I thought I remembered something about using the "choose" notation nCr, because I thought that the number of 1s in the binary representation of n was somehow related to Pascal's Triangle. But I can't find this anywhere, and I can't remember it. In fact, I might be making that up entirely. So the recursive or iterative versions may be the best we'll get.

For anybody who's wondering about the ceil_ld bit in there, it's a trick I worked out for the bonus question on Brian Harvey's Computer Science 61A final exam in 1997. If you have a recursive procedure f(i) which generates a list of length 2^i, and you want the nth item of such a list, you can use f(ceil(ld(n+1))) to get a list which will be long enough to contain it ("at least as long as you need"). I used that on the final exam, except that I wasn't allowed to use a library function to take logarithms, so I had to write another recursive procedure to evaluate ceil(ld(n+1)), which was fun, and indeed very concise in Scheme. (It doesn't require floating point at all -- do you see why?)

At http://www-inst.eecs.berkeley.edu/~cs61a/fa97/final, look for "Infinite Hanoi Stream", second category of correct solutions. My solution was one of the four in that category, although I think I described it in terms of logarithms. (Actually, I think the way I wrote it out was more complicated that the example given on-line, because I think I was fairly explicit about the logarithms and how they were calculated and why they were relevant.) That problem, which had to do with the recursive solution to the Towers of Hanoi problem, actually has a lot in common with these binary representation things -- in particular, having recursive rules that you can apply to lists to get other lists.

ceil_ld is a useful function for other things -- in fact, it also happens to be used in the iterative method mentioned above for solving the "number of ones" problems iteratively. It's just a useful thing to have around. ld is used a lot in information theory, and to a lesser extent in the analysis of data structures and algorithms. All the binary numbers between 0 and n can be written in at most ceil(ld(n)) bits apiece, for instance.

Latin and acronyms

ld is a mostly-obsolete and very useful notation for "logarithm to the base 2". It's short for Latin "logarithmus dualis" ("dual logarithm"), just as the much more familiar ln is short for "logarithmus naturalist" ("natural logarithm"). For computer people, ld is probably more useful than ln. Unfortunately, most people don't care for the notation, and end up writing the much more cumbersome log_2. Hmmmm.

I need to find out about publication of a really fun summary of Professor Robin Hartshorne's research into the history of "QED" and similar Latin acronyms in mathematics. I really enjoyed reading it (and I learned some cool acronyms).

Western Massachusetts

I am amazed to see rwatson talking about Northampton, MA, my home town, not to mention MHC and Bread and Circus -- the Bread and Circus, no doubt, in the Hampshire Mall, or more properly the "Dead Mall" right next to the Hampshire Mall. (Malls in Western Massachusetts are not anything like malls in California. Back in Western Massachusetts, a mall is an unusual thing and a trip to the mall is an event.) In Southern California, and to some extent in any vaguely suburban part of Northern California, there are malls all over the place, anywhere you go, and most of them rather larger than the Hampshire Mall.

Actually, I'm even more amazed to see a mention of Bela (I think only one "l"?) on Advogato. As a longtime vegetarian resident of Northampton, I ate there years ago, when it was called Fez; after it was bought by a new owner, the name changed to Bel{,l}a and the food improved. I ate there in the first week of January with a whole assortment of friends (only one of whom might be reading this), and we were all suitably impressed.

Northampton, not to mention Amherst, is an incredibly fun place to eat, not least if you are vegetarian. (Everyone thinks that Berkeley must be so great for vegetarians -- Northampton is much better.) I can think of about ten or twelve restaurants that I really wish existed in San Francisco or Berkeley -- for the quality of food, the very high proportion of vegetarian stuff, and the very low prices. I'm afraid that, given San Francisco rents, any Northampton restaurants transplanted out this way would probably become pretty expensive.

I'm hoping to be back in Northampton in May for the Smith graduation, and I'm really looking forward to the food! (Oh, yeah, and my family, and my friends who are graduating, and stuff...) :-) :-)

Free software

I wish I had actual free software projects to which I could make a regular contribution of code -- it would make me feel better about Advogato diary entries, for one thing. :-)

I'll get there, though, I think, and I'm not unhappy with the things I'm doing now.

Later today (Tuesday), I'll try to get to my first LUG meeting in a while.

2000-04-04

(14:19:07)

I'm really sad that I'm not out at Computers, Freedom, and Privacy in Toronto right now. The only reason that I'm not there is that I didn't know about it beforehand and therefore couldn't make plans to attend. (CFP started Tuesday.)

I should really be sure to go next year -- and I should be sure to go back to DefCon this year. That's the weekend of July 28. I solemnly promise Brita that I won't write her a letter from there, unless I also finish it while I'm still in Las Vegas. (Last year, I took the letter home with me -- and pretty much everywhere else thereafter -- and it grew, and grew, and grew...)

More on Latin and acronyms

I wrote to Professor Hartshorne asking for permission to publish a copy of his nice summary on-line. I hope he won't mind.

Junger v. Daley, free speech in computer software, etc.

Wendy Seltzer found the opinion in the Junger v. Daley appeal the same morning it came out -- wow, what great fodder for the people up at CFP 2000 to talk about!

The Northern District of Ohio granted summary judgment against Professor Peter D. Junger, when he sued the government, claiming that encryption export control rules were an infringement upon his freedom of speech. The court said that his claim was groundless and that computer software could not be protected as a form of speech.

Professor Junger appealed this ruling, and encryption enthusiasts have been waiting around for a good long time to see what would happen. So this morning brought the great news that the appeals court overruled the district court and found full First Amendment protection for computer source code.

Hey, two judicial circuits down, eleven to go... :-)

For those who don't follow such things, the other so-far successful case is Bernstein v. Department of Justice (9th Circuit, where I live -- in fact, the United States District Court for the Northern District of California is walking distance from my apartment, and from Linuxcare, and more or less from the EFF; if there should happen to be hearings again there, maybe I can attend them). The 6th Circuit ruling this morning is more or less in line with the previous N.D. Cal. and 9th Circuit rulings in Professor Bernstein's case, whereas the original N.D. Ohio ruling in Junger's case had been almost exactly the opposite.

I wish I had some time to write an Advogato article about this issue. We've got an increasingly large number of cases where a computer program becomes the focus of a legal controversy, and many people have been reluctant to bring the first amendment into such controversies. But in some cases -- like crypto export rules -- it seems to be the only option for programmers who'd like to challenge software-related laws or administrative decisions in court.

As Professor Junger is well aware, his own case could have a lot of relevance for the DeCSS cases (and eventually, I'm convinced, for the legality of projects like LiViD, which choose to implement CSS without a license from the DVD CCA).

Let me correct myself. There has been litigation over computer software for years and years. However, in the past, it was almost always companies suing other companies, claiming copyright infringement, trade secret infringement, trade dress infringement, unfair competition, and even (alas!) patent infringement -- all in connection with proprietary commercial software.

But now some very significant software is being written by individuals, or by free software projects, and the legal issues are becoming more evident -- individuals are getting sued, in some instances, for the software they write, because of what it can be used for. (And many people have predicted that it is only a matter of time until there are trade secret or patent lawsuits against free software that starts to compete with major proprietary packages.)

I have a lot more to say about this, but I fear that, if I keep going, I'll start to write that article I didn't have time for...

Anarchy and law

I still haven't had time to comment here. I hope that the discussion will still be of interest to people here by the time I get a chance to have a go at it.

CalLUG

I'm still planning to go to CalLUG this evening and to stay overnight in Berkeley. I wonder whether any actual Berkeley students will come to CalLUG. :-)

2000-04-06

(01:14:55)

Let's suppose that (because it gets mentioned as interesting for embedded systems and so on) I'd like to learn FORTH. What should I do?

I was reminded of this by graydon's diary entry -- but I've had the idea for a while, after hearing that FORTH is excellent for robotics, embedded systems, and firmware.

Certainly Nick is also pretty enthusiastic about learning FORTH. If we ever get the living room cleaned up, we might well buy a few breadboards and a few chips with FORTH implementations on them, and see what we can work out.

Dimwits

I think I have to agree with jrennie's view on "Dimwit" ratings -- unless they're some kind of Zork reference.

Pejorative ratings possibly aren't necessary. If they are necessary, possibly they shouldn't be public. "I don't want to go on record saying that you are a Master" is not quite the same as "I do want to go on record saying that you are a Dimwit"!

Public ratings are useful for some things, but not necessarily for making people feel welcome and appreciated. That's especially true if they're pejorative; I remember some interesting discussions with lilo about the value of courtesy.

Logarithms

I see I made a typo a couple of days ago and wrote "logarithmus naturalist" instead of "logarithmus naturalis". This is because the Markov chains inside my head are so much more accustomed to typing the word "naturalist" than the word "naturalis" that the probability of my actually typing "naturalis" when I want to is pretty low.

What I need to do is get a decent keyboard and then type

Philosophiae Naturalis Principia Mathematica
Philosophiae Naturalis Principia Mathematica
Philosophiae Naturalis Principia Mathematica
Philosophiae Naturalis Principia Mathematica
Philosophiae Naturalis Principia Mathematica

about a hundred times. And then I'll be better at typing the word "naturalis", so that I'll go and say something like "her roommate is a pretty accomplished naturalis" and then resent the fact that I'm so good at typing "naturalis" and no longer very good at typing "naturalist".

Strangely, I don't have a habit of typing "dualist" instead of "dualis", as far as I can tell -- maybe I just write more about naturalists than about dualists?

Northampton cuisine

OK, "Bela" has only one "l" -- I found them at Noho.com. They are on Masonic Street, near the fire station, and near a bookstore... I miss Northampton.

Free speech in software

kelly points out that the 6th Circuit suggests that intermediate scrutiny, not strict scrutiny, should be applied to source code. Hmmmm. I have re-read the decision, and am fairly confused about what the court actually held. I'll have to go and take a look at this case, and the Bernstein case, again.

I'm on a couple of mailing lists with Professor Junger, at the moment, so I might just be really lazy and ask him whether he'd care to give one of those lists his interpretation of the recent ruling. :-)

Latin acronyms

Neat: Professor Hartshorne's summary of Latin acronyms and formulas for concluding proofs (a history of competitors to "QED") is now on-line.

Berkeley and LUGs; friends

I went back to Berkeley Tuesday -- first off for CalLUG. There were fairly few people there, and the discussion was pretty technical. yosh and uzi gave me a ride over, and they demonstrated some wireless Ethernet PCMCIA cards (I guess they were WaveLan cards?), and then Josh got three network interfaces up in his laptop at once: eth0 (wireless Ethernet), eth1 (100baseT Ethernet, connected to my laptop via a crossover cable), and ppp0 (PPP link over a Ricochet modem). With routing enabled, it was pretty neat -- I could ping Manish's laptop, two hops away, and browse web pages on it. (Josh's machine served as a gateway between the wireless LAN and the very small normal Ethernet LAN.) We could log into Josh's machine and get Internet access, too; he didn't try to set up masquerading.

Some people feel that CalLUG is not doing a very good job of serving the general Berkeley student community these days. One problem is that many of the people who have been going to meetings for a long time are very experienced and know each other fairly well; as a result, we get together and start chatting randomly, often about fairly technical things. This might make new users feel somewhat excluded, despite the fact that most of us would be positively overjoyed if some new Linux user at a LUG started to ask questions.

I remember that some former CalLUG members, like Michael Kaminsky (a very active organizer of the group in the past, now at MIT), were very good at keeping discussion flowing, involving everyone, and keeping everyone interested. I worry about the new user experience in small LUGs like CalLUG -- it's not as though people are wandering around trying to find someone to talk to, since everyone is gathered in a circle. I hope that new attendees don't give up, and feel comfortable joining in... otherwise the LUGs in question won't make all so much progress for the community.

Clearly, LUGs are potentially worthwhile in any case, if you have friends who enjoy going to their meetings -- but LUGs can be more than that, including an opportunity for outreach and for meeting new people. That won't work if everybody new gets scared off. I don't think that I myself am very good about being inclusive in technical meetings, unless I make a conscious effort.

When I first went to a LUG meeting (CalLUG, in September 1997), I remember that I was pretty scared, and remained so until I discovered that I had a relatively large amount of practical Linux experience (acquired mostly in almost total isolation from "The Community"), which meant that what I had to say was of interest. If I hadn't already been using Linux for a while, I don't know that my LUG experience would have felt particularly friendly or inviting.

Of course I am very attached to CalLUG, as a former officer and regular attendee -- and I don't mean any criticism unique to CalLUG here. I am just concerned that LUG organizers or regular visitors should keep these things in mind.

After this, I had a good time with several friends, had some very nice conversations, and learned a lot of physics (some of which I was too tired to absorb correctly). Now I see why linear algebra is such a big deal -- the quantum physicists use it left and right (and left again).

I wrote a little commentary about some deficiencies in math education (its frequent failure to provide context for understanding the significance of things), but I'm going to leave that out of this diary for now.

Actually, I have to say that I was very tired during my entire visit to Berkeley, which made it less fun than it might have been. I appreciate the willingness of my friends to put up with me under the circumstances.

I went out to Solano Avenue by bike; it was fun.

Communication

Communication is really, really important. I think I'm starting to get that impression.

I also remember some conversations with lilo on the value of good communication; so we've got communication and courtesy as some nice principles lilo works on. Does he have a third "c" out there somewhere?

A new project for work

I'm going to be going away for a while to work on an interesting project I can't talk about at the moment. I'm looking forward to it (and to the opportunity to say what it is in the future).

I appreciate the confidence that my being sent to work on this indicates. I think it's going to be a very good experience.

The Beautiful Gift of Linux

I just found some copies of "The Beautiful Gift of Linux". I hadn't seen one of those in almost a year -- they were drawn by Wolfgang as a birthday present for me after Don Marti suggested that somebody really ought to do a Linux cartoon tract in the style of the evangelical comic strips people sometimes hand out on street corners. Wolfgang did an unbelievably beautiful job of this, and the result is "The Beautiful Gift of Linux" -- a fold-up cartoon of which I've handed out a few dozen copies. Most of them I've kept for myself: perhaps I should bring them to LUG meetings and see how people react.

Wolfgang wasn't clear on the "free speech versus free beer" point -- but, in all fairness, neither was Richard M. Stallman, once upon a time.

Once upon a time, you could find a scanned version of "The Beautiful Gift of Linux" up on the SuSE N.A. web page somewhere; I don't know whether it's still there. For the full effect, though, you need to find the little booklet in a cafe, or someone needs to hand you one on a street corner, or while preaching on Sproul Plaza.

Advogato diary

I'm having fun writing long diary entries here. While these entries certainly contain points which are relevant to free software and to the community here (LUGs, legal stuff, package status), they also contain lots of non-topical personal stuff. Some people enjoy reading this, and some people don't. Obviously, nobody forces anyone to read a particular person's Advogato diary -- but maybe it's better if Advogato diaries are just about free software?

I'm wondering whether I should start maintaining a diary on the web (as lots of people did before Advogato), without any specific promise that it will have to do with free software on a particular day. I could then link to that from my Advogato personal page and then reserve my comments on Advogato for things that relate directly to free software.

I mean, I enjoy the opportunity to write a diary on the web. And I enjoy having people I know, including many free software people, take a look at what I have to say. But I'm afraid that most of what I'm going to end up saying is not in the spirit of "today I fixed 6 bugs and made a new release of foonly-2.7.18" -- it may be more interesting to some people, and less interesting to some people. So maybe I should say that I'm a free software person, but I don't specifically keep a free software diary, and so my diary shouldn't be on Advogato. Any opinions?

Advogato certifications (again)

I think kelly is right to wish for something in between Apprentice and Journeyer. According to the posted definition of Journeyer, I shouldn't really be a Journeyer -- at least if we talk about things like programming and not other contributions like LUG, list, and outreach work. (The Bay Area is really good for this, as I've often said.) In principle, I could be certified for my Bootable Business Card work or something, but I think that people who certified me mostly certified me because they felt that I was involved with and committed to free software, not because they had read my code somewhere. (trivial-net-setup is not a very pretty bash script, and I would hardly encourage anyone to read it looking to learn bash style...)

And I've certified people on a similar basis -- not following, I'm afraid, the certification categories' definitions very literally. I use Journeyer because I think "this person is in between Apprentice and Master, as I understand them" -- and I do feel that the people i've certified all have a serious and regular commitment, and make serious and regular contributions of some kind.

But if there were something in between, it could be helpful; I'd want to be certified there myself, for the moment. I would also like to hear more clarification about whether we can really count non-code contributions. I absolutely believe, intuitively, that dria and lilo both deserve their Master certifications for their particular contributions -- but I think even I have more code in free software projects than either of those folks. :-)

It's tricky. It's very tricky. But I'm not too worried about it, because this whole business is just supposed to be an experiment, and not any kind of Official Pronouncement that a certain person is a Master of Free Software in a sense that another person is not. And as far as trying to get some basic consensus that someone is involved in or has done good work on behalf of free software, and is maybe worth listening to or working with, I think this system is pretty impressive and has a lot of promise.

The Advogato certification system is one of the most popular topics for diary entries. Maybe people here talk about it much more than is necessary, but it seems very fascinating. Maybe it's all those pretty colors -- I wonder whether some people are going to try to get certified to their favorite colors' levels. :-)

Thanks to the people who are saying interesting things about the certifications -- you're having me a lot of good stuff to think about.

monte

At CalLUG, I talked about some perverse possible uses of the two-kernel monte program -- including booting kernels from very unusual filesystems which are not supported by any standard Linux boot loader. (For example, you could probably boot a kernel from an NTFS, a reiserfs, an iso9660 FS, or over SMB. Yes, I realize that lilo can do some of these things if you set it up right, because it doesn't necessarily care about the filesystem on which the kernel image is stored.)

I think there's a lot of fun stuff possible there; I'm also wondering whether you can use a RAM disk as the source for the contents of a RAM disk -- something like "initrd=/dev/ram0", except from a running system that has a filesystem already mounted on /dev/ram0. This would be a whole lot of fun for experimentation -- you could keep tweaking a RAM disk image in RAM and keep rebooting instantly to see the effect of your tweaks, if by some miracle that trick could work. That way, you could keep casually experimenting with the system boot process until you broke it entirely! :-)

ISC on BIND: not a cracking tool

I told the ISC about how the Daily Cal got the nature of BIND really wrong -- in some reporter's or editor's head, the concept seems to have morphed from "a security hole in the BIND program was used to break into computers" to "the BIND program was used to break into computers" to "BIND is a program which intruders use to break into computers". (There was also a corresponding mistake about buffer overflows: in the article, BIND ended up looking like the tool used to exploit overflows, as opposed to the victim of overflow exploits.)

This bothered me so much -- in part because of the really ridiculous slight against the authors of BIND, who, instead of praise for their contributions to Internet infrastructure, receive criticism for supposedly writing cracking tools -- that I asked the ISC whether they wanted to write to the Daily Cal to correct the misperception.

Impressively, they did -- and David Conrad's letter appeared yesterday to correct the misreporting.

I just thought it was really sad that a really important free software project from Berkeley was being totally misrepresented by the Berkeley student newspaper. Thanks to the ISC for taking the time to write about it.

2000-04-06

(10:23:32)

They say that the difference between automatic gain control copy protection for VHS and Macrovision is that Macrovision includes the AGC stuff and other schemes, like color-stripe CP stuff.

I am not much of a video geek and still less am I a VHS geek. But my reading of the law now is that, at the end of this month, VCRs manufactured or sold in the United States will be expected to implement Macrovision. Only part of Macrovision is required by law if you've been manufacturing a VCR without CP support in the past. If you produce a new model of VCR which has not been manufactured before, it will be expected to implement both AGC and color-stripe CP, which is to say the main parts of Macrovision that people I talked to on mailing lists were familiar with.

I think they just didn't want to say "Macrovision" in the text of the law, because it would look weird to admit that it was illegal to refuse to comply with a particular company's proprietary standard if you want to build a VCR.

AGC is not inherently a kind of copy protection -- once, it was a feature, but some implementations could break copying under some circumstances, so I think it was originally a feature that you were allowed to turn off.

2000-04-07

(12:23:09)

I went down to Menlo Park Thursday for a consulting assignment. I was there for most of the day, and I took CalTrain, since I don't drive. I enjoyed the opportunity to take the train -- it was very relaxing, and I even took a nap briefly on the way down. I may end up using CalTrain a lot more often in the near future; I can certainly think of worse fates. (CalTrain is slow, compared to some other methods of transportation, but it seems clean, quiet, and spacious. And you're even allowed to eat on board!)

When I got back, I tried to clean up my room a bit. I haven't done anything to clean up in a little over a week -- for reasons that will be obvious to some people. I've made a little progress today, and maybe I can keep making progress.

I've got to try to deal with a whole lot of loose ends at work tomorrow (Friday). It should be interesting.

Birthdays

A happy birthday to Raph, with many happy returns and appreciation for Advogato and his other free software work.

That reminds me that some other people I know have birthdays in April. I just wish they would tell me who they are. :-)

Digital roots and divisibility

While on CalTrain, I worked on my old digital roots and divisibility theorem. I've now got really concise proofs of the theorem and its converse -- the end result is something like

The digital root test for divisibility always works if and only if you are testing for divisibility by a divisor of the largest digit in the base in which the numbers are written.

Suppose you have a base b and you want to know if some numbers written in that base are divisible by d.

My theorem says that, if b-1 is divisible by d, the digital root test will always tell you correctly whether or not a particular number is divisible by d. But if b-1 is not divisible by d, there are always some numbers for which the digital root test will give you the wrong answer. (There are also some numbers for which it will give you the right answer, but it won't work "in general".)

I'd still really like to write up a nice explanation of this and a nice proof which will be easy to follow. I first discovered this years ago, but it took me until this past January to get around to proving the converse. But I still haven't written it up, unfortunately. Now I want to learn some more LaTeX, so that I can typeset the equations in the proof. Maybe I'll look for Paul Abrahams's book on TeX.

Gee, I feel like I have a math diary rather than a free software diary. It's a good thing that so many programmers have at least a passing interest in math.

MPAA wants injunctions against linking

Some of you may have heard that the MPAA has asked the New York court in which it's suing Emmanuel Goldstein (now 2600) to forbid him by an injunction from linking to web pages which provide mirrors of DeCSS.

That's really, really, really, really, really, really scary. I read the MPAA's arguments on dvd-discuss (you can find them right away in HTML on Cryptome -- when does John Young sleep?), and wow, do they ever misunderstand the World Wide Web.

I posted some gripes about how links are actually only a particular form of markup (a particular attribute of a particular tag). From the point of view of HTML, this is not substantially different from saying that the marked-up text is in boldface or italics, or is a title. If you think all this is perfectly obvious, go read the MPAA's motion. :-)

There's a lot there.

Ouch

My jaw hurts this morning for no apparent reason.

Traversing an array randomly

I've got an idea for rakholh, which I'll try to write up later.

2000-04-08

(13:45:40)

I was not very productive on Friday, but I corresponded with some people. Some things happened that are stressful. I enjoyed reading mbp's writing; the parts that are not funny are still clear.

The fog has sort of come back in San Francisco after about a week of very clear, hot days. I don't expect it to rain here again any time soon, but we may well alternate between blue skies and fog. It's fog for the moment, and it's pretty.

This evening (Saturday), a friend I haven't seen in a long time will be coming over -- she's going to a conference at UCSF this weekend. This afternoon, I'm going to go in to work on the Bootable Business Card a bit.

As I live right about on the midpoint of a line segment connecting the Sound Factory with the San Francisco Buca di Beppo (hmmm, why don't I just say "halfway in between"?), I really wish I had remembered to try to go to the Mozilla party which stephane mentions. I suspect that it would have been a nice time and that I would have met some interesting people, although I'm not 21 yet, so I don't know that I could have been "legitimately" admitted.

I think that Stephane and Nick are picking up hell of influences from each other, which is kind of neat. They are still both welcome to use my VCR, even if it does implement Macrovision.

(As a non-native Californian, whenever I use "hell of", I'm making a joke.)

Anoles

StevenRainwater saw an anole recently. This reminds me that, when I was younger, I kept two anoles, a male anole named Flash and a female anole named Brasilia (because I'd done a report on Brazil in elementary school). I kept them in a terrarium with a heating rock and some plants and sticks (which I think must have been awfully boring). They were very cute (yeah, I think anoles are cute). Bush beans, wow. Where do you get those, anyway?

That was the only time I ever had a pet, and I haven't felt all so much affinity for lizards since then. I do still think that anoles are cute, though.

The biggest activity for people who have anoles sometimes seems to be explaining the difference between an anole and a chameleon. Oh, and arguing about whether anole rhymes with "hole" (so says my Macmillan Illustrated Almanac) or with Latin "noli" (which is used with an infinitive as one of two ways to form the negative imperative).

Math

I decided to put up a web page with notes and articles on mathematical recreations. This could be a nice counterpoint to the page I've been working on about Martin Gardner. I know that at least one person here knows who Martin Gardner is, so I may well end up asking around on Advogato for advice or comments on my Martin Gardner fan page.

I've now written a Python program that calculates Catalan numbers by counting pairwise parenthesizations. Python is a really cool language -- for recreational mathematics, and in general.

DVD stuff

I sent off a note about the fact that I'm falling behind in the details of the legal argumentation in the DVD cases. I think that my contributions in the future will mostly need to be confined to outreach and education. I'm still hoping to organize a lecture series at Berkeley, which should help.

Wolfgang

I sent off a postcard to Wolfgang, who deserves an awful lot more than that. She's staying at a relative's house under the pseudonym of "Kait"; some people who don't know her too well might suspect that "Wolfgang" is the pseudonym, but I'd dispute that.

Market value of domain names

A friend of mine may receive $100,000 for a domain name. (He's no squatter -- he's really using it for himself, but someone felt it was worth $100,000.) This is a strange world and a strange time.

The fact that domain names are being sold at all makes me fear that there is a problem in the DNS. Some of this has to do with the question of why the general public uses the root servers that were established so long ago -- and how the operators of those root servers view the process of allocating names.

It's very trendy to write a long essay on the problems of procedures for domain-name allocation -- so I might just have to do it.

2000-04-08

(16:26:15)

Still, maybe instead of asking to get rid of the status, I ought to try to live up to it?

I had a good conversation with lilo which was enjoyable and encouraging. We talked about journalism a bit. I think next time a journalist calls me (no, I haven't spoken to a reporter, not a one, since I was interviewed for American Lawyer a month ago about geeks' contributions to OpenLaw), I should maybe say

Are you a good reporter, or an eeeevil reporter?

Hey, raph, how long until we can use the <CITE> tag?

2000-04-09

(12:11:34)

I write long Advogato diary entries. (If you're reading this, it's most likely you've already noticed that.) If you find them interesting, you are welcome and invited to read them. I'm not prepared to answer dria's very worthwhile question about why I or anyway else does that. If you don't care for my diary entries, you can skip them; if you read the "Recent diary entries" page and are annoyed at the length of my entries that are getting in the way of reading other people's shorter entries, we could ask raph for a way to save user preferences on ignoring particular people's diaries, or I could try to implement something.

Thanks.

General

I didn't manage to go to work or to the CABAL meeting on Saturday, because I was busy cleaning so that my friend would have a place to sleep. (My room was really, really messy.)

I had some interesting conversations with her when she came by, and am amazed at the amount of co-incidental parallelism that can happen in people's lives. It's the sort of thing that makes people say that certain things are "in the air" -- or maybe it's the great human pattern-finding ability that makes people see things in clouds. However, since January, my friend's life and mine have been parallel in a fairly astonishing number of ways. (Mutatis mutandis -- she's female.)

San Francisco is really nice. I really like it here.

I had lunch with lilo and his family, which was nice, although too hurried (with the above-mentioned cleaning -- I had to run off in the middle).

The sky is clear again today -- the fog has gone away for the moment.

People in the allegedly lovely cities of Portland and Seattle (I haven't been to either, but really want to visit) are supposedly pretty annoyed that all of these Californians keep moving there and making it crowded. I think people in San Francisco have a sort of conflicted view of people moving here.

1. It's nice that people move here, because it validates our view that San Francisco is really great. See?
2. It's annoying that people move here, because it makes it more crowded and makes the rent go up.

Maybe San Franciscans are just really insecure -- "San Francisco is lovely and interesting. No, really. Just look at all the people who dream of moving here! Honest!"

Cherry blossoms

mjs made it to the Cherry Blossom Festival, and I didn't. Oops. I need to write these things down in advance.

I did get to see the Dragon Parade in San Francisco at the Chinese new year. That was really cool; it's also a source of some of the "parallelism that can happen in people's lives". (My friend who visited yesterday was there, too.)

But I remembered wanting to see the festival, and then not remembering when it was. I guess I should have thought "in the spring" and then made the connection that it's the spring now.

Culture clash

I looked into Wendell Berry's What Are People For? again, and was a little better prepared for what I would find there (which is not necessarily a good thing, but which is a bit less upsetting). Imagine that you grew up with most of your friends for several years being on a BBS, that you started programming when you were six, that the smallest town you ever lived in as a child had 30,000 people or so, that you chose where to go to college in part because of all of the historically important free software that had been written there (and also in part because of the coverage in the New York Times of the community of cypherpunks which hung around there), that you had pity on the anarchocapitalist boyfriend of Ellen Ullman in Close to the Machine because you were an anarchocapitalist programmer too ("Isn't everyone?" asks Ullman, in a tone I think she appropriated from my cousin in New York) and she didn't quite understand him or know what to make of him. Now imagine that you read Wendell Berry.

Some of my teachers in high school were real fans of Wendell Berry. One of them gave me a copy of "Why I Am Not Going to Buy a Computer" (which is in What Are People For?), and I was fairly distressed by it back then. I even tried to write a response, but I didn't do a very good job, because I missed most of Berry's points entirely. I found myself in agreement with his claim that a computer would not be useful to him (and further might cause him to do things which his current standards consider a waste of time), but I thought that Berry was missing the point that we don't use computers because they are useful.

(This is actually an interesting point about the difference between people who chose computing as a profession and people who chose computing as a culture or as a way of life. I was in the latter category for as long as I can remember. This conflict is easy to see -- between "just a job people" and "hobbyists", but also elsewhere. It's important, but it's not the particular culture clash that I want to talk about here.)

I think that people who suggest (as lots and lots of advertising from computer companies now does) that most people in the general non-computer-using public can benefit -- according to their existing values and standards -- from a computer are probably wrong. (There are fascinating exceptions. Cliff Stoll does a rather responsible job of discussing both sides in his technology-critic books, although he's so occupied with trying to refute rampant computer industry hype that he doesn't have much time to address other kinds of claims about the value of computers. So there is a lot that he leaves out or passes over -- including things like the free software community.)

I mean, sometimes the actual real-world usefulness of these things (almost by accident) impresses me. I was almost horrified when I first heard about WebVan -- I thought it must be a joke, a self-parody. What, we can't go to the store down the street any more and buy our own groceries, we have to get them delivered by ordering with a credit card through a dot com?

Then I learned through a mutual friend that a handicapped woman I know in Berkeley, who can't shop for her own groceries, uses WebVan regularly to shop for groceries and have them delivered -- exactly what she wants, conveniently, at a reasonable price. Oh. The WebVan people are doing some good and providing a genuinely beneficial service -- almost, it seems, as if by accident, to people not necessarily in their target market.

But Wendell Berry might say that, in the old days, handicapped people had to rely on their local communities and support networks to help them, and that this dependence was not necessarily bad: that the problem was when those networks were absent, collapsing, or dysfunctional, and not the bare fact that they existed at all. He would probably complain that our particular brand of consumerism rejoices when a community network of friends and family of the disabled can be obsoleted and replaced by some commercial web sites. And he would probably be worrying about whether or not all disabled students have credit cards and credit with which to buy groceries on-line. Maybe he has written about such things recently -- most of the essays in What Are People For? are as old as I am, and I know that Wendell Berry has kept working since then.

To return to my point about culture clash, above: imagine that you have this sort of background (perhaps shared by a number of Advogato readers), and that you then dip into Wendell Berry, who is talking about a world of small farmers respectful of the land not because of environmentalist ideology, but because they actually personally cared about it, and thought it was necessary for their lives... a world in which people cared about the particular places they came from and didn't all want to move to famous cities to do big, exciting things... a world in which people were dependent upon their friends and family and didn't think that was a good thing... a world in which nobody was inclined to think that the idea of technological progress was much more than dangerous hype.

I did also look into Virginia Postrel's The Future and Its Enemies; of course, she's got an index entry for Wendell Berry, as an Enemy of the Future. Well, he is, according to the way virtually everyone who even uses the web regularly probably imagines the future.

Even though I am much more comfortable with Postrel, who actually believes in progress and mistrusts the government and thinks computers are interesting and education is good and there is such a thing as "creating wealth" and we shouldn't be forced by our circumstances and environments to conform to the expectations of tradition if we chose not to -- Berry is a much better and much more interesting writer. Postrel is a decent modern writer who can collect evidence in a clippings file and present the evidence in a reasonable, logical, clear and readable way. Berry is a master essayist who understands and shows he understands how to think.

So here we have the problem (for those who by disposition or training believe in something akin to Postrel's Future) that the Enemy of the Future is at the very least smarter and a better writer than its proponent -- and maybe has a more subtle understanding, too. (I can see a lot of gaps in Berry's knowledge, but I'm sure he could see more in mine.)

Before Silicon Snake Oil, it seems, all the smart people believed in progress -- or, in particular, all the smart people I read believed in progress.

And we doubt not through the ages
One increasing purpose runs,
And the thoughts of men are widened
With the process of the suns.

Now I see other things, like Berry, who says that the loss of small farming communities in favor of high-tech and highly intellectual city life is not progress, but a great loss and a failing, through which we and the world are impoverished (some materially and all culturally and spiritually) and endangered.

I don't know how to deal with this culture clash. It's not as though I want to ignore Berry (or Stoll, or the other perennial NMH faculty favorite, the fascinating Daniel Quinn, some of whose ideas are not necessarily far from Berry's, though they are much more abstract and ambitious). He was serious and respectful enough not to resort to relativism: he thinks that modern urban culture is a real problem, for everyone, even for the people who live in the cities and work with the computers.

And I know some people, like Wolfgang, who are in fact drawn to living in the country and farming and not working with computers (much). I'm not: but I'm amazed that people are now telling me that I'm missing out on something there, and that they're serious about what they're saying.

New project tomorrrow

Tomorrow I'm expecting to go off and start working on some neat stuff. I look forward to it.

I regret both quiet periods and corporate secrecy -- I mean, I would rather be talking about all of these things than writing about how I can't talk about them. When there was only the Linux community and not also the Linux industry, we would go to user groups and talk about everything we knew, and share our knowledge and our speculations.

Now (well, some people have been living with this for longer), when we go to LUGs, we can't talk about everything we know, or everything we are working on. (In fact, that's one reason that there's so little free software stuff in my diary here -- a lot of what I work on with Linux is now confidential, at least temporarily.)

I feel that this aspect of the success of Linux is a very substantial loss: our friends and comrades of the community are now, for some of us, at least theoretically our business competitors.

Maybe this isn't a problem of the Linux industry but of the entire business world -- I never had a full-time professional job before this. Lots of people in the various free software companies never had full-time professional jobs before this. All the computer people I know have done consulting in college, and some professionally to support themselves, but the world of NDAs and strategic alliances and business press releases is really fairly new to many of us. And I think it would be safe to say that, for all those for whom it is new, it produces some discomfort.

It's nice to know that people are still communicating, collaborating, and being courteous. Rumors of the death of the free software community (or of the net) have been greatly exaggerated.

A note on "environmentalist ideology"

(For readers of this who may have been wondering what I meant when I said "environmentalist ideology" above.) Some people are environmentalists and think of themselves as such, and have a set of views related to the idea that the environmental and natural world should be preserved or protected. Wendell Berry has a view which is similar to this but isn't necessarily in the mainstream of American environmentalist thought: so I am just trying to distinguish him from that mainstream somehow. One of the main things I see that distinguishes him from many environmental activists is the value he personally places on farming and on the land -- in part because of what he perceives that they bring to him, his family, and his local communities.

There are many different reasons that people believe in environmentalism: shallow ecology, deep ecology, religious convictions, personal intuitions. It's hard to put Wendell Berry in any particular category here (particularly because he regularly expresses a disapproval of ideologies and abstract categories). I can just say that I've spent a fair amount of time in the company of urban student environmental activists, and most of their motivations and sensibilities seem pretty different from Berry's.

2000-04-09

(12:17:14)

2000-04-10

(09:45:13)

For browsers that don't allow nested BLOCKQUOTE tags, this is going to look weird -- since I quote dhd in a few places using BLOCKQUOTE, and the whole diary entry is itself displayed inside of BLOCKQUOTEs.

General

My friend came back after the conference (it was a two-day thing), and we had lunch, talked some more, and took naps. (I don't want to say "took a nap", in that it wasn't the same nap.)

Now I feel guilty for having corrupted my friend the alternative medicine student with my Evil Hacker Meal and Sleep Schedule. I mean, it was hard for me to get out of my habits of doing everything three to five hours later than normal (especially on weekends), and I fear that I've dragged her along with me this weekend. I hope she can recover her regular routine pretty quickly and without too much trouble.

Certification

dhd took me all the way from Master down to Dimwit in a single step!

No offense taken -- now I'm back at Journeyer level, which I think is more reasonable. I don't have an active campaign to subvert the certification system, even though I think it Has Interesting Problems as much as the next person. At the moment, I am personally content to give people accurate ratings, as I see it (except Dimwit), and hope that the certification system works well. I just don't want to end up in a category that I don't think is justifiable -- I was thinking "What if Ulrich Drepper were here, or the gcc maintainers, or the maintainers of Linux kernel subsystems, or the inventors of programming languages or cryptographic algorithms, or the authors of great classics of computer science theory and practice -- or (l'havdil) Professor Tanenbaum himself? And I'm supposed to be in the top category with these folks?".

I've spoken at conferences, I've served on committees, I've Gotten Paid For This Stuff, I might have a shell script or three in your wallet right now, I've even held protest signs out on picket lines for free software, but I'm not comfortable with any certification system that makes me indistinguishable from the people I'd like to recognize as Masters.

Many thanks to dhd for listening to my lament and even for certifying me as ... something other than Master. :-)

dhd's diary entry

While we're on the subject of dhd:

For that matter, I'm somewhat uncomfortable with hierarchies in general, or at least, ones that are socially constructed (as, no doubt about it, Advogato's is). I do not want to believe that the "in-crowd"/"out-crowd" dynamic that we're starting to see here is some inevitable part of the way a group of people will organize itself socially. In the particular case of "geeks" (my apologies to those who don't fit that description), who, presumably, have spent the better part of their lives in the out-crowd, it's distressing to see them responding by creating their own in-crowd.

This is the only basically troubling implication of the certification concept for me. I appreciated the perception that the technical world was more or less a meritocracy -- while it's really basically a meritocracy, with fewer barriers to entry than other parts of life, still there is a large effect of in-crowd-ness and cliquishness here and there in the community. (I am not at all immune to this.)

This isn't generally used maliciously or with a view to excluding people; I'm not convinced that it's totally avoidable, either. But it should be important that new "unknown" people can get involved and get listened to on the strength of their interest and skills -- and that may be difficult sometimes, here as in real life.

Being inclusive is good -- as everybody who's ever been excluded for trivial reasons can tell you.

I can keep a secret. I've found so far that most of the stuff that is truly secret is either not technical in nature, not free software, not generally crucial to the bettering of free software and society at large, or has strict time limits on its secrecy, so it's not too hard to deal with on a day-to-day basis.

I have a few things that I've been involved with and maybe am about to be involved with that are in the "time limits" category; this category is frustrating, though understandable.

Some people particularly like to

draw a distinction between "capitalism" and "free enterprise".

because of who invented the word "capitalist", and why. Or is this claim an urban legend?

This reminds me of my "what's really wrong with the WTO" concept, which might be a diary entry for another day.

Blah. I would have zero interest in this, except that I find the earnest way in which people have carried on the discussion to be somewhat quaint and amusing.

I'm not going to take you to task for calling this discussion "quaint" the way dria did. Just go look at the "Person Paper", if you haven't seen it. (Yes, I know you used "they" where I would have used "he" -- and you're a linguist and I'm more of a "once and future" linguist, so you get away with it.)

Some of the questions about sexist language or the origins of terms (which haven't been discussed so much on Advogato -- it's been more a matter of meta-discussions of whether or not this is an appropriate topic to worry about) get back to the famous Chomsky-Whorf debate. Does sexist language promote sexism? Can people think, and can they be counted on to think, outside of the apparent limitations or biases of their language?

It seems to me that perhaps Whorf is winning in popular culture, while Chomsky is winning in linguist circles -- but my most recent reading in linguistics is Steven Pinker, and he's got a strong and clear bias.

Oh, and I don't have a girlfriend. Applications with references and code samples are always accepted. Operators are standing by.

I know of someone whose web page says that she's accepting code samples from prospective boyfriends; maybe the two of you could trade some code and see if it's compatible. :-)

(Having done a huge amount of thinking this month about girlfriends, I'd bet that this is a downright awful way to find partners for romantic relationships in real life.)

Domain names

My friend of the much-coveted domain name has now sold it for a good deal of money; I did not see the final negotiated terms, which could be fairly different from the intermediate draft he showed me. In any case, it was a big thing. I have to say again that living here, now, is a weird and fascinating thing.

Berkeley DVD forum

Does anybody in the Bay Area (or who likes to commute to the Bay Area for interesting talks) have a preference for dates and times? I'm hoping that daw's schedule will resolve itself and that he, Professor Samuelson, and Robin Gross (and maybe some other folks -- with or without an affiliation with Berkeley) will be able to give some lectures on their perspectives on the present DVD mess.

I'd still appreciate nominations for speakers and suggestions for times. This event has already been delayed more than a month because all of the speakers we wanted were interested but too busy.

The Idea of a University

Sumana got Nick to reveal the answer: it's a book by John Henry Newman. (Hi, Sumana.)

Newsbet

Skud mentions Newsbet, which reminds me of a scaled-down version of Foresight Exchange (based on Robin Hanson's "idea futures" -- "Could Gambling Save Science?"): but Newsbet is actually going to let people bet with real money?

2000-04-11

(10:56:18)

Advice to people who have trouble with low blood sugar: carry around some non-perishable carbohydrates, or renew your supply of fruit that you can carry with you. You never know when you will be in San Francisco on a Sunday and everywhere you expected to eat will turn out to be closed.

No, this didn't happen to me, but I did see it happen to someone else; it didn't look like it was very much fun. Sorry about that.

If anyone ever found my PalmPilot, I would want to make something like TealMeal that could search by food group, time, and proximity. For example, "What's open near here that I like where I can get carbohydrates?" "What's open near here that I like where I can get salad?" "What's open near here that I like where I can get tofu?"

If nobody ever finds my Pilot, I should just make a map of places to eat in San Francisco. It's an old-fashioned idea, maybe, but useful.

I'm still slouching toward veganism, but my mental food groups are "carbohydrates", "vegetables", and "tofu" -- mainly just "carbohydrates". If I were vegan right now, I would probably eat more processed foods and sugars than any vegan I know -- most vegans seem to gravitate toward fresh fruits and vegetables, but I really don't.

Work and after

I learned about cool stuff Monday.

After that, I went to Fry's with jimd. We talked about Advogato certifications briefly.

Jim found some 802.11 wireless Ethernet cards. They look cool. An increasing number of people are picking up those things.

I found some 100baseTX Ethernet cards -- NetGear in a 10-pack. I figure it is a pretty good deal. People used to make money by buying these 10-packs and then selling the card individually to students in dorms (at some pretty substantial markup). I have no such scalper instinct. I just figure they'll come in handy sometime (especially because I didn't have a single PCI NIC or 100baseTX NIC anywhere, unless you count PCMCIA). If you know me, and you need a PCI NIC, and you see me in person sometime, let me know.

Getting to know me in order to get a NetGear card is discouraged: by the time you get to know me, I will probably already be out of them. :-)

Maybe I should say "Mention this offer, and receive a complimentary coffee mug!" -- Cliff Stoll, who ought to be famous enough that he can't get away with this, said he'd give coffee and a book to people he met who saw the offer on his web page. Only about one person a year took him up on it!

I also bought a few power strips (I was also running out of outlets), and tried and failed to find a standalone video camera with NTSC output and no recording capability. Everything I saw is either a camcorder, or a CCD camera with USB output. The cameras I want should cost less than $50, if only I could find one -- they'll probably be sold as "security cameras".

All of this reminds me that I saw an eeeevil journalist Monday. Come on, you good journalists -- we know you're out there!

What's the difference between a scalper and an arbitrageur? I know some people who say there is none. Maybe the scalper is the person who makes what is already cheap for you expensive (but convenient), while the arbitrageur is the person who makes what is expensive for you cheaper. (This is a psychological distinction rather than solid economic definition -- it's just that some people like the idea of arbitrage, but few people say they're very fond of scalpers.)

On the train

I poked at Zoe-Dan numbers a bit on the train Monday. (This is not an official term in the literature. When I write up something about them, I'll explain why they're called that.) I haven't done anything with them in a few years, but they're on my list of things to mention when I get around to writing up interesting math stuff. So I thought I would take a look again.

I proved that the nth Zoe-Dan number is between n and n! (inclusive), but that's not saying much, and I think we already knew tighter bounds than that seven years ago. So I'm rusty. (I'm not rusty, though -- but I hope he feels better soon, all around.) OK, I also worked out my old proof that Zoe-Dan numbers grow faster than any linear function. I'm still not anywhere near a closed-form formula or even a halfway useful procedure for calculating them.

I also wrote a short letter (on paper) to my ex-girlfriend, but I didn't send it. I don't think I'm going to. I'm not sure exactly what's constructive.

One of the quotations whose source I'm looking for is very close to

All things are permissible, but not all things are constructive.

Ryan

I was very tempted to find a title for Ryan (along the lines of "Ferret Master") to match his appointment of me as "the most verbose little bugger on advogato"... but I realized that he already has more titles than he quite knows what to do with. So I will refrain from appointing him anything else.

Perhaps Ryan is He Who Does Not Need Your Silly Titles.

Western Massachusetts

Wow, another person is from Western Massachusetts (or at least lives there at the moment).

Whatever happened to...

Erik Fair and clock.org? I was following Erik's home page for a while, because of various points of interest -- but he hasn't updated it in two years, and there hasn't been any new material about clock.org for about four years.

It sounded like a very good idea to me when I first heard of it. Is there such a thing as a large, reliable, and explicitly public NTP network? (I know that there are lots of NTP servers; are they co-ordinated and listed anywhere for the benefit of the time-using public?)

Is clock.org actually doing what it set out to do? I know that there are a few machines listed there which are public NTP servers.

2000-04-12

(00:08:27)

I talked to some people from the OSI about the future of that organization, which is always a controversial subject. I hope we'll have the opportunity to get together and talk about it further.

Nick is reliving his former life through radio dramas -- a much better radio fan than I in real life, he grew up listening to all sorts of wonderful serials on the radio and on cassette tapes. Now he's found sources for many such things on CD -- but I'll let him tell you about it.

Work

I worked on cool stuff Tuesday, and met interesting people.

After that, I again got to see Jim and Heather (who doesn't have an Advogato account that I know of). We talked about Linux and about the valuation of common stock. Perhaps some finance geek can tell me whether there is any basis for the valuation of common stock other than dividends (actual or anticipated) and speculative value.

On the train

On the way down, I talked to some people on the phone. One the way back, I managed to sleep a little; I wish that I'd managed to sleep more.

Taxes (and their connotations)

I looked over my tax forms. I've been too busy and too preoccupied this year to spend much energy so far on the political implications of taxes or tax day. I remember, when I was involved with the Libertarian Party, having wanted to have protests and such; this year, it's been more a matter of "OK, it's time to pay taxes, and hopefully sometime later on in the year I can do something to object to that".

The LP does, every year, organize anti-tax protests on April 15 (the day income taxes are due in the US) -- but they usually fail to convey any kind of coherent message to the public beyond "taxes are too high", which is (1) subjective, (2) boring, (3) the sole province and political capital of the Republican Party.

The problem is that interesting and worthwhile messages about political philosophy mostly don't fit in sound bites. The messages of the Republican Party (say, on taxes) fit into sound bites just about as well as those of the Democratic Party. Other people with somewhat more radical positions have ideas which are elaborately justified (I might provoke some stronger-than-intended reactions by saying "well justified", so maybe I can stay with "elaborately" at the moment), but it takes time to appreciate the detail of these justifications, as well as their premises.

So, these things like tax day protests are usually spectacular failures, as conceived, because they are really far "out there", as far as most people are concerned, and it takes a lot of effort to get many people to realize what the underlying message is or to take it at all seriously.

I'm sure that these protests are lots of fun for the demonstrators, though. That's got to be the main reason that so many Linux people in the Bay Area have turned out to protest various things -- it seems that Windows Refund Day had no long-term impact (except for making more people aware of some aspect of Microsoft business practices that they could dislike, and making more people curious about the legal status of shrinkwrap licenses), but it was lots and lots of fun.

... or you may be in a similar situation, and if you're in a situation like that, there's only one thing you can do, is walk into the Microsoft regional office wherever you are, just walk in, and say, "If you do not agree to the terms of this EULA, Manufacturer and Microsoft Licensing, Inc. ("MS") are unwilling to license the SOFTWARE PRODUCT to you. In such event, you may not use or copy the SOFTWARE PRODUCT, and you should promptly contact Manufacturer for instructions on return of the unused product(s) for a refund." And walk out.

I hope I will write a diary entry about the uses of protests at some point.

I'm going to assume that non-US Advogato readers already understand who Republicans, Democrats, and Libertarians are as well as they care to.

Back to taxes: It looks like my withholding for the year was about right, unless there are weird things around stocks. So I only have to worry about California taxes for the time being. (I'm going to file for an extension on both State and Federal, and take my paperwork to a tax accountant sometime over the summer.) I hope that Advogato readers who mean to file income tax in the United States are thinking about it already, in order to save themselves to much last-minute stress.

OK, so some people at Linuxcare and other places filed back in February, but I just can't imagine that kind of lack of procrastination.

Interactive fiction

I remembered appreciating the quality of some of the "interactive fiction" games written after the style of Zork; I really believe that writing a good IF game is a real artistic achievement. (Some people say that they are more like novels than like games.)

Unfortunately, I've never managed to write one myself, nor had the time and patience to solve any puzzle-oriented IF games.

I particularly remember SoFar (extra points to people who can figure out why), and Photopia. Photopia, which won the 1998 IF competition, had no serious puzzles, and you would always "win" (if you could call it that -- we should say "complete") it if you had basic adventure game experience. But that doesn't mean that it wasn't well-done: it's superb.

OK, so neither SoFar nor Photopia can be "won" (by producing an ending which is a satisfying triumph). In that respect, they have a different sort of depth than a more impressive game (and bigger time commitment) like NetHack (which is a non-IF adventure game).

I played Photopia through to the end many times (alternately trying to understand and trying to change the outcome). By then I felt more or less like I did when I read the beginning of book II of the Aeneid -- really sad, and for some of the same reasons.

Or maybe I felt the way Aeneas says he feels there, since I was, or seemed to be, helping to tell the story, but unable to alter it.

Et, si fata deum, si mens non laeva fuisset...

Other people's diaries

I've got some wisdom teeth, and hope I'll continue to (without causing myself other problems).

Happy birthday to joey and to anyone else with an April birthday. (I wish that my mysterious friends with April birthdays would let me know soon, before April is over.)

Paul, of numismatist law student fame, claimed that the new golden dollars are not only the same diameter and weight as Susan B. Anthony dollars, but even have the same electrical and electromagnetic properties as Susan B. Anthony dollars.

Speaking of Paul, chromatic says that the quotation about "all things" sounds like Romans 14. That's true, and I thought that quotation sounded very New Testament-ish, but I can't find that exact line anywhere in Romans (or anywhere else in the New Testament, or anywhere else at all, for that matter).

2000-04-13

(00:54:58)

Tuesday night, I read part of Umberto Eco's book about universal language, and ate some Mexican food (with too much dairy, I think -- I'd really feel better if I kept cutting back). My dinner Wednesday was a large salad-bar salad, which I think was a much better idea.

I have three books which talk about Raymon Lull: Gardner's Logic Machines and Diagrams, Eco's The Search for the Perfect Language, and Mackay's Extraordinary Popular Delusions and the Madness of Crowds.

Gardner talks about Lull as a logician; Eco treats him as a logician-linguist (with ambitions somewhat like those of Leibniz), and Mackay presents him as an alchemist! (None of them can agree on how to spell his name, but all of them agree that he was definitely a religious fanatic -- he wanted to be a martyr, and he eventually got his wish.)

There is actually some degree of overlap between the Gardner and Eco books, because most "logic machines" before the 1930s were part of projects to produce universal notations or languages as part of a program of formalizing and automating human reasoning.

I think this is a fascinating area. If you're interested in linguistics and philosophy, The Search for the Perfect Language is very highly recommended.

I had a whole variety of bad dreams, from kidnapping to the collapse of the Linux community. Fortunately, I'm not known for having especially prophetic dreams.

The SVLUG discussion list got a forwarded copy of a message about how FINCEN was interested in FreeNet and in people who run FreeNet servers. I assume that the FreeNet developers ought to take that as a compliment.

I tried, without much luck, to stay in touch with a few friends, mostly by writing to them.

There was continuing discussion about the legal status of all sorts of random things whose legal status gets discussed on mailing lists I'm on:

• DeCSS
• cphack
• the GPL
• shrinkwrap licenses
• gratuitous licenses
• FreeNet
• and more

Someone from American Lawyer called back to check a quotation about OpenLaw that I had given to a reporter a while ago. She seemed to have a hard time understanding that I wasn't a lawyer and that tenured Harvard law professors would want to get a bunch of programmers together to talk about DVDs.

Lawyers have a stronger tradition of worrying about professional training and qualifications than do programmers. Programmers can deal with the "No, I don't have a college degree, I've just been doing this since elementary school" thing, which might sound strange coming from a lawyer. Why is this?

It's worth noting that I'm already a vegetarian (i.e. don't eat animals) and am thinking of becoming a vegan (i.e. wouldn't eat animal products like milk or eggs). I wish mbp lots of luck in becoming a vegetarian, though.

I am skeptical of his "disemvowel" script: it (self-consciously) puts punctuation and other non-alphabetical stuff both above and below the line. But I don't have a copy of Microserfs, so I don't know whether that's intended or not.

Work

I worked on cool stuff.

In the meantime, I got asked a question which led me to a C question about pointers to functions, but I thought I should check K&R before mentioning it here, and, sure enough, 2nd ed., p. 119:

Since they are known to be functions, the & operator is not necessary, in the same way that it is not necessary before an array name.

I find this decision irritating, but it does answer my question, which concerned why there were so many different valid syntaxes for forming pointers to functions. (I still don't see why Kernighan and Ritchie, in their infinite wisdom, did it this way.)

I never had occasion to use explicit pointers to functions before, outside of someone else's sample code, which already did it correctly. (I still don't have occasion to use explicit pointers to functions, actually, but I wondered what was going on.)

On the train

People on trains in the Bay Area talk on cell phones about IP addresses and firewalls. OK, so I talk on cell phones about IP addresses and firewalls while I'm on trains, too, but that doesn't make it normal!

And they read books that use big-sigma notation or talk about elliptic-curve cryptography. OK, so I read books that use big-sigma notation or talk about elliptic-curve cryptography, too, but that doesn't make that normal, either.

Some people even play Quake on trains. I wouldn't do that -- Quake makes me feel motion-sick enough when I play it in a room indoors. If I didn't have so many other things to do, I would play NetHack on a train. (That sounds like something out of Dr. Seuss.)

Ryan wants to flee Canada?

Rob says that Ryan would be welcome in San Francisco if the latter got sick of Ottawa and wanted out. (Funny how that works -- I was just wondering myself whether it would be fun to live in Ottawa, but not for the "not living in the Bay Area" problem that tends to result from that sort of thing. Maybe we should have an exchange program or something.)

Anyway, if I remember correctly, Ryan's already tried living here, once upon a time. Want to give it another shot? We've got springtime, corruption, fog, pretty historic buildings, and way too many web companies (all of which advertise on the radio). We've also got chocolate, oceans, bridges, more fog, some rainbows, interesting people, universities, electric motor scooters, cable cars, a lot of interesting ethnic food, and a very dramatic gap between rich and poor. We've got islands and ferries and nearby redwoods, and bicycles and bicycle activists, and some other stuff, too. We've got Muni Metro, a strange microclimate, and almost no mosquitoes. We've got earthquakes, and fountains that have survived them. We've got a huge tourist industry and a lot of different kinds of mass transit. We've got a new stadium three blocks from our apartment that Nick is really unhappy about. (They play baseball there, by the way, Nick.) We've got neighborhoods in transition and people who are unhappy about that; we've got refugees from the East Coast (and Vietnam and China and Mexico, and elsewhere). We've got Linux user groups and fog.

It is legally difficult, although not impossible, to keep ferrets in the City. You can fight City Hall, but you will typically lose.

Most of those things could probably have been hyperlinks, but I thought I'd hold off this time.

On the other hand, it's always worth keeping in mind that

caelum, non animum mutant, qui trans mare currunt.

chromatic finds the quotation

Thanks to chromatic for finding 1 Corinthians 6:12 ("all things are ... but not all things are ..."). I'm impressed that my strategy of asking idle questions in an Advogato diary entry produced such a quick answer.

Now I should try to remember what else I was looking for.

libyou-have-no-right-to-distribute.so.0

I looked at a couple of proposed open source licenses for the OSI. I hope that a little bit of the backlog in license review will be gone soon, although that's not an unequivocally good thing. (The existence of the OSI has led more people to create licenses. A lot of those licenses are incompatible with one another. That might hinder code re-use between projects which are released under incompatible licenses.)

Upcoming events

zw: thanks for the reminder on Douglas Hofstadter's talk.

Now I need to know the dates of a few other things, so I can write them all down on a calendar. But I bet I can find them out for myself, more or less.

Unfortunately, my current work arrangement might make it difficult to get to talks at Berkeley in the early evening.

Massachusetts

Wow, more people who have lived in Massachusetts. I thought of going to Simon's Rock once myself, right before I started to enjoy high school. I remember visiting a very interesting class about Immanuel Kant and game theory. I subsequently enjoyed high school so much that I didn't at all regret not having gone off to Simon's Rock. Fortunately, though, it's there for people who need it.

The mind is its own plac