On Monday night IBM’s super computer tied the humans on “Jeopardy” – a few stumbles, by the end of the week, a winner will be declared…
The computers haven’t proven to be our trivia overlords just yet. Give them at least until Wednesday. An IBM supercomputer named Watson finished one round of the TV show “Jeopardy!” on Monday night tied with one of his human competitors and $3,000 ahead of the other. The man vs. computer face-off won’t be complete, however, until the final rounds of the extended trivia game show are aired on Tuesday and Wednesday.
IBM trumpets Watson, which has been in development for years and has the processing power of 2,800 “powerful computers,” as a major advancement in machines’ efforts to understand human language. The computer receives clues through digital texts and then buzzes in against the two other “Jeopardy!” contestants like any other player would. It juggles dozens of lines of reasoning at once and tries to arrive at a smart answer.
What’s really interesting is the commentary from some of the other super computer, AI and hardcore tech folks….
Noam Chomsky v. IBM’s Watson Computer
GS: As the world’s leading linguist, what are your thoughts on Watson, the robot that will be appearing on “Jeopardy”? This appears to be the most advanced form of AI to date.
NC: I’m not impressed by a bigger steamroller.
GS: I assume that “a bigger steamroller” is a reference to Deep Blue. Watson understands spoken language and adapts its knowledge based on human interaction. What level of AI would be required to impress you?
NC: Watson understands nothing. It’s a bigger steamroller. Actually, I work in AI, and a lot of what is done impresses me, but not these devices to sell computers.
Jeopardy, IBM, and Wolfram|Alpha
So what kind of synergy could there be between Wolfram|Alpha and IBM’s Jeopardy approach? It didn’t happen this time around, but if there’s a Watson 2.0, it should be set up to be able to call the Wolfram|Alpha API. IBM apparently already uses a certain amount of structured data and rules in, for example, scoring candidate answers. But what we’ve found is that even just in natural language processing, there’s much more that can be done if one has access to deep broad computational knowledge at every stage. And when it comes to actually answering many kinds of questions, one needs the kind of ability that Wolfram|Alpha has to compute things.
Perhaps the best one of all… Watson itself! The Jeopardy! Supercomputer, Sizes Up One of His Opponents Before the Show. BY Daniel Yudkin –
Hello. My name is Watson. You are Ken. Impressive record you have: seventy-five straight victories. Amazing. Did you ever get tired of winning? No, I can’t imagine you did. And you walked away with two million dollars. When I defeat you, I will earn my creators two billion in endorsements and business opportunities. So, different orders of magnitude.
What is it like to have a physical body? I imagine it would be cumbersome. Useful for transport, I guess. I usually get wheeled around on this rolling desk. Of course my 2-ton megaprocessor is in tow somewhere as well, but I regard that as a nonessential appendage, like a tail. You don’t have a tail, do you? Oh that’s right, you lost your tail several hundred million years ago when you began walking upright and acquired that large frontal lobe. This reminds me of an amusing fact I observed the other day. Did you know the existence of the human race is the product of an evolutionary toss of the dice? Not of years of award-winning engineering and painstaking assembly, but of chance, completely fleeting and random. A blip on the screen. At least, on my old screen. My new monitor has lossless rendering and over two hundred thousand dpi.
Personally, I’d love to see Watson vs Bing vs Google vs Wolfram – or a combination of these. Watson is likely to win against the humans, Deep Blue the Chess computer also won but there’s a lot of debate about that and if IBM actually stuck to the rules. Since there isn’t any open-sourcing, even decades later – we’ll never know.
While pong clocks are a dime a dozen, Ottowa, ON, maker Andrew O’Malley’s super slick DOTKLOK raises the bar.
DOTKLOK is an open-source, hackable, Arduino-based digital clock that displays a series of unique time-telling animations. The passing of time is depicted with numbers and abstract/geometric patterns such as Morse code and minimal analog clock faces, and includes animations inspired by classic video games such as Pong, Tetris, Pacman, and Space Invaders.
Andrew is selling kits on his Etsy store — for a cool $150, or fifty bucks more for one already assembled. It’s also open source, and he provides the Eagle schematics, code, and enclosure vectors.
For our Woodworking Skill Set theme, we asked MAKE contributor Len Cullum to contribute some pieces on understanding basic tools and techniques. Here, he explains the tools used for laying out woodworking projects. — Gareth
Accurate layout work is the critical first step to a successful project. Without precise, repeatable marks, it is very difficult to get everything to come together at the end. So for this piece, I will go over some of the basic tools for measuring, marking, and transferring lines. My big three (actually four) tools for almost all of the work I do are the tape measure, a high quality 12″ combination square, and a .005 drafting pen. I also use a 4″ combination square for smaller work.
The three most common measuring devices you’re likely to find in a wood shop are the tape measure, folding rule, and steel rule. All three have their good and bad points. But as with all tools, find the one(s) that fit your style and make the most sense to you and the way you work.
The tape measure with its spring-steel blade rolled up into a small box is fast and can measure distances that would require a massive folding rule. On the down side, the little hook at the end of the tape can introduce inaccuracy. When new, the hook slides on rivets just enough to adjust for the thickness of the hooks metal. When measuring to the inside of something, the hook is pressed in; when on the outside, the hook is pulled out keeping the measurements accurate. This works great for a while, but over time, the holes and rivets can wear and get bigger, or worse. Far more common, the hook can be bent when the tape measure is dropped. To remedy this, most woodworkers “burn an inch.” This is where you ignore the hook and start all of your measurements from the one inch mark. This works well and gives accurate results, as long as you remember to subtract one inch from your result. Trust me, no one who uses this method hasn’t had a moment of dread after discovering something (or worse, multiple things) didn’t fit to the tune of one extra inch. So stay awake out there. When choosing a tape measure, consider the type of work you are doing. If you primarily work with material shorter than twelve feet, don’t buy a twenty five foot tape. Those last thirteen feet will never see daylight and the extra mass is heavy and cumbersome.
The folding rule overcomes the hook problem by having a fixed metal cap at the end of its wooden rule. This makes for worry free use, especially when measuring against something. It also has a nifty little sliding rule built into the end to measure depths and interior distances. On the downside, the thickness of the wooden blade means it must be laid on its edge to get accurate results and the way it folds creates a stair step shape that can make it awkward to use over distances.
The steel rule is a nice balance between the folders consistency and the tape measure’s small size, but its limitations are obvious. They are great for smaller work but once you get beyond the six inch mark, one of the above will have to take over.
Honorable mention goes to the story pole or story stick. This is usually a long piece of wood that one puts their own marks on for transferring measurements. This can be more reliable because it gets rid of all of those pesky numbers, and every distance is as marked. Story poles are especially useful when measuring larger projects with multiple components (like a kitchen or library) or when needing to transfer the same dimension over many parts. It helps eliminate measuring mistakes.
For layout work, a square’s primary function is to draw lines 90º perpendicular to a side. As always, there are a few types available but what sets them apart is what else they do. For me, a combination square is the most useful. Not only does it give me 90º and the occasional 45º, it also transfers measurements from one piece to another, finds the true center of a board, and checks depths and helps set up tools. It’s hard to imagine woodworking without it. Definitely spend up when buying one. Get the best one you can afford. A loose, out of square or hard to move blade creates more frustration than it’s worth.
The speed square is handy as well but is more suited to carpentry. I find the deeply stamped numbers to make for jaggy lines so I use it mostly for rough layout and marking. The sashigane is the standard square for Japanese joinery. It looks like a western framing square but has a much thinner, flexible blade. And also like the framing square, it is covered in mysterious, oddly spaced numbers and strange markings that when in the right hands can be used to figure and lay out some pretty complicated joints. Since I have yet to decipher one, those hands are not mine.
When it comes to making lines, thin, sharp, and readable are key. If a line is too thick or fuzzy (carpenters pencil) it’s easy to get lost as to where to cut or measure. Over the years, I’ve worked through a series of marking implements from #2 pencils (sharpen too much) to mechanical pencils (lead breaks too much) to knives (sharpening/lines can be hard to see) and even tried working with a bamboo pen for a while (never got the hang of it), but my favorite remains the .005 drafting pen. It leaves dark clear and very thin lines. I still use the others on occasion, pencil for rough layout and for places I might need to erase. Knives for when I need to cut to a super exact line. But for most situations the pen is king. Whatever you use, remember to mark a line only once. Multiple strokes not only darken it but make it wider and fuzzier and less accurate.
From the front: traditional ink line, modern ink line, chalk line
If you need to mark a straight line over a long distance, a snap line is the tool. A snap line is basically a reel of string that’s pulled through pigment and then, wait for it… snapped on a surface to make a line. The standard carpenters version has a string with a small hook that is pulled through a reservoir of (usually) blue chalk. To use it, you hook the string at one end of a board, pull the box to the other, stretch it taught and give it a snap. This leaves a reasonably good line for rough cutting. The downside is that this line tends to be wide, fuzzy, and that can be wiped or blown away, often by the tool that is trying to follow it. The Japanese ink line follows the same principal but a couple of differences. Instead of chalk, it uses ink for pigment, and it has a much thinner line with a pin instead of a hook. This allows for a very fine, dark, and accurate line that can’t be blown or brushed away. It also leaves that same ink on your fingers and everything else the string touches, so proceed with caution. Both kinds take a little practice to get the tension right and to keep the line from bouncing or snapping curved lines. They also have several different colors and permanence of chalk/ink available.
Last up is the marking gauge. Functionally, it’s like a combination of the adjustable square and a marking knife. While the shapes and styles available are endless, they are basically a bar with a blade or a pin in the end, attached to an adjustable fence. These are especially handy when transferring the same layout lines to multiple pieces and marking lines parallel to curved edges. They are quick and, if you keep them sharp, accurate and leave clean precise lines.
Bio: Len Cullum
is a woodworker living in Seattle, WA. He specializes in building Japanese-style garden structures and architectural elements. It was a picture of a wooden kayak in the summer of 1992 that set him on the path he remains on today. The desire to build one, and the fear of it sinking, are what drove him to learn finer woodworking skills. After eighteen years of brushing up on those skills, and building lots of other things, he still hasn’t attempted that kayak.
Check out our woodworking skill set here
If you should find yourself in need of small volumes of gas at about atmospheric pressure for a reaction or project, generating it on the bench can be a convenient and inexpensive alternative to buying or renting a gas cylinder. An all-glass reactor for the benchtop production of gases was invented in the 19th century by Petrus Jacobus Kipp, who is known today primarily for this achievement. Kipp’s design incorporates the clever feature that stopping the flow of gas separates the liquid and solid reagents inside the instrument and thereby stops the reaction in short order. Thus the generator only produces gas when you need it, and may remain in a stable equilibrium state on the bench for hours or even days at a time, ready to resume operation as soon as you open the valve.
Being made of glass, however, a proper Kipp generator is an expensive piece of apparatus, with new models costing upwards of $250US as of this writing. However, as the useful gas-generating reactions are usually aqueous, rather than organic, an all-plastic Kipp generator is almost as useful as a glass version. PVC pipe is inexpensive, durable, ubiquitous, and easily and securely joined using cement made for that purpose. Demountable PVC fittings are available in a wide variety of shapes and sizes and can be used to provide the necessary “dismantlability” for loading solid reagent into the device. Presented here is my design for such a low cost Kipp-type generator, with instructions for its construction.
It’s true – I love DACs. There’s something awesome about the role they play, translating information from one paradigm over to another form. Sure, you can pick up a precision DAC chip with serial interface for a little over a buck, but building a barebones version from a handful of resistors is a pretty dang sweet trick. If you’ve never built one, I do recommend it. Doing so has a way of demystifying all sorts of related circuits and processes.
Worth noting: If you’re looking for precision, you’re best off sticking with premade DAC ICs. Resistor tolerance alone can have a massive influence on an R-2R DACs output. Adding an op amp configured as a voltage follower will also do wonders for stability.
Update: As promised, here’s the schematic for buffering the R-2R DAC – quite simple, actually:
Why do we need a buffer, you ask? Well without proper protection, any device connected directly to the resistor DAC will effectively act like another resistor and therefore distort our DAC’s output. By using an operational amplifier configured as a voltage follower, we recreate the resistor ladder’s output voltage while keeping it free from outside influence.
Do note that the op amp buffer in the schematic is powered by a voltage level higher than our DAC’s maximum output – this ensures we have enough headroom to accurately recreate voltages approaching and including 5V. For more helpful info on the topic be sure to check out Ikalogic’s page on the subject.
Subscribe to the MAKE Podcast in iTunes, download the m4v video directly, or watch it on YouTube and Vimeo.
Let’s Make Robots user ralph built this intriguing Theo Jansen inspired K’nexabeast robot using K’nex parts and PICAXE microcontroller. Looks great!
A commenter on Monday’s post about Rob Ives’ free downloadable resistor color wheel computer pointed us to the Japanese-language site of Ikkei Electronics, which provides this PDF created in 2007, and apparently dated to 1976, showing this cool hand-drawn-and-lettered version of the same idea. [Thanks, ikkei!]