Re:Babies have an instinctive understanding of 're
on
The Baby Bootstrap?
·
· Score: 1
and psychologists have a bear of a time understanding volition, desire, and attention.
How do we decide what exactly to attend to in the visual scenes in front of us? (The marketing types want to know this so they can feed us more advertising, the psychology types want to know this so they can figure out how attention is parcelled out) Example, "looming" is when something is approaching rapidly and may strike the body or head: the CNS attends to this quickly if stereopsis is present and causes the body to move and the neck and shoulders and even arms to move in reaction. This appears to be a hardwired reflex. Fear of snakes also appears to cause reflexive autonomic changes and appears to be hardwired into the blueprint of generating the brain.
Ah, those are exactly the things I was commenting about above...
That's what the "neural network" paradigm was all about. You have an arbitrary and fixed number of input node, you have an arbitrary and fixed number of output nodes. You create linkages between these nodes and "weight" them with some multiplicative factor. In some particular instantiations, you limit all inputs to be [-1... +1] and limit all weights to be within the range [-1... +1].
So with A input nodes and B output nodes, you've got a network of AxB interconnections between these input and output layers. The brain analogy is that the A layer is the input layer or receptor layer, the Blayer is the output or motor layer, and it is the interconnections between these neurons, the neural network composed of the axons and dendrites connecting these virtual neurons that does the thinking.
Example: create network as above. Place completely random numbers meeting the criteria of the model (e.g. within the range -1 weight B's output feeds forward to C, etc., and these are called intermediate layers.
Rumelhart and Mcllelland encoded spellings as triplets of letters (26x26x26), had a few (or one, I can't remember this now) intermediate layers, and an output layer corresponding to phonemes to be said. They effectively encoded the temporal aspect of the processing into the triplets, sidestepping a (what I consider the more intersting...) part of the problem. They trained this neural network by feeding it the spelling of words and adjusting the weights of the networks until the outputs were the desired ones.
Note that nowhere in this process do they explicitly tell the system that certain spelling combinations lead to specific pronunciations. They only "trained" the system by telling it if it's right or wrong. The systems weights incorporated this knowledge in these "Hebbian" synapses and neurons.
So this is associative processing, using only feed-forward mechanisms. Feedback, loops, and temporal processing are even more interesting...
alas not enough room in this margin to keep going.
Sure, that was the engine of thought behind stories such as WarGames and 9x109 names of god. Somehow, unfettered access to data and time with "neural networking" capacity to form links and create linkages to pieces of data ("associative memory") would be all that was needed to create intelligence, and perhaps even sentience.
Minsky came up wrong on the single layer perceptron, AI was wrong on the purely feed-forward neural-network systems, Rumelhart and McLelland got some good promo off of their feed forward net that could learn to pronounce idiosyncracies, and Sejnowski got a great job at the salk from the AI delusions. But no, it appears to not have gone anywhere... thus far.
and a Pastrami on Rye says that you'll be required to buy the list from a vendor who has both a sweetheart deal with the state government and a cozy relationship with the governor or someone on his staff. For similar examples, look at BearingPoint scandals in Florida and all of the privatization fiascos in Florida and California. Google-twins power activate!
Oh I so sincerely agree. The hardest thing to do as a consultant is to tell the client to NOT do something, or NOT use a particular toy, library, package, etc... especially if they've brought you in as a consultant to bolster their own plans rather than actually investigate and give honest consultation.
Great point. There are also a lot of doctors (in smaller locales) who start to feel the same way: that they are unreplaceable, they have to keep working every hour, that no-one else can do their job or that no-one else is available to do that job.
This always leads to burnout. As a doctor, you've got to have call coverage and be prepared to be able to have downtime. If you're in a situation like this, it's a sure sign that they need more than one person. Don't be the patsy
yes... except for the fact that the poster actually states that the recording device was taken apart inexpertly. So I don't think it's possible to reinsert the drive and get an operational configuration.
yeah, but they can compile a lot of accurate data from the contents of your email, like telling Aunt Millie what your flight number is and when you'll be arriving in Toledo can tell them your true town of origin, and reminding her of your phone number gives them (of course) your TRUE phone number, with the area code and exchange identifying your location fairly well.
Any yahoos or googlies who email you (even if you personally do not have a googly or yahooie account just is another conduit of information to them. Plus with google and the infamous gmail invites, they've almost got a set of useful directed graphs showing linkages between users as the invites propagated.
Privacy is so paramount, but some idiot emailing you can blow it.
You know, it's just a minor "url"-ification, just like the Harvard M.B.A.s-not-to-be who changed the text of the pointer URL to see if they were going to receive acceptance letters or not. It's the server side that had the problem, accepting malformed or un-authorized requests.
That's as bad as passing your login and password as cleartext on the URL request, isn't it?
Hmmm... differential GPS just depends on having a local waypoint to give you a very accurate position, beyond the 1-3 meter resolution of GPS.
Now why not keep track of the accelerations, integrate (SUM) over time to get the velocity, integrate once again to get the spatial location. You could keep a log of where the laptop goes while it's on. Hmm... I might have to buy one of these toys, make the software and put it in the passenger seat of my car and see what I can make it do...
I remember a circuit cellar article about a 3-d accelerometer, but I didn't feel like dinking around with a soldering iron that year. Looks like a new powerbook will let me accomplish that long-delayed task with software alone.
It's also like the Housing Code that's often just a "insert law by reference" based on a copyrighted document that was not created by the gov't. Ask to see a copy of the construction code at your city or county seat and you're told to go buy a copy, the gov't cannot give you or let you see a copy of these laws.
but if they decide that your card number has been linked to devious activity, they could set up the system to flag anyone using that card...
then then next time you use that card, you'll be flagged and perhaps before you roll your grocery cart out the door, mr. and ms. police officers will be there to escort you to the police station to remove the rest of your civil liberties.
Frequent Upgrade Cycle Kicked : not necessary
on
Mac mini Dissection
·
· Score: 1
in regards to grandparent's The ability to change devices is the difference between a computer anyone can continue using for years, and a computer that becomes useless after 2 years because one minor component fails and there's no way to replace it.
and parent's statement Instead of upgrading your machine every two years, you can simply buy a new Mini every two years. End-result is more or less the same, as is the expense.
The bonus with macs is that they stay usable for longer periods of time. You may not need to upgrade every two years; you might want to or lust for the newest things though.
Let me say that for some simpler things, my Powerbook G3-233 was doing just fine 'til its power supply went up in smoke (literally, LITERALLY!) and is being replaced by an iMac-333 (green) which I picked up for less than 90 bucks U.S.ian . It's got enough horsepower to run old matlab scripts on system 8.6, though I may have to up it to 9.x for the stupid Brother laserprinter driver which does a stupid system check and refuses to run under 8.x systems. This is one point where linux wins out: running gimp-print with the HL-1250 works fine for the 1250, 1650, 1440, and the 5040 under linux.
Built-in 10/100 BASE-T Ethernet and a 56K v.92 fax modem give
Not to say anything against the mini. I actually want one and will probably get one myself soon. It's kind of like the drugs that first get FDA approval. Let's wait till the masses have these and see what kind of symptoms and side-effects sprout. I've usually gotten Macs as 2nd generation items (nisei?) not as first generation items. Except for that Newton..., couldn't wait on that or on the original 128k and the 512k Palm Pilot when the logo on it was still US Robotics. m....
I agree with your rant about the idiocy of paying to be at the movie theaters and being subjected to commercials and adverts while stuck there. Don't forget to complain to the manager of the movie theater. It may not do much, but registering your feedback may help to quell the invasion of advertisements into the world: loos, the floorspace at supermarkets, the little LCD screens at the petrol-station pumps. It's getting ridiculous.
And along that topic, a friend of mine has a brother who works for Nielsen-VNU (Soundscan, Nielsen TV ratings, Nielsen demographics, etc) as a statistician who had an interesting comment to make when I made my rant about those adverts at movie theaters: Nielsen actually sends people out to the movie theaters to gauge the response of the audience to these commercials. Yikes! I wish we could influence this madness, but the world (USA corporatocracy) wants to view us as only consumers, stuck in our little media trough at the movie theater being force-fed the adverts that their real customers (the advertisers) want us to see.
Just like the idiotic medical system that wants to call us customers or clients rather than as patients... but I digress... wait this is/.... that is de rigeuer. ici, n'est-ce pas?
Good point. A couple of apple laptops have obscure PRAM resetting key combos, and some laptops require disconnecting the AC adaptor, removing the battery and THEN hitting the secret hidden reset button.
As an aside, I have a friend with an HP laptop with a sleep to disk mode which copies RAM contents to a partition on HD and shuts off in "sleep mode". When it is restarted, it recopies the data from the partition back into RAM and goes along on its merry way. This works fine for DOS and windows, but with linux you need to be in console mode when you go to sleep this way: X seems to screw it up. And don't even this of this sleep-to-HD function with KNOPPIX, even from a terminal console (I tried...), it won't work. It'll go to sleep but when it wakes up and tries to read the CD, it keeps doing "cloop compression errors...".
and Mach also named "Mach Bands", a psychophysical illusion when gradients of light and dark are next to each other: the edge of the light region appears even brighter and the edge of the dark region appears even darker than it truly is. A series of graduated intensity bands will appear to be non-linearly illuminated because of the contrast-enhancement performed by ours retinas and cortices... (retinae?)
What about saccades? When the eye moves rapidly over a long angular direction (which it does in tracking objects or changing your view) or a short angular direction (a.k.a. microsaccades, which happen multiple times a second), you get blurring which is normally suppressed by the visual attention system.
When you do saccades across long persistence displays like LCDs, you will not see any major aberration as the light source effectively stays on. When you saccade across medium to short persistence displays (P21 phosphors for short, your regular TV or CRT for medium), it is possible to notice that there is either a shearing or tearing artifact.
TV/CRT displays are scanned left-to-right at (say for 640x480 VGA at 80 Hz) 480*80=38400 times per second and scanned slow...ly up-to-down 80 times per second followed by that quick scan back up. Well you can try this at home (TV's at ~60 Hz show this a little more easily than most of our CRTs which are set at a less-likely-to-appear to flicker refresh of >80Hz):
look at an object to the left of the TV screen. Then rapidly switch what you're looking at to the right side of the TV screen. The image of the TV will no longer look rectangular but like a shortened-horizontally and sheared (top to the leftish, bottom to the rightish) parallelogram. If you do a right-to-left saccade, the image will appear longer horizontally and top to the rightish of the bottom.
Now the interesting thing happens with up-to-down saccades: if you go up-to-down at slower than or close to the same angular velocity as the scan line (depends on how close you're sitting to the screen) goes down the screen, the projected image will appear SHORTER-UP-TO-DOWN and if you actually match the scan-line's downward angular velocity, the TV image will seem to just be a poorly set up XF86 display of one pixel in height.
If you have an effectively ZERO-PERSISTENCE direct write display, since the laser is being used to draw directly on the retina (or to project on a screen) rather than an electron-train hitting chemicals causing them to phosphoresce with a certain limited time before they stop glowing (PERSISTENCE...), then fixation has to be maintained or the illusion of motion based on the projection's position is destroyed. Laser projection systems try do multiple lines scanned at once or other fancy projection scan patterns rather than the usual cathode-ray-gun approach, but the saccade problem continues to be an issue.
The saccade errors are the big to-do with projective laser displays for visible wavelengths, regardless of whether they are projected onto a screen or direct write onto the retina.
The other problem is... bah, it's enough already.
Re:So, can you hook up a Mac via a serial connecti
on
Netatalk 2.0.0 Released
·
· Score: 3, Interesting
Nice.
I've actually got an old 700 (quadra) and some SE-30s that want to talk to my serial laserwriter, but the laserwriter's fuser hardware has gotten melty and gummed up. Do you know of a good way to emulate an apple laserwriter on a serial port on a linux box and hook up the apple's RS-422 to the serial port and make it think that the linux-box is a post-script level one printer?
I've tried simple things like making a linux serial console and running GS on it but the mac's couldn't make it through. And my old computer with two serial ports on it which I could use to peek at serial protocols is stuck in storage (mold problems... water leakage everywhere... >$30k hardware damaged as attic-collapsed-from-water-weight... ) so I can't probe it to make a serial emulator. Is there a quick and dirty way to do this for older macs that don't have ethernet availability?
True, true. But the right problems can be efficiently parallel-parcelled out if you have the right hardware.
Remember the Connection Machine with 2^10 processors, with a hypercube topology allowing each node to communicate with 10 other nodes? True: 1980's architecture; true: tougher to scale the hardware up; true: they had parallelizable SCHEME implemented on it; true: they had the cute LEDs o' blinken lighten on one side panel. So I agree with you that software can hobble the best multi-processing system, but I think you can scale higher even in hardware given the right network topologies and communication paths.
The path-traversal algorithms for interprocessor communications was fairly cool and tolerant of partial link breakdowns. Each message passed had its ultimate destination address as the prefix. The receiving processor only had to bit flip one or more bits on it to choose its current path and send the message along the way.
If there were no hyper-cube-link outages, then a message would have to traverse at most 10 (ten) edges to get from any node to any other node. If there were any missing edges in the system, the network could route around it, because there would be up to 2^10 paths to get from any node A to any other node B, even more if you allowed paths of greater than 10 steps (such as hitting a network-edge outage as soon as it occurred.)
Hmm..., Nicholson Baker (Vox, and a few other books too) could have written the whole BOOK on just the cereal eating. Italo Calvino's book would have covered the guy getting ready to pull the cereal box out and all of his internal mental states leading to that moment would have been fully explored. (If On a Winter's Night, A Traveller...) (Se un Notte d'Iverno, un Viagiatorre... i think was the italian title).
Slashdot Mirror
and psychologists have a bear of a time understanding volition, desire, and attention.
How do we decide what exactly to attend to in the visual scenes in front of us? (The marketing types want to know this so they can feed us more advertising, the psychology types want to know this so they can figure out how attention is parcelled out) Example, "looming" is when something is approaching rapidly and may strike the body or head: the CNS attends to this quickly if stereopsis is present and causes the body to move and the neck and shoulders and even arms to move in reaction. This appears to be a hardwired reflex. Fear of snakes also appears to cause reflexive autonomic changes and appears to be hardwired into the blueprint of generating the brain.
Ah, if only we knew a few more answers...
Ah, those are exactly the things I was commenting about above...
... +1].
That's what the "neural network" paradigm was all about. You have an arbitrary and fixed number of input node, you have an arbitrary and fixed number of output nodes. You create linkages between these nodes and "weight" them with some multiplicative factor. In some particular instantiations, you limit all inputs to be [-1... +1] and limit all weights to be within the range [-1
So with A input nodes and B output nodes, you've got a network of AxB interconnections between these input and output layers. The brain analogy is that the A layer is the input layer or receptor layer, the Blayer is the output or motor layer, and it is the interconnections between these neurons, the neural network composed of the axons and dendrites connecting these virtual neurons that does the thinking.
Example: create network as above. Place completely random numbers meeting the criteria of the model (e.g. within the range -1 weight B's output feeds forward to C, etc., and these are called intermediate layers.
Rumelhart and Mcllelland encoded spellings as triplets of letters (26x26x26), had a few (or one, I can't remember this now) intermediate layers, and an output layer corresponding to phonemes to be said. They effectively encoded the temporal aspect of the processing into the triplets, sidestepping a (what I consider the more intersting...) part of the problem. They trained this neural network by feeding it the spelling of words and adjusting the weights of the networks until the outputs were the desired ones.
Note that nowhere in this process do they explicitly tell the system that certain spelling combinations lead to specific pronunciations. They only "trained" the system by telling it if it's right or wrong. The systems weights incorporated this knowledge in these "Hebbian" synapses and neurons.
So this is associative processing, using only feed-forward mechanisms. Feedback, loops, and temporal processing are even more interesting...
alas not enough room in this margin to keep going.
Sure, that was the engine of thought behind stories such as WarGames and 9x109 names of god. Somehow, unfettered access to data and time with "neural networking" capacity to form links and create linkages to pieces of data ("associative memory") would be all that was needed to create intelligence, and perhaps even sentience.
...
Minsky came up wrong on the single layer perceptron, AI was wrong on the purely feed-forward neural-network systems, Rumelhart and McLelland got some good promo off of their feed forward net that could learn to pronounce idiosyncracies, and Sejnowski got a great job at the salk from the AI delusions. But no, it appears to not have gone anywhere... thus far.
Later comment will be positive.
and a Pastrami on Rye says that you'll be required to buy the list from a vendor who has both a sweetheart deal with the state government and a cozy relationship with the governor or someone on his staff. For similar examples, look at BearingPoint scandals in Florida and all of the privatization fiascos in Florida and California. Google-twins power activate!
Oh I so sincerely agree. The hardest thing to do as a consultant is to tell the client to NOT do something, or NOT use a particular toy, library, package, etc... especially if they've brought you in as a consultant to bolster their own plans rather than actually investigate and give honest consultation.
Great point. There are also a lot of doctors (in smaller locales) who start to feel the same way: that they are unreplaceable, they have to keep working every hour, that no-one else can do their job or that no-one else is available to do that job.
This always leads to burnout. As a doctor, you've got to have call coverage and be prepared to be able to have downtime. If you're in a situation like this, it's a sure sign that they need more than one person. Don't be the patsy
yes... except for the fact that the poster actually states that the recording device was taken apart inexpertly. So I don't think it's possible to reinsert the drive and get an operational configuration.
yeah, but they can compile a lot of accurate data from the contents of your email, like telling Aunt Millie what your flight number is and when you'll be arriving in Toledo can tell them your true town of origin, and reminding her of your phone number gives them (of course) your TRUE phone number, with the area code and exchange identifying your location fairly well.
Any yahoos or googlies who email you (even if you personally do not have a googly or yahooie account just is another conduit of information to them. Plus with google and the infamous gmail invites, they've almost got a set of useful directed graphs showing linkages between users as the invites propagated.
Privacy is so paramount, but some idiot emailing you can blow it.
You know, it's just a minor "url"-ification, just like the Harvard M.B.A.s-not-to-be who changed the text of the pointer URL to see if they were going to receive acceptance letters or not. It's the server side that had the problem, accepting malformed or un-authorized requests.
That's as bad as passing your login and password as cleartext on the URL request, isn't it?
k say chalet
hey, that sounds like the circuit cellar article I mentioned above...
Bravo!
I read what you write.
I like what you write. I agree with your point in its entirety.
Hmmm... differential GPS just depends on having a local waypoint to give you a very accurate position, beyond the 1-3 meter resolution of GPS.
:)
Now why not keep track of the accelerations, integrate (SUM) over time to get the velocity, integrate once again to get the spatial location. You could keep a log of where the laptop goes while it's on. Hmm... I might have to buy one of these toys, make the software and put it in the passenger seat of my car and see what I can make it do...
I remember a circuit cellar article about a 3-d accelerometer, but I didn't feel like dinking around with a soldering iron that year. Looks like a new powerbook will let me accomplish that long-delayed task with software alone.
Must acquire cash for purchase NOW
It's also like the Housing Code that's often just a "insert law by reference" based on a copyrighted document that was not created by the gov't. Ask to see a copy of the construction code at your city or county seat and you're told to go buy a copy, the gov't cannot give you or let you see a copy of these laws.
Egads! They haven't updated their copyright date to MV yet have they?
Right,
but if they decide that your card number has been linked to devious activity, they could set up the system to flag anyone using that card...
then then next time you use that card,
you'll be flagged and perhaps before you roll your grocery cart out the door, mr. and ms. police officers will be there to escort you to the police station to remove the rest of your civil liberties.
in regards to grandparent's
The ability to change devices is the difference between a computer anyone can continue using for years, and a computer that becomes useless after 2 years because one minor component fails and there's no way to replace it.
and parent's statement
Instead of upgrading your machine every two years, you can simply buy a new Mini every two years. End-result is more or less the same, as is the expense.
The bonus with macs is that they stay usable for longer periods of time. You may not need to upgrade every two years; you might want to or lust for the newest things though.
Let me say that for some simpler things, my Powerbook G3-233 was doing just fine 'til its power supply went up in smoke (literally, LITERALLY!) and is being replaced by an iMac-333 (green) which I picked up for less than 90 bucks U.S.ian . It's got enough horsepower to run old matlab scripts on system 8.6, though I may have to up it to 9.x for the stupid Brother laserprinter driver which does a stupid system check and refuses to run under 8.x systems. This is one point where linux wins out: running gimp-print with the HL-1250 works fine for the 1250, 1650, 1440, and the 5040 under linux.
re 1 PCI slot is occupied by an ethernet card, un upgraded in 5 years. Gigabit is built into the mac mini
:
Actually NO.
The Mac mini has 10/100 according to Apple's own web site
Built-in 10/100 BASE-T Ethernet and a 56K v.92 fax modem give
Not to say anything against the mini. I actually want one and will probably get one myself soon. It's kind of like the drugs that first get FDA approval. Let's wait till the masses have these and see what kind of symptoms and side-effects sprout. I've usually gotten Macs as 2nd generation items (nisei?) not as first generation items. Except for that Newton..., couldn't wait on that or on the original 128k and the 512k Palm Pilot when the logo on it was still US Robotics. m....
Hey Michael,
/. ... that is de rigeuer. ici, n'est-ce pas?
I agree with your rant about the idiocy of paying to be at the movie theaters and being subjected to commercials and adverts while stuck there. Don't forget to complain to the manager of the movie theater. It may not do much, but registering your feedback may help to quell the invasion of advertisements into the world: loos, the floorspace at supermarkets, the little LCD screens at the petrol-station pumps. It's getting ridiculous.
And along that topic, a friend of mine has a brother who works for Nielsen-VNU (Soundscan, Nielsen TV ratings, Nielsen demographics, etc) as a statistician who had an interesting comment to make when I made my rant about those adverts at movie theaters: Nielsen actually sends people out to the movie theaters to gauge the response of the audience to these commercials. Yikes! I wish we could influence this madness, but the world (USA corporatocracy) wants to view us as only consumers, stuck in our little media trough at the movie theater being force-fed the adverts that their real customers (the advertisers) want us to see.
Just like the idiotic medical system that wants to call us customers or clients rather than as patients... but I digress... wait this is
Good point. A couple of apple laptops have obscure PRAM resetting key combos, and some laptops require disconnecting the AC adaptor, removing the battery and THEN hitting the secret hidden reset button.
As an aside, I have a friend with an HP laptop with a sleep to disk mode which copies RAM contents to a partition on HD and shuts off in "sleep mode". When it is restarted, it recopies the data from the partition back into RAM and goes along on its merry way. This works fine for DOS and windows, but with linux you need to be in console mode when you go to sleep this way: X seems to screw it up. And don't even this of this sleep-to-HD function with KNOPPIX, even from a terminal console (I tried...), it won't work. It'll go to sleep but when it wakes up and tries to read the CD, it keeps doing "cloop compression errors...".
C'est la vie.
and Mach also named "Mach Bands", a psychophysical illusion when gradients of light and dark are next to each other: the edge of the light region appears even brighter and the edge of the dark region appears even darker than it truly is. A series of graduated intensity bands will appear to be non-linearly illuminated because of the contrast-enhancement performed by ours retinas and cortices... (retinae?)
exactly, or if you saccade along with the scan line direction and effectively immobilize the spot on your retina. Similar effect.
Plus that so-called infinite resolution is limited by the fact that your fovea has at most 30-arc-second packing of the L and M sensitive cones...
Okay, just a few things about this and some problems.
... bah, it's enough already.
Microvision is the company doing this.
What about saccades? When the eye moves rapidly over a long angular direction (which it does in tracking objects or changing your view) or a short angular direction (a.k.a. microsaccades, which happen multiple times a second), you get blurring which is normally suppressed by the visual attention system.
When you do saccades across long persistence displays like LCDs, you will not see any major aberration as the light source effectively stays on. When you saccade across medium to short persistence displays (P21 phosphors for short, your regular TV or CRT for medium), it is possible to notice that there is either a shearing or tearing artifact.
TV/CRT displays are scanned left-to-right at (say for 640x480 VGA at 80 Hz) 480*80=38400 times per second and scanned slow...ly up-to-down 80 times per second followed by that quick scan back up. Well you can try this at home (TV's at ~60 Hz show this a little more easily than most of our CRTs which are set at a less-likely-to-appear to flicker refresh of >80Hz):
look at an object to the left of the TV screen. Then rapidly switch what you're looking at to the right side of the TV screen. The image of the TV will no longer look rectangular but like a shortened-horizontally and sheared (top to the leftish, bottom to the rightish) parallelogram. If you do a right-to-left saccade, the image will appear longer horizontally and top to the rightish of the bottom.
Now the interesting thing happens with up-to-down saccades: if you go up-to-down at slower than or close to the same angular velocity as the scan line (depends on how close you're sitting to the screen) goes down the screen, the projected image will appear SHORTER-UP-TO-DOWN and if you actually match the scan-line's downward angular velocity, the TV image will seem to just be a poorly set up XF86 display of one pixel in height.
If you have an effectively ZERO-PERSISTENCE direct write display, since the laser is being used to draw directly on the retina (or to project on a screen) rather than an electron-train hitting chemicals causing them to phosphoresce with a certain limited time before they stop glowing (PERSISTENCE...), then fixation has to be maintained or the illusion of motion based on the projection's position is destroyed. Laser projection systems try do multiple lines scanned at once or other fancy projection scan patterns rather than the usual cathode-ray-gun approach, but the saccade problem continues to be an issue.
The saccade errors are the big to-do with projective laser displays for visible wavelengths, regardless of whether they are projected onto a screen or direct write onto the retina.
The other problem is
Nice.
I've actually got an old 700 (quadra) and some SE-30s that want to talk to my serial laserwriter, but the laserwriter's fuser hardware has gotten melty and gummed up. Do you know of a good way to emulate an apple laserwriter on a serial port on a linux box and hook up the apple's RS-422 to the serial port and make it think that the linux-box is a post-script level one printer?
I've tried simple things like making a linux serial console and running GS on it but the mac's couldn't make it through. And my old computer with two serial ports on it which I could use to peek at serial protocols is stuck in storage (mold problems... water leakage everywhere... >$30k hardware damaged as attic-collapsed-from-water-weight... ) so I can't probe it to make a serial emulator. Is there a quick and dirty way to do this for older macs that don't have ethernet availability?
True, true. But the right problems can be efficiently parallel-parcelled out if you have the right hardware.
Remember the Connection Machine with 2^10 processors, with a hypercube topology allowing each node to communicate with 10 other nodes? True: 1980's architecture; true: tougher to scale the hardware up; true: they had parallelizable SCHEME implemented on it; true: they had the cute LEDs o' blinken lighten on one side panel. So I agree with you that software can hobble the best multi-processing system, but I think you can scale higher even in hardware given the right network topologies and communication paths.
The path-traversal algorithms for interprocessor communications was fairly cool and tolerant of partial link breakdowns. Each message passed had its ultimate destination address as the prefix. The receiving processor only had to bit flip one or more bits on it to choose its current path and send the message along the way.
If there were no hyper-cube-link outages, then a message would have to traverse at most 10 (ten) edges to get from any node to any other node. If there were any missing edges in the system, the network could route around it, because there would be up to 2^10 paths to get from any node A to any other node B, even more if you allowed paths of greater than 10 steps (such as hitting a network-edge outage as soon as it occurred.)
Hmm..., Nicholson Baker (Vox, and a few other books too) could have written the whole BOOK on just the cereal eating. Italo Calvino's book would have covered the guy getting ready to pull the cereal box out and all of his internal mental states leading to that moment would have been fully explored. (If On a Winter's Night, A Traveller...) (Se un Notte d'Iverno, un Viagiatorre... i think was the italian title).