That makes a lot more sense. In Omaha, you get four cards instead of two, and you have to use two cards from your set of four to make your final hand--whereas you can use all 5 from the table on Hold'Em.
Because of these differences, it's very easier and more frequent to find yourself in a position where you know you have "the nuts", the best possible hand, in an Omaha game. Whereas human players often miss some improbable hand (like thinking they're going to win because they have an ace-high flush, where a straight flush is possible) that's actually fairly likely in an Omaha game. I wouldn't be surprised at all to find that a bot that just waited until it had the nuts and bet like crazy then could wipe some of the idiots that play on-line.
Quite true that the Emperor may have been telling what he thought to be the truth. But: how would Vader know that? Confronted with the fact that the child he was told died is alive, which would you expect him to conclude: that the master of Sith made an honest mistake, or that he lied to Vader to set him off and make him a better tool? It's not like he's going to visit his evil master and ask, "hey, dude, when you told me my wife died before our kids were born, were you sure about that? Cause I think you made a mistake."
The endings of both tESB and RotJ had a new spin for me after watching Sith recently. By the end of Empire, Vader has figured out that the Emperor lied to him and his son was alive all along. What does he do? Try to get Luke's help to muscle the old man out and run things himself. Now it comes off not just as greed for more power, but as hoping for revenge for being lied to and otherwise manipulated.
Similary, the end of Jedi seems totally different to me now. Vader realizes his hopes of taking over as the new Emperor aren't going to happen because Luke just isn't good enough to pull it off, and he cracks. He gets pissed off and the old reflexes to kill the person responsible kick in, so he offs Palpatine in a rage the same way he routinely used to kill people in his younger days. It's not to atone for his sins or even to save his son. He's been getting pushed around by this guy for twenty years, looking for a weak moment to off him, and finally he gives up hope on a better ending and just finishes him off while there's a good window to do it.
I've also been very happy with the Verizon service and have used it extensively in both Baltimore and New York City when I get 60KB/s+ when connected to their EVDO network.
Only one real issue. Under Windows, the Verizon software loads a Venturi compression client by default. I had endless problems with the connection being unresponsible, especially when using SSH, when this software was running. Once I disabled it, smooth sailing since then.
I've had success using the appropriate matching Stowaway keyboard with various Palm devices. It folds up into pocket size but is quite usable even for moderate touch-typing speeds when unfolded.
In addition to the IIIe, there are a ton of cheap Stowaway keyboards for the many cheap and unwanted Handspring Visor models, most (all?) of which support USB (the IIIe is going to be serial without some sort of aftermarket upgrade). This is increasingly important as many laptops in particular are shipping nowadays without serial ports. I doubt you could put together a solution much cheaper than this while still having USB and a useful client for both Windows and Linux.
Of course it's stolen from 1984: now Microsoft has stooped to copying even Apple's advertising.
Actually, the real motivation here is that Microsoft doesn't want anyone else going around stealing other people's ideas because they'd like to have a monopoly on that, too.
I was in a doctor's office the other day and the doctor was putting all her notes into a desktiop PC in the exam room. It didn't sound right; I put my head closer to hear better, then told her it was a Western Digital drive, the bearings on it were starting to go bad, and it would likely last at most a few more months before failing. She looked at me like I was crazy. After I showed her how to use the Control Panel so she could confirm I was right about the brand, I had considerably more credibility with that prediction. She made sure all the data on that system was backed up.
Anyway, you answered your own question here. Don't buy that kind of Western Digital drives if you care about the noise. Any of the newer drives with fluid bearings will be much quieter. Maxtor's drives have them (for the few months the drive lasts, at least). Many of Seagates drives are even quieter, but they're not quite as fast as the Maxtor units. Even Western Digital is supposedly has their "Fluid Dynamic Bearing" models out now.
Couldn't resist doing the math. By the day the six-month lock-up was over, Raymond's stock was down to $45/share. If he did wait until June as he suggested he might, it could have been worth as little as $31/share. Still, that's at least 1/2 a million--if he went through with selling it.
As of today, 150,000 shares of VA are only worth about $200K. Quite a fall from $36M.
Re:Hey! buy the Jakarta Commons Cookbook
on
Apache Jakarta Commons
·
· Score: 5, Informative
Amazingly, this bit of self-promotion is actually more useful than the review. Tim O'Brien's book is my favorite on this subject and it's far, far better than the Will Iverson title. I particularly like how well it covers some of the extremely useful but not well documented APIs like the HttpClient, there are far more readable, usable examples that cover the kind of real-world issues you run into the minute you try to use that API than any other source I know of.
Larne Pekowsky: "Apache Jakarta and Beyond: A Java Programmer's Introduction" is also a decent title in this area, albeit a bit too broad to really do that good of a job on any topic. Even a 600 page book can't cover things like Eclipse or Ant in a chapter.
I have all three books on my Safari bookshelf right now; only O'Brien's is staying once I can remove the other two. The Iverson and Pekowsky titles have some interesting sections worth looking at once, while "Jakarta Commons Cookbook" I always want to keep around for reference, in the same fashion I already rely on the similarly useful "Java Cookbook".
They certainly were called Tiger Software before, which gives them even more reason to be ticked.
> NO ONE thinks of TigerDirect when the word Tiger is mentioned among Mac afficianados.
Uh, isn't that the point of the lawsuit--the fact that their name brand recognition is being destroyed here is exactly why they're suing.
While I hate all of this class of lawsuits, I find the knee-jerk comments people are making about this one reek of a double standard. It's only because people like Apple and (rightfully) are disgusted by Tiger Direct, the smaller company here, that the comments are playing out the way they are. Change the big and little guys here and community reaction would be totally different. I offer the following thought experiment: imagine what would be happening right now if instead it was Microsoft coming out with the new product, and they picked a standard English word like, say, "Snort" for it. Would the same people coming to Apple's defense and damning Tiger Direct be consistant by suggesting that it was OK for Microsoft to do that, because after all it's just a regular word nobody can own, right?
> Even if you search for "Tiger", you still get Tiger Direct before Apple.
Today. Six months from now, when there are links all over the web to pages about Tiger, and amazon.com associates all over the web are pointing to books like the inevitable "Mac OS X Tiger in a Nutshell", things will be different.
Right now, if you search for "panther" with Google, links #1 and #2 both point to pages within apple.com/macos/ Want to make a bet what position Tiger Direct is in once Google notices those pages have changed?
On the Comparison Matrix page that ends the article, look at the line for "USB extension cable". Those devices come with a short USB cable that solves the problem of the drive itself being too fat to plug into a crowded USB port.
I've been using the same Verizon broadband service as is described in the article for about six months now. When connecting to their EVDO network, I get a consistant raw rate of 500Kb/s on downloads and around 100Kb/s up. This is in the suburbs about 15 miles from the center of the coverage area I live near (Baltimore). While I don't have any firm numbers, the network seems even faster when I'm nearer to the center of their area, like when in NYC.
Note that I saw raw rate here because I'm not using the compression package Verizon provided. While it seemed to slightly improve performance when browsing the web, I found that something it was messing with made SSH sessions much less reliable. I turned it off the first day and haven't missed it.
While there are occasional hiccups that cause me to lose the connection (which are no more frequent than when I'm connected with 802.11b; screen is your friend here), I can assure you that when in the EVDO sections the SSH sessions I launch every day work perfectly even if I have two or three computers sharing the network connection via NAT.
> I don't know where you got the impression that the government pays your debts in bankruptcy protection.
Probably from the fact that the US government has help bailout a few large organizations in the past when it was deemed important enough; airlines and banks are the usual beneficiaries.
In this case, spammer Scott Ricters's company is filing for Chapter 11 bankruptcy. The debts aren't being repaid by anyone or even written off yet. This provides them with some protection from creditors while the company is reorganized with some supervision by the courts. See http://www.sec.gov/investor/pubs/bankrupt.htm for more information.
It somehow just feels right that when I try to access most of the pages on the Murphy's Laws site right now, like following your link, they return an error:
"Murphy's error 404 law: The chance of receiving a 404 error page is in inverse proportion to your need to find the page"
P/E ratios are a poor way to compare stocks in new growth companies, as they don't account for the rate at which earnings growth is accelerating which is far more important than the current earnings amount. P/E looks back, not forward.
If you look at the earnings rating that Investor's Business Daily computes, GOOG gets a score of 99, the maximum possible. I quote from IBD founder Bill O'Neil's book "How To Make Money in Stocks" to point out show short-sighted P/E thinking is with companies like this, picking one example most here are familiar with.
"American Online sold for over 100 times earnings in November 1994 before increasing 14,900% from 1994 to its top in December 1999."
I've made plenty of money buying companies with a P/E of near 100 and selling as it hit 500+. That said, I still wouldn't buy GOOG, but it's more because the market cap extrapolated from the relatively small number of public shares seems insane.
Do you really want to be tilting a PowerBook around when you've only got one hand free?
Re:MENSA is not THAT smart..
on
MSN Sponsors Mensa
·
· Score: 3, Informative
> Among many options, one only needs 1300 (out of 1600) on SAT. These days, MANY people easily get 1300.
You don't get a qualifying Mensa reading comprehension score with this statement. You need a >1300 if you took test before 9/30/74, back when it was hard. They clearly state that they haven't considered the SAT to correlate with IQ for over ten years now; 1994 was the last year they accepted those scores.
Cosmic Encounter remains my all my time favorite board game for larger groups--it takes about 6 people to really be fun and you can easily play with over 10 if you have enough expansion parts.
The only similarity to Magic is that each player draws a card at the beginning of the game that determines which type of alien you are and therefore your special abilities. Your style of play totally changes based on which of the aliens you are, which is part of what keeps it so fresh for repeats: it's never quite the same dynamic.
> PC clocks are rather crappy and temperature sensitive
Line voltage sensitive, too. With the way newer processors throttle their speeds around based on temperature and loading, and the way fans change their parameters based on temperature, I have little hope for this technique nailing any new system.
Let's see, what were the authors using in the lab where they tested machine to machine variations?
"All the machines were Micron PCs with 448MHz Pentium II Processors". Right. From this, we get the grand statement shortly afterward "The current results strongly support our claim that modern processors have relatively stable clock skews". Uh, sorry guys, you didn't use a single modern processor for this section; just some obsolete ones that run so cool they don't have any CPU clock or temperature varation. There's not a machine to be found in their entire test that features the kind of design we seen in acutal modern processors.
> Set half of your neighbors to the lowest supported frequency, and the other half to the highest...Put yours in the middle.
This is the worst possible solution. Most wireless routers default to the middle channel, 6. This means that the ones you won't be able to hack into will conflict if you pick the middle frequency as the one for yourself. And any new units added to the network after your hacking spree, or anyone who resets their units to the factory default are also going to conflict with you.
For your private frequency, you need to use 1 or 11 (I find 11 works better, less conflict with microwaves or something, don't know why). Then remap all your neighbors to either 6 or the one you didn't use. If you figure out who is who using a tool like Netstumbler on a site survey of sorts first, you can often remap things such that your neighbors work better than they did before you started because they interfere with one another less. That's what I di....thought about doing before realizing it would be illegal.
You need to buy a CRT because nothing else you can buy can be expected to have the kind of longevity you're looking for. The long-term service records of DLP, LCD, and especially Plasma screens are either short or unknown. You'll pay considerably more than $300 over 20 years for new backlights on an LCD/DLP for example (expect to pay that much every 4000 hours of watching or so). And while typical plasma units are claimed to last 30,000 hours which comes out to over 20 years for typical playtimes, if you're on the unlucky part of the bell curve a failure inside the main plasma area can mean scrapping the entire unit. And as you correctly devine, Plasma prices do not reflect a reasonable price/performance ratio.
So you need another CRT, and a new Mitsubishi unit to replace your old one would be perfect--those are still solid units likely to last you a long time and their display's technical quality is best in class. A glance at their web site shows that a current HD-capable set (having an outboard tuner is a better idea anyway) at 55", the WS-55315, weighs 215 pounds. Price on that model runs around $1700. What you can get in any alterate display technology for $1700 is laughable in comparison quality, longevity, and servicability wise.
To read about the sort of things that are important to getting a good quality picture out of HDTV, I would recommend picking up any random issue of Widescreen Review magazine that features some CRT TV reviews to see what parameters they test on. The main thing that cheap sets (and any Plasma unit for that matter) really screw up are color linearity/accuracy and black level.
I'm going to address CRTs as far as good brands go, because if you're not getting a CRT you're prioritizing something other than image quality under normal viewing conditions. As such the non-CRT recommendation process becomes very specific to your priorities and it's hard to give a good answer. I'm not trying to be snotty here, because I certainly understand that it's often the case that display quality is not even close to the main decision parameter. For example, the last TV box I bought was a small projector, which I knew perfectly well wasn't as good as a CRT. But I was living on a 5th floor walk-up apartment and not about to haul a good TV up there when I had a short-term lease.
Anyway, Mitsubishi's high-end Diamond CRTs have the best factory calibration regime I'm aware of to make sure they are faithfully displaying their inputs, and their less expensive models are invariably at the top of the accuracy (and build quality) heap as well. Usually on the expensive side in get what you pay for fashion. At lower price ranges, Toshiba CRT sets usually give the best accuracy relative to their price. Some of the Sony sets look very good, but talk to any statistically significant number of people who have dealth with Sony repair centers and you'll never consider one of their products again.
As always with TVs, displays in showrooms are totally bogus unless you are verifying color temperature and brightness/contrast/sharpness yourself across sets. Most showrooms sets are too bright and too blue, because those are the characteristics that make people prefer a TV at first glance in the same way that louder stereo equipment always seems better at first.
Slashdot Mirror
That makes a lot more sense. In Omaha, you get four cards instead of two, and you have to use two cards from your set of four to make your final hand--whereas you can use all 5 from the table on Hold'Em.
Because of these differences, it's very easier and more frequent to find yourself in a position where you know you have "the nuts", the best possible hand, in an Omaha game. Whereas human players often miss some improbable hand (like thinking they're going to win because they have an ace-high flush, where a straight flush is possible) that's actually fairly likely in an Omaha game. I wouldn't be surprised at all to find that a bot that just waited until it had the nuts and bet like crazy then could wipe some of the idiots that play on-line.
Quite true that the Emperor may have been telling what he thought to be the truth. But: how would Vader know that? Confronted with the fact that the child he was told died is alive, which would you expect him to conclude: that the master of Sith made an honest mistake, or that he lied to Vader to set him off and make him a better tool?
It's not like he's going to visit his evil master and ask, "hey, dude, when you told me my wife died before our kids were born, were you sure about that? Cause I think you made a mistake."
The endings of both tESB and RotJ had a new spin for me after watching Sith recently. By the end of Empire, Vader has figured out that the Emperor lied to him and his son was alive all along. What does he do? Try to get Luke's help to muscle the old man out and run things himself. Now it comes off not just as greed for more power, but as hoping for revenge for being lied to and otherwise manipulated.
Similary, the end of Jedi seems totally different to me now. Vader realizes his hopes of taking over as the new Emperor aren't going to happen because Luke just isn't good enough to pull it off, and he cracks. He gets pissed off and the old reflexes to kill the person responsible kick in, so he offs Palpatine in a rage the same way he routinely used to kill people in his younger days. It's not to atone for his sins or even to save his son. He's been getting pushed around by this guy for twenty years, looking for a weak moment to off him, and finally he gives up hope on a better ending and just finishes him off while there's a good window to do it.
Anyone care to wager the correct order these events will happen in?
1) First mod-chip to bypass firmware limitations of Apple x86 hardware released
2) Linux distribution boots on new Apple-x86 hardware
3) Mac OS X for Intel boots on generic x86 hardware
4) Windows hacked to boot on new Apple x86 hardware
5) Mac OS X for Intel hacked to run in emulated virtual x86 machine
Tiebreaker question: estimate the date when OS X for x86 runs under Virtual PC on a G5 running the current OS X.
I've also been very happy with the Verizon service and have used it extensively in both Baltimore and New York City when I get 60KB/s+ when connected to their EVDO network.
Only one real issue. Under Windows, the Verizon software loads a Venturi compression client by default. I had endless problems with the connection being unresponsible, especially when using SSH, when this software was running. Once I disabled it, smooth sailing since then.
I've had success using the appropriate matching Stowaway keyboard with various Palm devices. It folds up into pocket size but is quite usable even for moderate touch-typing speeds when unfolded.
In addition to the IIIe, there are a ton of cheap Stowaway keyboards for the many cheap and unwanted Handspring Visor models, most (all?) of which support USB (the IIIe is going to be serial without some sort of aftermarket upgrade). This is increasingly important as many laptops in particular are shipping nowadays without serial ports. I doubt you could put together a solution much cheaper than this while still having USB and a useful client for both Windows and Linux.
Of course it's stolen from 1984: now Microsoft has stooped to copying even Apple's advertising.
Actually, the real motivation here is that Microsoft doesn't want anyone else going around stealing other people's ideas because they'd like to have a monopoly on that, too.
I was in a doctor's office the other day and the doctor was putting all her notes into a desktiop PC in the exam room. It didn't sound right; I put my head closer to hear better, then told her it was a Western Digital drive, the bearings on it were starting to go bad, and it would likely last at most a few more months before failing. She looked at me like I was crazy. After I showed her how to use the Control Panel so she could confirm I was right about the brand, I had considerably more credibility with that prediction. She made sure all the data on that system was backed up.
Anyway, you answered your own question here. Don't buy that kind of Western Digital drives if you care about the noise. Any of the newer drives with fluid bearings will be much quieter. Maxtor's drives have them (for the few months the drive lasts, at least). Many of Seagates drives are even quieter, but they're not quite as fast as the Maxtor units. Even Western Digital is supposedly has their "Fluid Dynamic Bearing" models out now.
Couldn't resist doing the math. By the day the six-month lock-up was over, Raymond's stock was down to $45/share. If he did wait until June as he suggested he might, it could have been worth as little as $31/share. Still, that's at least 1/2 a million--if he went through with selling it.
As of today, 150,000 shares of VA are only worth about $200K. Quite a fall from $36M.
Amazingly, this bit of self-promotion is actually more useful than the review. Tim O'Brien's book is my favorite on this subject and it's far, far better than the Will Iverson title. I particularly like how well it covers some of the extremely useful but not well documented APIs like the HttpClient, there are far more readable, usable examples that cover the kind of real-world issues you run into the minute you try to use that API than any other source I know of.
Larne Pekowsky: "Apache Jakarta and Beyond: A Java Programmer's Introduction" is also a decent title in this area, albeit a bit too broad to really do that good of a job on any topic. Even a 600 page book can't cover things like Eclipse or Ant in a chapter.
I have all three books on my Safari bookshelf right now; only O'Brien's is staying once I can remove the other two. The Iverson and Pekowsky titles have some interesting sections worth looking at once, while "Jakarta Commons Cookbook" I always want to keep around for reference, in the same fashion I already rely on the similarly useful "Java Cookbook".
They certainly were called Tiger Software before, which gives them even more reason to be ticked.
> NO ONE thinks of TigerDirect when the word Tiger is mentioned among Mac afficianados.
Uh, isn't that the point of the lawsuit--the fact that their name brand recognition is being destroyed here is exactly why they're suing.
While I hate all of this class of lawsuits, I find the knee-jerk comments people are making about this one reek of a double standard. It's only because people like Apple and (rightfully) are disgusted by Tiger Direct, the smaller company here, that the comments are playing out the way they are. Change the big and little guys here and community reaction would be totally different. I offer the following thought experiment: imagine what would be happening right now if instead it was Microsoft coming out with the new product, and they picked a standard English word like, say, "Snort" for it. Would the same people coming to Apple's defense and damning Tiger Direct be consistant by suggesting that it was OK for Microsoft to do that, because after all it's just a regular word nobody can own, right?
> Even if you search for "Tiger", you still get Tiger Direct before Apple.
Today. Six months from now, when there are links all over the web to pages about Tiger, and amazon.com associates all over the web are pointing to books like the inevitable "Mac OS X Tiger in a Nutshell", things will be different.
Right now, if you search for "panther" with Google, links #1 and #2 both point to pages within apple.com/macos/ Want to make a bet what position Tiger Direct is in once Google notices those pages have changed?
On the Comparison Matrix page that ends the article, look at the line for "USB extension cable". Those devices come with a short USB cable that solves the problem of the drive itself being too fat to plug into a crowded USB port.
I've been using the same Verizon broadband service as is described in the article for about six months now. When connecting to their EVDO network, I get a consistant raw rate of 500Kb/s on downloads and around 100Kb/s up. This is in the suburbs about 15 miles from the center of the coverage area I live near (Baltimore). While I don't have any firm numbers, the network seems even faster when I'm nearer to the center of their area, like when in NYC.
Note that I saw raw rate here because I'm not using the compression package Verizon provided. While it seemed to slightly improve performance when browsing the web, I found that something it was messing with made SSH sessions much less reliable. I turned it off the first day and haven't missed it.
While there are occasional hiccups that cause me to lose the connection (which are no more frequent than when I'm connected with 802.11b; screen is your friend here), I can assure you that when in the EVDO sections the SSH sessions I launch every day work perfectly even if I have two or three computers sharing the network connection via NAT.
> I don't know where you got the impression that the government pays your debts in bankruptcy protection.
Probably from the fact that the US government has help bailout a few large organizations in the past when it was deemed important enough; airlines and banks are the usual beneficiaries.
In this case, spammer Scott Ricters's company is filing for Chapter 11 bankruptcy. The debts aren't being repaid by anyone or even written off yet. This provides them with some protection from creditors while the company is reorganized with some supervision by the courts. See http://www.sec.gov/investor/pubs/bankrupt.htm for more information.
It somehow just feels right that when I try to access most of the pages on the Murphy's Laws site right now, like following your link, they return an error:
"Murphy's error 404 law: The chance of receiving a 404 error page is in inverse proportion to your need to find the page"
P/E ratios are a poor way to compare stocks in new growth companies, as they don't account for the rate at which earnings growth is accelerating which is far more important than the current earnings amount. P/E looks back, not forward.
If you look at the earnings rating that Investor's Business Daily computes, GOOG gets a score of 99, the maximum possible. I quote from IBD founder Bill O'Neil's book "How To Make Money in Stocks" to point out show short-sighted P/E thinking is with companies like this, picking one example most here are familiar with.
"American Online sold for over 100 times earnings in November 1994 before increasing 14,900% from 1994 to its top in December 1999."
I've made plenty of money buying companies with a P/E of near 100 and selling as it hit 500+. That said, I still wouldn't buy GOOG, but it's more because the market cap extrapolated from the relatively small number of public shares seems insane.
Do you really want to be tilting a PowerBook around when you've only got one hand free?
> Among many options, one only needs 1300 (out of 1600) on SAT. These days, MANY people easily get 1300.
You don't get a qualifying Mensa reading comprehension score with this statement. You need a >1300 if you took test before 9/30/74, back when it was hard. They clearly state that they haven't considered the SAT to correlate with IQ for over ten years now; 1994 was the last year they accepted those scores.
Cosmic Encounter remains my all my time favorite board game for larger groups--it takes about 6 people to really be fun and you can easily play with over 10 if you have enough expansion parts.
The only similarity to Magic is that each player draws a card at the beginning of the game that determines which type of alien you are and therefore your special abilities. Your style of play totally changes based on which of the aliens you are, which is part of what keeps it so fresh for repeats: it's never quite the same dynamic.
> PC clocks are rather crappy and temperature sensitive
Line voltage sensitive, too. With the way newer processors throttle their speeds around based on temperature and loading, and the way fans change their parameters based on temperature, I have little hope for this technique nailing any new system.
Let's see, what were the authors using in the lab where they tested machine to machine variations?
"All the machines were Micron PCs with 448MHz Pentium II Processors". Right. From this, we get the grand statement shortly afterward "The current results strongly support our claim that modern processors have relatively stable clock skews". Uh, sorry guys, you didn't use a single modern processor for this section; just some obsolete ones that run so cool they don't have any CPU clock or temperature varation. There's not a machine to be found in their entire test that features the kind of design we seen in acutal modern processors.
> don't forget to disable their 802.11g protocols as well as any proprietary "get 100Mbps out of your 11Mbps card" crap.
Now that they might notice as a performance drop and investigate. Shouldn't give the neighbors any reason to notice what you've done.
> Set half of your neighbors to the lowest supported frequency, and the other half to the highest...Put yours in the middle.
This is the worst possible solution. Most wireless routers default to the middle channel, 6. This means that the ones you won't be able to hack into will conflict if you pick the middle frequency as the one for yourself. And any new units added to the network after your hacking spree, or anyone who resets their units to the factory default are also going to conflict with you.
For your private frequency, you need to use 1 or 11 (I find 11 works better, less conflict with microwaves or something, don't know why). Then remap all your neighbors to either 6 or the one you didn't use. If you figure out who is who using a tool like Netstumbler on a site survey of sorts first, you can often remap things such that your neighbors work better than they did before you started because they interfere with one another less. That's what I di....thought about doing before realizing it would be illegal.
You need to buy a CRT because nothing else you can buy can be expected to have the kind of longevity you're looking for. The long-term service records of DLP, LCD, and especially Plasma screens are either short or unknown. You'll pay considerably more than $300 over 20 years for new backlights on an LCD/DLP for example (expect to pay that much every 4000 hours of watching or so). And while typical plasma units are claimed to last 30,000 hours which comes out to over 20 years for typical playtimes, if you're on the unlucky part of the bell curve a failure inside the main plasma area can mean scrapping the entire unit. And as you correctly devine, Plasma prices do not reflect a reasonable price/performance ratio.
So you need another CRT, and a new Mitsubishi unit to replace your old one would be perfect--those are still solid units likely to last you a long time and their display's technical quality is best in class. A glance at their web site shows that a current HD-capable set (having an outboard tuner is a better idea anyway) at 55", the WS-55315, weighs 215 pounds. Price on that model runs around $1700. What you can get in any alterate display technology for $1700 is laughable in comparison quality, longevity, and servicability wise.
To read about the sort of things that are important to getting a good quality picture out of HDTV, I would recommend picking up any random issue of Widescreen Review magazine that features some CRT TV reviews to see what parameters they test on. The main thing that cheap sets (and any Plasma unit for that matter) really screw up are color linearity/accuracy and black level.
I'm going to address CRTs as far as good brands go, because if you're not getting a CRT you're prioritizing something other than image quality under normal viewing conditions. As such the non-CRT recommendation process becomes very specific to your priorities and it's hard to give a good answer. I'm not trying to be snotty here, because I certainly understand that it's often the case that display quality is not even close to the main decision parameter. For example, the last TV box I bought was a small projector, which I knew perfectly well wasn't as good as a CRT. But I was living on a 5th floor walk-up apartment and not about to haul a good TV up there when I had a short-term lease.
Anyway, Mitsubishi's high-end Diamond CRTs have the best factory calibration regime I'm aware of to make sure they are faithfully displaying their inputs, and their less expensive models are invariably at the top of the accuracy (and build quality) heap as well. Usually on the expensive side in get what you pay for fashion. At lower price ranges, Toshiba CRT sets usually give the best accuracy relative to their price. Some of the Sony sets look very good, but talk to any statistically significant number of people who have dealth with Sony repair centers and you'll never consider one of their products again.
As always with TVs, displays in showrooms are totally bogus unless you are verifying color temperature and brightness/contrast/sharpness yourself across sets. Most showrooms sets are too bright and too blue, because those are the characteristics that make people prefer a TV at first glance in the same way that louder stereo equipment always seems better at first.