For those fo you that don't know, Carver Mead is the scientist who constructed a "silicon retina" (a light-sensing chip that mimics the topology and information processing techniques of the retina.)
This appears to be an outgrowth of that work, using the analog VLSI techniques that were invinted for the retina project.
Yeah, well Microsoft only provides these free numbers for people to download updates, so they only need to contact Microsoft's servers. So there's no reason why their DNS would need to forward requests, or even be connected to the Internet.
You could employ the "sophisticated-super-compooter-image-recognition-s oftware" used by the friendly censor
groups!
I think the proper word is "sophistimacated":)
Actually, one of the techniques I implemented for my vision class has something like a 97% accuracy in distingushing between images that contain faces, and images that don't. The research actually is getting somewhere.
Now, this is truly a "moby hack" and I'm very impressed. It strikes me that Microsoft (or whoever is providing the telephone line) could simply filter out DNS requests that aren't trying to find one of their servers.
"they beam this information everywhere, all through the fucking air. You just gotta know how to grab it. Just got to know how to grab it." --Heat
So how long until some maker of domain servers sends them a cease-and-desist letter?
Sometimes the current legal climate, re DeCSS, the CueCat, et al, makes me wish all the good hackers knew how to stay underground instead of posting websites everywhere saying "700k 4t m3! 1 m4d3 4n 0p3n 50urc3 h4xxx0r!!!" It's the kind of thing that causes bad laws to be passed.
(before you flame, realize my tounge is planted halfway in cheek....)
And I'm actually sticking to VHS for now, the tapes are only half as much, and I don't really need a "choose your
own adventure" style porno.
Yeah, keeping a hand on the remote is a drag. Also, it's a drag having to hit a key to advance the picture in GQView. Someone should write a program that uses computer learning techniques to figure out what kind of pictures I like to linger on, and which ones I like to skip over...
Damn, I just got an idea for an Open Source Project!!!!
Someone should engrave the contents of css_descramble.c into a brass placard in front of the Jefferson Memorial.
We all know he would have supported us in this cause if he were alive.
OK, I get it so far....
Software designers are so infatuated with the fact that they can, that they don't stop to think if they should.
It takes a little practice to
use a standard transmission without lurching out of a stoplight. Why shouldn't computers be the same way?
80% of the people I know can't drive a stickshift. And yet they get around, because an automatic hass all the functionality they need. The real problem is that some commercial interfaces are designed to be easy to use in the showroom where the salemen points to the Start menu or whatever, instead of having to explain key concepts ("you see, there's these files, and they go in directories....") And computer salesmen are certainly not going to point at a user manual.
Now, a TV comes with a poorly translated manual that's hard to read, but at least it tells you how to change the channel! Computers today seem to come with no such thing. Just a 6 page booklet that tells you how to turn it on and maybe find the Help button if you're lucky. Thing is, while people will read a printed manual, reading online help is an unproductive pain in the neck for most people. It's a crying shame.
"look how easy to set up our computer is! why, the manual is only 4 pages long, and it's all pictures!"
Thank you. This place is populated by people who threw the baby out with the bathwater when they took up command-line interfaces. Interface design is a Good Thing and it's what we should be doing; interface misdesign is what we should be fighting.
If anyone has an old 68000-era Mac around, they should take a look at the owner's manual. That's a WELL-WRITTEN, NON-TECHNICAL book that tells you how to USE THE MACHINE. A far cry from the modern documentation that comes with today's machines: "here's how you plug it in, here's the power switch, press the question mark if you have questions." And then the computer companies tout their system as being easy to use! "Look, the manual is only six pages..."
User "stupidity" is usually due to this maddening cycle: rather than explain key concepts (not even the difference between * and a regexp, I mean key concepts like files go in directories they try to hide the concept and make it unnecesary (rather than put applications on your disk in an appropriately named directory, applications go under the Start menu.) That backfires when you ned to do something not in the Start menu. But by that point, the computer and the OS has already been sold and invested in by the user. So the incentive for computer/OS makers is to sell units that appear to be useful, rather than be useful in real life.
Me, I don't care how easy and lickable the interface is, is has to come with a MANUAL that TELLS you how to DO THINGS.:-P
That clears it up, at least somewhat... What
you want is a dual-paned file manager but with icon-based display, right? Cool. Good luck finding one.;)
The NeXT filemanager, Greg's Browser, and the MacOSX browser all work like this, IIRC, haven't used one of those in a while.
Oh, you meant find one for Linux, did you? Can't help you there, I don't even LIKE dual-pane filemanagers. At least they're better than single pane filemanagers. But I prefer to have my folders open in a new window when I'm given the choice. And the latest versions of Windows make it very hard to do, blech.
The filemanager is dead. Long live the filemanager. Bah.
Notice how a decent window manager puts the Close button at the opposite side from the other, less damaging buttons? Well, not all windowmanagers should be like that, but they should be. In MacOS The close button is on one side and minimize/maximize are at the other side. This is so that it's hard to accidentally close a window, and it's easier to focus on the button you want.
This is the principle at work here. "put all the buttons in the same place" is not a valid design principle (IMHO), because different butons do different things, and ot put them all in the same place just encourages confusion. Even worse is the recent toolbar-style trend, where you have a bunch of vague icons all in the same place, and you have to mouse over them and wait for the tooltip to pop up before you can tell what one of them does. (GNOME people, are you listening? Is it too much to ask to put text labels on my icons?)
'Course, it's all my opinion. I'm in favor of skinnability, but not without at least a usable reference implementation, and I have some faith in the peopel at Eaze, since so many of them have experience in interface design, of the kind where you hire non-authors to try the interface, and actually see how fast it takes to do things.
Actually, one of the better innovations I have seen is the use (in GDAM for example) of Glade xml files for interface descriptions. If you don't like it then bop into Glade and put the buttons where you like them. This is a Good Thing as far as skinnability goes, and I wish more people would use it.
Precious few make any money on IP. Most make it on live
shows. So they want the word out (mindshare) so attendence will be good. Musicians do not need IP to make a
living. The absence of IP will not make a dent in the lives of the vast majority of working musicians. It will make
it hard for those in the musical pork barrel industry that grafts itself onto musicians' backs.
This statement is untrue and I don't want to ever hear this argument again. I know several local musicians, many of whom are very good, much better than 90% of major-label artists--but none of them make a dime off of playing shows. The fact is that you have to sell tickets to turn a profit playing shows. Very few people have enough name recognition power to fill a venue. U2, Pink Floyd, LatestTeenPopSensation--these are the people who have been able to convince enough fans to give $20 or up to go se a show. But not Noe Venable--an incredible musician in my area. Do you know how she makes money? BY SELLING RECORDS AT HER SHOWS. Not by selling tickets. If you can come up with a way to sell tickets for a small artist, let me know. But most people won't pay to see someone they haven't heard of. However they will go to a free show, or to a club with a small cover and many musicians--and if they like it, they'll buy CDs there.
Now there are exceptions to this rule--the rave scene comes to mind, as most DJs make money by playing shows. But still, only a few rave promoters are able to turn a profit, and to be blunt, a pair of turntables and a collection of records costs much less than good musical inturements and studio time. And if someone playing at the rave has albums out, you can be sure they'll sell them.
Now these DJs must buy their records from someone, right? It's a rare electronica producer that has the ability to play a live show. How do these people make money? Why, by selling records!
Look into the costs of touring and playing concerts before you spout off misleading information. FEW artists may make money on album sales, but FEWER (much, much fewer) make any money on live shows.
"Hell, my normal usage is 1) Listening to an MP3
2) Coding in Visual Studio.
3) Doing graphics in 3D Studio
4) Doing textures in Photoshop.
This is without any skips from the MP3 player! "
I've *never* had my MP3 player skip on me in Linux. However it's not the point. The point is latency. You see, anyone can simply allocate a whoppin' big buffer and get rid of skips that way. It's easy, and it's the wrong solution.
Let's say I'm performing music live, using a sequencer and a software synth. If I hit a note on my keyboard, the software synth needs to respond to it and play that note. This needs to happen immediately, i.e. within milliseconds. A buffer does nothing for you here--you need to guarantee that the synth program geets a timeslice whenever it needs one, no matter what.
Your mp3 player doesn't need to respond immediately to any actions, so it can prebuffer as long as it needs to. Try adjusting the equalizer. Notice how long it takes for your changes to take effect? There's your prebuffering.
For true realtime performance--scientific and mission-critical applications, say, like targeting a brain tumor with a linear accelerator--you need hard realtime (i.e. microsecond latencies). This is provided by QNX and the RTLinux kernel.
For 'soft' realtime like my synthesizer, you need millisecond latencies. This is provided by BeOS, Ingo Molnar's Linux patches, and now the MonteVista patches. Not by Windows or WindowsNT.
For mp3 playing, you just need buffering, and you can have shit latencies. This can be implemented under any OS.
Need I remind anyone that the Apple Newton was using StrongARM five years ago?
Foo. Anyway, it's not like that trounced anyone in the marketplace... bringing us to the truth that architectore doesn't matter. Palm could be using PICs and they'd still have marketing clout.:-P
Connectix can still market in Intel virtual machine (akin to VMWare, Plex86, MOL, SheepShaver) for MacOS, which would still be quite useful for a lot of people. Same thing as Virtual PC, only you don't have to bother with translating the instruction set. So you wouldn't have to reboot, and you gain the nifty things that Connectix Virtual PC has, like full copy&paste and drag&drop between the Windows environment and the Mac. It'd still sell.
I'm not so sure about "many of the folks in the industry will admit freely to doing the same things". We have a
hard enough time getting IT manangers to admit to their bosses they're using Linux/FreeBSD on the mail server.
What makes people think that saying "Gee boss, I'm a hacker" is any easier?
Well, the "Industry" in question is the computer security industry. Whole different ballgame from the IT industry as a whole.
In California a child can be convicted of a felony for writing his name in wet cement. "Felony" has become a meaningless term, but we don't have anything to replace it. But it's a meaningless term that will follow you around for the rest of your life, and people will be denied opportunities because of a meaningless "felon" label attached to them.
Some people will be objective and look hard enough to discriminate between a serious crime and a bullshit "felony." But not enough people.
Slashdot code could be used to the same purpose. Submit a patent idea as a story, let others flesh it out. Any
open-source product that gets a cease and desist letter can send back a URL of where to pick up the discussion on
Slashdot.
Isn't this implemented at ShouldExist? Of course they use Scoop instead...
Slashdot Mirror
....each pound of which weighs over ten thousand pounds.
Darwin compiles on Intel hardware, genius.
This appears to be an outgrowth of that work, using the analog VLSI techniques that were invinted for the retina project.
Yeah, well Microsoft only provides these free numbers for people to download updates, so they only need to contact Microsoft's servers. So there's no reason why their DNS would need to forward requests, or even be connected to the Internet.
Look up Markus Weber's 2000 PhD thesis from Caltech, that should be able to get you started.
I think the proper word is "sophistimacated" :)
Actually, one of the techniques I implemented for my vision class has something like a 97% accuracy in distingushing between images that contain faces, and images that don't. The research actually is getting somewhere.
"they beam this information everywhere, all through the fucking air. You just gotta know how to grab it. Just got to know how to grab it." --Heat
Sometimes the current legal climate, re DeCSS, the CueCat, et al, makes me wish all the good hackers knew how to stay underground instead of posting websites everywhere saying "700k 4t m3! 1 m4d3 4n 0p3n 50urc3 h4xxx0r!!!" It's the kind of thing that causes bad laws to be passed.
(before you flame, realize my tounge is planted halfway in cheek....)
I knew that. What makes you think I didn't know that?
Yeah, keeping a hand on the remote is a drag. Also, it's a drag having to hit a key to advance the picture in GQView. Someone should write a program that uses computer learning techniques to figure out what kind of pictures I like to linger on, and which ones I like to skip over...
Damn, I just got an idea for an Open Source Project!!!!
OK, I get it so far....
Software designers are so infatuated with the fact that they can, that they don't stop to think if they should.
I don't get it. Are you trying to be ironic?
80% of the people I know can't drive a stickshift. And yet they get around, because an automatic hass all the functionality they need. The real problem is that some commercial interfaces are designed to be easy to use in the showroom where the salemen points to the Start menu or whatever, instead of having to explain key concepts ("you see, there's these files, and they go in directories....") And computer salesmen are certainly not going to point at a user manual.
Now, a TV comes with a poorly translated manual that's hard to read, but at least it tells you how to change the channel! Computers today seem to come with no such thing. Just a 6 page booklet that tells you how to turn it on and maybe find the Help button if you're lucky. Thing is, while people will read a printed manual, reading online help is an unproductive pain in the neck for most people. It's a crying shame.
"look how easy to set up our computer is! why, the manual is only 4 pages long, and it's all pictures!"
Riiight.
Thank you. This place is populated by people who threw the baby out with the bathwater when they took up command-line interfaces. Interface design is a Good Thing and it's what we should be doing; interface misdesign is what we should be fighting.
User "stupidity" is usually due to this maddening cycle: rather than explain key concepts (not even the difference between * and a regexp, I mean key concepts like files go in directories they try to hide the concept and make it unnecesary (rather than put applications on your disk in an appropriately named directory, applications go under the Start menu.) That backfires when you ned to do something not in the Start menu. But by that point, the computer and the OS has already been sold and invested in by the user. So the incentive for computer/OS makers is to sell units that appear to be useful, rather than be useful in real life.
Me, I don't care how easy and lickable the interface is, is has to come with a MANUAL that TELLS you how to DO THINGS. :-P
The NeXT filemanager, Greg's Browser, and the MacOSX browser all work like this, IIRC, haven't used one of those in a while.
Oh, you meant find one for Linux, did you? Can't help you there, I don't even LIKE dual-pane filemanagers. At least they're better than single pane filemanagers. But I prefer to have my folders open in a new window when I'm given the choice. And the latest versions of Windows make it very hard to do, blech.
The filemanager is dead. Long live the filemanager. Bah.
This is the principle at work here. "put all the buttons in the same place" is not a valid design principle (IMHO), because different butons do different things, and ot put them all in the same place just encourages confusion. Even worse is the recent toolbar-style trend, where you have a bunch of vague icons all in the same place, and you have to mouse over them and wait for the tooltip to pop up before you can tell what one of them does. (GNOME people, are you listening? Is it too much to ask to put text labels on my icons?)
'Course, it's all my opinion. I'm in favor of skinnability, but not without at least a usable reference implementation, and I have some faith in the peopel at Eaze, since so many of them have experience in interface design, of the kind where you hire non-authors to try the interface, and actually see how fast it takes to do things.
Actually, one of the better innovations I have seen is the use (in GDAM for example) of Glade xml files for interface descriptions. If you don't like it then bop into Glade and put the buttons where you like them. This is a Good Thing as far as skinnability goes, and I wish more people would use it.
This statement is untrue and I don't want to ever hear this argument again. I know several local musicians, many of whom are very good, much better than 90% of major-label artists--but none of them make a dime off of playing shows. The fact is that you have to sell tickets to turn a profit playing shows. Very few people have enough name recognition power to fill a venue. U2, Pink Floyd, LatestTeenPopSensation--these are the people who have been able to convince enough fans to give $20 or up to go se a show. But not Noe Venable--an incredible musician in my area. Do you know how she makes money? BY SELLING RECORDS AT HER SHOWS. Not by selling tickets. If you can come up with a way to sell tickets for a small artist, let me know. But most people won't pay to see someone they haven't heard of. However they will go to a free show, or to a club with a small cover and many musicians--and if they like it, they'll buy CDs there.
Now there are exceptions to this rule--the rave scene comes to mind, as most DJs make money by playing shows. But still, only a few rave promoters are able to turn a profit, and to be blunt, a pair of turntables and a collection of records costs much less than good musical inturements and studio time. And if someone playing at the rave has albums out, you can be sure they'll sell them.
Now these DJs must buy their records from someone, right? It's a rare electronica producer that has the ability to play a live show. How do these people make money? Why, by selling records!
Look into the costs of touring and playing concerts before you spout off misleading information. FEW artists may make money on album sales, but FEWER (much, much fewer) make any money on live shows.
I've *never* had my MP3 player skip on me in Linux. However it's not the point. The point is latency. You see, anyone can simply allocate a whoppin' big buffer and get rid of skips that way. It's easy, and it's the wrong solution.
Let's say I'm performing music live, using a sequencer and a software synth. If I hit a note on my keyboard, the software synth needs to respond to it and play that note. This needs to happen immediately, i.e. within milliseconds. A buffer does nothing for you here--you need to guarantee that the synth program geets a timeslice whenever it needs one, no matter what.
Your mp3 player doesn't need to respond immediately to any actions, so it can prebuffer as long as it needs to. Try adjusting the equalizer. Notice how long it takes for your changes to take effect? There's your prebuffering.
For true realtime performance--scientific and mission-critical applications, say, like targeting a brain tumor with a linear accelerator--you need hard realtime (i.e. microsecond latencies). This is provided by QNX and the RTLinux kernel.
For 'soft' realtime like my synthesizer, you need millisecond latencies. This is provided by BeOS, Ingo Molnar's Linux patches, and now the MonteVista patches. Not by Windows or WindowsNT.
For mp3 playing, you just need buffering, and you can have shit latencies. This can be implemented under any OS.
High Intensity Computer Colleges? by Cliff on Sunday October 10, @01:43AM EDT 322
2 Do Geeks Need College? by CmdrTaco on Monday April 12, @11:50AM EDT 365
2 Should Geeks Skip College? by CmdrTaco on Monday January 11, @09:45AM EDT 475
2 Rob Finishes College by CmdrTaco on Thursday December 17, @11:01AM EDT 134
Need I remind anyone that the Apple Newton was using StrongARM five years ago?
Foo. Anyway, it's not like that trounced anyone in the marketplace... bringing us to the truth that architectore doesn't matter. Palm could be using PICs and they'd still have marketing clout. :-P
(not that I have anything against PICs...)
Slashdot is a good source of links, but is fundamentally content-free. Reading it like this is just as useful.
Connectix can still market in Intel virtual machine (akin to VMWare, Plex86, MOL, SheepShaver) for MacOS, which would still be quite useful for a lot of people. Same thing as Virtual PC, only you don't have to bother with translating the instruction set. So you wouldn't have to reboot, and you gain the nifty things that Connectix Virtual PC has, like full copy&paste and drag&drop between the Windows environment and the Mac. It'd still sell.
Well, the "Industry" in question is the computer security industry. Whole different ballgame from the IT industry as a whole.
Some people will be objective and look hard enough to discriminate between a serious crime and a bullshit "felony." But not enough people.
Isn't this implemented at ShouldExist? Of course they use Scoop instead...