This entry will be a hodgepodge, or as they say in some circles, a “hotch potch.” (I think it’s a Britishism; Colin Wilson used that spelling many times.) Stuff has been piling up in the Contra file. Carol and I have been slighting housework for these past six months, she laid up after surgery on both feet, and me writing what has doubtless been the most difficult half-a-book I’ve ever written. We’ve been cleaning up, putting away, and generally getting back to real life. Real life never tasted so delicious.
One reason is rum horchata. I’m not one for hard liquor, mostly, and generally drink wine. (Beer tastes far too bitter to me.) But Rumchata got me in a second. It’s a dessert cordial no stronger than wine, with the result that you can actually taste the other ingredients, like vanilla, cream, and cinnamon. Highly recommended.
People ask me periodically what I’ve been reading. After soaking my behind in computer science for the past six or eight months, I’ve been studiously avoiding technology books. That said, I do endorse Degunking Windows 7 by my former co-author Joli Ballew. I actually used it to learn some of the Win7 details that weren’t obvious from beating my head on the OS. I wish it were a Coriolis book, but alas, it’s not. That doesn’t mean it’s not terrific.
True to my random inborn curiosity about everything except sports and opera, I’ve developed an interest in the chalk figures of southern England. The next time we get over there (soon, I hope, though probably not until summer 2015) we’re going to catch the Long Man of Wilmington, the White Horse of Uffington, and that very well hung (40 feet!) Cerne Giant. Other chalk figures exist, many of them horses. Some can be seen from Google Earth. A reasonable and cheap intro is Lost Gods of Albion by Paul Newman. The book’s been remaindered, and you can get a new hardcover for $3. I wouldn’t pay full price for it, but it was worth the hour and change it took to read. My primary complaint? It needs more pictures of chalk figures, duhh.
Quick aside: While researching kite aerial photography with my found-in-the-bushes GoPro Hero2 sports camera, I came upon an impressive video of the White Horse of Westbury taken from a double bow kite (rokkaku). I have the cam, and loads of kites. All I need now is a chalk figure. (I suspect I could coerce my nieces into drawing one for me.)
Far more interesting than Lost Gods of Albion was Gogmagog by Thomas Lethbridge. I lucked into a copy of the 1957 hardcover fairly cheap, but availability is spotty and you may have to do some sniffing around. If you’re willing to believe him, Lethbridge did an interesting thing back in the 1950s: He took a 19th century report that a chalk giant existed on a hillside in Wandlebury (near Cambridge) and went looking for it. His technique was dogged but straightforward: For months on end, he wandered around the hillside with a half-inch metal bar ground to a point, shoving it into the ground and recording how far it went in before it struck hard chalk. His reasoning was that the outlines of a chalk figure would be dug into the chalk, and thus farther down than undisturbed chalk. In time he had literally tens of thousands of data points, and used them to assemble a startling image of two gods, a goddess, a chariot, and a peculiar horse of the same sort as the Uffington White Horse.
Not everybody was convinced. Even though Lethbridge was a trained archaeologist, his critics claimed that he was a victim of pareidolia, and simply seeing the patterns he wanted to see in his thousands of hillside holes. The real problem was that Lethbridge was a pendulum dowser, and a vocal one: He published several books on the subject, which make a lot of claims that aren’t easily corroborated. Lethbridge claims that most people can dowse, and hey, it’s an experiment that I could make, if I decided it was worth the time. (It probably isn’t.)
The third book in my recent readings is The Physical Phenomena of Mysticism by Herbert Thurston, a Jesuit priest who spent a good part of his life collecting reports of peculiarly Catholic weirdnesses (stigmata, levitation, inedia, odor of sanctity, etc.) and presenting them in a manner similar to that of Charles Fort, if better written. Most of the articles were originally published in obscure theology journals, but were collected in 1952 in a volume that I’ve never seen for less than $100. Last year it was finally reprinted by White Crow Books and can be had for $18. I’m not sure what one can say about reports of people who have not eaten for forty years. Mysticism is a weird business, but physics is physics. The book is entertaining, and it’s given me some ideas for stories, particularly since I have a spiritually butt-kicking psychic little old Polish lady as a major chartacter in Old Catholics. (Vampires are just so 2007.)
If three books doesn’t seem like much, consider my habit of going back to books I’ve read and liked, and flipping through them to see what notations I’ve made in the margins. We all make them; when was the last time you deliberately went back to read and reconsider them? I’ve been dipping into Gary Taubes’ Good Calories, Bad Calories, Steven Pinker’s The Better Angels of Our Nature, Colin Wilson’s A Criminal History of Mankind, and Matt Ridley’s The Rational Optimist, and arguing with my own marginal notes. One can learn things arguing with oneself, and I’ve been known to change my mind based on things I scribbled in other people’s book’s ten or twelve years ago. (Before that I was too young to have anything like informed opinions.)
For example, I’ve gone back to calling it “global warming.” Climate is always changing, and the assumption that we know all the forces propelling those changes is just wrong–and in tribalist hands, willfully dishonest. Carbon dioxide has exactly one climate trick in its bag: It warms the atmosphere. That’s it. If the discussion is about carbon dioxide, it’s about global warming. Why climate changes is still so poorly understood (and so polluted by political hatred) that we may be decades before we even know what the major forcings are. In the meantime, I want predictions. If your model gives you climate data out fifty years, it will give you data out five. Publish those predictions. And if they prove wrong, be one of those people who really do #*%^*ing love science and admit it. Being wrong is how science works. Being political is how science dies.
I have a long-delayed electronics project back on the bench: Lee Hart’s CDP1802 Membership Card. I started it last summer, and set it aside when the Raspberry Pi gig turned up. It’s basically a COSMAC Elf in an Altoids tin. I had an Elf almost forty years ago. I programmed it in binary because that’s all there was in 1976. And y’know? I can still do it: F8 FF A2. F8 47 A5…
Some things really are eternal.
Based on the time frame you described I’m assuming you have a rev F membership card. I built the very first rev F board back in May 2013. Apparently I ordered right after whoever was the last person to get a rev E. My board works perfectly, although when I initially turned it on pin 1 on U8 was shorted to pins 4 and 5 on P2. Those pins are in line with each other when the boards are stacked (at least that’s the situation with rev F, I don’t know about rev G). With the membership card it’s fairly important to make sure the pins are cut as short as possible.
What I have been investigating lately is how to efficiently make the 1802 do basic floating point trig functions like sine and cosine. I’m assuming that the way to go about it is with CORDIC (http://en.wikipedia.org/wiki/CORDIC). Floating point trig code for the 1802 may already exist, but if it does I haven’t seen it. Anyone who has seen such code for the 1802 or knows of quality CORDIC references, I would be interested.
I recall back in the early 1980s, I knew a guy in California who was developing an emulator for various CPUs. His dev system was an Osborne portable, with its tiny 5″ screen, and his programming was all done in hex. Not only was he doing Z80 code in hex, but using it to develop emulators for other processors, such as the Zilog Super8 which was then my favorite. Eternal, it may be, but at that level, coding in hex is, IMHO, eternally tedious.
Alex, you should look at Jack Crenshaw’s Math Toolkit for Real-time Programming.
Bill, thanks for the book recommendation. I just checked and the UCF library has the e-book version of ‘Math Toolkit for Real-time Programming’, so I’ll be taking a look at that right away. Note that after a quick scan through the index I’m pretty sure there is no mention of CORDIC anywhere in the book (although I expect the book will be very useful anyway). That appears to be the trouble with CORDIC, very little information about it seems to have been published. Hopefully before too long someone will prove me wrong about that.
There’s an interesting pic at the top of Lee Hart’s COSMAC page (http://www.sunrise-ev.com/MembershipCard.htm). Who knows what the alien-like “elf” is about, but the robot and his master look suspiciously familiar.
By the way, Jeff, your Photo Gallery page has been messed up recently.
Yup. That’s me. Lee and I go way back. The original drawing was in pen and ink, by Chris Cloutier.
My photo page is basically dead and has been for awhile now. I’m looking for something to host photos (not on my hosting service; in this case, the cloud is fine) that gives me more or less the same functionality as Gallery. I’d like comments with moderation and the ability to divide the photos into galleries by topic; people who want to see photos of my robot Cosmo Klein may not want to see pictures of QBit. I had intended to work on this last summer, but then the Raspberry Pi gig came up, and then the zombification of XP, and the upstairs is getting new carpet on Friday. Been busy. As always, suggestions welcome.
I put together a Flickr account and uploaded a bunch of old photos of COSMAC stuff and my COSMAC-stuffed robot, Cosmo Klein. See:
https://www.flickr.com/photos/121958351@N02/sets/72157643274453323/
Thanks Jeff, I appreciate those photos. I also built a COSMAC-controlled robot back in the day. Would be interested in details of how Cosmo Klein worked (if and when you have a chance). I recall using CMOS bus buffers to drive 7475 latches for output ports, which drove 7406 buffers (paralleled for larger current, IIRC), which in turn drove relays operating the motors. For some motors, a low-current reed relay drove the larger motor relay. Not exactly elegant!
And neither was the code. What a weird instruction set! But it was only my second machine language at the time (8080 was first), so I didn’t know enough to know how odd it was. Discovering the 6800 family later was a revelation. From my present vantage point, I would say the 1802 instruction set was closer to the microcode of a microprogrammed CPU architecture than to a “normal” one. It worked OK for small apps where you could load all your pointers and call addresses into the registers and then leave them there, but was pretty darn cumbersome and inefficient for larger programs. Good memories, though, from those early adventures. I was in high school at the time.
I’ll check out your BYTE article to see how you managed such a programming feat. I’m not sure if it’s good or bad that I remember the instruction set well enough to attempt to understand what you did!
I did a lot with the 1802 back in the day. I had an article in Byte called “The COSMAC Doodler” which was a paint program in 256 *bytes* of code. (Byte, May 1980.) I designed and built something I called The COSMAC IMP (Inexpensive Matrix Printer) but by that time (late 1981) the IBM PC was getting all the press and nobody was doing long-form COSMAC articles anymore.
I have photos of a lot of that ancient stuff, and once I find a photo hosting site I’ll create a COSMAC gallery.
You spent the time to squeeze a drawing program into 195 bytes, then you didn’t bother to include antialiasing or layers or morphing? You still had 61 free bytes in the page! I guess they don’t make homebrew computing pioneers like they used to… 😉
(Cosmo Klein at your service! – yes, I (think I) still have every issue, along with DDJ, PCT/VD, and others…)
As for Climate and #$^@ing loving science. I found this history of Climate Science infiltration of non-scientists by Richard Lindzen interesting. http://arxiv.org/ftp/arxiv/papers/0809/0809.3762.pdf
Great to see your photos coming back online. They’ve been missed. My favorite is still “the Head of R&D”.
Also glad to hear you found time to tinker with the Membership Card. It’s up to rev.G now, which adds a serial port, and a second memory chip so you can have 64k of memory and both RAM and ROM at once. If you haven’t started assembly yet, let me know and I can send you the rev.G boards.
Cosmo Klein is rather well documented in “Captain Cosmo’s Whizbang”; a great little booklet you wrote on the 1802 way back in the late 1970’s. Its cover illustration of you and Cosmo (by Chris Cloutier) was the inspiration for the cover of the Membership Card manual. 🙂
I too built a robot with the 1802. It was basically a classic Logo turtle robot, but with the computer and Logo fully self-contained in the robot itself, all housed in a small Bud minibox. I called it “Itsabox” (It’s a box turtle robot; get it?).
You brought Itsabox to our house in Rochester in 1982 or 1983, and I remember running it around the picnic table on our back patio. Great gadget. Basically, PE’s Emily with some brains. My IC gradeschool classmates still remember me as “the robot guy” after 48 years, thanks to Emily. I coached my nephew Brian while he made a Meccano beambot in the Emily style in 1999. Still have both, both still work, though I need to re-mount the silicon solar cell in my Emily.
On the subject of Britannic archaelogy:
Alan Garner’s superb fantasies The Weirdstone of Brisingamen (1960) and The Moon of Gomrath (1963) reference a lot of what is just barely hidden in the British country landscape. Garner made use of The Old Straight Track (1925), an early exploration of “ley lines”.