In software, you are typically trying to make one person be both the domain expert (the mechanical engineer), and the tool and die maker (the implementation expert).
The component count in a useful embedded software product probably falls between that of an automobile engine, and that of an entire automobile. Some of the parts in that automobile are indeed simple standardized bolts and washers, but then there are all the others which, when they break and must be replaced, cost $382.27 at the dealer's parts counter.
Demonstrate that the mechanical engineering model miraculously brings a software product from concept to production, with fewer people and at lower cost than Ford or Boeing or Bechtel require to produce a tangible product.
Look also at the magnitude and time scale of the investment comparatively between the mechanical product and the software product, both in design, and in design verification.
As a fellow long-time coder (30 years), I must in large part agree, though I find it very difficult to part company with my "pathetic" need to print. I just wish I were printing something more informative than code as we know it.
Our tools are absurdly dull. In December, I visited the Getty Museum, and saw an exhibition of sculptures by Jean-Antoine Houdon. The sculptor uses a hammer and a chisel or two, and a nearly incomprehensible amount of vision and skill, to produce magnificent works. Slowly.
And here, I as a programmer, try to build works which, if not magnificent, are at least functional, by gluing individual grains of sand together under a magnifying glass, with a pair of sadly deformed Ubangistani discount tweezers, and some dubious glue.
If I use too much of my past experience in the work, my previous employer will sue me to ruin. I must either be indentured for too long to one master, or must behave as if my mind has been substantially erased of the most directly useful elements of my previous efforts.
Simonyi also makes a valuable observation by delineating the problem of domain expertise vs implementation expertise.
Every time it becomes my job to produce a piece of code which does something, I spend an awful lot of time trying to be conversant enough with some complex problem domain, to be able to do something useful within it.
This document makes me ANGRY. The weasly little DC lawyers who wrote it should be strung up by their whatevers and left to hang in a public place with a generous supply of large rocks nearby. Its tone is not that of something issuing from a checked and balanced organ of a constitutional and freedom loving republic.
Gems such as "The rules should make it clear that carriers are not permitted to argue in any petition for further extension of the Packet-Mode Deadline that the service for which a further extension is being sought is not covered by CALEA."
I'm sorry, but this is the language of paranoid totalitarianism, not of a democratic and constitutional republic. Much as I would like to blame this solely on Ashcroft, CALEA goes back an administration or two.
DoJ: "You must comply." Carrier: "With what standard." DoJ: "We can't tell you." Carrier: "What do you mean you can't tell me?" DoJ: "We're all about compliance, not about standards." Carrier: "But what am I supposed to comply with." DoJ: "What we want you to comply with." Carrier: "But what IS that?" DoJ: "That's not our department, were only enforce compliance."
I believe we must have accidentally outsourced the Department of Justice to a North Korean front entity.
It is all too frequent to confuse the bandwidth of a scope (or the frequency of a sampling process) with the fundamental operating frequency of the system under test.
If I have a digital system running with a 50MHz (nominally square wave) clock, and I put a 50MHz bandwidth scope onto the clock, I will see a 50MHz sine wave. That gives me some information, but not much more than the fact that the signal is present, and has some approximate amplitude.
To actually see the squareness (or lack of it) of the 50MHz square wave, I need to add an infinite series of odd harmonics (150MHz, 250MHz, 350MHz,...) of decreasing amplitude: sin(n*x)/n, n=1, 3, 5,...
To get a meaningful idea of whether I have impedance mismatches between source, transmission line, and load (causing reflections and thus clock glitches, or clock to signal skew or jitter), I could need a scope with at least 350MHz bandwidth to see what is really happening in those crucial few nanoseconds of the signal transitions. In digital terms, this would mean a sampling rate of at least 700MHz.
It means that you are ignorant of conventions of English outside the United States. British usage considers a company name as denoting a plural entity, thus "Picotech do" rather than "Picotech does."
Though I am American, I have adopted this usage, not out of affectation, but in recognition of the fact that companies are groups of human individuals, organized and operating in one portion of their lives toward some common purpose.
United States currency paper is made by Crane & Co. You can buy Crane's paper at good stationery stores. It is wonderful paper, but it is not their currency paper.
Crystals in the usual MHz frequency ranges are NOT especially expensive to obtain at custom frequencies. Circa 1972 I had one made in the 4MHz range and it cost me around $13.
The place which made them was not far away from me, and you could see the simple benchtop polishing machinery from the front counter.
The situation with 32768 Hz tuning fork quartz resonators for watches is a bit different. I suspect, though, that even these are susceptible to hacking.
Just open up that tiny tubular can, put the crystal into a vacuum deposition system, and evaporate a bit more metal onto it (increasing the mass of the fork arms) until it oscillates at 31903.94 Hz instead of 32768. Whether you are going to be able to pull the frequency down this way by as much as 2.64% is an open question. Then find a way to repackage it.
The way you did this back in the days of WW2 surplus crystals in screw-closed FT-243 holders, was to open up the holder and use a soft pencil to make a mark on the surface of the crystal, increasing the mass slightly and lowering its frequency. To increase the frequency, you would grind the crystal thickness down a little against plate glass with a very fine abrasive (e.g. toothpaste).
I have been a member of IEEE CS for a number of years, with a couple of gaps during times when the considerable expense seemed imprudent. I add and drop publications as my interests change. The expense can be stiff, especially if one maintains diverse interests. (IEEE once called me up and asked if I was breaking the rules by trying to use personal rate subscriptions to stock a library.)
I was previously a member of ACM. At one time CACM was something to look forward to every month, and it was visually elegant. Then they changed to computer typesetting and for some reason decided to use a hideous ragged-right layout, further mulching the pages with a rotten salad of silly display fonts.
Worse, the good fundamental articles disappeared and were replaced by institutional bleatings, e.g. "eating our seed corn", and promotion of the fads of the moment in so-called "software engineering".
I ascribe the demise of CACM to the editorial tenure of P.J. Denning. Perhaps I am not being fair in this, but that is the opinion that I formed at the time. It must be recognized that the field is vastly broader than it was during CACM's glory days of the 60s and 70s.
Emacs' origins are detailed in this 1979 paper by Bernard Greenberg, with a 1996 introduction. Its history is longer than one might think.
http://www.multicians.org/mepap.html
You cannot make any such conclusions from the face of the issued patent. You would have to examine the "wrapper" (file history of the application) to learn the real story.
Lemelson is infamous for these sorts of "submarine" patents, kept on the slowest of snail's paces in the examination process by amendment after amendment.
After enough iterations over the years, the application is finally allowed to gel on some concept just a little ahead of the documented state of the art, and the hook is set.
Then the patent gets issued and the shakedown begins.
Now that changes in patent law have applications being published some reasonable interval after being submitted, submarining should be much more difficult to perpetrate.
Tivo have made source code available since approximately day one of their product.
Even before they had an FTP site, they would ship promptly and for a very reasonable fee, source on CD-R.
The real guts of the product, including all substantial video-related drivers, are in loadable modules. The kernel and provided source have just enough hardware-specific code to calm the hardware down enough to allow the kernel to get started.
As far as I can tell, Tivo have done everything they need to under GPL.
[Disclosure of interests: I own a small amount of Tivo stock. When I ordered the source code way back when, they included a nice Tivo hat along with the CD.]
If you are going to post so-called extracts of others' work, it is good form to post cites, especially if the work is entirely copied from one source, with zero editorial contribution or interpretation, as your post appears to have been.
The above material appears to come from an article by Stephen J. Gislason, MD from the following page:
These figures are incorrect. The ATSC 8VSB digital TV signal has a 19.39Mbit/sec payload.
That will contain one HDTV program with current MPEG-2 encoding, or up to 6 standard definition programs with reasonably aggressive MPEG-2 encoding.
Using MPEG-4 or other advanced encoding, one might be able to get the 1.76Mbit/sec rate required to support 11 SD programs per 8VSB channel.
I don't know of anyone who believes that HD has much prospect of being acceptably encoded below 7-8Mbit/sec in the forseeable future. That means 2 HD programs per RF channel at most.
Incorrect. Consumer electronics frequently use ceramic capacitors with Z5U dielectric. This dielectric is only rated to +10C, and at 0C a capacitor made with it essentially ceases to function.
Re:How the heck do you pronounce "Qatar"?
on
4l-j4z333ra 0wn3d
·
· Score: 1
The problem with Qatar is that it contains two sounds which are not part of the English sound system. The "Q" is a glottal/uvular stop. The "t" is the uniquely Arabic emphatic T, which is pronounced with the tongue contacting considerably farther back than an English "t", with the effect of rounding and deepening the surrounding vowels.
The cadence and stress are akin to "cutter", said quickly.
Disclaimer: I'm only in my third semester of Arabic.
I don't know which is more offensive: the idiots who flood me with spam advertising this silly fraud, or the people who get fleeced by falling for it.
They ask $200 for what is actually a very ordinary high pass filter. One Web site offers these in bulk for $2.25 each in lots of 2500+. Buy a container load from China, and they are probably $0.80 each.
On one-way cable, these can be used to cut interference from AM, ham, and shortwave radio transmitters. On two-way cable, they will have the effect of killing the return path from the settop box to the cable company.
The device is in no way a "descrambler".
Some sort of vacuum gauge which was labelled as containing Radium 226, sensibly stored outdoors.
A vast stack of plastic barrels which had once contained artificial flavorings for ice cream. I opened the bung of one and felt as if I had been bowled down by a rotting peach 75 feet in diameter.
The absolute best pair of needlenose pliers which I have ever owned, bar none: Erem Swiss. Cost maybe $3.75. They are still my most frequently used pliers.
For a process to be meaningfully based upon statistics, one must be dealing with something which is meaningfully quantifiable, and which occurs in a sufficient number of comparably quantifiable instances to be stastically significant.
What about an IT process is meaningfully quantifiable?
What about an IT process occurs in statistically significant numbers of comparably quantifiable instances?
What do you have a million of to see where you stand relative to 3.04 defects? Transaction processing response times. What else? Not much that I can think of.
1. Digital signature combining identify of camera and operator, with date and time
2. An audit trail of the enhancement, showing exactly step-by-step the operations performed upon various areas of the picture, to produce the result brought into evidence, along with the identity of the operator
3. A demonstration of the progression of these operations applied to the original source material, resulting in the final image in evidence.
It would be good if the actual MOU were published somewhere.
It is almost certain that 5C protection is specified for the 1394 interface. And it probably also specifies Macrovision for any analog video outputs.
Sadly, there are no fewer than 4 broadcast digital systems in use in the U.S. today: digital cable (64QAM or 256QAM modulation), ATSC (for airwave broadcast: 8VSB modulation), DVB-S for Echostar, and DirecTV's system.
All but DirecTV are based upon MPEG-2 Transport packaging/framing (but most MPEG decoder chips can handle both MPEG-2 Transport and DirecTV framing). Each has its own modulation, Forward Error Correction, and interleaving scheme, due to the differing respective characteristics of the cable, airwave, and satellite channels.
Add to this several different conditional access systems. Perhaps this agreement specifies CableLabs' POD pluggable conditional access module framework.
I don't see any victory in this for advocates of weak protection.
I also don't see a universal digital receiver coming from this alone. It is possible, though, that some receiver manufacturers will include multi-standard tuners and demodulators. The added tuner and demodulator costs should be around $14, which would show up as around $50 extra at retail, by comparison with a digital cable only receiver.
I used to, in the '70s eagerly await the arrival of EN, but I stopped reading it around 1983. Every issue seemed to bring more news of large layoffs than of new developments.
The all time low came when one of the trade papers ran an article with the headline "Cat Fight Looms in Logic Analyzers". I don't recall if this was EN, but...
Perhaps there was a time after that when it improved for a while. I don't know - EET became the main source of my industry news and I never looked back. Now, with more of my information wants met online, I just spend a few minutes scanning through someone else's paper copy of EET.
Paper has the great advantage of being readable while on the toilet.
Any organization with the word "Coalition" in the name are socialists operating under some other guise. Same for any organization which uses the word "Toxics" [sic].
These organizations spread disinformation and hysteria, with the real intent of 1) stopping profitable business or 2) finding a high-leverage segment of the economy to burden with special taxes and fees. Why? To support their Marxian ideal of redistribution of income.
Most could not care less about the actual ecological effects. They want instead to make people living within a successful economic system, feel guilty that people living within a murderous and failing economic system can only make $0.17/hour.
A CRT screen contains phosphors, not "phosphorus". The U.S. Navy lists phosphorus as "highly toxic". Indeed - elemental phosphorus is nasty, but without phosphorus as phosphate, our metabolism would cease instantly. The confusion of the two words is either inexcusably ignorant, or deliberately fraudulent.
These people are loudmouth liars - they don't care whether they get ANYTHING correct, as long as they can gain more control.
Color CRT screen phosphors have no significant quantity of phosphorus in them. They contain Yttrium, Europium, Vanadium, Zinc, Sulfur, Silver, Copper, Gold, Aluminum, and Oxygen.
Some phosphors in older or specialized CRTs did contain Cadmium, which is toxic. Of course gazillions of tons of Cadmium go onto steel as corrosion inhibiting coatings, and Cadmium is fundamental to NiCd rechargeable batteries.
It is difficult to believe that the lead bound up in the leaded glass in CRTs is anything other than less hazardous than the elemental and exposed lead in solder in nearly every electronic device ever manufactured.
These shrill fraudulent twerps, leveraged through criminally ignorant, malicious, or ideologically aligned lawyers and regulators, managed, nonetheless, to make it impossible to simply throw away an old TV set or CRT monitor in California.
In the meantime, many of them are almost certainly hypocritically driving giant gas-sucking SUVs.
Slashdot Mirror
The component count in a useful embedded software product probably falls between that of an automobile engine, and that of an entire automobile. Some of the parts in that automobile are indeed simple standardized bolts and washers, but then there are all the others which, when they break and must be replaced, cost $382.27 at the dealer's parts counter.
Demonstrate that the mechanical engineering model miraculously brings a software product from concept to production, with fewer people and at lower cost than Ford or Boeing or Bechtel require to produce a tangible product.
Look also at the magnitude and time scale of the investment comparatively between the mechanical product and the software product, both in design, and in design verification.
Our tools are absurdly dull. In December, I visited the Getty Museum, and saw an exhibition of sculptures by Jean-Antoine Houdon. The sculptor uses a hammer and a chisel or two, and a nearly incomprehensible amount of vision and skill, to produce magnificent works. Slowly.
And here, I as a programmer, try to build works which, if not magnificent, are at least functional, by gluing individual grains of sand together under a magnifying glass, with a pair of sadly deformed Ubangistani discount tweezers, and some dubious glue.
If I use too much of my past experience in the work, my previous employer will sue me to ruin. I must either be indentured for too long to one master, or must behave as if my mind has been substantially erased of the most directly useful elements of my previous efforts.
Simonyi also makes a valuable observation by delineating the problem of domain expertise vs implementation expertise.
Every time it becomes my job to produce a piece of code which does something, I spend an awful lot of time trying to be conversant enough with some complex problem domain, to be able to do something useful within it.
Gems such as "The rules should make it clear that carriers are not permitted to argue in any petition for further extension of the Packet-Mode Deadline that the service for which a further extension is being sought is not covered by CALEA."
I'm sorry, but this is the language of paranoid totalitarianism, not of a democratic and constitutional republic. Much as I would like to blame this solely on Ashcroft, CALEA goes back an administration or two.
DoJ: "You must comply." Carrier: "With what standard." DoJ: "We can't tell you." Carrier: "What do you mean you can't tell me?" DoJ: "We're all about compliance, not about standards." Carrier: "But what am I supposed to comply with." DoJ: "What we want you to comply with." Carrier: "But what IS that?" DoJ: "That's not our department, were only enforce compliance."
I believe we must have accidentally outsourced the Department of Justice to a North Korean front entity.
If I have a digital system running with a 50MHz (nominally square wave) clock, and I put a 50MHz bandwidth scope onto the clock, I will see a 50MHz sine wave. That gives me some information, but not much more than the fact that the signal is present, and has some approximate amplitude.
To actually see the squareness (or lack of it) of the 50MHz square wave, I need to add an infinite series of odd harmonics (150MHz, 250MHz, 350MHz, ...) of decreasing amplitude: sin(n*x)/n, n=1, 3, 5, ...
To get a meaningful idea of whether I have impedance mismatches between source, transmission line, and load (causing reflections and thus clock glitches, or clock to signal skew or jitter), I could need a scope with at least 350MHz bandwidth to see what is really happening in those crucial few nanoseconds of the signal transitions. In digital terms, this would mean a sampling rate of at least 700MHz.
Though I am American, I have adopted this usage, not out of affectation, but in recognition of the fact that companies are groups of human individuals, organized and operating in one portion of their lives toward some common purpose.
The paper is made principally of cotton fiber.
The place which made them was not far away from me, and you could see the simple benchtop polishing machinery from the front counter.
The situation with 32768 Hz tuning fork quartz resonators for watches is a bit different. I suspect, though, that even these are susceptible to hacking.
Just open up that tiny tubular can, put the crystal into a vacuum deposition system, and evaporate a bit more metal onto it (increasing the mass of the fork arms) until it oscillates at 31903.94 Hz instead of 32768. Whether you are going to be able to pull the frequency down this way by as much as 2.64% is an open question. Then find a way to repackage it.
The way you did this back in the days of WW2 surplus crystals in screw-closed FT-243 holders, was to open up the holder and use a soft pencil to make a mark on the surface of the crystal, increasing the mass slightly and lowering its frequency. To increase the frequency, you would grind the crystal thickness down a little against plate glass with a very fine abrasive (e.g. toothpaste).
I was previously a member of ACM. At one time CACM was something to look forward to every month, and it was visually elegant. Then they changed to computer typesetting and for some reason decided to use a hideous ragged-right layout, further mulching the pages with a rotten salad of silly display fonts.
Worse, the good fundamental articles disappeared and were replaced by institutional bleatings, e.g. "eating our seed corn", and promotion of the fads of the moment in so-called "software engineering".
I ascribe the demise of CACM to the editorial tenure of P.J. Denning. Perhaps I am not being fair in this, but that is the opinion that I formed at the time. It must be recognized that the field is vastly broader than it was during CACM's glory days of the 60s and 70s.
Emacs' origins are detailed in this 1979 paper by Bernard Greenberg, with a 1996 introduction. Its history is longer than one might think. http://www.multicians.org/mepap.html
Yeah, but it's vastly better than the totally hideous CDE which they had been pushing.
You cannot make any such conclusions from the face of the issued patent. You would have to examine the "wrapper" (file history of the application) to learn the real story.
Lemelson is infamous for these sorts of "submarine" patents, kept on the slowest of snail's paces in the examination process by amendment after amendment.
After enough iterations over the years, the application is finally allowed to gel on some concept just a little ahead of the documented state of the art, and the hook is set.
Then the patent gets issued and the shakedown begins.
Now that changes in patent law have applications being published some reasonable interval after being submitted, submarining should be much more difficult to perpetrate.
Even before they had an FTP site, they would ship promptly and for a very reasonable fee, source on CD-R.
The real guts of the product, including all substantial video-related drivers, are in loadable modules. The kernel and provided source have just enough hardware-specific code to calm the hardware down enough to allow the kernel to get started.
As far as I can tell, Tivo have done everything they need to under GPL.
[Disclosure of interests: I own a small amount of Tivo stock. When I ordered the source code way back when, they included a nice Tivo hat along with the CD.]
The above material appears to come from an article by Stephen J. Gislason, MD from the following page:
http://www.vegan-straight-edge.org.uk/nocoffee.htm
A quick search shows that Gislason is CEO of Alpha Nutrition, which would like to sell you nutritional supplements.
These figures are incorrect. The ATSC 8VSB digital TV signal has a 19.39Mbit/sec payload.
That will contain one HDTV program with current MPEG-2 encoding, or up to 6 standard definition programs with reasonably aggressive MPEG-2 encoding.
Using MPEG-4 or other advanced encoding, one might be able to get the 1.76Mbit/sec rate required to support 11 SD programs per 8VSB channel.
I don't know of anyone who believes that HD has much prospect of being acceptably encoded below 7-8Mbit/sec in the forseeable future. That means 2 HD programs per RF channel at most.
Each of these two patents appears to have exactly one claim which could even remotely be considered broad enough to be applicable to eBay.
Most of the independent claims are cluttered with enough extraneous and hoary specifics to make them just laughably inapplicable.
When you get further down to the broadened independent claims without so many specifics, it still doesn't look much like eBay.
If there is any logic to this verdict, it must be tortured indeed.
Incorrect. Consumer electronics frequently use ceramic capacitors with Z5U dielectric. This dielectric is only rated to +10C, and at 0C a capacitor made with it essentially ceases to function.
Make it a petajoule and I'll be impressed.
The problem with Qatar is that it contains two sounds which are not part of the English sound system. The "Q" is a glottal/uvular stop. The "t" is the uniquely Arabic emphatic T, which is pronounced with the tongue contacting considerably farther back than an English "t", with the effect of rounding and deepening the surrounding vowels. The cadence and stress are akin to "cutter", said quickly. Disclaimer: I'm only in my third semester of Arabic.
I don't know which is more offensive: the idiots who flood me with spam advertising this silly fraud, or the people who get fleeced by falling for it. They ask $200 for what is actually a very ordinary high pass filter. One Web site offers these in bulk for $2.25 each in lots of 2500+. Buy a container load from China, and they are probably $0.80 each. On one-way cable, these can be used to cut interference from AM, ham, and shortwave radio transmitters. On two-way cable, they will have the effect of killing the return path from the settop box to the cable company. The device is in no way a "descrambler".
Ahhh yes Apex.
During one visit there 20+ years ago I saw:
A small mountain of old airliner seats.
Some sort of vacuum gauge which was labelled as containing Radium 226, sensibly stored outdoors.
A vast stack of plastic barrels which had once contained artificial flavorings for ice cream. I opened the bung of one and felt as if I had been bowled down by a rotting peach 75 feet in diameter.
The absolute best pair of needlenose pliers which I have ever owned, bar none: Erem Swiss. Cost maybe $3.75. They are still my most frequently used pliers.
What about an IT process is meaningfully quantifiable?
What about an IT process occurs in statistically significant numbers of comparably quantifiable instances?
What do you have a million of to see where you stand relative to 3.04 defects? Transaction processing response times. What else? Not much that I can think of.
1. Digital signature combining identify of camera and operator, with date and time
2. An audit trail of the enhancement, showing exactly step-by-step the operations performed upon various areas of the picture, to produce the result brought into evidence, along with the identity of the operator
3. A demonstration of the progression of these operations applied to the original source material, resulting in the final image in evidence.
CableLabs article
It would be good if the actual MOU were published somewhere.
It is almost certain that 5C protection is specified for the 1394 interface. And it probably also specifies Macrovision for any analog video outputs.
Sadly, there are no fewer than 4 broadcast digital systems in use in the U.S. today: digital cable (64QAM or 256QAM modulation), ATSC (for airwave broadcast: 8VSB modulation), DVB-S for Echostar, and DirecTV's system.
All but DirecTV are based upon MPEG-2 Transport packaging/framing (but most MPEG decoder chips can handle both MPEG-2 Transport and DirecTV framing). Each has its own modulation, Forward Error Correction, and interleaving scheme, due to the differing respective characteristics of the cable, airwave, and satellite channels.
Add to this several different conditional access systems. Perhaps this agreement specifies CableLabs' POD pluggable conditional access module framework.
I don't see any victory in this for advocates of weak protection.
I also don't see a universal digital receiver coming from this alone. It is possible, though, that some receiver manufacturers will include multi-standard tuners and demodulators. The added tuner and demodulator costs should be around $14, which would show up as around $50 extra at retail, by comparison with a digital cable only receiver.
The all time low came when one of the trade papers ran an article with the headline "Cat Fight Looms in Logic Analyzers". I don't recall if this was EN, but...
Perhaps there was a time after that when it improved for a while. I don't know - EET became the main source of my industry news and I never looked back. Now, with more of my information wants met online, I just spend a few minutes scanning through someone else's paper copy of EET.
Paper has the great advantage of being readable while on the toilet.
These organizations spread disinformation and hysteria, with the real intent of 1) stopping profitable business or 2) finding a high-leverage segment of the economy to burden with special taxes and fees. Why? To support their Marxian ideal of redistribution of income.
Most could not care less about the actual ecological effects. They want instead to make people living within a successful economic system, feel guilty that people living within a murderous and failing economic system can only make $0.17/hour.
A CRT screen contains phosphors, not "phosphorus". The U.S. Navy lists phosphorus as "highly toxic". Indeed - elemental phosphorus is nasty, but without phosphorus as phosphate, our metabolism would cease instantly. The confusion of the two words is either inexcusably ignorant, or deliberately fraudulent.
These people are loudmouth liars - they don't care whether they get ANYTHING correct, as long as they can gain more control.
Color CRT screen phosphors have no significant quantity of phosphorus in them. They contain Yttrium, Europium, Vanadium, Zinc, Sulfur, Silver, Copper, Gold, Aluminum, and Oxygen.
Some phosphors in older or specialized CRTs did contain Cadmium, which is toxic. Of course gazillions of tons of Cadmium go onto steel as corrosion inhibiting coatings, and Cadmium is fundamental to NiCd rechargeable batteries.
It is difficult to believe that the lead bound up in the leaded glass in CRTs is anything other than less hazardous than the elemental and exposed lead in solder in nearly every electronic device ever manufactured.
These shrill fraudulent twerps, leveraged through criminally ignorant, malicious, or ideologically aligned lawyers and regulators, managed, nonetheless, to make it impossible to simply throw away an old TV set or CRT monitor in California.
In the meantime, many of them are almost certainly hypocritically driving giant gas-sucking SUVs.