I understand that the new-HD-triggers-activation problem can be avoided by making sure the new drive has the same volume ID as the old one. There is a tool to fix that floating around the net.
Much more sensible to use one of the tools to neuter XP entirely, so it stops lifting its leg on you every time you change the hardware. Fuck M$, it's none of their damn business what I do with MY hardware!
Then again, it was pretty clear to me (from back when I followed the XP newsgroups) that XP's hardware support was designed primarily to run on DELL systems, which in turn are designed to be NOT upgradeable. So WPA's assumption was that no one would ever do more than the most minor upgrades.
[I like and prefer Windows, but WPA is THE factor that makes me so hope for a really mature linux desktop that I can switch myself and my users to without pain... *sigh*...maybe I'll spend tomorrow playing with this shiny new SuSE CD that I just got from Novell.]
I'm wondering if embedding the stylesheet is actually good Coding Zen -- might be that what bandwidth it eats with every search is more than offset by the number of requests a separate stylesheet might get (even allowing for caching), thus in terms of sheer number of hits on their servers, embedded *might* be more efficient.
[reference: Abrash's Zen of ASM, where he points out that sometimes performance is much better if you have a big slow routine executed once, rather than a small fast routine executed a million times -- even tho at first glance, the small routine *seems* so much more efficient.]
Except there was more to it... Holland's Powers That Be told them in so many words that they were to hie themselves to the port and take the next ship out, or walk into the ocean, their choice. (Much as a fed-up England had informed them that they could either leave the country, or go to prison, their choice.) Hence America was a chance destination.
Holland only took them at all due to a treaty they'd made with Cromwell.
I think you're right... and I suspect most of the founding fathers, were they to air their 1776 views in 2005's political arena, would be regarded as somewhere between looney and dangerous:(
Some years ago, AOL instituted a keyword filtering process, and "breast" was one of the words you couldn't say in an AOL chat room. If you typed "breast", you were automatically disconnected.
This went over poorly in AOL's own breast cancer chat room...
Or at least, anyone who *admits* to being an atheist... then again, you won't get elected to Congress, or the Presidency, or appointed to the Supreme Court, if you profess such a blatantly "anti-religious" viewpoint.
And exactly so... I wish religion, government, and other coercive forces would just stay in their place and out of my life. A doomed hope!
Well, I wouldn't call Puritans nonconformists. They wanted everyone to conform.. to THEIR ideals, forcibly if need be. (Which is why they were kicked out of England, and then out of Holland -- they didn't leave either *voluntarily*.)
Unfortunately, that's a lot of the foundation for America's mindset; even lo these several centuries later, Puritanism rears its ugly head. Not that America is alone in having such issues, but it does run completely counter to what we *supposedly* stand for.
BTW, what *did* Jefferson believe? I don't recall anything about that from my gleefully-dirt-digging high school US-history teacher (if she'd known, she would have told us!)
Interesting... I just tried your workaround link in NS3 (no js) and wound up at the plain HTML page (which is the one *I* want:) If you get that page and don't want it, following the preferences link might put you back on the js model, since it complains that setting preferences requires a js-enabled browser. Just a thought.
I find that I expect scientific data to be expressed in Centigrade (do I sound old or what?:)...er, I mean Celsius, and everyday life in Fahrenheit. I think it's a matter of the appropriate granularity for the job intended:
Celsius naturally fits into calculations in chemistry and physics (where water is frequently used as a constant). Why do math involving 32 and 212 when you can do math involving 0 and 100?
Fahrenheit give more precision for everyday life without having to resort to decimals, but more importantly for everyday use, it divides nicely into blocks of "about that much" for stuff like how hot to set the oven and whether you need an extra sweater.
And if it involves superconductors or deep space, somehow my brain wants the numbers in Kelvin. It's no damned wonder it had a disk-full error in college; it was keeping all that data in triplicate!:)
And personally, I don't give a damn if it's "valid code" or not, so long as it works in every browser.
I just checked GMail with my *preferred* Netscape 3.04 (js and images off) and it worked great, plus was much faster -- and considerably more *readable* (the JS version has some annoying width errors that force sidescrolling in Mozilla).
I can now recommend GMail to my visually-impaired and hardware-challenged users, with every expectation that it will work fine for them.
Tagline alert!! You've, er, hit the nail right on the head.
Actually, that's a lot of the problem with marketing. "We have hammers. You have nails. What? They're not nails?? [*POUND*POUND*POUND*] They are now!!"
Egads, like ELN's "new" pig-slow webmail that's 100% javascript-driven, and broken in about half the browsers out there, didn't cause enough complaints? (The ELN newsgroups were flooded with 'em for over a year after the switch.) Sure, break it for half the browsers that still work by turning it into flash. I can't think of a more useless, or more potentially-dangerous way to do webmail. (Given that the user has absolutely no control over flash's behaviour, other than turning it off entirely with something like the mozdev addon.)
ELN's old webmail was FAST and had the same functionality and worked in any browser all the way down to lynx. Why wasn't that good enough? Why didn't they put it back in place after the customer base screamed bloody murder over the new webmail?
A: because it was an expensive package (IIRC, list price is $10,000) they'd already bought and paid for, so by damn it was staying no matter HOW much everyone hated it.
As to the increased mailbox -- I have to wonder if evil varieties of webmail are DESIGNED to force people who need webmail to send their mail somewhere else, thus rendering the size of the mailboxes moot.
[I've been an ELN customer for over 8 years, but there are times when I'd like to smack 'em upside the head.]
I'd say copyright infringement is more akin to writing a bad check. And the usual penalty for bad checks is that you have to pay your bank an overdraft fee ($5 to $20 depending on the bank), plus you have to pay the "infringed" party three times the value of the original check. This is fair and reasonable -- the culprit gets dinged proprotionally to the actual offense, and the victim gets a certain compensation for their time and trouble, as well as recovering the original money owed.
But copyright infringement penalties demand is that you pay not only for the original loss and the PITA for recovering that loss, but ALSO that you pay for every POTENTIAL loss.
This is the same as if the victim of a bad check gets to recover not only the value of the original check, but ALSO the *potential* value AND the *potential* INCREASE in value of anything the victim *might* have purchased with the money that the culprit deprived them of the use of.
IOW, it penalizes based on the SPECULATIVE value of the money represented by the infringed item, rather than based on the ACTUAL value.
I didn't get the "we done took it down" message; I just got a shitload of ColdFusion errors. And I concur -- every damn time I've seen a CF site go under heavy load, it's proceeded to produce tons of such errors. Bah!!
As you imply, one way to make Powerful Folks' heads whip around is when it Happens To Them. Maybe some enterprising techie could apply these tracking techniques to certain Powerful Folks, and make the data public. After all, if our lives are no longer private, why should theirs be??
What's sauce for the goose is sauce for the gander...
What are your thoughts about how spoofable it might be, say with a firewall *designed* to produce "random skew" ??
You mentioned Comcast... I can see nasty ISPs using this to tie service to a single computer, and forcing folk to pay extra if they have more than one machine connected.
Someone above mentioned that the numbers indicate it should be accurate enough "to uniquely identify 1 computer in a million". Given that, combined with other identifiable info it should be enough to get a hard ID.
However, I'm wondering how easy it might be to spoof, such as with a firewall *designed* to produce random skew. Thoughts, anyone??
Has anyone scoured the available sources for said purported easter egg? I'd think it could be as simple as "type in undocumented switch, get back undocumented Hello World".
I have no idea re CP/M, but in one program that I do know about, there is an easter egg that consists of only one line of source code, and if it weren't obfuscated, the whole thing would be only a few dozen characters long.
I have used M$DOS versions 3.2, 5.0, 6.00, 6.2x, 7.0, and 7.1; PCDOS 3.1, and DRDOS versions 6.0, DR/NWDOS7.0 (patch 9 -- DR/NWDOS7 had *FIFTEEN* patches to address serious or fatal bugs), and DRDOS 7.01 and 7.03.
Of the lot, M$DOS 6.00 was the most stable and bug-free -- in fact, the OS itself NEVER crashed in all the years I used it (my main DOS machine back then *averaged* almost 2 years between reboots, and then only for hardware issues). And it never exhibited any weird conflicts or self-induced memory leaks.
Conversely, DRDOS does crash occasionally, even with no apps running -- the DRDOS memory manager, while more capable and versatile, has some issues (its DPMI host doesn't leak as much as CWSDPMI or PMode, but it still leaks). The main advantage was that if an application crashed on M$DOS, it generally locked up the system, whereas if the same app crashed on DRDOS, you might get back a prompt (but memory would still be messed up, and it was best to salvage your work if you could, then reboot). Also, DRDOS had more oddball conflicts, again primarily due to memory manager quirks. And I found that the DRDOS EMM386 won't run at all on my P3 systems (it insists on hogging some memory addresses that the chipset uses, and its -exclude switch does not work).
Win3.1 was somewhat more stable atop DR/NWDOS7, but I think this was because 1) Win3.1x would not run at all unless the DRDOS DPMI host was loaded, due to some deficiency in DRDOS's EMM386 which I no longer remember the specifics of, and 2) likely the DRDOS DPMI host was more stable than the Win3.1x DPMI host (which apparently didn't load if it found a host already present).
I also benchmarked M$DOS 6.00 vs DR/NWDOS 7.x -- and found that M$DOS ran about twice as fast on the same hardware. DRDOS was still noticeably slower on a 486, tho not so much on midrange Pentiums. Again this was primarily a memory manager issue, but partly poor I/O. Some of this issue went away as of patch 15.
Novell lost the source for their 15 patches, so when it became DRDOS7 again, they had to start over with the 7.00 source. And DRDOS 7.01 was terrible (unstable and cranky). But by 7.03 they'd gotten the bugs worked out, and back up to the state of Novell's edition with all 15 patches, plus had got rid of a major bug in the DPMI host. End result -- DRDOS 7.03 is certainly the best of its line. It's unfortunate that it seems to be the last.
So.. here's what I settled on as the most stable, but not necessarily best-performing setups:
For pure DOS with no need for a DPMI host: M$DOS 7.0 (or 6.0 if no need for FAT32; they are otherwise functionally identical. Why not 6.2x? it has an I/O bug that is in neither 6.00 nor 7.x.) Note: I use this even on *very* old systems (XT thru 486), because it can produce the most available memory, and is the best performer of all DOS versions I've worked with.
For DOS and Win3.1x: DRDOS 7.03 or NWDOS 7.0 (patch 9 or later *required*; patch 15 preferred) mainly to provide the DPMI host. This will be more stable than Win3.1x on M$DOS 6, but will also run noticeably slower.
For DOS apps that need a DPMI host, or for the base under Win95 on first-generation Pentiums: base OS M$DOS 7.0, with the DRDOS 7.03 EMM386 and DPMI host.
PII and above, and for FAT32 support: M$DOS 7.x, and 3rd party DPMI hosts as needed.
If I had to pick one DOS as a mission-critical OS, where being absolutely crash-free was the #1 criterion, it would be M$DOS 6.00 or 7.0.
Did I mention that I resent rebooting, and never do so unless forced into it?:)
Slashdot Mirror
I understand that the new-HD-triggers-activation problem can be avoided by making sure the new drive has the same volume ID as the old one. There is a tool to fix that floating around the net.
...maybe I'll spend tomorrow playing with this shiny new SuSE CD that I just got from Novell.]
Much more sensible to use one of the tools to neuter XP entirely, so it stops lifting its leg on you every time you change the hardware. Fuck M$, it's none of their damn business what I do with MY hardware!
Then again, it was pretty clear to me (from back when I followed the XP newsgroups) that XP's hardware support was designed primarily to run on DELL systems, which in turn are designed to be NOT upgradeable. So WPA's assumption was that no one would ever do more than the most minor upgrades.
[I like and prefer Windows, but WPA is THE factor that makes me so hope for a really mature linux desktop that I can switch myself and my users to without pain... *sigh*
To be rapidly offset by the volume of complaints from people whose browsers don't support PNG and don't understand why the logo isn't being displayed.
I'm wondering if embedding the stylesheet is actually good Coding Zen -- might be that what bandwidth it eats with every search is more than offset by the number of requests a separate stylesheet might get (even allowing for caching), thus in terms of sheer number of hits on their servers, embedded *might* be more efficient.
[reference: Abrash's Zen of ASM, where he points out that sometimes performance is much better if you have a big slow routine executed once, rather than a small fast routine executed a million times -- even tho at first glance, the small routine *seems* so much more efficient.]
Except there was more to it... Holland's Powers That Be told them in so many words that they were to hie themselves to the port and take the next ship out, or walk into the ocean, their choice. (Much as a fed-up England had informed them that they could either leave the country, or go to prison, their choice.) Hence America was a chance destination.
Holland only took them at all due to a treaty they'd made with Cromwell.
I think you're right... and I suspect most of the founding fathers, were they to air their 1776 views in 2005's political arena, would be regarded as somewhere between looney and dangerous :(
:)
LOL at the examples, especially that last one!
Oh no, I understood -- I just found it interesting that the workaround still falls back to the plain HTML version if it encounters a non-JS browser.
:)
I *want* the plain-HTML version, so I was happy
Ah, thanks for the link. [reads random sections] Interesting. Too bad we don't have more Jeffersons running the world these days... :(
Some years ago, AOL instituted a keyword filtering process, and "breast" was one of the words you couldn't say in an AOL chat room. If you typed "breast", you were automatically disconnected.
This went over poorly in AOL's own breast cancer chat room...
Or at least, anyone who *admits* to being an atheist ... then again, you won't get elected to Congress, or the Presidency, or appointed to the Supreme Court, if you profess such a blatantly "anti-religious" viewpoint.
And exactly so... I wish religion, government, and other coercive forces would just stay in their place and out of my life. A doomed hope!
Interesting quotes, and sadly, he's dead-on :(
... hey! Utah is available....
BTW, I'm mostly a republican, wholly an atheist, and firmly believe that the religious right should all be quarantined
Well, I wouldn't call Puritans nonconformists. They wanted everyone to conform.. to THEIR ideals, forcibly if need be. (Which is why they were kicked out of England, and then out of Holland -- they didn't leave either *voluntarily*.)
Unfortunately, that's a lot of the foundation for America's mindset; even lo these several centuries later, Puritanism rears its ugly head. Not that America is alone in having such issues, but it does run completely counter to what we *supposedly* stand for.
BTW, what *did* Jefferson believe? I don't recall anything about that from my gleefully-dirt-digging high school US-history teacher (if she'd known, she would have told us!)
Interesting... I just tried your workaround link in NS3 (no js) and wound up at the plain HTML page (which is the one *I* want :) If you get that page and don't want it, following the preferences link might put you back on the js model, since it complains that setting preferences requires a js-enabled browser. Just a thought.
I find that I expect scientific data to be expressed in Centigrade (do I sound old or what? :) ...er, I mean Celsius, and everyday life in Fahrenheit. I think it's a matter of the appropriate granularity for the job intended:
:)
Celsius naturally fits into calculations in chemistry and physics (where water is frequently used as a constant). Why do math involving 32 and 212 when you can do math involving 0 and 100?
Fahrenheit give more precision for everyday life without having to resort to decimals, but more importantly for everyday use, it divides nicely into blocks of "about that much" for stuff like how hot to set the oven and whether you need an extra sweater.
And if it involves superconductors or deep space, somehow my brain wants the numbers in Kelvin. It's no damned wonder it had a disk-full error in college; it was keeping all that data in triplicate!
And personally, I don't give a damn if it's "valid code" or not, so long as it works in every browser.
I just checked GMail with my *preferred* Netscape 3.04 (js and images off) and it worked great, plus was much faster -- and considerably more *readable* (the JS version has some annoying width errors that force sidescrolling in Mozilla).
I can now recommend GMail to my visually-impaired and hardware-challenged users, with every expectation that it will work fine for them.
Tagline alert!! You've, er, hit the nail right on the head.
Actually, that's a lot of the problem with marketing. "We have hammers. You have nails. What? They're not nails?? [*POUND*POUND*POUND*] They are now!!"
Egads, like ELN's "new" pig-slow webmail that's 100% javascript-driven, and broken in about half the browsers out there, didn't cause enough complaints? (The ELN newsgroups were flooded with 'em for over a year after the switch.) Sure, break it for half the browsers that still work by turning it into flash. I can't think of a more useless, or more potentially-dangerous way to do webmail. (Given that the user has absolutely no control over flash's behaviour, other than turning it off entirely with something like the mozdev addon.)
ELN's old webmail was FAST and had the same functionality and worked in any browser all the way down to lynx. Why wasn't that good enough? Why didn't they put it back in place after the customer base screamed bloody murder over the new webmail?
A: because it was an expensive package (IIRC, list price is $10,000) they'd already bought and paid for, so by damn it was staying no matter HOW much everyone hated it.
As to the increased mailbox -- I have to wonder if evil varieties of webmail are DESIGNED to force people who need webmail to send their mail somewhere else, thus rendering the size of the mailboxes moot.
[I've been an ELN customer for over 8 years, but there are times when I'd like to smack 'em upside the head.]
I'd say copyright infringement is more akin to writing a bad check. And the usual penalty for bad checks is that you have to pay your bank an overdraft fee ($5 to $20 depending on the bank), plus you have to pay the "infringed" party three times the value of the original check. This is fair and reasonable -- the culprit gets dinged proprotionally to the actual offense, and the victim gets a certain compensation for their time and trouble, as well as recovering the original money owed.
But copyright infringement penalties demand is that you pay not only for the original loss and the PITA for recovering that loss, but ALSO that you pay for every POTENTIAL loss.
This is the same as if the victim of a bad check gets to recover not only the value of the original check, but ALSO the *potential* value AND the *potential* INCREASE in value of anything the victim *might* have purchased with the money that the culprit deprived them of the use of.
IOW, it penalizes based on the SPECULATIVE value of the money represented by the infringed item, rather than based on the ACTUAL value.
http://www.negativland.com/albini.html
I didn't get the "we done took it down" message; I just got a shitload of ColdFusion errors. And I concur -- every damn time I've seen a CF site go under heavy load, it's proceeded to produce tons of such errors. Bah!!
As you imply, one way to make Powerful Folks' heads whip around is when it Happens To Them. Maybe some enterprising techie could apply these tracking techniques to certain Powerful Folks, and make the data public. After all, if our lives are no longer private, why should theirs be??
What's sauce for the goose is sauce for the gander...
What are your thoughts about how spoofable it might be, say with a firewall *designed* to produce "random skew" ??
You mentioned Comcast... I can see nasty ISPs using this to tie service to a single computer, and forcing folk to pay extra if they have more than one machine connected.
Someone above mentioned that the numbers indicate it should be accurate enough "to uniquely identify 1 computer in a million". Given that, combined with other identifiable info it should be enough to get a hard ID.
However, I'm wondering how easy it might be to spoof, such as with a firewall *designed* to produce random skew. Thoughts, anyone??
Has anyone scoured the available sources for said purported easter egg? I'd think it could be as simple as "type in undocumented switch, get back undocumented Hello World".
I have no idea re CP/M, but in one program that I do know about, there is an easter egg that consists of only one line of source code, and if it weren't obfuscated, the whole thing would be only a few dozen characters long.
I have used M$DOS versions 3.2, 5.0, 6.00, 6.2x, 7.0, and 7.1; PCDOS 3.1, and DRDOS versions 6.0, DR/NWDOS7.0 (patch 9 -- DR/NWDOS7 had *FIFTEEN* patches to address serious or fatal bugs), and DRDOS 7.01 and 7.03.
:)
Of the lot, M$DOS 6.00 was the most stable and bug-free -- in fact, the OS itself NEVER crashed in all the years I used it (my main DOS machine back then *averaged* almost 2 years between reboots, and then only for hardware issues). And it never exhibited any weird conflicts or self-induced memory leaks.
Conversely, DRDOS does crash occasionally, even with no apps running -- the DRDOS memory manager, while more capable and versatile, has some issues (its DPMI host doesn't leak as much as CWSDPMI or PMode, but it still leaks). The main advantage was that if an application crashed on M$DOS, it generally locked up the system, whereas if the same app crashed on DRDOS, you might get back a prompt (but memory would still be messed up, and it was best to salvage your work if you could, then reboot). Also, DRDOS had more oddball conflicts, again primarily due to memory manager quirks. And I found that the DRDOS EMM386 won't run at all on my P3 systems (it insists on hogging some memory addresses that the chipset uses, and its -exclude switch does not work).
Win3.1 was somewhat more stable atop DR/NWDOS7, but I think this was because 1) Win3.1x would not run at all unless the DRDOS DPMI host was loaded, due to some deficiency in DRDOS's EMM386 which I no longer remember the specifics of, and 2) likely the DRDOS DPMI host was more stable than the Win3.1x DPMI host (which apparently didn't load if it found a host already present).
I also benchmarked M$DOS 6.00 vs DR/NWDOS 7.x -- and found that M$DOS ran about twice as fast on the same hardware. DRDOS was still noticeably slower on a 486, tho not so much on midrange Pentiums. Again this was primarily a memory manager issue, but partly poor I/O. Some of this issue went away as of patch 15.
Novell lost the source for their 15 patches, so when it became DRDOS7 again, they had to start over with the 7.00 source. And DRDOS 7.01 was terrible (unstable and cranky). But by 7.03 they'd gotten the bugs worked out, and back up to the state of Novell's edition with all 15 patches, plus had got rid of a major bug in the DPMI host. End result -- DRDOS 7.03 is certainly the best of its line. It's unfortunate that it seems to be the last.
So.. here's what I settled on as the most stable, but not necessarily best-performing setups:
For pure DOS with no need for a DPMI host: M$DOS 7.0 (or 6.0 if no need for FAT32; they are otherwise functionally identical. Why not 6.2x? it has an I/O bug that is in neither 6.00 nor 7.x.) Note: I use this even on *very* old systems (XT thru 486), because it can produce the most available memory, and is the best performer of all DOS versions I've worked with.
For DOS and Win3.1x: DRDOS 7.03 or NWDOS 7.0 (patch 9 or later *required*; patch 15 preferred) mainly to provide the DPMI host. This will be more stable than Win3.1x on M$DOS 6, but will also run noticeably slower.
For DOS apps that need a DPMI host, or for the base under Win95 on first-generation Pentiums: base OS M$DOS 7.0, with the DRDOS 7.03 EMM386 and DPMI host.
PII and above, and for FAT32 support: M$DOS 7.x, and 3rd party DPMI hosts as needed.
If I had to pick one DOS as a mission-critical OS, where being absolutely crash-free was the #1 criterion, it would be M$DOS 6.00 or 7.0.
Did I mention that I resent rebooting, and never do so unless forced into it?
Paterson claims that Evans falsely accused Kildall of being the "inventor" of DOS
[emphasis mine] Well, that would explain the defamation suit. ;)
[disclaimer: I like and still use DOS. Oh dear... are we DOS users indemnified here?? or will we all be defamed too??]