Wow! Programmers are working on a new program called KDE which will be released this spring! That's what the article says. I can't wait to try it out. Maybe by next year they'll report on the 2000 USA elections.
Like most others here, I'm very curious about what really happened to Streamcast's Morpheus network. But in practical terms, I settled for trying for the new "Preview Edition". The Musiccity web site last night said that it would be available "in two hours" and indeed after that page was unchanged for more than two hours, the new edition was on Download.com. I had been thinking about rebooting into Linux but this gave me another reason to stay in Windows. That and my kids' wanting to play more Jimmy Neutron this morning....
The new client is really Gnucleus -- if you mouse over the "M" logo in the Tray, that's what it shows. The client is much more primitive than the old FastTrack one. It doesn't include an integral player, so you can't listen to files as they upload, unless maybe you have WinAmp or something running. It gives no clue about who the other end of a file is, so you can't choose one that's more likely to work, and it doesn't report the MP3 bit rate or ID3 info that you can usually see inside the FastTrack client. The failure rate is high -- most attempts to download just quit after ten seconds, though some wait and Retry and a few actually work. FastTrack was much more reliable in that regard.
It also keeps popping up Internet Exploder windows. That's really annoying; I rarely use IE (only for "IE only" sites). It's mostly ads, I'm sure, but the current popups don't even work, causing another annoying distraction.
Being Gnutella based, it probably has scaling problems. I'm on a broadband link, which helps, but I know about the basic math problem with Gnutella's original architecture and I don't know what has been done to fix it, in Gnucleus, Limewire or whatever. Again, FastTrack worked really well, and I hope they can merge its best concepts with Gnutella.
I realize they had to get this out in a hurry. It's only a "Preview" so it shouldn't be viewed as a finished product. But it does weaken the competitive position of Morpheus.
Cisco knows that IPv6 is a lose; they have to support it, but don't have to push it hard.
IPv6 is a bad job, period. Most Slashdotters probably don't know its provenance. It has been around for about a decade. IETF created it as a compromise. IETF insider Steve Deering had created a poor-quality hack called SIP (Steve's IP) while insider Paul Francis (aka Tsuchiya) created one called PIP (Paul's IP). How bad? SIP, for instance, assigned all addresses by countries, based on population, and thus gave a shorter prefix to North Korea than to South Korea because it was a bit more populous in his almanac. IPv6 is PIP and SIP glommed together.
Just before the time it was adopted, IETF had adopted a different replacement for IP, TUBA (which I think was also called IPv8). TUBA used a profile of the OSI Connectionless Network Protocol (CLNP). Cisco had already implemented it, along with CLNP's routing protocol, IS-IS. CLNP was elegant and flexible -- some of the OSI work stank, but CLNP and TP4 were gems. The only reason TUBA was dropped was because Vint Cerf, the Chauncey Gardner of the Internet (not really so smart, but he's famous for Being There), changed his vote and dropped TUBA support.
Had Vint not been so perfidious, IPv8 would have been phased in before the public Internet boom of the mid-1990s. The code has been in Cisco and other vendor equipment for a decade.
IPv6, on the other hand, has a wasteful 16-octet address field (only 8 octets are useful at a time) and does little else to solve IP's problems. It does NOT provide QoS (that's an urban legend) or security any better than IPv4 with its existing options. And given the inefficient assignment of IPv4 adresses in the past, the 32-bit field has a lot of life left.
Think about VoIP: With IPv4, the header has 8 address octets, while the payload has to be short in order to minimize delay. And it's bloody inefficient. With IPv6, the header has 32 address octets while the payload is the same. It's a bleedin' joke! IPv6 is just plain wasteful.
At 8 kbps, the KT voice was better, sure. And at 4 kbps it was roughly intelligible. But the 32 and 64 kbps tests were of questionable materials. The 32k was of electronic rock, and the 64 of synth. How do you take electronic synthesizer music and judge what it's supposed to sound like?
Let me hear the same tests on acoustical instruments (say, flute and piano) or an orchestra. Then maybe it'll count.
I have more scratched CDs than I know what to do with. This is not a problem for a typical adult with a few dozen CDs in jewel cases. It is a serious problem if a) the CDs are played a lot in the car, where the driver can't put them away while driving, and/or b) the kids play games, and don't exactly handle CDs with kid gloves. I've learned all sorts of tricks for getting around scratches, but none work that well. The best, btw, is an LG CD-RW drive that simply reads through them better than anything else I own. I burn replacements using CloneCD or some other program. In a serious case of a scratched music disk, I convert to MP3 and burn back, substituting a "napster'd" copy of the unsalvageable track; I consider that fair use!
Dataplay uses a 3"-floppy-like cartridge. Nice, but a bit small in capacity. BluRay's 5" CD in a cart will be very helpful. Caddys may have been a schlep but on balance I miss them.
Read the araticle. Wal-Mart IS more efficient. They are a distribution company, and they distribute things (from manufacturer to consumer) very efficiently by means of very good IT.
They are by no means alone in using cheap labor -- Gap and Nike do, and usually have much higher prices. I do not condone the labor violations that Wal-Mart and other stores may be committing (nor am I rendering judgement), but the whole point of the Technology Review article was that Wal-Mart automates its supply chain better than its competitors, so inventory is more carefully ordered and moved around.
>They make software to sell hardware. thats the whole *goal* of the company.
Which is why Apple will always be a marginal player, with marginal finances, selling to a fan club.
IBM used to be the same way, but in the mid-1960s they saw the light. It didn't hurt them one bit. Their hardware and software businesses each had to look out for themselves, without being codependent. Novell used to be a hardware company, but saw the light, and only then became a major player. VA Linux Systems... well, never mind.
But the point is that software is potentially more profitable than hardware, while hardware is a commodity. Even if it's (in Apple's case) a commodity in an odd-shaped box (commoddity? -- this was originally a typo but I like it). Jobs is so hung up on his boxes, and his 1950s view of the computer industry, that he's isolating himself.
The point I'm making is that the implementation is not primarily based on academic theory. It's based on the Internet's own culture, which puts experimentation in front of architecture, implementation in front of design. It's based on empiricism, on doing what works, and for the most part doesn't give a goat's bzadeh about Kleinrock's theory. Even when it should pay more heed!
Yes, his work on queueing theory is important, at least to people concerned with math and network analytics. And if anybody gives a damn about analyzing the performance of a packet-switched network mathematically, then it'll fall back onto Kleinrock's work.
But what passes for packet switching nowadays -- "The Internet" -- is most certainly not the result of careful analysis! It works by brute force. It's inefficient. It is badly monitored and mostly unmetered. So Kleinrock's analysis, which might be useful, is ignored.
Anybody with half a sense of the math wouldn't dare try to cram constant-rate streaming traffic, like telephony or broadcasting, onto the IP Internet. It's inefficient as hell. Economy of scale is what makes it seem to work, compared to economy of specialization (what ATM would excel at). But that's the Internet's current ruling ethos -- if it seems to work, do it, even to excess.
The original inventors -- Davies, Baran, and the BBN crew -- were not doing mathematical optimization. They were hacking (in the good sense) something together and observing what worked. Kleinrock is like a guy who invents a great network management system that never gets turned on, but who still claims credit for the network that his system might have been able to manage.
And puh-LEEZ, don't give Vint any of the credit. He's made a great living as the Chauncy Gardiner of the Internet.
In Congress, "initiative" refers to a funding measure. Gore led the charge for the funding for the NSFnet. That was indeed an "initiative". And the NSFnet was *the* major Internet backbone after the ARPAnet and before commercialization.
Finding a different meaning of a word which doesn't apply is simply obfuscation.
Mr. Z is correct -- moderators, promote the note I'm replying to! Cringely is wrong about layers. Layer 1 does all the bit stuff, including modulation and even ATM cells. (Layer 2 is about user-sized frames and error detection.)
But that's not what matters. Shannon matters. You can't defeat Shannon, and Cringely admits it. So let's see... Shannon basically says that the limit of bps is proportionate to the product of bandwidth times the log2 of the signal to noise ratio. So if you have an infinite SNR, you can have infinite bandwidth. But getting 33 Mbps (around the top end of DOCSIS cable modems) requires good SNR. My cable modem right now has 36 dB SNR and is running QAM64; DOCSIS adapts speed to line quality.
So even if wavelets were better than QAM (and I can't say, because Cringely doesn't tell enough to know if this is real or a scam), there's just not that much more you can do in 36 dB! (Shannon limit of 6 MHz at 36 dB is around 6M*12=72 Mbps.)
What VeriZontal Wireless is introducing is the so-called "1XRTT" form of CDMA2000, which is one of the flavors of "third generation" (3G) cellular telephony. While there has been a lot of noise about 3G around the world, and European carriers have shelled out tens of billions of licenses (dotcom-style investment) for new 3G spectrum (putting them deeply into debt), VeriZontal Wireless and Sprint PCS are instead taking the "just do it" approach.
There are two distinct technical flavors (air interfaces) to 3G, both based on CDMA. The GSM (most of world) and IS-136-TDMA (Cingular, ATT-W) carriers, with existing TDMA networks, are migrating to WCDMA. The CDMA carriers (Sprint, VZW, Korea) are migrating to CDMA2000. (Qualcomm favors CDMA2000, but makes patent royalties off of both. They really did invent it.) The CDMA2000 spec in turn has multiple variants. The "1XRTT" flavor is simply a software change to the way existing CDMAone carriers are allocated among calls. The peak speed is only 144 kbps (ten times what CDMA one gives you) but there's no forklift upgrade, and no new spectrum needed. Of course it needs new handsets to make use of the new features, but the base stations are backwards compatible. Very graceful, 3G on the cheap.
So VZW and Sprint are both rolling out 1XRTT this year. VZW announced faster, but they're both gated, in practice, by the availability of handsets and similar remote devices from the (mostly Korean) makers. The CDMA and GSM carriers are instead phasing in a "2 1/2G" technology, EDGE, as a sort of bridge to WCDMA. They'll need separate networks, or a forklift upgrade, to do 3G. Since WCDMA doesn't share spectrum with TDMA, they can't do the easy phase-in that CDMA gives you.
But don't think of 3G as a substitute for fast wireline. A 144 kbps call basically eats ten voice calls' worth of network bandwidth. So it will be expensive! Packetized data, by the byte, will be cheaper, but really aimed more at low-bandwidth things like email than high-bandwidth things like music or ordinary web browsing. (Look up EDGE pricing on the GSM networks to get an idea; it's in dollars/MB). This is a premium service for users who need it.
>The process of licensing conveys ownership and its rights as specified in the license for the duration of the license.
Wrong! Licensing conveys RIGHT TO USE, not "ownership". BIG difference. Only some hardcore extreme right wing groups, and perhaps some doctrinaire Libertarians, think otherwise.
Said right to use is not necessarily exclusive. Every gasoline car creates sparks that interfere with radios, but usually below certain limits. Do we ban cars? Do we ban computers, which spew lots of RF? Part 15 defines what we can do without a license. UWB simply modifies that rule to allow intentional radiation at levels below those allowed for unintentional.
Think of it as an easement, if you insist n property analogies. But it's only an analogy, because it's not property.
I hardly think the "property" argument jibes with the "public interest, convenience and necessity" traditional mandate of FCC regulation. Older "narrowband" (and CDMA) licensed services are sensitive to interference from each other. UWB promises not to interfere, so what's the beef? That they didn't pass a Morris Code exam? That allowing phone patches (old amateur radio tradition predating even Carterphone) will take away "the phone company's" guaranteed revenue? That my radio waves can't "overfly" your land without paying your for the privilege?
The debate over UWB centers on the difference between intentional and unintentional transmission. UWB advocates want to be allowed to intentionally transmit at levels *below* those autorized in Part 15 for unintentional radiators. Sounds fair, except of course that the sum of lots of them might seriously raise the noise floor in some portions of the spectrum. That's a valid technical debate, but not a property debate, unless it degrades performance of licensed services.
Cringely, of course, did make major mistakes in his article. UWB doesn't use "ALL" frequencies (the proverbial "DC to daylight"), just a lot more than "traditional" spread spectrum. And its power/range tradeoff is about the same as other spread spectrum. And PCS goes a LOT farther than 1 km, outside of the densest urban enviroments, if its towers are high enough.
(BTW, I have an Extra Class ham ticket, and know the Morse quite well.)
There's no risk at all to broadcast sats. The TV stations use geostationary (GEO) satellites, in orbit 23k miles above the equator. They all orbit the earth synchronously, so there are no collisions in that belt.
This and the other amateur (OSCAR) sats are in low earth orbit (LEO), 200-300 miles up. They fly around for a few years and eventually fall back and burn up. They don't come within >22,000 miles of your precious TV broadcast relay.
First off, I'm skeptical about "losing" the mediaone.net domain; it has been in use since, oh, 1997 or so, so a cybersquatting claim is dubious at best. But if ATT-B decided they wanted an *excuse* to can it, then they might have decided to "lose" it. Anybody know more?
MediaOne was the name US West made up for the stuff it bought, mostly from Continental Cable. And Continental had already set up cable modem service as "Continental Express". But MediaOne really grew it.
Comcast has now agreed to buy ATT-B. Theoretically, the new corporation will be "AT&T Comcast", but there will be a separate AT&T Corp. (the original one) doing telephone and its own ISP stuff. So will "AT&T Comcast" keep the AT&T or ATTBI name forever, or will they decide next year to change it to "@comcast..." or @attcomcast..."?
Me, I'm glad that my real mail is still on a private ISP that lets me POP from anywhere. (I use a MediaOne cable modem.) My wife's email is on Yahoo, which allows both POP and webmail access. Yahoo just announced that their SMTP server will demand authentication (I can understand that, though I don't think Eudora Light can do it) but they still seem okay for POP, and the price is right. My gradeschooler is not happy that his email will be changing, especially if he loses his "name" portion. They'll honor your current @mediaone.net if there's no dupe already on attbi, but a lot of former @home users are already using that space.
While doing a study of energy consumption, it became obvious that notebooks are far more energy-efficient than ATX desktops, and are widely used in mostly-fixed applications. (My office PC is a Stinkpad, with an external kbd and mouse.) They are however virtually unexpandable. A form factor that I would like to see takes the best of both worlds:
The case would be smaller than a micro-ATX, a bit bigger than the SV24 (which seems rather hard to work on). The motherboard would support a couple of slots, perhaps riser-style. Sound and "good for most" video would be on the mobo. (This isn't a gamers' system but an office/home/general one.) The CPU would be a low-power one, like Crusoe, C3 or Intel's low-power line but socketed. There'd also be a little battery. Not to run it for 3 hours on a plane, but to do the "UPS" role. The power supply might be external, but a small one might fit inside. It would have a small fan, and room for a regular 3" HD and 5" CD/DVD. Even a floppy might fit.
Typical power consumption would be >30W. (I am an Athlon user, but this design would need a.13-Duron or the like, clocked down, to do the trick. This isn't AMD's strong space.) It would be cool, quiet, light, and not take up much space. A handle on top might be nice, as would a place for an antitheft locking cord (like notebooks have).
BTW, "lunchboxcomputer.com", noted above, builds big costly heavy industrial lunchboxes. But they did claim the name first.
That would be true IF that's what MIPS really meant, and I'm not referring to the already posted "meaningless indication" joke. But it's not.
MIPS was a measurement created (I suspect by CMP) back in the 1960s. It was the amount of processing power that a CPU had in terms of IBM 360/50 machine instructions. (Millions of IBM Instructions per Second.) Not cycles of the machine being measured, but normalized against a 360/50's work/clock being "1".
The 360/50 was a classic CISC machine, with the kind of complex addressing modes that only a BAL programmer could love. RISC demonstrated that simple instructions generated by a compiler could often outperform microcode. But that came later: As IBM developed the 360 and 370 lines, work per clock cycle varied. MIPS was normalized.
At DEC, we faced demands for comparison between the VAX and 360 families. (Apples to squash, really, but you know how people want simple comparisons.) In raw CPU capability, an early CISC VAX-11 was not far from a 360/50 in work/cycle. But the measurement we used was the VUPS (VAX unit of processor speed). Again, it was a performance measurement, not a clock timer.
They have funny wording in their release about data that is practically random. Well, that can be parsed to mean that in practice, the data is random and therefore it can be replaced by any other random string. After all, it's random! Not mathematically random in the entopy sense, but used by an application which wants any old string of random numbers.
So sure, I can send a message saying, "generate me 1000 random digits". Great compression. Useless in practice, of course.
In any case, these guys sound like a get-rich-quick scheme, trying to fool people, and not the only one of that type I can think of.
Read it. The law says that spammers have to provide an "opt-out" address, and users have to send their real email addresses to them, the spammers.
Duh.
So now the spammers will have a list of valid, guaranteed active email accounts. To sell, which is what opt-out addresses in spam are for. Not to opt out, but to verify that they're real.
And since this is a state law, the spammer can get away with this by being out of state. Not that spammers ever care about the law. The law merely encourages users to ACT LIKE IDIOTS and send real email addresses to spammers who will then use them as verified, premium spambait!
Doesn't Bertlesmann, the German Big 5 record conglomerate, own Napster now? It's not Sean's anymore.
This leads me to think that Napster is, like PressPlay and MusicNet, designed to fail. It is designed to allow subscribers to listen to music on their computers, not elsewhere. So it's value to the user is grossly inferior to, say, FastTrack/Kazaa/Morpheus or the old Napster.
The major labels are willing to put money into failures, because they want to prove a point, that online sales are not going to work. They want to divide the world into two camps, those who purchase costly mechanical artifacts with stored music (CDs, etc.) licensed to their possessors (good), and those who copy off the net (bad).
A real online music distribution system could be concocted, with the artist getting directly paid based on upload volume, and the consumer paying, say, a monthly subscription fee that is allocated among artists proportionately (not fee per play). But that would undercut the major labels, who won't let it happen. And the DMCA + WIPO let them keep it from happening.
Digital cable doesn't use the same bandwidth for everything.
DVD is a pretty good MPEG-2 stream, what, 6 Mbps? Digital cable assigns that much bandwidth to some channels but less to others. Sports, for instance, need more bandwidth than talking heads. CSPAN probably compresses very low. Movies? It can vary by channel.
Slashdot Mirror
Wow! Programmers are working on a new program called KDE which will be released this spring! That's what the article says. I can't wait to try it out.
Maybe by next year they'll report on the 2000 USA elections.
Like most others here, I'm very curious about what really happened to Streamcast's Morpheus network. But in practical terms, I settled for trying for the new "Preview Edition". The Musiccity web site last night said that it would be available "in two hours" and indeed after that page was unchanged for more than two hours, the new edition was on Download.com. I had been thinking about rebooting into Linux but this gave me another reason to stay in Windows. That and my kids' wanting to play more Jimmy Neutron this morning.... The new client is really Gnucleus -- if you mouse over the "M" logo in the Tray, that's what it shows. The client is much more primitive than the old FastTrack one. It doesn't include an integral player, so you can't listen to files as they upload, unless maybe you have WinAmp or something running. It gives no clue about who the other end of a file is, so you can't choose one that's more likely to work, and it doesn't report the MP3 bit rate or ID3 info that you can usually see inside the FastTrack client. The failure rate is high -- most attempts to download just quit after ten seconds, though some wait and Retry and a few actually work. FastTrack was much more reliable in that regard. It also keeps popping up Internet Exploder windows. That's really annoying; I rarely use IE (only for "IE only" sites). It's mostly ads, I'm sure, but the current popups don't even work, causing another annoying distraction. Being Gnutella based, it probably has scaling problems. I'm on a broadband link, which helps, but I know about the basic math problem with Gnutella's original architecture and I don't know what has been done to fix it, in Gnucleus, Limewire or whatever. Again, FastTrack worked really well, and I hope they can merge its best concepts with Gnutella. I realize they had to get this out in a hurry. It's only a "Preview" so it shouldn't be viewed as a finished product. But it does weaken the competitive position of Morpheus.
Cisco knows that IPv6 is a lose; they have to support it, but don't have to push it hard.
IPv6 is a bad job, period. Most Slashdotters probably don't know its provenance. It has been around for about a decade. IETF created it as a compromise. IETF insider Steve Deering had created a poor-quality hack called SIP (Steve's IP) while insider Paul Francis (aka Tsuchiya) created one called PIP (Paul's IP). How bad? SIP, for instance, assigned all addresses by countries, based on population, and thus gave a shorter prefix to North Korea than to South Korea because it was a bit more populous in his almanac. IPv6 is PIP and SIP glommed together.
Just before the time it was adopted, IETF had adopted a different replacement for IP, TUBA (which I think was also called IPv8). TUBA used a profile of the OSI Connectionless Network Protocol (CLNP). Cisco had already implemented it, along with CLNP's routing protocol, IS-IS. CLNP was elegant and flexible -- some of the OSI work stank, but CLNP and TP4 were gems. The only reason TUBA was dropped was because Vint Cerf, the Chauncey Gardner of the Internet (not really so smart, but he's famous for Being There), changed his vote and dropped TUBA support.
Had Vint not been so perfidious, IPv8 would have been phased in before the public Internet boom of the mid-1990s. The code has been in Cisco and other vendor equipment for a decade.
IPv6, on the other hand, has a wasteful 16-octet address field (only 8 octets are useful at a time) and does little else to solve IP's problems. It does NOT provide QoS (that's an urban legend) or security any better than IPv4 with its existing options. And given the inefficient assignment of IPv4 adresses in the past, the 32-bit field has a lot of life left.
Think about VoIP: With IPv4, the header has 8 address octets, while the payload has to be short in order to minimize delay. And it's bloody inefficient. With IPv6, the header has 32 address octets while the payload is the same. It's a bleedin' joke! IPv6 is just plain wasteful.
The comparisons were questionable.
At 8 kbps, the KT voice was better, sure. And at 4 kbps it was roughly intelligible. But the 32 and 64 kbps tests were of questionable materials. The 32k was of electronic rock, and the 64 of synth. How do you take electronic synthesizer music and judge what it's supposed to sound like?
Let me hear the same tests on acoustical instruments (say, flute and piano) or an orchestra. Then maybe it'll count.
The protective cover is the BEST thing about it!
I have more scratched CDs than I know what to do with. This is not a problem for a typical adult with a few dozen CDs in jewel cases. It is a serious problem if a) the CDs are played a lot in the car, where the driver can't put them away while driving, and/or b) the kids play games, and don't exactly handle CDs with kid gloves. I've learned all sorts of tricks for getting around scratches, but none work that well. The best, btw, is an LG CD-RW drive that simply reads through them better than anything else I own. I burn replacements using CloneCD or some other program. In a serious case of a scratched music disk, I convert to MP3 and burn back, substituting a "napster'd" copy of the unsalvageable track; I consider that fair use!
Dataplay uses a 3"-floppy-like cartridge. Nice, but a bit small in capacity. BluRay's 5" CD in a cart will be very helpful. Caddys may have been a schlep but on balance I miss them.
Read the araticle. Wal-Mart IS more efficient. They are a distribution company, and they distribute things (from manufacturer to consumer) very efficiently by means of very good IT.
They are by no means alone in using cheap labor -- Gap and Nike do, and usually have much higher prices. I do not condone the labor violations that Wal-Mart and other stores may be committing (nor am I rendering judgement), but the whole point of the Technology Review article was that Wal-Mart automates its supply chain better than its competitors, so inventory is more carefully ordered and moved around.
>They make software to sell hardware. thats the whole *goal* of the company.
Which is why Apple will always be a marginal player, with marginal finances, selling to a fan club.
IBM used to be the same way, but in the mid-1960s they saw the light. It didn't hurt them one bit. Their hardware and software businesses each had to look out for themselves, without being codependent. Novell used to be a hardware company, but saw the light, and only then became a major player. VA Linux Systems... well, never mind.
But the point is that software is potentially more profitable than hardware, while hardware is a commodity. Even if it's (in Apple's case) a commodity in an odd-shaped box (commoddity? -- this was originally a typo but I like it). Jobs is so hung up on his boxes, and his 1950s view of the computer industry, that he's isolating himself.
The point I'm making is that the implementation is not primarily based on academic theory. It's based on the Internet's own culture, which puts experimentation in front of architecture, implementation in front of design. It's based on empiricism, on doing what works, and for the most part doesn't give a goat's bzadeh about Kleinrock's theory. Even when it should pay more heed!
Kleinrock is claiming too much credit.
Yes, his work on queueing theory is important, at least to people concerned with math and network analytics. And if anybody gives a damn about analyzing the performance of a packet-switched network mathematically, then it'll fall back onto Kleinrock's work.
But what passes for packet switching nowadays -- "The Internet" -- is most certainly not the result of careful analysis! It works by brute force. It's inefficient. It is badly monitored and mostly unmetered. So Kleinrock's analysis, which might be useful, is ignored.
Anybody with half a sense of the math wouldn't dare try to cram constant-rate streaming traffic, like telephony or broadcasting, onto the IP Internet. It's inefficient as hell. Economy of scale is what makes it seem to work, compared to economy of specialization (what ATM would excel at). But that's the Internet's current ruling ethos -- if it seems to work, do it, even to excess.
The original inventors -- Davies, Baran, and the BBN crew -- were not doing mathematical optimization. They were hacking (in the good sense) something together and observing what worked. Kleinrock is like a guy who invents a great network management system that never gets turned on, but who still claims credit for the network that his system might have been able to manage.
And puh-LEEZ, don't give Vint any of the credit. He's made a great living as the Chauncy Gardiner of the Internet.
In Congress, "initiative" refers to a funding measure. Gore led the charge for the funding for the NSFnet. That was indeed an "initiative". And the NSFnet was *the* major Internet backbone after the ARPAnet and before commercialization.
Finding a different meaning of a word which doesn't apply is simply obfuscation.
That's impossible. Creationists aren't scientists; the two categories are mutually exclusive.
Mr. Z is correct -- moderators, promote the note I'm replying to! Cringely is wrong about layers. Layer 1 does all the bit stuff, including modulation and even ATM cells. (Layer 2 is about user-sized frames and error detection.)
But that's not what matters. Shannon matters. You can't defeat Shannon, and Cringely admits it. So let's see... Shannon basically says that the limit of bps is proportionate to the product of bandwidth times the log2 of the signal to noise ratio. So if you have an infinite SNR, you can have infinite bandwidth. But getting 33 Mbps (around the top end of DOCSIS cable modems) requires good SNR. My cable modem right now has 36 dB SNR and is running QAM64; DOCSIS adapts speed to line quality.
So even if wavelets were better than QAM (and I can't say, because Cringely doesn't tell enough to know if this is real or a scam), there's just not that much more you can do in 36 dB! (Shannon limit of 6 MHz at 36 dB is around 6M*12=72 Mbps.)
http://slashdot.org/article.pl?sid=02/01/27/144241 &mode=thread
Geez, the thread isn't even cold yet!
What VeriZontal Wireless is introducing is the so-called "1XRTT" form of CDMA2000, which is one of the flavors of "third generation" (3G) cellular telephony. While there has been a lot of noise about 3G around the world, and European carriers have shelled out tens of billions of licenses (dotcom-style investment) for new 3G spectrum (putting them deeply into debt), VeriZontal Wireless and Sprint PCS are instead taking the "just do it" approach.
There are two distinct technical flavors (air interfaces) to 3G, both based on CDMA. The GSM (most of world) and IS-136-TDMA (Cingular, ATT-W) carriers, with existing TDMA networks, are migrating to WCDMA. The CDMA carriers (Sprint, VZW, Korea) are migrating to CDMA2000. (Qualcomm favors CDMA2000, but makes patent royalties off of both. They really did invent it.) The CDMA2000 spec in turn has multiple variants. The "1XRTT" flavor is simply a software change to the way existing CDMAone carriers are allocated among calls. The peak speed is only 144 kbps (ten times what CDMA one gives you) but there's no forklift upgrade, and no new spectrum needed. Of course it needs new handsets to make use of the new features, but the base stations are backwards compatible. Very graceful, 3G on the cheap.
So VZW and Sprint are both rolling out 1XRTT this year. VZW announced faster, but they're both gated, in practice, by the availability of handsets and similar remote devices from the (mostly Korean) makers. The CDMA and GSM carriers are instead phasing in a "2 1/2G" technology, EDGE, as a sort of bridge to WCDMA. They'll need separate networks, or a forklift upgrade, to do 3G. Since WCDMA doesn't share spectrum with TDMA, they can't do the easy phase-in that CDMA gives you.
But don't think of 3G as a substitute for fast wireline. A 144 kbps call basically eats ten voice calls' worth of network bandwidth. So it will be expensive! Packetized data, by the byte, will be cheaper, but really aimed more at low-bandwidth things like email than high-bandwidth things like music or ordinary web browsing. (Look up EDGE pricing on the GSM networks to get an idea; it's in dollars/MB). This is a premium service for users who need it.
>The process of licensing conveys ownership and its rights as specified in the license for the duration of the license.
Wrong! Licensing conveys RIGHT TO USE, not "ownership". BIG difference. Only some hardcore extreme right wing groups, and perhaps some doctrinaire Libertarians, think otherwise.
Said right to use is not necessarily exclusive. Every gasoline car creates sparks that interfere with radios, but usually below certain limits. Do we ban cars? Do we ban computers, which spew lots of RF? Part 15 defines what we can do without a license. UWB simply modifies that rule to allow intentional radiation at levels below those allowed for unintentional.
Think of it as an easement, if you insist n property analogies. But it's only an analogy, because it's not property.
I hardly think the "property" argument jibes with the "public interest, convenience and necessity" traditional mandate of FCC regulation. Older "narrowband" (and CDMA) licensed services are sensitive to interference from each other. UWB promises not to interfere, so what's the beef? That they didn't pass a Morris Code exam? That allowing phone patches (old amateur radio tradition predating even Carterphone) will take away "the phone company's" guaranteed revenue? That my radio waves can't "overfly" your land without paying your for the privilege?
The debate over UWB centers on the difference between intentional and unintentional transmission. UWB advocates want to be allowed to intentionally transmit at levels *below* those autorized in Part 15 for unintentional radiators. Sounds fair, except of course that the sum of lots of them might seriously raise the noise floor in some portions of the spectrum. That's a valid technical debate, but not a property debate, unless it degrades performance of licensed services.
Cringely, of course, did make major mistakes in his article. UWB doesn't use "ALL" frequencies (the proverbial "DC to daylight"), just a lot more than "traditional" spread spectrum. And its power/range tradeoff is about the same as other spread spectrum. And PCS goes a LOT farther than 1 km, outside of the densest urban enviroments, if its towers are high enough.
(BTW, I have an Extra Class ham ticket, and know the Morse quite well.)
There's no risk at all to broadcast sats. The TV stations use geostationary (GEO) satellites, in orbit 23k miles above the equator. They all orbit the earth synchronously, so there are no collisions in that belt.
This and the other amateur (OSCAR) sats are in low earth orbit (LEO), 200-300 miles up. They fly around for a few years and eventually fall back and burn up. They don't come within >22,000 miles of your precious TV broadcast relay.
First off, I'm skeptical about "losing" the mediaone.net domain; it has been in use since, oh, 1997 or so, so a cybersquatting claim is dubious at best. But if ATT-B decided they wanted an *excuse* to can it, then they might have decided to "lose" it. Anybody know more?
MediaOne was the name US West made up for the stuff it bought, mostly from Continental Cable. And Continental had already set up cable modem service as "Continental Express". But MediaOne really grew it.
Comcast has now agreed to buy ATT-B. Theoretically, the new corporation will be "AT&T Comcast", but there will be a separate AT&T Corp. (the original one) doing telephone and its own ISP stuff. So will "AT&T Comcast" keep the AT&T or ATTBI name forever, or will they decide next year to change it to "@comcast..." or @attcomcast..."?
Me, I'm glad that my real mail is still on a private ISP that lets me POP from anywhere. (I use a MediaOne cable modem.) My wife's email is on Yahoo, which allows both POP and webmail access. Yahoo just announced that their SMTP server will demand authentication (I can understand that, though I don't think Eudora Light can do it) but they still seem okay for POP, and the price is right. My gradeschooler is not happy that his email will be changing, especially if he loses his "name" portion. They'll honor your current @mediaone.net if there's no dupe already on attbi, but a lot of former @home users are already using that space.
While doing a study of energy consumption, it became obvious that notebooks are far more energy-efficient than ATX desktops, and are widely used in mostly-fixed applications. (My office PC is a Stinkpad, with an external kbd and mouse.) They are however virtually unexpandable. A form factor that I would like to see takes the best of both worlds:
.13-Duron or the like, clocked down, to do the trick. This isn't AMD's strong space.) It would be cool, quiet, light, and not take up much space. A handle on top might be nice, as would a place for an antitheft locking cord (like notebooks have).
The case would be smaller than a micro-ATX, a bit bigger than the SV24 (which seems rather hard to work on). The motherboard would support a couple of slots, perhaps riser-style. Sound and "good for most" video would be on the mobo. (This isn't a gamers' system but an office/home/general one.) The CPU would be a low-power one, like Crusoe, C3 or Intel's low-power line but socketed. There'd also be a little battery. Not to run it for 3 hours on a plane, but to do the "UPS" role. The power supply might be external, but a small one might fit inside. It would have a small fan, and room for a regular 3" HD and 5" CD/DVD. Even a floppy might fit.
Typical power consumption would be >30W. (I am an Athlon user, but this design would need a
BTW, "lunchboxcomputer.com", noted above, builds big costly heavy industrial lunchboxes. But they did claim the name first.
MIPS was a measurement created (I suspect by CMP) back in the 1960s. It was the amount of processing power that a CPU had in terms of IBM 360/50 machine instructions. (Millions of IBM Instructions per Second.) Not cycles of the machine being measured, but normalized against a 360/50's work/clock being "1".
The 360/50 was a classic CISC machine, with the kind of complex addressing modes that only a BAL programmer could love. RISC demonstrated that simple instructions generated by a compiler could often outperform microcode. But that came later: As IBM developed the 360 and 370 lines, work per clock cycle varied. MIPS was normalized.
At DEC, we faced demands for comparison between the VAX and 360 families. (Apples to squash, really, but you know how people want simple comparisons.) In raw CPU capability, an early CISC VAX-11 was not far from a 360/50 in work/cycle. But the measurement we used was the VUPS (VAX unit of processor speed). Again, it was a performance measurement, not a clock timer.
They have funny wording in their release about data that is practically random. Well, that can be parsed to mean that in practice, the data is random and therefore it can be replaced by any other random string. After all, it's random! Not mathematically random in the entopy sense, but used by an application which wants any old string of random numbers. So sure, I can send a message saying, "generate me 1000 random digits". Great compression. Useless in practice, of course. In any case, these guys sound like a get-rich-quick scheme, trying to fool people, and not the only one of that type I can think of.
Read it. The law says that spammers have to provide an "opt-out" address, and users have to send their real email addresses to them, the spammers.
Duh.
So now the spammers will have a list of valid, guaranteed active email accounts. To sell, which is what opt-out addresses in spam are for. Not to opt out, but to verify that they're real.
And since this is a state law, the spammer can get away with this by being out of state. Not that spammers ever care about the law. The law merely encourages users to ACT LIKE IDIOTS and send real email addresses to spammers who will then use them as verified, premium spambait!
Doesn't Bertlesmann, the German Big 5 record conglomerate, own Napster now? It's not Sean's anymore.
This leads me to think that Napster is, like PressPlay and MusicNet, designed to fail. It is designed to allow subscribers to listen to music on their computers, not elsewhere. So it's value to the user is grossly inferior to, say, FastTrack/Kazaa/Morpheus or the old Napster.
The major labels are willing to put money into failures, because they want to prove a point, that online sales are not going to work. They want to divide the world into two camps, those who purchase costly mechanical artifacts with stored music (CDs, etc.) licensed to their possessors (good), and those who copy off the net (bad).
A real online music distribution system could be concocted, with the artist getting directly paid based on upload volume, and the consumer paying, say, a monthly subscription fee that is allocated among artists proportionately (not fee per play). But that would undercut the major labels, who won't let it happen. And the DMCA + WIPO let them keep it from happening.
I don't care how secure they think it is. Give Danny Ocean three weeks and he'll get anything he wants from there.
(Or George Clooney, in a pinch. Yeah, I liked the movie. Cash vault, sure.)
Digital cable doesn't use the same bandwidth for everything.
DVD is a pretty good MPEG-2 stream, what, 6 Mbps? Digital cable assigns that much bandwidth to some channels but less to others. Sports, for instance, need more bandwidth than talking heads. CSPAN probably compresses very low. Movies? It can vary by channel.