Older blog entries for mojotoad (starting at number 19)

Hiatus maximus.

Went on vacation, recently, mountain biking in Colorado. Fun was had by all, and Susan certainly expanded her biking envelope.

My friends Than and Chris recently embarked on their globe circling year-long tour. I'm green with envy...I've been wanting to do something similar in spirit for around 10 years now, and I'm pretty sure that it was a conversation on the very topic that was the catalyst for Chris to strike up the deal with Than. I was invited, but always felt my own year-long wanderlust was best stomped alone, so I declined. They will have a great time, I'm certain.

As a small concession, I decided I'd at least start in the opposite direction. Perhaps we can say hello along the way.

I noticed some interesting commentary in Paul's diary regarding genetic engineering. Most of his concerns seem to be centered around the interaction of genetically enhanced species with the "natural" world. This is certainly a valid concern, and experimental tweakers should always keep that in mind.

However, I believe evolution is not, by definition, perfect. Evolution has blind spots. Perhaps divergent paths is a better description. Certain paths, once embarked upon by evoloution within a particular species branch, will likely not switch to a new branch after a certain degree of complexity has evolved. A great example of this is the retina. Mammals have blind spots over the optic nerve since the nerve bundle distributes on the inner surface of the retina. Squid, on the other hand, have no blind spot since their optic nerve splays on the outer surface of the orb, achieving full coverage over that portion of the retina.

Obviously, mammalian brains are great at compensating for their blind spots, but it is clear that there is a better solution that, at least in this case, has been demonstrated by another branch of evolution. The odds of any mammal leaping to that branch of occular evolution via mutation are slim to non-existant, because a) complexity has defined the bounds of the problem, and b) compensation in other areas (the mammalian visual cortex) have largely negated any evolutionary advantage such a radical mutation would produce.

Of course I realize that 'b' offers endless speculation on whether nature ever produces anything "truly superior" between isolated branches of evolution. This is of course dependent on context. Using this same example, nobody would feel the disadvantage during the day. Night, however, is an entirely different story -- probably everyone here is familiar with the superiority of peripheral vision when staring into the darkness.

This is of course expounded upon in great and wondrous detail in Godel, Escher, and Bach, where Douglas Hofstadter quite dramatically illustrates the phenomenon using natural number theory, courtesty of Godel's brilliant work.

In the Godel nutshell, your derivable truths are limited by your axioms. In a sufficiently complex truth system, there are provable truths that are underivable...by definition. And that, of course, assumes you've defined the right axioms to properly approach the problem.

Anyway, fascinating topic of conversation. My personal take is that genetic engineering should be approached with caution and wisdom. There are potentially enormous benefits to be gained. I believe the radical ones will be through this "branch hopping" type of modification; a modification that properly lives somewhere between natural evolution and contact with alien flora.

The expanded argument could benefit from the nanotech analogy: somebody will figure it out, so it would be best to figure it out first if for no other reason than to be able to properly understand the problem and defend yourself.

Jabba the Hutt needs a home!

It's sad. We spent a lot of time building a full sized Jabba the Hutt for halloween 98, and he has been languishing in my friend's garage ever since. He did make a brief appearance for halloween 99. But my friends are selling the house, and we don't have anywhere to put him. I had been lobbying to take him to Burning Man for a send-off in a blaze of glory, but that trip never materialized.

If you are curious, you can check out his chops. And oh, if you live near Houston and would like to adopt a fat alien, by all means let me know.

We might be able to fob him off onto the Houston Art Car community for a fate more noble than the land fill, but we'll see.

Florida was a blast, very relaxing and beautiful. I ate many many baked oysters, yum.

Making progress on the new HTML::TableExtract, it's about ready for release.

Hot Damn!

Got my kernels kickin'
Got my Zooms zoomin'
Got my masqs rading'
Got my hops in hand
Sittin' by the pool,
Posting on Guacamole
With a laptop lapping.

Of course, the mosquitoes are out and it's past midnight. This was just a range test. Zoom wireless cards rock.

Time to sync up on the CVS tree and start cranking out my next module releases. I've been neglecting them lately, mostly due to external distractions but partly due to getting the laptop kicking. I'm ready to start cranking, now, though.

I suppose I can be one of those people in the coffee houses now, face aglow from the laptop betwixt sips of joe. We'll see. Mostly I'm just glad for any change of scenery from my cluttered bedroom, where I tend to hack from a frankenstein desktop constructed from the hastily disassembled components of my mp3 box. The micro ATX form factor looks like a postage stamp in the case. Anyway, it's nice to be able to wander about with the laptop.

Off to the Redneck Riviera next week for vacation. Laguna Beach, Florida, which is between Destin and Panama City Beach. Beautiful beaches, awesome seafood (oysters...mmmm...).

Can't decide whether to take projects along with me. I should probably go analog for a while.

The car (87 Volvo 240 GL, what a trooper) is in the shop for a tune-up, so I'll be biking around town tomorrow. Unfortunately I don't think my satchel is big enough for the laptop in addition to a change of clothes, shower paraphernalia, etc. To stink or to hack, that is the question. I suppose I will have mercy on my coworkers.

I haven't done squat for coding lately.

I did get my laptop in, however. It's a Sony VAIO PCG-F420, and is quite a nice little toy.

Unfortunately, even though it has a fully capable parallel port, I can't seem to get a PLIP connection working on it...my mothership computer has an internal modem, the laptop has a useless winmodem...so I was hoping to use PLIP for bootstrapping debian off of my main computer. All the interupts, etc., yadda yadda are set correctly, but no dice. Finally I said to hell with it, and ordered a couple of ZoomAir wireless cards. They're more fun, anyway.

Also got my new monitor in, today. XF86Setup is being a bitch, though, because the vga16 driver does not display properly on the new monitor (driven by a voodoo3), and that's the only way it allows you to run...grrr. I need to RTFM on XFree86. I wish debian used xconfigurator. I guess I'll find out whether it does anything that debian dislikes during my reading.

metaguacamole: Regarding the mozilla image thing, I've been tracking it on the various sites (mostly kuro5hin, but it also looks like the shit finally hit the /.fan).

Anyway, it's sort of encouraging to see the course of the conversation here on guacamole. The alarmist rants hardly even registered before a discussion of the ethical issues involved in the world of producer and consumer emerged; in particular, how privacy issues survive in the crucible between the sometimes conflicting goals of the two. (give credit where credit is due, eivind probably first kicked that can with a position he must have known would be contrary to most readers; I suppose it's possible that it was a bit of a troll, but he at least makes some intelligent efforts to back up his assertions with things other than ad hoc button stomping).

It's why I keep coming back, anyway.

As for diaries...various distractions lately...this have been distracting from coding as well. sigh. These distractions have not been without merit, however; drum lessons are proceeding well. I'm pretty sure I detected faithful execution of a rhumba/bolero beat emanating from my flailings today. It could be a monkey/typewriter thing, though.

There is something distinctly nice about a coding frenzy and a flagon of mead, so long as you keep an eye on the metric of clear thought.

I ran into some unexpectedly interesting problems whilst wrangling with the general issues surrounding context-free HTML table extraction. I need to crack open the CS books and figure out what the real name is for something I loosely refer to as "delayed conditional nodal inheritance". Hrmm.

I am a bit bummed that the Espresso PC has a less than stellar built-in sound card, leastwise according to the arstechnica review. I was hoping to use one of these puppies to reimplement the core of my home grown mp3 extravaganza box since I had to cannibalize the prior motherboard for my central computing needs recently. Ah well.

Mmmph. <slurp>. Mmm. Triple latte good. Mmmm. <slurp>.

It's storming outside. Sounds very nice; I have the windows cracked open so I will no doubt sleep quite soundly.

I've been retooling HTML::TableExtract in a major way. I've fixed header extractions to account for the nastiness you get from colspan and rowspan effects, so that the columns you extract are the columns you would expect when looking at the table visually. (such is the fair of those who deal with sparse trees representing grids). In more exciting realms, I've been implementing search chains which allow you to yank tables relative to other tables using lists of checkpoints, in terms headers, depths, counts, or some arbitrary chain thereof. Time permitting, the new release should be set loose within a couple of days after I've tested it to my satisfaction.

Viva data mining, HTML context free. Other than being in a table somewhere on a page, of course.

Ugh. Can't sleep. I think I need some meditation techniques to turn off code mode at will. As it stands I just lie there with my eyes closed, brain on overdrive.

The crawfish boil went well. As promised, there were whole heads of garlic tossed into the pot; I felt sorry for anyone who walked up late and had to walk that gauntlet of breath.

Saw Frequency over the weekend. As long as you resolve not to dwell on time paradox, it's a highly enjoyable movie. (they subscribe to the "malleable present" type of theory, where you can alter your present by diddling with your past...and as the cause of the alteration, you retain "dual memories" of your past -- the original, and the new version, which at least avoids the doppleganger problem that movies like Back to the Future suffer from, where some poor bastard with the memories of the new past gets displaced into lah-lah land). Anyway. I could go on at great length about the three major Hollywould time paradox approaches. In the case of this movie, just assume that there will be paradox and enjoy the roller coaster.

I guess I'm just going to stay up all night. I'll perk back up when my normal waking hours roll around, then crash and burn sometime after lunch.

I've been skulking around at The Perl Monks site lately, after having seen several references spring up around here. Lo and behold, I bumped into chromatic when discussing a question I posed.

(if you are curious, the question has to do with scalar evaluation of a perl hash when it is tied...it always returns 0, so "if %hash" does not mean the same thing for a tied hash as opposed to a native hash, assuming it has entries. You can get around it by saying "if keys %hash", but I'm a fan of consistency)

Mudbugs tomorrow! I can't wait. I hear tell Botz is planning on adding mushrooms and whole heads of garlic into the batches, along with the regular new potatoes and corn. The gills on mushrooms really soak up those spices. Mmmm! I am glad I no longer wear contacts; it's tempting to wipe away the tears.

Read Souls In the Great Machine recently. Pretty fantastic science fiction romp, if you ask me.

10 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!