I mean, when you allow technology like this, which the public does not understand, to become central to the democratic process, you empower a small technological elite who understands the technology and its limitations.
Not meaning any disrespect (since I amd not referring to the previous author, but our community as a whole) but losing control by not understanding technology is no different than losing controil by not understanding history (i.e. how things really work in the long term), politics (how things really work in the short term) or not understanding, or even bothering to read (*gasp*) THE LAW (how things are *supposed to work - and what we are talking about here)
This bill has been discussed here since it was introduced in May, and the 1999 hearing have been on line much longer. The House passed this law two weeks ago and the final version on the law has been online since June 8.
I urge all slashdotters to read this bill (or at least this summary), so we can discuss this intelligently.
Perhaps more importantly, take a look at this list of other currently and recently pending laws on digital security and e- cyber- computer-whatever before Congress. As the people who understand the technology, we need to play an active role in forming these laws and discussion, which means we need to have a clue about the laws under discussion. (Hint: the subject is *never* ipchains or alernative files systems, etc. Those are just incidentals to the law.)
BTW, when reading the linked laws/discussion, please remember even 'dead' legislation is likely to come back. The current bill was formed from the 1999 'Bliley Bill" (HR 1714), and 1999 Senate 761.
I do not in any way mean to detract from the accomplishments presented here (though, of course, it's been presented in varying shades of "Any time now" for decades. I made my first hologram as a Explorer scout in 1977 or so, and at that time it was already a staple of such classics as "The Adolescence of PI")
However, the storage capacity cited is 10GB/cm^3 cubic centimeter) not cm^2 (square centimeter) as you stated. By comparison, given how thin the magnetic films are, I doubt the 75GB HDD even has 1 cc of active storage volume so holograms do not approach the volumetric density of magnetic media.
The key of course is that holograms offer the promise of true volumetric storage, where magnetic media is limited to the thin film [though who can forget the hedelberg group who used a thin film -- namely a roll of commercial adhesive tape to as an optical medium for up to 10GB a few minths ago?)
So, since it comes down to form factor, I'm not excited. The problem with the HDD is the overall associated mechanism, and with 1 GB matchbook 10G-resistant HDDs out there *today*, I'm not sure when I'll ever be excited. There is no reason to expect we'll be carrying 'naked' (or packaged) holographic media, any more than we carry naked (or packaged) *high density* platters today -- and holographic drives may well be larger, more expensive, more fragile, etc. than HDDs in 2003, as well. In 2003, you won't be able to *buy* a new HDD as puny as 10GB, if indeed they are still maanufacturing that size, today.
The external hologram surface may turn out to be sensitive to damage (though I can hope for the use of confocal optics, etc. to image the volume despite surface imperfections or contamination) and the volume may be vulnerable to sunlight.
The hologram technology used here showed promise because it can be multiplexed with different laser colors and at different angles, but the 'clarity' of the signal goes down with the square of the number of channels, until it is unintelligible. This does not bode well for rapid breakthroughs (though if we could predict them, they wouldn't be breakthroughs). Precise alignment is necessary to assure high density, reliable readings. it seems likely that the 2003 holographic drive will be larger, more expensive, and offer no appreciable advantage (aside from ?magnetic insensitivity?, if that counts)
Aside from the probability of actually seeing a production drive someday, I think that several other holographic technologies are more promising. and none of the holographic technologies show signs of exceeding the practical capacities of straight optical media in the predictable future -- i.e. the next three years. Standards, not technical capabilities, block DVD-R from coming out *this year*
Hey, I want my petabyte encrypted keychain as much as the next guy... but, you know, 'fire' still has many unparalleled uses, in the nuclear age. And I'd rather grill than irradiate my dead cow this weekend anyway
Five major points about system memory compression:
1. Why do you want to do it Is it because RAM is expensive? Okay, but RAM prices would have to climb much higher to make it worth the new boards, new architectures, and intrinsic problems
Is it because your system will 'run faster with more RAM'? Don't count on it. Trading RAM latency for apparent RAM size will mean that a given apparent RAM size would run slower than when uncompressed (i.e. 64->128MB compressed is slower than 128MB uncompressed), and the performance gain would be variable for a given *true* RAM size (= larger apparent size), and may disappear in certain settings (Is 64MB->128 MB faster than 64MB compressed? Depends.)
Remember, CPUs are data hungry critters, and feeding them at one end (and emptying them at the other) is already one of the biggest challenges of modern system design.
2. Transparency is not enough. We need ultra-transparency Remember: any general purpose compression yields variable results with different data (and changing a bit will change the 'actual size' of a block, and hence the physical location oif the bytes within the block. Compression confounds the 1:1 correspondence between the physical and logical memory address, and the relationships between different memory addresses, so we'll need to de/compress entire blocks and cache them (more on this later)
Without ultra-transparency, optimizing low level code becomes fraught with emergent effects. the most important thing about RAM is the capability for RANDOM access a lot of people have forgotten there was any other kind -- serial memory, bubble memory, etc. -- bucket-brigaded bits demanded very different algorithms for efficiency!
Think of the pitfalls of straight CPU/mobo caching designs. 'Cache thrashing' can bring some fast algorithms to a molasses crawl, precisely because caching disrupts the relationship between between contiguous bytes: A slower algorthm that reuses bytes in cache is preferred over a 'mathematically faster' one that relies on massive sequential reads. Compression thrashing can do the same, and will have (multiple) cache problems on top of this..
3. Where did I put that block? There is no assurance that you will be able to return a block to the same DRAM location you got it -- change a few bytes, and it may be larger (in physical length) than it was, despite having the same virtual length. This implies RAM fragmentation -- and all the associated housekeeping. And where are you going to store the record keeping? In *another* local cache? In RAM?
You'll need lots of hardware housekeeping here. It's do-able, but without a level of sophistication that approaches predictive brancing and pipelining, you can count on extensive unexpected 'emergent effects' -- code that's slow for non-obvious reason, or bugs.
4. Strangely, on-chip cache may be the best place to use hardware RAM compression! a) Hardware compression would be a small addition to the CPU circuitry, and can be run at the full multiplier speed. b) No need for separate chips or mobo redesign c) on-chip cache costs hundreds of dollars a meg (price a 512K PIII vs a 2MB PIII Xeon of the same clock speed) so extraordinary measures can be taken. It will probably improve chip yield, too. d) integrating compression with the prediction/pipelining/cache management/etc of the CPU, can make it more transparent e) L1/L2 is where you will get *huge* payoff, by 'keeping baby fed'. f) there's much less latency (chipset, PC trace, L1, L2, L3 cache) on-chip vs. off chip, so Adding an off-chip layer adds more latency than adding an on-chip layer
5. The fundamental performance rule is: trade excess performance in one place to improve inferior performance elsewhere RAM is no longer a source of 'excess performance', it's a pressure point. Every extra 10MBps in RAM throughput shows up in the benchmarks (unlike, say HDD busses like ATA33/66/100, where doubling or tripling speed does little for system performance.
Where's the excess speed in modern systems? It's inside the CPU -- which runs at a multiplied clock speed, has vast optimization, and is always starved for data. it's also the place where adding RAM does the most good.
However (!) using compressed on-chip cache will require an intensive study and redesign of cache theory, unless this possibility has already been explored in conjunction with the development of the current CPU features like predictive/pipelining/VLIW. 2-,4-, 8-way associative simply won't hack it when the cache doesn't have 1:1 virtual/physical data correspondence!
The article doesn't seem to reveal the identity of the "major software house" but it's probably good to have a test case.
Let's not assume that the "major software house" is our enemy. (I can see you didn't) It is not unheard of for the ACLU, for example, to deliberately bring a 'friendly test case' with the cooperation of both plaintiff and defendant when it wants to establish a point of law or strike down an unconstitutional law.
This is often necessary because the ACLU does not have 'standing' (direct involvement) and the courts are set up to administer justice in actual cases, not to pontificate on philosophical issues.
However, that does not mean that the case will not be vigorously argued by both sides. A test case is of little use if it doesn't
There is no problem updating the GPL to fix loopholes, because almost all GPL'ed code is specifically licensed under 'Version X or later'...
This is correct in substance (good work!), but I feel I should clarify the details, because they are not all good. First let's look at the relevant section of the license:
8.... this License incorporates the limitation as if written in the body of this License.
9. The Free Software Foundation may publish revised and/or new versions of the General Public License from time to time. Such new versions will be similar in spirit to the present version, but may differ in detail to address new problems or concerns.
Each version is given a distinguishing version number. If the Program specifies a version number of this License which applies to it and "any later version", you have the option of following the terms and conditions either of that version or of any later version published by the Free Software Foundation. If the Program does not specify a version number of this License, you may choose any version ever published by the Free Software Foundation.
Firstly, an 'unspecified' GPL opens up the software to the loopholes in all versions of GPL. Why was this done? Probably as a protection against one or more versions being struck down or interpreted to limit 'openness'. It is not entirely clear whether a free license may be revoked (invalidated for an individual product) by the author, since 'valuable consideration' (value recieved for value granted) is not specified, aside fraom a general intent of helping Free Software. Under general contract law, a contract with no valuable consideration may be later voided at the choice of the granting party - or sometimes either party. 'Goodwill' may or may not be 'valuable consideration' depending on the circumstances (usually, "did the party reasonably expect direct presonal gain as a result of this goodwill" - getting "laid or paid", not warm fuzzies)
Secondly, the license does not say 'the latest version' of GPL, but 'any later version' according to the user's choice - and the most likely reason for creating a v4 or v3.1 is to address loopholes in v3. Let us not forget that a corporation could also be a 'user' of the source.
The general wording in GPL encourages propagation of loopholes. Why? Because copyleft is itself a sort of 'loophole' in current 'normal' copyright use. Open software is best served by granting many rights and secondarily served by invoking a few carefull chosen restrictions. I think the FSF chose wisely in this case, but it was a devil's choice between keeping the primary source open (no matter what happens in court) and keeping all derivatives open forever.
This is just one reason (of several) why little in the GPL is clearly "no problem". Compromises were made. Good ones, but blind ones, in the absense of court rulings. A court ruling could be very specific to the case, or could begin to give us the information we need to start crafting a better license (but no single case will probably suffice, certainly not the first. Judges can be wary of untrodden turf -- if they see it as such -- most judges *don't* want to make precedent.)
I have been very surprised that the issue of alternate provider access question gets so little attention and activist concern. Phone and power were one luxuries, as bandwidth is today, yet few would argue that these utilities are not 'essential services', as critical to commerce and as potentially life-saving as the (generally gov't maintained) roadways. It would also be difficult to argue that infrastructure maintanence (long term) is been better in the effective monopoly that takes care of our crumbling roadways than in the competitive phone and power grids.
This morning, President Clinton made an inte rnet-only speech. If you weren't on line, you missed it. Sure, it'll be a sound-bite on tonight's news, but in a few months it'll be as routine and boring as Space Shuttle lauches. Communications and information have always been essential legs of democracy. We cannot tolerate a situation where a different message can be so easily fed to the educated and connected 'haves' while being withheld from 'have-nots who may barely be able to afford a $35 second-hand 486. Also,in our history, even localized monopolies have rarely existed without abuses serious enough to bring public outcry and legislative action (the exception being government monopolies, like roads and mail -a fate I don't want to see for the internet)
(Recall that in the Kennedy-Nixon debates, low bandwidth radio listeners generally felt Nixon won, while high-bandwidth TV viewers gave the victory to Kennedy. Like it or not (I don't), in too many ways to count, the medium impinges the message to such a extent that Marshal McLuhan concluded "the Medium is the Message.")
Internet access will become more crucial in the next decade. Business will not merely adopt but change its nature with widespread telecommutting, just as it did with automotive commuting in the past half century. It is difficult to imagine what life was like when there was only "urban" and "rural" (often separated solely by a cityline boundary) and one rarely shopped outside ones neighborhood, creating a near-captive market.
Yet we let the cable companies continue using an argument that has fallen in essentially every other industry: ownership of the Last Mile.
In the 1700's, most good long distance roads were private due to the cost of construction and maintanance. This was more a service model than a product model, paid through private tolls, bond levies, or subscription. This was a major US industry! Transportion companies (stagecoaches, shippers, etc.) chose the best roads and bridges to patronize, based on their budget and needs, but land travel was rarely the best choice for long diatance transport -- rivers and oceans were. (We all know the story of how San francisco "49'er" miners found it cheaper to send their laundry to Hawaii than to send it overland out of the gold rush zone.
To show how different this system was: when the car entered the scene, Auto clubs sprang up to create maps and routings of the road network: community- or neighborhood- maintained and private Such maps/routes had previously existed only in the heads of professional transportation workers. Each club painted its own mark and route numbers on landmarks along recommended routes to/from major destinations. (The handbook might read "To enter route #7 to Lancaster, turn left at the creek after the barn by the apple tree. Look for our red Beaver mark!")
As the auto became more important, the government looked at the recent success of the railways, and chose to move away from a commercial model for roads. (Today, we have much more experience regulating large companies and infrasctructure networks than we did in the days of the robber barons) More government roads were built outside town limits, and state highways and federal Interstates (for defense) were built.
At that time, most rural areas were largely self-sufficient for food (and it wasn't until almost WWII that the US ceased to be primarily rural). However, one early need for roads was the provision of medical services (physicians were amoing the earliest adopters of cars in the days of the house call) Similarly, today phone access is all but guaranteed, because they are a lifeline to medical services - try calling an ambulance without one!
For amny reasons, phoines were considered a copelling public good, and phone companies (local bush-lines, medium companies, and of course, Ma Bell, who overshadowed all others) were granted certain legislative advantages to encourage them to wire every home in the country that wanted it. The system paid for itself many times over, and now the local access companies no longer have a true monopoly. They charge for use and maintanance of the 'last mile' copper, but actual services do not have to be purchased from them. You can choose your Long Distance carrier, your DSL ISP, etc., and the local phone compnay collects a few bucks as an access charge.
Later, a similar sequence of events took place in power generation, which uses a 'product delivery' model, than a service industry model, but has the same 'last mile' copper infrastructure. In many states (including mine), the consumer is free to choose any alternate electricity supplier, and pay a few bucks to the local utility to 'deliver' the power to my home. The actual behind-the-scenes work is an elaborate mesh of accounting reimbursements -- reselling bulk power purchases across disparate and sharing the discount with consumers, and the actual power flow is unchanged. Bulk purchases make life easier for the utilities, which is why a factory may pay almost half what you do for their off-peak usage. The middleman, who hires accountants instead of linemen and reactor workers, is happy with taking a few percent of the discount as profit, and passing the rest to the consumers. It's a crazy process, but it seems to work.
Today, look at the monopolies (roads, rail, mail, etc.) and look at the 'alternative provider' industries with the same last-mile problem (phone, power private delivery, etc.) Which do you think provide better services, and are better positioned for future services? A 'sole access' monopoly is not even always more profitable before it is forced to open access -- if Bell had remained the sole phone monopoly, I doubt we would have the range of services and alternative networks that we have today. The money that the many Baby Bells make from their many diverse businesses far outshadows hat they could have made solely from stodgey phone services.
No one likes being made to get up off their butt, but it's often for the best. Contrary to the cable industry's arguments, the captive audience of 'sole access', will always be there tomorrow -- so the majority of cable districts do not have a date set for providing cable internet access. It simply isn't the service they provide. If those customers could -by law- go to other service providers over those cable lines, they'd quickly move to capture the market that was slipping between their fingers
I wanted to write a filk: "Listen to the rhythm of the line printers".
Hey, Bubbala, for a fellow/. er, anything.... after all, Orpheus (in Greek Mythology) was not just the guy who snagged the rad Eurydice, he was a mega-jammin' musician first
Listen to the Rhythm of the Line Printer To the tune of "Rhythm Of The Falling Rain" by The Cascades
Listen to the rhythm of the Line Printer It giving me a terminal headache I wish that I had thought before I hit 'Enter' Cause the feedpath screws up when I 'Break'
The user queue is lengthy for for the batch laser, And management won't buy a broken part. So I'm stuck with this Goliath of a Line Printer That grew up drawing ASCII art.
Hey, please tell me now does that seem fair? The sysops all are underpaid, but they don't care I can't track a cracker when log output takes half a day
The user queue is lengthy for for the batch laser, And management won't buy a broken part. So I'm stuck with this Goliath of a Line Printer That grew up drawing ASCII art.
There's Pain in my temples and it just won't go. I'm going to light the paper, set this place aglow Take the mainframe apart, then maybe our budget will grow
Listen to the rhythm of the Line Printer It giving me a terminal headache I wish that I had thought before I hit 'Enter' Cause the feedpath screws up when I 'Break'
Oh listen to the falling rain pitter-patter, from the sprinklers, ooo-oo-oo Listen, listen....
As I have noted in earlier, more detailed posts, like this one, this agreement is simply a classic example of the 'patent sharing' (not always entirely voluntary) that is standard practice within the Japanese patent system. I'm not saying that it is entirely innocuous -- much of the Japanese system of doing business could be framed in terms that would provoke extreme outrage from much of the Western geek community (and indeed the business community). However, I *am* saying that it is not only part of 'business as usual' today, but has been an element of the electronics industry, back before the first cheap (and shoddy) Japanese transistor radios -- i.e. all our lives! It's not a disturbing new trend, if anything, it's milder than it once was.
Yes, Rambus isn't Japanese -- but look at its list of partners (recently divided up, regrettably into separate pages for each technology)
This system has several major goals: preserving hierarchy and the Japanese business structure (which is very different than ours, with strong, almost monopolistic vertical integration, pan-industry consortia, and many intermediate layers of supernumary distributors), 'maintaining relationships' in the Japanese sense, perhaps most importantly allowing Japanese industrial development in the face of foreign patents (in the early decades).
It is an expression of their culture, which retains very significant feudal elements. It would be disruptive and disrespectful to expect them to instantly adopt *our* values, as if we had some intrinsic superiority. If they even attempted this, to a greater degree than MITI already does, it could massively disrupt their economy and society (e.g. the supernumary distributors are a major part of the economy and cannot be eliminated easily; also, their approach to lifetime career and company affiliation could not be more alien to Silicon valley, where even founders like Steven Jobs leave, compete, return, etc.)
Sadly, my first reaction on seeing the article on the front page was "Does Trixie included?"
Don't waste your time guys, according to the seller, Trixie is now a 50 year-old lap dancer in Kansas City, site of the Slashdot Anti-PETA BBQ. He did try to offer me a deal on a slightly used Chim-Chim, but I didn't want to know what 'slightly used' entailed.
A of course NAL, bit my understanding of this decision is that it was just upholding an injunction. The full trial on all the information hasn't been heard yet, this is just a ruling on a motion about how things will remain during a trial.
Although, my understanding of injunctions was that they were supposed to be quick by their nature and purpose, and not drag on for years. I didn't even know that they could be appealed...
While I've only begun my research, here are a few resources that you -- and others should probably look into.
(It takes me a while to dig and digest info before I reply. I wonder how many people come and go before I can post)
I've lined up webspace with CGI capability, etc. and no advertising via a friend who has an under-utilized site at a hosting service. We've written up an agreement to cover the basics and assure the site stays up, even if there are some hassles along the way. I'll only have to relocate if I cause *operational* problems at his site. He likes the project, and corporate pressure (for example) wouldn't faze him.
He asked me not to announce the URL until I had a debugged site to put up (to give him a chance to give security a once-over first) and since this article is going to drop off the radar soon, I got a GeoCities page, where I will set up a FAQ, a feedback section, and announce the regular URL when it goes beta (Has any living website ever left beta?)
So after, say, Saturday (24 June 2000) , you may want to check in at the temporary page to see how things are coming. Bookmark it for future reference.
The single most immediately pressing question is "What do we call it?" It may not be the actual Award name (which will be the first order of business on the site, along with categories), but I can't keep calling it 'Coxsackie' forever (Coxsackie virus causes Hand-Foot-Mouth Disease. It's my personal term for projects I know I'll regret volunteering for. It also makes a moderately satisfying curse)
Which is a non-issue since by region price fixing is illegal under the WTO.
This is an excellent point! Why haven't we heard more about this on the many/. debates?
I'd appreciate any citations you might have, or even the names of documents (The WTO, which whom, I have been recently corresponding, does not index all of its downloadable documents for maximum user-friendliness. You often have to know which documents you want, download, and then search them yourself)
"Enforcement of antitrust laws is rigorous, as witnessed by the large number of ongoing investigations and actions taken to combat price-fixing, predatory pricing and exclusionary pacts involving major U.S. and foreign companies. Enforcement of laws protecting intellectual property rights (IPRs) is also rigorous, so as to ensure adequate returns for investment in innovation."
How's that for wry wit? I think it could apply to the MPAA as well as it does to MS-AP (Microsoft Appeals - their real business for the time being)
I believe that the Academy Awards are a fine idea. I don't fault the Nobel Prizes for their choice of subject matter (Economics, for ecample, must've seemed a shining star when the award was created, yet the Nobel's own awards show its unfulfilled promise -- i.e. in Physics or Medicine consecutive laureates always agree on 99.9%+ of their subject; in economics, they may disagree on their most central tenets)
However, I think that this noble concept -- to advance the art by recognizing its finest work -- has been caught in a common quandary. changing too much is disruptive (as in the religious debates of baseball and other sports over decades old rules changes and equipment technology) and difficult. Consistency has value in the judging process and allows competitors to know precisely where they stand. The patina of age and tradition also serves the dignity and value of the awards.
But change is also necessary. Perhaps the conflict is intrinsic. Perhaps, despite the Academy's efforts to recognize advances in the science and technology of film, it is inevitable that an award for 'film' will pass the way of national awards for artistic heiroglyphics or penmanship. there are already major motion pictures in theatrical release that are displayed digitally on electronic screens. It is difficult or impossible to argue that inherently pixellated films like Toy Story (an Oscar winner) are any more film-like than an Australian indy production that was disqualified because its live actors were recorded on videotape instead of emulsion.
The numerous flaws of the Academy's voting system are well known (e.g. it is universally agreed that most members have not seen even a sizable minority of the candidates, and vote based solely on publicity). We at Slashdot have seen similar issues -- to the extent that we scrutinize the process and read 'hidden' discussions like sid=lostkarma, sid=moderation, and sid=metamoderation.
As far as the internet ruling goes: it is merely an explicit elaboration of a rule that has existed from the beginning. The foreign film rule, however, seems to be an accomodation of changing realities that do not violate fundamental tenets of the Oscars (it is a recognition that LA is not the alpha and omega of the film world) I cannot condemn them for their decisions.
We really do need to establish an award for Internet Art, that will stand alongside the Emmys and Webbies of the future. The idea is not original to me. I've read it in this very thread.
However, since this project will not create itself, if there is sufficient interest, I am willing to commit the resources (time, money, access, programming) necessary, including a website to be established by the end of the July 4th holiday weekend. I would appreciate input and assistance in identifying the categories and nominees, criteria and structure, fixing on a name, locating suitable judges, etc. (though public voting can and should play a role, I am not sure the 'standard' web voting site is suitable to be the sole element of voting at present. I have seen too many abuses and flaws in that system) and other areas. I look forward to such feedback, and anticipate turning to Slashdot frequently as this noncorporate venture proceeds.
I wish that they had a picture of the unit, but what I am envisioning from the description is a box with little more than a 'off' switch and a display in front (it doesn't need either, but they are cheap, and would aid marketing at this price point - $250 w/o system purchase)
The real magic would be the 'back panel', standard output jacks to mate with a home stereo, which unfortunately will probably just be phono plugs and clamp-ons for bare speaker wire. It's really too bad, because believe me, if they added stereo 'balanced input' jacks, I might even suggest it some people I know. getting a decent balanced in to a computer, even one with a studio-style music card (not a Soundblaster) is pretty expensive.But that would make the device and I/O device, and I see no suggestion that it's anything but an output device.
I also wouldn't immediately assume it's Ethernet or anything like that. It could be USB.
It's primary advantages (if I envision it correctly) are:
0) Fiddle free operation: i could (and have, in the course of helping people with home studios) create some of all of the advantages below without adding components, but I like hacking. And because it was important to get it 100%, I noticed how many apparent solutions weren't 100% for various reasons
1) Skip protection -- face it, HDDs are *slow* to change tracks. I doubt there's an IDE HDD around that can't be pushed to 20-30ms *worst case* track seeks. Don't bother checking your specs, NO ONE publishes worst case numbers, and even their 'typical' and 'capability' (best case) figures are unstandardized between manufacturers and suspect, technically, as savvy HDD review sites like Storage Review will happily show you, via exhaustive testing and comparisons. So when you're doing a few things at once, skips aren't unheard of, even at modest CPU loads. I rarely notice them, and they don't bother me, but they are there. A meg of buffer at the output device is cheap and easy anti-skip pretection.
2) Better sound reproduction -- Yeah, you could buy a better sound card, but as great as those can sound, a dedicated device can be better, especially when driving a stereo amp/speakers . I'm no sound snob personally, but I help a lot of home studio musicians, and the difference is easy to hear, even for me. Similarly, few PC speakers come close to a goo set of home stereo speakers. Even some $300 'big name' (you'd recognize it instantly) USB speakers I tested recently were very disappointing.
3) Device segregation: There are actually good reasons why a normal user mught be better off keeping a 'game style' (FMsynth/MIDI/WAV) card like a soundblaster as his primary audio device, rather than a killer studio-style card -- I have a friend who has separate semi-pro MIDI and digital audio cards, and a cheap Ensoniq for apps/surfing, so he doesn't have to power up his whole studio for normal computing. Such a set-up tends to create confusion in apps that expect a SB compatible, too. Sure, you could do this with USB speaker drivers, but see above re: those
4) Looks cool on shelf: don't knock it
5) allows full utilization of existing Home Entertainment components (which, in many houses doesn't revolve around the PC, as strange as that sounds) Unfortunately, this is merely a theoretical advantage. This product seems clearly targeted at the market that will buy first (computer users) and will probably not even talk to other home stereo components, aside from passing on the audio analog signals.
However, having said all that, I am utterly unmoved by this product, and I expect most users will be, too. What I'd like is a "digital entertainment station" - bidirectional, so it would be able to pipe me radio, TV, good balanced mike input etc. as well as piping my decompressing and D/A'ing mp3 (and WAV) stream to the speakers.
There is a market out there for this. Lots of companies sell this stuff for Beaucoup bucks. But Dell is going for a mass market portal model for revenue and expansion
Forgive me Lubeck Streyer (aka 'lubie-babie', when I was in molecular bio) and St. Lehninger, for the sins I am about to commit...
I really feel I ought to explain what they are calling 'sugars' here -- it's a biochemical term that (very crudely) boils down to 'a chain of carbons with water attached' -- only the 'water' has broken into two parts ( HOH => H + OH ) and these two parts connect to the carbon, instead of each other. you can think of a 'sugar' as a chain of carbon groups that look like HO-C-H and are connected to each other at by the carbons like this: ......._____........H............ ....../.....|....HO-C-H..___......__....... ...HO-C-H...|........\../...|.../..\...... ...HO-C-H...|......HO-C.....|....\__/\..... ...HO-C-H...|......HO-C-H...|.........\__.. ...HO-C-H...|......HO-C-H...|........./..\. ...HO-C-H...|......HO-C-H...|.........\__/. ....H-C-OH..|.......H-C-OH..|.............. ......\_____|..........\____|...Table Sugar
The big loop just indicates that the carbons are generally in a ring. The second figure indicates that not all of the carbons are always in the ring. The last HO-C-H group is backwards to indicate that major difference between many sugars of the same size is simply whether each -OH group points up or down when we lay the ring flat. "up-up-down-up-up-up" is one sugar, but "up-up-up-down-up-up" is different sugar (they may seem like reflections, but trust me, in a 3-D ring, they aren't)
The 'well-known sugars' (most of which you've never heard of) have carbon chain lengths from 3 carbons (e.g. triose) up to seven carbons (e.g. sedheptulose). However, the 'familiar sugars' are usually based on a six carbon (glucose, fructose, etc.) or five-carbon (ribose) ring. Table sugar (sucrose) consists of *two* six carbon sugars connected together. Chains of sugars longer than two can be very 'un-sugar-like' -- cellulose (wood fiber) is nothing but long linked chains of glucose (blood sugar) while glycogen (a stored fuel in your liver) is also just branched chains of glucose, but is very different physically.
Glucose (blood sugar) or fructose (fruit sugar) are C6-H12-O6. Table sugar (sucrose) is C12-H22-O12. In space conditions, it might be useful to think of carbon chain lengths as being like stacked blocks -- the kind children play with. Generally stacking two blocks is easy, but six is more than three times as hard (it tends to fall apart easily)
The so-called "sugar" they found in space is two carbons long (glycoaldehyde C2-H4-O2) and is very unlike the six-carbon (okay, 5-7) sugars we usually think of. In biochemistry, it isn't generally called a sugar at all. Three carbons was a sort of bottom limit to be sugar like, because the 'ends' often have an extra hydrogen, and a two carbon 'sugar' would be nothing but 'end' and can't form a ring. It's not very 'sugar-like'. It is an extremely simple molecule, that would be easy to make ("stack") by random, and it looks like this (where the = means a double bond) ......H.H.... -- glycoaldehyde, ...HO-C-C=O.. -- the so-called ......H.H.... -- "space sugar"
[Honestly, each time I tried to submit this from a Windoze machine -four tries- the box crashed! Then I switched to a Linux machine and the ISP went down. Fortunately I have a backup ISP. Chill, Bill, it ain't that incriminating!]
Here's the specific section, 15 US Code 29(b), covering "what happens next". It's short and clear.
TITLE 15 - COMMERCE AND TRADE CHAPTER 1 - MONOPOLIES AND COMBINATIONS IN RESTRAINT OF TRADE Sec. 29. Appeals (b) Direct appeals to Supreme Court
An appeal from a final judgment pursuant to subsection (a) of this section shall lie directly to the Supreme Court, if, upon application of a party filed within fifteen days of the filing of a notice of appeal, the district judge who adjudicated the case enters an order stating that immediate consideration of the appeal by the Supreme Court is of general public importance in the administration of justice. Such order shall be filed within thirty days after the filing of a notice of appeal. When such an order is filed, the appeal and any cross appeal shall be docketed in the time and manner prescribed by the rules of the Supreme Court. The Supreme Court shall thereupon either (1) dispose of the appeal and any cross appeal in the same manner as any other direct appeal authorized by law, or (2) in its discretion, deny the direct appeal and remand the case to the court of appeals, which shall then have jurisdiction to hear and determine the same as if the appeal and any cross appeal therein had been docketed in the court of appeals in the first instance pursuant to subsection (a) of this section.
Well, we simply can't conduct this thread without a link to the actual ruling at the DoJ site. (PDF only)
Here's the actual text, as sloppily transcribed by ol' stumblefingers (me). Beware of possible typos.
[BTW, is 'general public importance sufficient to meet the standards of 15 USC 29(b) -- gotta check!]
Upon consideration of the joint motion of the plaintiffs for certification pursuant to 15 USC 29(b), and the opposition of Microsoft thereto, it is, this 20th day of June, 2000.
ORDERED, that the Joint motion by the plaintiffs for certification pursuant to 15 USC 29(b) is granted and that this court hereby certifies that immediate consideration by the Supreme court of the appeal taken herein is of general public importance in the administration of justice' and it is
FURTHER ORDERED, that the Finalk Judgement of June 7, 2000, is stayed in its entirety until the appeal therefrom is heared and decided, unless the stay is earlier vacated by an appellate court.
Since the article specifically Refers to the Aussie situation (mandatory porn filtering by ISPs), here's what the final report of the Australian Government (National Office for the Information Economy) has to say about the weaknesses of this approach in their review of blocking technologies entitled Access Prevention Techniques for Internet Content Filtering (Google cache) :
The quest to detect pornography is often more concerned with images than text and getting computers to recognise a pornographic image is equally, if not more, difficult than the task of distinguishing between erotica and other literature. There have been efforts to characterise pornographic pictures based on the amount of 'flesh tone' in the images and on the poses struck by the subjects. Computers have difficulty distinguishing between art and pornography, say between a Rubens painting and a Penthouse centrefold, and this approach is not generally regarded as being effective. Video and other streaming media further complicate the filtering task by supplying a constant flow of images to be examined for undesirable content.
Furthermore, they complain:
This approach is affected by the same issue as profile filtering in that an image - or a fair percentage of an image - needs to be loaded before it can be analysed, during which time it may be displayed on a user's screen..
Of course, this second problem only applies to an Aussie-type ISP restriction. Geocities did this years ago (don't know if they still do): scanning their own HDDs (Free user pages), deleting 'questionable graphics' (with or without human review) and waiting for the page authors to complain about any mistakes,
The Article had so many technical, philosophical, mathematical and other misconceptions in the article (just a few listed below), that it could pass for a modestly well crafted troll. It had 'something for everyone' (i.e. anyone should be able to poke *some* hole in it, with a moment's thought), making it both an 'obvious troll' and 'good bait'.
At first, I thought the author was sincere, but then I noted that he actually reversed and misrepresented its flaws as *strengths* (e.g. 'the birthday effect' in namespace collisions)
How did this article get on the front page of SlashDot? <sarcasm> Is it supposed to be a sly social analysis, a wry deconstructionist experiment or dry Gallic humor? I wonder why it's under censorship rather than crypto -- could it be footnote #6, below? Must be, else this ubmission would never be the cream of the crop. </sarcasm>
[1] "Free speech" is only meaningful when it can be widely heard. Perfect encryption without public decryption is like locking yourself in a trunk and throwing away the key. If every Joe Sixpack and Dexter Tapedglasses can read your message without prior arrangement, so can Joe Gannon and Janet Reno. if JS and DT can't read it, it ain't 'free speech', its 'private communications'.
[2] The only privacy insight here is the obvious fact that "encrypted files may look like garbage" (regardless of encryption method) However *cleverly* encrypted files, e.g. steganography, may look like something utterly harmless. Which approach is safer/more secure for the originator, the storage site, and the recipient? Especially in the light of laws like England's mandatory key surrender on (proper) demand. Someday, keeping massive Porn databases may be your duty as a patriot!;-> How else can we stop the jackbooted thugs from finding/blocking our 21st century Federalist Papers?]
[3] While independently assigned padnames of 8 bytes may offer 2^64 names, there is a 50% chance of collision after relatively few pads are generated (i.e. millions). The birthday problem the article mentions doesn't suggest high freedom from collisions (as he implies), it means collisons are much likelier than we expect: if there are 24 people in a room, it's *probable* (>50%) that there'll be a birthday collision (shared birthday) even though there are 366 possible days in the dataspace. He cites this as proof that collisions will *not* be a problem
[4] The system loses the ability to decode as more random pads are created/shared and collisions begin to occur. Since pad generation is uncontrolled, this method would become an information hole -- if you used the 'wrong' #6930d3ed740d54de for a given file, you'd get gibberish -- yet all pad #6930d3ed740d54de are equally valid. The system he calls "A whole Mess 'O' Pads" would degrade to "A Whole Mess" (of bits) -- an effective information hole.
[5] At its best, this inept rendition of a one-time pad is a Geek Pig Latin [GPL??], reducing the encription value of a theoretically UNBREAKABLE 128K one-time Pad to a *theoretical* maximum of 2^(64*n) combinations [where n is the number of OTPs XOR'd together. and a minimum that is no more than x^n combinations for brute force cracking [where x=number of published pads, n= number of XORs]
You can best think of it as a poor key generation method, where the true key is not the 128K pad, but the far shorter 'instructions' -- the keynames to XOR together. The example he gave (6 XORs, 8 byte keynames) amounts to the same security as XORing against a 384 bit key, as far as a brute force attack is concerned. This is the same security as XORing against "Netscape Engineers are Weenies! They really are!!" (48 bytes)
[6] Perhaps this article can be most charitably read as an experiment in information darwinism, but not in the Dawkinsian 'meme' sense: the speaker who uses this method is 'too dumb to be listened to' (and silenced by disappearing into the 'Whole Mess'O'Bits) -- akin to the sardonic 'too dumb to live'. (This is supported by his assertion that he is not sure free speech is a good thing)
Of course, it'd be in the best interest of GM to just support the hardware out-of-the-box, but we'd need to pressure them to make sure the drivers are under the GPL.
(Dateline: Indianapolis) As part of the pre-race publicity for the Indy 500, actress Natalie Portman (Phantom Menace, Better Living through Circuitry, Heat, Leon) whose movies often include elaborate race/chase scenes, drove a ceremonial lap around legendary Indy Track in a General Motors Indy car.
The brief trip was not without incident. An on-board video camera wrenched itself loose under commands from remote viewers. GM mechanic Alfor Hominy was puzzled, "I didn't even think it was supposed to swivel. Fixed-view is all we need for Indy."
In a related story, Richard Stallman held a press conference praising GM's use of a 'free as in speech' operating system in their cars, but it was abruptly cut short when one reporter asked about the plethora of closed source drivers in the new operating system.
"er... uh... yes... that's true, but..." he said, before breaking down and sobbing "she's not automatically open under GPL!" He was escorted from the stage by Kernel Developer Linus Torvalds, who later explained. Richard has devoted this life to Free Software. The idea of a driver he can't freely access, manipulate, and modify to his wishes rather upsets him."
To dispel the embarrassment, Transmeta CEO Dave Ditzel declared the cash bar "free as in beer", which much to the disappointment of the assembled press, turned out to mean $2.50 for a Lite
My initial reaction was that somebody was feling the heat and trying to return the HDDs to stop or deprioritize the investigation. The history of HumInt is full of ridiculous stories of inside sources who were almost caught (but weren't) because an investigation moved on or was deprioritized. The Article suggests DoE was thinking on the same lines.
Despite their safe recovery, Energy Secretary Bill Richardson said Friday afternoon he would continue an investigation into the matter, and he promised disciplinary action would result. Investigators are treating the area where the drives were found as a crime scene, and Richardson said "their authenticity" was still being evaluated.
Maybe the theft was discovered sooner than expected (perhaps due to the post-fire audit) or maybe the original plan was to copy and return the HDD, but of course 'return' is less pressing than the original theft (and almost as risky). You have the data, everything else is icing... until you realize your source is almost compromised!
"authenticity" - I doubt anyone was foolish enough to return a copy of the HDD, so this suggests some clean-room (in the original sense) technique for checking if an archival (written then stored) HDD was ever read, or if *all* of it were read (Normal use would only read parts as needed. If all of it had been read - especially recently - it would strongly suggest copying.)
Does anyone out there have any info on the current capabilities of 'clean-room' HDD analysis? We know the Gov't has impressive recovery methods for recovering long erased and multiply re-written files from off-track traces.
While its not a major budget item for a large company, it's certainly something they would consider, especially since the cost of power consumption (not just the power itself) is considerably hiogher than you estimated
Air conditioning has to dispose of the heat once it leaves the machine. It takes more than 1 watt of energy to move 1 watt of waste heat (if you could move 1 watt of heat for less than one watt of input, on a large scale, you could do stuff that verged on 'free energy' -- i.e. use 200W to move 1000W of waste heat into a 50% efficient heat engine to generate 500W, then, if you really want to push it, use the 500W to... well you get the idea) While new HVAC plants are quite efficient, using 1.5W to remove 1W of waste heat is not unreasonable for the national installed base (HVAC updating is very expensive) when you consider all ancillary costs
An office worker generates only 100 cal (400 joules) per hour of waste body heat. A desktop computer (w/ monitor) can generate up to 225 kcal (250W*3600sec = 900K joules). There's a lot more forced hot air streaming out of your PC than convecting gently from your body. Of course humans and PC's are hardly the only heat sources in a building, but you get the idea:
Buying big A/C capacity cost $100K's. A maintanence contract cost $10K's/yr. It cost $10K's to power the PCs and 1.5 times as much to powert the AC. An AC failure can cost $10's-100K's per hour as Important CPUs are overheated by waste heat from idle CPUs. And not one dollar of all that directly generates new revenue. CFOs hate that.
It adds up to a cost that management may want to cut -- especially when a company is expanding, and the choice is "more AC capacity and costs" vs. "making do with existing AC capacity"
Office leasing makes it worse. Why invest (or co-invest) in infrastructure for a building you don't own?
Over the past 6-12 months, we've seen a lot of articles on the 'patent wars' between Japanese companies (and when you consider the RamBus partners, this is pretty much a treaty in these ongoing wars)
To understand the significance (or lack thereof) in RamBus having patents on both sides of the SDRAM/RDRAM divide, you have to understand the outlook of the Japanese patent system (both its examiners and users)
In Japan, companies tend to patent much smaller details and modifications than are considered patentable in, say, the US. If you patent an "iridium-based' lightbulb, you can expect other companies to patent 'iridium based' bulbs with different bases, shapes, colors, multiple filaments, higher output, lower voltage, etc. "Iridium filament bulb with bayonet attachent for use in battery-powered flashlights" or "Iridium filament stage lights" would not be considered patentable in most of Europe or the US, but I've seen exactly this sort of patent on special bulbs in Japan.
Before WWII, this allowed japanese manufacturers to efectively bypass international patents in the domestic Japan market: "Maybe GE has the patent on that bulb, but we have the patent on green ones!" This would hold up in a Japanese court -- not that many companies tried to enforce their rights in Japanese courts back then. it wasn't a major market, and "Made in Japan" meant cheap and disposable.
In time, The Japanese system led to a strangling network of interlocking patents. Any big company could push you to license your patents, else they'd threaten to enforce the myriad tiny patents they or their corporate allies ('Kureitsu' relations is 'schoolyard politics') held to paralyze you or block your use of your own patent.
Result, lots of sharing and intertwined relationships, which is, of course, how the Japanese preferred to do business, anyway.
Ask anyone who tried to open a Japanese market (or business presence) in the 70's/80's. The culture shock was like a 2x4 to the head. It felt like socially condoned fraud, theft, monopolism, -- and a lot of words I can't print here. Comparisons to the Mafia were common (behind closed doors)
This is simply business-as-usual in Japanese business. It's not corruption or even scheming, it's their culture. "The nail that protrudes must be hammered down" is one of their most famous proverbs, and many of their well-accepted (by society in general) applications of it would be megabuck civil rights violations here.
It's like a Soviet and an American arguing economics in the 50's -- the premises are simply so different that you can't make a single logical step without considering them
Well, there's a difference between opening up the protocol and opening up th AOL servers to other clients (but I presume you could set up you own server, if the project is valuable to your organization -- presumably within Lotus SameTime.)
Personally, when I read the title of your post, the first thing I thought was "OmiGawd - IM spam! There are certainly enough addresses in harvestable form out there. Spammers don't seem to care who they annoy for that.001% sales rate.
Take it a step further - a bot that 'listens' for certain keyphrases, and interjects product recommendations.
DaveThomas: What sort of POS terminals do you think we buy for our 5000 retail outlets? FrankPerdue: We use distributors, and don't do POS JSmith@PhoneyOutfit: For price, reliability, and service, you can't beat CrapCo. An industy secret! Best in the business!
The mind boggles at the possibilities. Especially since opening up the protocol increases the possibility that a spammer, using a custom spam program, could fake being a (e.g. an older) AIM client
Disclaimer: I haven't used AIM in a year, and only used it for limited purposes, like instant communication when collaborating on projects. I may not recall its specific capabilities entirely correctly.
Slashdot Mirror
I mean, when you allow technology like this, which the public does not understand, to become central to the democratic process, you empower a small technological elite who understands the technology and its limitations.
Not meaning any disrespect (since I amd not referring to the previous author, but our community as a whole) but losing control by not understanding technology is no different than losing controil by not understanding history (i.e. how things really work in the long term), politics (how things really work in the short term) or not understanding, or even bothering to read (*gasp*) THE LAW (how things are *supposed to work - and what we are talking about here)
This bill has been discussed here since it was introduced in May, and the 1999 hearing have been on line much longer. The House passed this law two weeks ago and the final version on the law has been online since June 8.
I urge all slashdotters to read this bill (or at least this summary), so we can discuss this intelligently.
Perhaps more importantly, take a look at this list of other currently and recently pending laws on digital security and e- cyber- computer-whatever before Congress. As the people who understand the technology, we need to play an active role in forming these laws and discussion, which means we need to have a clue about the laws under discussion. (Hint: the subject is *never* ipchains or alernative files systems, etc. Those are just incidentals to the law.)
BTW, when reading the linked laws/discussion, please remember even 'dead' legislation is likely to come back. The current bill was formed from the 1999 'Bliley Bill" (HR 1714), and 1999 Senate 761.
I do not in any way mean to detract from the accomplishments presented here (though, of course, it's been presented in varying shades of "Any time now" for decades. I made my first hologram as a Explorer scout in 1977 or so, and at that time it was already a staple of such classics as "The Adolescence of PI")
However, the storage capacity cited is 10GB/cm^3 cubic centimeter) not cm^2 (square centimeter) as you stated. By comparison, given how thin the magnetic films are, I doubt the 75GB HDD even has 1 cc of active storage volume so holograms do not approach the volumetric density of magnetic media.
The key of course is that holograms offer the promise of true volumetric storage, where magnetic media is limited to the thin film [though who can forget the hedelberg group who used a thin film -- namely a roll of commercial adhesive tape to as an optical medium for up to 10GB a few minths ago?)
So, since it comes down to form factor, I'm not excited. The problem with the HDD is the overall associated mechanism, and with 1 GB matchbook 10G-resistant HDDs out there *today*, I'm not sure when I'll ever be excited. There is no reason to expect we'll be carrying 'naked' (or packaged) holographic media, any more than we carry naked (or packaged) *high density* platters today -- and holographic drives may well be larger, more expensive, more fragile, etc. than HDDs in 2003, as well. In 2003, you won't be able to *buy* a new HDD as puny as 10GB, if indeed they are still maanufacturing that size, today.
The external hologram surface may turn out to be sensitive to damage (though I can hope for the use of confocal optics, etc. to image the volume despite surface imperfections or contamination) and the volume may be vulnerable to sunlight.
The hologram technology used here showed promise because it can be multiplexed with different laser colors and at different angles, but the 'clarity' of the signal goes down with the square of the number of channels, until it is unintelligible. This does not bode well for rapid breakthroughs (though if we could predict them, they wouldn't be breakthroughs). Precise alignment is necessary to assure high density, reliable readings. it seems likely that the 2003 holographic drive will be larger, more expensive, and offer no appreciable advantage (aside from ?magnetic insensitivity?, if that counts)
Aside from the probability of actually seeing a production drive someday, I think that several other holographic technologies are more promising. and none of the holographic technologies show signs of exceeding the practical capacities of straight optical media in the predictable future -- i.e. the next three years. Standards, not technical capabilities, block DVD-R from coming out *this year*
Hey, I want my petabyte encrypted keychain as much as the next guy... but, you know, 'fire' still has many unparalleled uses, in the nuclear age. And I'd rather grill than irradiate my dead cow this weekend anyway
Five major points about system memory compression:
1. Why do you want to do it
Is it because RAM is expensive? Okay, but RAM prices would have to climb much higher to make it worth the new boards, new architectures, and intrinsic problems
Is it because your system will 'run faster with more RAM'? Don't count on it. Trading RAM latency for apparent RAM size will mean that a given apparent RAM size would run slower than when uncompressed (i.e. 64->128MB compressed is slower than 128MB uncompressed), and the performance gain would be variable for a given *true* RAM size (= larger apparent size), and may disappear in certain settings (Is 64MB->128 MB faster than 64MB compressed? Depends.)
Remember, CPUs are data hungry critters, and feeding them at one end (and emptying them at the other) is already one of the biggest challenges of modern system design.
2. Transparency is not enough. We need ultra-transparency
Remember: any general purpose compression yields variable results with different data (and changing a bit will change the 'actual size' of a block, and hence the physical location oif the bytes within the block. Compression confounds the 1:1 correspondence between the physical and logical memory address, and the relationships between different memory addresses, so we'll need to de/compress entire blocks and cache them (more on this later)
Without ultra-transparency, optimizing low level code becomes fraught with emergent effects. the most important thing about RAM is the capability for RANDOM access a lot of people have forgotten there was any other kind -- serial memory, bubble memory, etc. -- bucket-brigaded bits demanded very different algorithms for efficiency!
Think of the pitfalls of straight CPU/mobo caching designs. 'Cache thrashing' can bring some fast algorithms to a molasses crawl, precisely because caching disrupts the relationship between between contiguous bytes: A slower algorthm that reuses bytes in cache is preferred over a 'mathematically faster' one that relies on massive sequential reads. Compression thrashing can do the same, and will have (multiple) cache problems on top of this..
3. Where did I put that block?
There is no assurance that you will be able to return a block to the same DRAM location you got it -- change a few bytes, and it may be larger (in physical length) than it was, despite having the same virtual length. This implies RAM fragmentation -- and all the associated housekeeping. And where are you going to store the record keeping? In *another* local cache? In RAM?
You'll need lots of hardware housekeeping here. It's do-able, but without a level of sophistication that approaches predictive brancing and pipelining, you can count on extensive unexpected 'emergent effects' -- code that's slow for non-obvious reason, or bugs.
4. Strangely, on-chip cache may be the best place to use hardware RAM compression!
a) Hardware compression would be a small addition to the CPU circuitry, and can be run at the full multiplier speed.
b) No need for separate chips or mobo redesign
c) on-chip cache costs hundreds of dollars a meg (price a 512K PIII vs a 2MB PIII Xeon of the same clock speed) so extraordinary measures can be taken. It will probably improve chip yield, too.
d) integrating compression with the prediction/pipelining/cache management/etc of the CPU, can make it more transparent
e) L1/L2 is where you will get *huge* payoff, by 'keeping baby fed'.
f) there's much less latency (chipset, PC trace, L1, L2, L3 cache) on-chip vs. off chip, so Adding an off-chip layer adds more latency than adding an on-chip layer
5. The fundamental performance rule is: trade excess performance in one place to improve inferior performance elsewhere
RAM is no longer a source of 'excess performance', it's a pressure point. Every extra 10MBps in RAM throughput shows up in the benchmarks (unlike, say HDD busses like ATA33/66/100, where doubling or tripling speed does little for system performance.
Where's the excess speed in modern systems? It's inside the CPU -- which runs at a multiplied clock speed, has vast optimization, and is always starved for data. it's also the place where adding RAM does the most good.
However (!) using compressed on-chip cache will require an intensive study and redesign of cache theory, unless this possibility has already been explored in conjunction with the development of the current CPU features like predictive/pipelining/VLIW. 2-,4-, 8-way associative simply won't hack it when the cache doesn't have 1:1 virtual/physical data correspondence!
The article doesn't seem to reveal the identity of the "major software house" but it's probably good to have a test case.
Let's not assume that the "major software house" is our enemy. (I can see you didn't) It is not unheard of for the ACLU, for example, to deliberately bring a 'friendly test case' with the cooperation of both plaintiff and defendant when it wants to establish a point of law or strike down an unconstitutional law.
This is often necessary because the ACLU does not have 'standing' (direct involvement) and the courts are set up to administer justice in actual cases, not to pontificate on philosophical issues.
However, that does not mean that the case will not be vigorously argued by both sides. A test case is of little use if it doesn't
There is no problem updating the GPL to fix loopholes, because almost all GPL'ed code is specifically licensed under 'Version X or later'...
... this License incorporates the limitation as if written in the body of this License.
This is correct in substance (good work!), but I feel I should clarify the details, because they are not all good. First let's look at the relevant section of the license:
8.
9. The Free Software Foundation may publish revised and/or new versions of the General Public License from time to time. Such new versions will be similar in spirit to the present version, but may differ in detail to address new problems or concerns.
Each version is given a distinguishing version number. If the Program specifies a version number of this License which applies to it and "any later version", you have the option of following the terms and conditions either of that version or of any later version published by the Free Software Foundation. If the Program does not specify a version number of this License, you may choose any version ever published by the Free Software Foundation.
Firstly, an 'unspecified' GPL opens up the software to the loopholes in all versions of GPL. Why was this done? Probably as a protection against one or more versions being struck down or interpreted to limit 'openness'. It is not entirely clear whether a free license may be revoked (invalidated for an individual product) by the author, since 'valuable consideration' (value recieved for value granted) is not specified, aside fraom a general intent of helping Free Software. Under general contract law, a contract with no valuable consideration may be later voided at the choice of the granting party - or sometimes either party. 'Goodwill' may or may not be 'valuable consideration' depending on the circumstances (usually, "did the party reasonably expect direct presonal gain as a result of this goodwill" - getting "laid or paid", not warm fuzzies)
Secondly, the license does not say 'the latest version' of GPL, but 'any later version' according to the user's choice - and the most likely reason for creating a v4 or v3.1 is to address loopholes in v3. Let us not forget that a corporation could also be a 'user' of the source.
The general wording in GPL encourages propagation of loopholes. Why? Because copyleft is itself a sort of 'loophole' in current 'normal' copyright use. Open software is best served by granting many rights and secondarily served by invoking a few carefull chosen restrictions. I think the FSF chose wisely in this case, but it was a devil's choice between keeping the primary source open (no matter what happens in court) and keeping all derivatives open forever.
This is just one reason (of several) why little in the GPL is clearly "no problem". Compromises were made. Good ones, but blind ones, in the absense of court rulings. A court ruling could be very specific to the case, or could begin to give us the information we need to start crafting a better license (but no single case will probably suffice, certainly not the first. Judges can be wary of untrodden turf -- if they see it as such -- most judges *don't* want to make precedent.)
I have been very surprised that the issue of alternate provider access question gets so little attention and activist concern. Phone and power were one luxuries, as bandwidth is today, yet few would argue that these utilities are not 'essential services', as critical to commerce and as potentially life-saving as the (generally gov't maintained) roadways. It would also be difficult to argue that infrastructure maintanence (long term) is been better in the effective monopoly that takes care of our crumbling roadways than in the competitive phone and power grids.
This morning, President Clinton made an inte rnet-only speech. If you weren't on line, you missed it. Sure, it'll be a sound-bite on tonight's news, but in a few months it'll be as routine and boring as Space Shuttle lauches. Communications and information have always been essential legs of democracy. We cannot tolerate a situation where a different message can be so easily fed to the educated and connected 'haves' while being withheld from 'have-nots who may barely be able to afford a $35 second-hand 486. Also,in our history, even localized monopolies have rarely existed without abuses serious enough to bring public outcry and legislative action (the exception being government monopolies, like roads and mail -a fate I don't want to see for the internet)
(Recall that in the Kennedy-Nixon debates, low bandwidth radio listeners generally felt Nixon won, while high-bandwidth TV viewers gave the victory to Kennedy. Like it or not (I don't), in too many ways to count, the medium impinges the message to such a extent that Marshal McLuhan concluded "the Medium is the Message.")
Internet access will become more crucial in the next decade. Business will not merely adopt but change its nature with widespread telecommutting, just as it did with automotive commuting in the past half century. It is difficult to imagine what life was like when there was only "urban" and "rural" (often separated solely by a cityline boundary) and one rarely shopped outside ones neighborhood, creating a near-captive market.
Yet we let the cable companies continue using an argument that has fallen in essentially every other industry: ownership of the Last Mile.
In the 1700's, most good long distance roads were private due to the cost of construction and maintanance. This was more a service model than a product model, paid through private tolls, bond levies, or subscription. This was a major US industry! Transportion companies (stagecoaches, shippers, etc.) chose the best roads and bridges to patronize, based on their budget and needs, but land travel was rarely the best choice for long diatance transport -- rivers and oceans were. (We all know the story of how San francisco "49'er" miners found it cheaper to send their laundry to Hawaii than to send it overland out of the gold rush zone.
To show how different this system was: when the car entered the scene, Auto clubs sprang up to create maps and routings of the road network: community- or neighborhood- maintained and private Such maps/routes had previously existed only in the heads of professional transportation workers. Each club painted its own mark and route numbers on landmarks along recommended routes to/from major destinations. (The handbook might read "To enter route #7 to Lancaster, turn left at the creek after the barn by the apple tree. Look for our red Beaver mark!")
As the auto became more important, the government looked at the recent success of the railways, and chose to move away from a commercial model for roads. (Today, we have much more experience regulating large companies and infrasctructure networks than we did in the days of the robber barons) More government roads were built outside town limits, and state highways and federal Interstates (for defense) were built.
At that time, most rural areas were largely self-sufficient for food (and it wasn't until almost WWII that the US ceased to be primarily rural). However, one early need for roads was the provision of medical services (physicians were amoing the earliest adopters of cars in the days of the house call) Similarly, today phone access is all but guaranteed, because they are a lifeline to medical services - try calling an ambulance without one!
For amny reasons, phoines were considered a copelling public good, and phone companies (local bush-lines, medium companies, and of course, Ma Bell, who overshadowed all others) were granted certain legislative advantages to encourage them to wire every home in the country that wanted it. The system paid for itself many times over, and now the local access companies no longer have a true monopoly. They charge for use and maintanance of the 'last mile' copper, but actual services do not have to be purchased from them. You can choose your Long Distance carrier, your DSL ISP, etc., and the local phone compnay collects a few bucks as an access charge.
Later, a similar sequence of events took place in power generation, which uses a 'product delivery' model, than a service industry model, but has the same 'last mile' copper infrastructure. In many states (including mine), the consumer is free to choose any alternate electricity supplier, and pay a few bucks to the local utility to 'deliver' the power to my home. The actual behind-the-scenes work is an elaborate mesh of accounting reimbursements -- reselling bulk power purchases across disparate and sharing the discount with consumers, and the actual power flow is unchanged. Bulk purchases make life easier for the utilities, which is why a factory may pay almost half what you do for their off-peak usage. The middleman, who hires accountants instead of linemen and reactor workers, is happy with taking a few percent of the discount as profit, and passing the rest to the consumers. It's a crazy process, but it seems to work.
Today, look at the monopolies (roads, rail, mail, etc.) and look at the 'alternative provider' industries with the same last-mile problem (phone, power private delivery, etc.) Which do you think provide better services, and are better positioned for future services? A 'sole access' monopoly is not even always more profitable before it is forced to open access -- if Bell had remained the sole phone monopoly, I doubt we would have the range of services and alternative networks that we have today. The money that the many Baby Bells make from their many diverse businesses far outshadows hat they could have made solely from stodgey phone services.
No one likes being made to get up off their butt, but it's often for the best. Contrary to the cable industry's arguments, the captive audience of 'sole access', will always be there tomorrow -- so the majority of cable districts do not have a date set for providing cable internet access. It simply isn't the service they provide. If those customers could -by law- go to other service providers over those cable lines, they'd quickly move to capture the market that was slipping between their fingers
I wanted to write a filk: "Listen to the rhythm of the line printers".
/. er, anything.... after all, Orpheus (in Greek Mythology) was not just the guy who snagged the rad Eurydice, he was a mega-jammin' musician first
Hey, Bubbala, for a fellow
Listen to the Rhythm of the Line Printer
To the tune of "Rhythm Of The Falling Rain" by The Cascades
Listen to the rhythm of the Line Printer
It giving me a terminal headache
I wish that I had thought before I hit 'Enter'
Cause the feedpath screws up when I 'Break'
The user queue is lengthy for for the batch laser,
And management won't buy a broken part.
So I'm stuck with this Goliath of a Line Printer
That grew up drawing ASCII art.
Hey, please tell me now does that seem fair?
The sysops all are underpaid, but they don't care
I can't track a cracker when log output takes half a day
The user queue is lengthy for for the batch laser,
And management won't buy a broken part.
So I'm stuck with this Goliath of a Line Printer
That grew up drawing ASCII art.
There's Pain in my temples and it just won't go.
I'm going to light the paper, set this place aglow
Take the mainframe apart, then maybe our budget will grow
Listen to the rhythm of the Line Printer
It giving me a terminal headache
I wish that I had thought before I hit 'Enter'
Cause the feedpath screws up when I 'Break'
Oh listen to the falling rain
pitter-patter, from the sprinklers, ooo-oo-oo
Listen, listen....
Symphony for Dot Matrix?
Man-oh-man, it must be Friday, 'cuz I read something totally different
Sympathy for Dot Matrix
(to the tune of "Sympathy for the Devil" By Mick Jagger/Keith Richards)
Please allow me to introduce myself
I'm a past that you must face
I've been around for a long, long year
Stole many a man's soul and faith
I was around when TRS-80s
Had their moment of dubious fame
Made Damn sure that Tandy
Washed their hands and sealed your fate.
CHORUS
Pleased to meet you
Hope you guess my name
But what's puzzling you
Is the nature of my game
I stuck around in adding machines
When the computer saw the time to change
I handled carbons and NCR's
As the lasers screamed in vain
Built like a tank
Held a general's rank
When line printers raged
And the toner stank
CHORUS
I watched with glee
While compatibility
and the price you paid
Were the laser's grave
You always knew
What screwed your CRT
It was EMI
From the DMP
Let me please introduce myself
I'm a past that you must face
And I'm the best for preprinted forms
That can't be filled in any other way
CHORUS
Just as every box is a terminal
Most of your print queue is text
I'm noisy as Hell
Just call me Lucifer
'Cause for some jobs I'm still the best
So if you meet me
Have some courtesy
Have some sympathy, and some taste
Use all your well-learned politesse
Or I'll lay your forms to waste.
CHORUS
Tell me baby, what's my name
Tell me honey, baby guess my name
Tell me baby, what's my name
Tell you one time, you're to blame
Ooo, who -- Ooo, who -- Ooo, who -- Ooo, who -- who
Ooo, who -- Ooo, who -- Ooo, who -- Ooo, who -- who
Oh, yeah
What's my name
Tell me, baby, what's my name
Tell me, sweetie, what's my name
Ooo, who -- Ooo, who -- Ooo, who -- Ooo, who -- who
Ooo, who -- Ooo, who -- Ooo, who -- Ooo, who -- who
Oh, yeah
As I have noted in earlier, more detailed posts, like this one, this agreement is simply a classic example of the 'patent sharing' (not always entirely voluntary) that is standard practice within the Japanese patent system. I'm not saying that it is entirely innocuous -- much of the Japanese system of doing business could be framed in terms that would provoke extreme outrage from much of the Western geek community (and indeed the business community). However, I *am* saying that it is not only part of 'business as usual' today, but has been an element of the electronics industry, back before the first cheap (and shoddy) Japanese transistor radios -- i.e. all our lives! It's not a disturbing new trend, if anything, it's milder than it once was.
Yes, Rambus isn't Japanese -- but look at its list of partners (recently divided up, regrettably into separate pages for each technology)
This system has several major goals: preserving hierarchy and the Japanese business structure (which is very different than ours, with strong, almost monopolistic vertical integration, pan-industry consortia, and many intermediate layers of supernumary distributors), 'maintaining relationships' in the Japanese sense, perhaps most importantly allowing Japanese industrial development in the face of foreign patents (in the early decades).
It is an expression of their culture, which retains very significant feudal elements. It would be disruptive and disrespectful to expect them to instantly adopt *our* values, as if we had some intrinsic superiority. If they even attempted this, to a greater degree than MITI already does, it could massively disrupt their economy and society (e.g. the supernumary distributors are a major part of the economy and cannot be eliminated easily; also, their approach to lifetime career and company affiliation could not be more alien to Silicon valley, where even founders like Steven Jobs leave, compete, return, etc.)
_________________
Is the chick on the picture included?
Sadly, my first reaction on seeing the article on the front page was "Does Trixie included?"
Don't waste your time guys, according to the seller, Trixie is now a 50 year-old lap dancer in Kansas City, site of the Slashdot Anti-PETA BBQ. He did try to offer me a deal on a slightly used Chim-Chim, but I didn't want to know what 'slightly used' entailed.
A of course NAL, bit my understanding of this decision is that it was just upholding an injunction. The full trial on all the information hasn't been heard yet, this is just a ruling on a motion about how things will remain during a trial.
Although, my understanding of injunctions was that they were supposed to be quick by their nature and purpose, and not drag on for years. I didn't even know that they could be appealed...
While I've only begun my research, here are a few resources that you -- and others should probably look into.
(It takes me a while to dig and digest info before I reply. I wonder how many people come and go before I can post)
The history of the ACLU challenge to COPA, including the filings, injunctions and hearings.
The Appellee brief (the basis of appeal) in the recently decided case, so you can see the issues that were specifically raised
And of course, no discussion can be complete without COPA itself
I've lined up webspace with CGI capability, etc. and no advertising via a friend who has an under-utilized site at a hosting service. We've written up an agreement to cover the basics and assure the site stays up, even if there are some hassles along the way. I'll only have to relocate if I cause *operational* problems at his site. He likes the project, and corporate pressure (for example) wouldn't faze him.
He asked me not to announce the URL until I had a debugged site to put up (to give him a chance to give security a once-over first) and since this article is going to drop off the radar soon, I got a GeoCities page, where I will set up a FAQ, a feedback section, and announce the regular URL when it goes beta (Has any living website ever left beta?)
So after, say, Saturday (24 June 2000) , you may want to check in at the temporary page to see how things are coming. Bookmark it for future reference.
The single most immediately pressing question is "What do we call it?" It may not be the actual Award name (which will be the first order of business on the site, along with categories), but I can't keep calling it 'Coxsackie' forever (Coxsackie virus causes Hand-Foot-Mouth Disease. It's my personal term for projects I know I'll regret volunteering for. It also makes a moderately satisfying curse)
Ideas, people! I need ideas!
Which is a non-issue since by region price fixing is illegal under the WTO.
/. debates?
This is an excellent point! Why haven't we heard more about this on the many
I'd appreciate any citations you might have, or even the names of documents (The WTO, which whom, I have been recently corresponding, does not index all of its downloadable documents for maximum user-friendliness. You often have to know which documents you want, download, and then search them yourself)
This process often sidetracks me into chasing down the wrong papers. I do hopwever, often stumble across sardonics gems like the following (from the most recent (1999) WTO report on conditions in the US):
"Enforcement of antitrust laws is rigorous, as witnessed by the large number of ongoing investigations and actions taken to combat price-fixing, predatory pricing and exclusionary pacts involving major U.S. and foreign companies. Enforcement of laws protecting intellectual property rights (IPRs) is also rigorous, so as to ensure adequate returns for investment in innovation."
How's that for wry wit? I think it could apply to the MPAA as well as it does to MS-AP (Microsoft Appeals - their real business for the time being)
I believe that the Academy Awards are a fine idea. I don't fault the Nobel Prizes for their choice of subject matter (Economics, for ecample, must've seemed a shining star when the award was created, yet the Nobel's own awards show its unfulfilled promise -- i.e. in Physics or Medicine consecutive laureates always agree on 99.9%+ of their subject; in economics, they may disagree on their most central tenets)
However, I think that this noble concept -- to advance the art by recognizing its finest work -- has been caught in a common quandary. changing too much is disruptive (as in the religious debates of baseball and other sports over decades old rules changes and equipment technology) and difficult. Consistency has value in the judging process and allows competitors to know precisely where they stand. The patina of age and tradition also serves the dignity and value of the awards.
But change is also necessary. Perhaps the conflict is intrinsic. Perhaps, despite the Academy's efforts to recognize advances in the science and technology of film, it is inevitable that an award for 'film' will pass the way of national awards for artistic heiroglyphics or penmanship. there are already major motion pictures in theatrical release that are displayed digitally on electronic screens. It is difficult or impossible to argue that inherently pixellated films like Toy Story (an Oscar winner) are any more film-like than an Australian indy production that was disqualified because its live actors were recorded on videotape instead of emulsion.
The numerous flaws of the Academy's voting system are well known (e.g. it is universally agreed that most members have not seen even a sizable minority of the candidates, and vote based solely on publicity). We at Slashdot have seen similar issues -- to the extent that we scrutinize the process and read 'hidden' discussions like sid=lostkarma, sid=moderation, and sid=metamoderation.
As far as the internet ruling goes: it is merely an explicit elaboration of a rule that has existed from the beginning. The foreign film rule, however, seems to be an accomodation of changing realities that do not violate fundamental tenets of the Oscars (it is a recognition that LA is not the alpha and omega of the film world) I cannot condemn them for their decisions.
We really do need to establish an award for Internet Art, that will stand alongside the Emmys and Webbies of the future. The idea is not original to me. I've read it in this very thread.
However, since this project will not create itself, if there is sufficient interest, I am willing to commit the resources (time, money, access, programming) necessary, including a website to be established by the end of the July 4th holiday weekend. I would appreciate input and assistance in identifying the categories and nominees, criteria and structure, fixing on a name, locating suitable judges, etc. (though public voting can and should play a role, I am not sure the 'standard' web voting site is suitable to be the sole element of voting at present. I have seen too many abuses and flaws in that system) and other areas. I look forward to such feedback, and anticipate turning to Slashdot frequently as this noncorporate venture proceeds.
I wish that they had a picture of the unit, but what I am envisioning from the description is a box with little more than a 'off' switch and a display in front (it doesn't need either, but they are cheap, and would aid marketing at this price point - $250 w/o system purchase)
The real magic would be the 'back panel', standard output jacks to mate with a home stereo, which unfortunately will probably just be phono plugs and clamp-ons for bare speaker wire. It's really too bad, because believe me, if they added stereo 'balanced input' jacks, I might even suggest it some people I know. getting a decent balanced in to a computer, even one with a studio-style music card (not a Soundblaster) is pretty expensive.But that would make the device and I/O device, and I see no suggestion that it's anything but an output device.
I also wouldn't immediately assume it's Ethernet or anything like that. It could be USB.
It's primary advantages (if I envision it correctly) are:
0) Fiddle free operation: i could (and have, in the course of helping people with home studios) create some of all of the advantages below without adding components, but I like hacking. And because it was important to get it 100%, I noticed how many apparent solutions weren't 100% for various reasons
1) Skip protection -- face it, HDDs are *slow* to change tracks. I doubt there's an IDE HDD around that can't be pushed to 20-30ms *worst case* track seeks. Don't bother checking your specs, NO ONE publishes worst case numbers, and even their 'typical' and 'capability' (best case) figures are unstandardized between manufacturers and suspect, technically, as savvy HDD review sites like Storage Review will happily show you, via exhaustive testing and comparisons. So when you're doing a few things at once, skips aren't unheard of, even at modest CPU loads. I rarely notice them, and they don't bother me, but they are there. A meg of buffer at the output device is cheap and easy anti-skip pretection.
2) Better sound reproduction -- Yeah, you could buy a better sound card, but as great as those can sound, a dedicated device can be better, especially when driving a stereo amp/speakers . I'm no sound snob personally, but I help a lot of home studio musicians, and the difference is easy to hear, even for me. Similarly, few PC speakers come close to a goo set of home stereo speakers. Even some $300 'big name' (you'd recognize it instantly) USB speakers I tested recently were very disappointing.
3) Device segregation: There are actually good reasons why a normal user mught be better off keeping a 'game style' (FMsynth/MIDI/WAV) card like a soundblaster as his primary audio device, rather than a killer studio-style card -- I have a friend who has separate semi-pro MIDI and digital audio cards, and a cheap Ensoniq for apps/surfing, so he doesn't have to power up his whole studio for normal computing. Such a set-up tends to create confusion in apps that expect a SB compatible, too. Sure, you could do this with USB speaker drivers, but see above re: those
4) Looks cool on shelf: don't knock it
5) allows full utilization of existing Home Entertainment components (which, in many houses doesn't revolve around the PC, as strange as that sounds) Unfortunately, this is merely a theoretical advantage. This product seems clearly targeted at the market that will buy first (computer users) and will probably not even talk to other home stereo components, aside from passing on the audio analog signals.
However, having said all that, I am utterly unmoved by this product, and I expect most users will be, too. What I'd like is a "digital entertainment station" - bidirectional, so it would be able to pipe me radio, TV, good balanced mike input etc. as well as piping my decompressing and D/A'ing mp3 (and WAV) stream to the speakers.
There is a market out there for this. Lots of companies sell this stuff for Beaucoup bucks. But Dell is going for a mass market portal model for revenue and expansion
Forgive me Lubeck Streyer (aka 'lubie-babie', when I was in molecular bio) and St. Lehninger, for the sins I am about to commit...
.......H............ ....HO-C-H..___. .....__....... ........\../...| ... /..\...... ......HO-C.....| ....\__/\..... ......HO-C-H...| .........\__.. ......HO-C-H...| ........./..\. ......HO-C-H...| .........\__/. .......H-C-OH..| .............. ..........\____| ...Table Sugar
I really feel I ought to explain what they are calling 'sugars' here -- it's a biochemical term that (very crudely) boils down to 'a chain of carbons with water attached' -- only the 'water' has broken into two parts ( HOH => H + OH ) and these two parts connect to the carbon, instead of each other. you can think of a 'sugar' as a chain of carbon groups that look like HO-C-H and are connected to each other at by the carbons like this:
......._____.
....../.....|
...HO-C-H...|
...HO-C-H...|
...HO-C-H...|
...HO-C-H...|
...HO-C-H...|
....H-C-OH..|
......\_____|
The big loop just indicates that the carbons are generally in a ring. The second figure indicates that not all of the carbons are always in the ring. The last HO-C-H group is backwards to indicate that major difference between many sugars of the same size is simply whether each -OH group points up or down when we lay the ring flat. "up-up-down-up-up-up" is one sugar, but "up-up-up-down-up-up" is different sugar (they may seem like reflections, but trust me, in a 3-D ring, they aren't)
The 'well-known sugars' (most of which you've never heard of) have carbon chain lengths from 3 carbons (e.g. triose) up to seven carbons (e.g. sedheptulose). However, the 'familiar sugars' are usually based on a six carbon (glucose, fructose, etc.) or five-carbon (ribose) ring. Table sugar (sucrose) consists of *two* six carbon sugars connected together. Chains of sugars longer than two can be very 'un-sugar-like' -- cellulose (wood fiber) is nothing but long linked chains of glucose (blood sugar) while glycogen (a stored fuel in your liver) is also just branched chains of glucose, but is very different physically.
Glucose (blood sugar) or fructose (fruit sugar) are C6-H12-O6. Table sugar (sucrose) is C12-H22-O12. In space conditions, it might be useful to think of carbon chain lengths as being like stacked blocks -- the kind children play with. Generally stacking two blocks is easy, but six is more than three times as hard (it tends to fall apart easily)
The so-called "sugar" they found in space is two carbons long (glycoaldehyde C2-H4-O2) and is very unlike the six-carbon (okay, 5-7) sugars we usually think of. In biochemistry, it isn't generally called a sugar at all. Three carbons was a sort of bottom limit to be sugar like, because the 'ends' often have an extra hydrogen, and a two carbon 'sugar' would be nothing but 'end' and can't form a ring. It's not very 'sugar-like'. It is an extremely simple molecule, that would be easy to make ("stack") by random, and it looks like this (where the = means a double bond)
......H.H.... -- glycoaldehyde,
...HO-C-C=O.. -- the so-called
......H.H.... -- "space sugar"
You can find more info at these pages:
The structure and function of macromolecules (an outline)
Some sketches of various sugars (let the pictures load before scrolling, or you'll lose your place)
MircoSoap ph33rz me!
[Honestly, each time I tried to submit this from a Windoze machine -four tries- the box crashed! Then I switched to a Linux machine and the ISP went down. Fortunately I have a backup ISP. Chill, Bill, it ain't that incriminating!]
Here's the specific section, 15 US Code 29(b), covering "what happens next". It's short and clear.
TITLE 15 - COMMERCE AND TRADE
CHAPTER 1 - MONOPOLIES AND COMBINATIONS IN RESTRAINT OF TRADE
Sec. 29. Appeals
(b) Direct appeals to Supreme Court
An appeal from a final judgment pursuant to subsection (a) of this section shall lie directly to the Supreme Court, if, upon application of a party filed within fifteen days of the filing of a notice of appeal, the district judge who adjudicated the case enters an order stating that immediate consideration of the appeal by the Supreme Court is of general public importance in the administration of justice. Such order shall be filed within thirty days after the filing of a notice of appeal. When such an order is filed, the appeal and any cross appeal shall be docketed in the time and manner prescribed by the rules of the Supreme Court. The Supreme Court shall thereupon either (1) dispose of the appeal and any cross appeal in the same manner as any other direct appeal authorized by law, or (2) in its discretion, deny the direct appeal and remand the case to the court of appeals, which shall then have jurisdiction to hear and determine the same as if the appeal and any cross appeal therein had been docketed in the court of appeals in the first instance pursuant to subsection (a) of this section.
Well, we simply can't conduct this thread without a link to the actual ruling at the DoJ site. (PDF only)
Here's the actual text, as sloppily transcribed by ol' stumblefingers (me). Beware of possible typos.
[BTW, is 'general public importance sufficient to meet the standards of 15 USC 29(b) -- gotta check!]
Upon consideration of the joint motion of the plaintiffs for certification pursuant to 15 USC 29(b), and the opposition of Microsoft thereto, it is, this 20th day of June, 2000.
ORDERED, that the Joint motion by the plaintiffs for certification pursuant to 15 USC 29(b) is granted and that this court hereby certifies that immediate consideration by the Supreme court of the appeal taken herein is of general public importance in the administration of justice' and it is
FURTHER ORDERED, that the Finalk Judgement of June 7, 2000, is stayed in its entirety until the appeal therefrom is heared and decided, unless the stay is earlier vacated by an appellate court.
(signed)
Since the article specifically Refers to the Aussie situation (mandatory porn filtering by ISPs), here's what the final report of the Australian Government (National Office for the Information Economy) has to say about the weaknesses of this approach in their review of blocking technologies entitled
Access Prevention Techniques for Internet Content Filtering (Google cache) :
The quest to detect pornography is often more concerned with images than text and getting computers to recognise a pornographic image is equally, if not more, difficult than the task of distinguishing between erotica and other literature. There have been efforts to characterise pornographic pictures based on the amount of 'flesh tone' in the images and on the poses struck by the subjects. Computers have difficulty distinguishing between art and pornography, say between a Rubens painting and a Penthouse centrefold, and this approach is not generally regarded as being effective. Video and other streaming media further complicate the filtering task by supplying a constant flow of images to be examined for undesirable content.
Furthermore, they complain:
This approach is affected by the same issue as profile filtering in that an image - or a fair percentage of an image - needs to be loaded before it can be analysed, during which time it may be displayed on a user's screen..
Of course, this second problem only applies to an Aussie-type ISP restriction. Geocities did this years ago (don't know if they still do): scanning their own HDDs (Free user pages), deleting 'questionable graphics' (with or without human review) and waiting for the page authors to complain about any mistakes,
This method described has almost no merit at all.
;-> How else can we stop the jackbooted thugs from finding/blocking our 21st century Federalist Papers?]
The Article had so many technical, philosophical, mathematical and other misconceptions in the article (just a few listed below), that it could pass for a modestly well crafted troll. It had 'something for everyone' (i.e. anyone should be able to poke *some* hole in it, with a moment's thought), making it both an 'obvious troll' and 'good bait'.
At first, I thought the author was sincere, but then I noted that he actually reversed and misrepresented its flaws as *strengths* (e.g. 'the birthday effect' in namespace collisions)
How did this article get on the front page of SlashDot? <sarcasm> Is it supposed to be a sly social analysis, a wry deconstructionist experiment or dry Gallic humor? I wonder why it's under censorship rather than crypto -- could it be footnote #6, below? Must be, else this ubmission would never be the cream of the crop. </sarcasm>
[1] "Free speech" is only meaningful when it can be widely heard. Perfect encryption without public decryption is like locking yourself in a trunk and throwing away the key. If every Joe Sixpack and Dexter Tapedglasses can read your message without prior arrangement, so can Joe Gannon and Janet Reno. if JS and DT can't read it, it ain't 'free speech', its 'private communications'.
[2] The only privacy insight here is the obvious fact that "encrypted files may look like garbage" (regardless of encryption method) However *cleverly* encrypted files, e.g. steganography, may look like something utterly harmless. Which approach is safer/more secure for the originator, the storage site, and the recipient? Especially in the light of laws like England's mandatory key surrender on (proper) demand. Someday, keeping massive Porn databases may be your duty as a patriot!
[3] While independently assigned padnames of 8 bytes may offer 2^64 names, there is a 50% chance of collision after relatively few pads are generated (i.e. millions). The birthday problem the article mentions doesn't suggest high freedom from collisions (as he implies), it means collisons are much likelier than we expect: if there are 24 people in a room, it's *probable* (>50%) that there'll be a birthday collision (shared birthday) even though there are 366 possible days in the dataspace. He cites this as proof that collisions will *not* be a problem
[4] The system loses the ability to decode as more random pads are created/shared and collisions begin to occur. Since pad generation is uncontrolled, this method would become an information hole -- if you used the 'wrong' #6930d3ed740d54de for a given file, you'd get gibberish -- yet all pad #6930d3ed740d54de are equally valid. The system he calls "A whole Mess 'O' Pads" would degrade to "A Whole Mess" (of bits) -- an effective information hole.
[5] At its best, this inept rendition of a one-time pad is a Geek Pig Latin [GPL??], reducing the encription value of a theoretically UNBREAKABLE 128K one-time Pad to a *theoretical* maximum of 2^(64*n) combinations [where n is the number of OTPs XOR'd together. and a minimum that is no more than x^n combinations for brute force cracking [where x=number of published pads, n= number of XORs]
You can best think of it as a poor key generation method, where the true key is not the 128K pad, but the far shorter 'instructions' -- the keynames to XOR together. The example he gave (6 XORs, 8 byte keynames) amounts to the same security as XORing against a 384 bit key, as far as a brute force attack is concerned. This is the same security as XORing against "Netscape Engineers are Weenies! They really are!!" (48 bytes)
[6] Perhaps this article can be most charitably read as an experiment in information darwinism, but not in the Dawkinsian 'meme' sense: the speaker who uses this method is 'too dumb to be listened to' (and silenced by disappearing into the 'Whole Mess'O'Bits) -- akin to the sardonic 'too dumb to live'. (This is supported by his assertion that he is not sure free speech is a good thing)
Of course, it'd be in the best interest of GM to just support the hardware out-of-the-box, but we'd need to pressure them to make sure the drivers are under the GPL.
(Dateline: Indianapolis) As part of the pre-race publicity for the Indy 500, actress Natalie Portman (Phantom Menace, Better Living through Circuitry, Heat, Leon) whose movies often include elaborate race/chase scenes, drove a ceremonial lap around legendary Indy Track in a General Motors Indy car.
The brief trip was not without incident. An on-board video camera wrenched itself loose under commands from remote viewers. GM mechanic Alfor Hominy was puzzled, "I didn't even think it was supposed to swivel. Fixed-view is all we need for Indy."
In a related story, Richard Stallman held a press conference praising GM's use of a 'free as in speech' operating system in their cars, but it was abruptly cut short when one reporter asked about the plethora of closed source drivers in the new operating system.
"er... uh... yes... that's true, but..." he said, before breaking down and sobbing "she's not automatically open under GPL!" He was escorted from the stage by Kernel Developer Linus Torvalds, who later explained. Richard has devoted this life to Free Software. The idea of a driver he can't freely access, manipulate, and modify to his wishes rather upsets him."
To dispel the embarrassment, Transmeta CEO Dave Ditzel declared the cash bar "free as in beer", which much to the disappointment of the assembled press, turned out to mean $2.50 for a Lite
My initial reaction was that somebody was feling the heat and trying to return the HDDs to stop or deprioritize the investigation. The history of HumInt is full of ridiculous stories of inside sources who were almost caught (but weren't) because an investigation moved on or was deprioritized. The Article suggests DoE was thinking on the same lines.
Despite their safe recovery, Energy Secretary Bill Richardson said Friday afternoon he would continue an investigation into the matter, and he promised disciplinary action would result. Investigators are treating the area where the drives were found as a crime scene, and Richardson said "their authenticity" was still being evaluated.
Maybe the theft was discovered sooner than expected (perhaps due to the post-fire audit) or maybe the original plan was to copy and return the HDD, but of course 'return' is less pressing than the original theft (and almost as risky). You have the data, everything else is icing... until you realize your source is almost compromised!
"authenticity" - I doubt anyone was foolish enough to return a copy of the HDD, so this suggests some clean-room (in the original sense) technique for checking if an archival (written then stored) HDD was ever read, or if *all* of it were read (Normal use would only read parts as needed. If all of it had been read - especially recently - it would strongly suggest copying.)
Does anyone out there have any info on the current capabilities of 'clean-room' HDD analysis? We know the Gov't has impressive recovery methods for recovering long erased and multiply re-written files from off-track traces.
While its not a major budget item for a large company, it's certainly something they would consider, especially since the cost of power consumption (not just the power itself) is considerably hiogher than you estimated
Air conditioning has to dispose of the heat once it leaves the machine. It takes more than 1 watt of energy to move 1 watt of waste heat (if you could move 1 watt of heat for less than one watt of input, on a large scale, you could do stuff that verged on 'free energy' -- i.e. use 200W to move 1000W of waste heat into a 50% efficient heat engine to generate 500W, then, if you really want to push it, use the 500W to... well you get the idea) While new HVAC plants are quite efficient, using 1.5W to remove 1W of waste heat is not unreasonable for the national installed base (HVAC updating is very expensive) when you consider all ancillary costs
An office worker generates only 100 cal (400 joules) per hour of waste body heat. A desktop computer (w/ monitor) can generate up to 225 kcal (250W*3600sec = 900K joules). There's a lot more forced hot air streaming out of your PC than convecting gently from your body. Of course humans and PC's are hardly the only heat sources in a building, but you get the idea:
Buying big A/C capacity cost $100K's. A maintanence contract cost $10K's/yr. It cost $10K's to power the PCs and 1.5 times as much to powert the AC. An AC failure can cost $10's-100K's per hour as Important CPUs are overheated by waste heat from idle CPUs. And not one dollar of all that directly generates new revenue. CFOs hate that.
It adds up to a cost that management may want to cut -- especially when a company is expanding, and the choice is "more AC capacity and costs" vs. "making do with existing AC capacity"
Office leasing makes it worse. Why invest (or co-invest) in infrastructure for a building you don't own?
Over the past 6-12 months, we've seen a lot of articles on the 'patent wars' between Japanese companies (and when you consider the RamBus partners, this is pretty much a treaty in these ongoing wars)
To understand the significance (or lack thereof) in RamBus having patents on both sides of the SDRAM/RDRAM divide, you have to understand the outlook of the Japanese patent system (both its examiners and users)
In Japan, companies tend to patent much smaller details and modifications than are considered patentable in, say, the US. If you patent an "iridium-based' lightbulb, you can expect other companies to patent 'iridium based' bulbs with different bases, shapes, colors, multiple filaments, higher output, lower voltage, etc. "Iridium filament bulb with bayonet attachent for use in battery-powered flashlights" or "Iridium filament stage lights" would not be considered patentable in most of Europe or the US, but I've seen exactly this sort of patent on special bulbs in Japan.
Before WWII, this allowed japanese manufacturers to efectively bypass international patents in the domestic Japan market: "Maybe GE has the patent on that bulb, but we have the patent on green ones!" This would hold up in a Japanese court -- not that many companies tried to enforce their rights in Japanese courts back then. it wasn't a major market, and "Made in Japan" meant cheap and disposable.
In time, The Japanese system led to a strangling network of interlocking patents. Any big company could push you to license your patents, else they'd threaten to enforce the myriad tiny patents they or their corporate allies ('Kureitsu' relations is 'schoolyard politics') held to paralyze you or block your use of your own patent.
Result, lots of sharing and intertwined relationships, which is, of course, how the Japanese preferred to do business, anyway.
Ask anyone who tried to open a Japanese market (or business presence) in the 70's/80's. The culture shock was like a 2x4 to the head. It felt like socially condoned fraud, theft, monopolism, -- and a lot of words I can't print here. Comparisons to the Mafia were common (behind closed doors)
This is simply business-as-usual in Japanese business. It's not corruption or even scheming, it's their culture. "The nail that protrudes must be hammered down" is one of their most famous proverbs, and many of their well-accepted (by society in general) applications of it would be megabuck civil rights violations here.
It's like a Soviet and an American arguing economics in the 50's -- the premises are simply so different that you can't make a single logical step without considering them
Well, there's a difference between opening up the protocol and opening up th AOL servers to other clients (but I presume you could set up you own server, if the project is valuable to your organization -- presumably within Lotus SameTime.)
.001% sales rate.
Personally, when I read the title of your post, the first thing I thought was "OmiGawd - IM spam! There are certainly enough addresses in harvestable form out there. Spammers don't seem to care who they annoy for that
Take it a step further - a bot that 'listens' for certain keyphrases, and interjects product recommendations.
DaveThomas: What sort of POS terminals do you think we buy for our 5000 retail outlets?
FrankPerdue: We use distributors, and don't do POS
JSmith@PhoneyOutfit: For price, reliability, and service, you can't beat CrapCo. An industy secret! Best in the business!
The mind boggles at the possibilities. Especially since opening up the protocol increases the possibility that a spammer, using a custom spam program, could fake being a (e.g. an older) AIM client
Disclaimer: I haven't used AIM in a year, and only used it for limited purposes, like instant communication when collaborating on projects. I may not recall its specific capabilities entirely correctly.