The map tiles are all PNG images. Check out the page info in Firefox, and look up the Media tab. Like "Google Suggest", this is making a bunch of requests for fragments of information on the backend. It looks real nice, but the programming is fairly straight-forward.
I was hoping that it was using SVG maybe, even though it's not standard yet. If a portal used SVG for something widespread like a mapping service, then all the browsers would be forced to make it standard, hopefully.
Totally agree. Not only that, but if you have one great idea and get millions, then you've got seed money to bankroll your next great idea. Then instead of getting a payoff of 10 more million, you could easily create another company and be worth 10x or 100x more. I think that if you're really smart, then you know how NOT smart your bosses are, i.e. the founders. If your founders are worth hundreds or thousands of millions (like the Google, Yahoo, or Microsoft founders), then you might still feel undervalued. Once you've got financial security, then your ego still needs to be fed.
If your motivation is to do great work, and you're already rich, then you might as well just teach and do research at a leisurely pace, armed with a bunch of eager (and poor) grad students. In the research world you can make a long-term, pervasive impact without worrying about giving your competitors ammunition to kill you, as you would by publishing your research from inside a company.
Basically, I think that keeping the really good employees is always a problem, whether you give them a little or a lot of money.
There are certainly hundreds of cases of prior art, and Tripwire is probably one of them. It computes and maintains a database of hashes for all the files on a file system to check for intrusions and corruption. The wiki entry says it first surfaced in 1992.
Just curious, since I'm not an international lawyer, what happens if somebody sues a Chinese company. Can't China just claim that they will not honor any software patents on any software or on Linux specifically? It's not like they have a history of respecting other countries IP rights.
The US could complain to the WTO or somebody, but they are toothless. China is too big to start a trade war with.
Poland just recently decided against supporting software patents in the EU. Does that mean they will not respect other countries' patents on software or just that they will not go along with Europe issuing them?
This would be fantastic. I've seen websites where people describe how to make projection screen TVs using LCD panels, mirrors, high lumen lamps, fresnel lenses, etc. They seem to be cheap and easy, but they are hot, and require fans and lots of ventilation.
What would be better is a high brightness array of LED lights (white of course) to use as the white light source. Some DIY forums have noted that the problem is focusing the light from the array into a controllable direction and even intensity. This is hard because you need so many LEDs to get enough brightness. If this could be cheaply solved, then it would make a great DIY projector.
Anybody have any luck with this or know any good websites for it?
Fuel-cell technology was advanced for use in the space shuttle.
Great advances were made to make the heat shield for the shuttle. I'm sure the glue (even though imperfect) was a big advance also.
To get people to Mars and back will require many more breakthroughs, since they will be in space for a few months at least. Problems include developing very efficient energy systems like better solar cells and batteries. They may have to get rocket fuel from the ice on Mars. (I don't know how they would extract the oxygen or hydrogen, but I remember that was a suggestion.)
Life support systems will have to recycle everything possible, for food, air, human waste. Shielding would have to be advanced enough to protect from cosmic rays. Some form of human hibernation may eventually be used for very long term human space flight. All of these advances would have great everyday application, such as
UV-protection, medicine, clean energy, garbage reduction and efficient recycling, hydrogen fuel production and certainly many many others.
Oh, and without the manned space program, we probably wouldn't have gotten 2001: A Space Odyssey, Star Wars, Star Trek, Blade Runner,
Planet of the Apes, the Alien series,....
Instead we would have had movies about feeding starving children. Who the hell wants to see movies about fields of wheat?
One thing that bothers me about new interface ideas is that they suck screen space without providing more information. XP is a great example of that. More flash with less utility. I love flash also, but I want something that gives me lots of useful info in an easy to digest form, with less scrolling and button clicks.
The better the interface is, the more likely I could use the same thing on a PDA, cellphone, or laptop instead of only on a 17 inch display workstation.
3-D perspective displays could provide more information density if they are able to provide shrunken 3-D like views that allowed you to fit more onto the screen. From the sample images, it does look like there is some screen space savings, by showing windows vertically rotated. But lots of space is wasted elsewhere. As a work in progress, it doesn't look like it would be a more productive environment yet.
This may not be practical for general traffic, but I could see it being very useful for places where one can control a fleet of individual cooperative vehicles. This could be on a factory floor with robotic delivery vehicles (e.g. in an Amazon-type warehouse), baggage haulers on airport runways, at airports with the airplanes themselves to get to runways, construction sites with heavy machinery, companies with fleets of similar vehicles like at UPS, FedEx, Walmart, military sites with tanks and humvees (using encrypted channels of course), etc.
There are lots of places where you have a need for traffic control with big or many vehicles, in tight spaces. Such resource allocation is a huge part of many problems. That's where they should market this first, I think.
I stand corrected, I hope. I vaguely recall what I read in the Oddpost FAQ about why it required IE 5. If the needed functionality in already built into standard Mozilla/et al, then great. It should be an easy (or not difficult) port then.
Yahoo/Oddpost could try to make it work on Mozilla/Firefox, etc. but I think that it requires some particular extensions built into IE 5+, to do stuff relating to SOAP and drag/drop, I think. At a minimum, some mechanism is required to talk to the server to avoid doing full web page refreshes.
In other words, porting it is not simply a matter of porting to a different dialect of javascript, CSS, and the DOM.
Therefore, Mozilla/Firefox, should have an extension and plugin that provides the same functionality required by Oddpost. Afterall, Mozilla users have already gone through the trouble of installing a foreign browser, so installing some good extensions is no big deal. Since Yahoo is very widely used, these nonstandard extensions would be very widely applicable.
The required functionality could probably be done using a java applet running invisibly in the browser whose sole purpose is to communicate with the mail servers. But this requires launching the java VM which is heavy. That's why a lightweight extension that mimics the needed IE 5+ functionality might be preferable.
I totally agree. I still believe that there is intelligent life out there, but we'll never find it.
For a universe of 15 billions years in age, intelligent life on earth is a super tiny fraction of that. That's probably the same (order of magnitude) for other solar systems. Statistically speaking, we'll probably have better luck winning the Lotto while being simultaneously struck by lightning, than we will finding radio signals from life forms that overlap temporally with humanity's lifespan and are near enough to detect.
In the mean time, we could be folding proteins to help cure SARS, modelling global warming effects, run fusion simulations or other kinda more useful things.
You don't have to use the keypad. A cellphone can be equipped with an accelerometer which can be used to sense movement and changes in orientation of the cell phone. That could let you twist and tilt the phone to get a different view of a scene. The cell phone could therefore be equipped to know where it is with GPS and how it is oriented and moved, using motion detectors, a compass, and accelerometer.
I imagine that a store or restaurant could supply a 3-D view of their storefront with their online yellow pages entry. Your GPS-enabled cell phone would find the store you want, provide a 2-D of it's location and display a 3-D model of the street scene and store front; so your cell phone or mobile computer could show you exactly what your destination looks like.
This may be overkill for finding the nearest Taco Bell, but the Postal Service, EMS and ambulances, FedEx, social workers, car rental agencies, the tourism industry, etc., and of course, police and military, could definately benefit from having these gadgets for customers/patrons.
I remember hearing something about how buckeyballs, or some nano-structure, absorbs photons very easily and that it ignites like flash powder. But I don't remember any details.
Anybody know what a mass of nanotubes looks like? And buckeyballs? Soot, which is black, contains lots of buckeyballs I think. And diamonds are colorless. So how would the nanotube structure affect the color?
Sue the PTO for incompetence?, negligence? dereliction of duty? How about restraint of trade? Criminal neglect? There's got to be something legalistic way to get the patent reviewers fired or jailed.
Hell, if Martha goes to jail for trying to cover her tracks to save less than 100K, then surely lots of heads should roll at the PTO. Their incompetence costs the high tech industries probably billions of dollars in unnecessary licensing for trivial patents and for legal fees.
A number of posts have suggested suing the PTO. But my question is for what and how?
I just noticed your reply, many days later. Anyway, I got the 50+ years part wrong; it should have been 40+ years from my fingertips. From what I remember of my computer language and AI classes, Fortran and Lisp were the first or among the first languages and I thought they started in the 60s. And Fortran compilers are supposed to be among the best at advanced optimization.
But basically I meant that compilers have been ahead-of-time for many decades and their writers have developed non-local optimization techniques; non-local in that they use inter-procedure analysis such as finding code without useful side-effects or return values and therefore enabling further optimization. I don't think the early pcode (bytecode) interpreters had runtime compiling or optimization, but my memory is foggy about it.
Thanks for the link. I trust Technology Review a lot more than Wired for deeper research news. (Although it is run by MIT so I wonder about the journalistic impartiality.) Anyway, I wasn't necessarily picking on the Media Lab. For those who are interested, their research page lists their highly diverse set of projects.
The entire Media Lab seems to follow that same pattern of pursuing fluffy PR-friendly pseudo-science. Wired had this to say about it: The Lab that Fell to Earth. (It's an ironic criticique given that Wired is very fluffy tech news.)
Contrast that to the MIT AI and CS Lab, which does and has done outstanding work, in hard AI, theory, robotics, vision, and so on.
Still, the Media Lab just seems like the most fun place to work.
The C# way (and the way some java tools worked before the JIT VM) is that byte-code is distributed but the client-side machine compiles it ahead of time and saves the compiled code. There is no difference to the sandbox security, with respect to ahead-of-time compilation, as long as the cached code on disk is secure. JIT VM always compile it as the code runs and then discards the compiled byte-code.
So basically what I want is that the compiled code should be cached and stored on disk, e.g. like a browser loading cached pages when they are still up to date. Not everything needs to be cached, since profiling (that is behind the current Hotspot technology) could be used to identify those parts of the code that should be aggressively optimized. So those critical areas should automatically be aggressively optimized, which takes time and that should be cached. That's what I would ideally like to see in java.
Also, as for unsafe constructs, I've read about them in C# but haven't developed with it. However, I have used the java native interface. It is really ugly and very cumbersome and intrinsically system-dependent. Sometimes it is necessary to use JNI for performance and for low-level interfaces to the machine. I want something that's low-level but easier to use than JNI for those rare occasions where you've gotta have it.
Yeah, I expect that an open source project would break things much more. Eclipse breaks the plugin.xml format subtley everytime. EMacs libraries always have bad interaction. Mozilla's XUL API has given me headaches, and is horribly, horribly, horribly documented.
The big plus side to open sourcing is perhaps the language could be forced to match the nice features of C#, like unsafe constructs and precompilation, both for performance reasons. There's only so much JIT optimization you can do. But precompiling (like GCJ, but intrinsic to the VM) would provide greater opportunities for large scale full source tree optimizations. Compiler writers have been doing this stuff for 50+ years.
Anybody who knows they have a great new algorithm (PhD or not), does not need to join a company to implement it. Why should they join a company and merely become millionares, when they could become billionares. Brin and Page are probably good engineers, but not business geniuses. They found adult supervision to manage the business. The inventor of the next better mousetrap can do the same thing. If they wan't babes, bling-bling, or research dollars, you can do lots more with a billion than a million.
If there are targetted ads attached to email, then advertisers or anybody else could pay for certain advertising keywords just to see if people are writing email about it. Some small percentage of them will click on the ad and then the ad sponsor will know something about the contents of your email. Say Microsoft or SCO wants to see if people are talking about linux or about switching to linux. They could bid for keywords like "linux" or "linux AND switch OR switching" (though I don't know if booleans are used in the ad selection). So some curious email readers will click on the ad and thereby will give MS/SCO a quantitive statistical sampling of people talking about switching to linux (or at least using one of the keywords).
Extrapolate this to any words that somebody would be willing to pay to watch, regarding politics, religion, cults, music, or whatever other creepy corners your paranoia guides you to.
The important difference between targetting ads to web pages vs email is that web pages are designed for wide publication. The contents of email is usually meant to be private.
Hypothetically, if web pages were identified with an 8-bit code of 0x01 along with a 32-bit identifier, then one could just assign another code to signify web pages. e.g. codes 0x00-0x7f could be web page codes, 0x80 for PDFs, 0x81 for Gifs, etc. Each code would be combined with a 32-bit int identifier that is unique relative to that code, giving a 40-bit identifier space.
As for the space required, they must have gone to beyond 32-bits for on-disk identifiers. URL's and cached pages easily take a lot more space than a 5-byte to 8-byte (64-bits) identifier, so they've definately got the storage. For archival purposes, 64-bits is ample space and small.
But a good reason to keep identifier sizes small is so that they don't take up much RAM space. That's why variable sized IDs would be useful. They are a simple fast form of compression. UTF8 is a variable sized encoding that uses 8 bits to encode the vast majority of characters used in English (ASCII) and uses between 2-bytes and 4-bytes for other less common character codes (symbols and other language characters). This is done by using the top 2-bits of the first byte to indicate how large that variable-sized character is. (I don't remember the details, however.) The effect is that on average for English, most strings would consume slightly more than 8 bits per character.
The same principle would work for any variable sized identifier, e.g. useful for DOC Ids or word/term ids. The most common web pages (yahoo, hotmail, msn, nytimes, etc) would have very high page rank and could be given small ids', eg. 16-bits (2-bit code, 14-bit id). Same thing for frequent words, "whether", "while", "with", "over", or closed-class words. Compress them to small ids.
Anyway the point is that you could have an effective id space of much greater than 32 bits and yet use much less than 32-bits per identifier on average. Every search engine must have dispensed with the 32-bit barrier by their beta phase, unless they're run by idiots. Maybe that's Microsoft's problem.
Slashdot Mirror
I was hoping that it was using SVG maybe, even though it's not standard yet. If a portal used SVG for something widespread like a mapping service, then all the browsers would be forced to make it standard, hopefully.
If your motivation is to do great work, and you're already rich, then you might as well just teach and do research at a leisurely pace, armed with a bunch of eager (and poor) grad students. In the research world you can make a long-term, pervasive impact without worrying about giving your competitors ammunition to kill you, as you would by publishing your research from inside a company.
Basically, I think that keeping the really good employees is always a problem, whether you give them a little or a lot of money.
There are certainly hundreds of cases of prior art, and Tripwire is probably one of them. It computes and maintains a database of hashes for all the files on a file system to check for intrusions and corruption. The wiki entry says it first surfaced in 1992.
I have a paper published by ACM Press and it is not available. The link goes to the publisher site where you can subscribe for access.
The US could complain to the WTO or somebody, but they are toothless. China is too big to start a trade war with.
Poland just recently decided against supporting software patents in the EU. Does that mean they will not respect other countries' patents on software or just that they will not go along with Europe issuing them?
What would be better is a high brightness array of LED lights (white of course) to use as the white light source. Some DIY forums have noted that the problem is focusing the light from the array into a controllable direction and even intensity. This is hard because you need so many LEDs to get enough brightness. If this could be cheaply solved, then it would make a great DIY projector.
Anybody have any luck with this or know any good websites for it?
3-D perspective displays could provide more information density if they are able to provide shrunken 3-D like views that allowed you to fit more onto the screen. From the sample images, it does look like there is some screen space savings, by showing windows vertically rotated. But lots of space is wasted elsewhere. As a work in progress, it doesn't look like it would be a more productive environment yet.
That's a cool idea. Add flashing lights all around the bumper cars, with screaming sound effects, just to get people in the mood.
There are lots of places where you have a need for traffic control with big or many vehicles, in tight spaces. Such resource allocation is a huge part of many problems. That's where they should market this first, I think.
I stand corrected, I hope. I vaguely recall what I read in the Oddpost FAQ about why it required IE 5. If the needed functionality in already built into standard Mozilla/et al, then great. It should be an easy (or not difficult) port then.
In other words, porting it is not simply a matter of porting to a different dialect of javascript, CSS, and the DOM.
Therefore, Mozilla/Firefox, should have an extension and plugin that provides the same functionality required by Oddpost. Afterall, Mozilla users have already gone through the trouble of installing a foreign browser, so installing some good extensions is no big deal. Since Yahoo is very widely used, these nonstandard extensions would be very widely applicable.
The required functionality could probably be done using a java applet running invisibly in the browser whose sole purpose is to communicate with the mail servers. But this requires launching the java VM which is heavy. That's why a lightweight extension that mimics the needed IE 5+ functionality might be preferable.
In the mean time, we could be folding proteins to help cure SARS, modelling global warming effects, run fusion simulations or other kinda more useful things.
I imagine that a store or restaurant could supply a 3-D view of their storefront with their online yellow pages entry. Your GPS-enabled cell phone would find the store you want, provide a 2-D of it's location and display a 3-D model of the street scene and store front; so your cell phone or mobile computer could show you exactly what your destination looks like.
This may be overkill for finding the nearest Taco Bell, but the Postal Service, EMS and ambulances, FedEx, social workers, car rental agencies, the tourism industry, etc., and of course, police and military, could definately benefit from having these gadgets for customers/patrons.
Anybody know what a mass of nanotubes looks like? And buckeyballs? Soot, which is black, contains lots of buckeyballs I think. And diamonds are colorless. So how would the nanotube structure affect the color?
Hell, if Martha goes to jail for trying to cover her tracks to save less than 100K, then surely lots of heads should roll at the PTO. Their incompetence costs the high tech industries probably billions of dollars in unnecessary licensing for trivial patents and for legal fees.
A number of posts have suggested suing the PTO. But my question is for what and how?
But basically I meant that compilers have been ahead-of-time for many decades and their writers have developed non-local optimization techniques; non-local in that they use inter-procedure analysis such as finding code without useful side-effects or return values and therefore enabling further optimization. I don't think the early pcode (bytecode) interpreters had runtime compiling or optimization, but my memory is foggy about it.
Contrast that to the MIT AI and CS Lab, which does and has done outstanding work, in hard AI, theory, robotics, vision, and so on.
Still, the Media Lab just seems like the most fun place to work.
So basically what I want is that the compiled code should be cached and stored on disk, e.g. like a browser loading cached pages when they are still up to date. Not everything needs to be cached, since profiling (that is behind the current Hotspot technology) could be used to identify those parts of the code that should be aggressively optimized. So those critical areas should automatically be aggressively optimized, which takes time and that should be cached. That's what I would ideally like to see in java.
Also, as for unsafe constructs, I've read about them in C# but haven't developed with it. However, I have used the java native interface. It is really ugly and very cumbersome and intrinsically system-dependent. Sometimes it is necessary to use JNI for performance and for low-level interfaces to the machine. I want something that's low-level but easier to use than JNI for those rare occasions where you've gotta have it.
The big plus side to open sourcing is perhaps the language could be forced to match the nice features of C#, like unsafe constructs and precompilation, both for performance reasons. There's only so much JIT optimization you can do. But precompiling (like GCJ, but intrinsic to the VM) would provide greater opportunities for large scale full source tree optimizations. Compiler writers have been doing this stuff for 50+ years.
Of course, I'm poor and dumb, so what do I know.
Extrapolate this to any words that somebody would be willing to pay to watch, regarding politics, religion, cults, music, or whatever other creepy corners your paranoia guides you to.
The important difference between targetting ads to web pages vs email is that web pages are designed for wide publication. The contents of email is usually meant to be private.
As for the space required, they must have gone to beyond 32-bits for on-disk identifiers. URL's and cached pages easily take a lot more space than a 5-byte to 8-byte (64-bits) identifier, so they've definately got the storage. For archival purposes, 64-bits is ample space and small.
But a good reason to keep identifier sizes small is so that they don't take up much RAM space. That's why variable sized IDs would be useful. They are a simple fast form of compression. UTF8 is a variable sized encoding that uses 8 bits to encode the vast majority of characters used in English (ASCII) and uses between 2-bytes and 4-bytes for other less common character codes (symbols and other language characters). This is done by using the top 2-bits of the first byte to indicate how large that variable-sized character is. (I don't remember the details, however.) The effect is that on average for English, most strings would consume slightly more than 8 bits per character.
The same principle would work for any variable sized identifier, e.g. useful for DOC Ids or word/term ids. The most common web pages (yahoo, hotmail, msn, nytimes, etc) would have very high page rank and could be given small ids', eg. 16-bits (2-bit code, 14-bit id). Same thing for frequent words, "whether", "while", "with", "over", or closed-class words. Compress them to small ids.
Anyway the point is that you could have an effective id space of much greater than 32 bits and yet use much less than 32-bits per identifier on average. Every search engine must have dispensed with the 32-bit barrier by their beta phase, unless they're run by idiots. Maybe that's Microsoft's problem.