We use it here too, and have folks working from home who are happy with it. It's commercial, however, whereas CVS is, of course, free as in beer and speech. (Of course, if you're considering VSS, hey, that's commercial too:-)
On the scalability side, I haven't seen too many problems with CVS as things got bigger. Speaking for myself, I'd look at Perforce, but if it's already been decided From On High that it's gonna be either CVS or VSS, I'd go with CVS.
> I betcha it was the NSA who did this, trying to put their backdoor on Windows systems!!
Actually, the first thing that went through my mind was "I'm glad NSA is gonna be all over this."
The number of users likely to click "yes" to the question "Always trust certificates from Microsoft Corporation" is staggeringly high. In the absence of a viable CRL (certificate revocation) capability in browsers, these certs, if (when?) they fall into the wrong hands, are dangerous weapons.
If the "wrong hands" are organized criminals, the stability of the banking system could be at risk. If the "wrong hands" are agents of another government, it could get even worse.
> Do they want to wait until the next cracker to deface the front page of Altavista or Yahoo adds an ActiveX virus and wipes out (quite easily) ten million machines?
Stop thinking "cracker", "portal page", and "0wn j00", and start thinking "criminal", "financial institution", and... well, "0wn" is the right word, isn't it?
Nobody takes the kind of risk this guy took without a reasonable expectation of reward. The individual(s) who got the certs is probably not the group who ultimately intends to use them.
> So even though VeriSign lists the certificates in their CRL, they don't provide a way for the browser's CRL-checking mechanism to check it. Looks like its still VeriSign's fault.
s/still/operating to spec, and as designed from Day One, this design flaw always was/g
> from what little I read the person who fraudulently claimed to represent Msft might be in some serious trouble.
...and this stops others from misusing the fraudulently-obtained certificate... how?
Sure, MSFT will have a patch out for W95 through XP. But given the number of Solaris boxen out there still running Sendmail 8.6 - how likely is it that every Joe-average windoze luzer in the world will apply the patch?
Someone's gonna make a lot of money off this cert. Illegally, yeah. But it's gonna happen. Given the bits about the organized cracking attempt made on the banks recently, this scares the living fuck out of me.
> I have to laugh when people (non-technical people) complain about how a computer is obsolete the day you buy it.
Ditto. I just flashed the 0112 (beta) BIOS into my old TX-97 motherboard. My old P166MMX now supports 2.0V-2.2V CPUs. $30 for a K6-III at Fry's (which I originally bought for my I-opener!), and my old P166 box is now chugging along at 400 MHz with 256K on-die cache. If I overclock the PCI bus, I can take it to 500.
For $30, I turn "that old computer I don't use anymore" into an all-in-one software-based DVD player, MP3 jukebox, and decent gaming box for the price of a pizza and a case of beer. (And I find a good use for the TV-out jack on the video card, which I never bothered to use, 'cuz hey, I can't play DVDs on a P166;-)
> word processing, email, and browsing.
In defence of your nontechnical friends, if they're "compulsive upgraders", their "old" computers probably have gotten slower, even if they're doing the same tasks. Netscape 3.0 to IE5.5? Word97 to Orifice2K? Win95 OSR2 to Win2000 or WinME? Yeah, if they've done these upgrades, their perfectly adequate P166 boxen probably are grinding under the strain, especially if they cheaped-out and got 32M of RAM.
Computers don't get slower as time passes, but software does.
> It reads like one of those godawful free "alternative" newspapers: obsequious defenses of Bill Clinton that would embarass James Carville, columnists who say "fuck" a lot and endless, tedious blathering about sex.
Yeah. I read it because it's enjoyable - most of the time, I disagree with their editorial stance, but I figure that's a Good Thing; it never hurts to see things from the other side's point of view. I'll miss that.
(Of course, that said, I've pretty much skipped their political commentary too, about the time of the Hyde stories. Yeah, Salon, we know you're a mouthpiece of the Democratic party. We get it. Honest.;-)
> Turning off JavaScript gets rid of most of the crap. I normally run with JavaScript turned off.
Ditto. Hell, I didn't even know Salon was running Javashit popup ads.
When Salon went from the "one page per article" to the "break up long articles into five pages, each one or two PgDn presses long" model, the only reason I continued reading is because I could click on "Print This" and get the article in a readable format. (Yeah, idjit web dee-zyners, some of us like to scroll! Pressing the spacebar is a lot easier than mousing to a one-digit-wide URL and clicking... especially if I've got a coffee in my mousing hand!)
Of course, I'm part of the problem. I read Salon for the content, not the ads, and I have Junkbuster blocking cookies and image-autoloading off. And while I enjoy it, it ain't worth $30 to me.
But all they need to do to get rid of me is take away "view the whole article with one click".
Hmm... Now that I think about it, something like Junkbuster and a non-null sblock.ini should probably qualify.
After all, it's something its users use to block content they deem offensive;-)
If it ever looks like the bill will pass, all we have to do is point this out and supply a suitable blockfile. The DMA will spend a small fortune on lobbyists to kill it.
Pisses off the control freaks. Costs the DMA money. Sounds like win-win all around.
> I don't know if you know or not, but you do not have to use EarthLink's browser with their ISP
service. Its not like AOL, which a gateway to the internet (or it was last time I checked).
While this is true today (I, too, set up M$ DUN manually when dealing with any ISP), it may not be in the future.
There have been persistent rumors over the past month or so that ELNK is about to be bought out by MSN.com.
Earthpink's business goal is to become the next AOL. "Sandbox" is an apt word - they market themselves as "the real Internet" (the anti-AOL), but the reality is that they're trying to be AOL.
> What would happen if a Senator (or CongressCritter) read DeCSS out loud on the floor of the
Senate or House? Would it then be a matter of Public Record?
Never mind DeCSS, how 'bout OT III?
(Oh, wait, someone did that in Sweden, which is what led to the lawsuit against Zenon, which is what led to the DMCA threat against Slashdot, and wow, we've come full circle;-)
> If the President doesn't use encryption in his e-mail to friends and family, then Joe User certainly
doesn't.
1) Dubya knows enough about security to know not to send needs-to-be-secure information over insecure networks. (His Dad taught him better than that at his previous gig;-)
2) His personal emails, which don't need to be as secure, may or may not be encrypted. FOIA is what we're talking about, however, and it may require (I guess it did require, in the case of the emails mentioned) that - encrypted or not, the keys be turned over.
3) This isn't about Joe User - it's about Mr. President. Mr. President's emails, like any government employee's, may very well be public property and therefore subject to an FOIA request. This kinda sucks (if you're Mr. President), but it's not inconsistent with the Fourth Amendment.
As a public servant (leaving politics out of this for a moment), Dubya knew this when he took the job. If his snail-mails are subject to FOIA requests, then his e-mails ought to be similarly subject.
> [MP3s were useless on a 386 with a 150M hard drive...] Remember, it is not about what is, it's about what will be.
Yep. Great example for the "where's the innovation" article.
DivX;-) is another bit of "evolution" - but it (and MPEG-2, and MPEG-1 before it) had to wait until CPUs were fast enough to decode the compressed video on-the-fly.
Just like (early Microsoft.AVI formats) we had to wait for I/O channels to handle the bandwidth of 160x100 uncompressed video.
I look back at my '286 with the 40M hard drive, and realize I have individual songs that exceed that drive's capacity.
(To say nothing of my Apple ][ and its 128K of RAM, 64K of which could be accessed at once by bank-switching... ah, 143K on a 5.25" diskette! For only $7.00 per disk!)
>I'd hate to be the admin on a mail server with users passing around 700mb files via email. Most
of the trading is done on IRC, usenet, and P2P applications...
I dunno, being a USENET admin where they're posting 700M files isn't much fun either;-)
> In looking at research being done now, I again see only a path of
incremental improvement.
Unless your name is Einstein ("what happens as you get faster to the speed of light?") or Wright ("Hmm, if we stick static wings on this bicycle, maybe we can fly better than the guys trying to flap"), or Ford ("Hey, what if we made a machine to build a bunch of identical parts, and used humans to assemble the parts into lots of cars, instead of building carriages by hand?"), you're only likely to see incremental improvement.
Incremental improvement isn't bad. The automobile, passenger air travel, and yes, 300-baud modems through 56K modems are all examples of incremental improvement.
Some technologies which seem like breakthroughs (MP3 vs. uncompressed.WAV) are merely incremental -- I remember when you got "graphics" by downloading uncompressed memory dumps of video RAM. Then there was.GIF (lossless compression, quick to view on a '286). Then - when CPUs permitted it -.JPG (lossy compression that required a 386 or 486 to render in 2-3 seconds per image).
But yes, evolution of technology does take years.
I like games like CivII and Alpha Centauri, where the "waiting" of 10 years between breakthrough and application can be over in a night. Unfortunately, real life ain't like that - if it takes 10 years of "game time", it takes 10 years of "real time".
As for comparisons between the airplane and the rocket for space exploration... well, until there's somewhere there worth going to, or something there worth bringing back, nobody's gonna build the technology to make it worthwhile. This is (IMHO) sad, verging on the tragic, but true.
> Only problem is america seems to have a heavy anti-bemani sentiment.. finding an
arcade on the east coast with bemani games is near impossible, while in cali, there are
dozens:(
Although bemani is cool as hell - the games are much larger than their 1980s-style cash-cow predecessors, and cost a lot more.
I don't have current pricing, but I'd imagine these machines are comparable to the driving games. Maybe $10000 vs. $15-20K. But still a lot of money for the small-time "hometown arcade" operator to gamble, particularly with the cultural differences wrt Dance Dance Revolution, Beatmania, and so on.
> Arcade games can be broken down into three categories, in
my experiences: Racing, Shooting (gun to monitor) and fighting. It's my experience that when the gaming
industry finds a hit, it takes years to recover, and it's a damned shame.
I don't think it's coming back.
The real reason isn't so much that apart from racing and shooting games, everything's doable at home, but that arcade operators can no longer make a buck.
1981: Buy a Missile Command for $2000-3000. There's nothing at home that can match it, except for people who own $2000 computers - that's what a 6502-based computer cost back then, and it didn't even have graphics! There's nothing even comparable until the Atari 800, and it's not like very many kids own one.
So you make your money back in 3-6 months, and then it's all profit for the next year or so.
1998: Buy Mortal Kombat 4 for $4500. About the same as your $2000 Missile Command after inflation. Problem is, the home version comes out a month or two later, and the kids stay at home. At $0.50 per play, you barely make enough to pay for the machine.
2001: Don't buy Mortal Kombat VIII - you learned your lesson with MK3 and MK4. Instead, buy the latest racing game for $20000. If it turns out to be the next "Cruisin'" or "Virtua Racing", you might make money.
If it doesn't, you've got a $20000 piece of junk on your hands. It takes up four times the space of a regular machine. It's just that after barely breaking even on four MK3 machines a few years ago, you figure at least this thing won't be available in the home for another six months.
But even at a buck a play, it's no different than having four normal machines at $0.25 per play in terms of revenue per square foot. You just hope that the kids don't get bored with before before you pay off the initial cost.
Arcade machines are no longer cash cows - they've become very expensive gambles. There's a reason Sega owns the Playdium -- it's because nobody else is willingto risk it. In fact, I'd bet that Playdium and the other "really big arcades" of today would be losing money if they paid the prices the average arcade operator would have to pay. The "play center" arcades are loss-leaders in order to generate interest in the console markets.
If you make consoles or console-based games, arcades aren't "good business", they're just part of the marketing budget. For everyone else, arcades aren't good business. Period.
> Fuck that. It reminds me
of some old SNES and NES games with large levels. Oh, sure you could continue, but you'd have to
continue from the beginning of the level and do it ALLL over again just to get killed by the boss at the end
one more time.
So - this is game design we're talking about - get rid of the "boss".
IMHO game design started to go downhill when the "boss" concept came into vogue. The term we used was "quartersucker", in that the game was typically designed to allow you an easy romp through most of the level before utterly slaughtering you at the boss stage.
A positive monotonic learning curve is a good thing - the game gets harder as the player gets better, and it does so at a reasonably uniform rate.
The games of the 80s featured this - but the difficulty usually maxed out at some point where the CPU was incapable of throwing more stuff at the player and maintaining frame rates (or keeping track of everything in the 64K RAM typically available to CPUs of that day.) Difficulty typically maxed out after 15-20 minutes, after which time the length of play was limited only by the player's bladder capacity and/or number of lives.
The boss concept introduced a discontiunity into the learning curve, allowing all players to enjoy the game up to a certain point, then ending the game (or forcing the addition of more quarters).
With modern CPUs, we could remove the upper limit on difficulty imposed by the 8-bit CPUs of the olden days, and do away with the boss and the continue concept. Gameplay can become "infinitely" hard; limited only by the player's nervous system.
Think of the 100-meter dash - a "game" that's thousands of years old. There's an upper limit to how fast even the most steroid-enhanced runner can run 100m. But nobody knows what that is, which is why we still revere folks who set records. (Or Konami's "Track and Field" with a CPU fast enough that your final score won't max out when you use a vibrating buzzer to tap the buttons for you - the game becomes a test of engineering skills;-)
Getting around the problem you describe ("boring easy levels") is easy enough - just allow the player to start on a harder level, and give a bonus (score or time) for doing so. (as in Atari's Tempest, 1982).
Getting around the other problem you describe ("starting over from the beginning every time you get killed in 10 seconds by that goddamn boss") goes away when you get rid of the boss.
My candidate for "worst offender" in the "continue / boss" category: Rabbit Punch. Great game, funny as all hell. You "continue" through the entire game, utterly failing to build up any skill whatsoever, until the final level, at which point you can no longer continue, and have to restart the level from scratch. There's a special spot in hell waiting for the game designer who did that.
> how many people do you think actually *know* what the registry *is* ?
The registry is "the thing that gets fixed when you click on this.reg file that lets you connect up with a new Napster ID".
Only a clueless newbie would click on a.reg file without inspecting its contents. But - the people who don't know what the registry is are precisely the people clueless enough to blindly do so in order to get back onto Napster.
So yeah, if all they did was nuke users, they would ultimately end up with a situation where all Joe Windoze knows (and to accomplish his task, all he needs to know) about "the registry" is that "it's what gets fixed by clicking on the 'i want my napster back' icon"...
For those not following along - it's called a DDoS attack on the IRS.
That's right - this cult successfully DDoSed the IRS' legal team by ordering its members to file individual suits against IRS in such numbers that the IRS caved in and granted the cult tax-exempt status.
Just because it's done with legal papers and not TCP/IP packets doesn't change the fact that it's a DDoS attack.
For more $cieno DDoS activity, search for "scientology spam usenet attack" or some such. Basically, it was the largest and most coordinated HipCrime-style attack on in USENET history.
Slashdot Mirror
Not even an article moderation approval header referring to Cats? Damn, you guys do play hardball! ;-)
We use it here too, and have folks working from home who are happy with it. It's commercial, however, whereas CVS is, of course, free as in beer and speech. (Of course, if you're considering VSS, hey, that's commercial too :-)
On the scalability side, I haven't seen too many problems with CVS as things got bigger. Speaking for myself, I'd look at Perforce, but if it's already been decided From On High that it's gonna be either CVS or VSS, I'd go with CVS.
Actually, the first thing that went through my mind was "I'm glad NSA is gonna be all over this."
The number of users likely to click "yes" to the question "Always trust certificates from Microsoft Corporation" is staggeringly high. In the absence of a viable CRL (certificate revocation) capability in browsers, these certs, if (when?) they fall into the wrong hands, are dangerous weapons.
If the "wrong hands" are organized criminals, the stability of the banking system could be at risk. If the "wrong hands" are agents of another government, it could get even worse.
Stop thinking "cracker", "portal page", and "0wn j00", and start thinking "criminal", "financial institution", and... well, "0wn" is the right word, isn't it?
Nobody takes the kind of risk this guy took without a reasonable expectation of reward. The individual(s) who got the certs is probably not the group who ultimately intends to use them.
s/still/operating to spec, and as designed from Day One, this design flaw always was/g
Sure, MSFT will have a patch out for W95 through XP. But given the number of Solaris boxen out there still running Sendmail 8.6 - how likely is it that every Joe-average windoze luzer in the world will apply the patch?
Someone's gonna make a lot of money off this cert. Illegally, yeah. But it's gonna happen. Given the bits about the organized cracking attempt made on the banks recently, this scares the living fuck out of me.
Serves you right for putting it next to my box, which I was cooling with liquid hydrogen!
Ditto. I just flashed the 0112 (beta) BIOS into my old TX-97 motherboard. My old P166MMX now supports 2.0V-2.2V CPUs. $30 for a K6-III at Fry's (which I originally bought for my I-opener!), and my old P166 box is now chugging along at 400 MHz with 256K on-die cache. If I overclock the PCI bus, I can take it to 500.
For $30, I turn "that old computer I don't use anymore" into an all-in-one software-based DVD player, MP3 jukebox, and decent gaming box for the price of a pizza and a case of beer. (And I find a good use for the TV-out jack on the video card, which I never bothered to use, 'cuz hey, I can't play DVDs on a P166 ;-)
> word processing, email, and browsing.
In defence of your nontechnical friends, if they're "compulsive upgraders", their "old" computers probably have gotten slower, even if they're doing the same tasks. Netscape 3.0 to IE5.5? Word97 to Orifice2K? Win95 OSR2 to Win2000 or WinME? Yeah, if they've done these upgrades, their perfectly adequate P166 boxen probably are grinding under the strain, especially if they cheaped-out and got 32M of RAM.
Computers don't get slower as time passes, but software does.
Yeah. I read it because it's enjoyable - most of the time, I disagree with their editorial stance, but I figure that's a Good Thing; it never hurts to see things from the other side's point of view. I'll miss that.
(Of course, that said, I've pretty much skipped their political commentary too, about the time of the Hyde stories. Yeah, Salon, we know you're a mouthpiece of the Democratic party. We get it. Honest. ;-)
Ditto. Hell, I didn't even know Salon was running Javashit popup ads.
When Salon went from the "one page per article" to the "break up long articles into five pages, each one or two PgDn presses long" model, the only reason I continued reading is because I could click on "Print This" and get the article in a readable format. (Yeah, idjit web dee-zyners, some of us like to scroll! Pressing the spacebar is a lot easier than mousing to a one-digit-wide URL and clicking... especially if I've got a coffee in my mousing hand!)
Of course, I'm part of the problem. I read Salon for the content, not the ads, and I have Junkbuster blocking cookies and image-autoloading off. And while I enjoy it, it ain't worth $30 to me.
But all they need to do to get rid of me is take away "view the whole article with one click".
> 1) This is terrible!
> therefore
> 2) Something must be done!
You forgot: 2.5) for great just^W^Wthe children!
>
> # route add default reject
Hmm... Now that I think about it, something like Junkbuster and a non-null sblock.ini should probably qualify.
After all, it's something its users use to block content they deem offensive ;-)
If it ever looks like the bill will pass, all we have to do is point this out and supply a suitable blockfile. The DMA will spend a small fortune on lobbyists to kill it.
Pisses off the control freaks. Costs the DMA money. Sounds like win-win all around.
While this is true today (I, too, set up M$ DUN manually when dealing with any ISP), it may not be in the future.
There have been persistent rumors over the past month or so that ELNK is about to be bought out by MSN.com.
Earthpink's business goal is to become the next AOL. "Sandbox" is an apt word - they market themselves as "the real Internet" (the anti-AOL), but the reality is that they're trying to be AOL.
Never mind DeCSS, how 'bout OT III?
(Oh, wait, someone did that in Sweden, which is what led to the lawsuit against Zenon, which is what led to the DMCA threat against Slashdot, and wow, we've come full circle ;-)
Passphrase: "seineew era starcomed"
Who needs quantum computers? ;-)
1) Dubya knows enough about security to know not to send needs-to-be-secure information over insecure networks. (His Dad taught him better than that at his previous gig ;-)
2) His personal emails, which don't need to be as secure, may or may not be encrypted. FOIA is what we're talking about, however, and it may require (I guess it did require, in the case of the emails mentioned) that - encrypted or not, the keys be turned over.
3) This isn't about Joe User - it's about Mr. President. Mr. President's emails, like any government employee's, may very well be public property and therefore subject to an FOIA request. This kinda sucks (if you're Mr. President), but it's not inconsistent with the Fourth Amendment.
As a public servant (leaving politics out of this for a moment), Dubya knew this when he took the job. If his snail-mails are subject to FOIA requests, then his e-mails ought to be similarly subject.
I think you've got it backwards.
1) The "hacker-kiddies" ain't gonna sue for trademark infringement, and more importantly:
2) Without the name-recognition of the geek set, where are you gonna generate the "buzz" required for a successful financing?
Yep. Great example for the "where's the innovation" article.
DivX ;-) is another bit of "evolution" - but it (and MPEG-2, and MPEG-1 before it) had to wait until CPUs were fast enough to decode the compressed video on-the-fly.
Just like (early Microsoft .AVI formats) we had to wait for I/O channels to handle the bandwidth of 160x100 uncompressed video.
I look back at my '286 with the 40M hard drive, and realize I have individual songs that exceed that drive's capacity.
(To say nothing of my Apple ][ and its 128K of RAM, 64K of which could be accessed at once by bank-switching... ah, 143K on a 5.25" diskette! For only $7.00 per disk!)
I dunno, being a USENET admin where they're posting 700M files isn't much fun either ;-)
Unless your name is Einstein ("what happens as you get faster to the speed of light?") or Wright ("Hmm, if we stick static wings on this bicycle, maybe we can fly better than the guys trying to flap"), or Ford ("Hey, what if we made a machine to build a bunch of identical parts, and used humans to assemble the parts into lots of cars, instead of building carriages by hand?"), you're only likely to see incremental improvement.
Incremental improvement isn't bad. The automobile, passenger air travel, and yes, 300-baud modems through 56K modems are all examples of incremental improvement.
Some technologies which seem like breakthroughs (MP3 vs. uncompressed .WAV) are merely incremental -- I remember when you got "graphics" by downloading uncompressed memory dumps of video RAM. Then there was .GIF (lossless compression, quick to view on a '286). Then - when CPUs permitted it - .JPG (lossy compression that required a 386 or 486 to render in 2-3 seconds per image).
But yes, evolution of technology does take years.
I like games like CivII and Alpha Centauri, where the "waiting" of 10 years between breakthrough and application can be over in a night. Unfortunately, real life ain't like that - if it takes 10 years of "game time", it takes 10 years of "real time".
As for comparisons between the airplane and the rocket for space exploration... well, until there's somewhere there worth going to, or something there worth bringing back, nobody's gonna build the technology to make it worthwhile. This is (IMHO) sad, verging on the tragic, but true.
Although bemani is cool as hell - the games are much larger than their 1980s-style cash-cow predecessors, and cost a lot more.
I don't have current pricing, but I'd imagine these machines are comparable to the driving games. Maybe $10000 vs. $15-20K. But still a lot of money for the small-time "hometown arcade" operator to gamble, particularly with the cultural differences wrt Dance Dance Revolution, Beatmania, and so on.
I don't think it's coming back.
The real reason isn't so much that apart from racing and shooting games, everything's doable at home, but that arcade operators can no longer make a buck.
1981: Buy a Missile Command for $2000-3000. There's nothing at home that can match it, except for people who own $2000 computers - that's what a 6502-based computer cost back then, and it didn't even have graphics! There's nothing even comparable until the Atari 800, and it's not like very many kids own one.
So you make your money back in 3-6 months, and then it's all profit for the next year or so.
1998: Buy Mortal Kombat 4 for $4500. About the same as your $2000 Missile Command after inflation. Problem is, the home version comes out a month or two later, and the kids stay at home. At $0.50 per play, you barely make enough to pay for the machine.
2001: Don't buy Mortal Kombat VIII - you learned your lesson with MK3 and MK4. Instead, buy the latest racing game for $20000. If it turns out to be the next "Cruisin'" or "Virtua Racing", you might make money.
If it doesn't, you've got a $20000 piece of junk on your hands. It takes up four times the space of a regular machine. It's just that after barely breaking even on four MK3 machines a few years ago, you figure at least this thing won't be available in the home for another six months.
But even at a buck a play, it's no different than having four normal machines at $0.25 per play in terms of revenue per square foot. You just hope that the kids don't get bored with before before you pay off the initial cost.
Arcade machines are no longer cash cows - they've become very expensive gambles. There's a reason Sega owns the Playdium -- it's because nobody else is willingto risk it. In fact, I'd bet that Playdium and the other "really big arcades" of today would be losing money if they paid the prices the average arcade operator would have to pay. The "play center" arcades are loss-leaders in order to generate interest in the console markets.
If you make consoles or console-based games, arcades aren't "good business", they're just part of the marketing budget. For everyone else, arcades aren't good business. Period.
So - this is game design we're talking about - get rid of the "boss".
IMHO game design started to go downhill when the "boss" concept came into vogue. The term we used was "quartersucker", in that the game was typically designed to allow you an easy romp through most of the level before utterly slaughtering you at the boss stage.
A positive monotonic learning curve is a good thing - the game gets harder as the player gets better, and it does so at a reasonably uniform rate.
The games of the 80s featured this - but the difficulty usually maxed out at some point where the CPU was incapable of throwing more stuff at the player and maintaining frame rates (or keeping track of everything in the 64K RAM typically available to CPUs of that day.) Difficulty typically maxed out after 15-20 minutes, after which time the length of play was limited only by the player's bladder capacity and/or number of lives.
The boss concept introduced a discontiunity into the learning curve, allowing all players to enjoy the game up to a certain point, then ending the game (or forcing the addition of more quarters).
With modern CPUs, we could remove the upper limit on difficulty imposed by the 8-bit CPUs of the olden days, and do away with the boss and the continue concept. Gameplay can become "infinitely" hard; limited only by the player's nervous system.
Think of the 100-meter dash - a "game" that's thousands of years old. There's an upper limit to how fast even the most steroid-enhanced runner can run 100m. But nobody knows what that is, which is why we still revere folks who set records. (Or Konami's "Track and Field" with a CPU fast enough that your final score won't max out when you use a vibrating buzzer to tap the buttons for you - the game becomes a test of engineering skills ;-)
Getting around the problem you describe ("boring easy levels") is easy enough - just allow the player to start on a harder level, and give a bonus (score or time) for doing so. (as in Atari's Tempest, 1982).
Getting around the other problem you describe ("starting over from the beginning every time you get killed in 10 seconds by that goddamn boss") goes away when you get rid of the boss.
My candidate for "worst offender" in the "continue / boss" category: Rabbit Punch. Great game, funny as all hell. You "continue" through the entire game, utterly failing to build up any skill whatsoever, until the final level, at which point you can no longer continue, and have to restart the level from scratch. There's a special spot in hell waiting for the game designer who did that.
The registry is "the thing that gets fixed when you click on this .reg file that lets you connect up with a new Napster ID".
Only a clueless newbie would click on a .reg file without inspecting its contents. But - the people who don't know what the registry is are precisely the people clueless enough to blindly do so in order to get back onto Napster.
So yeah, if all they did was nuke users, they would ultimately end up with a situation where all Joe Windoze knows (and to accomplish his task, all he needs to know) about "the registry" is that "it's what gets fixed by clicking on the 'i want my napster back' icon"...
That's right - this cult successfully DDoSed the IRS' legal team by ordering its members to file individual suits against IRS in such numbers that the IRS caved in and granted the cult tax-exempt status.
Just because it's done with legal papers and not TCP/IP packets doesn't change the fact that it's a DDoS attack.
For more $cieno DDoS activity, search for "scientology spam usenet attack" or some such. Basically, it was the largest and most coordinated HipCrime-style attack on in USENET history.