Or do you mean you don't like Mac UI's? Well guess what? It's the same app whether you recompiled it or not. People are actually complaining about this for firefox - they don't follow Mac's standards for accelerators. Again, recompile + X.
Stack behaviour is static (ie, stack effects are constant. m popped, n pushed, all the time - unlike Forth) in the JVM. That means it is possible, but sort of hard, to statically translate the stack code to register code. Push/Pop don't HAVE to be used. IIRC, it could be relatively complex in Forth (http://www.complang.tuwien.ac.at/projects/forth.h tm it might be discussed in this paper http://www.complang.tuwien.ac.at/papers/ertl93.ps. gz, but i'm far from sure, and don't feel like making sure at the time:p), but that might be related tot he fact that forth words can have arbitrary stack effects (both data and return), complicating even the common case. So, intuitively, while that sort of optimisation can be used in a real compiler, i'm not sure a JIT one can afford it. Maybe that's a place where gcj could make up for their lack of profiling data.
Woah, cool! Please post more info on your journal, that's exactly the sort of thing i'm interested in:)
Well, current products sometimes try and hide their head in the sand and pretend that audio feedback is just as good. That's always an option;)
I believe Sony claims that an LCD screen + piezo-electric vibrations is enough for basic tactile feedback, and had a demo some time ago.
I've thought of using a LCD with small pin-like buttons (a couple mm of diameter) going through holes at short regular intervals. Obviously, you lose a lot on the graphic front, but, it seems to me, should gain a lot on the usability as an input device front. The idea wouldn't be to use each button separately (too much precision required), but rather to press on adjacent ones simultaneously.... A bit like matrix keypads when you press on 3 buttons in a L (the fourth corner of the rectangle is automatically sensed as pressed), well, it'd use that effect to have a button associated to each square (1x1 square). You get to use most of the screen as normal, but can still have luverly clickety action:)
Apart from that... I don't think LCDs like being locally raised or lowered mechanically in general;)
The CPU listed is a Nios @ 100 MHz. Nios is Altera's soft core, so either they prototyped on an FPGA, and will use the same design to fab, or, much more likely since it's obviously a low volume application, they still have the fpga in the production equipment. Given that there aren't many (heh) GPUs for that kind of device, my guess is they have the CPU and the GPU on the same FPGA. Most newer FPGAs have capabilities for partial reconfiguration. Why didn't they leave the possibility of using it to their users, or if they did, why aren't they touting this as a feature? I suppose the feature wouldn't be as useful as if it could drive devices directly, but it still could be used for more exotic UI/Graphic capabilities.
If only the thing was smaller, i.e., ~ the size of a graphing calculator, and didn't use _14_ W. There's this touchpad/keyboard that i keep reading about on/. whenever someone mentions RSI. I think they sell their keyboards at something like $400-500. It would be interesting if THIS one was cheaper:)
Re:yet more confusion between ibook and powerbook
on
Apple Announces New iBooks
·
· Score: 4, Informative
Lighter, more professional looking*, better FSB, possibility to go and haggle for a lower price due to the new ibook being so close in perf? More built-ins: 802.11g, BT, larger HD. More video RAM, which i guess is more important with Quartz. Better case (aluminium > plastic:). Oh, and the PB is slightly smaller in every dimension. Note that i only made the comparison for the 12.1" models, some points may not apply on larger ones.
*That was actually important for my Corpo sister. She can't go on a board meeting with a shiny, cutesy iBook.
New Nintendo Handheld. Two screens, lower one is touch sensitive. The hardware received good reviews (see penny-arcade, for example). A handheld where FPSes are playable! And that i can use to Operation, Electronic Millenium Edition! What's not to 3?:)
There are many interesting options for input devices in a wearable computer. For the short term, i'm going to buy a throat mic (around Christmas, said the poor broke college student:), but the frogpad (http://www.frogpad.com/) has received good reviews, and the last time i checked, thinkgeek still had some halfkeyboards... On a "slightly" longer term, gesture recognition should work ok with a webcam:)
Maybe it's only the tools that are to blame (only this one time, hehe). Might something like Teddy http://www-ui.is.s.u-tokyo.ac.jp/~takeo/teddy/tedd y.htm be better suited to you? You only have to draw the outline, and the shapes are assumed to be round; You can then cut them as you want.
I've heard Shade (a popular modeller in Japan; Gunnm's author uses it http://jajatom.moo.jp/E-top/Egunnm/3DCG01/cg%20gal lery%20top.html) had a module reminiscent of Teddy in one of its newer versions. Of course, Shade seems to be impossible to find outside Planet Japan, so i'm not sure how much that helps, apart from letting you know that there are alternatives...
IT's CSS+Javascript+Tables. CSS might be supported by links, tables are, but not JS. So to support links/lynx, they'd have to completely change the architecture. Doubtful. In other words, they CAN support text-only browsers, it's only that those browsers have other limitations than only not having pictures.
A vector processor is SIMD; a cluster is MIMD. MIMD is a superset of SIMD, but, _theoretically_, less effective. The advantage of SIMD is its simplicity, but when MIMD is effectively simpler to use than SIMD, vector processors are obviously only used when one has to.
There are a ocuple problems with that plan, even assuming that the sun shines all year long, 7/7.
Hydroelectricity is very demanding in large, unoccupied spaces - you need huge basins, and the new lake can be filled with heavy metals (that one may have been fixed - i'm not that well-informed). Extracting energy from falls isn't generally as efficient as extracting it from a dam. Moreover, depending on our (Hydro-Québec's) energy is a Bad Plan: we have our very own green nuts too! We still have to find a place to build new dams. Their main argument is that we don't need more power... Which, considering the population's growth rate might have some truth in it, but doesn't take into account that oil is still a major source of energy for us, and the more we can use electricity, the less we'll need it.... And the same is true for our clients. Even if Hydro-Q only wanted to build new dams to make money, but for each clean MWh they export, someone is producing one less dirty MWh. But, as it is, we're going to have to _import_ energy in a couple years, and if we don't build our own dams or relatively clean natural gas power plant (nuclear is taboo, obv), i'm pretty sure the electricty we'll buy will come from dirtier sources. But that's the perverse effect of Kyoto for you: we're not trying to reduce green house gas emission as a planet - we'll simply OUTSOURCE the pollution:)
Sp erhm... Just follow China's lead and buy our uranium instead:)
The Zaurus C3xx was successfully used in a wearable (twiddler and BOB-II). The SL-6000 should work even better, since it has USB host built-in. You might want to consider something like a frogpad, if you have the $$. (frogpad+SL6k+earphones pretty much = instant unconspicuous wearable... we're getting there:)
Slashdot Mirror
OK. So you like Mac UI's? Use it.
Or do you mean you don't like Mac UI's? Well guess what? It's the same app whether you recompiled it or not. People are actually complaining about this for firefox - they don't follow Mac's standards for accelerators. Again, recompile + X.
So, recompile your apps and use a X server?
Polydactily is a rare but dominant mutation in humans.
Stack behaviour is static (ie, stack effects are constant. m popped, n pushed, all the time - unlike Forth) in the JVM. That means it is possible, but sort of hard, to statically translate the stack code to register code. Push/Pop don't HAVE to be used. IIRC, it could be relatively complex in Forth (http://www.complang.tuwien.ac.at/projects/forth.h tm it might be discussed in this paper http://www.complang.tuwien.ac.at/papers/ertl93.ps. gz, but i'm far from sure, and don't feel like making sure at the time :p), but that might be related tot he fact that forth words can have arbitrary stack effects (both data and return), complicating even the common case. So, intuitively, while that sort of optimisation can be used in a real compiler, i'm not sure a JIT one can afford it. Maybe that's a place where gcj could make up for their lack of profiling data.
Woah, cool! Please post more info on your journal, that's exactly the sort of thing i'm interested in :)
;)
:)
;)
Well, current products sometimes try and hide their head in the sand and pretend that audio feedback is just as good. That's always an option
I believe Sony claims that an LCD screen + piezo-electric vibrations is enough for basic tactile feedback, and had a demo some time ago.
I've thought of using a LCD with small pin-like buttons (a couple mm of diameter) going through holes at short regular intervals. Obviously, you lose a lot on the graphic front, but, it seems to me, should gain a lot on the usability as an input device front. The idea wouldn't be to use each button separately (too much precision required), but rather to press on adjacent ones simultaneously.... A bit like matrix keypads when you press on 3 buttons in a L (the fourth corner of the rectangle is automatically sensed as pressed), well, it'd use that effect to have a button associated to each square (1x1 square). You get to use most of the screen as normal, but can still have luverly clickety action
Apart from that... I don't think LCDs like being locally raised or lowered mechanically in general
The CPU listed is a Nios @ 100 MHz. Nios is Altera's soft core, so either they prototyped on an FPGA, and will use the same design to fab, or, much more likely since it's obviously a low volume application, they still have the fpga in the production equipment. Given that there aren't many (heh) GPUs for that kind of device, my guess is they have the CPU and the GPU on the same FPGA. Most newer FPGAs have capabilities for partial reconfiguration. Why didn't they leave the possibility of using it to their users, or if they did, why aren't they touting this as a feature? I suppose the feature wouldn't be as useful as if it could drive devices directly, but it still could be used for more exotic UI/Graphic capabilities.
/. whenever someone mentions RSI. I think they sell their keyboards at something like $400-500. It would be interesting if THIS one was cheaper :)
If only the thing was smaller, i.e., ~ the size of a graphing calculator, and didn't use _14_ W. There's this touchpad/keyboard that i keep reading about on
Lighter, more professional looking*, better FSB, possibility to go and haggle for a lower price due to the new ibook being so close in perf? :). Oh, and the PB is slightly smaller in every dimension. Note that i only made the comparison for the 12.1" models, some points may not apply on larger ones.
More built-ins: 802.11g, BT, larger HD. More video RAM, which i guess is more important with Quartz. Better case (aluminium > plastic
*That was actually important for my Corpo sister. She can't go on a board meeting with a shiny, cutesy iBook.
New Nintendo Handheld. Two screens, lower one is touch sensitive. The hardware received good reviews (see penny-arcade, for example). A handheld where FPSes are playable! And that i can use to Operation, Electronic Millenium Edition! What's not to 3? :)
I knew i'd forgotten something: Belkin's nostromo n50.
There are many interesting options for input devices in a wearable computer. For the short term, i'm going to buy a throat mic (around Christmas, said the poor broke college student :), but the frogpad (http://www.frogpad.com/) has received good reviews, and the last time i checked, thinkgeek still had some halfkeyboards... On a "slightly" longer term, gesture recognition should work ok with a webcam :)
Maybe it's only the tools that are to blame (only this one time, hehe). Might something like Teddy http://www-ui.is.s.u-tokyo.ac.jp/~takeo/teddy/tedd y.htm be better suited to you? You only have to draw the outline, and the shapes are assumed to be round; You can then cut them as you want.
l lery%20top.html) had a module reminiscent of Teddy in one of its newer versions. Of course, Shade seems to be impossible to find outside Planet Japan, so i'm not sure how much that helps, apart from letting you know that there are alternatives...
I've heard Shade (a popular modeller in Japan; Gunnm's author uses it http://jajatom.moo.jp/E-top/Egunnm/3DCG01/cg%20ga
With all this talk about goto as the ultimate control flow, i'd say it's pretty clear someone doesn't grok continuations.
Are you sure it's not showing you 68 threads? Some of your users could be doing something weird...
IT's CSS+Javascript+Tables. CSS might be supported by links, tables are, but not JS. So to support links/lynx, they'd have to completely change the architecture. Doubtful. In other words, they CAN support text-only browsers, it's only that those browsers have other limitations than only not having pictures.
A vector processor is SIMD; a cluster is MIMD. MIMD is a superset of SIMD, but, _theoretically_, less effective. The advantage of SIMD is its simplicity, but when MIMD is effectively simpler to use than SIMD, vector processors are obviously only used when one has to.
It sounds more like the dispatch mechanism in CLOS.
There are a ocuple problems with that plan, even assuming that the sun shines all year long, 7/7.
:)
:)
Hydroelectricity is very demanding in large, unoccupied spaces - you need huge basins, and the new lake can be filled with heavy metals (that one may have been fixed - i'm not that well-informed). Extracting energy from falls isn't generally as efficient as extracting it from a dam. Moreover, depending on our (Hydro-Québec's) energy is a Bad Plan: we have our very own green nuts too! We still have to find a place to build new dams. Their main argument is that we don't need more power... Which, considering the population's growth rate might have some truth in it, but doesn't take into account that oil is still a major source of energy for us, and the more we can use electricity, the less we'll need it.... And the same is true for our clients. Even if Hydro-Q only wanted to build new dams to make money, but for each clean MWh they export, someone is producing one less dirty MWh. But, as it is, we're going to have to _import_ energy in a couple years, and if we don't build our own dams or relatively clean natural gas power plant (nuclear is taboo, obv), i'm pretty sure the electricty we'll buy will come from dirtier sources. But that's the perverse effect of Kyoto for you: we're not trying to reduce green house gas emission as a planet - we'll simply OUTSOURCE the pollution
Sp erhm... Just follow China's lead and buy our uranium instead
Tablets aren't touch sensitive... Touch sensitive devices are usually less expensive, though, so you might be in luck :)
Preview is your friend.
The Zaurus C3xx was successfully used in a wearable (twiddler and BOB-II). The SL-6000 should work even better, since it has USB host built-in. You might want to consider something like a frogpad, if you have the $$. (frogpad+SL6k+earphones pretty much = instant unconspicuous wearable... we're getting there :)
Nor does the patent (OR, not and). It's extremely broad, including just about any input or output device as long as it is collar-mounted.
Check the nomadic radio link. It's not jsut vapourware - it was used in research for a couple years (not anymore, since the creator got his degree).
nt
GPUs don't do IEE floats. That might be bad for his purpose...
Network. USB.