Cavalcade of Mammals

A compter security hack has appropriated the word nematode to describe “good” network worms. Dave Aitel says

“We don’t want people to think this is impossible. It’s entirely possible to create and use beneficial worms and it’s something businesses will be deploying in the future.”

So much for ever being able to Google “nematode” again! Spoiler! Bastard! And it’s so wrong. Aitel has a apparently heard of Caenorhabditis elegans, everyone’s favorite worm and entirely harmless, and jumped to the conclusion that every nematode is beneficial to humans. So wrong, so sad; a little knowledge is a dangerous thing. If this is the depth of his thinking I certainly wouldn’t take his advice on computer security.

I just read a review on cutting-edge new sequencing technology (Shendure et al., 2004). There are several approaches that were new to me. One that caught my imagination incorporates “polony technology,in which PCR is performed in situ in an acrylamide gel” for DNA amplification. A related technique using emulsion has been developed by the Volgelstein lab.

My idea of developing a method using reversibly terminating nucleotides has also occurred to many other people! Apparently finding a way to do the reversibly termination has been a roadblock. I certainly didn’t have a way to do it. They have worked out approaches to detection of incoration, the other half of the method, and also a part I didn’t develop.

Very interesting tech. According to the paper, even nanopore sequencing is close to working!

The paper talks a bit about using ULCS for personal genome sequencing (PGP, everthing gets an acronym), about the whys and what it will mean. It contain the usual throw away consideration of ethics and consequences. The papers says this “will require high levels of informed consent and security”. In practice, your personal seq info and related disease susceptibility info *will* get spread to interested parties. Just look at who calls the shots; after more than a decade of attention to genetic privacy and overwhelming public support, “no US federal laws that ban genetic discrimination for medical insurance or in the workplace”.

How I would love to seqeunce 1Gb a week!

From PZ Myers, unsourced:

“0.1% of all the species that have existed are currently extant, and the average lifetime of a species is roughly 10 million years”

Bio stats page

I’ve messed with the site’s css style sheets. If it looked OK before, there’s no change. If the layout used to suck, it should be fine now. Css is such a tar baby, I won’t tell you how long it took me.

And I went and downloaded 400+ new mammal pics, should be enough to keep the site in mammals all year!

Here’s an idea: wall moisture sensor. Water damage can be hard to spot early on and by the time you notice it the damage and rot may be extensive. So add a cheap moisture sensor.

The sensor is a 5cm spike with a nickel-sized head. To install it, you push it through the drywall. It expands a bit inside the wall and this helps fix it in place. Also, the back side of the sensor head is adhesive covered so it sticks to the plasterboard and seals the hole.

Now on to the working bits. It contains a cheap humidity sensor IC, a control chip, and a battery. On the head there is a tiny solar cell, a pair of contacts, and a red LED. The solar cell gets a bit of energy from room lighting and keeps the unit working for more than a decade. When humidity is detected, the LED flashes. Touch the contacts and the LED lights to show it works. The sensor and a control chip are embedded in the spike.

It should be cheap to make and long lasting. Home owners can buy a couple and pop them into walls they want to monitor–walls of finished basements, walls containing plumbing, exterior walls, etc. Installed, the sensor is unobtrusive.

If you think this could be useful and are interested in manufacturing/selling this send an email.

       [   ]-plasterboard
 R     [   ]
 o   ||[   ]
 o   ||========* -sensor spike
 m   ||[   ]
       [   ]  Inner wall
       [   ]

A high-end CPU these days uses nearly 100W of power. But it doesn’t have nearly the computing power of a human brain. AI is in good part, perhaps mainly, a software problem, but raw computing power seems lacking too. So how many of today’s CPUs would it take to build a computer with human intelligence? Say at least 10,000 Opterons.

This 10K CPU computer system would use 1MW of power. So how does that compare to a human brain? A person runs on 2000 Kcal / day.

At 86400 s / day this is 23 cal / s x 4.187 calories / J = 97 J / s. J / s are equal to Watts so this is 97 W.

Say 40% of the body’s energy is used by the brain. Then a person’s brain uses 40 W, as much a weak light bulb. Which is order-of-magnitude correct–your head is a little cooler than a weak bulb but the bulb is smaller.

So a human intelligent computer would use 1MW of power while a person’s brain uses 40W. The human brain is 25,000X more effcient than a computer. This says a few things about today’s computers. They are terribly inefficient, at least for the types of computations an AI needs to do. And today’s AI software design doesn’t capture the organization of biological computers. We have 1000 CPU systems today, and could build 10K CPU systems. But no system today is as clever as a mouse. Today’s AI may have crested the house fly brain goal post. But the lack is clearest in the hardware architecture.

Read an article by Confessions of a Community College Dean on Textbooks

Some students have been complaining about the price of textbooks for certain classes. Curious, I went to the bookstore and roamed the stacks, seeing just what they’re paying.

Wow.

The students are right. The costs are absurd.

Intro to [a popular foreign language] — $200.

He starts out getting it, then rambles on about how not much can be done by the College.

Come on now, why would an introductory language class *need* a new textbook? Language instruction changes very slowly. Why not pick the best from the last decade’s editions and use only used textbooks?

And your complaints about a system where the college buys the books seem poorly thought out. The *point* of such a system is to put the college in the crunch zone. And the college is better able to handle it–pressure publishers, reuse books, or pick inexpensive titles.

Here’s a picture from a story making the rounds because it turned up in a London fishmarket. It’s Cymothoa exigua, a fish parasite.

The 3.5cm creature had grabbed onto the fish’s tongue and slowly ate away at it until only a stub was left.

It then latched onto the stub and became the fish’s “replacement tongue”.

The bugs are usually found off the coast of California, so it’s possible the fish was imported to the UK.

Here are two pictures from the London story:

Here’s a pic from a web site about parasites:

Two pics from the Australian Museum:

Be glad you have hands to pull it off you if it bites your tongue. The poor fish is stuck–his fins can’t reach in and grab it!

I installed Expression Profiler Next Generation (EPNG) on my site. My interest is in canabalizing parts of it to show gene expression clusters. Turns out that part uses SVG which is a poor choice–SVG requires a browser plugin, a separate install, and isn’t even available for Mozilla/Firefox (the Adobe plugin for Netscape 4.x seems to work on the Mac, but not on the PC).

But on to EPNG… It’s a beast, a Frankenstein monster of code, composed of bits of a host of web technologies stitched together. It’s Perl based, which is what got me interested in it. And it has a database backend (Postgresql or Oracle, though I was able to port it to Mysql). And there’s some Javascript, everyone uses Javascript. But the developers were not content, they use gobs of Perl modules, including some that require libraries (PGPLOT), and a set of cirucular module requires (the stats modules)! R is used for the stats, that makes sense, and some R modules, and then SVG because the fewer users the better, and SVG/Javascript to take advantage of the SVG, and some Java, just to mix it up. And some AJAX is thrown in to pull in dataset histories.

So it’s a large and sprawling. I found it hard to install. Admittedly, I made it harder than expected by porting it to Mysql, but it was a pain even apart from that. Among the install joys are the release not working–the database schema doesn’t match the code! Boy that was a fun set of bugs to find.

And one of the modules required, XML::LibXML::SVG doesn’t exist! Google doesn’t know about it! It looks to be a private or temporary module, perhaps it got converted to SVG–most of the API and function is shared by the two. I converting the code to use SVG, but it was a huge bother!

Detailed EPNG install log.

Here are great photos of deep ocean animals (copyright
ExploreTheAbyss.Com).

Here’s a Deep-sea Pompeii Worm that lives in boiling hot hydrothermal vents:

And here’s a cephlopod. I’m guessing the part that looks like a giant brain is the gut/main body: