A few years ago, Uncle Charlie (the FCC) came up with a wacky scheme. They allowed local telephone companies (LECs) to impose a "Primary Interexchange Carrier Charge" (PICC, as in "pixie") on every line, with the bill sent to the presubscribed LD carrier. It was I think $0.53 on the primary residential line but higher on second lines and business lines. This is what got LD companies to start imposing minimums or monthly fees -- they passed it right along, only since they didn't get hteir PICC bills itemized by line, they couldn't pass them directly, so they were averaged.
If you didn't have a PIC (e.g., no LD carrier), then you the subscriber got the PICC bill. No escape.
This stupid scheme was finally dropped, I think, in July of this year. The End User Common Line charge (monthly "access charge") will go up instead; it has been capped at $3.50 on primary resi lines and $6 on others. THis is NOT a charge for LD service, but a charge for that portion of local service which is technically assigned to the FCC's jurisdiction. Don't ask. (Smith v. Illinois Bell, US Sup. Ct. 1927.)
That's the state regulator's (DTE) letter to consumers about slamming. There's a bunch more on their web site, and a state law against it, as well as federal rules against it.
State law requires Third Party Verification by an authoritized "TPV" vendor, *or* tape recording of your phone call, *or* written authorization. It is not applicable only to ILECs (VeriZontal) but to CLECs like MediaOne/AT&T too. There are stiff penalties. Of course AT&T knows this, as they were on the losing end of zillions of MCI slams a decade or so ago; that's what led to the stricter rules. It's probably a third-party telemarketing company they retained who is desperate to make quota, but it's still AT&T's responsibility.
BTW I use AT&T/MediaOne's phone service in the Boston area too, and it's quite good. I also use AT&T long distance, with online billing (also a pretty good rate, 9c days 5c nights, no fee; I don't think they advertise it but it's available for the asking; international however is a different story). This hasn't stopped AT&T (the long distance side) from calling me asking me to switch *to* AT&T, from sending me checks whose indorsement consitutes authorization to switch me to AT&T, etc. So their internal records are (to use a famous Boston term for a young codfish) scrod.
AT&T's a big company. The consumer LD side is incredibly profitable but shrinking as LD rates decline. So Wall Street is unhappy, and AT&T's considering spinning the whole thing off. AT&T Broadband is still basically two companies, MediaOne and ex-TCI. They're not well coordinated. AT&T Broadband is talking about offering their own LD service, but for now doesn't seem to have any LD billing.
The FCC is opening a dangerous door if they think they have authority over a protocol or applications-layer operation, such as AIM. They regulate raw pipes, not applications or for that matter IP (which is technically "information service", not "telecommunications").
If AOL's behavior is in some way "anticompetitive", there are other agencies (DoJ, for instance) who do have general say over that matter. And mergers do invite scrutiny. So if there's some specific DoJ question to be answered around their IM networks, then fine. The FCC does have some authority over cable systems, which AOL is trying to buy. But the FCC should NOT consider IM to be theirs to regulate.
Capital cost is not the only expense. Certainly a lot of networks (including the 19th century railways) were only profitable after being picked up at a discount in bankruptcy, but Iridium's operating expense is astronomical too.
Motorola set it up so that they wouldn't lose. They got paid for the satellites and paid to run them. Somebody else could probably run them much cheaper but Motorola doesn't want that to happen; that could make them look bad in the bankruptcy court.
Note this description of a SoDoMI feature: <begin> You can do things like super-distribution, for example, where you can e-mail the song and say, "If you get 10 of your best friends to buy it, I'll give you free tickets to the Britney Spears concert next month." So you get on AOL and you e-mail the thing to 50 of your best friends and so on. And with InterTrust you can go down as many levels as you want, so they can e-mail it to 50 of their best friends, turning the consumer into a distributor of sorts. We think offers like that are going to be very compelling. <end> Yep. Spam strangers and win prizes. Wonderful.
What kind of dumb-arse retro computer crap is this Jobs guy up to anyway? The headline talks about notebooks with no keyboard, but then using handwriting instead. Not "choice of input device". Well, handwriting (designed for bird quills) is an obsolete transitional technology. If you're going to replace keyboard then you should go back to cuneiform! Yeah, that's the ticket. Provide a stylus and chisel, and have the user hammer patterns into the screen. Much nicer than scratching at a pad like a chicken with a goose feather caught in its foot.
The point of MacOS X is to have the robustness of Unix (vs. the older primitive floppy-based kernel) with the "user friendliness" of the Mac. Sounds like a good deal to me, if you're looking to attract the types of people who favor Macs (and who by inference generally wouldn't be comfortable in a classical Unix environment).
I don't have any screenshots, but I did try out AbiWord recently. PLUG-UGLY FONTS! I'm talking about little ones, your basic 10-12 point fonts on an 1152-pixel wide 17 inch screen. KDE's native Helvetica looks okay, but the rest of the font collection is mostly unreadable. It is an amazing and sad contrast to, say, Win98. And need I say more about how bad a job Netscrape does?
Now my Linux is raw Mandrake 6.1 (7.0 won't run on my system; it has problems with both the PS/2 mouse and the Tulip ethernet). You note, "And in fact can be fixed with a 10 second session with XF86Config". Can you be more specific?
If the distros do a bad job, then they're to blame, but it is not something obvious for a user to fix. All help appreciated.
Unix was designed for programmers, scientists and engineers back when PHBs and secretaries didn't use computers. So should non-scientists remain in the grip of Bill, or should they have alternatives? That's really the question.
Unix ain't great for everybody. I've long complained that the design of the shells, not to mention a lot of other details, imply a "boy's club" mentality, wherein shared secrets (non-obvious commands, for instance) are the requisites of membership. The Linux programmers are, of course, upper-level members of the society. Geek pride and all that. So I don't think that Unix in anything resembling its raw form is what the Rest Of Us need.
But give Apple credit (and btw I'm not a Mac fan) where due: In MacOS X, they're merging a Unix kernel with a Mac API and GUI and it looks like they'll pull it off. The robustness of the Unix kernel keeps it running, the Unix API helps programmers add capabilities, and the Mac layers make it palatable to mere muggles.
I think KDE and to a lesser extent GNOME are doing something similar, but they're far from ready for prime time. Frankly the font rendering on KDE sucks (well, it's based on antiquated X11 technology, after all). Until such core display issues are fixed, Linux will simply be too ugly for widespread use.
Give Bill credit where due: Win98 does a fantastic job of displaying text. The open source community needs to learn from them.
BTW I'd be very open to a non-Unix non-Bill OS too. As Miguel pointed out in his recent essay, Unix sucks. It was a great experiment but we're stuck with some obsolete ideas. I wish somebody would really rethink things and build an OS that takes advantage of what we've learned since Ritchie et al started their important work 30+ years ago.
Just wondering... this is a pretty absurd theft. Since the registry is American, does the "former" domain holder have the right to sue the registry or Brazilian team for that matter? That would put the case before an American jury. Or is WIPO somehow above American civil law? (anybody can sue anybody for anything here, no?)
I don't know why hjames scored 0 (moderated down, not A.C.?) but he had the point right. Sony had no involvement in the audio cassette patent. Philips patented it around 1963, with a license policy that was basically "free to anyone but you must conform to the spec". That made it a hit, guaranteeing compatibility. Eventually (post-patent expiration) there were some oddball half-speed cassette drives for voice transcription.
Sony is famous for the Beta fiasco. They didn't license it to other makers, so JVC came up with VHS. Just different enough to escape the patents, and licensed to any and all comers. Sony got 100% of a dying market; JVC with its slightly-inferior VHS ended up winning.
So will Sony follow its own precedent or have they learned?
You're flat-out wrong. Sprint is NOT owned by "MCI"! MCI Worldcom owns UUNET, and is *trying* to buy Sprint. But it's far from a done deal. When Worldcom (UUNET) bought MCI, they were required to sell off MCI's Internet (to Cable & Wireless). It's almost certain that the same would happen with Sprint; they'd keep UUNET or Sprint's ISP but not both.
If they did get both, then peering WOULD break, because the combined UUNET/Sprint would own more than 50% of the backbone and it would have no incentive to do peering with anyone. It would literally be without peer, and a monopoly-power situation on the backbone would occur. Fortunately, this problem is well understood on both sides of the Atlantic, and neither the US DOJ nor the EU are likely to allow it.
The Tier 1 backbone ISPs are today competing in a free market. No price regulation, and no monopolies. So the price is what the market will bear. "Peering" is done for mutual self interest. UUNET needs good access to Sprint's subscribers, and Sprint needs access to UUNET's, so they peer for free. That's how the handful of Tier 1s work -- it's in their mutual self interest.
Now you have little guys who have no content to speak of. So it means little to UUNET or Genuity to have good access to them. Yet they want free peering just like the big boys! Hey, I wish my house had free peering over a T3 too, but somebody has to pay for the Tier 1s! They <B>sell</B> backbone service to smaller ISPs. If it were free, where would the money come from? Repeat: This is a free market. The price is not set by the government, but by haggling until both sides' mutual self-interest is met. Not the way some Americans might be used to it, but the real world is full of bargaining. BTW, smaller backbone ISPs (Tier 2s, who almost by definition are the ones whose web sites call themselves "Tier 1", don't always pay a fixed price for Tier 1 upstreaming. They negotiate based on mutual self-interest. As they get bigger, they acquire leverage.)
Ditto to the smaller country players who whine about paying for bandwidth to US peering points. If they had content that the US providers though it was more vital to have, then they might split the cost.
Disclosing source is not the same as "open source". For instance, Digital Equipment Corp. long distributed source code with its VAX/VMS operating system. Typically on microfiche, to be sure, but the idea was that a user should be able to look at the source in order to figure out its behavior. I'm not sure if all of the comments were left in the public fiches, though.
Copyright remains, and they still didn't give away the right of unlicensed use. But it helped keep things reliable. Hell, Windows NT is directly derived from VMS (David N Cutler led both efforts). So certainly MS could disclose Windows sources without giving anything away.
Nutmeg and lime, mostly. No "cola nut" flavor to speak of, nor coca leaf, the original raison'd'etre of CocaCola. Some orange juice, perhaps, in Coke, and perhaps a hint of chocolate in Pepsi.
At least that's what I've read in published "decompilations" of the sodas. The label will be interesting, but then who knows how close it'll taste to the Real Thing.
Not just Sprint. That started as the Southern Pacific Communications Company (SPCC) and got the name "Sprint" in an employee contest. SP was bought by GTE, who later sold it (in two steps) to United Telecommunications, an old telco, which adopted the Sprint name.
Railroads always have signaling bandwidth. The old SPCC/Sprint was built out of trackside microwave, later replaced by fiber optics. MCI got a lot of its rights-of-way by burying fiber optic cables along railroad lines. Qwest was created by somebody who owned the Southern Pacific for a while (Anschultz) and who kept the right to bury cable under it when he sold the rest.
So it's natural for a country like India, with lots of railway, to look towards the rail network as the basis of telecommunications. If it hasn't been done by now, it's probably because of the politics between competing ministries.
Solar cells came later... ECD began in the 1960s with amorphous discrete transistors, whose raison d'etre was to be cheaper to make than crystalline silicon ones. But they were not quite as reliable and, uh, silicon got better and cheaper faster. Eventually he hit upon solar cells and NIMH. But I found it fascinating that he went something like 30 years without a profitable quarter! And that was probably a fluke. Investors have poured money into Ovshinsky's dreams for years. I'm not surprised that ECD was the main holding of the Steadman Fund, long famous as the country's worst-performing mutual fund. (It's not even allowed in some states.) So again I take his inventions with a grain of salt. They're probably based on real physics and chemistry, are demonstrable in the lab, and exciting to investors. But only rarely do they make it to market. I'm trying to visualize his life as a movie. Sort of a Tom Edison Story starring Woody Allen. Tucker as a comedy.
I looked on the web site of a distributor (shown on the LEGO site...) and the one in Pittsburgh lists it at $49. That's the single-user license, of course; they also sell site licenses, etc., for the school market.
Zoom isn't dead either.... Companies who were in the modem business can adapt, making DSL modems, cable modems, ISDN adapters, wireless modems, home routers, etc. So while analog-line modems are on the decline, the companies who make them can reorient their skills towards replacement products. Just as, for instance, H-P reoriented its computer folks away from minicomputers (the 1000, 3000, etc.) and towards PCs and workstations.
The interesting question in a spinoff is to see just what assets are spun off, and what are kept. The old "USR" corporate boundaries are not likely to be respected.
Out Standing In Its Field (an obituary)
on
R.I.P. Iridium
·
· Score: 3
Iridium may have finally met its untimely end. Motorola's dream of a satphone system, paid for by foolish investors but not customers, never really had a chance.
Besides the price, Iridium had performance issues. It was based on GSM, but its link power budget limited the net bit rate to 2400, which meant that the voice quality wasn't so hot. And its data applications are somewhat limited, though by no means nonexistent (not everything is fast web browsing!).
But it didn't work in a car. It didn't work indoors. It didn't even work well through leafy trees, or if buildings were in the way. It was for people who were "out standing in their field."
They designed it before cell phones were widely available around the world, and continued to build it for business travelers even after that market no longer needed them. They designed it with complex sat-to-sat relaying, bypassing cheap terrestrial fiber optics. Again, a 1985 design decision gone bad. These errors add up.
It was called Iridium because that precious metal is element number 77, and there were to be 77 satellites in the original constellation. They later lowered it to 66, a decision which might have strained its performance budget even more than it saved on cost... but they didn't change which element it was named for. How fitting, though, that element 66 is named Dysprosium. (Other slashdotters are invited to check out its etymology.)
I tip my virtual hat to whoever it was (on Slashdot) who said something like this: The trouble with Linux is that 98% of users make the other 2% look bad.
Whenever I go on abUsenet to ask a technical question, I pretty much expect to be confronted, if I get an answer at all, with off-point jabs. That is, if I ask a question that reveals that I'm, yes, using (alert alert!) *Windows* for anything at all, then a vocal subset of Linux geeks won't answer my Linux questions, but will instead excoriate me for not going 100% Windows-free. Dumb stuff like that.
It's not a new phenomenon. I used to read one of the "net.religion" newsgroups about 15 years ago, and the attacks between people who disagreed on theological issues were vicious. They're pretty much the same now, though I gave up wading through the chaff looking for grains of wheat about a decade ago. I think that is the type of "old days" on line discussion that set the tone, too often, for many of today's online communities.
I do however subscribe to some mailing lists where users get rather hostile to those who breach norms of civility, and where a list owner can if necessary dump a jerk off the list. That's a more pleasant environment. And I think web site moderators should *moderate* aggressively. (Slash is a pretty good medium for this; it filters out some of the flames for me, though of course they hang around at 0 or 1 before a moderator sees them.)
I think I *am* the target Corel user. I'm a regular computer user, literate, but not a Unix Guy. I've played with Linux on and off for over five years (since Yggdrasil) but it's not my home OS. (We have to use Windoze at work, and I moved to it at home when OS/2 finally became useless.) So I eagerly awaited Corel Linux and went through the serious effort to download it and burn a CDROM. (I didn't have enough hard disk space left to make it easy.)
Corel's installer has nice eye candy, but it gets the GUI concept wrong -- a typical decent Windows install wizard (even Win98 itself) has lots of defaults but lets you change things. Corel has not enough control. It then wiped out my MBR, deciding, without asking, that LILO belongs there. And of course it *requires* you to repartition and then format a new partition during an EXT2 install, which is dumb (I already had one; I was replacing Mandrake 6.0.)
Yes, the new file manager is cute, but when I log in as a user (not root), it hides "system" and only shows me my accounts. The KDE menus leave out most KDE apps; root can go to/usr/X11R6/bin but a normal user can't, and there's no documentation on adding to the start menu. So lots of stuff just doesn't show up. Worst of all, on my system, logging out of KDE hangs the system hard. And those are just a few beefs.
So yesterday I plugged in a new Mandrake 6.1 disk. Yes, install took a long time, because I went through the RedHat "custom" package selection and there were a zillion or so packages to choose from. Once I finished the kid-in-a-candy-shop routine, it went right in. Oh, it would have been a bit hard without previous RedHat and Mandrake experience, and it's ugly (a VGA16 installer would help a lot on the eye candy side), and it still forgets to do sndconfig, but I ended up with a nice working system. AND the Mandrake 6.1 installer puts everything onto the Start menu. With a menu editor -- but I can't see how to delete entries I don't want. (KDE Bug: IF there are more entries than fit vertically on screen, they get lost off the bottom. Windows gets that right.)
So Mandrake wins my thanks as a nice easy system to get running and one that works the way I'd expect it to. Corel 1.0 is just a bad joke, demoware that looks good on a properly-packaged system, and maybe right for some particular hardware/user taste combinations, but it needs lots of work.
The public doesn't know real-time from batch
on
Apocalypse Not
·
· Score: 4
As I summarize it, computers generally view time in one of three ways:
1) Real-time. Computers run machines without human intervention. 2) Interactive. People sit in front of computers and see them operate. (PCs, etc.) 3) Batch. Computers do work without human intervention, but just pound out paper monthly or so.
Many/most real-time systems don't give a hoot what year it is; they work in seconds or milliseconds. So the nut cases were worried about failures of real-time systems (electric, water, etc.) but those rarely cared much about date. Maybe some paleo-PCs will have reboot problems. Most of these were remediated; this is where the hype and reality were way out of sync.
Interactive systems can have visible clock errors. But humans can work around them. Few Y2K bugs showed up. Big surprise, not.
Batch systems are most likely to use old COBOL code with spotty source decks. Those are the ones that had the most Y2K problems, and those problems won't all turn up for a while. Mostly they're miscaluclated interest, payments, etc. They're not apocalyptic and can be fixed after the fact when somebody sees a billing error. This stuff did cost a fortune to fix and it had to be done, but it's not the stuff of bad TV movies.
Uh, I know how/. readers like to use the word "geek" in its new meaning, but do y'all know what it previously meant (and still does)?
A carnival side-show performer who bites the heads off of live chickens. Geek!
Now disregarding that meaning, I certainly think we need to credit Edison, Einstein, Leonardo, Galileo, Tesla, and all those other great geniuses and hackers!
Seriously... When I got my ticket in 1965, there were three classes that counted, Novice, Tech and General. Extra was just honorary. The ARRL wanted to sell more study guides so they pushed "Incentive Licensing", with reduced General privs and more exams. It largely killed ham radio in the USA; growth has been very low ever since. The new scheme is a lot like pre-1965s, except that now the top code test is 5 WPM, the three licenses are a more modern mix, and you need a second theory test to go from tech to general. So it's going to hurt study guide sales, but help ham radio. - fred k1io
Slashdot Mirror
A few years ago, Uncle Charlie (the FCC) came up with a wacky scheme. They allowed local telephone companies (LECs) to impose a "Primary Interexchange Carrier Charge" (PICC, as in "pixie") on every line, with the bill sent to the presubscribed LD carrier. It was I think $0.53 on the primary residential line but higher on second lines and business lines. This is what got LD companies to start imposing minimums or monthly fees -- they passed it right along, only since they didn't get hteir PICC bills itemized by line, they couldn't pass them directly, so they were averaged.
If you didn't have a PIC (e.g., no LD carrier), then you the subscriber got the PICC bill. No escape.
This stupid scheme was finally dropped, I think, in July of this year. The End User Common Line charge (monthly "access charge") will go up instead; it has been capped at $3.50 on primary resi lines and $6 on others. THis is NOT a charge for LD service, but a charge for that portion of local service which is technically assigned to the FCC's jurisdiction. Don't ask. (Smith v. Illinois Bell, US Sup. Ct. 1927.)
http://www.magnet.state.ma.us/dpu/telecom/slamlett .pdf
That's the state regulator's (DTE) letter to consumers about slamming. There's a bunch more on their web site, and a state law against it, as well as federal rules against it.
State law requires Third Party Verification by an authoritized "TPV" vendor, *or* tape recording of your phone call, *or* written authorization. It is not applicable only to ILECs (VeriZontal) but to CLECs like MediaOne/AT&T too. There are stiff penalties. Of course AT&T knows this, as they were on the losing end of zillions of MCI slams a decade or so ago; that's what led to the stricter rules. It's probably a third-party telemarketing company they retained who is desperate to make quota, but it's still AT&T's responsibility.
BTW I use AT&T/MediaOne's phone service in the Boston area too, and it's quite good. I also use AT&T long distance, with online billing (also a pretty good rate, 9c days 5c nights, no fee; I don't think they advertise it but it's available for the asking; international however is a different story). This hasn't stopped AT&T (the long distance side) from calling me asking me to switch *to* AT&T, from sending me checks whose indorsement consitutes authorization to switch me to AT&T, etc. So their internal records are (to use a famous Boston term for a young codfish) scrod.
AT&T's a big company. The consumer LD side is incredibly profitable but shrinking as LD rates decline. So Wall Street is unhappy, and AT&T's considering spinning the whole thing off. AT&T Broadband is still basically two companies, MediaOne and ex-TCI. They're not well coordinated. AT&T Broadband is talking about offering their own LD service, but for now doesn't seem to have any LD billing.
The FCC is opening a dangerous door if they think they have authority over a protocol or applications-layer operation, such as AIM. They regulate raw pipes, not applications or for that matter IP (which is technically "information service", not "telecommunications").
If AOL's behavior is in some way "anticompetitive", there are other agencies (DoJ, for instance) who do have general say over that matter. And mergers do invite scrutiny. So if there's some specific DoJ question to be answered around their IM networks, then fine. The FCC does have some authority over cable systems, which AOL is trying to buy. But the FCC should NOT consider IM to be theirs to regulate.
Capital cost is not the only expense. Certainly a lot of networks (including the 19th century railways) were only profitable after being picked up at a discount in bankruptcy, but Iridium's operating expense is astronomical too.
Motorola set it up so that they wouldn't lose. They got paid for the satellites and paid to run them. Somebody else could probably run them much cheaper but Motorola doesn't want that to happen; that could make them look bad in the bankruptcy court.
Note this description of a SoDoMI feature:
<begin>
You can do things like super-distribution, for example, where you can e-mail the song and say, "If you get 10 of your best friends to buy it, I'll give you free tickets to the Britney Spears concert next month." So you get on AOL and you e-mail the thing to 50 of your best friends and so on. And with InterTrust you can go down as many levels as you want, so they can e-mail it to 50 of their best friends, turning the consumer into a distributor of sorts. We think offers like that are going to be very compelling.
<end>
Yep. Spam strangers and win prizes. Wonderful.
What kind of dumb-arse retro computer crap is this Jobs guy up to anyway? The headline talks about notebooks with no keyboard, but then using handwriting instead. Not "choice of input device". Well, handwriting (designed for bird quills) is an obsolete transitional technology.
If you're going to replace keyboard then you should go back to cuneiform! Yeah, that's the ticket. Provide a stylus and chisel, and have the user hammer patterns into the screen. Much nicer than scratching at a pad like a chicken with a goose feather caught in its foot.
The point of MacOS X is to have the robustness of Unix (vs. the older primitive floppy-based kernel) with the "user friendliness" of the Mac. Sounds like a good deal to me, if you're looking to attract the types of people who favor Macs (and who by inference generally wouldn't be comfortable in a classical Unix environment).
I don't have any screenshots, but I did try out AbiWord recently. PLUG-UGLY FONTS! I'm talking about little ones, your basic 10-12 point fonts on an 1152-pixel wide 17 inch screen. KDE's native Helvetica looks okay, but the rest of the font collection is mostly unreadable. It is an amazing and sad contrast to, say, Win98. And need I say more about how bad a job Netscrape does?
Now my Linux is raw Mandrake 6.1 (7.0 won't run on my system; it has problems with both the PS/2 mouse and the Tulip ethernet). You note, "And in fact can be fixed with a 10 second session with XF86Config". Can you be more specific?
If the distros do a bad job, then they're to blame, but it is not something obvious for a user to fix. All help appreciated.
Actually, it's a good question.
Unix was designed for programmers, scientists and engineers back when PHBs and secretaries didn't use computers. So should non-scientists remain in the grip of Bill, or should they have alternatives? That's really the question.
Unix ain't great for everybody. I've long complained that the design of the shells, not to mention a lot of other details, imply a "boy's club" mentality, wherein shared secrets (non-obvious commands, for instance) are the requisites of membership. The Linux programmers are, of course, upper-level members of the society. Geek pride and all that. So I don't think that Unix in anything resembling its raw form is what the Rest Of Us need.
But give Apple credit (and btw I'm not a Mac fan) where due: In MacOS X, they're merging a Unix kernel with a Mac API and GUI and it looks like they'll pull it off. The robustness of the Unix kernel keeps it running, the Unix API helps programmers add capabilities, and the Mac layers make it palatable to mere muggles.
I think KDE and to a lesser extent GNOME are doing something similar, but they're far from ready for prime time. Frankly the font rendering on KDE sucks (well, it's based on antiquated X11 technology, after all). Until such core display issues are fixed, Linux will simply be too ugly for widespread use.
Give Bill credit where due: Win98 does a fantastic job of displaying text. The open source community needs to learn from them.
BTW I'd be very open to a non-Unix non-Bill OS too. As Miguel pointed out in his recent essay, Unix sucks. It was a great experiment but we're stuck with some obsolete ideas. I wish somebody would really rethink things and build an OS that takes advantage of what we've learned since Ritchie et al started their important work 30+ years ago.
Just wondering... this is a pretty absurd theft. Since the registry is American, does the "former" domain holder have the right to sue the registry or Brazilian team for that matter? That would put the case before an American jury. Or is WIPO somehow above American civil law? (anybody can sue anybody for anything here, no?)
I don't know why hjames scored 0 (moderated down, not A.C.?) but he had the point right. Sony had no involvement in the audio cassette patent. Philips patented it around 1963, with a license policy that was basically "free to anyone but you must conform to the spec". That made it a hit, guaranteeing compatibility. Eventually (post-patent expiration) there were some oddball half-speed cassette drives for voice transcription.
Sony is famous for the Beta fiasco. They didn't license it to other makers, so JVC came up with VHS. Just different enough to escape the patents, and licensed to any and all comers. Sony got 100% of a dying market; JVC with its slightly-inferior VHS ended up winning.
So will Sony follow its own precedent or have they learned?
You're flat-out wrong. Sprint is NOT owned by "MCI"! MCI Worldcom owns UUNET, and is *trying* to buy Sprint. But it's far from a done deal. When Worldcom (UUNET) bought MCI, they were required to sell off MCI's Internet (to Cable & Wireless). It's almost certain that the same would happen with Sprint; they'd keep UUNET or Sprint's ISP but not both.
If they did get both, then peering WOULD break, because the combined UUNET/Sprint would own more than 50% of the backbone and it would have no incentive to do peering with anyone. It would literally be without peer, and a monopoly-power situation on the backbone would occur. Fortunately, this problem is well understood on both sides of the Atlantic, and neither the US DOJ nor the EU are likely to allow it.
The Tier 1 backbone ISPs are today competing in a free market. No price regulation, and no monopolies. So the price is what the market will bear. "Peering" is done for mutual self interest. UUNET needs good access to Sprint's subscribers, and Sprint needs access to UUNET's, so they peer for free. That's how the handful of Tier 1s work -- it's in their mutual self interest.
Now you have little guys who have no content to speak of. So it means little to UUNET or Genuity to have good access to them. Yet they want free peering just like the big boys! Hey, I wish my house had free peering over a T3 too, but somebody has to pay for the Tier 1s! They <B>sell</B> backbone service to smaller ISPs. If it were free, where would the money come from? Repeat: This is a free market. The price is not set by the government, but by haggling until both sides' mutual self-interest is met. Not the way some Americans might be used to it, but the real world is full of bargaining. BTW, smaller backbone ISPs (Tier 2s, who almost by definition are the ones whose web sites call themselves "Tier 1", don't always pay a fixed price for Tier 1 upstreaming. They negotiate based on mutual self-interest. As they get bigger, they acquire leverage.)
Ditto to the smaller country players who whine about paying for bandwidth to US peering points. If they had content that the US providers though it was more vital to have, then they might split the cost.
Disclosing source is not the same as "open source". For instance, Digital Equipment Corp. long distributed source code with its VAX/VMS operating system. Typically on microfiche, to be sure, but the idea was that a user should be able to look at the source in order to figure out its behavior. I'm not sure if all of the comments were left in the public fiches, though.
Copyright remains, and they still didn't give away the right of unlicensed use. But it helped keep things reliable. Hell, Windows NT is directly derived from VMS (David N Cutler led both efforts). So certainly MS could disclose Windows sources without giving anything away.
Nutmeg and lime, mostly. No "cola nut" flavor to speak of, nor coca leaf, the original raison'd'etre of CocaCola. Some orange juice, perhaps, in Coke, and perhaps a hint of chocolate in Pepsi.
At least that's what I've read in published "decompilations" of the sodas. The label will be interesting, but then who knows how close it'll taste to the Real Thing.
Not just Sprint. That started as the Southern Pacific Communications Company (SPCC) and got the name "Sprint" in an employee contest. SP was bought by GTE, who later sold it (in two steps) to United Telecommunications, an old telco, which adopted the Sprint name.
Railroads always have signaling bandwidth. The old SPCC/Sprint was built out of trackside microwave, later replaced by fiber optics. MCI got a lot of its rights-of-way by burying fiber optic cables along railroad lines. Qwest was created by somebody who owned the Southern Pacific for a while (Anschultz) and who kept the right to bury cable under it when he sold the rest.
So it's natural for a country like India, with lots of railway, to look towards the rail network as the basis of telecommunications. If it hasn't been done by now, it's probably because of the politics between competing ministries.
Solar cells came later... ECD began in the 1960s with amorphous discrete transistors, whose raison d'etre was to be cheaper to make than crystalline silicon ones. But they were not quite as reliable and, uh, silicon got better and cheaper faster. Eventually he hit upon solar cells and NIMH. But I found it fascinating that he went something like 30 years without a profitable quarter! And that was probably a fluke. Investors have poured money into Ovshinsky's dreams for years. I'm not surprised that ECD was the main holding of the Steadman Fund, long famous as the country's worst-performing mutual fund. (It's not even allowed in some states.) So again I take his inventions with a grain of salt. They're probably based on real physics and chemistry, are demonstrable in the lab, and exciting to investors. But only rarely do they make it to market. I'm trying to visualize his life as a movie. Sort of a Tom Edison Story starring Woody Allen. Tucker as a comedy.
I looked on the web site of a distributor (shown on the LEGO site...) and the one in Pittsburgh lists it at $49. That's the single-user license, of course; they also sell site licenses, etc., for the school market.
Zoom isn't dead either.... Companies who were in the modem business can adapt, making DSL modems, cable modems, ISDN adapters, wireless modems, home routers, etc. So while analog-line modems are on the decline, the companies who make them can reorient their skills towards replacement products. Just as, for instance, H-P reoriented its computer folks away from minicomputers (the 1000, 3000, etc.) and towards PCs and workstations.
The interesting question in a spinoff is to see just what assets are spun off, and what are kept. The old "USR" corporate boundaries are not likely to be respected.
Besides the price, Iridium had performance issues. It was based on GSM, but its link power budget limited the net bit rate to 2400, which meant that the voice quality wasn't so hot. And its data applications are somewhat limited, though by no means nonexistent (not everything is fast web browsing!).
But it didn't work in a car. It didn't work indoors. It didn't even work well through leafy trees, or if buildings were in the way. It was for people who were "out standing in their field."
They designed it before cell phones were widely available around the world, and continued to build it for business travelers even after that market no longer needed them. They designed it with complex sat-to-sat relaying, bypassing cheap terrestrial fiber optics. Again, a 1985 design decision gone bad. These errors add up.
It was called Iridium because that precious metal is element number 77, and there were to be 77 satellites in the original constellation. They later lowered it to 66, a decision which might have strained its performance budget even more than it saved on cost... but they didn't change which element it was named for. How fitting, though, that element 66 is named Dysprosium. (Other slashdotters are invited to check out its etymology.)
I tip my virtual hat to whoever it was (on Slashdot) who said something like this: The trouble with Linux is that 98% of users make the other 2% look bad.
Whenever I go on abUsenet to ask a technical question, I pretty much expect to be confronted, if I get an answer at all, with off-point jabs. That is, if I ask a question that reveals that I'm, yes, using (alert alert!) *Windows* for anything at all, then a vocal subset of Linux geeks won't answer my Linux questions, but will instead excoriate me for not going 100% Windows-free. Dumb stuff like that.
It's not a new phenomenon. I used to read one of the "net.religion" newsgroups about 15 years ago, and the attacks between people who disagreed on theological issues were vicious. They're pretty much the same now, though I gave up wading through the chaff looking for grains of wheat about a decade ago. I think that is the type of "old days" on line discussion that set the tone, too often, for many of today's online communities.
I do however subscribe to some mailing lists where users get rather hostile to those who breach norms of civility, and where a list owner can if necessary dump a jerk off the list. That's a more pleasant environment. And I think web site moderators should *moderate* aggressively. (Slash is a pretty good medium for this; it filters out some of the flames for me, though of course they hang around at 0 or 1 before a moderator sees them.)
I think I *am* the target Corel user. I'm a regular computer user, literate, but not a Unix Guy. I've played with Linux on and off for over five years (since Yggdrasil) but it's not my home OS. (We have to use Windoze at work, and I moved to it at home when OS/2 finally became useless.) So I eagerly awaited Corel Linux and went through the serious effort to download it and burn a CDROM. (I didn't have enough hard disk space left to make it easy.)
/usr/X11R6/bin but a normal user can't, and there's no documentation on adding to the start menu. So lots of stuff just doesn't show up. Worst of all, on my system, logging out of KDE hangs the system hard. And those are just a few beefs.
Corel's installer has nice eye candy, but it gets the GUI concept wrong -- a typical decent Windows install wizard (even Win98 itself) has lots of defaults but lets you change things. Corel has not enough control. It then wiped out my MBR, deciding, without asking, that LILO belongs there. And of course it *requires* you to repartition and then format a new partition during an EXT2 install, which is dumb (I already had one; I was replacing Mandrake 6.0.)
Yes, the new file manager is cute, but when I log in as a user (not root), it hides "system" and only shows me my accounts. The KDE menus leave out most KDE apps; root can go to
So yesterday I plugged in a new Mandrake 6.1 disk. Yes, install took a long time, because I went through the RedHat "custom" package selection and there were a zillion or so packages to choose from. Once I finished the kid-in-a-candy-shop routine, it went right in. Oh, it would have been a bit hard without previous RedHat and Mandrake experience, and it's ugly (a VGA16 installer would help a lot on the eye candy side), and it still forgets to do sndconfig, but I ended up with a nice working system. AND the Mandrake 6.1 installer puts everything onto the Start menu. With a menu editor -- but I can't see how to delete entries I don't want. (KDE Bug: IF there are more entries than fit vertically on screen, they get lost off the bottom. Windows gets that right.)
So Mandrake wins my thanks as a nice easy system to get running and one that works the way I'd expect it to. Corel 1.0 is just a bad joke, demoware that looks good on a properly-packaged system, and maybe right for some particular hardware/user taste combinations, but it needs lots of work.
As I summarize it, computers generally view time in one of three ways:
1) Real-time. Computers run machines without human intervention.
2) Interactive. People sit in front of computers and see them operate. (PCs, etc.)
3) Batch. Computers do work without human intervention, but just pound out paper monthly or so.
Many/most real-time systems don't give a hoot what year it is; they work in seconds or milliseconds. So the nut cases were worried about failures of real-time systems (electric, water, etc.) but those rarely cared much about date. Maybe some paleo-PCs will have reboot problems. Most of these were remediated; this is where the hype and reality were way out of sync.
Interactive systems can have visible clock errors. But humans can work around them. Few Y2K bugs showed up. Big surprise, not.
Batch systems are most likely to use old COBOL code with spotty source decks. Those are the ones that had the most Y2K problems, and those problems won't all turn up for a while. Mostly they're miscaluclated interest, payments, etc. They're not apocalyptic and can be fixed after the fact when somebody sees a billing error. This stuff did cost a fortune to fix and it had to be done, but it's not the stuff of bad TV movies.
Uh, I know how /. readers like to use the word "geek" in its new meaning, but do y'all know what it previously meant (and still does)?
A carnival side-show performer who bites the heads off of live chickens. Geek!
Now disregarding that meaning, I certainly think we need to credit Edison, Einstein, Leonardo, Galileo, Tesla, and all those other great geniuses and hackers!
Seriously... When I got my ticket in 1965, there were three classes that counted, Novice, Tech and General. Extra was just honorary. The ARRL wanted to sell more study guides so they pushed "Incentive Licensing", with reduced General privs and more exams. It largely killed ham radio in the USA; growth has been very low ever since. The new scheme is a lot like pre-1965s, except that now the top code test is 5 WPM, the three licenses are a more modern mix, and you need a second theory test to go from tech to general. So it's going to hurt study guide sales, but help ham radio. - fred k1io