SBC trots this out whenever they want something from regulatory authorities. But they don't actually install it.
Read this 1999 article about SBC's 'Project Pronto'.
" According to SBC, when the expanded deployment program is completed [in three years] customers will be able to receive minimum downstream connection speedsof 1.5 megabits per second, with more than 60 percent eligible to receive guaranteed speeds of 6 megabits a second." Right.
SBC's new "Project Lightspeed" isn't about the Internet at all. It's just cable TV, implemented using Windows Media 9 over DSL using Scientific-Atlanta set-top boxes. The system doesn't use the Internet at all. It has its own infrastructure, which is a Microsoft-implemented multicast implementation.
Why Addison-Wesley rejected this
on
The New C Standard
·
· Score: 4, Interesting
It's clear why Addison-Wesley rejected this. The author is a terrible writer. He has a fondness for run-on sentences, fails to use commas appropriately, and makes several grammatical mistakes per page.
I'm up to page 132, and so far, it's an introductory cognitive psychology text. A bad one.
There are 1616 pages of this drivel. Even for someone interested in programming language design, this is a painful read.
There's room for a good book in this area, but this isn't it. A more useful approach might be to start from Amit Yoran's statement that "About 95% of software bugs come from 19 common, well-understood" programming mistakes", and evaluate language designs against that.
And almost everybody gets this wrong, including the people asking the question, because they have no idea how stuff is made.
The real reason manhole covers are round reflects late 19th century manufacturing technology. In the late 19th century, casting worked fine, but the only power tools were lathes, planers, and steam hammers. Milling machines and welders were in the future.
Given that toolset, a round manhole cover is an easy thing to make - cast, chuck in lathe, finish machine in one setup. A manhole cover ring, which needs a little finish machining to clean up the inside of the ring, is also straightforward.
Simple, cheap, and suitable for volume production.
Making a rectangular plate with 1890s technology is harder than making a round one. It would probably require four passes through a power planer, which is a more expensive machine than a lathe.
Making a rectangular manhole frame with that toolset is really tough. You can't use a lathe to do the finish machining. It's tough to get a planer into the inside of a rectangle. You'd need a specialized planer with a long reach, and it would take at least four setups to do the job, probably eight to get into each corner from both directions.
Today, you'd cut four straight sections and weld the parts together, which is how rectangular frames are made today. But that option didn't exist in 1890.
Take a look at a steam locomotive from that era. All big metal parts consist of cast surfaces, flat machined surfaces, circular machined surface.
Anything else was really difficult to make.
Re:ICMP flaw #1 on Linux: it's in the kernel
on
Examining ICMP Flaws
·
· Score: 5, Informative
Sigh.
As someone who once implemented ICMP (in 1982, before BSD, even), I should say something.
First, ICMP is a layer 3 protocol, like TCP and UDP.
ICMP is IP protocol #1; TCP is #6 and UDP is #17.
Second, it's quite feasible to put ICMP in user space. I'm writing this on a QNX system where it's in user space. My 1982 implementation was also in user space, as part of 3COM's UNET. Linux doesn't do it that way, but it's not fundamental that ICMP must be in the kernel. It needs to have a mechanism to pass messages to the other protocols, but that's a local message passing problem. But I'm not going to rehash the ever-growing monolithic kernel issue here.
Third, we knew about many of those vulnerabilities back in the 1980s, but weren't as concerned about them because the Internet was a DoD/NSF operation.
Destination Unreachable and Source Quench messages used to be taken more seriously than they are now.
Destination Unreachable told you where the network was down, and Source Quench told you where it was congested, basic network management info back then. Today, nobody does network management that way and many TCP stacks don't do much, if anything, with ICMP information. I used to encourage the use of Source Quench for congestion management (see my RFC on this, from 1984), but it's far less appropriate today. Back then, we were concerned about packet loss through transmission errors, a frequent occurence with leased-line synchronous modems. So, when a packet was lost, the question was whether you should retransmit rapidly (appropriate for an error) or slowly (appropriate for congestion). Source Quench could disambiguate that situation. Today, it's assumed that packets are lost almost entirely through congestion, since the lower levels are of much better quality than they used to be.
In the early days of Go, I used to see people carrying the Go prototype around Palo Alto. It was a book-sized box with a rubber cover for the screen, using an Intel 386 processor. With pen input. Remember that laptops barely existed then. Way ahead of its time.
Go made two big mistakes. One was negotiating with Microsoft, and the other was partnering with AT&T.
Back then, AT&T was trying to figure out what to do after deregulation. There was an AT&T PC (running UNIX, no less), AT&T minicomputers (running on 48VDC, just like telco gear), and for a brief period, AT&T retail computer stores. AT&T insisted that Go port their system over to some wierd AT&T CPU, which they did. Then they had some designer firm design a cool-looking but inconvenient case for the thing, with plastic plug-on modules on the side.
The prototype was better than the "production version".
The notorious Black Box Hosting ("Our offshore bullet proof web hosting plans allow bulk email hosting, spam friendly web hosting and bulletproof host.") is still up.
They claim to be in "some province in the highlands of China", and their netblock (219.148.32.234) comes up as "CHINANET HEBEI PROVINCE NETWORK".
There's no indication on the spammerforums of any fears about China-based hosting yet.
I'm not saying that I think that "bullet makers" should be held responsible for the actions of a select few of their customers, but I do think that there is a certain amount of discretion that companies riding the razor's edge ought to employ.
Or, in the words of George Bernard Shaw:
What on earth is the true faith of an Armorer?
UNDERSHAFT. To give arms to all men who offer an honest price for
them, without respect of persons or principles: to aristocrat and
republican, to Nihilist and Tsar, to Capitalist and Socialist, to
Protestant and Catholic, to burglar and policeman, to black man
white man and yellow man, to all sorts and conditions, all
nationalities, all faiths, all follies, all causes and all
crimes.
Note some key, well-identified problems that haven't been solved.
Energy production. None of the great ideas of the last fifty years have panned out. Fission is more dangerous than expected. Fusion gets further away every year. Solar cells still cost too much and have lousy efficiency. Oil shale remains marginal and messy. And we're running out of oil.
That's the biggest problem out there, and there's nothing in the pipeline that looks really promising.
Space travel Space flight with chemical rockets just barely works. So much weight reduction is necessary that rockets are too fragile to be reliable. Chemical fuels just don't have the energy density to make it really work. This was known in 1950, yet we still don't have nuclear rockets that work.
Artificial intelligence We're stuck. Nobody has a clue how to do it, really. Half a century of banging on the problem, and we basically have the ideas of the 1960s with more CPU power behind them. We have enough CPU power now that we should be able to do a low-end mammal brain, at least. And we can't. It's embarassing.
I saw the moron gram too. Actually, BNC on audio gear is rare, but it does show up in broadcast equipment and ham gear. BNC audio interconnects were more common 20 years ago than they are now. Consumers have now been educated that BNC = video and RCA = audio, so if you violate that convention, you get phone support calls.
There's a tendency in the RF world to run everything through BNC connectors, whether you need to or not. Signal generators and scopes usually come with BNC connectors, so if you have those, you tend to have lots of BNC-BNC cables around the bench. Plus the little drawer of T-connectors, angle connectors, and adapters.
Hence its popularity in the ham, broadcast, and scientific instrument worlds.
The main problem with RCA connectors is that they bend and become loose as they wear out. That's why they're avoided in PA gear. XLR connectors self-align better and latch into place.
Actually, I do servomotor control, which has most of the problems of audio but with bigger currents. Keeping the huge chopped motor currents from inducing noise into nearby analog sensors is a major headache. But with extra capacitors and inductors, it's a solveable problem.
In any case, without a scope you can't do anything but guess.
The ARRL Handbook is a good source for info about power supply filtering.
BNC on audio gear is rare, but not unheard of.
There's a trend away from it, though, because consumers are used to BNC for video and RCA for audio.
If you have balanced output, XLRs are appropriate. But most consumer-grade (and even most audiophile gear) doesn't have balanced outputs. And, actually, BNC connectors have better frequency response; they're coaxial all the way through, and nearly flat to 50MHz at least. If you have access to a time-domain reflectometer, you can see the difference. Not that it really matters for audio.
For a good laugh, see these RCA cables. Palladium wires with solid silver RCA plugs. "You will enjoy a pitch black background, deep, yet lightning fast bass, smooth midrange, and most importantly, seemingly limitless top end extension. Though not at all bright or fatiguing in any manner, Pure Palladium's sparkling highs allow for the presence of the often coveted sense of air as well as glorious imaging and soundstage. This interconnect possesses the ability to untangle even the most complex pieces of music." $1,550.00 for a pair of 1.5 meter cables.
Any common video cable with BNC connectors will do better than that.
That article makes me embarassed to be an IEEE member.
Those "special $50 capacitors" sound like a rip-off.
There are grades of capacitors, but no small-value cap costs $50 from Digi-Key.
Another amusing point is the mania for expensive RCA jacks in the audiophile world. Any BNC connector, which is what you see on pro audio gear (and most video gear), has better high-frequency response than the fanciest RCA jack. And BNC jacks latch, so they don't come loose. Yet the audiophile nuts are still equipping their overpriced amps with RCA jacks.
Really, if you're going to do stuff like this, the first step is to put a scope on the power supply outputs and watch them under load. If you see noise or changes under load, it's time to do power supply work. You may need to juggle capacitors or add inductors, like ferrite beads. It's quite common to see some digital noise spikes getting into the power to the analog circuitry, and you've got to get rid of that. But there's nothing mysterious about how to do it.
Without measurement tools, though, you don't get anywhere.
Any entity (person, corporation, or partnership) should be able to get one domain name cheaply. After that, ICANN should impose a charge of about $250 per year, to kill off those "link farm" operations.
I can't find the exact quote offhand, but Dijkstra said something like a necessary precondition for being a good computer scientist is absolute command over your native language.
The problems of business administration in general and data base management in particular are much too difficult for people that think in IBMerese, compounded with sloppy English.
About the use of language: it is impossible to sharpen a pencil with a blunt axe. It is equally vain to try to do it with ten blunt axes instead.
Besides a mathematical inclination, an exceptionally good mastery of one's native tongue is the most vital asset of a competent programmer.
Clark said he would like to see two things addressed in any replacement for the current internet. The first is a coherent security architecture. The second is a healthy economic infrastructure for network service providers, who will need a bigger piece of the pie in the new internet than the one they are getting now if they are going to help pay for building it.
This guy must be getting support from a telco.
Telecommunications providers
hate the Internet. Not only is the Internet too cheap, it's not set up for detailed billing. The US Internet backbone cost about $1bn to build, and costs about $100 million per year to run. For something that handles over 100 million users, that's nothing. All the intelligence is in the end nodes, so telcos don't get to add "value added services" for which they can overcharge.
What telcos want is an environment they control, like cell phones. With charges for everything from ring tones to SMS messages.
That's what Clark is talking about here.
The telcos tried this idea back in the 1980s, and it was called TP4, or "ISO 8073 COTP Connection-Oriented Transport Protocol - X.224" X.224 is very much like TCP, but without the adaptive retransmit machinery to work well over unreliable links. You're supposed to run X.224 over a reasonably reliable virtual circuit provided by a telco. For which you pay by the packet, like X.25 or ISDN. Bad idea. Windows NT4 actually had support for X.224, and some older Cisco routers understand it, but it's dead.
This is not a place we, as users, want to go.
Sun spent $20 million to promote a free product.
on
James Gosling on Java
·
· Score: 0
Of course Java succeeded. Sun gave it away and spent $20 million promoting it at launch time.
It's a good language. Not a great language, but an good one. Which is good enough.
The Java libraries range from mediocre to terrible, but that's a separate issue.
That's exactly what Microsoft tells the huge number of business users still running Windows 2000.
It's not a troll; it's reality.
Microsoft keeps adding stuff to Windows that allows external programs to initiate activity from the network. Windows Messenger Service. Universal Plug and Play. Windows Update. Active Management. AutoPlay. Now, RSS. And they consistently have them turned on by default. This guarantees a large supply of future security holes.
In ten years, they haven't even been able to secure Outlook.
Department of Homeland Security appointments in the computer security area are disappointing. Amit Yoran, head of the "National Cyber Security Division" at DHS, quit in disgust. He was replaced by a lawyer and TV producer. The "National Cyber Security Division" seems to have been pushed down to a lower level of the DHS bureauracy.
The path planner goes slower and generates paths that are initially ambiguous when faced with multiple alternatives. That's no surprise. I'm working on the steering control program for our DARPA Grand Challenge vehicle, and it does that, too. Doesn't mean it's not "digital".
This is a good reason to keep Sun from putting Java components into OpenOffice. As eWeek puts it, "The Java code in the newest OpenOffice, however, does not compile well with open-source Java compilers like the GCJ".
That gives Sun leverage to restrict the free use of OpenOffice in future.
Now that we're finally out from under Microsoft Office, we don't want to be in a similar position with Sun. Users don't want Sun to be in a position to change the rules at a future time. That's the whole point of open source.
Actually, Usenet is doing quite well. The spam battle has been won; there's very little spam in the technical groups. Serious workers in difficult fields are on there. Check out, say, "comp.games.development.programming.algorithms", where the people who write physics engines discuss how to do it. Or "comp.std.c++.moderated", where proposed changes to C++ are discussed. Usenet has far lower advertising content than the Web, where, today, "content" seems to be a little box in the middle of the page, surrounded by blinking ads.
The decision is worth reading. A key point that influenced the court was that Streamcast was profiting from the infringement by putting in adware and spyware and deriving revenue thereby. And Grokster was advertising itself as a "Napster replacement":
"Evidence indicates that
ih[i]t was always [StreamCast(TM)s] intent to use [its OpenNap
network] to be able to capture email addresses of [its]
initial target market so that [it] could promote [its]
StreamCast Morpheus interface to them."
"Grokster launched its own OpenNap system called Swaptor and inserted digital codes into its Web site so that
computer users using Web search engines to look for
'Napster' or 'free filesharing' would be directed to the
Grokster Web site, where they could download the Grokster software."
This evidence of intent to profit from infringement seems to be what lost the case for them.
Animation studio edifice complex
on
Lucas's New HQ
·
· Score: 1
Lucas had to catch up. Disney built a complex in Burbank in the 1990s with columns modelled after the Seven Dwarfs and other theme-park features. (This was followed by a layoff, a pay cut, and the closure of Disney Animation in Orlando.)
Pixar has a modern building in Marin with a really big atrium. Dreamworks/PDI moved into a huge glass and steel complex at the edge of the Bay in Redwood City originally built for Excite@Home.
Lucasfilm/ILM had some boring industrial buildings in San Raphael. So this is just catchup.
Slashdot Mirror
Read this 1999 article about SBC's 'Project Pronto'. " According to SBC, when the expanded deployment program is completed [in three years] customers will be able to receive minimum downstream connection speedsof 1.5 megabits per second, with more than 60 percent eligible to receive guaranteed speeds of 6 megabits a second." Right.
SBC's new "Project Lightspeed" isn't about the Internet at all. It's just cable TV, implemented using Windows Media 9 over DSL using Scientific-Atlanta set-top boxes. The system doesn't use the Internet at all. It has its own infrastructure, which is a Microsoft-implemented multicast implementation.
It's not about Internet access at all. All you can get is what they want to send you. Lightspeed will block access to Internet video.
I'm up to page 132, and so far, it's an introductory cognitive psychology text. A bad one.
There are 1616 pages of this drivel. Even for someone interested in programming language design, this is a painful read.
There's room for a good book in this area, but this isn't it. A more useful approach might be to start from Amit Yoran's statement that "About 95% of software bugs come from 19 common, well-understood" programming mistakes", and evaluate language designs against that.
Jacob Nielson's cartoon says this best.
The real reason manhole covers are round reflects late 19th century manufacturing technology. In the late 19th century, casting worked fine, but the only power tools were lathes, planers, and steam hammers. Milling machines and welders were in the future.
Given that toolset, a round manhole cover is an easy thing to make - cast, chuck in lathe, finish machine in one setup. A manhole cover ring, which needs a little finish machining to clean up the inside of the ring, is also straightforward. Simple, cheap, and suitable for volume production.
Making a rectangular plate with 1890s technology is harder than making a round one. It would probably require four passes through a power planer, which is a more expensive machine than a lathe. Making a rectangular manhole frame with that toolset is really tough. You can't use a lathe to do the finish machining. It's tough to get a planer into the inside of a rectangle. You'd need a specialized planer with a long reach, and it would take at least four setups to do the job, probably eight to get into each corner from both directions. Today, you'd cut four straight sections and weld the parts together, which is how rectangular frames are made today. But that option didn't exist in 1890.
Take a look at a steam locomotive from that era. All big metal parts consist of cast surfaces, flat machined surfaces, circular machined surface. Anything else was really difficult to make.
As someone who once implemented ICMP (in 1982, before BSD, even), I should say something.
First, ICMP is a layer 3 protocol, like TCP and UDP. ICMP is IP protocol #1; TCP is #6 and UDP is #17.
Second, it's quite feasible to put ICMP in user space. I'm writing this on a QNX system where it's in user space. My 1982 implementation was also in user space, as part of 3COM's UNET. Linux doesn't do it that way, but it's not fundamental that ICMP must be in the kernel. It needs to have a mechanism to pass messages to the other protocols, but that's a local message passing problem. But I'm not going to rehash the ever-growing monolithic kernel issue here.
Third, we knew about many of those vulnerabilities back in the 1980s, but weren't as concerned about them because the Internet was a DoD/NSF operation. Destination Unreachable and Source Quench messages used to be taken more seriously than they are now. Destination Unreachable told you where the network was down, and Source Quench told you where it was congested, basic network management info back then. Today, nobody does network management that way and many TCP stacks don't do much, if anything, with ICMP information. I used to encourage the use of Source Quench for congestion management (see my RFC on this, from 1984), but it's far less appropriate today. Back then, we were concerned about packet loss through transmission errors, a frequent occurence with leased-line synchronous modems. So, when a packet was lost, the question was whether you should retransmit rapidly (appropriate for an error) or slowly (appropriate for congestion). Source Quench could disambiguate that situation. Today, it's assumed that packets are lost almost entirely through congestion, since the lower levels are of much better quality than they used to be.
Go made two big mistakes. One was negotiating with Microsoft, and the other was partnering with AT&T. Back then, AT&T was trying to figure out what to do after deregulation. There was an AT&T PC (running UNIX, no less), AT&T minicomputers (running on 48VDC, just like telco gear), and for a brief period, AT&T retail computer stores. AT&T insisted that Go port their system over to some wierd AT&T CPU, which they did. Then they had some designer firm design a cool-looking but inconvenient case for the thing, with plastic plug-on modules on the side.
The prototype was better than the "production version".
Proxy Mailing Servers -- 1 server for $499/month
Allowed Software
Proxy mailing software like the following are allowed to be used:
There's no indication on the spammer forums of any fears about China-based hosting yet.
So, thus far, any crackdown is vaporware.
Or, in the words of George Bernard Shaw:
What on earth is the true faith of an Armorer?
UNDERSHAFT. To give arms to all men who offer an honest price for them, without respect of persons or principles: to aristocrat and republican, to Nihilist and Tsar, to Capitalist and Socialist, to Protestant and Catholic, to burglar and policeman, to black man white man and yellow man, to all sorts and conditions, all nationalities, all faiths, all follies, all causes and all crimes.
The hard problems are not being cracked.
There's a tendency in the RF world to run everything through BNC connectors, whether you need to or not. Signal generators and scopes usually come with BNC connectors, so if you have those, you tend to have lots of BNC-BNC cables around the bench. Plus the little drawer of T-connectors, angle connectors, and adapters. Hence its popularity in the ham, broadcast, and scientific instrument worlds.
The main problem with RCA connectors is that they bend and become loose as they wear out. That's why they're avoided in PA gear. XLR connectors self-align better and latch into place.
Actually, I do servomotor control, which has most of the problems of audio but with bigger currents. Keeping the huge chopped motor currents from inducing noise into nearby analog sensors is a major headache. But with extra capacitors and inductors, it's a solveable problem.
In any case, without a scope you can't do anything but guess.
The ARRL Handbook is a good source for info about power supply filtering.
If you have balanced output, XLRs are appropriate. But most consumer-grade (and even most audiophile gear) doesn't have balanced outputs. And, actually, BNC connectors have better frequency response; they're coaxial all the way through, and nearly flat to 50MHz at least. If you have access to a time-domain reflectometer, you can see the difference. Not that it really matters for audio.
For a good laugh, see these RCA cables. Palladium wires with solid silver RCA plugs. "You will enjoy a pitch black background, deep, yet lightning fast bass, smooth midrange, and most importantly, seemingly limitless top end extension. Though not at all bright or fatiguing in any manner, Pure Palladium's sparkling highs allow for the presence of the often coveted sense of air as well as glorious imaging and soundstage. This interconnect possesses the ability to untangle even the most complex pieces of music." $1,550.00 for a pair of 1.5 meter cables.
Any common video cable with BNC connectors will do better than that.
That article makes me embarassed to be an IEEE member.
Those "special $50 capacitors" sound like a rip-off. There are grades of capacitors, but no small-value cap costs $50 from Digi-Key.
Another amusing point is the mania for expensive RCA jacks in the audiophile world. Any BNC connector, which is what you see on pro audio gear (and most video gear), has better high-frequency response than the fanciest RCA jack. And BNC jacks latch, so they don't come loose. Yet the audiophile nuts are still equipping their overpriced amps with RCA jacks.
Really, if you're going to do stuff like this, the first step is to put a scope on the power supply outputs and watch them under load. If you see noise or changes under load, it's time to do power supply work. You may need to juggle capacitors or add inductors, like ferrite beads. It's quite common to see some digital noise spikes getting into the power to the analog circuitry, and you've got to get rid of that. But there's nothing mysterious about how to do it. Without measurement tools, though, you don't get anywhere.
Any entity (person, corporation, or partnership) should be able to get one domain name cheaply. After that, ICANN should impose a charge of about $250 per year, to kill off those "link farm" operations.
That's from "Programming as a discipline of mathematical nature", in which Dikjstra writes "A programmer must be able to express himself extremely well, both in a natural language and in the formal systems."
In a ranting mood, Djykstra once wrote these one-liners:
Then came PowerPoint.
This guy must be getting support from a telco.
Telecommunications providers hate the Internet. Not only is the Internet too cheap, it's not set up for detailed billing. The US Internet backbone cost about $1bn to build, and costs about $100 million per year to run. For something that handles over 100 million users, that's nothing. All the intelligence is in the end nodes, so telcos don't get to add "value added services" for which they can overcharge.
What telcos want is an environment they control, like cell phones. With charges for everything from ring tones to SMS messages. That's what Clark is talking about here.
The telcos tried this idea back in the 1980s, and it was called TP4, or "ISO 8073 COTP Connection-Oriented Transport Protocol - X.224" X.224 is very much like TCP, but without the adaptive retransmit machinery to work well over unreliable links. You're supposed to run X.224 over a reasonably reliable virtual circuit provided by a telco. For which you pay by the packet, like X.25 or ISDN. Bad idea. Windows NT4 actually had support for X.224, and some older Cisco routers understand it, but it's dead.
This is not a place we, as users, want to go.
It's a good language. Not a great language, but an good one. Which is good enough.
The Java libraries range from mediocre to terrible, but that's a separate issue.
Microsoft keeps adding stuff to Windows that allows external programs to initiate activity from the network. Windows Messenger Service. Universal Plug and Play. Windows Update. Active Management. AutoPlay. Now, RSS. And they consistently have them turned on by default. This guarantees a large supply of future security holes.
In ten years, they haven't even been able to secure Outlook.
Department of Homeland Security appointments in the computer security area are disappointing. Amit Yoran, head of the "National Cyber Security Division" at DHS, quit in disgust. He was replaced by a lawyer and TV producer. The "National Cyber Security Division" seems to have been pushed down to a lower level of the DHS bureauracy.
The path planner goes slower and generates paths that are initially ambiguous when faced with multiple alternatives. That's no surprise. I'm working on the steering control program for our DARPA Grand Challenge vehicle, and it does that, too. Doesn't mean it's not "digital".
VA Linux (remember them?) at peak had a bigger market cap than the ITER will cost.
Now that we're finally out from under Microsoft Office, we don't want to be in a similar position with Sun. Users don't want Sun to be in a position to change the rules at a future time. That's the whole point of open source.
Actually, Usenet is doing quite well. The spam battle has been won; there's very little spam in the technical groups. Serious workers in difficult fields are on there. Check out, say, "comp.games.development.programming.algorithms", where the people who write physics engines discuss how to do it. Or "comp.std.c++.moderated", where proposed changes to C++ are discussed. Usenet has far lower advertising content than the Web, where, today, "content" seems to be a little box in the middle of the page, surrounded by blinking ads.
This evidence of intent to profit from infringement seems to be what lost the case for them.
Lucasfilm/ILM had some boring industrial buildings in San Raphael. So this is just catchup.