OO was never designed for speed or efficiency, only ease of modelling business systems.
OO was designed for speed and efficiency for implementing complex applications such as operating systems.
(There, fixed that for you.)
I'm thinking of the OO that predated C++ or Java -- the stuff in Smalltalk and Lisp from the 1970s. (Which still works better than the newer stuff...)
Let's see, I have an original Macintosh that still boots, a TRS-80 circa 1977 that I think works (but haven't tried it lately), and an H19 terminal (kit) circa 1979 that I still actually use as the console for one of my DEC Alpha computers.
I also have a prototype (wire wrapped) Lisp Machine circa 1980 from when I worked at the AI Lab; this was running fine when I last turned it off about 12 years ago, but I would have to do serious maintenance on it (mostly on the disk drive) before attempting to run that one again, for fear of destroying something. I also have some Symbolics machines from the late 80s that probably work.
I think I lost the bare 3x1" circuit board and alligator clips 300 baud modem that I originally used on the H19.
I will prove that anything written in a higher-level language will not be as fast as my implementation of it in C.
I recall a shootout in 1994 where the problem was in Genetic Programming; the result was that Lisp slightly beat GCC. I believe there are plenty of other (and more recent) examples which are of more mundane things where Lisp nearly matches, matches, or exceeds the performance of C programs. I think I saw one involving simple string manipulation recently where Lisp beat C due to poor assumptions by the compiler about memory alignment or something. My memory is a hazy about all these details because Lisp is always fast enough and C is otherwise unsuitable to the problems I work on.
Bring it on.(*) Caveat: It must be a small challenge involving a relatively simple task.
If your claim is merely that you can, by constraining the problem to something suitably trivial, come up with programs in C that are faster than Lisp (or whatever), then I have no doubt that you will in some cases be right. (You might be right less often than you expect, though, because compilers are probably smarter than you think you are.)
If you are concerned only with certain trivial programs and the performance thereof, that's fine, but that's not interesting to most people who consider programming in Lisp.
This reminds me of the LISP/FORTRAN benchmark shootouts from about 1974 where we got sick of people saying that Lisp wasn't good for numerical programming. So the compiler developers spent a month tuning the MACLISP compiler for math, and then LISP was beating FORTRAN all the time after that.
Note also that it makes little sense to compare the speed of languages - you are always comparing the speed of particular implementations (compilers). People usually pick GCC for C, of course. And lots of languages today are single-implementation languages. But Lisp is an ANSI language standard that has lots of (open source and commercial) implementations. (Well, and there are also a number of other Lisp dialects that are not the ANSI one, with all their attendant implementations. In fact, I am sure that I could create a Lisp implementation that will beat whatever C shootout problem you would like to propose to prove your point that one language is faster than another, for that matter. I don't think it would prove anything useful except that your assertion is misguided.)
As for comparing web server performance, there are lots of issues to consider. I didn't read TFOA - not that interested as I already have Lisp-based web servers I like.
By the way, web servers in Lisp are nothing new, they were out there from the beginning. A super-fancy server called CL-HTTP was in use before PHP was invented, for example. There are quite a few web servers written in Lisp, some attached to Apache or IIS, and many as complete standalone servers. Now apparently some dude has coded up a server that outperforms some other server according to some metrics. Yawn.
When I find that my servers are running too slow, I just buy another server. Because they're never running so slow that this is unreasonable. The performance of the Lisp based servers is always comparable to the alternatives in my experience.
I know of incidents at a major educational institute in which grad students were forced to do things like put their hands into dangerous chemicals, causing pain and mutilation. They were terrified that their world-renowned professors would destroy their lives if they reported the incidents.
There are some sick (not to mention criminal) fucks out there controlling these dangerous lab environments. Some of them are Nobel prize winners.
"Lisa, so you won't be scared, I'll show you some of the tools I'll be using. This is the scraper, this is the poker, and this happy little fellow is called the gouger. Now the first thing I'll be doing is chiseling some teeth out of your jawbone. Now hold still while I gas you."
The phone rings and a desperate call, but the team is already engaged and Rick Moranis is left to answer the call. Attempting to prove himself worthy of being a Ghostbuster, he goes alone to encounter a demon who sends him back in time several years before the original Gozer encounter.
Rick does get some action with Sigourney Weaver this time, but some other things go horribly wrong. But there is plenty of action and special effects, and by the end of the movie the team is formed albeit with some slight changes. Ghostbusters Headquarters looks like the Apple Store. Complete with Macbook Pros. And lens flare. Lots of lens flare.
It is by the fizz of Diet Cherry Coke that thoughts acquire speed, the lips acquire a stain the stain becomes a warning it is by will alone I set my mind in motion It is by will alone I set my mind in motion...
When I was an EMT, we had a guy who was throwing up due to overeating and not chewing his food properly, and he had a heart attack, likely from the increase in blood pressure/pulse rate while vomiting.
He shouldn't have eaten that last, one, small, wafer.
Obama promised that his administration would be more transparent. They are pandering to Hollywood and the RIAA, and setting up for full monitoring of all citizens Internet communications. This much seems pretty transparent to me!
So what's the problem, then?
I can't think of any argument that this has any financial impact on the author.
Suppose there is a company that derives profit from the statistical analysis of submitted papers. The student is losing their right to license their papers to that company. Meanwhile, the company profits from having free access to a paper that was otherwise never published. Seems straightforward to me.
This bill does NOT give the government the power to monitor teh Intertubes. It does two things: Secretary of Commerce as the power to map all networks, public and private; President has an OFF switch for the Internet. The reason that all networks are affected is that it speaks about any "United States critical information system or network".
Due to the language of "system", think it probably also extends to the phone network, and perhaps to any computer system that they claim is "United States critical".
I don't understand the part about granting powers to the Government "without regard to any provision of law, regulation, rule, or policy restricting
such access". Can you pass a law that says a branch of the Government is exempt from any laws? Well, I guess you can: many state laws specifically exclude the police.
The bill also prepares for spending a lot of money on grants and studies.
Most interesting thing to me is that it makes it unlawful for anyone not licensed by the federal government to engage in the "cyber security" business inside the USA.
The summary makes it sound like satellites are going to track the airplanes, but that's not what is going on at all.
What this is really about is that the airplane's transponder (simply a radio that transmits about 200 miles around) will broadcast not only the plane's ID tag, but also it's GPS position. Satellites only come into this system in the sense that the airplane has a GPS receiver on board, and GPS is of course satellites. So each airplane broadcasts not only who it is, but where it is. The other new part is that all the airplanes will recieve and process that information to give the pilots a picture of who else is flying around near them. Furthermore, ground radar stations will broadcast on the transponder channel as a proxy for those aircraft that are not equipped to transmit their GPS.
Historically, planes have always transmitted an ID code (mainly, a manually assigned code from the air traffic center who is most recently responsible for them). The next big thing was for the transponder to also include the aircraft's altitude. Now, these are called "transponders" because they only transmit when polled by a ground station's radar sweep. And until recently, only the ground controllers received the transponder hits from the aircraft. About 10 years ago, planes (expensive airliners, mainly) started receiving and processing the nearby transponder responses as well, so that they could see what other planes were at their altitude. This is a collision-avoidance system. So now that planes are equipped with GPS comes the revolution: they can transmit their precise location to each other, and also to the controllers, and everyone can see a complete picture of where all the nearby planes are. This will ultimately enable pilots to fly more efficient routes, allowing more freedom for the controllers and pilots to work things out dynamically.
In the mid-to-late 80s, a "laptop" was the size of a substantial suitcase and weighed ~20 pounds. Examples were the Osborne and Compaq "luggables". Which name I like because it brings forth the proper picture of a big piece of luggage. That was 20 years ago, nowhere near 40.
OK, try more than 30 years ago...
The first luggable computer that I used was the IBM 5100 in 1976. It weighed about 50 lbs, and had a screen that was 16x64 (though 16x32 was more readable and all we needed for APL code). The 16-bit CPU emulated an IBM mainframe, and it ran an OS that offered APL or BASIC (selected from another toggle switch). I used it for APL programming. It had 64 KB of memory and a cartridge tape drive.
IBM link: http://www-03.ibm.com/ibm/history/exhibits/pc/pc_2.html
Someone's good photos: http://wandel.ca/ibm5100/
This was arguably the first personal computer. (Another contender that I had used years earlier would be the Datapoint 2200, but it was larger, and wasn't pitched as a PC, and it wasn't intended to be portable.)
Tags: historical olderthanyouthink nothingisnew GOML
This is about the same size as one of the original TCP/IP stacks on the Internet, on the ITS operating system (advanced timesharing) on the PDP-10 (small mainframe) back in 1982.
he admitted himself that he can't even use a computer! He has to ask his wife for help!
You do realize that the reason he doesn't "use a computer" and "needs help" is that he cannot use his hands and arms to type on a keyboard, because of the permanent damage and disfigurement when he was tortured for defending your right to post on Slashdot, right? And he does have people who use computers for him (people to press buttons, staff of experts to do research for him, create his web pages, read his ten thousand non-spam emails each day, as well as his snail mail, etc.), you know.
Anyway, this is not McCain's bill -- it is sponsored by Charles Schumer [D-NY], and co-sponsors include John Kerry, Dianne Feinstein, Hillary Clinton, and a bunch of other people including...Barack Obama.
Alright, couple things. Yes. It is true that the vast majority of the general public don't want to learn how a computer works.
But I see some fault by manufacturers too. C
The fact that you see any "fault" whatsoever in the desire of the general consumer to have an out-of-the-box perfectly running computer with intuitive user interfaces and the familiar assortment of applications...is why Linux will never succeed there. Linux is for geeks, by geeks, and the geeks don't even begin to comprehend this. Unix-based computing is already available and embraced by the general public. It's called "Macintosh", because Apple fully comprehends this, and did something about it.
The phone will have text entry modes: Abc, ABC, 123,
but will not have the "smart words" input completion
feature. Andriod developers can not learn to use contractions...
Someone posts a one-line flamebait message, and they are modded up because the moderator's politics. I post a polite response stating facts and asking the OP if they have something reasonable to say to back up their flamebait. The result is that I am modded as flamebait.
And if you're not going to war, but just happen to be mistaken for an uniformed combatant. Sorry, you're fucked. We're not required to give you a proper trial, we can just assume you are guilty. Isn't this logic brilliant?
If you were there, you're at war. The point of war is to kill the other people. Once war starts, everybody and everything is fair game. So, yes, you're fucked. Sorry, and have nice day.
"Innocent until proven guilty. Although that statement is ignored just as often in the US as it is in England, laws that we pass try to at least give the impression that we respect it."
umm, Guantanamo Bay?
Guantanamo Bay is where we are holding certain people who we captured during a war, people who were violating the rules of war (and thus not even part of the Geneva Convention). But the subject of this discussion is about citizens being arrested and held in their own country. Which US citizens are in Guantanamo Bay?
Increasing the amount of time that someone can be held is only vaguely related to demanding the disclosure of encryption keys. (The vague relation is a very dangerous one -- people's freedom and the encroachment of police state.) If the Government wants to lump these things together, it's an obvious power ploy and a bad idea. If either of these ideas (incarceration time, encryption) actually has merit, perhaps they should be debated seperately? Would that strategy be better, or worse, in terms of the gradual erosion of freedom. Super-laws that eliminate the normal protections of the people's rights under law are by definition a poor idea, no matter how pragmatic they seem.
it's seen as a way to serve the community by providing emergency communication in times of need. Do you really need more than that?
Ham radio is often touted as comprising a potentially critical emergency/disaster communications system, when the telephone (or Internet) infrastructure is down.
When the public electrical power system is compromised, most hams will be unable to communicate, because all they have is a little handy talkie that can barely reach the repeater base stations (which are themselves offline). Sure, some repeaters have battery backups. But so does the public telephone network and the Internet. And what volume of traffic and kind of traffic do you suppose the hams will be transmitting, even if the repeater facilities are operational?
If one is suggesting that ham radio is critical for the authorities to communicate, wouldn't it be better to rework the official emergency infrastructure so that it doesn't go to pot in the event of the kind of emergencies that it's intended to respond to? (Is it really, in fact, so vulnerable, that the officials will all be helpless, wandering around and hoping that they encounter some amateur radio hobbyist who can save the day?)
Maybe the reality is that ham radio isn't a significant part of the official emergency infrastructure, but only augments it. Primarily, for private messages. In that case, what sense does Morse Code make? I am sure we can make rugged (even EMP resistant) computer-radio systems that work reliably in poor reception conditions and so on. Why not allow anybody at all to access the citizen-use emergency frequencies using a modern encoding system such as voice or data packets, and without needing any license?
On the other hand, if ham radio is just for education and entertainment, why not limit it to exclusively Morse Code transmissions? Or at least that the operators only use "experimental" home-built equipment (which used to be tubes, but nowadays might be mostly-software kits)? These hobby hands would be for licensed hobbyists and expiermenters, not for people who are walking around with what amount to low-quality party-line mobile telephones they bought at the store.
My comments above were written with the USA in mind, not third-world or wilderness countries that don't have modern communications systems everywhere. But those places have perfectly good satellite phones, anyway. Maybe the civil defense authorities everywhere should issue some kind of satellite phone, so there's always going to be one nearby (in your village, out on your ranch, or on your suburban block, and in someone's nearby vehicle) for when you really need it. And there could be regional emergency telephone centers that switch the calls, point-to-point, from the various satellite, land-line, and mobile calls. These distributed emergency centers could be made to withstand any emergency situation. Of course, the message capacity would be very limited, but not so limited as the ham radio system.
I can't transmit below 50MHz because I can't seem to learn a 150 year old communications method.
Most hams seem to take the attitude of "I had to learn it, I'm glad I did, but I don't use it" when it comes to code. If my current difficulty learning code continues, I really doubt I'll use it, unless my life depends on it (which, since no one else will be able to understand, will be a useless excercise). Too many bad memories.
Who says code isn't high tech? Isn't Morse Code used primarily by astronauts and submariners? You know, banging on the hull? Or interpreting the message from the prodigal-cum-alien space probe?
Slashdot Mirror
OO was never designed for speed or efficiency, only ease of modelling business systems.
OO was designed for speed and efficiency for implementing complex applications such as operating systems. (There, fixed that for you.) I'm thinking of the OO that predated C++ or Java -- the stuff in Smalltalk and Lisp from the 1970s. (Which still works better than the newer stuff...)
Let's see, I have an original Macintosh that still boots, a TRS-80 circa 1977 that I think works (but haven't tried it lately), and an H19 terminal (kit) circa 1979 that I still actually use as the console for one of my DEC Alpha computers. I also have a prototype (wire wrapped) Lisp Machine circa 1980 from when I worked at the AI Lab; this was running fine when I last turned it off about 12 years ago, but I would have to do serious maintenance on it (mostly on the disk drive) before attempting to run that one again, for fear of destroying something. I also have some Symbolics machines from the late 80s that probably work. I think I lost the bare 3x1" circuit board and alligator clips 300 baud modem that I originally used on the H19.
I will prove that anything written in a higher-level language will not be as fast as my implementation of it in C.
I recall a shootout in 1994 where the problem was in Genetic Programming; the result was that Lisp slightly beat GCC. I believe there are plenty of other (and more recent) examples which are of more mundane things where Lisp nearly matches, matches, or exceeds the performance of C programs. I think I saw one involving simple string manipulation recently where Lisp beat C due to poor assumptions by the compiler about memory alignment or something. My memory is a hazy about all these details because Lisp is always fast enough and C is otherwise unsuitable to the problems I work on.
Bring it on.(*) Caveat: It must be a small challenge involving a relatively simple task.
If your claim is merely that you can, by constraining the problem to something suitably trivial, come up with programs in C that are faster than Lisp (or whatever), then I have no doubt that you will in some cases be right. (You might be right less often than you expect, though, because compilers are probably smarter than you think you are.)
If you are concerned only with certain trivial programs and the performance thereof, that's fine, but that's not interesting to most people who consider programming in Lisp. This reminds me of the LISP/FORTRAN benchmark shootouts from about 1974 where we got sick of people saying that Lisp wasn't good for numerical programming. So the compiler developers spent a month tuning the MACLISP compiler for math, and then LISP was beating FORTRAN all the time after that.
Note also that it makes little sense to compare the speed of languages - you are always comparing the speed of particular implementations (compilers). People usually pick GCC for C, of course. And lots of languages today are single-implementation languages. But Lisp is an ANSI language standard that has lots of (open source and commercial) implementations. (Well, and there are also a number of other Lisp dialects that are not the ANSI one, with all their attendant implementations. In fact, I am sure that I could create a Lisp implementation that will beat whatever C shootout problem you would like to propose to prove your point that one language is faster than another, for that matter. I don't think it would prove anything useful except that your assertion is misguided.)
As for comparing web server performance, there are lots of issues to consider. I didn't read TFOA - not that interested as I already have Lisp-based web servers I like.
By the way, web servers in Lisp are nothing new, they were out there from the beginning. A super-fancy server called CL-HTTP was in use before PHP was invented, for example. There are quite a few web servers written in Lisp, some attached to Apache or IIS, and many as complete standalone servers. Now apparently some dude has coded up a server that outperforms some other server according to some metrics. Yawn.
When I find that my servers are running too slow, I just buy another server. Because they're never running so slow that this is unreasonable. The performance of the Lisp based servers is always comparable to the alternatives in my experience.
I know of incidents at a major educational institute in which grad students were forced to do things like put their hands into dangerous chemicals, causing pain and mutilation. They were terrified that their world-renowned professors would destroy their lives if they reported the incidents. There are some sick (not to mention criminal) fucks out there controlling these dangerous lab environments. Some of them are Nobel prize winners.
"Lisa, so you won't be scared, I'll show you some of the tools I'll be using. This is the scraper, this is the poker, and this happy little fellow is called the gouger. Now the first thing I'll be doing is chiseling some teeth out of your jawbone. Now hold still while I gas you."
The phone rings and a desperate call, but the team is already engaged and Rick Moranis is left to answer the call. Attempting to prove himself worthy of being a Ghostbuster, he goes alone to encounter a demon who sends him back in time several years before the original Gozer encounter. Rick does get some action with Sigourney Weaver this time, but some other things go horribly wrong. But there is plenty of action and special effects, and by the end of the movie the team is formed albeit with some slight changes. Ghostbusters Headquarters looks like the Apple Store. Complete with Macbook Pros. And lens flare. Lots of lens flare.
It is by the fizz of Diet Cherry Coke that thoughts acquire speed, the lips acquire a stain the stain becomes a warning it is by will alone I set my mind in motion It is by will alone I set my mind in motion...
When I was an EMT, we had a guy who was throwing up due to overeating and not chewing his food properly, and he had a heart attack, likely from the increase in blood pressure/pulse rate while vomiting.
He shouldn't have eaten that last, one, small, wafer.
Obama promised that his administration would be more transparent. They are pandering to Hollywood and the RIAA, and setting up for full monitoring of all citizens Internet communications. This much seems pretty transparent to me! So what's the problem, then?
I can't think of any argument that this has any financial impact on the author.
Suppose there is a company that derives profit from the statistical analysis of submitted papers. The student is losing their right to license their papers to that company. Meanwhile, the company profits from having free access to a paper that was otherwise never published. Seems straightforward to me.
This bill does NOT give the government the power to monitor teh Intertubes. It does two things: Secretary of Commerce as the power to map all networks, public and private; President has an OFF switch for the Internet. The reason that all networks are affected is that it speaks about any "United States critical information system or network". Due to the language of "system", think it probably also extends to the phone network, and perhaps to any computer system that they claim is "United States critical". I don't understand the part about granting powers to the Government "without regard to any provision of law, regulation, rule, or policy restricting such access". Can you pass a law that says a branch of the Government is exempt from any laws? Well, I guess you can: many state laws specifically exclude the police. The bill also prepares for spending a lot of money on grants and studies. Most interesting thing to me is that it makes it unlawful for anyone not licensed by the federal government to engage in the "cyber security" business inside the USA.
Why do you say is your IM buddy really a chatbot? Earlier you said something about your motherboard.
What this is really about is that the airplane's transponder (simply a radio that transmits about 200 miles around) will broadcast not only the plane's ID tag, but also it's GPS position. Satellites only come into this system in the sense that the airplane has a GPS receiver on board, and GPS is of course satellites. So each airplane broadcasts not only who it is, but where it is. The other new part is that all the airplanes will recieve and process that information to give the pilots a picture of who else is flying around near them. Furthermore, ground radar stations will broadcast on the transponder channel as a proxy for those aircraft that are not equipped to transmit their GPS.
Historically, planes have always transmitted an ID code (mainly, a manually assigned code from the air traffic center who is most recently responsible for them). The next big thing was for the transponder to also include the aircraft's altitude. Now, these are called "transponders" because they only transmit when polled by a ground station's radar sweep. And until recently, only the ground controllers received the transponder hits from the aircraft. About 10 years ago, planes (expensive airliners, mainly) started receiving and processing the nearby transponder responses as well, so that they could see what other planes were at their altitude. This is a collision-avoidance system. So now that planes are equipped with GPS comes the revolution: they can transmit their precise location to each other, and also to the controllers, and everyone can see a complete picture of where all the nearby planes are. This will ultimately enable pilots to fly more efficient routes, allowing more freedom for the controllers and pilots to work things out dynamically.
See also: [Levitt, 1981] David Levitt. A Melody Description System for Jazz Improvisation. Masterâ(TM)s thesis, MIT, Cambridge, MA, 1981
In the mid-to-late 80s, a "laptop" was the size of a substantial suitcase and weighed ~20 pounds. Examples were the Osborne and Compaq "luggables". Which name I like because it brings forth the proper picture of a big piece of luggage. That was 20 years ago, nowhere near 40.
OK, try more than 30 years ago... The first luggable computer that I used was the IBM 5100 in 1976. It weighed about 50 lbs, and had a screen that was 16x64 (though 16x32 was more readable and all we needed for APL code). The 16-bit CPU emulated an IBM mainframe, and it ran an OS that offered APL or BASIC (selected from another toggle switch). I used it for APL programming. It had 64 KB of memory and a cartridge tape drive. IBM link: http://www-03.ibm.com/ibm/history/exhibits/pc/pc_2.html Someone's good photos: http://wandel.ca/ibm5100/ This was arguably the first personal computer. (Another contender that I had used years earlier would be the Datapoint 2200, but it was larger, and wasn't pitched as a PC, and it wasn't intended to be portable.) Tags: historical olderthanyouthink nothingisnew GOML
This is about the same size as one of the original TCP/IP stacks on the Internet, on the ITS operating system (advanced timesharing) on the PDP-10 (small mainframe) back in 1982.
he admitted himself that he can't even use a computer! He has to ask his wife for help!
You do realize that the reason he doesn't "use a computer" and "needs help" is that he cannot use his hands and arms to type on a keyboard, because of the permanent damage and disfigurement when he was tortured for defending your right to post on Slashdot, right? And he does have people who use computers for him (people to press buttons, staff of experts to do research for him, create his web pages, read his ten thousand non-spam emails each day, as well as his snail mail, etc.), you know. Anyway, this is not McCain's bill -- it is sponsored by Charles Schumer [D-NY], and co-sponsors include John Kerry, Dianne Feinstein, Hillary Clinton, and a bunch of other people including ...Barack Obama.
Alright, couple things. Yes. It is true that the vast majority of the general public don't want to learn how a computer works.
But I see some fault by manufacturers too. C
The fact that you see any "fault" whatsoever in the desire of the general consumer to have an out-of-the-box perfectly running computer with intuitive user interfaces and the familiar assortment of applications...is why Linux will never succeed there. Linux is for geeks, by geeks, and the geeks don't even begin to comprehend this. Unix-based computing is already available and embraced by the general public. It's called "Macintosh", because Apple fully comprehends this, and did something about it.
The phone will have text entry modes: Abc, ABC, 123, but will not have the "smart words" input completion feature. Andriod developers can not learn to use contractions...
Well, I'm disillusioned.
Increasing the amount of time that someone can be held is only vaguely related to demanding the disclosure of encryption keys. (The vague relation is a very dangerous one -- people's freedom and the encroachment of police state.) If the Government wants to lump these things together, it's an obvious power ploy and a bad idea. If either of these ideas (incarceration time, encryption) actually has merit, perhaps they should be debated seperately? Would that strategy be better, or worse, in terms of the gradual erosion of freedom. Super-laws that eliminate the normal protections of the people's rights under law are by definition a poor idea, no matter how pragmatic they seem.
-
When the public electrical power system is compromised, most hams will be unable to communicate, because all they have is a little handy talkie that can barely reach the repeater base stations (which are themselves offline). Sure, some repeaters have battery backups. But so does the public telephone network and the Internet. And what volume of traffic and kind of traffic do you suppose the hams will be transmitting, even if the repeater facilities are operational?
- If one is suggesting that ham radio is critical for the authorities to communicate, wouldn't it be better to rework the official emergency infrastructure so that it doesn't go to pot in the event of the kind of emergencies that it's intended to respond to? (Is it really, in fact, so vulnerable, that the officials will all be helpless, wandering around and hoping that they encounter some amateur radio hobbyist who can save the day?)
Maybe the reality is that ham radio isn't a significant part of the official emergency infrastructure, but only augments it. Primarily, for private messages. In that case, what sense does Morse Code make? I am sure we can make rugged (even EMP resistant) computer-radio systems that work reliably in poor reception conditions and so on. Why not allow anybody at all to access the citizen-use emergency frequencies using a modern encoding system such as voice or data packets, and without needing any license?On the other hand, if ham radio is just for education and entertainment, why not limit it to exclusively Morse Code transmissions? Or at least that the operators only use "experimental" home-built equipment (which used to be tubes, but nowadays might be mostly-software kits)? These hobby hands would be for licensed hobbyists and expiermenters, not for people who are walking around with what amount to low-quality party-line mobile telephones they bought at the store.
My comments above were written with the USA in mind, not third-world or wilderness countries that don't have modern communications systems everywhere. But those places have perfectly good satellite phones, anyway. Maybe the civil defense authorities everywhere should issue some kind of satellite phone, so there's always going to be one nearby (in your village, out on your ranch, or on your suburban block, and in someone's nearby vehicle) for when you really need it. And there could be regional emergency telephone centers that switch the calls, point-to-point, from the various satellite, land-line, and mobile calls. These distributed emergency centers could be made to withstand any emergency situation. Of course, the message capacity would be very limited, but not so limited as the ham radio system.