.. from a company called Earth Class Mail. They receive your mail, send you an image of the envelope, and let you tell them what to do with it: shred it, recycle it, open and scan contents and send PDF, deposit check, etc. The company was the subject of a sort-of documentary last year.
This kind of prediction/action by Adobe raises a bunch of issues.
One is that, even in 10 years, the Internet may not penetrate everywhere on the planet with sufficient bandwidth to make these applications feasible. So a part of the market (admittedly, probably a small part) is cut off.
More importantly, there are environments where the possibility, let alone the reality, of data leaving the premises is not acceptable. The most obvious example is the (large, and apparently growing) classified world. Will it become impossible to make a classified briefing with the latest version of Adobe's products?
It seems to me that you have to separate out why Ajax is spreading among developers, and why Ajax-based applications are popular with users. These are not totally independent, of course, but worth thinking of in different ways.
I see Ajax-based applications as being very reminiscent of the what used to be called "full-duplex" applications. Unix, because it was based on using teletypes for I/O to the user, and because teletypes were inherently full-duplex, seemed much more interactive, at least with some applications. Nothing quite like Ajax, but a step in that direction. Conventional main-frame apps, based on either half-duplex (I type, then I hit carriage return, and the keyboard locks until the system responds) or electronic versions of that (such as with the famous 3270 displays, which would lose characters if you typed when the system wrote to the screen), were much more... well, boring.
So, it seems to me that, from the user's viewpoint, Ajax can allow the app builder to effectively decouple user input and system output, and make the whole "flow" between system and user be much more continuous, and less synchronized. Another way of seeing this is thinking of an overseas phone call in the days of poor channel allocators, which really made it necessary to stop talking when the other person started, or neither of you would hear the other. Nothing at all like a really engaged, face-to-face, conversation.
Wait a minute! Are you saying that IBM bid a machine based on Woodcrest? If IBM bid anything here, it would have been either a BlueGene, or, perhaps, something like the ASCI machines, which are conventional PowerPCs with a fast interconnect. Hard - no - impossible - to believe that IBM would have bid an Intel processor.
I've been getting 3 or 4 of these a day for at least a month now. The text can always be found in some file of an old book provided by the Gutenberg Project, which is making non-copyright texts available through volunteer effort.
I think the theory about using this stuff to untrain spam filters is very plausible. But it's difficult to see how it will work. There's no common text among these e-mails; in order to send effective spam, there'll have to be at least some text which is the same across multiple mails, and that will tend to expose it.
As many have pointed out, there are power and bandwidth issues around this idea. Not everything that CAN be done is worth doing. This seems like one that isn't worth doing. There was a tiny (less than 256 bytes of code, as I recall!) web server done at least 5 years ago on something like a PIC controller at U of Mass (?). So this doesn't seem very impressive.
What's the advantage to having a web server where there's uncertain connectivity, limited resources generally, and high communication costs?
More interesting would be a stationary web server and an interesting way of updating information on it from the mobile unit while conserving bandwidth and minimizing the effects of intermittent connectivity. So, perhaps I could clip the phone to my shirt pocket and have it send back to the server a photo every 5 minutes. (Of course, if noone ever visits the web site, a solution like this will use MORE power than the server on the phone, if noone ever connects to it...)
I believe MS outlined 7 different versions for different markets... home, enterprise, small business, entertainment center, etc. Why wouldn't they configure the firewall in each of these by default to be what's appropriate for its target market, rather than letting the desires of the Fortune 500 wag my mother's machine in a less than completely safe way? Given the world's recent experience with various forms of malware, erring on the side of safety certainly seems to be justified.
From the very sketchy descriptions, my guess is that this algorithm doesn't really have to do with search per se, but rather with figuring out what the multiple meanings or contexts associated with a term are. So, if I search for "American Revolution" the interesting thing would be to realize that that is a broad and many-faceted topic. So the cool thing for this algorithm to do would be for it to look at the search results which are returned, and then realize that some of the more specific aspects of the results might be:
battles political aspects leadership England and King George III
and so on, and then let you choose which of these was the best fit for what you are looking for, and show you the results which are related to that aspect, and then repeat this process on that subset, so that if you chose "political aspects", it might further offer you
Federalist papers Continental Congress actions etc.
The difficult problem, which perhaps Allon has solved, but as far as I know noone else has, is automating figuring out of these aspects.
Does anyone know of a more substantive description than the rehashed stuff which appears in 100 news sites..? A paper, a patent application, anything? Do we know where Bill Gates learned about it?
Your SSN is visible to many people who ought not really to have it; it's a time bomb. The one that bothers me is every clerk in a doctor's office, where the SSN is present on every insurance form, in most cases. (Some companies have conciously stopped using SSNs, and that's a great idea...).
I'm surprised that noone has, apparently, mentioned the Aspen project, done in the mid 70s by the MIT Architecture Machine Lab group, the precursor to the Media Lab.
The Aspen project photographed the entire city of Aspen, taking a picture every 10 (?) feet on every street, looking left and right, forward and back, during each of 4 seasons, and making most turns. The images were put on a videodisk (12-inch, analog, precursor to DVDs), and software was written which, in conjunction with special video switching hardware and 2 copies of the disk spinning, made it possible to "drive" throughout the city. Aspen was chosen for this project because it's nice and rectilinear, and the size they wanted. The skiing and good food were irrelevant, I'm sure.
Some of the buildings (e.g., city hall) in addition had slide shows of the interior, so you could virtually "enter" the building; in one case there was even a video interview (with the mayor, or some such).
The system ultimately died when the special hardware was destroyed because the owners were sick and tired of demoing it. There's a paper somewhere describing all this. It's not as well known as it deserves to be.
I think the reviewer got his facts a bit screwed up. The thing that saved IBM, after the depression started and it continued manufacturing, was the start of the Social Security System (I think in 1933; 1941 would have been a long time to wait...).
The WW II connection is that IBM turned over its manufacturing plants to the government to make war materiel at a 1% profit. Carbines, gun sights, small cannons, other things, were all made in IBM's plants in Poughkeepsie, Endicott, and elsehwere.
Would someone plese tell me how the RIAA can distinguish between a song which I downloaded from Kazaa or some place like that, and the same song RIPped from a CD I own? I think even the RIAA would agree that the latter is legal. But if it looks like I violated the copyright, then I must have done so?
Does anyone know what the impetus behind these DMCA+ laws is coming from? They seem enough different in intent that I doubt it's the Disney+friends music world. They've already been passed in half a dozen states (including Michigan), and are under consideration in several more. I'm writing my state legislators, but I think the time has come to mount a campaign to roll back the DMCA in its entirety. It's clear (to me, anyway) that it's bad law. We were smart enough to undo prohibition, although it took about 12 years. Maybe we can correct this error more promptly.
Given that modern day law did not specifically anticipate free software, is there any modification (not limited to copyright law) of the law which you would like to see enacted in order to advance the cause of free and open source software?
Disney and others aren't shy about buying the changes they want; why shouldn't we at least ask?
I just yesterday finished reading "Complications", having bought it and started reading it before a long interaction, involving complications in my own case, with the medical establishment. So I can speak from experience, rather than hearsay.
There are good and bad doctors everywhere. I don't know the statistics, and they're probably unknowable. But I do observe that our expectations are raised when we read about the wonders of modern medicine and the peaks of what can be accomplished. That makes the cases in which miracles are not accomplished, or worse, all the more disappointing. My own situations involves a surgeon who didn't react promptly or decisively enough to a complication. He's now off the case, but those who have replaced him have done very well and have been a pleasure to deal with.
Gawande's story makes worthwhile reading for anyone who deals with -- or may have to deal with -- medical professionals. I think that means pretty much everyone, sooner or later. The book will give you an appreciation of what MDs are up against, but also for the wide range of how different doctors DO deal with their patients, disease, and the uncertainties of the field.
Someone suggested SLAC. I'd add to that Fermilab, in Batavia, Illinois, not too far from Chicago. They have a circular ring (4mi circumference, I think), and a buffalo herd to keep the grass short..
American first put this up in May. They sent me some e-mail suggesting I update my preferences at the site, but it turned out primarily to be a ploy to get me to agree to the new agreement. I refused, and wrote to them and told them they were insane to expect people to (a) read (b) understand (c) care and (d) agree to a license which when printed, as I recall, is 14 or 15 pages of fairly small print, and which as people have noted here gives them everything and the user nothing. Here's their helpful answer. I still haven't agreed to it, or flown AA since (and I'm AAdvantage Gold). Note the nonsense about how it's all to protect their user's privacy.
We regret your objections to the Terms and Conditions associated with the use of AA.com. When we launched our new site, we strengthened our Terms and Conditions to ensure the protection of our customers' privacy. When using our site, your privacy is very important to us.
The new verbiage contained in the agreement is certainly not meant to alarm our customers or dissuade anyone from using our site. The simple issue of securing our customers' interest and privacy effectuated the enhancement of our Terms and Conditions on AA.com. However, we are concerned by your apprehension and have forwarded your comments to our managers at AA.com for review.
We will continue to make every effort to ensure that our customers receive a convenient, personalized Web experience on AA.com. Thank you for contacting us.
.. and it's getting worse. I recently logged onto the American Airlines site. They asked me to agree to a usage agreement which, in printed form, is 5 pages long...of pretty small type.
A lot of it has to do with what use I'm permitted to make of the site.. like not using "mileage aggregation services" and the like. But life is too short, and I've just abandoned my use of the site. But I hope this isn't a sign of worse to come.
Seems to me it's pretty clear what a DRM OS is. I think all the chest beating in this discussion above is really pretty hollow. We certainly know how to build such a system technically, and I'd assert that we know how to build it such that the cost of defeating it is pretty large, although we may not be able to do that for an appropriate cost. That's because it's a matter not only of software, but of hardware, packaging, etc.
What I can't understand is how they got the patent. Pretty much everything they describe is well known prior art from the area of multi-level secure operating systems. Those systems have (had? not sure if any multi-user systems of this ilk are still in use) the ability to authenticate multiple users, store multiple classes of information, expose information only to those users who were authorized by policy to see them, differentiated between the right to view and the right to print, prevented copying from one level of secure environment to a lesser (or different) level, and so on. What the MS patent claims is pretty much the same thing.
The Polish Bombes are long gone... But you can see one of the American ones at the National Cryptologic Museum, a stone's throw from NSA at Fort Meade. And of course, there's the Colossus rebuild at Bletchley Park, well worth the trip.
Slashdot Mirror
.. from a company called Earth Class Mail. They receive your mail, send you an image of the envelope, and let you tell them what to do with it: shred it, recycle it, open and scan contents and send PDF, deposit check, etc. The company was the subject of a sort-of documentary last year.
This kind of prediction/action by Adobe raises a bunch of issues.
One is that, even in 10 years, the Internet may not penetrate everywhere on
the planet with sufficient bandwidth to make these applications feasible. So
a part of the market (admittedly, probably a small part) is cut off.
More importantly, there are environments where the possibility, let alone the
reality, of data leaving the premises is not acceptable. The most obvious example
is the (large, and apparently growing) classified world. Will it become impossible
to make a classified briefing with the latest version of Adobe's products?
.. not yet a patent. Look for it as a patent in 2-3 years. Maybe never.
It seems to me that you have to separate out why Ajax is spreading among developers, and why Ajax-based applications are popular with users. These are not totally independent, of course, but worth thinking of in different ways.
... well, boring.
I see Ajax-based applications as being very reminiscent of the what used to be called "full-duplex" applications. Unix, because it was based on using teletypes for I/O to the user, and because teletypes were inherently full-duplex, seemed much more interactive, at least with some applications. Nothing quite like Ajax, but a step in that direction. Conventional main-frame apps, based on either half-duplex (I type, then I hit carriage return, and the keyboard locks until the system responds) or electronic versions of that (such as with the famous 3270 displays, which would lose characters if you typed when the system wrote to the screen), were much more
So, it seems to me that, from the user's viewpoint, Ajax can allow the app builder to effectively decouple user input and system output, and make the whole "flow" between system and user be much more continuous, and less synchronized. Another way of seeing this is thinking of an overseas phone call in the days of poor channel allocators, which really made it necessary to stop talking when the other person started, or neither of you would hear the other. Nothing at all like a really engaged, face-to-face, conversation.
Wait a minute! Are you saying that IBM bid a machine based on Woodcrest? If IBM bid
anything here, it would have been either a BlueGene, or, perhaps, something like the
ASCI machines, which are conventional PowerPCs with a fast interconnect. Hard - no - impossible -
to believe that IBM would have bid an Intel processor.
Does anyone know who else bid on this contract? Was BlueGene a contender?
...
It would be interesting to know the other bids and their performance
I've been getting 3 or 4 of these a day for at least a month now. The text can
always be found in some file of an old book provided by the Gutenberg
Project, which is making non-copyright texts available through volunteer
effort.
I think the theory about using this stuff to untrain spam filters is very plausible.
But it's difficult to see how it will work. There's no common text among these
e-mails; in order to send effective spam, there'll have to be at least some text which
is the same across multiple mails, and that will tend to expose it.
As many have pointed out, there are power and bandwidth issues around this idea. Not everything that CAN be done is worth doing. This seems like one that isn't worth doing. There was a tiny (less than 256 bytes of code, as I recall!) web server done at least 5 years ago on something like a PIC controller at U of Mass (?). So this doesn't seem very impressive.
What's the advantage to having a web server where there's uncertain connectivity, limited resources generally, and high communication costs?
More interesting would be a stationary web server and an interesting way of updating information on it from the mobile unit while conserving bandwidth and minimizing the effects of intermittent connectivity. So, perhaps I could clip the phone to my shirt pocket and have it send back to the server a photo every 5 minutes. (Of course, if noone ever visits the web site, a solution like this will use MORE power than the server on the phone, if noone ever connects to it...)
I believe MS outlined 7 different versions for different markets... home, enterprise, small business, entertainment center, etc. Why wouldn't they configure the firewall in each of these by default to be what's appropriate for
its target market, rather than letting the desires of the Fortune 500 wag my
mother's machine in a less than completely safe way? Given the world's recent
experience with various forms of malware, erring on the side of safety certainly seems to be justified.
From the very sketchy descriptions, my guess is that this algorithm doesn't really have to do with search per se, but rather with figuring out what the multiple meanings or contexts associated with a term are. So, if I search for "American Revolution" the interesting thing would be to realize that that is a broad and many-faceted topic. So the cool thing for this algorithm to do would be for it to look at the search results which are returned, and then realize that some of the more specific aspects of the results might be:
battles political aspects leadership England and King George III
and so on, and then let you choose which of these was the best fit for what you are looking for, and show you the results which are related to that aspect, and then repeat this process on that subset, so that if you chose "political aspects", it might further offer you
Federalist papers Continental Congress actions etc.
The difficult problem, which perhaps Allon has solved, but as far as I know noone else has, is automating figuring out of these aspects.
Does anyone know of a more substantive description than the rehashed stuff which appears in 100 news sites..? A paper, a patent application, anything? Do we know where Bill Gates learned about it?
Surely there should be some real experience to report by now, rather than just a press release from the manufacturer...
Your SSN is visible to many people who ought not really to have it; it's a time bomb. The one that bothers me is every clerk in a doctor's office, where the SSN is present on every insurance form, in most cases. (Some companies have conciously stopped using SSNs, and that's a great idea...).
What is the incidence of well-meaning but misinformed people introducing incorrect information? Do you make any attempt to track this?
Related, what is the incidence of what appears to be intentional sabotage by introducing incorrect information? Can you distinguish?
I'm surprised that noone has, apparently, mentioned the Aspen project, done in the mid 70s by the MIT Architecture Machine Lab group, the precursor to the Media Lab.
The Aspen project photographed the entire city of Aspen, taking a picture every 10 (?) feet on every street, looking left and right, forward and back, during each of 4 seasons, and making most turns. The images were put on a videodisk (12-inch, analog, precursor to DVDs), and software was written which, in conjunction with special video switching hardware and 2 copies of the disk spinning, made it possible to "drive" throughout the city. Aspen was chosen for this project because it's nice and rectilinear, and the size they wanted. The skiing and good food were irrelevant, I'm sure.
Some of the buildings (e.g., city hall) in addition had slide shows of the interior, so you could virtually "enter" the building; in one case there was even a video interview (with the mayor, or some such).
The system ultimately died when the special hardware was destroyed because the owners were sick and tired of demoing it. There's a paper somewhere describing all this. It's not as well known as it deserves to be.
I think the reviewer got his facts a bit screwed up. The thing that saved IBM, after the depression started and it continued manufacturing, was the start of the Social Security System (I think in 1933; 1941 would have been a long time to wait...).
The WW II connection is that IBM turned over its manufacturing plants to the government to make war materiel at a 1% profit. Carbines, gun sights, small cannons, other things, were all made in IBM's plants in Poughkeepsie, Endicott, and elsehwere.
If you could change one thing about copyright or patent law in the United States, what would it be?
Would someone plese tell me how the RIAA can distinguish between a song which I downloaded from Kazaa or some place like that, and the same song RIPped from a CD I own? I think even the RIAA would agree that the latter is legal. But if it looks like I violated the copyright, then I must have done so?
Does anyone know what the impetus behind these DMCA+ laws is coming from? They seem enough different in intent that I doubt it's the Disney+friends music world. They've already been passed in half a dozen states (including Michigan), and are under consideration in several more. I'm writing my state legislators, but I think the time has come to mount a campaign to roll back the DMCA in its entirety. It's clear (to me, anyway) that it's bad law. We were smart enough to undo prohibition, although it took about 12 years. Maybe we can correct this error more promptly.
Given that modern day law did not specifically anticipate free software, is there any modification (not limited to copyright law) of the law which you would like to see enacted in order to advance the cause of free and open source software?
Disney and others aren't shy about buying the changes they want; why shouldn't we at least ask?
I just yesterday finished reading "Complications", having bought it and started reading it before a long interaction, involving complications in my own case, with the medical establishment. So I can speak from experience, rather than hearsay.
There are good and bad doctors everywhere. I don't know the statistics, and they're probably unknowable. But I do observe that our expectations are raised when we read about the wonders of modern medicine and the peaks of what can be accomplished. That makes the cases in which miracles are not accomplished, or worse, all the more disappointing. My own situations involves a surgeon who didn't react promptly or decisively enough to a complication. He's now off the case, but those who have replaced him have done very well and have been a pleasure to deal with.
Gawande's story makes worthwhile reading for anyone who deals with -- or may have to deal with -- medical professionals. I think that means pretty much everyone, sooner or later. The book will give you an appreciation of what MDs are up against, but also for the wide range of how different doctors DO deal with their patients, disease, and the uncertainties of the field.
Check out http://www.atomictourist.com
Someone suggested SLAC. I'd add to that Fermilab, in Batavia, Illinois, not too far from Chicago. They have a circular ring (4mi circumference, I think), and a buffalo herd to keep the grass short..
.. and it's getting worse. I recently logged onto the American Airlines site. They asked me to agree to a usage agreement which, in printed form, is 5 pages long...of pretty small type.
A lot of it has to do with what use I'm permitted to make of the site.. like not using "mileage aggregation services" and the like. But life is too short, and I've just abandoned my use of the site. But I hope this isn't a sign of worse to come.
What I can't understand is how they got the patent. Pretty much everything they describe is well known prior art from the area of multi-level secure operating systems. Those systems have (had? not sure if any multi-user systems of this ilk are still in use) the ability to authenticate multiple users, store multiple classes of information, expose information only to those users who were authorized by policy to see them, differentiated between the right to view and the right to print, prevented copying from one level of secure environment to a lesser (or different) level, and so on. What the MS patent claims is pretty much the same thing.
The Polish Bombes are long gone... But you can see one of the American ones at the National Cryptologic Museum, a stone's throw from NSA at Fort Meade. And of course, there's the Colossus rebuild at Bletchley Park, well worth the trip.