Older blog entries for chalst (starting at number 91)

mikehearn: Really, really excellent post, topping a series of excellent posts. The discussion of COM and object models has been one the most interesting topics here at advogato in recent weeks. Just a point about KDE hostility to object model standardisation on a GTK standard: I think we are starting to put behind us the days of religious conflict between the Gnome and KDE camps. Making GObject palatable to the KDE project means finding practical reasons that it is worth KDE developer's time to embrace it. It's harder to make that case when you already have an object model you are happy with.

Murphy, Prince of Darkness There's a paper by Wietse Venema outlining four important and misunderstood kinds of security risk, that I recently rediscovered, entitled Murphy's law and computer security. This is a nice, well-illustrated paper that I strongly recommed.

16 Sep 2003 (updated 20 Apr 2007 at 08:56 UTC) »
Best of the Blogosphere A personal perspective
A valuable activity we can do in our diaries is help our friends and colleagues sift what is good from what is not so good. I've asked a few times for people to submit their lists of what they like, but never with any success, so, hoping to provoke by example, is my tentative first attempt...

The best weblogs, by category:

  1. Politics: Josh Marshall of Talking Points Memo. Sample on Wesley Clark's chances.
  2. General commentators: Eugene Volokh and David Bernstein at the Volokh Conspiracy. Two posts on "Jews for Jesus" by David and by Eugene. Postscript: After an interesting start, it's a shame to report that Berstein seems to have degenerated into a cliched war hawk culture warrior. But conspiritors Orin Kerr & Jim Lindgren have consistently & richly rewarded reading.
  3. Computer Science: Ehud Lamm's Lambda the Ultimate. Sample on A Logic File System.
  4. On intellectual property: Larry Solum. Sample on Water Wells and MP3 Files: The Economics of Intellectual Property, the latest post in a dialog with Eugene Volokh. A must read.
  5. Here at Advogato: Bram Cohen. Sample on Bram's law. Postscript: ncm wrote some reflections worth reading on Bram's Law.
  6. Media commentator: Felix Salmon. Sample on Korba the Dread.
  7. Interviewer: Kevin Drum, at Calpundit. Sample interview with Paul Krugman.
  8. Polemicist: Oliver Kamm. Sample on Chomsky and third-world liberation.
  9. Political theorist: Norman Geras. Sample on top ten lists.

Nominations are sought for the categories of:

  • Best Slashdot journal;
  • Best Euroblog (ie. a group blog covering several different EU nations).
(in other words, I can't seem to sift the wheat from the chaff on these topics).

Updates Changed the samples for Josh Marshall and Oliver Kamm, changed nominations sought.

I have posted an article about today's anniversary to the frontpage.

tk raises three points, all interesting, about a three sentence paragraph from my last post:

  1. What is reason? I don't think reason is reducible to logical inference. Before I give a proper response, a counter question: why are Aristotle and Frege missing from your Boole, Hilbert, Brouwer, Gentzen, etc list? Especially in the context of Objectivism, omitting Aristotle seems perverse!
  2. Respect for people and respect for their beliefs. I think if you really respect someone, you should respect their beliefs. Of course you should not accept every belief you respect; for most people this is a sure route to inconsistency.
  3. On cults: I have more to say, but in view of todays date I will wait before I reply.

I had a short dialog with Bob Solovay following Bram's and my recent comments on playing games with sentences of ZF set theory. I'll post something about this in the not too distant future, I just want to say I have at very long last solved the puzzle thanks to a kind hint of Bob's.

9 Sep 2003 (updated 9 Sep 2003 at 05:20 UTC) »
Meandering Thoughts on Film Criticism
I found a fascinating link chasing a link from the latest bout of Andrew Sullivans conflict between his homosexuality and his attachment to the Catholic Church, namely The Vatican Film list at Steven D. Greydanus' DecentFilms.com. Interesting for two reasons, firstly how much the Catholic Church approves of challenging, counter-orthodox ideas in its film choices, and secondly how good I find the religiously-motivated film criticism of Greydanus to be. In particular, I find the reviewer shows an admirable job of treating high brow, low brow and middle brow films on their merits, as very few film critics succeed in doing, and I think the reviewer's religious orientation is a key to his success in doing this.

Felix Salmon, a friend and a damn good blogger to boot, has written about Crap writing about mainstream movies, but he concentrates mostly on the commercial reasons why there is little in the way of good quality reviews of mainstream movies. There is another issue: even when you look film reviewers who work outside these business pressures, good reviewing of low brow films is much harder to find than good reviewing of high brow films. Here's an example: the only review I found to be at least moderately insightful of of Eminem/Curtis Hanson's 8-Mile was a guest review at Greydanus' site. This is clearly an important popular movie: why the lack of thought-provoking reviews? Reviewer after reviewer makes the connection between the film's form and the convention of Hollywood sport's movies, many talk about its punchy portrayal of urban despair in Detroit, and say almost nothing beyond this. Even Jonathan Rosenbaum's normally reliable Chicago Reader reviews gives only a capusle review, despite regarding it as one of the 10 most important film releases of 2002.

Chomsky in the firing line
Brad De Long wrote a nice put-down of irritating attempts to reason with those far gone on Chomsky idolisation. His characterisation of Chomsky as an intellectual totalitarian is spot on, and not just in his political work. I note that the wikipedia entry on Chomsky is excessively deferential to the great man.

I eagerly await Oliver Kamm's forthcoming overview of Chomsky's political work. I put some links in his comments page:

  1. A linguist-list review of Randy Harris' landmark study of The Linguistics Wars.
  2. A review by Randy Harris of Barksy's brown-nosing biography of Chomsky.
  3. Another review of Barsky's book by John Goldsmith.
I'd be grateful for any more references to intellectually honest exposes of the Chomsky myth in any area of his work.

Even more links

  • Dan Drezner writes about the progress of graduate students in The Art of Criticism.
  • Objectivism reaches the shores of advogato, courtesy of shlomif. I encountered Neo-Tech before, a few years ago: it looked like some cult-cum-confidence trick to me back then; is this unfair? I don't regard objectivism as a whole either as wholly absurd (at least one person I respect highly believes in it), or pernicious (reason-centered credos can't be entirely bad in my book), but I find Rand's invocation of Aristotle to be comically bad in a great many more respects.
  • I've recently rediscovered John Baez's This Week's Finds in Mathematical Physics: I used to read this devotedly when I was doing my doctorate, but lost the habit when I finished. I've restarted: it's a pleasure, and I recommend it to anyone with a background in algebra, geometry or physics.
  • I have a feeling that there is an important link between the ideas the nanocorp promoters have and the sociological ideas that drive free/open software promoters, and I'd like to see more attempts to get the two communities to talk. garym is a distinguished member of the set who understands and applies both sets of ideas; I wish he saw more success in applying them. In response to his five pronged dilemma, I think 1-3 are bad options, 4 and 5 are good, totally different, and the scenario needs more information to decide. I guess option 4 is how the textbook solved this puzzle?
  • Larry Solum has posted a much-needed index to his writings on copynorms.
  • I've found it difficult to make up my mind about whether the US was justified in invading Iraq, which I couldn't decide beforehand. An important milestone in the post war aftermath will be the publication of the Kay report later this month; I suggest this CEIP article is a useful perspective on what to expect; from following Josh Marshall's TPM, I think this article reaches correct conclusions.
  • I still read Michael Glazer's diary, because he occasionally throws up interesting links, even while foaming at the mouth. This questionnaire amused me: the author doesn't give a "pass mark" for being a leftist, but since I am a zero scoring leftist, I assume the figure must be low... Actually, excepting the obvious possible mark for voting Nader, does anyone here score above zero? What for?
  • From the bad taste department: Everything you wanted to know about Hamas but were afraid to ask.
Postscript: ReciprocalityTheory
A slightly whacky idea that has been doing the rounds has been Alan Carter and Company's work on Reciprocality Theory. I say slightly whacky, because while I think the ideas are almost certainly wrong, I think there is interesting new data gathered here and the ideas might point in the direction of sound science, especially since the authors have made efforts to make their theories empirically testable. Probably the best line of investigation into their ideas lies in their idea of the MzeroParasite, on which I there opined:
It's an interesting hypothesis, and it would be valuable to see it taken up by competent neuroscientists, but I'm afraid Alan Carter's writing style means that a neuroscientist working under usual working pressures is going to find it hard to justify the work. My guess is that his theory is probably `interestingly wrong': plenty of health workers don't think the currently accepted etiology and treatment practice of ADHD is right, and if Carter's hypothesis is wrong it may still be pointing in the right directions. Carter needs to write up his stuff in a way that his thesis is interesting to the ungenerously skeptical neuroscientists, which I think needs a cooler, leaner paper.
18 Aug 2003 (updated 9 Sep 2003 at 00:39 UTC) »
Macros and error messages
In a post I otherwise agree with, raph makes the claim:
Macro substitution, while an appealingly powerful abstraction, is absolutely hopeless when it comes to mission-critical error recovery.
If raph is talking about m4 (which in the context of a discussion of autoconf's failings is possible), this seems fair. But the paragraph suggests this applies to all languages that support macros. lukeg points out that CMUCL supports sophisticated error messages for macro expansion. Even in their absence, I can say from my own macro-heavy programming style in various LISPs that I don't find debugging macros to be especially difficult: most debugging time is spent on the procedural side. Also C programmers make heavy use of macros, and again one doesn't hear from them how difficult their (admittedly less sophisticated) macros are to debug. The difference between these two languages and m4, is that for them it is a matter of using transformations on program source to make source management easier and provide shortcuts to a more intuitive programming idiom. In the case of m4, it is the source of all computational power.

Nerdy, inane, and barely grammatical
Via Dan Drezner, I see that The Economist passes opinion on the weblog phenomenon. Interesting figure: there are about 750 000 webloggers in the world today.

Postscript: Google anomaly
I've stumbled across an anomalous search query on Google: at the moment it gives two results, with the promise of more but clicking on "More results from hnn.us" gives almost identical results but for more "site:hnn.us" strings being added to the query. Odd.

15 Aug 2003 (updated 18 Aug 2003 at 15:19 UTC) »
Bad cases make bad laws
With the recent assertion by SCO's chief counsel that `The GPL is invalid' based on an clearly absurd interpretation of copyright law, their case has completely severed what few strands of credibility it once possessed. Some people still argue that this is an important lawsuit, because of the possibility of related actions. This is to neglect the useful lawyer's maxim that bad cases make bad law. I don't think that significant precedents on IP will come from this farce, and the only non-prurient interest I see left is to understand why a well-regarded law firm like Boies, Schiller and Flexner has let its reputation be abused in this way.

Voting machines
Bram argued in his 13th August diary entry against the accuracy of electronic voting machines: in my last post, I tacitly made the assumption that if electronic voting systems are ever deployed, they will perform better than mechanical systems under normal circumstances. I realise that there are grounds to believe this an optimistic assumption; still it is important to bear in mind how bad mechanical systems are: apparently almost 1 in 30 votes is miscast or ignored. Also, there is no reason why electronic voting should not create a paper trail, although, I agree, most current proposals do not.

Bram's minefield
In Bram's 6th August entry Bram introduces a `game' played over ZFC: even in the simplified case of large cardinal axioms (LCAs), there may be no fact of the matter of which is stronger: if you look at the diagram of important LCAs in Kanamori's "The Higher Infinite", they form a partial order, not a linear order (I don't know whether this reflects known independence results or just ignorance). I do think as a kind of thought experiment, your exercise is illuminating; I like to think about how different mathematical pholosophy might be if results came to known in different order, might we regard a much weaker or stronger set theory as the default?

The Knight's Tour I
First in a series of posts on a combinatoric problem. A knights tour of an MxN chessboard is a sequence of squares of the board, where there is a valid knight's moves between each square and its successor, and where each square of the chessboard occurs exactly once. The tour is closed if its first square can be reached from its last. Can you prove that there are no solutions for 2x2, 3x3, and 4x4 boards, or closed solutions for the 5x5 board?

Erratum: The original version of this entry asked to prove there are no solutions to 5x5 boards, which is false: there are solutions, only no closed solutions.

13 Aug 2003 (updated 13 Aug 2003 at 10:11 UTC) »
raph: Indeed, most convincing, thanks. The IACREOT website is interesting in a perverse sort of way, the organisation seems to have once been a functional professional organisation now twisted by the drive for revenue. I feel much the same way about the ACM: I have a membership, but I refuse to endorse them. While good scientific work is done under the ACM umbrella, the typical price of ACM conferences is an outrage, and academically exclusionary, and I think academics should as far as possible see if they can do their work outside the ACM. Kent Pitman wrote an amusing polemic against the Communications of the ACM, but I can't find the link now.
Postscript: I found the Kent Pitman article, it was a comp.lang.lisp post. Rather nice polemic, and I basically agree; while I don't think it is practical for most computer science academics to boycott ACM conferences and journals, I do think awareness needs to be raised about what is wrong with the ACM today.
12 Aug 2003 (updated 28 Oct 2009 at 12:32 UTC) »
Electronic voting systems
Steve at Begging to Differ argues against the security concerns voiced regarding electronic voting systems, thinking that on average the new machines will be better than the old. IMO, he's completely missed the point. I think that while electronic voting systems reduces the overall chance of error in elections, and gives a possibility of completely correct votes, they do so at the cost of making massive deliberate distortions of election results much more likely. Tampering with punched cards is a time consuming process, that must be done on a one-by-one basis, and carries a risk of detection something like proportional to the size of the fraud. Finding a backdoor in an election machine may allow tampering with as much of the vote as the fraudster thinks is credible, and can be replicated on every similar machine the fraudster has access to (conceivably remote access, in the worst designs).

A good side effect of the debate is that perhaps the wider public will start to be suspicious of the security by obscurity argument.

raph: How do you know there were voting machine representatives in that meeting? It wasn't in the Denver Post article or Dan Gilmour's blog. Also, Gilmour misstates the Denver Post article, which says she was ejected because she was not a public official, rather different than she wasn't credentialled.

Postscript: They don't call it that over there either
Eugene Volokh has has a nice puzzle: for how many countries are the native names for the countrry unrelated to the name in english? The answer is five, and an interesting question is how does that figure change for other European languages? By my reckoning there are four for german. We have fluent French and Spanish speakers (at least) here: what are the figures for these languages? Are there interesting differences to cases for english (there aren't in the case of german). This resource may be useful.

Postscript #2: At last...
At last, the definitive, credible exposure of the civilisation-hating Maoists at the BBC:

The BBC is pathologically hostile to the Government and official opposition, most British institutions, well-polished shoes, American policy in almost every field, stamp-collectors, Israel, moderation in Ireland, lightly boiled new potatoes, the colour mauve, freedom of speech in the Isle of Wight, all Western religions, most odd numbers between 32 and 57, sun-roofs on Ford Escorts, Crunchy Nut Cornflakes, most manifestations of the free market economy and household gadgets of every kind.
3 Aug 2003 (updated 3 Aug 2003 at 17:50 UTC) »
Intellectual Property
This entry is all information (ie. links), and no thoughts.

The third installment (here are the first and second) in Larry Solum's series on IP has been posted. This is a really excellent post, concentrating on the question of whether the RIAA has any chance of effectively eliminating file sharing (by which he means file sharing of proprietary works, irritatingly Larry doesn't acknowledge the existence of file-sharing of non-proprietary works).

Larry links to the Intellectual Property Page, and excellent resource that I don't think I ever encountered before. Also Randy Barnett at the Volokh Conspiracy provides brief commentary on Larry's post, where he interestingly comments that he will be developing an anti-IP argument in a seminar series he will be giving at Boston University.

Text and hrefs
I'm interested in what opinions exist on how to mix text and hrefs in HTML. As far as possible I try to make the text in an HTML document between anchor tags be informative as to what the href is, and to be a determiner phrase (eg. "two arguments against rhubarb", or "the definitive reference to custard"). Is there anything like a style-guide to these sorts of issues?

zhaoway: you are being rather mysterious. Don't worry about Erik, he only bites posters to comp.lang.lisp. In any case, isn't it better to get bitten early than suffer deeper bites later?

tk: Indeed you are right, it's not so difficult, and I'm not sure what Mike thought the problem was. Perhaps the point is that you can't parse Haskell/python using lex/yacc without the use of persistent state in the lexer. The right data structure to handle indentations is a linked list; you can then count the number of times you pop the list to see how many OUTDENTs to issue.

Terrorism and many eyes
Oliver Kamm writes about the idea of a futures market where one can bet on future terrorist attacks. He argues the plan is incompetent, because it tries to make predictions in the absence on public information. However, he says a similar idea would work, namely using market mechanisms to identify risk areas for terrorist attacks.

There's an interesting point of similiarity between this idea and the idea that open source projects tend, others things being equal, to be less buggy than closed projects. Both have the criticism that attackers can use the published information to find vulnerabilities, and the counter-argument is comparitively complex.

82 older entries...

New Advogato Features

New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.

Keep up with the latest Advogato features by reading the Advogato status blog.

If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!