If you are a physicist then I would honestly like to know why my above post is so terrible. Seems to extrapolate in brave asses hangin' off the cliff Slasdot style. Since I've only been able to read Scientific American and Nature articles about this, and only had up to organic chemistry in college I certainly don't have a background in the field but do have a thirst for knowledge. Please respond with valid information, anonymous coward or not.
I got a question, hope somebody can poke holes in it. Say you have an unlimited number of states, and we ignore problems with how much energy or time you might need to get to that resolution.
Suppose you encode everything about a computer, its RAM contents and operating rules into a piece of data, basically a long number. Say you are doing something like dumping the VMWare PC emulator program and its memory buffers into this piece of data, along with your own program and also a bunch of other locations which are telltale bits that would only be set to true if certain instructions (your program) are executed in a certain order, so you can in a sense freeze a sequence of calculations, an overall machine state.
So in the end the last telltale will finally be set only if the results of the calculation which suposedly had been executed by this hypothetical (virtual) computer, was provably the answer you seek, i.e. the factors of a big prime number which could be multiplied together to show they are the right answer. A self-referential logic filter.
My conjecture (gleefully made without more knowledge of quantum physics than is available in lay publications..) is this. Could you use this huge number as a filter or reference beam to collapse the waveform of your recording medium, and read out the state of the virtual computer with the output of your program, in a picosecond?
It would seem that any Turing machine from a Cray to a ribosome (an rna tape device), could be simulated in this way, though smaller memory footprint/instruction set machines would be easier since they could be represented with less eigenstates. I wonder how many states would be the least amount necessary to simulate something useful.. if a full hardware abstraction is not needed and you can get away with just a language definition and virtual machine (yes like Java VM).
Would this mean you could run any program that can fit into the virtual machine in picosecond time? And if so, could you not in fact build a computer of any arbitrary capabilities by simply writing a pseudocode definition of how it ought to work? Final scary question.. The interior of a cell is a controlled environment and until the cell is queried by some process it is conceivable that some ribosomes could exist in superimposed states. Put another way, if you could solve the isolation problem it might end up to be cheaper to build the eigenstate computer with common cellular apparatus than by using big expensive lasers. What conclusions can you draw from this?
I think this is what was meant by a prediction I once came across.. that the coming century would create a new science of computing which is to today's computers as nuclear energy is to fire.
Like I said I hope someone can poke holes in this. The biggest problem seems to be universal laws about information, for example I understand that the recent sending of a light pulse at 300 times the ordinary speed of light was only possible because the leading edge of the pulse had enough information to reconstruct the rest of the pulse, suggesting that you could not send an entire packet of bits faster than the speed of light.
Thought I'd add this thought I had since other posts here seem related.
It seems the terrain in this battle could be altered significantly if server vulnerabilities were used to insert a file (mpaa.html or decss.html) into the root directory of a large number of servers across the net. (Alternately an otherwise invisible text link or transparent gif link in a bottom corner of the page might work as well, to find mirrors, like the PI character in the movie the Net). The html file could contain source and multiple executables of the DeCSS code, and this file could be given a variety of related names or placed in subfolders, stegographed into gifs, inserted into Javascript, encrypted with a public key, etc., so that it is less detectable by administrators and mpaa search engine investigators while promising an unending morph of target and terrain.
Maybe if the MPAA ever gives up the server admins would be notified of the security holes they have.. but I am thinking that admins might even intentionally add this to their own systems since they could say it was placed there by a hacker. Leaving a security hole open intentionally might be a plausible defense in that case.
This thought came to me when I was considering a perl script published in a well known network security secrets book. A friend inexperienced in perl or hacking tried offhand to test his clients' NT servers and suddenly was granted System permissions three times out of three, score 3 for linux. This seems a far more interesting game for script kiddies too, why bother defacing some clueless government's page when they could screw the MPAA instead. There is also less damage on the part of those defaced.
Some more on Ruby - since this is probably the answer to what a Perl type language would look like if written by Japanese speaking programmers.
The big draw of Ruby as I understand it is that it is very easy to write network utilities, is highly object oriented, and programs are short and sweet. The network module has gotten much stronger, the initial weakness of which kept me away in the beginning (and of course I am more comfortable with Perl). Every string for example is an object with built in methods you can call after the dot. It also works with GTK which might interest Slashdot readers. The Brave GNU World article (translated itself into several languages) mentions it just before covering a Bonobo component.. The article does mention that Ruby was designed to be a successor of Perl, and borrows from Perl, Python, Lisp and Smalltalk, and Eiffel, and also runs on a number of platforms. Hence the name Ruby, which makes sense if you read Perl as a pearl.. also brings up another good point that there is a ruby Apache module, and that ruby supports multithreading even on MS-Dos?!?! This I did not know!
A recent Japanese magazine I bought introduced web programming by using Ruby exclusively, going through the RFC documents for HTTP, and building protocol sniffers. Fun stuff! Ruby people are looking for help in spreading the language around the world so this might be more reason to check it out. In fact that magazine (OpenDesign, in Japanese, by CQ Shuppan) was entirely about that and also I-Mode telephone protocols and programming.
Brave GNU World article 8/14/00:
http://www.gnu.org/brave-gnu-world/issue-18.en.h tml
English homepages:
http://www.ruby-lang.org/en/index.html
http://ruby-lang.jaist.ac.jp:8000/en/doc.html
http://kt-www.jaist.ac.jp:8000/~ttate/lang/ruby/
I am a programmer in Tokyo and think much of the richness and draw of Perl is lost on young Japanese programmers. Advanced programmers are just that, there is no problem, and they read English well. Advanced languages and AI apps make more use of language, but it seems that new paradigms are not necessarily coming from language but from study of semantics and dynamics, which are universal. Some might take off better in other countries, for example toontalk is easy to undestand by Japanese.
Some of the draws of Perl are "There Is Always More Than One Way To Do It" and of course the amazing CPAN multiplexed archive and module search/installation system. TIAMTOWTDI doesn't just mean writing verbs in a different order. Where it comes from flexible thinking English may help but certainly has no monopoly.. but there is also more vocabulary and more ways to say something in English than in other languages, which may affect programming.
Of course people who don't read English well are either unaware of the CPAN system or have difficulty making use of it.
One response may have been the development of Ruby in Japan, which borrows from several different languages including Perl, but is at first seems to have confusing operators and syntax. It is highly object oriented, looking more like C or ASP than Perl, though it is meant to be used as Perl and relies heavily on external modules of which there are few.
One good landmark is legal writing in Japan. When well done it is much like a program, and does not suffer much of the archaic nature of American legal writing. It is crisp, and when well written rings like a bell, like most well written Japanese text (style is why it is hard to write well, the characters are not so bad).
Perl might not have been developed at all, or developed in the same way, if it had been done by Japanese speakers or by people brought up in Japanese culture.. but there is no problem with qualityamong serious programmers, the main loss is the fun English readers get.. like understanding the meaning of "State of the Onion" and the hilarity of jolly code writing which uses terms like bless or die in semi poetic context.
The spring action is very cool, as is their means of testing it and monitoring wear and tear (though
not sor sure 20 times is enough to be sure).
They did not invent the bearing idea. Because nanoscale fabrication is still in its beginnings, commercial concerns are busy engineering films of nanoscale particles. One big player is Sharp. A nanoengineering scientist who has his own startup described nanoscale bearing manufacture as a grail many have been seeking, of course he had his own angle on it. The first profitable use is expected to be in lubrication of disk drive surfaces, so that the head actually rests on the disk. I understood the idea to be rolling tubes rather than round spheres as bearings.
There is also a big danger, in that these particles will get into anything including you and me, apparently being quite lethal.
I also know a company which makes machines to sinter fine powder into 3d objects with lasers. The powder is very fine on microscale (steel and soon titanium, but best quality is achieved with less conductive nylon). These guys were pulling so many all nighters, on such a small budget, that in developing the next material they often did not use breathing masks but this is probably the limit to how small a particle you want to have around you without major protection.
Some people may remember the Source, a Prime computer running before public Internet access. It had a bad natural language parsing system for the perennial dungeon adventure but the text descriptions were fantastic. You might have something like it on your linux box, type "adventure".
Infocom had an excellent parsing system that dealt with ambiguity and was very impressive in the adventure context, starting with Zork on the Apple II. They sold an entire series of games based around this parser so it must have been sufficient for the task at hand.
A company called Picture Network Interactive (PNI) built an online photo rental agency which they tried to use to take over the industry (I was on the receiving side in Japan and heard the presentation). Supposedly they used a natural language parsing system to describe photos based on something they (or someone they bought out) used to build a database application for the White House. Supposedly it could handle English, Romance languages, and even Japanese (that was probably vapor), this was sometime around 1993 I believe.
Infoseek's search engine does some language parsing, this was one of their two original selling points (the other being that they would sell you a keyword for targeted ads).. They told me at the time that the company they bought this technology off was called HNC (?) apparently a big IPO the previous year, which means around 1995 or thereafter I believe. Now that I think about it, HNC could have been involved with PNI not Infoseek oh well.
Muse Technologies' synthetic environment for data visualization software provides voice command support but this is for very limited grammar and vocabulary even if it is a Sandia spinoff (musetech.com).
Translation of spoken language, through the phone or via consumer items, has been promised for a while. AT&T has a very nice system. It seems that much is possible if you limit the topic, grammar, and vocabulary.
But simple grammar should not be so hard, and interactive q&a can resolve ambiguity if the system recognizes it.. something between configure.sh and clippy. Actually I wouldn't mind clippy as much if it had broader knowledge.
The UMap analyzer for web queries is intense and combines text analysis and production of gorgeous clickable, customizable live maps. The site is being changed but a quick look at a screenshot is at http://www.kartograph.com/uk/index1024.htm (umap.com had a downloadable program mere months ago, still living on my harddrive.) Very very nice. Kartograph is for lotus notes. It isn't parsing so much as analyzing correspondences between the meanings and statistics of word appearances, I think.
I came across a morpohological analyzer recently but can't find the link.. maybe look in google.com, here are two links I found there.
Here is a link to a bunch of natural language tools. http://web.syr.edu/~mdtaffet/nlp_sites_for_instr uctors.html
Maybe this is useful for snlp.. http://www.rxrc.xerox.com/research/mltt/fsnlp/mo rph.html
Would certainly like to find a perl module or c source code for an snlp facility!
I have seen posters of the English version of Mononoke Hime in Japan, and think it either was played here or will be here on tapes. (someone else please provide better info).
Is it possible that the dubbed version could be available here in Japan on DVD without the Japanese version being available?
If a major film is released on DVD must the security be able to handle any forseeable development of deCss style systems in the medium term?
A recent Scientific American article by the discoverer of the fastest possible quantum computer search algorithm describes the issues well. It could have been Shor.
Unfortunately I have neither the magazine nor yet able to read the above article. Apparently the inventor had proven that his own algorithm was the fastest possible one, though the definition of searching may have been stretched to fit the probabilistic algorithm now in the news.
I do not remember if the basic idea was to assign one qubit to each letter of a name, or just one qubit to each item in a database, but it seemed quite powerful even with a small number of qubits.
A more interesting question in the short term might be whether a <=15 qubit computer can make serious inroads on big factoring problems.
This is just the latest salvo in a year long campaign by the government and individuals who want to jumpstart the economy with Internet technology.
I have discussed several times the issue with Jean-Paul Smets, a representative of the French ministry of industry in Lorraine, who is working with the secretary for information technology under the prime minister. A number of linux software and hardware companies are starting up and one unit (the linbox, based on SuSE linux) with vmware and windows and macos installed is available preinstalled for individual or large clusters.
This was installed last year in a local school and a minister I believe was in attendance. Surely exhaustion with the NSA and Microsoft may have something to do with it but there is also the economy, lack of performance relative to Silicon Valley (well maybe not as bad after the market tanked), and the French idea of freedom.. plus the straight capitalist reasoning which is what a company is based on, i.e. profits. The system I just mentioned is supposedly the cheapest way to implement an application server system (can't say myself) and individual terminals can net boot which is probably a very good thing even if French kids don't break pencils off in floppy drives like I've seen in the U.S. I know Jean-Paul hopes the Metz-Nancy area can develop into an open-source silicon valley style community which can ignite a domestic and global scale business, and he is tech-minded enough (and if I remember right did music at IRCAM) to get it. Email me if you are interested in finding out more about his work, especially if you are French.
Check this out-- keep an online version of the documentation in a CVS tree. Then everyone using the fscking stuff can fix errors, share code fixes, note bugs, and search locally and on the web site. Next step: find a way to let you integrate all manuals you have (or have subscribed to) onto your harddisk (I'd settle for just dropping a bunch of html tarballs in and mangling it myself) so I can have a knowledge base. I did this with much of Microsoft and Apple sites and was willing to spend time editting local HTML pages to keep navigation and research topics together. If everyone is having the same problem, some kind of database and CVS-type solution seems imperative. Oh yeah, integrated IRC (/join Chapter 26 Null Pointers in Windows) would be nice too thanx.
Short answer: If funds are tight, hire top notch manual writers and do the best job you can digitally with HTML. But high density computer books and essential manuals should be in print plus non-proprietary format on CD. Website-based manuals should provide a tarball and updates (spent *lots* of time spidering manuals that way..) Microsoft should stop wasting all those trees, the manuals suck so badly they should just be used as search engine fodder with someone smart making a hierarchical list of links so you can use it fast.
Absolutely essential to have one or the other when you need it. All printed docs should come with a searchable version obviously. The Inside Macintosh tomes just had to be on CD or else.. but the software and font size absolutely sucked (I used it to hack Quicktime for Windows and while the Apple Quicktime online manual site was fantastic, it cost a lot of money to use online when I really just wanted a tarball of the site!!) When I learned Japanese I used Nelson's for years and got to the point where I normally opened the book to the right page better than theory (well not quantum) would allow. Even tried to scan it, but some books, especially algorithmic or with graphics, you need in print.
That said, everything by Microsoft should be free on the CD not print.. I have found it all to be horrible, low satisfaction per kilo, and best used for searching. Course they should be shot just for the horrible Java-based site they have. As if paper is the only way to control piracy. Serious manuals and computer books should all have websites with errata and ads to support the book.
Reading for this comment: end of related article from an above post, search for "Bohm" (apparently a major quantum theory person) where it talks about an actual scientific theory for a thought-like quantum field: http://www.techweb.com/printableArticle?doc_id=TWB 19980515S0019
There are other ways to disrupt a random system than spooky quantum effects in microtubules though personally I think it more likely than not that something quantum will turn up too. If that is how the world works, it seems improbable that the brain completely fails to harness quantum effects after a billion years of trial and error. For that matter it seems likely that quantum effects may have helped along the evolution process..
Regardless of whether posited phenomenon is based on human electromagnetic radiation or quantum effects, at least you should be able to build a gadget that can tell if it is being looked at intensely by the Princeton project's top performers. Another way for immediate effects to be generated is related to a post I made a week ago to Slashdot that got cancelled.
A Canadian fringe physicist was arrested last month (again, he keeps getting his equipment confiscated). He is said to have found all kinds of bizarre reproducible physical effects from the interaction of different kinds of electromagnetic signal generators, microwave towers, and geographical areas, effects normally called supernatural. It's been dubbed the Poltergeist Effect and apparently can be effected with standard house current.
In a different site I found a science teacher/experimenter who shows how to make charged masses of air with static electrical fields and documents all kinds of bizarre effects like a strong "wall" of charged air (accidentally created under a roll of moving plastic film) and visualizing threads of electrical energy emitted from the hairs on the skin. It appears that if you have a very complex jumble of electrical and magnetic fields in a space with charged air of the right humidity, it is perfectly reasonable that a person's own bioelectrical field and physical structures could be enough to alter it chaotically.
So differences between subjects could be either microtubule level, or capacitance/emission level. Unreproducible results could be related to local geomagnetic anomalies or saturation of the area with a high level of RF energy (which was a key factor in the Poltergeist Effect apparently).
Not that I'm against the quantum stuff, personally I think it is more likely that there is something to it, and I think it would be great. But you have to give a hand to the PEAR crew for trying desperately to explain it. It just seems pretty juvenile to discount "billions of trials" with pseudo-authoritarian shlock.
..if you had a GPS receiver attached to your computer and you kept cpu load (& temperature) pretty stable, could you do away with the atomic clock in some cases? (within 1 to 10 milliseconds with NTP and luck..)
cf: Network Time Protocol http://www.eecis.udel.edu/~ntp/ntp_spool/html/in dex.htm
It seems (just from reading web pages) that highly accurate positioning information and noise control is critical to small arrays of dishes, and that plus atomic-clock-level time accuracy, normally keyed from satellites like Project Argus, are some limiting factors for large interferometers.
If you had two or more ccd cameras on small (3"-8" dia. telescopes) and a single pc though, I'm curious about whether you could build a working optical interferometer or phased array (it might work until a gnat walking by jiggled one by 1/2 wavelength..). Mostly I've heard about gain, sensitivity, and patch size for radio astronomy but if optical is so great what is the simplest working thing you can make with this? Could you do interferometry on this in realtime on a Pentium III (phased array math sounds a lot harder) and resynchronize very often so you could move the set around?
I'm curious how much sheer processing power (such as it is on consumer electronics..) can make something interesting (multipixel imaging hopefully) with minimal input. Can this be demonstrated in optics with pinhole cameras, or in radio range with fresnel lenses or just (doh) jumbled lengths of wire? I'm thinking of interesting uses within the solar system or on Earth.
Last question, if you had a GPS
cf. "Phasing Several Smaller Antennas" http://www.setileague.org/askdr/interfer.htm
Columbus Optical SETI (a 10" telescope!) http://www.coseti.org/introcoseti.htm
"..it might not be ridiculous to suggest that eliciting the help of thousands of enthusiastic amateur astronomers would considerable aid.." http://www.coseti.org/radobs31.htm http://www.coseti.org/tecspmap.htm (tech index)
The xtal.org link from axe-inc includes info about the kernel in Japanese..
XTAL is pronounced "Crystal"
Used on all 32bit Sharp zaurii Used on Sharp's Communication Pal and Browser Board (?? couldn't find on Sharp's page) Is foundation of "SSS-Core" OS joint research project with Tokyo University Used in Takaoka Seisakujo's terminal (http://www.takaoka.co.jp/).. This seems to refer to a standalone SPARC-based CSV X Terminal which claims to be the fastest in the world (300L XStones) and the cheaper CSV model shown at http://www.takaoka.co.jp/system/SysXMiNT.html .
Used in both embedded computers and desktops Research on cluster machines using Crystal at Tokyo University in SSS-Core project
AXE-TCP has been used for 6Mbps satellite telecom and supports IPSEC NFS client for sharing files with unix systems, and Samba server for sharing with Windoze HTTPD -- HTTP server run off XTAL, and user interface provided in HTML
The guy comes across as an idiot, granted, though his points are individually correct. But even in the visions of hallowed science fiction writers the library remains a powerful cultural and technological force. Anyone can read on a private terminal (and get research done quickly like Heinlein's Friday who gets to "feed the elephant" (her thirst for knowledge). But look at the Encyclopedia Galactica or Asimov and other authors' visions of libraries. They have experts dedicated to protecting and promoting knowledge, and often higher technology too.
Libraries first need to guarantee that everyone will have access, first to the Internet and second to a comprehensive optical database of the full text of as many books as possible, to allow *all* libraries to have equivalent access to knowledge. The great universities have such amazing libraries and so many books it seems impossible with current technology and funds to make a dent in it with digitization.. but with requiring all publishers to make an SGML version available to libraries (at increased cost perhaps) will be a good start.
Libraries can then embark on developing more advanced systems than most individuals can use for reading books or similar experiences.. perhaps a high quality audio library and playback room, or in the future a CAVE-like immersive environment at any rate, 500 MHz Pentiums are just not the pinnacle of technology. Perhaps high bandwidth and CD or DVD writers would make sense in a library in a more affluent area for example. When I was a small child I remember seeing filmstrips at school, and at the library. That was multimedia! Microfilm readers were also amazing.
If libraries today have microfilm it seems a small step to having a CD burner. Some of the first CD-ROMs available were just collections of great literature and where are they now? If done right it will be very damaging to publishers and authors, so perhaps compensation on a per-view basis is necessary. I know of at least one person who has decided to pursue a study of library science because of the promise of library technology. I also have been involved in photo library digitization and studied the work of the Bettmann Archive (now belonging to Gates) where scanning could barely keep pace with the speed at which the paper media crumbled away. For the Chief Librarian to be so unwilling to consider technologies makes him the biggest impediment to bridging the current reality of inadequate libraries and the hope-filled vision of libraries of the future. Libraries should be meccas of knowledge and with universities, be the most powerful enemy of the tv culture this Librarian fears so much.
Re:AskJesus.Org, By TheSpark - An AskJeves Parrody
on
AskJeeves Interview
·
· Score: 1
I developed a perl-based system a few years ago called the Translator Engine to solve the problem of translating large websites. When you get into the thousands of pages you end up getting lots of different teams working on the site and it becomes very difficult to manage a single version let alone multiple language versions.
Since there is a previously existing site, there is no option to build templates or install a database, so the idea was to make a tool for building a static translated site. The system I built descends into a folder containing an entire site, and rips all content and tags into separate files which could ultimately (not implemented since rates changed) be emailed to translators living in areas where translation was cheaper. Otherwise you need html-savvy translators whereas you really just want translators to see plain text.
At any time you could rebuild the site automatically and see how far you had gotten. Initially MacPerl was used but because of the huge filesystem overhead it was moved to a Sun. It was done before much of the current CGI libraries were debugged so if I was going to use it now I would probably want to rewrite it to use them with MySql.
No matter how you implement translation remember that when an update or addition is made you need to have all other languages updated as well. If you can use a database which separates plain text and provides an index of tracts which are pending translation this would be very useful.
Another system I have (which is currently in use) is a multilingual search engine suite called EyeLatitude. It currently is being used in Japanese and English and uses separate resource files, with a web based administration interface generated separately before installation. It gets dicey trying to type multibyte strings into Perl and controlling the encoding type on output so it is easier to use the correct encoding once in the resource file. Separate files are also used for different functional modules, for example a timestamp routine includes a separate file with a micro-dictionary which lets the module "talk" in plain Japanese. Running that module with a different argument ("en" instead of "jp", gotten from embedded comments or hidden form input fields in the html) makes the module spit back a differently formatted string. A simple version of the engine uses a cache (with HTML tags ripped out) for small sites, since word stemming trees don't always work for these languages.
Perl XML might be a very good option for you now, but in the end your bottleneck is not in your ripping system but in how fast translators' work can be accomplished and incorporated into the site. PHP code mixed in will confuse them too. So you should have a database involved, but a better idea to do it in Perl and have the option to provide the client a database-driven site or if they don't want a database, a static site built out of it on your side before uploading to the server. And remember, be sure to use binary fields and quoting if you do database programming! Otherwise your first double byte translation could crash mysteriously.
Check out the available Perl libraries for ripping HTML and test how they work with your pages, particularly do they preserve all the accents or multibyte characters in your text, and do they preserve Javascript or PHP correctly. They should be up to the job. If you do have a large corporate site for which you want to manage translation and updating over the long term, I am willing to provide such a service. I don't know of any other such service around.
For people who develop technology for display, storage, and annotation of electronic texts, a paradox arises. When such a technology (like LZW, Qt, or RSA) is so successful as to be a de facto standard, it has become so critical a part of our media that by Stallman's philosophy it also becomes a prime target for replacement by free software, particularly if liscensing continues to restrict copying and modification. Even if ordinary users were happy up to that point, an unhackable technology is a menace (most obviously to the media hacker who has become a major force of social progress). The software development community determines the author (previously a free man playing by the rules he knew), and the patent or copyright law which supported his success, to be an enemy.
As with MP3, the cause of free software is going to continue to lose opportunities until free software spreads so widely as to become the de facto standard, and/or commerce or government can be convinced to enshrine your (Mr. Stallman's) ideas in practice or law.
Big investors in companies that are trying to build the infrastructure of the future might feel threatened by legions of hackers waiting to overthrow their model, so recourse to patent law might be a justifiable strategy. Can a reasonable standard for free software (including documentability, modification, inviolable modules, and distribution) be developed which corporate interests could relatively easily be convinced to incorporate into their standard practices?
Adhering to the Code would make the company a good "Free Software Citizen" and somehow protect them from being righteously targeted by the GNU replacement ethic, or give them brownie points on a citizenship scale applied to software packages, shades of McCarthy. Annual certification might fund free software. Perhaps independent verification that source code does not hide major security risks. Another carrot that is so draconian I'm afraid to suggest it, is a strong free software liscensing system which might require cryptography, viral code, or reporting over the network reminiscent of SBA tactics.
A key lies in today's news that Netscape 6 and IE for the Mac uphold HTML 4.0 but IE for Windows does not. Some questions that need to be asked are: What initiatives would convince an Adobe or a Sony to make their key technology free software? How can Sony be convinced to scrap the idea of zoning DVDs.. and is SDMI for free software, or against? How do we gain the hearts and minds of those who are building our eBook future into embedded modules, before the embedding is done? Is it possible for the FSF or the EFF to approach the American Library Association to get funding for free software intiatives which would uphold freedom worldwide and in the long term possibly save libraries money? Are you (Mr. Stallman) ready to take a step toward leading the developers' community in a pragmatic and understandable direction, or are you content to write essays and act as the custodian of the riches which free software has achieved thus far? Can freedom for any nation or man be purchased with draconian implementations which offend that freedom?
Solutions will require motivated people who can see both sides of the coin: Copyright and patents will still have worth in their original meanings, but the DMCA and compromised hardware must be made obsolete. Will you work toward giving corporations good reasons to get into free software besides raping and reselling it?
Ignores reality of libraries and shareware
on
RMS On eBooks
·
· Score: 2
Good timing. The U.S. Government on April 13 awarded over 2 billion dollars collected from carriers to the American Library Association for the smallest libraries to stay online and up to date. (link)
Libraries are a hallowed U.S. tradition with a lot of strong backers and the entire balkanization of what is now printed matter will not come to pass without a very big fight. Possibly one day the Library of Congress could play a new roll in distribution (perhaps a media key allowing reading on a library slate, not a great solution but a minimum one).
In the end the author and publisher need a way to make a profit or there won't be any e-books. Stallman's closing remark that copyright will be obsolete is inane. Why should anyone be forced to submit to an idea of the masses that the product of their labor should be free or anything else? Some authors might accept variable or no payment depending on the reader's level of enjoyment but if it works out like shareware it hardly seems like empowering the author. Currently the phone company makes more than the author on shareware downloads.
Micropayment + Transcopyright, an open source culmination of decades of work in the field (Ted Nelson) is one possible strategy in a universe of them.. and Stallman is focusing on shouting when he could be lining up allies. There is no reason why an open source or other software solution not created by publishers could not take hold, if it addresses the needs of authors and publishers. Legal provisions allowing fair (personal or editorial) use need to be covered by new technologies, especially when "solutions" like zoning of DVDs make it impossible to read certain "texts". When infrastructure makes it possible to charge for consumption of media dynamically, that will open the doors of accessibility to many more authors who will depend on some kind of copyright law (perhaps the software code will exceed the legal code) to make their living.
Last I heard, libraries buy their books. So a limit on the number of times a book is read sounds unworkable. But if prices fall naturally (by economics, not some cracker's idea of fairness) to a dollar a book, there is no reason why payment cannot be made up front or from a dedicated account. So I think Stallman's fears are based on an assumption of frozen technology, and that more technology will allow authors to apply all kinds of payment schemes including different levels of payment, annotation, and other characteristics, as envisioned by Nelson and others. It would be more interesting to do a serious analysis of the work done in this field and work toward a solution than to put blinders on and be alarmist. There may be dangers but there are great possibilities.
On the danger side I see communications carriers and credit card companies enforcing stiff inescapable charges, and companies with vested interests in video and audio taking the initiative with things which look more like entertainment titles than books. On the positive side, how about asking the ALA (or O'Reilly, or the EFF,...) for some of that 2 billion and start experimenting openly (not necessarily GPL) with Transmeta slates? That way people will be able to hack at this problem for a long, long time. It may take that long for copyright owners to all shift to a Napster/Stallman/Shareware-esque style of compensation (or not) based on its own merits.
I am no fan of the DMCA. But if Stallman wants to overturn the DMCA he should quit talking about trying to make copyright obsolete and put some energy into figuring out what initiative he could start or join to build a reasonable business model which can be influenced by the community of believers in electronic freedom. If his thinking stands solely on the ideas of GPL and uncontrolled dissemination of works he will lose credibility among those of us who live in a market economy and with it the opportunity to lead.
Would increase price but sounds like the modem is for looking up an online TV guide and setting when to record TV to disk. (The CNet article mentions access revenues a la WebTV but that doesn't sound like such a big market any more.) With this you might be able to view different camera angles of the Super Bowl with personalized 3D effects. Might also give advertisers feedback.. who needs cookies when you have a serial number and strong encryption?
Off the deeper end, could also become the hub of the home network (Sony wants a home server), allowing other devices to share the modem or 3D graphics capabilities of the PS2.. XHOST +sony_videophone
Might actually get people to program their VCRs too if it has your favorite animated character walk you through it (download your own!)
Personally I'd rather have ten times the current RAM and buy an external hard drive myself.
Isn't it more likely that someone who hates Microsoft (um, narrows down the field quite a bit..) intentionally inserted this? Anybody want to check other dlls for backwards text with a dictionary checker?
>> a firing offense for the as yet unidentified employees.
Slashdot Mirror
If you are a physicist then I would honestly like to know why my above post is so terrible. Seems to extrapolate in brave asses hangin' off the cliff Slasdot style. Since I've only been able to read Scientific American and Nature articles about this, and only had up to organic chemistry in college I certainly don't have a background in the field but do have a thirst for knowledge. Please respond with valid information, anonymous coward or not.
I got a question, hope somebody can poke holes in it. Say you have an unlimited number of states, and we ignore problems with how much energy or time you might need to get to that resolution.
Suppose you encode everything about a computer, its RAM contents and operating rules into a piece of data, basically a long number. Say you are doing something like dumping the VMWare PC emulator program and its memory buffers into this piece of data, along with your own program and also a bunch of other locations which are telltale bits that would only be set to true if certain instructions (your program) are executed in a certain order, so you can in a sense freeze a sequence of calculations, an overall machine state.
So in the end the last telltale will finally be set only if the results of the calculation which suposedly had been executed by this hypothetical (virtual) computer, was provably the answer you seek, i.e. the factors of a big prime number which could be multiplied together to show they are the right answer. A self-referential logic filter.
My conjecture (gleefully made without more knowledge of quantum physics than is available in lay publications..) is this. Could you use this huge number as a filter or reference beam to collapse the waveform of your recording medium, and read out the state of the virtual computer with the output of your program, in a picosecond?
It would seem that any Turing machine from a Cray to a ribosome (an rna tape device), could be simulated in this way, though smaller memory footprint/instruction set machines would be easier since they could be represented with less eigenstates. I wonder how many states would be the least amount necessary to simulate something useful.. if a full hardware abstraction is not needed and you can get away with just a language definition and virtual machine (yes like Java VM).
Would this mean you could run any program that can fit into the virtual machine in picosecond time? And if so, could you not in fact build a computer of any arbitrary capabilities by simply writing a pseudocode definition of how it ought to work? Final scary question.. The interior of a cell is a controlled environment and until the cell is queried by some process it is conceivable that some ribosomes could exist in superimposed states. Put another way, if you could solve the isolation problem it might end up to be cheaper to build the eigenstate computer with common cellular apparatus than by using big expensive lasers. What conclusions can you draw from this?
I think this is what was meant by a prediction I once came across.. that the coming century would create a new science of computing which is to today's computers as nuclear energy is to fire.
Like I said I hope someone can poke holes in this. The biggest problem seems to be universal laws about information, for example I understand that the recent sending of a light pulse at 300 times the ordinary speed of light was only possible because the leading edge of the pulse had enough information to reconstruct the rest of the pulse, suggesting that you could not send an entire packet of bits faster than the speed of light.
Thought I'd add this thought I had since other posts here seem related.
It seems the terrain in this battle could be altered significantly if server vulnerabilities were used to insert a file (mpaa.html or decss.html) into the root directory of a large number of servers across the net. (Alternately an otherwise invisible text link or transparent gif link in a bottom corner of the page might work as well, to find mirrors, like the PI character in the movie the Net). The html file could contain source and multiple executables of the DeCSS code, and this file could be given a variety of related names or placed in subfolders, stegographed into gifs, inserted into Javascript, encrypted with a public key, etc., so that it is less detectable by administrators and mpaa search engine investigators while promising an unending morph of target and terrain.
Maybe if the MPAA ever gives up the server admins would be notified of the security holes they have.. but I am thinking that admins might even intentionally add this to their own systems since they could say it was placed there by a hacker. Leaving a security hole open intentionally might be a plausible defense in that case.
This thought came to me when I was considering a perl script published in a well known network security secrets book. A friend inexperienced in perl or hacking tried offhand to test his clients' NT servers and suddenly was granted System permissions three times out of three, score 3 for linux. This seems a far more interesting game for script kiddies too, why bother defacing some clueless government's page when they could screw the MPAA instead. There is also less damage on the part of those defaced.
Some more on Ruby - since this is probably the answer to what a Perl type language would look like if written by Japanese speaking programmers.
The big draw of Ruby as I understand it is that it is very easy to write network utilities, is highly object oriented, and programs are short and sweet. The network module has gotten much stronger, the initial weakness of which kept me away in the beginning (and of course I am more comfortable with Perl). Every string for example is an object with built in methods you can call after the dot. It also works with GTK which might interest Slashdot readers. The Brave GNU World article (translated itself into several languages) mentions it just before covering a Bonobo component.. The article does mention that Ruby was designed to be a successor of Perl, and borrows from Perl, Python, Lisp and Smalltalk, and Eiffel, and also runs on a number of platforms. Hence the name Ruby, which makes sense if you read Perl as a pearl.. also brings up another good point that there is a ruby Apache module, and that ruby supports multithreading even on MS-Dos?!?! This I did not know!
A recent Japanese magazine I bought introduced web programming by using Ruby exclusively, going through the RFC documents for HTTP, and building protocol sniffers. Fun stuff! Ruby people are looking for help in spreading the language around the world so this might be more reason to check it out. In fact that magazine (OpenDesign, in Japanese, by CQ Shuppan) was entirely about that and also I-Mode telephone protocols and programming.
Brave GNU World article 8/14/00:
http://www.gnu.org/brave-gnu-world/issue-18.en.
English homepages:
http://www.ruby-lang.org/en/index.html
http://ruby-lang.jaist.ac.jp:8000/en/doc.html
http://kt-www.jaist.ac.jp:8000/~ttate/lang/ruby
I am a programmer in Tokyo and think much of the richness and draw of Perl is lost on young Japanese programmers. Advanced programmers are just that, there is no problem, and they read English well. Advanced languages and AI apps make more use of language, but it seems that new paradigms are not necessarily coming from language but from study of semantics and dynamics, which are universal. Some might take off better in other countries, for example toontalk is easy to undestand by Japanese.
Some of the draws of Perl are "There Is Always More Than One Way To Do It" and of course the amazing CPAN multiplexed archive and module search/installation system. TIAMTOWTDI doesn't just mean writing verbs in a different order. Where it comes from flexible thinking English may help but certainly has no monopoly.. but there is also more vocabulary and more ways to say something in English than in other languages, which may affect programming.
Of course people who don't read English well are either unaware of the CPAN system or have difficulty making use of it.
One response may have been the development of Ruby in Japan, which borrows from several different languages including Perl, but is at first seems to have confusing operators and syntax. It is highly object oriented, looking more like C or ASP than Perl, though it is meant to be used as Perl and relies heavily on external modules of which there are few.
One good landmark is legal writing in Japan. When well done it is much like a program, and does not suffer much of the archaic nature of American legal writing. It is crisp, and when well written rings like a bell, like most well written Japanese text (style is why it is hard to write well, the characters are not so bad).
Perl might not have been developed at all, or developed in the same way, if it had been done by Japanese speakers or by people brought up in Japanese culture.. but there is no problem with qualityamong serious programmers, the main loss is the fun English readers get.. like understanding the meaning of "State of the Onion" and the hilarity of jolly code writing which uses terms like bless or die in semi poetic context.
The spring action is very cool, as is their means of testing it and monitoring wear and tear (though
not sor sure 20 times is enough to be sure).
They did not invent the bearing idea. Because nanoscale fabrication is still in its beginnings, commercial concerns are busy engineering films of nanoscale particles. One big player is Sharp. A nanoengineering scientist who has his own startup described nanoscale bearing manufacture as a grail many have been seeking, of course he had his own angle on it. The first profitable use is expected to be in lubrication of disk drive surfaces, so that the head actually rests on the disk. I understood the idea to be rolling tubes rather than round spheres as bearings.
There is also a big danger, in that these particles will get into anything including you and me, apparently being quite lethal.
I also know a company which makes machines to sinter fine powder into 3d objects with lasers. The powder is very fine on microscale (steel and soon titanium, but best quality is achieved with less conductive nylon). These guys were pulling so many all nighters, on such a small budget, that in developing the next material they often did not use breathing masks but this is probably the limit to how small a particle you want to have around you without major protection.
Some people may remember the Source, a Prime computer running before public Internet access. It had a bad natural language parsing system for the perennial dungeon adventure but the text descriptions were fantastic. You might have something like it on your linux box, type "adventure".
r uctors.html
o rph.html
Infocom had an excellent parsing system that dealt with ambiguity and was very impressive in the adventure context, starting with Zork on the Apple II. They sold an entire series of games based around this parser so it must have been sufficient for the task at hand.
A company called Picture Network Interactive (PNI) built an online photo rental agency which they tried to use to take over the industry (I was on the receiving side in Japan and heard the presentation). Supposedly they used a natural language parsing system to describe photos based on something they (or someone they bought out) used to build a database application for the White House. Supposedly it could handle English, Romance languages, and even Japanese (that was probably vapor), this was sometime around 1993 I believe.
Infoseek's search engine does some language parsing, this was one of their two original selling points (the other being that they would sell you a keyword for targeted ads).. They told me at the time that the company they bought this technology off was called HNC (?) apparently a big IPO the previous year, which means around 1995 or thereafter I believe. Now that I think about it, HNC could have been involved with PNI not Infoseek oh well.
Muse Technologies' synthetic environment for data visualization software provides voice command support but this is for very limited grammar and vocabulary even if it is a Sandia spinoff (musetech.com).
Translation of spoken language, through the phone or via consumer items, has been promised for a while. AT&T has a very nice system. It seems that much is possible if you limit the topic, grammar, and vocabulary.
But simple grammar should not be so hard, and interactive q&a can resolve ambiguity if the system recognizes it.. something between configure.sh and clippy. Actually I wouldn't mind clippy as much if it had broader knowledge.
The UMap analyzer for web queries is intense and combines text analysis and production of gorgeous clickable, customizable live maps. The site is being changed but a quick look at a screenshot is at http://www.kartograph.com/uk/index1024.htm (umap.com had a downloadable program mere months ago, still living on my harddrive.) Very very nice. Kartograph is for lotus notes. It isn't parsing so much as analyzing correspondences between the meanings and statistics of word appearances, I think.
I came across a morpohological analyzer recently but can't find the link.. maybe look in google.com, here are two links I found there.
Here is a link to a bunch of natural language tools.
http://web.syr.edu/~mdtaffet/nlp_sites_for_inst
Maybe this is useful for snlp..
http://www.rxrc.xerox.com/research/mltt/fsnlp/m
Would certainly like to find a perl module or c source code for an snlp facility!
I have seen posters of the English version of Mononoke Hime in Japan, and think it either was played here or will be here on tapes. (someone else please provide better info).
Is it possible that the dubbed version could be available here in Japan on DVD without the Japanese version being available?
If a major film is released on DVD must the security be able to handle any forseeable development of deCss style systems in the medium term?
A recent Scientific American article by the discoverer of the fastest possible quantum computer search algorithm describes the issues well. It could have been Shor.
Unfortunately I have neither the magazine nor yet able to read the above article. Apparently the inventor had proven that his own algorithm was the fastest possible one, though the definition of searching may have been stretched to fit the probabilistic algorithm now in the news.
I do not remember if the basic idea was to assign one qubit to each letter of a name, or just one qubit to each item in a database, but it seemed quite powerful even with a small number of qubits.
A more interesting question in the short term might be whether a <=15 qubit computer can make serious inroads on big factoring problems.
Always felt queasy about FSF's suggestion that people sign licenses over to them. Perhaps this is the best example of why someone would want to do so?
This is just the latest salvo in a year long campaign by the government and individuals who want to jumpstart the economy with Internet technology.
I have discussed several times the issue with Jean-Paul Smets, a representative of the French ministry of industry in Lorraine, who is working with the secretary for information technology under the prime minister. A number of linux software and hardware companies are starting up and one unit (the linbox, based on SuSE linux) with vmware and windows and macos installed is available preinstalled for individual or large clusters.
This was installed last year in a local school and a minister I believe was in attendance. Surely exhaustion with the NSA and Microsoft may have something to do with it but there is also the economy, lack of performance relative to Silicon Valley (well maybe not as bad after the market tanked), and the French idea of freedom.. plus the straight capitalist reasoning which is what a company is based on, i.e. profits. The system I just mentioned is supposedly the cheapest way to implement an application server system (can't say myself) and individual terminals can net boot which is probably a very good thing even if French kids don't break pencils off in floppy drives like I've seen in the U.S. I know Jean-Paul hopes the Metz-Nancy area can develop into an open-source silicon valley style community which can ignite a domestic and global scale business, and he is tech-minded enough (and if I remember right did music at IRCAM) to get it. Email me if you are interested in finding out more about his work, especially if you are French.
Check this out-- keep an online version of the documentation in a CVS tree. Then everyone using the fscking stuff can fix errors, share code fixes, note bugs, and search locally and on the web site. Next step: find a way to let you integrate all manuals you have (or have subscribed to) onto your harddisk (I'd settle for just dropping a bunch of html tarballs in and mangling it myself) so I can have a knowledge base. I did this with much of Microsoft and Apple sites and was willing to spend time editting local HTML pages to keep navigation and research topics together. If everyone is having the same problem, some kind of database and CVS-type solution seems imperative. Oh yeah, integrated IRC (/join Chapter 26 Null Pointers in Windows) would be nice too thanx.
Short answer: If funds are tight, hire top notch manual writers and do the best job you can digitally with HTML. But high density computer books and essential manuals should be in print plus non-proprietary format on CD. Website-based manuals should provide a tarball and updates (spent *lots* of time spidering manuals that way..) Microsoft should stop wasting all those trees, the manuals suck so badly they should just be used as search engine fodder with someone smart making a hierarchical list of links so you can use it fast.
Absolutely essential to have one or the other when you need it. All printed docs should come with a searchable version obviously. The Inside Macintosh tomes just had to be on CD or else.. but the software and font size absolutely sucked (I used it to hack Quicktime for Windows and while the Apple Quicktime online manual site was fantastic, it cost a lot of money to use online when I really just wanted a tarball of the site!!) When I learned Japanese I used Nelson's for years and got to the point where I normally opened the book to the right page better than theory (well not quantum) would allow. Even tried to scan it, but some books, especially algorithmic or with graphics, you need in print.
That said, everything by Microsoft should be free on the CD not print.. I have found it all to be horrible, low satisfaction per kilo, and best used for searching. Course they should be shot just for the horrible Java-based site they have. As if paper is the only way to control piracy. Serious manuals and computer books should all have websites with errata and ads to support the book.
Reading for this comment: end of related article from an above post, search for "Bohm" (apparently a major quantum theory person) where it talks about an actual scientific theory for a thought-like quantum field: http://www.techweb.com/printableArticle?doc_id=TWB 19980515S0019
There are other ways to disrupt a random system than spooky quantum effects in microtubules though personally I think it more likely than not that something quantum will turn up too. If that is how the world works, it seems improbable that the brain completely fails to harness quantum effects after a billion years of trial and error. For that matter it seems likely that quantum effects may have helped along the evolution process..
Regardless of whether posited phenomenon is based on human electromagnetic radiation or quantum effects, at least you should be able to build a gadget that can tell if it is being looked at intensely by the Princeton project's top performers. Another way for immediate effects to be generated is related to a post I made a week ago to Slashdot that got cancelled.
A Canadian fringe physicist was arrested last month (again, he keeps getting his equipment confiscated). He is said to have found all kinds of bizarre reproducible physical effects from the interaction of different kinds of electromagnetic signal generators, microwave towers, and geographical areas, effects normally called supernatural. It's been dubbed the Poltergeist Effect and apparently can be effected with standard house current.
In a different site I found a science teacher/experimenter who shows how to make charged masses of air with static electrical fields and documents all kinds of bizarre effects like a strong "wall" of charged air (accidentally created under a roll of moving plastic film) and visualizing threads of electrical energy emitted from the hairs on the skin. It appears that if you have a very complex jumble of electrical and magnetic fields in a space with charged air of the right humidity, it is perfectly reasonable that a person's own bioelectrical field and physical structures could be enough to alter it chaotically.
So differences between subjects could be either microtubule level, or capacitance/emission level. Unreproducible results could be related to local geomagnetic anomalies or saturation of the area with a high level of RF energy (which was a key factor in the Poltergeist Effect apparently).
Not that I'm against the quantum stuff, personally I think it is more likely that there is something to it, and I think it would be great. But you have to give a hand to the PEAR crew for trying desperately to explain it. It just seems pretty juvenile to discount "billions of trials" with pseudo-authoritarian shlock.
..if you had a GPS receiver attached to your computer and you kept cpu load (& temperature) pretty stable, could you do away with the atomic clock in some cases? (within 1 to 10 milliseconds with NTP and luck..)
n dex.htm
cf: Network Time Protocol
http://www.eecis.udel.edu/~ntp/ntp_spool/html/i
It seems (just from reading web pages) that highly accurate positioning information and noise control is critical to small arrays of dishes, and that plus atomic-clock-level time accuracy, normally keyed from satellites like Project Argus, are some limiting factors for large interferometers.
If you had two or more ccd cameras on small (3"-8" dia. telescopes) and a single pc though, I'm curious about whether you could build a working optical interferometer or phased array (it might work until a gnat walking by jiggled one by 1/2 wavelength..). Mostly I've heard about gain, sensitivity, and patch size for radio astronomy but if optical is so great what is the simplest working thing you can make with this? Could you do interferometry on this in realtime on a Pentium III (phased array math sounds a lot harder) and resynchronize very often so you could move the set around?
I'm curious how much sheer processing power (such as it is on consumer electronics..) can make something interesting (multipixel imaging hopefully) with minimal input. Can this be demonstrated in optics with pinhole cameras, or in radio range with fresnel lenses or just (doh) jumbled lengths of wire? I'm thinking of interesting uses within the solar system or on Earth.
Last question, if you had a GPS
cf. "Phasing Several Smaller Antennas"
http://www.setileague.org/askdr/interfer.htm
World's tiniest fiber optic spectrometer (!)
http://www.coseti.org/oceanop1.htm
Columbus Optical SETI (a 10" telescope!)
http://www.coseti.org/introcoseti.htm
"..it might not be ridiculous to suggest that eliciting the help of thousands of enthusiastic amateur astronomers would considerable aid.."
http://www.coseti.org/radobs31.htm
http://www.coseti.org/tecspmap.htm (tech index)
Sorry, 300K (300,000) XStones not 300L.
The xtal.org link from axe-inc includes info about the kernel in Japanese..
XTAL is pronounced "Crystal"
Used on all 32bit Sharp zaurii
Used on Sharp's Communication Pal and Browser Board (?? couldn't find on Sharp's page)
Is foundation of "SSS-Core" OS joint research project with Tokyo University
Used in Takaoka Seisakujo's terminal (http://www.takaoka.co.jp/).. This seems to refer
to a standalone SPARC-based CSV X Terminal which claims to be the fastest in the world (300L XStones) and the cheaper CSV model shown at http://www.takaoka.co.jp/system/SysXMiNT.html .
Used in both embedded computers and desktops
Research on cluster machines using Crystal at Tokyo University in SSS-Core project
Supported CPUs: 80x86, Pentium, ARM6, ARM7, PowerPC 601, 603, 603e, PowerPC 403, SuperSparc, UltraSparc, SH3, 680X0
AXE-TCP has been used for 6Mbps satellite telecom
and supports IPSEC
NFS client for sharing files with unix systems, and Samba server for sharing with Windoze
HTTPD -- HTTP server run off XTAL, and user interface provided in HTML
The guy comes across as an idiot, granted, though his points are individually correct. But even in the visions of hallowed science fiction writers the library remains a powerful cultural and technological force. Anyone can read on a private terminal (and get research done quickly like Heinlein's Friday who gets to "feed the elephant" (her thirst for knowledge). But look at the Encyclopedia Galactica or Asimov and other authors' visions of libraries. They have experts dedicated to protecting and promoting knowledge, and often higher technology too.
Libraries first need to guarantee that everyone will have access, first to the Internet and second to a comprehensive optical database of the full text of as many books as possible, to allow *all* libraries to have equivalent access to knowledge. The great universities have such amazing libraries and so many books it seems impossible with current technology and funds to make a dent in it with digitization.. but with requiring all publishers to make an SGML version available to libraries (at increased cost perhaps) will be a good start.
Libraries can then embark on developing more advanced systems than most individuals can use for reading books or similar experiences.. perhaps a high quality audio library and playback room, or in the future a CAVE-like immersive environment at any rate, 500 MHz Pentiums are just not the pinnacle of technology. Perhaps high bandwidth and CD or DVD writers would make sense in a library in a more affluent area for example. When I was a small child I remember seeing filmstrips at school, and at the library. That was multimedia! Microfilm readers were also amazing.
If libraries today have microfilm it seems a small step to having a CD burner. Some of the first CD-ROMs available were just collections of great literature and where are they now? If done right it will be very damaging to publishers and authors, so perhaps compensation on a per-view basis is necessary. I know of at least one person who has decided to pursue a study of library science because of the promise of library technology. I also have been involved in photo library digitization and studied the work of the Bettmann Archive (now belonging to Gates) where scanning could barely keep pace with the speed at which the paper media crumbled away. For the Chief Librarian to be so unwilling to consider technologies makes him the biggest impediment to bridging the current reality of inadequate libraries and the hope-filled vision of libraries of the future. Libraries should be meccas of knowledge and with universities, be the most powerful enemy of the tv culture this Librarian fears so much.
Viewed slashdot with it.. fucking hilarious!
I developed a perl-based system a few years ago called the Translator Engine to solve the problem of translating large websites. When you get into the thousands of pages you end up getting lots of different teams working on the site and it becomes very difficult to manage a single version let alone multiple language versions.
Since there is a previously existing site, there is no option to build templates or install a database, so the idea was to make a tool for building a static translated site. The system I built descends into a folder containing an entire site, and rips all content and tags into separate files which could ultimately (not implemented since rates changed) be emailed to translators living in areas where translation was cheaper. Otherwise you need html-savvy translators whereas you really just want translators to see plain text.
At any time you could rebuild the site automatically and see how far you had gotten. Initially MacPerl was used but because of the huge filesystem overhead it was moved to a Sun. It was done before much of the current CGI libraries were debugged so if I was going to use it now I would probably want to rewrite it to use them with MySql.
No matter how you implement translation remember that when an update or addition is made you need to have all other languages updated as well. If you can use a database which separates plain text and provides an index of tracts which are pending translation this would be very useful.
Another system I have (which is currently in use) is a multilingual search engine suite called EyeLatitude. It currently is being used in Japanese and English and uses separate resource files, with a web based administration interface generated separately before installation. It gets dicey trying to type multibyte strings into Perl and controlling the encoding type on output so it is easier to use the correct encoding once in the resource file. Separate files are also used for different functional modules, for example a timestamp routine includes a separate file with a micro-dictionary which lets the module "talk" in plain Japanese. Running that module with a different argument ("en" instead of "jp", gotten from embedded comments or hidden form input fields in the html) makes the module spit back a differently formatted string. A simple version of the engine uses a cache (with HTML tags ripped out) for small sites, since word stemming trees don't always work for these languages.
Perl XML might be a very good option for you now, but in the end your bottleneck is not in your ripping system but in how fast translators' work can be accomplished and incorporated into the site. PHP code mixed in will confuse them too. So you should have a database involved, but a better idea to do it in Perl and have the option to provide the client a database-driven site or if they don't want a database, a static site built out of it on your side before uploading to the server. And remember, be sure to use binary fields and quoting if you do database programming! Otherwise your first double byte translation could crash mysteriously.
Check out the available Perl libraries for ripping HTML and test how they work with your pages, particularly do they preserve all the accents or multibyte characters in your text, and do they preserve Javascript or PHP correctly. They should be up to the job. If you do have a large corporate site for which you want to manage translation and updating over the long term, I am willing to provide such a service. I don't know of any other such service around.
For people who develop technology for display, storage, and annotation of electronic texts, a paradox arises. When such a technology (like LZW, Qt, or RSA) is so successful as to be a de facto standard, it has become so critical a part of our media that by Stallman's philosophy it also becomes a prime target for replacement by free software, particularly if liscensing continues to restrict copying and modification. Even if ordinary users were happy up to that point, an unhackable technology is a menace (most obviously to the media hacker who has become a major force of social progress). The software development community determines the author (previously a free man playing by the rules he knew), and the patent or copyright law which supported his success, to be an enemy.
As with MP3, the cause of free software is going to continue to lose opportunities until free software spreads so widely as to become the de facto standard, and/or commerce or government can be convinced to enshrine your (Mr. Stallman's) ideas in practice or law.
Big investors in companies that are trying to build the infrastructure of the future might feel threatened by legions of hackers waiting to overthrow their model, so recourse to patent law might be a justifiable strategy. Can a reasonable standard for free software (including documentability, modification, inviolable modules, and distribution) be developed which corporate interests could relatively easily be convinced to incorporate into their standard practices?
Adhering to the Code would make the company a good "Free Software Citizen" and somehow protect them from being righteously targeted by the GNU replacement ethic, or give them brownie points on a citizenship scale applied to software packages, shades of McCarthy. Annual certification might fund free software. Perhaps independent verification that source code does not hide major security risks. Another carrot that is so draconian I'm afraid to suggest it, is a strong free software liscensing system which might require cryptography, viral code, or reporting over the network reminiscent of SBA tactics.
A key lies in today's news that Netscape 6 and IE for the Mac uphold HTML 4.0 but IE for Windows does not. Some questions that need to be asked are: What initiatives would convince an Adobe or a Sony to make their key technology free software? How can Sony be convinced to scrap the idea of zoning DVDs.. and is SDMI for free software, or against? How do we gain the hearts and minds of those who are building our eBook future into embedded modules, before the embedding is done? Is it possible for the FSF or the EFF to approach the American Library Association to get funding for free software intiatives which would uphold freedom worldwide and in the long term possibly save libraries money? Are you (Mr. Stallman) ready to take a step toward leading the developers' community in a pragmatic and understandable direction, or are you content to write essays and act as the custodian of the riches which free software has achieved thus far? Can freedom for any nation or man be purchased with draconian implementations which offend that freedom?
Solutions will require motivated people who can see both sides of the coin: Copyright and patents will still have worth in their original meanings, but the DMCA and compromised hardware must be made obsolete. Will you work toward giving corporations good reasons to get into free software besides raping and reselling it?
Libraries are a hallowed U.S. tradition with a lot of strong backers and the entire balkanization of what is now printed matter will not come to pass without a very big fight. Possibly one day the Library of Congress could play a new roll in distribution (perhaps a media key allowing reading on a library slate, not a great solution but a minimum one).
In the end the author and publisher need a way to make a profit or there won't be any e-books. Stallman's closing remark that copyright will be obsolete is inane. Why should anyone be forced to submit to an idea of the masses that the product of their labor should be free or anything else? Some authors might accept variable or no payment depending on the reader's level of enjoyment but if it works out like shareware it hardly seems like empowering the author. Currently the phone company makes more than the author on shareware downloads.
Micropayment + Transcopyright, an open source culmination of decades of work in the field (Ted Nelson) is one possible strategy in a universe of them.. and Stallman is focusing on shouting when he could be lining up allies. There is no reason why an open source or other software solution not created by publishers could not take hold, if it addresses the needs of authors and publishers. Legal provisions allowing fair (personal or editorial) use need to be covered by new technologies, especially when "solutions" like zoning of DVDs make it impossible to read certain "texts". When infrastructure makes it possible to charge for consumption of media dynamically, that will open the doors of accessibility to many more authors who will depend on some kind of copyright law (perhaps the software code will exceed the legal code) to make their living.
Last I heard, libraries buy their books. So a limit on the number of times a book is read sounds unworkable. But if prices fall naturally (by economics, not some cracker's idea of fairness) to a dollar a book, there is no reason why payment cannot be made up front or from a dedicated account. So I think Stallman's fears are based on an assumption of frozen technology, and that more technology will allow authors to apply all kinds of payment schemes including different levels of payment, annotation, and other characteristics, as envisioned by Nelson and others. It would be more interesting to do a serious analysis of the work done in this field and work toward a solution than to put blinders on and be alarmist. There may be dangers but there are great possibilities.
On the danger side I see communications carriers and credit card companies enforcing stiff inescapable charges, and companies with vested interests in video and audio taking the initiative with things which look more like entertainment titles than books. On the positive side, how about asking the ALA (or O'Reilly, or the EFF,...) for some of that 2 billion and start experimenting openly (not necessarily GPL) with Transmeta slates? That way people will be able to hack at this problem for a long, long time. It may take that long for copyright owners to all shift to a Napster/Stallman/Shareware-esque style of compensation (or not) based on its own merits.
I am no fan of the DMCA. But if Stallman wants to overturn the DMCA he should quit talking about trying to make copyright obsolete and put some energy into figuring out what initiative he could start or join to build a reasonable business model which can be influenced by the community of believers in electronic freedom. If his thinking stands solely on the ideas of GPL and uncontrolled dissemination of works he will lose credibility among those of us who live in a market economy and with it the opportunity to lead.
A hard drive connected to your TV?..
Would increase price but sounds like the modem is for looking up an online TV guide and setting when to record TV to disk. (The CNet article mentions access revenues a la WebTV but that doesn't sound like such a big market any more.) With this you might be able to view different camera angles of the Super Bowl with personalized 3D effects. Might also give advertisers feedback.. who needs cookies when you have a serial number and strong encryption?
Off the deeper end, could also become the hub of the home network (Sony wants a home server), allowing other devices to share the modem or 3D graphics capabilities of the PS2.. XHOST +sony_videophone
Might actually get people to program their VCRs too if it has your favorite animated character walk you through it (download your own!)
Personally I'd rather have ten times the current RAM and buy an external hard drive myself.
Isn't it more likely that someone who hates Microsoft (um, narrows down the field quite a bit..) intentionally inserted this? Anybody want to check other dlls for backwards text with a dictionary checker?
>> a firing offense for the as yet unidentified employees.