You don't have to use the keyboard exclusively. It just has to be one of the options available to you at any given moment.
Secondly, there are a lot of "keyboards" out there that don't have any tactile feedback yet sell well enough: think of touchscreen ticket machines, for example (ooh er, perhaps there aren't any in the US since there's precious little public transport).
Also, consider decoupling the projector mechanism from the detector: the projected keyboard is simply a guide for the operator. The detector is a robust, solid-state item that can be made extremely rugged. The "keyboard" could be etched on steel for vandal-proof phones with internet access keyboards, for example. Or keypads in extreme temperature/chemical environments. They don't have to be one of NEC's never-see-the-light-of-day toys.
Finally, the sensor doesn't have to be configured as a keyboard at all. It could be a mousepad area, where you don't need a mouse: any object would do.
Odd that the NEC article would show links to a Canesta solution, with a news release way back in 2002. I hesitate to post the links because both sites will get slashdotted instantly, and the latter site has a silly activex, but this shows a VKB solution for a PDA. VKB itself can be seen here.
I've seen and used VKB technology. It's very compact and an elegant solution. Unlike Canesta, it doesn't need a stereoscopic view of the target area, which means it really could fit into a pen sitting in a charger or other holder.
A couple of years ago I was driven around Shenzhen in one of Chairman Mao's limos called, I think, a Red Flag. It looked like the world's heaviest Yugo, with some severely arm-wrenching doors. The glass was rendered bullet-proof by, as far as I could tell, using unfeasibly thick plates of the stuff.
The car had been converted to battery power using a vast number of Li-ion cells in the trunk and, indeed, everywhere else. The trunk had a bank of fans that had to be on when recharging. Someone had added some retro mag-alloy wheels.
The interior was blue rather than red. There was a front bench seat, with the (now redundant) gear shift peeping up through a slot in the cushion. Changing gear must have been a very Freudian experience.
We drove around, gradually getting slower and slower, yet no matter how slow we went, no-one would pass us.
Re:DVD's are obsolete
on
DVD-Rs go 8x
·
· Score: 1
CDs/DVDs are outdated,... I just got a Kingston 128mb USB key chain (1.1 nontheless),
If you've got 38 spare keychains, there's this movie you should see. I should be able to load them up in less than 12 hours.
On the other hand, it would herald a welcome return to traditional QA techniques. Wheeltappers from the rail industry could be retrained to walk around chips, tapping suspect components with their little hammers.
If Qube hacking is simply a quest to get a small but useful computer into a pretty case, then Via announced its Eden-N processor last month, dissipating 7W at 1GHz and 4W at 533MHz. Samples shipping now, so I assume a Nano-ATX board will be available soon.
The 866BASE gets a P3, 2 ethernet ports, and the usual interfaces on a 91mm x 96mm board.
Plenty of opportunities for packing a nice computer into a small case.
After upgrading all the HDs on a small network using software tools (it took a long time and a couple of NT systems wouldn't migrate), I bought this, a hardware unit. It works with a computer or stand-alone. Either plug its USB2 cable into a Windows PC, or plug the donor and recipient drives into each side, in which case it's OS-agnostic, of course. It costs about $150 which is way cheaper than other products I've seen.
Using direct disk-to-disk cloning, it does 120GB in about 20 minutes. Full range of copy/cloning products here.
When not being used to upgrade drives, I use it as a simple disaster recovery backup: plug into an idle computer and dump the drive onto any cheap drive I have laying around.
The cloning software seems to be Easy Ghost in ROM, but I've not seen any problems after some 15 cloned drives.
Its benefits for me have been speed and reliability. The downside is that it doesn't seem to be available in the US but there are surely similar solutions.
All 10 national BBC radio stations are streamed on the internet, plus the World Service, plus BBC local radio. See here for the stations. BBC7 streams old comedy, plays and books exclusively. Some of the material on all stations remains available for up to 7 days, depending on copyright.
If you're listening live, you can choose RealPlayer or Media Player. I think audio on demand is limited to RealPlayer.
With Total Recorder, you can capture the stream (and any other digital or analog audio stream in your computer) and save the resulting file(s) in any format you want. It has powerful automatic file naming, editing and scheduling tools, so since BBC7 repeats its output twice or more in a 24-hour cycle, I capture one stream to disk automatically divided into 15-minute chunks at 48kbps/22500kHz stereo, then schedule on-demand audio for other BBC programmes I want. The software gets the URL, starts RealAudio, does the capturing and shuts everything down nicely afterwards. It's an outstanding product.
Highlights: 8MB flash memory (4MB for data), 512kB RAM, runs at 24MHz or 48MHz, touchscreen uses Decuma handwriting recognition (Decuma is a Swedish company with Sony VC money; Decuma is also used in Sony Clies & cellphones). FM/Midi sound, vibrator; PIM, mail client supports POP/SMTP.
Can communicate with SD form-factor PHS card, Wi-Fi card, Bluetooth etc. Tri-color LED, sound and vibration alerts for incoming data.
I figure it's a much more useful device in Japan where connectivity is ubiquitous, than the US. But it will have a tough time competing with the likes of the J-SH53 and its successors.
The REX family were designed and manufactured by Citizen. Known in Japan as the DataSlim, there were 7 models released between 1997 and 2001. Later versions could input data. It was great being able to synch in the PC card slot.
The first OEM customer was Sterling Plastics (i.e. Rolodex). See here for the Japanese ones.
So Citizen certainly knows how to make this type of product.
Re:Field of view crop & other digital oddities
on
Digital 35mm SLRs?
·
· Score: 1
Putting it another way, your expensive 200mm zoom becomes a heavy 125mm lens on the digital SLR
And, indeed, putting it the wrong way. Quick slap with the cluestick there. Of course, it's a wonderful 320mm zoom. Mea culpa. Told you you shouldn't come to Slashdot for your answers.
Field of view crop & other digital oddities
on
Digital 35mm SLRs?
·
· Score: 1
As others have said, there are surely sites better than Slashdot from which to obtain good information. But I won't let that stop me.
Caveat: I've not used the Canon, I'm basing my opinions on reviews only, so E&OE.
If you're new to SLR photography, this Canon is a nice place to start if you use it with the EF-S lens. If you're adding this digital SLR to your existing Canon kit, then you may find it difficult to cope with the 1.6x field of view crop: the sensor is smaller than a 35mm frame, so you need a 22mm lens on the digital SLR to get the same field of view as a 35mm lens on a 35mm SLR. Putting it another way, your expensive 200mm zoom becomes a heavy 125mm lens on the digital SLR
So, in that respect, if you have an existing lens collection that you're comfortable with, I'd go for a digital body with a full-frame 35mm sensor.
You'll notice that your worries will change from "do I have enough film?" to "do I have enough batteries?"
Regarding "how many pixels are enough?", people seem to think the sweet spot is between 8MP and 16MP, but with digital photos you don't get grain (but you do get noise), so I personally think 5MP is a decent starting point.
Penultimately, the default sharpness in Sony, Minolta and other digital cameras seems to be a tad too sharp: I routinely dial back the in-camera sharpening algorithm and do any sharpening outside the camera.
Finally, if you just like the idea of lots of pixels and quite a wide zoom range, you may want to at least hold the Sony F828, which is an 8MP camera due in November or December. I've played with one and, as a 707 owner, I like it a lot: like many others with this series of cameras, I tend to shoot using the camera back display, holding the camera like a TLR.
As the digital index builds up, we will rapidly come across the situation where the electronic book is searchable, but the printed form is out of print. If this service ultimately allows single copies to be printed for delivery, it will be an outstanding demonstration of print-on-demand technology as advocated by the Print On Demand Initiative and others.
I'd love to be able to browse a giant back catalog, knowing that an original or facsimile copy could definitely be delivered to me.
Since these buildings tend to have active or passive dampers to counteract oscillation, perhaps one measure of how "usefully" high a building is could be the height of the mass damper above ground level.
The Taipei 101 has a pendulum-style damper, I think. I wonder how it's controlled if its motion is detected as contributing to oscillation rather than dampening it?
Interestingly, the spire has its own TMDs. I guess I'm a TMD kind of guy: there's something suitably retro about dragging 800 tons of stuff up as high as possible, simply to have it wobble or slide around.
I was never taught anything in school about China's exploration
Well that's not the fault of the Chinese, I think.
In the early 1400's Cheng Ho (Zheng He) made 7 voyages towards the western world, reaching Africa. He took more than 27,000 people in 320+ ships with him on some voyages. The largest ships weighed 1500 tons, were 180 meters long, and held 1000 people. See here to see how vastly larger a treasure ship was than the Santa Maria, built 87 years later.
An ideograph tends to be a complete word in itself. The only thing you can do with an ideograph is add or remove strokes. With many thousands of ideographs represented by between 1 and about 16 strokes with the majority being 9 strokes or less, messing with strokes changes the meaning of the ideograph itself. The ideographs have to be learnt by rote, since they contain no phonetic information as an aid to pronunciation. This is in distinct contrast to Romance and similar languages, where understanding the meaning of an unknown word typically involves attempting to pronounce it.
As an indication of the difference between, say, English and Japanese, Japanese handwriting recognition tends to be based on detecting the order in which the strokes for a character are laid down, which relies on everyone learning the same way. Recognition of Roman letters tends to be based on stroke and pool analysis (i.e. zero to two or so strokes plus any enclosed spaces).
My personal experience of speed reading suggests that the envelope shape of the word is very important. I figure the scrambled letters of the sample texts could be replaced by similar letters having the same "mass", as it were, and the text would still be reasonably comprehensible.
You could try a set of programmable foot switches, such as Kinesis. You need to program the actions from a Windows machine, but the unit can then be used with other OSes. I used this together with a small programmable keypad a few years back.
We had a similar system when I was in the UK. Core time was 10am to 4pm. Outside that, Flexitime was 7:56 to 18:Iforget. If we stated we were making full use of flexitime, then the rules were that we could be no more than 8 hours up or 8 hours down at the end of the month. If we were 8 hours up, we had to take a day off rather than claim 8 hours overtime. If we were 8 hours down, we could lose full flexitime rights.
We were expected to consult with our team and amend our hours to be reasonably coherent: an all-early-arriving team would expect new members to arrive early, for example.
Since it was my first work experience, I thought it was normal. I've since learnt that it was an outstanding system for all concerned: the company calculated its employees were working an extra 12 minutes per day for free on average, and we thought we were being treated as humans.
That "ergo" sure does contain a lot of assumptions.
Yup 8-). In an ideal world, more stable dyes are developed over time. More stable dyes mean you can write to them faster, so one indicator of this stability in the market is the existence of faster media. This inherent stability suggests a longer media life. So, given the choice between storing data on an older 1x CDR media or a newer 32x CDR media, I would choose the latter.
Also worth considering is that CDR dyes are sensitive to radiation at 780 nm. That's heat. So we have to protect a heat-sensitive disc from, well, heat.
DVD-R dyes are sensitive to radiation at 635-650 nm. That's red light. So we have to protect the discs from light, which is somewhat easier.
Newer products are sensitive to blue light, so are even easier to protect from light.
CDR is a single polycarbonate substrate with a varnish top coating to protect it. The label side, therefore, is the weak link and CDRs can die simply because the top coating ages. DVD-x is two polycarbonate substrates bonded together. So one side is as strong as the other. Polycarbonate technology is very mature. So DVDs should be inherently more robust than CDR.
Finally, many disk writers deliberately operate at higher power than specified for CDR and DVD-R. So the media may be receiving more energy during the write phase than it's supposed to. So a newly-burnt disk may come out the burner effectively a significant way through its supposed storage life.
But, all things being equal, I stick by my assertion that it's all down to the dye, and because all things aren't equal, it's why I archive to phase-change media!
Slashdot Mirror
Secondly, there are a lot of "keyboards" out there that don't have any tactile feedback yet sell well enough: think of touchscreen ticket machines, for example (ooh er, perhaps there aren't any in the US since there's precious little public transport).
Also, consider decoupling the projector mechanism from the detector: the projected keyboard is simply a guide for the operator. The detector is a robust, solid-state item that can be made extremely rugged. The "keyboard" could be etched on steel for vandal-proof phones with internet access keyboards, for example. Or keypads in extreme temperature/chemical environments. They don't have to be one of NEC's never-see-the-light-of-day toys.
Finally, the sensor doesn't have to be configured as a keyboard at all. It could be a mousepad area, where you don't need a mouse: any object would do.
I've seen and used VKB technology. It's very compact and an elegant solution. Unlike Canesta, it doesn't need a stereoscopic view of the target area, which means it really could fit into a pen sitting in a charger or other holder.
4 users >> 2 users (except for large values of 2)
More info, greater bandwidth, fewer ads, etc at Form Factors and, of course, Intel.
The car had been converted to battery power using a vast number of Li-ion cells in the trunk and, indeed, everywhere else. The trunk had a bank of fans that had to be on when recharging. Someone had added some retro mag-alloy wheels.
The interior was blue rather than red. There was a front bench seat, with the (now redundant) gear shift peeping up through a slot in the cushion. Changing gear must have been a very Freudian experience.
We drove around, gradually getting slower and slower, yet no matter how slow we went, no-one would pass us.
CDs/DVDs are outdated, ... I just got a Kingston 128mb USB key chain (1.1 nontheless),
If you've got 38 spare keychains, there's this movie you should see. I should be able to load them up in less than 12 hours.
On the other hand, it would herald a welcome return to traditional QA techniques. Wheeltappers from the rail industry could be retrained to walk around chips, tapping suspect components with their little hammers.
The 866BASE gets a P3, 2 ethernet ports, and the usual interfaces on a 91mm x 96mm board.
Plenty of opportunities for packing a nice computer into a small case.
US OEM version available at Diskology.
Using direct disk-to-disk cloning, it does 120GB in about 20 minutes. Full range of copy/cloning products here.
When not being used to upgrade drives, I use it as a simple disaster recovery backup: plug into an idle computer and dump the drive onto any cheap drive I have laying around.
The cloning software seems to be Easy Ghost in ROM, but I've not seen any problems after some 15 cloned drives.
Its benefits for me have been speed and reliability. The downside is that it doesn't seem to be available in the US but there are surely similar solutions.
If you're listening live, you can choose RealPlayer or Media Player. I think audio on demand is limited to RealPlayer.
With Total Recorder, you can capture the stream (and any other digital or analog audio stream in your computer) and save the resulting file(s) in any format you want. It has powerful automatic file naming, editing and scheduling tools, so since BBC7 repeats its output twice or more in a 24-hour cycle, I capture one stream to disk automatically divided into 15-minute chunks at 48kbps/22500kHz stereo, then schedule on-demand audio for other BBC programmes I want. The software gets the URL, starts RealAudio, does the capturing and shuts everything down nicely afterwards. It's an outstanding product.
Highlights: 8MB flash memory (4MB for data), 512kB RAM, runs at 24MHz or 48MHz, touchscreen uses Decuma handwriting recognition (Decuma is a Swedish company with Sony VC money; Decuma is also used in Sony Clies & cellphones). FM/Midi sound, vibrator; PIM, mail client supports POP/SMTP.
Can communicate with SD form-factor PHS card, Wi-Fi card, Bluetooth etc. Tri-color LED, sound and vibration alerts for incoming data.
I figure it's a much more useful device in Japan where connectivity is ubiquitous, than the US. But it will have a tough time competing with the likes of the J-SH53 and its successors.
The first OEM customer was Sterling Plastics (i.e. Rolodex). See here for the Japanese ones.
So Citizen certainly knows how to make this type of product.
And, indeed, putting it the wrong way. Quick slap with the cluestick there. Of course, it's a wonderful 320mm zoom. Mea culpa. Told you you shouldn't come to Slashdot for your answers.
Caveat: I've not used the Canon, I'm basing my opinions on reviews only, so E&OE. If you're new to SLR photography, this Canon is a nice place to start if you use it with the EF-S lens. If you're adding this digital SLR to your existing Canon kit, then you may find it difficult to cope with the 1.6x field of view crop: the sensor is smaller than a 35mm frame, so you need a 22mm lens on the digital SLR to get the same field of view as a 35mm lens on a 35mm SLR. Putting it another way, your expensive 200mm zoom becomes a heavy 125mm lens on the digital SLR
So, in that respect, if you have an existing lens collection that you're comfortable with, I'd go for a digital body with a full-frame 35mm sensor.
You'll notice that your worries will change from "do I have enough film?" to "do I have enough batteries?"
Regarding "how many pixels are enough?", people seem to think the sweet spot is between 8MP and 16MP, but with digital photos you don't get grain (but you do get noise), so I personally think 5MP is a decent starting point.
Penultimately, the default sharpness in Sony, Minolta and other digital cameras seems to be a tad too sharp: I routinely dial back the in-camera sharpening algorithm and do any sharpening outside the camera.
Finally, if you just like the idea of lots of pixels and quite a wide zoom range, you may want to at least hold the Sony F828, which is an 8MP camera due in November or December. I've played with one and, as a 707 owner, I like it a lot: like many others with this series of cameras, I tend to shoot using the camera back display, holding the camera like a TLR.
I'd love to be able to browse a giant back catalog, knowing that an original or facsimile copy could definitely be delivered to me.
The Taipei 101 has a pendulum-style damper, I think. I wonder how it's controlled if its motion is detected as contributing to oscillation rather than dampening it?
Interestingly, the spire has its own TMDs. I guess I'm a TMD kind of guy: there's something suitably retro about dragging 800 tons of stuff up as high as possible, simply to have it wobble or slide around.
Until elevators came along, there was no point in building high, so the first expressions of national can-do were horizontal: bridges.
That image is well worth panning through: the sheer number of galaxies in the background is astonishing. I guess it's a big place, after all.
Well that's not the fault of the Chinese, I think.
In the early 1400's Cheng Ho (Zheng He) made 7 voyages towards the western world, reaching Africa. He took more than 27,000 people in 320+ ships with him on some voyages. The largest ships weighed 1500 tons, were 180 meters long, and held 1000 people. See here to see how vastly larger a treasure ship was than the Santa Maria, built 87 years later.
I concur, though my tool of choice is the humble Leathermaster. I'm working my way through 3 video recorders at the moment, component by component.
As an indication of the difference between, say, English and Japanese, Japanese handwriting recognition tends to be based on detecting the order in which the strokes for a character are laid down, which relies on everyone learning the same way. Recognition of Roman letters tends to be based on stroke and pool analysis (i.e. zero to two or so strokes plus any enclosed spaces).
My personal experience of speed reading suggests that the envelope shape of the word is very important. I figure the scrambled letters of the sample texts could be replaced by similar letters having the same "mass", as it were, and the text would still be reasonably comprehensible.
You could try a set of programmable foot switches, such as Kinesis. You need to program the actions from a Windows machine, but the unit can then be used with other OSes. I used this together with a small programmable keypad a few years back.
We were expected to consult with our team and amend our hours to be reasonably coherent: an all-early-arriving team would expect new members to arrive early, for example.
Since it was my first work experience, I thought it was normal. I've since learnt that it was an outstanding system for all concerned: the company calculated its employees were working an extra 12 minutes per day for free on average, and we thought we were being treated as humans.
Yup 8-). In an ideal world, more stable dyes are developed over time. More stable dyes mean you can write to them faster, so one indicator of this stability in the market is the existence of faster media. This inherent stability suggests a longer media life. So, given the choice between storing data on an older 1x CDR media or a newer 32x CDR media, I would choose the latter.
Also worth considering is that CDR dyes are sensitive to radiation at 780 nm. That's heat. So we have to protect a heat-sensitive disc from, well, heat.
DVD-R dyes are sensitive to radiation at 635-650 nm. That's red light. So we have to protect the discs from light, which is somewhat easier.
Newer products are sensitive to blue light, so are even easier to protect from light.
CDR is a single polycarbonate substrate with a varnish top coating to protect it. The label side, therefore, is the weak link and CDRs can die simply because the top coating ages. DVD-x is two polycarbonate substrates bonded together. So one side is as strong as the other. Polycarbonate technology is very mature. So DVDs should be inherently more robust than CDR.
Finally, many disk writers deliberately operate at higher power than specified for CDR and DVD-R. So the media may be receiving more energy during the write phase than it's supposed to. So a newly-burnt disk may come out the burner effectively a significant way through its supposed storage life.
But, all things being equal, I stick by my assertion that it's all down to the dye, and because all things aren't equal, it's why I archive to phase-change media!