When it rains - it pours. Seems I'm going to be a lot more busy this holiday that I had planned. Connectivity will be patchy at best over the comming days.
Yesterday I made the comment that Jon Searle's chapter in "Are We Spiritual Machines?" was rubbish. Here's why:
I'm going to run a commentary on Searle's chapter and, in doing this, I'm leaving a lot out of course. You are welcome to read the chapter in full to get the context for these quotes. I purposefully avoided reading Kurzweil's reply.
Here is the difference: Kasparov was consciously looking at a chessboard, studying the position and trying to figure out his next move. He was also planning his overall strategy and no doubt having peripheral thoughts about earlier matches, the significance of victory and defeat, etc. We can reasonably suppose he had all sorts of unconscious thoughts along the same lines.
From that I would like to continue a little: Kasparov was studying the consequences of a set of possible moves. His subconscious has filtered most of the illegal moves before his conscious even considers them. The impressive pattern matching ability of his brain is making links with previous games and recalling which moves proved advantageous in those situations.
Does anything think that is unreasonable? Where is the magical factor in that?
The computer has a bunch of meaningless symbols that the programmers use to represent the positions of the pieces on the board. It has a bunch of equally meaningless symbols that the programmers use to represent options for possible moves. The computer does not know that the symbols represent chess pieces and chess moves, because it does not know anything. As far as the computer is concerned, the symbols could be used to represent baseball plays or dance steps or numbers or nothing at all.
Does Searle think that my visual cortex is aware of what chess is when it highlights edges and surfaces? The pattern matching systems of the brain are very general (try staring at clouds or even a blank wall for a while) and certainly aren't designed for processing chess positions. Why is it that symbol processing deserves such contempt?
I'm sure Searle would argue that there is a conscious kernel of the brain which is somehow `above' the mundane tasks of the brain. So is a child, playing for the first time with a rule book, not really playing chess? The child is doing the same as the computer is, looking at the rules for valid configurations; judging those configurations and then moving. At what point does the child actually start to really play chess?
Let us call it the Chess Room Argument. Imagine that a man who does not know how to play chess is locked inside a room, and there he is given a set of, to him, meaningless symbols. Unknown to him, these represent positions on a chessboard. He looks up in a book what he is supposed to do, and he passes back more meaningless symbols. We can suppose that if the rule book, i.e., the program, is skillfully written, he will win chess games. People outside the room will say, "This man understands chess, and in fact he is a good chess player because he wins." They will be totally mistaken. The man understands nothing of chess; he is just a computer. And the point of the parable is this: If the man does not understand chess on the basis of running the chess-playing program, neither does any other computer solely on that basis.
This is the Chinese Room argument (for which, Searle is best known). Usually the man in the room is processing Chinese symbols.
Again, the fallacy in this argument occurs in two places. Firstly, no neuron in my brain understands anything. If I filled a Chess Room with 1000 people, handling the task as a team, why do you expect any of them to understand it? The understanding is an emergent property of the actions of all the people and the records they are making.
The second problem is very similar to the one I outlined above. I'm sure this poor man will quickly become quite fast at processing the rules if he spends any reasonable length of time doing it. At some point he may recognise that the data he is handing out look like 2d coordinates and he could draw a grid inside his box to help him keep track with symbols for the pieces. With no previous knowledge of chess he could become quite adapt as his brain starts recognising patterns and caching the results so he doesn't repeat the same calculations. At what point is he really playing chess then?
Imagine that I, who do not know Chinese, am locked in a room with a computer program for answering written questions, put to me in Chinese, by providing Chinese symbols as answers. If properly programmed I will provide answers indistinguishable from those of native Chinese speakers, but I still do not understand Chinese. And if I don't, neither does any other computer solely on the basis of carrying out the program.
Reread the reply to the last quote. The faults are the same.
Kurzweil assures us that Deep Blue was actually thinking. Indeed he suggests that it was doing more thinking than Kasparov. But what was it thinking about? Certainly not about chess, because it had no way of knowing that these symbols represent chess positions. Was it perhaps thinking about numbers? Even that is not true, because it had no way of knowing that the symbols assigned represented numerical values. The symbols in the computer mean nothing at all to the computer. They mean something to us because we have built and programmed the computer so that it can manipulate symbols in a way that is meaningful to us.
I'm left wondering what, exactly, chess pieces mean to us that makes Deep Blue so fundamentally different. If I were to rename and redesign the pieces (without changing the rules) I'm still playing chess. Now if I number them all and replace them with bits of paper, I have to remember what the numbers mean, but I'm still playing chess. Now if I get rid of the board and imagine all the numbers on a grid in my head, I'm still playing chess. If I don't imagine a grid, but remember the positions also as numbers, I'm still playing chess (but I don't get to use the pattern matching systems of my brain). What's so different between myself and the computer now?
He confuses the computer simulation of a phenomenon with a duplication or re-creation of that phenomenon. This comes out most obviously in the case of consciousness. Anybody who is seriously considering having his "program and database" downloaded onto some hardware ought to wonder whether or not the resulting hardware is going to be conscious.
(in this, Searle is talking about downloading yourself into a computer by mapping and simulating your brain - Diaspora style)Exactly what is so magical about neurons? Most animals have less advanced myelin sheaths than humans do, and so have slower moving nerve impulses. This small change doesn't disqualify them from Searle's elite club.
It's unfortunate that all higher animals share the same basic nerve structure, however, as it isn't possible to point to another example. But we have a pretty good understanding of how nerves work, so I can pick a single neuron and replaced it with a device that records incoming impulses and can trigger outgoing impulses. This device communicates via radio to a computer which controls it and simulates the neuron I replaced. (this is just in theory, I'm not saying I could do this today).
If I continued to replace neuron's with perfect simulations of them I doubt even Searle would suggest that I'm altering anything about my brain functionally. So either I can replace my whole brain that way and simulate it all in the computer (at which point the actual neuron replacing devices can be discarded) or he's suggesting that there is something so special about neurons that they cannot even theoretically be simulated.
Actual human brains cause consciousness by a series of specific neurobiological processes in the brain. What the computer does is a simulation of these processes, a symbolic model of the processes. But the computer simulation of brain processes that produce consciousness stands to real consciousness as the computer simulation of the stomach processes that produce digestion stands to real digestion. You do not cause digestion by doing a computer simulation of digestion. Nobody thinks that if we had the perfect computer simulation running on the computer, we could stuff a pizza into the computer and it would thereby digest it. It is the same mistake to suppose that when a computer simulates the processes of a conscious brain it is thereby conscious.
I'm afraid I'm not even going to dignify that analogy with a reply. It's just absurd. Unfortunately for Searle, the absurdity is less hidden in this case than in his arguments above.
"This thesis says that all problems that a human being can solve can be reduced to a set of algorithms, supporting the idea that machine intelligence and human intelligence are essentially equivalent."
That definition is simply wrong. The actual thesis comes in different formulations (Church's is different from Turing's, for example), but the basic idea is that any problem that has an algorithmic solution can be solved on a Turing machine, a machine that manipulates only two kinds of symbols, the famous zeroes and ones.
(The part in italics is Searle quoting Kurzweil)
In what way is my neural net fundamentally different to a computer? It may well have some impressive emergent features but if you are suggesting (again) that neurons are fundamentally different from a UTM? At this people people usually start muttering the word "quantum" as an answer. Firstly, we have quantum computers anyway and, secondly, I'm pretty sure that a UTM can simulate a quantum computer and that the quantum aspect is just a matter of speed. (can anyone confirm/deny this?).
The rest of the chapter is just Searle sniping at Kurzweil and he doesn't put forward anything new.
The conclusion is at the top in the Summary box really. I won't repeat it here.
Joey has a great tribute to Gene. I actually have no recolection of signing Joey's book, but I'm glad that I've met him unknowingly! 
Crumbs - I'm playing a part in Aaron's dreams
LiveJournal has become the latest RSS agent. You can now add RSS feeds as friends (for example theregister).
"This is nothing. In a few years, kids are going to be demanding septal electrodes." Timothy Leary. Hasn't happened yet though Tim. The links at the bottom of the page are highly worth a read.
Have been reading Jon Searle's chapter in "Are We Spiritual Machines?". It's well written, but just .. wrong. I'm sorry, but his arguments are just rubbish.
(quick warning, kurzweilai.net is a pile of crap in terms of presentation, but the content is good)
It seems that Gene shot himself
Dubya losing the benefit of the doubt from the National Post (Canadian) [via Keith]. He still has a 70% approval rating though.
Notes on Fitz. Still vapourware at the moment but with people like Raph involved, it could be quite something.
The detailed designs of the MS TCP/IP stack [via coderman]. It has a little market-droid speak in it at times ("strategic enterprise network transport for its platforms" <- bullshit alert), but it has enough real content to be worth a skim. A recent traffic sniff of a root DNS server showed that Dynamic DNS requests to alter the root (from MS's screwed up implimentation) made up a significant fraction of the traffic.
Pretty graphs. It's sites like these that make 56K users feel it.
The person with the file:///dev/null turns out to be Ian Hill:
In ELinks you can set a fixed referer in the options menu. In fact it tries its best to stop you sending *real* referers by flagging that option as "Default - Insecure" It also lets you have no Referer: at all!.
ELinks is a modified version of Links which does SSL (apparently, I can't test it though as it wont use a proxy).
I'm pretty sure that standard Links does SSL too, but it doesn't have the Referer thing. Finially a brower which actually follows that obscure bit of the RFC
Paul Graham assures me (via email) that stuff is still happening to Arc (his new Lisp). Unfortunately the code seems to be non-public at the moment.
Channel 4 haven't been showing reruns of Ally McBeal in the afternoons this week so I'm going to have to watch the deaded new series tonight.
Wired has a piece on Gene's death. They suggest it's suicide too.
I pulled my comment that I suspected suicide when the Washington Post text came out and the family said it was an accident. Maybe the family think there's something shameful in it? I'm saddened.
Salvia - not for the conservatives in our readership.
Zooko's mail is still mounting up (165 messages now) and the logs are filled with the retries. I've made Zooko a local on metis now - I'll send him the mbox when he's alive again. Unfortunately, since qmail has queued the other 165 as remote it won't deliver them locally. I guess I could point zooko.com to metis in /etc/hosts but I don't want to play about like that. Hopefully Zooko is home soon.
Airhook looks very cool (read the page for the links at least). It's a TCP replacement protocol with a number of advantages and it can be used as a library using UDP.
Are capital letters in URLs considered harmful? I have a couple of failed attempts to access the Hitch Hikers text with the wrong capitalisation. Do some things assume lower case?
A quick congrats to the person with a referer of file:///dev/null (email me!). You are going to send the deep-link protection sites nuts with that!. Although the HTTP 1.1 spec says browsers SHOULD provide a way to disable sending Referer headers, I don't know of any that do. Personally I use privoxy which sets the Referer header to match the host name (for crappy anti-deep-link sites).
I guess that makes me hypocritical since I love reading the weird searches that turn up IV!
Gene Kan has died. I feel a little ashamed for missing it, but everyone seems to have. I'm sorry Gene.
[link1] [link2] [link3] [link4] [link5]
Still no solid word on how such a young and healthy guy passed away.
Update: added a fifth link, which says it was an accident
Zooko has been away for ages and his mail server is down. Since IV is the backup MX the queue has been filling for a while and it's now at 101 messages and counting. I've just upped the queue lifetime to 2 weeks to make sure they don't die, but still no sign of Zooko
Full book: "Are We Spiritual Machines?" posted by Kurzweil. Also available in dead-tree format.
An old (1993) talk by Vernor Vinge about the Infosingularity.
When Knuth wrote the Art of Computer Programming he used a fictional assembly language to do the examples in, called MIX. Well, MMIX - the 64-bit updated version of MIX - has been around for a while. You can get the documentation here
Well, someone has ported GCC to MMIX and it works pretty well. Grab the latest GCC 3.1 and binutils 2.12.1 and build gcc with the --target=mmix option to configure and it all goes swimmingly.
Not sure why you would want to do this, but it works
# 1 "test.c" ! mmixal:= 8H LOC Data_Section .text ! mmixal:= 9H LOC 8B .p2align 2 LOC @+(4-@)&3 .global main main IS @ SUBU $254,$254,8 STOU $253,$254,0 ADDU $253,$254,8 LDOU $253,$254,0 INCL $254,8 POP 0,0 .data ! mmixal:= 8H LOC 9B
I think I need some simple project to code on. Either helping with an existing one or something nice and short term that I can see the end of. (or someone could give me a real job - heh, yea right).
All my ideas are far too far out and I keep cycling. At one point I'm thinking that I'm nuts and will never manage any of what I'm planning so I cut the plan down hugely. Then I start thinking and designing and pretty soon I'm right back to where I started, by a different path.
This is getting really annoying because, by the time I'm at to the `cut it all back' point I'm planning a Turing-capable AI.
Landscape (there is a page for Landscape in the sitetree - but it sucks) mearly involves:
And that's a sane idea by my standards!
Oh crap
Quoted in another New Scientist article. This one doesn't link to IV unfortunately.
Very good text on the up and comming 64-bit chips: POWER4 and Itanium2. (the author discounts Hammer, which is a shame because I like it from what I've heard)
Seth Schoen has more notes on Palladium. [via Wes]
Seems the Earth will expire by 2050. I wonder if we hit info-singularity before we all die out?
RAVE Act: Reducing Americans' Vulnerability to Ecstasy. Sometimes I plain just don't get the USA. I guess a whole multitude of factors just increases the wanker/square meter count there.
Seems that all the new Evas work is going into evas2, which rasterman pointed me to. My UTF8 support is already in there . (and it means I don't have to patch Imlib now)
Evas2 seems to be a lot cleaner code but it's not as mature at the moment. For one, deleted objects don't actually disappear, which makes for some interesting effects to be sure. The API naming is a lot cleaner thou, it now has consistent noun-verb style everywhere.
Raster asked why his Japanese ttfs didn't work with UTF8. It seems that they map their characters into the Latin-1 region rather than their Unicode regions. I have no idea why. I suppose there must be some old encoding which they are using for when everything was ASCII.
JWZ was asking how to fix the sidebar in his calendar (read the LJ post for the full specs). I suggested that he could hide CSS from NS4 and IE3 by putting it in a @media screen block - which works really nicely.
Except that in order for NS4 to render it correctly you have to put the fixed sidebar in a table. This breaks konqueror unfortunately, which doesn't render the sidebar at all. Damm NS4
Janis Ian on music copyright. Nope, I've never heard of her either but it's a good text (even if it is preaching the choir here).
Sunshine Project on biological weapons.
IRC chat with Ray Kurzweil and Venor Vinge.
I have dates for flying to S. Korea now (16th and 25th of August) so I ambled down to my GP today for any injections that I need. Seems I only need Hep A and the needle wasn't too bad. (I don't like needles).
(yes, Ian (if you read this) that means I can't make your party - sorry!)
Also Lionhead, the sponsors of the UK IOI team, have offered me a job for the summer. I kindof wanted to stay around this year, but since my other job prospects are somewhat, erm, crap, I might take it up.
Added UTF8 support to Evas today. (Evas is the Enlightenment canvas). Unfortunately the ALPHA_SOFTWARE backend uses Imlib to render text, so I have to add UTF8 support to that too.
Thankfully, the M$ core fonts (one of the few good things ever to come out of That Company) have a Unicode table in them. Wrote ttfdump to display the code points in a given font and some of the M$ fonts have a scattering of U2200 glyphs (the maths symbols). Unfortunately, Wolfram's maths fonts put all the glyphs in the private use area, prat.
Will Knight pointed me to this paper on searching Gnutella networks. Some commentry:
The motivation for this metric is that in P2P systems, the most notable overhead tends to be the processing loadthat the network imposes on each participant.
The processing load is their most notable overhead? They must have one hell of a bandwidth or be really bad coders.
If a PC has to handle many network interrupts when it joins the P2P network, the user will be forced to take the PC off the P2P network to get "real" work done.
Because of interrupt load? I'm wondering if this paper has been translated from another language. Once again, the limiting factor is bandwidth and certainly not interrupts.
Other than that odd misunderstanding the rest of the paper is very good. Some points:
Unstructured networks fail to find uncommon documents generally, but walkers are very bad at it. Consider a million node network where a document only exists on one node. With 64 walkers, state keeping and an average of 1 second for a hop, you are looking at over 4 hours search time.
Not to mention that the suggested `talk back' limit of once per 4 hops would generate 16 packets a second to the searcher; enough to take up a notable chunk of a modem's bandwidth.
The paper also suggests random replication without considering why. I would hazard a guess of the following:
Random walkers are going to tend to end up at the well connected nodes in a power-law network and hang around there. Thus path replication will hit the high order nodes which random replication (as they define it) will tend to hitmore low order nodes. I would suspect that measuring the number of copies in the network would show that random replication gives the highest number.
The paper also suggests that networks should be random. This is very nice but not all nodes are created equal and those on a T1 line can handle more messages per second than modem users. This bandwidth inequality (and the distribution of bandwidth) will force a power-law network to some extent.
Well documented, but I'm just pointing it out.
EuroPython presentations [via Lambda]
Note on Coding Theory [via Raph]
Greedy people scheme:
limit freedoms for profit.
Resistance prepares.
Orwell on the decay of the English language. Still relevant, if not more so.
MI5: "Civil liberties are a Communist front" (1951)
Starting reading through JWZ's rants last night while waiting for CSI to come on. About 2/3 of the way down now
Ryan Lacky has written a long post about TCPA to the cypherpunks list. Ryan:
This feels rather unfulfilling. I even avoided posting it to other lists so as to limit the spread of crack and confine said crack to its standard resting place.
EROS doesn't exactly build cleanly and I don't know if my fixes break anything critial yet, but they do get the code to compile at least
(the EROS install page is here)
--- Makefile Wed Jul 4 19:15:29 2001 +++ Makefile Sun Jun 30 18:36:31 2002 @@ -103,7 +103,8 @@ XENV_LIBXML2 = libxml2-2.3.13 XENV_LIBXSLT = libxslt-0.13.0 -XENV_TOOLS = xenv-make xenv-binutils xenv-gcc xenv-libxml2 xenv-libxslt +#XENV_TOOLS = xenv-make xenv-binutils xenv-gcc xenv-libxml2 xenv-libxslt +XENV_TOOLS = xenv-binutils xenv-gcc xenv-libxml2 xenv-libxslt xenv: $(XENV_TOOLS) @@ -120,12 +121,6 @@ -rm -rf build xenv-binutils: - -rm -rf build - -mkdir build - cat pieces/$(XENV_BINUTILS)/tgz-part.* | $(ZCAT) - | (cd build; tar xf -) - (cd build/$(XENV_BINUTILS); ./configure \ - --prefix=$(EROS_XENV) \ - --target=$(EROS_TARGET)-unknown-linux) $(MAKE) -C build/$(XENV_BINUTILS) all install @echo @echo "BUILD SUCCEEDED... removing build subdir"
--- bfd.h Sun Jun 30 19:16:33 2002
+++ bfd.h Sun Jun 30 19:16:21 2002
@@ -98,7 +98,7 @@
#define TRUE_FALSE_ALREADY_DEFINED
#endif /* MPW */
#ifndef TRUE_FALSE_ALREADY_DEFINED
-typedef enum bfd_boolean {false, true} boolean;
+typedef enum bfd_boolean {FALSE, TRUE} boolean;
#define BFD_TRUE_FALSE
#else
/* Use enum names that will appear nowhere else. */
e-lang has a great thread on TCPA/Palladium. See, for example, this message.
I would quite like to switch to using EROS but I'm too lazy really. I have a spare 20GB IDE drive bolted into my case which isn't even powered at the moment. Maybe I should download it and give it a go.
Final exam finished this afternoon. Hung up my blazer for the last time. This feels weird.
On the upside looks like I'm a tech reviewer for the new edition of UNIX Power Tools.
Gun crime up 49% and this is since we banned handguns. I'm sure ESR would have something pro-gun to say about this, I'm not going to.
Those CA results are rubbish past 8 bits. It was a 1 char typo in the code which I'm happy about because they make sense now. Maybe some more analysis soon
Been working on a secure version of GRUB which keeps checksums on a write protected floppy and won't boot a trojan kernel. It will also check any number of other files before booting (say /sbin/init and friends). Working pretty well
Ian Hill gets a submission on /.
Metis (this box) got a nice update over the last couple of days. Switched from apache to publicfile on the internal interface, bind to tinydns for the DNS server and installed gr
A NewScientist story linked to imperialviolet.org . Thanks Will. (Will Knight is one of only 2 clueful journalists I've ever talked to. The other being Andrew from theregister).
I'm troubled
Thinking about the CA results from yesterday I'm pretty convinced that I'm doing something silly. I verified the results for 3 bits by hand and it all came out right. But still, it makes a mess of my thinking and I want those results to be wrong.
I'll go over the code again at some point
Wes doesn't like the way Theo is handling the latest OpenSSH bug. What would you do different Wes?
If Theo says "here's the fix" it's then a rush between sysadmins and blackhats as to which get to a host first. sshd cannot be chrooted or run as non-root so all cracks are total and you're looking at a reinstall. Privsep isn't really ready for the prime-time but it does make people mostly immune without revealing the bug (thou with the added focus people won't be long in finding it independantly). It's a bad situation, but Theo is handling it well
SSH Remote Exploit. All upgrade to OpenSSH 3.3 (see this for Debian)
Maximum of 44 outputs for CA's upto 27 bits now confirmed. (which means that 25 and 27 have 44 outputs since the number cannot go down). I don't expect to finish processing 29. I don't think the answer would be a shock anyway.
IBM laptops get TCPA chips. Now the chip actually sounds a little useful:
If they weren't fuckware chips it could be a good thing. As it is...
Forgot to upload Gilmore's essay. Fixed.
Been playing about with Rule 30 CAs (see ANKOS page 53). You might want to read Sunday's post first (if you have not already).
As I said, Rule 30 CA seems to map very regular numbers onto odd sequence numbers. The lack of type A randomness in the input number and the presence of it in the output data, gives rise to seeming `added complexity'.
Today I'm taking a look at that mapping.
The input to the CA is the starting line expressed as a number. A black block is a 1, white is 0. The input size is always odd. The output is the first n bits down the centerline (not including the bit from the starting line). For example
That is a size 3 CA with starting value 6 (101 in binary). The centerline is down the middle (the white column) and the result is 1 (001 in binary).
Now we can cycle through all the possible input values for a given size and record the output. Rather than give a huge table I've taken a tally count of the outputs and plotted output value against count
Results from 5 bits
Results from 7 bits
Results from 9 bits
Results from 15 bits
We can certainly see that the mapping is not one way. In fact there are very few possible output values and all the inputs map to a small number of possible outputs. The maximum count in each case is 2^{n-3} + 2^{n-2} (where n is the size of the input in bits).
The number of different outputs is also noteworthy. It has a maximum of 44 upto 23 bits of input. (of course it gets quite time consuming to run tests with greater than 23 bits of input).
Size of input (x) vs number of different outputs
I don't have any conclusions at the moment, but at least I would be worried about using Rule 30 CAs as random number generators given that all the seeds seem to map to (at most) 44 sequences
Cool! I'm on Ryan's LJ friends list. Now if only he posted weird and wonderful stuff in friends only posts
A few more scans from the prom. (photos from weezel)
toad.com is dead so I've posted a local copy of John Gilmore's great essay "What's wrong with copy protection" since some people might need the anti-memes for this
Given the price of flights to LV I'm not going to make DefCon 10 this year. I might well go down to the UKUUG exhibition though. Anyone want to meet up? Mail me.
| / | Root |
| Alternate | The Weird and Wonderful |
| Backlinks | What are backlinks |
| John Gilmore | What's Wrong with Copy Protection |
| Archives | Blog Archives |
| One | Archive 1 |
| Two | Archive 2 |
| Three | Archive 3 |
| Four | Archive 4 |
| Five | Archive 5 |
| Six | Archive 6 |
| Seven | Archive 7 |
| Eight | Archive 8 |
| Nine | Archive 9 |
| Ten | Archive 10 |
| Eleven | Archive 11 |
| Twelve | Archive 12 |
| Thirteen | Archive 13 |
| Fourteen | Archive 14 |
| Fifteen | Archive 15 |
| Sixteen | Archive 16 |
| Seventeen | Archive 17 |
| Eighteen | Archive 18 |
| Nineteen | Archive 19 |
| Twenty | Archive 20 |
| Twenty One | Archive 21 |
| Twenty Two | Archive 22 |
| Twenty Three | Archive 23 |
| Twenty Four | Archive 24 |
| Twenty Five | Archive 25 |
| Twenty Six | Archive 26 |
| Twenty Seven | Archive 27 |
| Twenty Eight | Archive 28 |
| Twenty Nine | Archive 29 |
| Photos | Poor People Caught on Film |
| Jack and the Beanstalk | Jack and the Beanstalk |
| RIP Scan | Results of a Stage Scan Fire |
| Yosemite | Yosemite National Park |
| Projects | Incomplete things from the lab |
| Seagull's Bane | Linux Automounter |
| bttrackd | BitTorrent Tracker |
| CAPTCHA | CAPTCHA CGI script |
| Conserv | Console Serving |
| Deerpark | Using Tor with Firefox/1.1 (Deerpark) |
| DNSFix | Fixing DNS |
| Xovers | XTA Crossover Control |
| IAFS | Archive Org Storage |
| JBIG2 | JBIG2 Encoder |
| Verify | PGP Key Verifier |
| MaxFlow | Maximal Flow in Python |
| PyBloom | Bloom Filters in Python |
| pyGnuTLS | Python wrapping of GnuTLS |
| Sxmap | Apache SuEXEC Map |
| Hellard | Union Server Notes |
| Recordings | Free recordings |
| ICSM Choir | St Paul's Church |
| School | Ancient School Stuff |
| Writings | Who knows |
| Cap Systems | Capability Systems |
| Intro | Introduction to me |
| Suprema | JMC2 Group Project |
| MP Letters | Letters I've written to my MP |
| Sound | Sound With Dramsoc |
| SyncThreading | The wonders of user-land threads |