"What do they see as the causes of America's very high (relative to other "first world" or developed nations) rates of violence, and what do they propose to do about it?"
Sure, I'll bite. Uh, how about
High rates of poverty and near-poverty by "first-world" standards?
School systems that are often operated like understaffed prisons, inhibiting socialization and acculturating children to violence.
High rates of incarceration due to Draconic drug laws, making violent criminals out of non-violent ones.
The almost total destruction of the public mental health care system?
The low rate of two-parent families with a full-time parent caregiver?
The relatively widespread and easy availability of weapons of all types?
I could probably think of other likely causes, but I'm tired of typing.
How much of each of the above contribute? I don't know, and I don't think anyone does. How would we fix them? In most cases, it would require significant social investment of money and effort.
What I'd be willing to bet on, though, is that effects of fixing any one of the above problems would dwarf any positive effect of banning violent video games...
I don't think WYSIWYG means what you think it means.
If Word paginates differently on the screen than the printer, then it ain't WYSIWYG. We can rhen argue about whether WYSIWYG is even a good idea, but at any rate it appears Word doesn't have it.
Uh, 1440 two-up duplex letter-size pages per minute. That's 5760 letter-size page impressions per minute! And keep in mind that a lot of print-on-demand books (the target market, presumably) are smaller than letter-size, and few are larger.
Xylene, friend. Sold in US paint-supply stores as "Goof Off". A bit toxic, highly carcinogenic (cancer-causing); wear chemical-resistant rubber gloves and wash everything quite thoroughly with dish soap and warm water afterward. I just re-watched "Who Killed Roger Rabbit", and they can say what they want about the composition of "The Dip": it was xylene.
Xylene will take the paint off your keycaps no prob, though. I just tried it on an old keyboard's "Caps Lock" key. You have to get through a thick layer of lacquer on a good-quality keyboard; it took me about 2 minutes of rewetting and rubbing with a paper towel. The label is gone now.
Actually, I meant "Clark-Hanlon".:-) I've always believed, as do others that "Hanlon's Razor" is actually due to Robert Heinlein. So it made sense to me to mis-spell Arthur Clarke's name, too, for consistency.:-)
Yes. You have to be a little careful of odd/even distances, and because the proper metric is not Euclidean distance but whichever is longer of the x and y axis distances, and of potential non-intersection (consider e.g. two stones dropped in the same location at different times), but your basic idea is sound. Anyone who tried to simulate to solve this missed the spirit of these competitions---they're really about mathematical modeling more than computer programming.
IANAL, but I believe a relicense would allow the copyright holder to distribute the binaries of new versions without distributing the changed source for those versions. You can't retroactively withdraw the GPL from code you've published, but you can quit using the GPL for code you hold the copyright to.
Don't ever sign your copyrights away unless you really seriously trust the assignee, or unless you are well-paid in return.
"Can anyone explain to me, clearly and convincingly, exactly how the average joe office worker's life benefits from the capabilities of Excel in 2005 versus Lotus 123 in, say, 1990, excluding Y2000 fixes, speed and memory?"
To pick one example, pivot tables. Pivot tables enable Excel to do a form of OLAP that a lot of businesses are getting a lot of mileage out of.
Note that I'm not a big fan of Excel, or indeed of spreadsheets in general. But I'm not sure the form of the critique given above is fair. Once VisiCalc established the basic spreadsheet idea, the rest almost necessarily consisted of minor improvements. So what? It's hardly like spreadsheets are the only financial tool commonly used by office workers in 2005---a very different situation from 1990. Modern expert-system tax preparation, for example, has revolutionized some classes of office work. There's innovation still going on; it's evident in VisiCalc-like groundbreaking products for the most part, though; not so much in feature enhancements.
I have 3-5 of them as a students. I'll see if they're interested. I have the skillset you're looking for, but I'm not willing to move to Boston and I doubt you'd want to pay what I'm worth.
There's nothing braindead about m4. I've used it as a metalanguage tool to design 3 or 4 highly successful little application-specific languages. See, for example, the first version of what eventually became XML-XCB.
What's braindead is csh---I wouldn't use it as a scripting language on my most desperate day. Horrible syntax and semantics, a variety of error-prone features, and a really buggy reference implementation (last I checked).
If only we could have it both ways. If only there was some way that folks that needed stability could have some kind of stable Linux kernel, while folks who wanted to experiment with the latest and greatest features could have some kind of experimental kernel.
Perhaps we could use some kind of numbering scheme that separated the two; for example using "odd" version numbers like 2.7 for the experimental kernel series and even numbers like 2.6 or 2.8 for the stable series. One might imagine that periodically the matured changes in the experimental series could be merged back into the stable series, starting new series for both stable and experimental.
Maybe Linux should have instituted a process like this years ago. Then they'd have some experience with it by now, and could have it running smoothly instead of messing around with new development processes like they currently are.
Folks' ideas of what a (good) CS Ph.D. is are really cracking me up, and are at the heart of why they are IMHO undervalued by industry. Have any of the respondents actually worked with a group of Ph.D.-level CS folks for any length of time? It sure doesn't sound like it. I've worked in almost every computing context you can imagine, and the Ph.D. research groups are the scary-best bunch of them all.
It's certainly true that Ph.D.s often require care and feeding, and are sometimes a lousy fit for mundane programming tasks. But those tasks aren't what is driving Google's new ventures: it's innovative ideas, conceived and implemented by a team headed by incredibly clueful people. This is what Ph.D. researchers are skilled and trained to do.
I recently consulted for a growing company that had been stagnating in bringing a tech app online for several years. I came in and in about 4 months solved a bunch of business and organizational problems, rewrote a bunch of their code to be 10-100x more efficient yet much simpler, and got them on the road to profitability. I don't think those folks will be saying anything bad about Ph.D.s anytime soon.
I'd miss Mad Magazine. Not that I read it anymore, but my kid might soon. Three months ago I would have missed MapQuest, but (I think this is on-topic) I just switched over to Google Maps for most everything. Shrug.
"Google offers nothing that cannot be cloned by MSN or Yahoo within a year."
Wrong. They own a pile of really skilled Ph.D.-level employees, and are hiring them at a rate unprecedented even during the glory days of Bell Labs and IBM. Google has bet the farm on the idea that putting some of the nation's smartest people in a productive work environment will make the company money. Essentially, they've taken a long position in Ph.D. futures. So far, the gamble seems to have paid off. Google has been launching services with big upside potential and low risk at an incredible clip, and the market has rewarded them for it.
I've been saying for 15 years that CS Ph.Ds are massively undervalued by the market. (Disclaimer: I have one now.) I personally think Google is about to demonstrate the truth of this proposition. But win or lose, at least it's an interesting business model.
"If HF, the natural limitations of the frequency mean that you need a HUMONGOUS 'chunk of spectrum' for a high-speed transmission. Wider than most modern transmitters can do, reasonably."
I've got a transmitter that will do 30MHz. Right now. 60 if I go quadrature. 12 bits effective. How much do you want?
"Not trying to discourage you but I saw your posting about a $600 cost to manufacture your dream device today -- so where is it?"
Click on the link already. Here's a direct link to where you can order from Ettus Research. I have 6 of them right now and a bunch of students playing with them. They work great.
"Somehow I missed clicking on your link, but I assume it's a software-defined radio kit. Seen 'em... most have horrible sensitivity and can't meet basic 3rd order harmonic specs to keep from interfereing with things around them without add-on filters, yet. You end up eating massive amounts of DSP horsepower to chew through the garbage that the non-selective receiver hears, and putting out spurious and cruddy signals on the transmitter unless you add filtering or keep the power levels very low."
You've given a great critique of a hypothetical device I might have owned. The actual Ettus Research device has great specs. Because there's enough bandwidth to massively oversample in normal applications, there's no need for fancy analog filtering. Because there's a current-generation FPGA on board, transmission and reception don't need to load the CPU unreasonably.
This is not 80s spread-spectrum. Computers and electronics have advanced a bit since then, as witnessed by 802.11, CDMA, etc. There are already hundreds, if not thousands, of real experimenters playing with the Ettus device. It currently works as an HDTV receiver, and my students are well along in a variety of other applications as well.
Anyway, thanks for giving me something to investigate as far as current regs in your other post. I'll ask around.
Just for the record, I want to do Viterbi/Trellis Coding in some constellation in frequency/amplitude space and/or frequency/phase space. This is the modem thing, and seems to be pretty high bandwidth from what I can tell.
I think the problem here is that I'm severely limited in bitrate in most of the bands, independent of the efficiency of the encoding. Am I wrong about this?
As usual, these arguments always end with a lot of agreement. Still...
I hope you followed the link in my grandparent post. It's an open-hardware open-software device that can easily accomodate any standard analog modulation scheme and a huge range of digital ones. It's reasonably priced, and you can easily mess with it yourself. This is, or at least should be, the future of amateur radio. If it's available $600 in onesies with small-shop production today, it's going to be $60 once these things hit the streets. You can't keep folks out of the ham bands at that point. Instead, give them a reasonable world to come into.
If you want to go halfway around the world on tiny power, digital is the only way to go. A good spread-spectrum modulator with plenty of FEC will beat the stuffings out of current tech with 1/10th the power in the same bandwidth. As others have pointed out, it also dominates CW. Using multiple amplitudes, phases, and frequencies just works a lot better than CW modulation in the same spectrum.
I think you're a little confused about how digital amateur radio would work. Think 802.11 ad hoc mode: multiple voice and data channels TDMed and CDMed into a single chunk of spectrum. "Sweeping the dial" is just sweeping a source address filter in this world. Slow switchover would of course be necessary; the good news is that allocating just a small sub-band in each ham band would be a great way to get started and wouldn't impact existing operations much.
Limiting the use of any part of the spectrum to radios that can be built "made by the user" seems wrong. (What does this mean, anyway? It isn't like folks are building their own tubes (mostly) or growing their own transistors. By any reasonable definition I can think of, our applications software atop the USRP we're using for a lot of work constitutes "making a radio", even though there's no soldering at all.)
Right now, most of our transmission work is in the "amateur 802.11" band down at channel 1. Being able to do 802.11-style DSSS at reasonable power levels is quite nice, but we'd like to do exotic modulation and work on bands with better propagation. Right now we can't.
There are undoubtedly parts of the country where a few of the bands are crowded. In Oregon, they're hardly overutilized. In any case, digital modulation uses the spectrum better, and should help relieve whatever overcrowding exists.
As to the relatively silly and pointless tasks you mention, yes. As I said earlier, I could certainly do both of them. I don't see why I should be required to, though. Who is it helping?
In 2005 you can chatter with your friends in Istanbul or Sao Paolo by voice quite nicely, through the twin miracles of cheap long distance cellphones and the Internet. Let the amateur bands be used for more interesting things.
"On the other hand, maybe with the development of the digital technologies, the analog radio technology potential members are just not bothering looking into it."
Ya think?
I do a bunch of radio-related research, and hold a Novice license. I have easily passed the technical portion of Tech Plus and General practice tests. However, I haven't the time to devote to learning Morse, and I haven't the slightest inclination to memorize a bunch of frequency bands that are readily available in tabular form. As a result, I must rely on my colleagues with more time and energy for key portions of my work. This is a pain for them, and accomplishes nothing positive for society as far as I can see.
I also can't use reasonable (digital) modulation schemes in any amateur band. Sending high-speed data would be really nice, and sending voice as digital data is way more spectrum-efficient than any allowed analog modulation method, but no...
By all means, let's get rid of the Morse requirement, and change the test to cover more meaningful material. Let's make room for reasonable digital amateur transmissions. Either that, or give the valuable and currently mostly-dead amateur bands to someone who will make more sensible use of them.
"We can eliminate a huge portion of the search space by only searching passwords of length x"
Nope! This is the beauty of iterative deepening. The number of possible (7-bit) passwords of length x is 128**x. The total number of possible passwords of length less than x is also 128**x. So, if you just search passwords in order of increasing length, it will cost you at most twice as much to search all passwords of length less than or equal to x as to just search passwords of length equal to x.
The vulnerability is just that you can watch ssh connections going by on the network and find accounts with short passwords.
You and Charlie have my sincere admiration both for your handling of your literary work and for your friendly and open participation in the web culture. You're also both excellent writers.:-)
I'm really surprised, BTW, that professional publishering houses don't own print-on-demand systems (e.g. InstaBook) just for producing ARCs. AFAIK these things have plenty of capacity and are cheap enough to operate that you should be able to give out high-quality facsimiles as needed...
"In this respect, UNIX/Linux can be considered more secure than Windows, but only until a competent administrator restricts local users to non-admin-equivalent accounts. Then things rapidly return to something amazingly close to equality."
But a competent Windows administrator can't and won't restrict local users to non-admin-equivalent accounts. While this setup may be amazingly more secure than the Windows default, it is also amazingly less functional. Many Windows programs, including some from Microsoft, require administrative privileges to operate. One of the truly important differences in the UNIX/Linux world is that programs are written to operate with limited privilege. If Microsoft (very hypothetically) were to crack down somehow and insist Windows apps do the same, then the foundation would be in place for real Windows security improvement.
Slashdot Mirror
"What do they see as the causes of America's very high (relative to other "first world" or developed nations) rates of violence, and what do they propose to do about it?"
Sure, I'll bite. Uh, how about
- High rates of poverty and near-poverty by "first-world" standards?
- School systems that are often operated like understaffed prisons, inhibiting socialization and acculturating children to violence.
- High rates of incarceration due to Draconic drug laws, making violent criminals out of non-violent ones.
- The almost total destruction of the public mental health care system?
- The low rate of two-parent families with a full-time parent caregiver?
- The relatively widespread and easy availability of weapons of all types?
I could probably think of other likely causes, but I'm tired of typing.How much of each of the above contribute? I don't know, and I don't think anyone does. How would we fix them? In most cases, it would require significant social investment of money and effort.
What I'd be willing to bet on, though, is that effects of fixing any one of the above problems would dwarf any positive effect of banning violent video games...
I don't think WYSIWYG means what you think it means.
If Word paginates differently on the screen than the printer, then it ain't WYSIWYG. We can rhen argue about whether WYSIWYG is even a good idea, but at any rate it appears Word doesn't have it.
Uh, 1440 two-up duplex letter-size pages per minute. That's 5760 letter-size page impressions per minute! And keep in mind that a lot of print-on-demand books (the target market, presumably) are smaller than letter-size, and few are larger.
This is a scary fast printer.
Xylene, friend. Sold in US paint-supply stores as "Goof Off". A bit toxic, highly carcinogenic (cancer-causing); wear chemical-resistant rubber gloves and wash everything quite thoroughly with dish soap and warm water afterward. I just re-watched "Who Killed Roger Rabbit", and they can say what they want about the composition of "The Dip": it was xylene.
Xylene will take the paint off your keycaps no prob, though. I just tried it on an old keyboard's "Caps Lock" key. You have to get through a thick layer of lacquer on a good-quality keyboard; it took me about 2 minutes of rewetting and rubbing with a paper towel. The label is gone now.
Actually, I meant "Clark-Hanlon". :-) I've always believed, as do others that "Hanlon's Razor" is actually due to Robert Heinlein. So it made sense to me to mis-spell Arthur Clarke's name, too, for consistency. :-)
Yes. You have to be a little careful of odd/even distances, and because the proper metric is not Euclidean distance but whichever is longer of the x and y axis distances, and of potential non-intersection (consider e.g. two stones dropped in the same location at different times), but your basic idea is sound. Anyone who tried to simulate to solve this missed the spirit of these competitions---they're really about mathematical modeling more than computer programming.
Cool. It's the "Clark-Hanlon Corollary". You shouldn't have posted AC---I'd love to have added your name to it when quoting it in the future.
IANAL, but I believe a relicense would allow the copyright holder to distribute the binaries of new versions without distributing the changed source for those versions. You can't retroactively withdraw the GPL from code you've published, but you can quit using the GPL for code you hold the copyright to.
Don't ever sign your copyrights away unless you really seriously trust the assignee, or unless you are well-paid in return.
"Can anyone explain to me, clearly and convincingly, exactly how the average joe office worker's life benefits from the capabilities of Excel in 2005 versus Lotus 123 in, say, 1990, excluding Y2000 fixes, speed and memory?"
To pick one example, pivot tables. Pivot tables enable Excel to do a form of OLAP that a lot of businesses are getting a lot of mileage out of.
Note that I'm not a big fan of Excel, or indeed of spreadsheets in general. But I'm not sure the form of the critique given above is fair. Once VisiCalc established the basic spreadsheet idea, the rest almost necessarily consisted of minor improvements. So what? It's hardly like spreadsheets are the only financial tool commonly used by office workers in 2005---a very different situation from 1990. Modern expert-system tax preparation, for example, has revolutionized some classes of office work. There's innovation still going on; it's evident in VisiCalc-like groundbreaking products for the most part, though; not so much in feature enhancements.
I have 3-5 of them as a students. I'll see if they're interested. I have the skillset you're looking for, but I'm not willing to move to Boston and I doubt you'd want to pay what I'm worth.
There's nothing braindead about m4. I've used it as a metalanguage tool to design 3 or 4 highly successful little application-specific languages. See, for example, the first version of what eventually became XML-XCB.
What's braindead is csh---I wouldn't use it as a scripting language on my most desperate day. Horrible syntax and semantics, a variety of error-prone features, and a really buggy reference implementation (last I checked).
"Does nobody edit these submissions?"
You're new here, aren't you.
Maybe the mod just thought there was a kernel of truth there somewhere. Not sure they were right. :-)
I started with Linux at 1.2.13, doing professional consulting. I guess that makes me a newbie also. :-) Oh for the good ol' days.
If only we could have it both ways. If only there was some way that folks that needed stability could have some kind of stable Linux kernel, while folks who wanted to experiment with the latest and greatest features could have some kind of experimental kernel.
Perhaps we could use some kind of numbering scheme that separated the two; for example using "odd" version numbers like 2.7 for the experimental kernel series and even numbers like 2.6 or 2.8 for the stable series. One might imagine that periodically the matured changes in the experimental series could be merged back into the stable series, starting new series for both stable and experimental.
Maybe Linux should have instituted a process like this years ago. Then they'd have some experience with it by now, and could have it running smoothly instead of messing around with new development processes like they currently are.
Oh well. Just a crazy dream I had.
I first heard about this at about 11:30 EDT on NPR. By now, Spaceflight Now has a preliminary discussion.
I hope Stephen King's OK.
Folks' ideas of what a (good) CS Ph.D. is are really cracking me up, and are at the heart of why they are IMHO undervalued by industry. Have any of the respondents actually worked with a group of Ph.D.-level CS folks for any length of time? It sure doesn't sound like it. I've worked in almost every computing context you can imagine, and the Ph.D. research groups are the scary-best bunch of them all.
It's certainly true that Ph.D.s often require care and feeding, and are sometimes a lousy fit for mundane programming tasks. But those tasks aren't what is driving Google's new ventures: it's innovative ideas, conceived and implemented by a team headed by incredibly clueful people. This is what Ph.D. researchers are skilled and trained to do.
I recently consulted for a growing company that had been stagnating in bringing a tech app online for several years. I came in and in about 4 months solved a bunch of business and organizational problems, rewrote a bunch of their code to be 10-100x more efficient yet much simpler, and got them on the road to profitability. I don't think those folks will be saying anything bad about Ph.D.s anytime soon.
I'd miss Mad Magazine. Not that I read it anymore, but my kid might soon. Three months ago I would have missed MapQuest, but (I think this is on-topic) I just switched over to Google Maps for most everything. Shrug.
"Google offers nothing that cannot be cloned by MSN or Yahoo within a year."
Wrong. They own a pile of really skilled Ph.D.-level employees, and are hiring them at a rate unprecedented even during the glory days of Bell Labs and IBM. Google has bet the farm on the idea that putting some of the nation's smartest people in a productive work environment will make the company money. Essentially, they've taken a long position in Ph.D. futures. So far, the gamble seems to have paid off. Google has been launching services with big upside potential and low risk at an incredible clip, and the market has rewarded them for it.
I've been saying for 15 years that CS Ph.Ds are massively undervalued by the market. (Disclaimer: I have one now.) I personally think Google is about to demonstrate the truth of this proposition. But win or lose, at least it's an interesting business model.
"If HF, the natural limitations of the frequency mean that you need a HUMONGOUS 'chunk of spectrum' for a high-speed transmission. Wider than most modern transmitters can do, reasonably."
I've got a transmitter that will do 30MHz. Right now. 60 if I go quadrature. 12 bits effective. How much do you want?
"Not trying to discourage you but I saw your posting about a $600 cost to manufacture your dream device today -- so where is it?"
Click on the link already. Here's a direct link to where you can order from Ettus Research. I have 6 of them right now and a bunch of students playing with them. They work great.
"Somehow I missed clicking on your link, but I assume it's a software-defined radio kit. Seen 'em... most have horrible sensitivity and can't meet basic 3rd order harmonic specs to keep from interfereing with things around them without add-on filters, yet. You end up eating massive amounts of DSP horsepower to chew through the garbage that the non-selective receiver hears, and putting out spurious and cruddy signals on the transmitter unless you add filtering or keep the power levels very low."
You've given a great critique of a hypothetical device I might have owned. The actual Ettus Research device has great specs. Because there's enough bandwidth to massively oversample in normal applications, there's no need for fancy analog filtering. Because there's a current-generation FPGA on board, transmission and reception don't need to load the CPU unreasonably.
This is not 80s spread-spectrum. Computers and electronics have advanced a bit since then, as witnessed by 802.11, CDMA, etc. There are already hundreds, if not thousands, of real experimenters playing with the Ettus device. It currently works as an HDTV receiver, and my students are well along in a variety of other applications as well.
Anyway, thanks for giving me something to investigate as far as current regs in your other post. I'll ask around.
Just for the record, I want to do Viterbi/Trellis Coding in some constellation in frequency/amplitude space and/or frequency/phase space. This is the modem thing, and seems to be pretty high bandwidth from what I can tell.
I think the problem here is that I'm severely limited in bitrate in most of the bands, independent of the efficiency of the encoding. Am I wrong about this?
As usual, these arguments always end with a lot of agreement. Still...
I hope you followed the link in my grandparent post. It's an open-hardware open-software device that can easily accomodate any standard analog modulation scheme and a huge range of digital ones. It's reasonably priced, and you can easily mess with it yourself. This is, or at least should be, the future of amateur radio. If it's available $600 in onesies with small-shop production today, it's going to be $60 once these things hit the streets. You can't keep folks out of the ham bands at that point. Instead, give them a reasonable world to come into.
If you want to go halfway around the world on tiny power, digital is the only way to go. A good spread-spectrum modulator with plenty of FEC will beat the stuffings out of current tech with 1/10th the power in the same bandwidth. As others have pointed out, it also dominates CW. Using multiple amplitudes, phases, and frequencies just works a lot better than CW modulation in the same spectrum.
I think you're a little confused about how digital amateur radio would work. Think 802.11 ad hoc mode: multiple voice and data channels TDMed and CDMed into a single chunk of spectrum. "Sweeping the dial" is just sweeping a source address filter in this world. Slow switchover would of course be necessary; the good news is that allocating just a small sub-band in each ham band would be a great way to get started and wouldn't impact existing operations much.
Limiting the use of any part of the spectrum to radios that can be built "made by the user" seems wrong. (What does this mean, anyway? It isn't like folks are building their own tubes (mostly) or growing their own transistors. By any reasonable definition I can think of, our applications software atop the USRP we're using for a lot of work constitutes "making a radio", even though there's no soldering at all.)
Right now, most of our transmission work is in the "amateur 802.11" band down at channel 1. Being able to do 802.11-style DSSS at reasonable power levels is quite nice, but we'd like to do exotic modulation and work on bands with better propagation. Right now we can't.
There are undoubtedly parts of the country where a few of the bands are crowded. In Oregon, they're hardly overutilized. In any case, digital modulation uses the spectrum better, and should help relieve whatever overcrowding exists.
As to the relatively silly and pointless tasks you mention, yes. As I said earlier, I could certainly do both of them. I don't see why I should be required to, though. Who is it helping?
In 2005 you can chatter with your friends in Istanbul or Sao Paolo by voice quite nicely, through the twin miracles of cheap long distance cellphones and the Internet. Let the amateur bands be used for more interesting things.
"On the other hand, maybe with the development of the digital technologies, the analog radio technology potential members are just not bothering looking into it."
Ya think?
I do a bunch of radio-related research, and hold a Novice license. I have easily passed the technical portion of Tech Plus and General practice tests. However, I haven't the time to devote to learning Morse, and I haven't the slightest inclination to memorize a bunch of frequency bands that are readily available in tabular form. As a result, I must rely on my colleagues with more time and energy for key portions of my work. This is a pain for them, and accomplishes nothing positive for society as far as I can see.
I also can't use reasonable (digital) modulation schemes in any amateur band. Sending high-speed data would be really nice, and sending voice as digital data is way more spectrum-efficient than any allowed analog modulation method, but no...
By all means, let's get rid of the Morse requirement, and change the test to cover more meaningful material. Let's make room for reasonable digital amateur transmissions. Either that, or give the valuable and currently mostly-dead amateur bands to someone who will make more sensible use of them.
"We can eliminate a huge portion of the search space by only searching passwords of length x"
Nope! This is the beauty of iterative deepening. The number of possible (7-bit) passwords of length x is 128**x. The total number of possible passwords of length less than x is also 128**x. So, if you just search passwords in order of increasing length, it will cost you at most twice as much to search all passwords of length less than or equal to x as to just search passwords of length equal to x.
The vulnerability is just that you can watch ssh connections going by on the network and find accounts with short passwords.
You and Charlie have my sincere admiration both for your handling of your literary work and for your friendly and open participation in the web culture. You're also both excellent writers. :-)
I'm really surprised, BTW, that professional publishering houses don't own print-on-demand systems (e.g. InstaBook) just for producing ARCs. AFAIK these things have plenty of capacity and are cheap enough to operate that you should be able to give out high-quality facsimiles as needed...
"In this respect, UNIX/Linux can be considered more secure than Windows, but only until a competent administrator restricts local users to non-admin-equivalent accounts. Then things rapidly return to something amazingly close to equality."
But a competent Windows administrator can't and won't restrict local users to non-admin-equivalent accounts. While this setup may be amazingly more secure than the Windows default, it is also amazingly less functional. Many Windows programs, including some from Microsoft, require administrative privileges to operate. One of the truly important differences in the UNIX/Linux world is that programs are written to operate with limited privilege. If Microsoft (very hypothetically) were to crack down somehow and insist Windows apps do the same, then the foundation would be in place for real Windows security improvement.