Re:Useful floppy-disk or CD-RW replacement, maybe
on
Quarter-sized CD's?
·
· Score: 2
It's a little smaller than the floppy, but looks similar. The disk itself sits towards one end, so it's rectangular, not square, and thus narrower than a floppy.
I got the impression that they were marketing the technology to all comers, including computers. But of course it's within their power to blow it. I've often seen good technology get mismarketed. Indeed that's probably the usual fate....
Useful floppy-disk or CD-RW replacement, maybe
on
Quarter-sized CD's?
·
· Score: 2
I have seen Dataplay prototypes and played with an actual disk. It's cute. The actual disk is encapsulated in a sleeve that looks like a 3-inch floppy, only a bit smaller. Critical in its favor is the metal shutter door. You do not scratch these on the tabletop or in the car. They're proteted. That's good. That's what's wrong with CDs -- even after error correction, they can easily fail, because they're so exposed. Stinkin' jewel cases are far more cumbersome than the cardboard sleeves that old LPs came in, but DataPlay solves this with a hard shuttered enclosure. (Anybody remember CD caddies?) And it still fits into a shirt pocket or Walk-creature-sized machine.
Now the DRM is an option that, of course, all of the prerecorded music companies will invoke. And I can't comment on its crackability. But if you take a drive by itself, and some blanks, then you can ignore the DRM, because it's your own drive. And it fits into a laptop, or a desktop, and can replace a floppy, while being a whole lot nicer than Zip or LS-120.
It's a tiny bit smaller in capacity than CD-RW, and I wonder if that's its weakness. Just a tiny bit more diameter on the disk would give it a lot more area, and it could have been 1-2 GB and still smaller than a floppy.
It'll be interesting to see how they spin this, and how the public responds.
CAP is basically obsolete. It was the first ADSL technology, and has worked okay in some cases. But it has one big disadvantage. CAP causes crosstalk. It puts a lot of energy into some narrow spectra, and it leaks into adjacent wire pairs. The FCC is cracking down on this and CAP might not pass future standards. If a telco deploys one DSL line out of a hundred subscribers, CAP will work, but as the subscsriber density rises, CAP causes trouble
DMT spreads the signal differently, and thus has less crosstalk. It is thus the "standard". It doesn't always perform as well and is more latent though.
Fact: There is no government-granted cable monopoly. There are two cable companies passing my house; I use one of them (AT&T), and my next-door neighbor uses the other (RCN). Municipalities have been prohibited by federal law from granting exclusive franchises since 1992; even before then, there were few exclusives. Financially, it is generally a lousy investment to be the second cable company in a given place, which is why overbuilding is so rare. But lack of competition does not equate to government monopoly.
But in any case, if you want to regulate an ISP because it is so good, then what else do you regulate? If it MUST by law deliver me ten pitches a day to enlarge my penis size or get a mortgage from some crook, then what else must an ISP do by law? It would be a victory for lawyers (since ISPs would need a lot of them) but most ISPs would simply go out of business rather than be subject to regulation.
EFF has it wrong this time. They make the statement that e-mail is "protected speech". That's a legal issue in the USA, which means that the government doesn't have the right to block it. But private parties are also not required to pay to relay it.
The Internet is not regulated as a telecom service. The FCC doesn't regulate ISPs, just the telecom services they buy. Nobody regulates mail servers. It's a free market, and it works. Now in a free market, you have competition. If your ISP uses MAPS and you don't like it, then you're free to go elsewhere. If your ISP is RBL'd, you're free to go elsewhere. There are lots of free e-mail services out there. See for instance http://www.emailaddresses.com/ . Now I wish my own "primary" e-mail provider, the one I ping many times a day, used one of these services, because I'm spammed to death and sick of it! If somebody couldn't get through, they almost certainly would find another way to reach me. Like I have a phone too, not to mention other e-mail addresses.
So given the fact that there is no anti-spam legislation, and negligible likelihood of effective anti-spam legislation within the next few years, then the free market approach (you know, the one the spammers cite to block anti-spam legislation) is to allow anti-spam filters at the ISPs. The ISPs will install them if it's good for business, and block spammers if being blackholed is bad for business.
Indeed one of the reasons that the Internet is not regulated as a "telecommunications service" is that it does not offer to provide transport of information "without change in form or content" -- an ISP may change things, of which blocking spam is one example. It would be quite a different story if a telecomm provider attempted to do the same thing -- their mission is to pass the bits unchanged, down there below layer 3.
And please don't tell me how easy it is to build an anti-spam filter on your private mail server. 99.9% of end users do no not run mail servers; ISPs, who have full-time bandwidth, run them for us.
Wireline routing protocols like OSPF are probably not well suited to wireless; there are too many assumptions that break. Radio links vary in quality and are less reliable than wireline; there's also mobility, as a given address might move around within the network. Or even if it stands still, its *logical* location -- which node it's getting best coverage from -- might change. Conventional TCP/IP doesn't handle that well.
One of the projects I've worked on in the distant past that directly addresses wireless networking is RSPF (radio SPF). This is a route-determination ("interior gateway") protocol designed for heterogeneous radio links. Originally just for ham radio AX.25, and very compact compared to OSPF or BGP, it could be adapted for other things like this. RSPF code, albeit experimental, is included in debian and SuSe distros. It adds a routing layer within the "subnet", so it doesn't have to look like a reliable fully-connected LAN.
Still, it's not clear how well a neighborhood network would work in practice. 802.11a, for instance, is limited to indoor use -- it's down in the 5.2 GHz low-power end of the UNII band, which is shared with satellites. The 5.7 GHz end allows outdoor use and more power, but cheap radios are still elusive. And it's sensitive to foliage fade. That's probably where NANs would make the most sense though. 802.11{b}, at 2.4 GHz, is cheap and has some passable range, especially the lower non-b speed. But 2.4 GHz is shared with microwave ovens, cordless phones, and other junque, which makes it tricky to use in urban areas.
I don't know waht Pair is doing. I do know that I can't run Linux Mandrake 8.1 on my VIA KT133A-based Athlon. Mostly it works, but mounting the SCSI-emulated CD-RW causes kernel panic. Various discussions on abUsenet a.o.l.m end up with "VIA chips are buggy". Funny, Windows has no trouble with them. But who knows what is happening there if Pair is running Linux.
Maybe the problem is Linux and a lack of time spent learning how to work with VIA chips.
While it's true that AMD CPUs are, uh, sensitive to cooling, I don't see that as a show-stopper. When you buy the parts to build your own Athlon system, as I did recently, you get plenty of warning to NOT TURN THIS ON WITHOUT A HEATSINK (yes, they shout, as they should).
Other CPUs are also very sensitive. What's rather surprising is how well Intel's P4 thermal shutdown works. I suspect AMD will get around to doing something similar. But in the meantime, I've attached a nice quiet (3800 RPM, not the 7200 RPM version) ThermoEngine to my Thunderbird, and it cruises at around 100 degrees F. Some newer/bigger heatsinks bolt to the motherboard, rather than clip on to the socket, which I suppose helps if you're really paranoid about its falling off. I use Motherboard Monitor to keep track of the temp via the Win98 system tray, and wish Linux distros would include similar capability out of the box (yeah, I know there's a way to build it in yourself...).
But then I do admit to using a 1 GHz Tbird rather than a faster one because I don't want that excess heat or power consumption.
Preface: I'm amazed at how poorly Slashdotters read the question. The post is about a 45 mile hop in rural Canada -- this is not the usual suburban nerd's home connection. No FCC, no RBOCs, and no, you can't just trench 45 miles of fiber optics for C$125k. (That's about what one mile of urban trenching costs, or maybe ten miles of rural Ditch-Witch burial.)
This type of application can, I'm sure even in Canada, use licensed point-to-point microwave. This allows lots higher power than 802.11 (forget the "b" which means higher speeds for even shorter distances). Typical rule-of-thumb is that frequencies under 10 GHz can go up to 30 miles (okay, say 45 km) on a single hop, if you can get line of sight. The site in question might need a repeater along the way. The terrain is all-important.
It probably is possible to get some microwave radios on that budget, though a repeater would possibly blow the limit. Harris, for instance, has a good selection, and a free program, Starlink, on their web site, which does path calculations for various radio - antenna combinations. (You can source the radios elsewhere, but Starlink is obviously geared to match Harris' own radios.) These would probably deliver 3 to 45 Mbps, depending on the radio in question. Industry Canada (which regulates spectrum matters) would probably be able to point you in the right direction for licensing and frequency/path coordination.
In New England, of which Maine is a part, "City" is defined by its type of government. A "Town" has a town meeting, either "open" (all voters can come) or "representative" (vote for your neighborhood's representatives, so TM is more like a legislature). I'm not sure about Maine, which spun off of Massachusetts in 1820, but in Massachusetts, a Town doesn't have a mayor; the Selectmen are the executive. A "City" in contrast has a mayor and council, no town meeting.
So we have Cities typically ranging from 10k population up, but Towns can be quite large too. Last year, several Towns in Massachusetts adopted City government. (But the largest, with over 60k pops, refuses.) Maine's cities are all small by Ontario standards, but they tend to be regional centers, and if they were Towns, they'd be too large for open meeting anyway.
Back to the original topic -- I think it'sa shame that Slashdotters overlay their aversion to Sprawl atop midcoast Maine, which really doesn't look a bit like Orange County.
Cringely is really off the wall this time. Yes, there are lots of failed providers of broadband, but there are others doing okay. Mostly small ones who don't have NASDAQ ticker symbols and big publicity.
@Home failed because it was a bad business. They had a nice gig doing the ISP stuff for the cable industry, but they got caught up in dotcom mania and bought the third-rate search engine Excite for a ridiculous amount. Excite never had a prayer of breaking even, so the whole thing was weighted down. Excite was also irrelevant to @Home's mission, which was to provide the cablecos with an ISP back end.
The data CLECs who tanked had bad business plans too. They mostly spent too much on collocation cages (needed before 1998 to access the loops) and they went into each others' markets, so a single telco CO would have half a dozen of them dividing the market among them. They also designed for a high breakeven, assuming that the others would have no market share. And they had big expense structures. So they tanked.
Cablecos do not need @Home any more. They can create in-house ISPs, as MediaOne did (ignore the @Home label, which is a borrowed trademark used because AT&T now owns them). They can and will also learn to work with ISPs, providing (without being forced) choice in ISP service. That does require some serious network reconfiguration, and since @Home had exclusive contracts with most of the cablecos into 2002, the cablecos aren't ready to open up. But with @Home finally being put out of its misery, the cablecos might finally recognize that they should work with other ISPs.
MediaOne's network has been called Continental Express, Highway One, Road Runner and finally @Home, but it is really something built locally and quite separate from @Home. The former-TCI parts of AT&T Broadband actually use the @Home plant, and are affected by @Home's problems. The former-MediaOne portions are not; they use the @Home trademark and little more, and are still taking orders and installing. And yeah, it works very nicely.
I expect the other @Home cablecos to have a fix in place very, very fast. Either AT&T will "fix" @Home or something else will be done.
Java incompatible with Netscrape 4.7x
on
Mozilla 0.9.5
·
· Score: 2
I have used Mozilla 0.9.3 and it's mostly pretty nice... but there's one show stopper that I don't think is being addressed.
My office uses Lotus Notes for mail (ugh!). Its Domino server lets me read and send mail from a browser, so I don't have to pollute my home PC with that godawful "client". Of course I have to get through the firewall to get there, but at least on Windows they gave me the necessary IPsec client (Nortel) that works with the SecureID card. (That's another reason I can't depend on Linux so far, but that subject it off topic.)
The Domino mail client uses Java to provide useful menu items like "next message" and "reply". This works on Netscape 4.7x and on Internet Exploder, but not on Mozilla. Obviously there's something different about their Javas. Maybe Domino uses an older version and the one in Mozilla isn't backwards compatible?
So until Mozilla can talk to Domino, I'll still need the unstable but well-understood old Netscape client. Suggestions for fixing this are of course welcome. (Flames about still using Windows for anything are not. My current problems with Mandrake 8.1 are off topic too.)
Natural gas doesn't smell by itself. Distributors add methyl mercaptan to it in order to make it smell, so that leaks can be found. (Walk up and down the street in front of my house any day of the year and you'll smell it coming from the rather porous old Boston Gas/Keyspan pipes. They make repairs when their leak detector shows the concentration getting scary.) Mercaptan was chosen because it, well, has a strong and distinctive smell. Acetylene smell similar but is itself explosive.
This practice began after a very unfortunate incident in the 1930s. The public school in London, Texas had been heated by gas that was being, uh, skimmed off of a pipeline passing from a nearby well. The connection wasn't exactly professional. A leak sprang, and gas accumulated in the basement, reaching serious concentrations without being noticed. It blew the school sky-high, killing about 200 people, including most of the children, largely wiping the town off the map. (The town, near Tyler, was renamed New London; it now has about 900 inhabitants.)
People nowadays appreciate methane's properties a bit better. A little cartridge to power a fuel cell should not be a problem.
First off, and speaking as a regular Slashdot karma whore who loves bashing Bill, I think it's cruel to Microsoft to equate VZ and SBC to them! But anyway...
These ILECs do not have a legal monopoly on anything any more. You and I can string wire under or over the street next to theirs. It's just bad business to try. Several companies such as Worldcom (ex-MFS, Brooks, MCI), AT&T (ex-TCG), Metromedia, Hyperion, XO and ELI dumped a few billion dollars burying fiber under the streets in major business districts. Often right next to each other, guaranteeing that nobody would have enough business to be profitable. That's one reason why the telecom sector is so weak now. Local competition was opened nationwide in 1996 and there was so much cheap capital around at the time that a lot was squandered.
There is still a growing CLEC sector, but the failures of Covad (operating in Chapter 11), Rhythms, Northpoint, Vitts, BBO and others have demonstrated how hard it is to operate as a pure-play DSL provider. Many of the surviving CLECs, like Allegiance and McLeod USA, combine their own facilities with leased ILEC wire and do both voice and data. Rhythms leased the wire but only got the data revenue from it.
CLEC survival depends upon the FCC enforcing rules that require the Bells and other ILECs to interconnect and lease critical facilities to them. That has been rather lacking as of late. It also depends upon Congress *not* changing the rules and, for instance, passing the Dingell-Tauzin bill, which would essentially shut CLECs out of the DSL business and make it an unregulated ILEC monopoly.
This isn't entirely new. There were "microsats" flown in the 1980s, some sponsored by the Amateur Satellite Corp. (AMSAT), and some university sats like Webersat (from Utah).
With today's smaller and more powerful chips, of course, it's a lot easier to do more in a small package.
The "telco geek" was right when he noted that there'd be too much NEXT/FEXT (crosstalk). It only works because all of the signals are loud and soft together. The CO transmits at a higher frequency than the modem (frequency split) so the CO-end signals are weak at the modem and, more importantly, the modem-end signals are all weak at the CO. Run it backwards and the CO-end modem would clobber the signals coming in from remotes. The cable on the street was designed for voiceband, and has too much crosstalk at the 50 kHz-1 MHz band used for ADSL.
BUT there are alterantive technologies. SDSL is popular for business; that's Covad's big item. It was designed for symmetry, using echo cancellation on the same frequencies rather than frequency splitting. Of course it's more expensive -- SDSL cannot share the wire with analog phones, so you need a dedicated loop. Plus it is sold to businesses, which tend to use it more intensively than consumers (average bytes/month), so the ISP price is set accordingly.
A technology I really like is Paradyne's MVL. It uses "ping-pong" signaling on the wire, avoiding voiceband frequencies so it can share the loop like ADSL but not SDSL. MVL (also used, with modification, in AGCS' SuperLine) transmits a frame in one direction at a time. If there's nothing to send, it sends a minimum-length burst. So the bandwidth is adaptively directional, allowing you to have a client or server without takling to your provider.... It avoids crosstalk by staying below 120 kHz. Its peak speed is 1 MHz, depending on loop length, but that's still adequate for most users. Oh yeah, unlike ADSL, MVL will work on crufty loops with bridge taps, over 25,000 feet (at degraded speed), just so long as it's unloaded (no series inductors).
Not many telcos use this stuff, since it's not "standard", but a few do; I know a CLEC startup that does. (I've no interest in Paradyne; I just like the technology.) Rumor has it that Rhythms had some, but alas that will probably be confirmed by their pending asset auction....
Out of the box, Linux usually looks lousy. At least it did for my last few installs, including Mandrake 7.1 and 7.2 and RedHat 8.0. The font rendering was plug-ugly, compared to Windoze. Indeed it was barely readable, especially in Netscape.
Now the main problem, of course, was that the profoundly defective AbiSuite fonts were installed in the font path. (Why do Linux distributions still do this?) Thanks to Google Groups, I found out about it and could remove the offending line. After that, the fonts were merely mediocre, maybe as good as Windows 3.0, though it's hard to compare the monitors of those days to now.
After a session of Linux, rebooting to Windows is a treat to the eyes. Not that Windows is better than Linux for everything, but XFree86 seems to have terribly primitive font rendering, while Windows pays close attention to appearances. I do typically insert Windows fonts into Linux, which are better than the usual X fonts, but Linux has still not got the best font rendering engines. It makes a real difference in readability when looking at the small fonts some web sites use.
The X Window System was a clever invention for its day, the early 1980s and Project Athena, where the goal was a "1-1-1" X server (display terminal) system (1 M byte RAM, 1 M pixel screen, 1 MHz CPU). But Linux would benefit from a modern replacement. (What ever happened to Hungry's "Y"?)
Depends on your shop. Where I work, Powerpoint is *everything*. It's a consulting firm which likes to deliver its 6-figure-plus reports to clients in Powerpoint form. They have people who are really good at some Powerpoint design formats, and can get quite a bit onto a page if they want to. Personally, I prefer the linear style of Word with a (very) few pictures, but I'm the exception there.
Outlook, on the other hand, isn't even installed. They use Notes for mail (which is execrable, but safe from viruses) and groupware functions (where it has potential).
For me to jettison 'doze and use Linux would require real two-way filters into and out of PPT, Word and Excel. I haven't found AbiWord to be very good at it. Gnumeric isn't complete either. And even different versions of Powerpoint (which really sucks) aren't compatible with each other; at work, it's all Office 97, and no talk about "upgrades". I suspect that's the most common version nowadays; we exchange a lot of stuff with clients using Office 97 formats.
... but not everybody in rural areas is poor or needs a big subsidy. Rural telco subsidies go to wealthy people's ski chalets in the Rockies, to beach cottages, and to The Donald's various retreats.
And at the end of the day, local telephone rates in these rural areas are actually lower than in town, even though the rural telco is only getting, say, 20-30% of its revenue from the local ratepayers' monthly bills. So the subsidies can be reduced, and targeted more carefully (e.g., more "low income support" in lieu of "high cost support"), while still preserving "universal service".
Unlicensed networks like Ricochet aren't allowed enough power/range to be profitable -- that's the idea! Metricom was caught between having enough base stations to meet needs and having enough customers to finance them. The numbers never converged. Unlicensed wireless works best in rural areas, especially flat ones.
The money for sky-borne relays has dried up with the rest of the industry. Don't hold you breath for those airships. Even if the FCC were to grant spectrum to them.
And it has been illegal for a municipality to grant exclusivity to a cable franchise since 1992. States grant telco franchises, which have been non-exclusive since 1996. There are overbuilders, but the economics aren't great. Check out RCN's quarterlies. From a financial perspective, access to existing ILEC loops is the most viable competitive solution.
I agree that rural areas get too much of a subsidy. But even then, the subsidy mechanism should be made available to competitive providers. That's *theoretically* possible but CLECs have a very hard time getting it. Western Wireless and McLeodUSA may have qualified.
Well, no, your history is wrong. Perhaps you're so young that you're confusing Canada with Cuba?
Canadian telephone service wasn't a Crown Corporation (like, say, British Telecom's forebears). Bell Canada was once affiliated with AT&T, though spun off some decades ago. Several western provinces owned their own telcos. And some mom'n'pop independents still exist in parts of ON and QC.
Broadband's easier in Canada in part because there's less sprawl. It's a big country but there's a clear city/country break. ADSL doesn't work more than around 15,000 wire feet from the DSLAM (in the CO). Canada's population is largely clustered in cities and towns; large-lot-zoned suburbs (which create long loops) don't rule as they do in much of the USA. So average loop lengths are under control, and you can reach half of the country's population within reasonable range of a hundred COs or so.
I'm amazed that/. wastes its front page space pointing to junk like this. It's just some undergraduate practicing his P.R. flak skills by rewriting some tired Bell company propaganda, washed liberally with conservative ideology.
But where's the free market when the Bells (and ohter ILECs) were granted their monopolies, which prevented anybody else, until 1996, from putting in competing facilities at all?
But where's the free market for "innovation" when the "wireless" options cited by the college kid author are, indeed, virtually nonexistent, under a government spectrum policy (remember, the airwaves are REGULATED) that is now aimed at maximizing license auction revenues? That results in high cost-per-bit cellular clones and ever-more-concentrated commercial broadcast groups. Wireless unlicensed options are very limited, by design. Satellite is limited by both spectrum availablity and the speed of light -- "innovation" isn't going to change the value of c.
Where's the free market when an incumbent monopolist is allowed to use their monopoly power (the stuff John D Rockefeller was notorious for) to crush any competition? Where's the free market when the Bell companies use every trick in the book to prevent living up to their legal obligations?
There are, of course, two different views of "free market". One is that the government shouldn't interfere with monopolies. The other is that the government has to limit monopoly power in order to let market forces work. Clearly the undergrad author is in the former camp, the "let's bend over and let the monopolies rule us" camp.
Servers were never allowed out on cable
on
Broadband Crackdown
·
· Score: 5, Informative
The @Home customer agreements never allowed servers, particularly web servers. There's a valid technical reason, too: Cable bandwidth is asymmetric. There's typically a downstream pool of about 27 Mbps (depending on settings) shared among all users, while the upstream pool is more often in the 2 Mbps or less range. This comes about because upstream has to fit into the narrow patches of usable spectrum below 40 MHz, while downstream just fits among the TV channels between 50 and 750 MHz.
So stick a server out there, get Slashdotted (or even just get mildly popular), and the upstream bandwidth is wiped out for your whole neighborhood (technically, the area of your optical conversion node and CMTS channel). This is a big risk, so the cable companies don't take it. Instead, they do give you some free hosting space at their data centers.
VeriZontal has no such excuse -- ADSL has little upstream bandwidth (they typically provision only 90 kbps) but it's your very own, and they end up with a huge surplus of upstream bandwidth at the back of the DSLAM, where all of the traffic is aggregated. It's downstream that can congest easily. They're just being shmucks as usual. But if their customer agreement doesn't allow servers, then that's the deal -- commercial-grade DSL services allow servers.
The real problem they're addressing (even VZ) is Code Red II. Web servers that get infected will probe their own networks like crazy looking for others to infect. This creates congestion. So shutting off port 80 stops the worm. Crude but effective. See the recent LinuxPlanet column about Charter for how a cable company won't admit that its infected servers are causing huge congestion. The author suggests blocking port 80!
Don't you find this a bit scary? Big Brother wants to be able to delete books when they no longer suit his purpose. He wants to delete references to unpersons. He will say paper books are doubleplusungood. The Ministry of Truth will order all books to be in revocable format.
Slashdot Mirror
It's a little smaller than the floppy, but looks similar. The disk itself sits towards one end, so it's rectangular, not square, and thus narrower than a floppy.
I got the impression that they were marketing the technology to all comers, including computers. But of course it's within their power to blow it. I've often seen good technology get mismarketed. Indeed that's probably the usual fate....
I have seen Dataplay prototypes and played with an actual disk. It's cute. The actual disk is encapsulated in a sleeve that looks like a 3-inch floppy, only a bit smaller. Critical in its favor is the metal shutter door. You do not scratch these on the tabletop or in the car. They're proteted. That's good. That's what's wrong with CDs -- even after error correction, they can easily fail, because they're so exposed. Stinkin' jewel cases are far more cumbersome than the cardboard sleeves that old LPs came in, but DataPlay solves this with a hard shuttered enclosure. (Anybody remember CD caddies?) And it still fits into a shirt pocket or Walk-creature-sized machine.
Now the DRM is an option that, of course, all of the prerecorded music companies will invoke. And I can't comment on its crackability. But if you take a drive by itself, and some blanks, then you can ignore the DRM, because it's your own drive. And it fits into a laptop, or a desktop, and can replace a floppy, while being a whole lot nicer than Zip or LS-120.
It's a tiny bit smaller in capacity than CD-RW, and I wonder if that's its weakness. Just a tiny bit more diameter on the disk would give it a lot more area, and it could have been 1-2 GB and still smaller than a floppy.
It'll be interesting to see how they spin this, and how the public responds.
CAP is basically obsolete. It was the first ADSL technology, and has worked okay in some cases. But it has one big disadvantage. CAP causes crosstalk. It puts a lot of energy into some narrow spectra, and it leaks into adjacent wire pairs. The FCC is cracking down on this and CAP might not pass future standards. If a telco deploys one DSL line out of a hundred subscribers, CAP will work, but as the subscsriber density rises, CAP causes trouble
DMT spreads the signal differently, and thus has less crosstalk. It is thus the "standard". It doesn't always perform as well and is more latent though.
Fact: There is no government-granted cable monopoly. There are two cable companies passing my house; I use one of them (AT&T), and my next-door neighbor uses the other (RCN). Municipalities have been prohibited by federal law from granting exclusive franchises since 1992; even before then, there were few exclusives. Financially, it is generally a lousy investment to be the second cable company in a given place, which is why overbuilding is so rare. But lack of competition does not equate to government monopoly.
But in any case, if you want to regulate an ISP because it is so good, then what else do you regulate? If it MUST by law deliver me ten pitches a day to enlarge my penis size or get a mortgage from some crook, then what else must an ISP do by law? It would be a victory for lawyers (since ISPs would need a lot of them) but most ISPs would simply go out of business rather than be subject to regulation.
The Internet is not regulated as a telecom service. The FCC doesn't regulate ISPs, just the telecom services they buy. Nobody regulates mail servers. It's a free market, and it works. Now in a free market, you have competition. If your ISP uses MAPS and you don't like it, then you're free to go elsewhere. If your ISP is RBL'd, you're free to go elsewhere. There are lots of free e-mail services out there. See for instance http://www.emailaddresses.com/ . Now I wish my own "primary" e-mail provider, the one I ping many times a day, used one of these services, because I'm spammed to death and sick of it! If somebody couldn't get through, they almost certainly would find another way to reach me. Like I have a phone too, not to mention other e-mail addresses.
So given the fact that there is no anti-spam legislation, and negligible likelihood of effective anti-spam legislation within the next few years, then the free market approach (you know, the one the spammers cite to block anti-spam legislation) is to allow anti-spam filters at the ISPs. The ISPs will install them if it's good for business, and block spammers if being blackholed is bad for business.
Indeed one of the reasons that the Internet is not regulated as a "telecommunications service" is that it does not offer to provide transport of information "without change in form or content" -- an ISP may change things, of which blocking spam is one example. It would be quite a different story if a telecomm provider attempted to do the same thing -- their mission is to pass the bits unchanged, down there below layer 3.
And please don't tell me how easy it is to build an anti-spam filter on your private mail server. 99.9% of end users do no not run mail servers; ISPs, who have full-time bandwidth, run them for us.
Wireline routing protocols like OSPF are probably not well suited to wireless; there are too many assumptions that break. Radio links vary in quality and are less reliable than wireline; there's also mobility, as a given address might move around within the network. Or even if it stands still, its *logical* location -- which node it's getting best coverage from -- might change. Conventional TCP/IP doesn't handle that well.
One of the projects I've worked on in the distant past that directly addresses wireless networking is RSPF (radio SPF). This is a route-determination ("interior gateway") protocol designed for heterogeneous radio links. Originally just for ham radio AX.25, and very compact compared to OSPF or BGP, it could be adapted for other things like this. RSPF code, albeit experimental, is included in debian and SuSe distros. It adds a routing layer within the "subnet", so it doesn't have to look like a reliable fully-connected LAN.
Still, it's not clear how well a neighborhood network would work in practice. 802.11a, for instance, is limited to indoor use -- it's down in the 5.2 GHz low-power end of the UNII band, which is shared with satellites. The 5.7 GHz end allows outdoor use and more power, but cheap radios are still elusive. And it's sensitive to foliage fade. That's probably where NANs would make the most sense though. 802.11{b}, at 2.4 GHz, is cheap and has some passable range, especially the lower non-b speed. But 2.4 GHz is shared with microwave ovens, cordless phones, and other junque, which makes it tricky to use in urban areas.
I don't know waht Pair is doing. I do know that I can't run Linux Mandrake 8.1 on my VIA KT133A-based Athlon. Mostly it works, but mounting the SCSI-emulated CD-RW causes kernel panic. Various discussions on abUsenet a.o.l.m end up with "VIA chips are buggy". Funny, Windows has no trouble with them. But who knows what is happening there if Pair is running Linux.
Maybe the problem is Linux and a lack of time spent learning how to work with VIA chips.
While it's true that AMD CPUs are, uh, sensitive to cooling, I don't see that as a show-stopper. When you buy the parts to build your own Athlon system, as I did recently, you get plenty of warning to NOT TURN THIS ON WITHOUT A HEATSINK (yes, they shout, as they should).
Other CPUs are also very sensitive. What's rather surprising is how well Intel's P4 thermal shutdown works. I suspect AMD will get around to doing something similar. But in the meantime, I've attached a nice quiet (3800 RPM, not the 7200 RPM version) ThermoEngine to my Thunderbird, and it cruises at around 100 degrees F. Some newer/bigger heatsinks bolt to the motherboard, rather than clip on to the socket, which I suppose helps if you're really paranoid about its falling off. I use Motherboard Monitor to keep track of the temp via the Win98 system tray, and wish Linux distros would include similar capability out of the box (yeah, I know there's a way to build it in yourself...).
But then I do admit to using a 1 GHz Tbird rather than a faster one because I don't want that excess heat or power consumption.
Preface: I'm amazed at how poorly Slashdotters read the question. The post is about a 45 mile hop in rural Canada -- this is not the usual suburban nerd's home connection. No FCC, no RBOCs, and no, you can't just trench 45 miles of fiber optics for C$125k. (That's about what one mile of urban trenching costs, or maybe ten miles of rural Ditch-Witch burial.)
This type of application can, I'm sure even in Canada, use licensed point-to-point microwave. This allows lots higher power than 802.11 (forget the "b" which means higher speeds for even shorter distances). Typical rule-of-thumb is that frequencies under 10 GHz can go up to 30 miles (okay, say 45 km) on a single hop, if you can get line of sight. The site in question might need a repeater along the way. The terrain is all-important.
It probably is possible to get some microwave radios on that budget, though a repeater would possibly blow the limit. Harris, for instance, has a good selection, and a free program, Starlink, on their web site, which does path calculations for various radio - antenna combinations. (You can source the radios elsewhere, but Starlink is obviously geared to match Harris' own radios.) These would probably deliver 3 to 45 Mbps, depending on the radio in question. Industry Canada (which regulates spectrum matters) would probably be able to point you in the right direction for licensing and frequency/path coordination.
In New England, of which Maine is a part, "City" is defined by its type of government. A "Town" has a town meeting, either "open" (all voters can come) or "representative" (vote for your neighborhood's representatives, so TM is more like a legislature). I'm not sure about Maine, which spun off of Massachusetts in 1820, but in Massachusetts, a Town doesn't have a mayor; the Selectmen are the executive. A "City" in contrast has a mayor and council, no town meeting.
So we have Cities typically ranging from 10k population up, but Towns can be quite large too. Last year, several Towns in Massachusetts adopted City government. (But the largest, with over 60k pops, refuses.) Maine's cities are all small by Ontario standards, but they tend to be regional centers, and if they were Towns, they'd be too large for open meeting anyway.
Back to the original topic -- I think it'sa shame that Slashdotters overlay their aversion to Sprawl atop midcoast Maine, which really doesn't look a bit like Orange County.
Cringely is really off the wall this time. Yes, there are lots of failed providers of broadband, but there are others doing okay. Mostly small ones who don't have NASDAQ ticker symbols and big publicity.
@Home failed because it was a bad business. They had a nice gig doing the ISP stuff for the cable industry, but they got caught up in dotcom mania and bought the third-rate search engine Excite for a ridiculous amount. Excite never had a prayer of breaking even, so the whole thing was weighted down. Excite was also irrelevant to @Home's mission, which was to provide the cablecos with an ISP back end.
The data CLECs who tanked had bad business plans too. They mostly spent too much on collocation cages (needed before 1998 to access the loops) and they went into each others' markets, so a single telco CO would have half a dozen of them dividing the market among them. They also designed for a high breakeven, assuming that the others would have no market share. And they had big expense structures. So they tanked.
Cablecos do not need @Home any more. They can create in-house ISPs, as MediaOne did (ignore the @Home label, which is a borrowed trademark used because AT&T now owns them). They can and will also learn to work with ISPs, providing (without being forced) choice in ISP service. That does require some serious network reconfiguration, and since @Home had exclusive contracts with most of the cablecos into 2002, the cablecos aren't ready to open up. But with @Home finally being put out of its misery, the cablecos might finally recognize that they should work with other ISPs.
MediaOne's network has been called Continental Express, Highway One, Road Runner and finally @Home, but it is really something built locally and quite separate from @Home. The former-TCI parts of AT&T Broadband actually use the @Home plant, and are affected by @Home's problems. The former-MediaOne portions are not; they use the @Home trademark and little more, and are still taking orders and installing. And yeah, it works very nicely.
I expect the other @Home cablecos to have a fix in place very, very fast. Either AT&T will "fix" @Home or something else will be done.
I have used Mozilla 0.9.3 and it's mostly pretty nice... but there's one show stopper that I don't think is being addressed.
My office uses Lotus Notes for mail (ugh!). Its Domino server lets me read and send mail from a browser, so I don't have to pollute my home PC with that godawful "client". Of course I have to get through the firewall to get there, but at least on Windows they gave me the necessary IPsec client (Nortel) that works with the SecureID card. (That's another reason I can't depend on Linux so far, but that subject it off topic.)
The Domino mail client uses Java to provide useful menu items like "next message" and "reply". This works on Netscape 4.7x and on Internet Exploder, but not on Mozilla. Obviously there's something different about their Javas. Maybe Domino uses an older version and the one in Mozilla isn't backwards compatible?
So until Mozilla can talk to Domino, I'll still need the unstable but well-understood old Netscape client. Suggestions for fixing this are of course welcome. (Flames about still using Windows for anything are not. My current problems with Mandrake 8.1 are off topic too.)
Methane doesn't stink. It's basically odorless.
Natural gas doesn't smell by itself. Distributors add methyl mercaptan to it in order to make it smell, so that leaks can be found. (Walk up and down the street in front of my house any day of the year and you'll smell it coming from the rather porous old Boston Gas/Keyspan pipes. They make repairs when their leak detector shows the concentration getting scary.) Mercaptan was chosen because it, well, has a strong and distinctive smell. Acetylene smell similar but is itself explosive.
This practice began after a very unfortunate incident in the 1930s. The public school in London, Texas had been heated by gas that was being, uh, skimmed off of a pipeline passing from a nearby well. The connection wasn't exactly professional. A leak sprang, and gas accumulated in the basement, reaching serious concentrations without being noticed. It blew the school sky-high, killing about 200 people, including most of the children, largely wiping the town off the map. (The town, near Tyler, was renamed New London; it now has about 900 inhabitants.)
People nowadays appreciate methane's properties a bit better. A little cartridge to power a fuel cell should not be a problem.
First off, and speaking as a regular Slashdot karma whore who loves bashing Bill, I think it's cruel to Microsoft to equate VZ and SBC to them! But anyway...
These ILECs do not have a legal monopoly on anything any more. You and I can string wire under or over the street next to theirs. It's just bad business to try. Several companies such as Worldcom (ex-MFS, Brooks, MCI), AT&T (ex-TCG), Metromedia, Hyperion, XO and ELI dumped a few billion dollars burying fiber under the streets in major business districts. Often right next to each other, guaranteeing that nobody would have enough business to be profitable. That's one reason why the telecom sector is so weak now. Local competition was opened nationwide in 1996 and there was so much cheap capital around at the time that a lot was squandered.
There is still a growing CLEC sector, but the failures of Covad (operating in Chapter 11), Rhythms, Northpoint, Vitts, BBO and others have demonstrated how hard it is to operate as a pure-play DSL provider. Many of the surviving CLECs, like Allegiance and McLeod USA, combine their own facilities with leased ILEC wire and do both voice and data. Rhythms leased the wire but only got the data revenue from it.
CLEC survival depends upon the FCC enforcing rules that require the Bells and other ILECs to interconnect and lease critical facilities to them. That has been rather lacking as of late. It also depends upon Congress *not* changing the rules and, for instance, passing the Dingell-Tauzin bill, which would essentially shut CLECs out of the DSL business and make it an unregulated ILEC monopoly.
This isn't entirely new. There were "microsats" flown in the 1980s, some sponsored by the Amateur Satellite Corp. (AMSAT), and some university sats like Webersat (from Utah).
With today's smaller and more powerful chips, of course, it's a lot easier to do more in a small package.
The "telco geek" was right when he noted that there'd be too much NEXT/FEXT (crosstalk). It only works because all of the signals are loud and soft together. The CO transmits at a higher frequency than the modem (frequency split) so the CO-end signals are weak at the modem and, more importantly, the modem-end signals are all weak at the CO. Run it backwards and the CO-end modem would clobber the signals coming in from remotes. The cable on the street was designed for voiceband, and has too much crosstalk at the 50 kHz-1 MHz band used for ADSL.
BUT there are alterantive technologies. SDSL is popular for business; that's Covad's big item. It was designed for symmetry, using echo cancellation on the same frequencies rather than frequency splitting. Of course it's more expensive -- SDSL cannot share the wire with analog phones, so you need a dedicated loop. Plus it is sold to businesses, which tend to use it more intensively than consumers (average bytes/month), so the ISP price is set accordingly.
A technology I really like is Paradyne's MVL. It uses "ping-pong" signaling on the wire, avoiding voiceband frequencies so it can share the loop like ADSL but not SDSL. MVL (also used, with modification, in AGCS' SuperLine) transmits a frame in one direction at a time. If there's nothing to send, it sends a minimum-length burst. So the bandwidth is adaptively directional, allowing you to have a client or server without takling to your provider.... It avoids crosstalk by staying below 120 kHz. Its peak speed is 1 MHz, depending on loop length, but that's still adequate for most users. Oh yeah, unlike ADSL, MVL will work on crufty loops with bridge taps, over 25,000 feet (at degraded speed), just so long as it's unloaded (no series inductors).
Not many telcos use this stuff, since it's not "standard", but a few do; I know a CLEC startup that does. (I've no interest in Paradyne; I just like the technology.) Rumor has it that Rhythms had some, but alas that will probably be confirmed by their pending asset auction....
Out of the box, Linux usually looks lousy. At least it did for my last few installs, including Mandrake 7.1 and 7.2 and RedHat 8.0. The font rendering was plug-ugly, compared to Windoze. Indeed it was barely readable, especially in Netscape.
Now the main problem, of course, was that the profoundly defective AbiSuite fonts were installed in the font path. (Why do Linux distributions still do this?) Thanks to Google Groups, I found out about it and could remove the offending line. After that, the fonts were merely mediocre, maybe as good as Windows 3.0, though it's hard to compare the monitors of those days to now.
After a session of Linux, rebooting to Windows is a treat to the eyes. Not that Windows is better than Linux for everything, but XFree86 seems to have terribly primitive font rendering, while Windows pays close attention to appearances. I do typically insert Windows fonts into Linux, which are better than the usual X fonts, but Linux has still not got the best font rendering engines. It makes a real difference in readability when looking at the small fonts some web sites use.
The X Window System was a clever invention for its day, the early 1980s and Project Athena, where the goal was a "1-1-1" X server (display terminal) system (1 M byte RAM, 1 M pixel screen, 1 MHz CPU). But Linux would benefit from a modern replacement. (What ever happened to Hungry's "Y"?)
Depends on your shop. Where I work, Powerpoint is *everything*. It's a consulting firm which likes to deliver its 6-figure-plus reports to clients in Powerpoint form. They have people who are really good at some Powerpoint design formats, and can get quite a bit onto a page if they want to. Personally, I prefer the linear style of Word with a (very) few pictures, but I'm the exception there.
Outlook, on the other hand, isn't even installed. They use Notes for mail (which is execrable, but safe from viruses) and groupware functions (where it has potential).
For me to jettison 'doze and use Linux would require real two-way filters into and out of PPT, Word and Excel. I haven't found AbiWord to be very good at it. Gnumeric isn't complete either. And even different versions of Powerpoint (which really sucks) aren't compatible with each other; at work, it's all Office 97, and no talk about "upgrades". I suspect that's the most common version nowadays; we exchange a lot of stuff with clients using Office 97 formats.
No doubt too late to be noticed...
... but not everybody in rural areas is poor or needs a big subsidy. Rural telco subsidies go to wealthy people's ski chalets in the Rockies, to beach cottages, and to The Donald's various retreats.
And at the end of the day, local telephone rates in these rural areas are actually lower than in town, even though the rural telco is only getting, say, 20-30% of its revenue from the local ratepayers' monthly bills. So the subsidies can be reduced, and targeted more carefully (e.g., more "low income support" in lieu of "high cost support"), while still preserving "universal service".
Unlicensed networks like Ricochet aren't allowed enough power/range to be profitable -- that's the idea! Metricom was caught between having enough base stations to meet needs and having enough customers to finance them. The numbers never converged. Unlicensed wireless works best in rural areas, especially flat ones.
The money for sky-borne relays has dried up with the rest of the industry. Don't hold you breath for those airships. Even if the FCC were to grant spectrum to them.
And it has been illegal for a municipality to grant exclusivity to a cable franchise since 1992. States grant telco franchises, which have been non-exclusive since 1996. There are overbuilders, but the economics aren't great. Check out RCN's quarterlies. From a financial perspective, access to existing ILEC loops is the most viable competitive solution.
I agree that rural areas get too much of a subsidy. But even then, the subsidy mechanism should be made available to competitive providers. That's *theoretically* possible but CLECs have a very hard time getting it. Western Wireless and McLeodUSA may have qualified.
Well, no, your history is wrong. Perhaps you're so young that you're confusing Canada with Cuba?
Canadian telephone service wasn't a Crown Corporation (like, say, British Telecom's forebears). Bell Canada was once affiliated with AT&T, though spun off some decades ago. Several western provinces owned their own telcos. And some mom'n'pop independents still exist in parts of ON and QC.
Broadband's easier in Canada in part because there's less sprawl. It's a big country but there's a clear city/country break. ADSL doesn't work more than around 15,000 wire feet from the DSLAM (in the CO). Canada's population is largely clustered in cities and towns; large-lot-zoned suburbs (which create long loops) don't rule as they do in much of the USA. So average loop lengths are under control, and you can reach half of the country's population within reasonable range of a hundred COs or so.
Monopolies don't help. Unregulated monopolies really don't help!
I'm amazed that /. wastes its front page space pointing to junk like this. It's just some undergraduate practicing his P.R. flak skills by rewriting some tired Bell company propaganda, washed liberally with conservative ideology.
But where's the free market when the Bells (and ohter ILECs) were granted their monopolies, which prevented anybody else, until 1996, from putting in competing facilities at all?
But where's the free market for "innovation" when the "wireless" options cited by the college kid author are, indeed, virtually nonexistent, under a government spectrum policy (remember, the airwaves are REGULATED) that is now aimed at maximizing license auction revenues? That results in high cost-per-bit cellular clones and ever-more-concentrated commercial broadcast groups. Wireless unlicensed options are very limited, by design. Satellite is limited by both spectrum availablity and the speed of light -- "innovation" isn't going to change the value of c.
Where's the free market when an incumbent monopolist is allowed to use their monopoly power (the stuff John D Rockefeller was notorious for) to crush any competition? Where's the free market when the Bell companies use every trick in the book to prevent living up to their legal obligations?
There are, of course, two different views of "free market". One is that the government shouldn't interfere with monopolies. The other is that the government has to limit monopoly power in order to let market forces work. Clearly the undergrad author is in the former camp, the "let's bend over and let the monopolies rule us" camp.
The @Home customer agreements never allowed servers, particularly web servers. There's a valid technical reason, too: Cable bandwidth is asymmetric. There's typically a downstream pool of about 27 Mbps (depending on settings) shared among all users, while the upstream pool is more often in the 2 Mbps or less range. This comes about because upstream has to fit into the narrow patches of usable spectrum below 40 MHz, while downstream just fits among the TV channels between 50 and 750 MHz.
So stick a server out there, get Slashdotted (or even just get mildly popular), and the upstream bandwidth is wiped out for your whole neighborhood (technically, the area of your optical conversion node and CMTS channel). This is a big risk, so the cable companies don't take it. Instead, they do give you some free hosting space at their data centers.
VeriZontal has no such excuse -- ADSL has little upstream bandwidth (they typically provision only 90 kbps) but it's your very own, and they end up with a huge surplus of upstream bandwidth at the back of the DSLAM, where all of the traffic is aggregated. It's downstream that can congest easily. They're just being shmucks as usual. But if their customer agreement doesn't allow servers, then that's the deal -- commercial-grade DSL services allow servers.
The real problem they're addressing (even VZ) is Code Red II. Web servers that get infected will probe their own networks like crazy looking for others to infect. This creates congestion. So shutting off port 80 stops the worm. Crude but effective. See the recent LinuxPlanet column about Charter for how a cable company won't admit that its infected servers are causing huge congestion. The author suggests blocking port 80!
Don't you find this a bit scary? Big Brother wants to be able to delete books when they no longer suit his purpose. He wants to delete references to unpersons. He will say paper books are doubleplusungood. The Ministry of Truth will order all books to be in revocable format.