does the "normal" signature algorithm take into account the recipient? i'm guessing it doesn't. then the message digest pre-signature will be the same for each message the spammer sends, which would negate the real compute time problem.
finally, i thought that network capacity was the real constraint on spammers, not compute time. already they slightly vary each message to try to get around kill lists.
sorry to be a naysayer - i get as much spam as everyone else.
option 1: just a signature.
that's the same problem as an email address. easy for a spammer to use someone else's.
option 2: what i think you intended: pr-key signed message digest.
who checks the signature before dropping it in the bit bucket? each end-user? could work, but how does the proggy know where to get the public key?
keygen for each spam doesn't really matter if you are a spammer. who cares if someone fakes your fake signature? make fake key pairs that aren't really sets of primes. or just gen 1415 real primes and use each x,y combo to generate 1,000,405 key pairs. sure, that would be cryptographic suicide, but again, why would a spammer care if they are just going to ignore the existing laws?
okay, so maybe you then slow down the servers that register pks so that people can't register keys so quickly. but just one "rogue" signature server and you are SOL. don't have all the signature servers in your list--well, then you start dropping legimate mail. and, you create a chokepoint like DNS for people to start holding you up for charges.
what's wrong with SPF? that seems like an idea that can work.
Only major brokers (i.e. members) have the direct access to make an error like this possible. It's not something a normal person using Etrade could accomplish, and not something Etrade should allow someone to do. Of course, things should be automated. But there are different levels of automation. Letting the computer trade by program without intervention should be at the top of the pyramid and should not be done lightly. And if you expect to reap the benefits of such high automation, then you should also be willing to pay the costs. Like your analogy, if you have a really important message, then maybe just sending an email without also calling to check it got there is not enough. That's all I'm saying.
Cancelling trades, even real errors, is the worst policy for NASDAQ. Here's why:
#1 It destroys faith in the markets. A lot of the people who bought stock to provide liquidity had their buys cancelled, but not their sales. Therefore they lost money. This is a rare, but not the first time, this has happened. All those liquidity providers will be a little slower to stabilize the markets in the future because of the risk someone comes and cancels their trades later. Than means when you go to sell your 1000 shares of SCOX when the market is down and SCOX is down more, but you need the money to buy your house, that bid you are counting on might not be there, or it might be a lower price that it would otherwise be. And this happens thousands of times compared to the occasional "real" error, so the cost of this "what if my trade gets cancelled worrying" is very high and very real.
#2 Negates personal (and corporate) responsibility by the people who caused the problem (which turned out to be a problem for a lot of people, not just them). People should think a bit before they hook up hundreds of millions of dollars to an automated machine, especially one that does things like "sell if I'm down 5%." Then they should think again. And if they handle that much money, we should be able to rely that they are sophisticated entities that should absorb their own errors. If not, someone needs to take their computers away and give them back their Nintendos to play with instead.
#3 Nasdaq only cancelled orders executed through certain systems under their direct purview, not all trades done during that time regardless of system. Half a solution really is worse than none at all. Pretending the rest of the market doesn't exist (it's not part of us, so it's not our problem) is not a high quality solution.
If you want serious reliability, it is possible. You have to think, and you have to be willing to pay for it, one way or another, if it's really that important.
I tried to find a good link to Jay Forrester's Reliability of Components article but the only thing I could find was the IEEE which wanted me to buy it (again--I can't find it). If anyone knows of a valid link, or has the pdf, please respond or email me.
You sound like you know what you are talking about, but it still doesn't make that much sense.
Was the client long one million shares of the stock, and they went to send 1000 orders to trade out of the stock, all at the same time, rather than 1000 orders for 1000 shares each spaced two minutes apart (would take days).
Or was the error that the client was only long 1000 shares, and then they autobailed their 1000 shares but the execution for some reason didn't feed back to their inventory, so it keep trying to sell the "same" 1000 shares over and over (and would have until the stock hit zero).
If it was more the former, that would fit what you explained. If it was more the latter, that wouldn't explain why the sales never flattened their inventory and thereby (mathematically) stopped the autobail process.
If the cursor moves randomly... NOMAD If it is just horribly buggy... SOBAD, MYBAD If every time you try to run the shell, it spams you with paid advertisements... POPAD If it has problems with for loops... CANTADD If its syntax drives you to drink... OLDGRANDAD And if hackers use it to root the system and trigger a catastrophe... NORAD
What's good for the goose is not always good for the gander. Taxing things at a price that doesn't precisely reflect its negative costs to society, if any, makes the pie smaller for everyone.
If you don't want email at zero marginal cost, then feel free to start your own "premium" service that does charge you, and everyone else, and runs on a private or pseudo-private network.
Taxing should be the last refuge because it is such a blunt instrument.
Maybe the interests of efficiency dictate that some smaller standard fossil-fuel power plants should be connected to the net. Some peakers may me turned on and off every hour or even more frequently as needed, and they tend to be smaller, too, thus making it even less economical to keep someone there all the time just to throw a manual switch.
But internal nuclear control systems open to net traffic, or secondarily vulernable in any way to DOSing? That's insanity.
Anyone reading this work at a nuke station or on their control systems? Please tell me the critical control and monitoring systems are on a sealed, unconnected network. I know that people are effectively still dependent on the public network for inter-plant communcation, but at least no one can "open the pod bay doors" on the nuke plant except the people in the control room. Also please tell me that they removed the floppy and CD-rom drives and locked down the units.
Think like the NSA does. Separate computers for the public and private network. I'd like my nuclear reactors to be at least that secure.
When there is no clear market price for what you are donating, then original cost, or if ascertainable, replacement cost, would be strong guide in determining the value of the goods provided. Depreciated original cost (whatever the book value would have been) might also be a good starting estimate.
Donating is sometimes a great idea. The problem is that since the company went out of business, it's almost certain they lost money, thus they have no taxes to offset. If was a C corp, then unless that business revives itself without a change of control within the statutory period for reclaiming tax losses, the "losses" will never offset any taxes inside the corp.
It would be very interesting to hear someone who knows says who "owns" the code if the corporation was formally wound-up without some of the intangible assets having been sold. Theoretically the liquidators should have gotten some value for it. It may have even been sold to the law firm or other insider for $0.01.
I would suggest taking the "I'm poor but would like to work on this / see the code live" approach. That will lead well into either the "why doesn't the current holder give it away to something like the FSF and take the deduction" or the "what's the lowest price I can pay for it" discussion or the licensing/royalty discussion, which also sounds like a good idea.
Feynman explains the "why" and in such a natural way that you're much more likely to remember it later, especially if the original poster is learning it by himself. (BTW, the exercises reprints accompanying the lectures are available -- look for scsi-guy's ebay store, for one.) The lectures wouldn't work that well to teach yourself during a semester; you can't juggle four or five real classes, all the homework, and still internalize Feyman's lectures, which were supposed to be presented over two years.
It raises the costs a lot (too much unless you are treating yourself or better yet, can get them from the library), but I found that listening to the audio tapes while reading the lectures to be a very powerful experience. There's no substitute for the subtlety, emphasis, and cadence that a great teacher brings to a subject he has mastered.
The beauty of the Feynman lectures is that they are very accessible. You don't have to "get deep" to get understand the main points.
As far as the depth goes--either Feynman himself or the foreword mentioned that they were a bit much for many of the students, but that as the course progressed, many of the seats vacated by the students were taken over by the other Caltech faculty.
One extremely well-written textbook is Calculus in Context. It is modestly below the level of my high school and college math texts, but years later, when I want to remind myself how to do something and why it's the right thing to do, it's a great reference.
If anyone reads this and can give a lead on where to buy copies of the calculus text that Feynman himself mentioned in "Surely You're Joking..." I'd really appreciate it.
Common sense here. When you send a normal, unsolicited message to someone, you no longer own the message or have any rights to it. Think about about the volumes of published letters of historical figures. That's why you are supposed to think twice before hitting the send button, the modern equivalent of having Mary Lincoln let the President's scathing letters season overnight.
Suppose I send you an email calling you a bastard, and put some legalese at the end. It just doesn't matter. The fact is that I called you a bastard, and you can tell anyone you want about it, and show them the email to prove it.
Obviously the situation is different if there is a formal relationship of trust or duty between the parties. Likewise, if the government wants you not to disclose something (e.g., a letter authorizing a wiretap order you as an SP are required to perform)--that's another story.
That's a fair comment. It also tells you how bad the problem is that NYS has had to this step.
Maybe it would make more sense if NYS at least kept the old policy with respect to its own state schools (presumably most likely to be serving the public as intended), but it might be against the law. (IANAL.)
The big public (i.e. they have listed stocks) companies are DeVry (c'mon, techies should know this one), Strayer, Corinthian College, Career Education Corp, Education Management Corp, and Apollo Group (owner of UoP) and parent of separately-listed UoP Online.
UoP is the "gold standard" because they only do degree programs. The rest have greater or lesser participation in "diploma" programs, which could be anything from art school to diesel mechanics. (Think Sally Struthers, and I'm not talking about hungry kids.)
Two-thirds of the all the for-profit enrollment goes to these institutions. The rest mostly go to numerous privately-owned for-profit colleges.
The big guys all have online programs to some extent, while the little guys are also developing them thanks to online service providers like microcap EVCI, which used to be a videoconference company but now licenses software and acts a service provider for online education to many colleges, including some of the big ones.
All the big colleges are expanding by buying up the smaller institutions. However, already owning 2/3 of the space, they are now finding it tougher to expand profitably and have started buying things like Caribbean medical schools (Ross U.). Because of the way Title IV federal funding for education works, it is much more favorable to by a branch already in operation that to open a new one. To continue to expand, they have to gain students from the non-profit colleges, namely the community colleges.
At quick glance one can't tell a for-profit from a non-for-profit unless you check it out. And it's not clear that you should care too much--many non-profits are run basically for the benefit of administrators and faculty--that's who gets the economic profit!
The big difference used to be the aggressive recruiting by the for-profits, which has since been disallowed because the institutions would price whatever program (degree or diploma) at the level of the government loans and just sing people up, telling them that they didn't have to front any money. Then the poor bastards would graduate (or more often, not) 18 or 24 months later none the wiser, default on their loans, and the institution would still get paid, because the loans are government guaranteed (besides which, they already collected their money). New York state is now changing the law to at least withhold 1/3 of funds until the student actually graduates; it's a small hardship for students to raise the cash ( a few thousand) but will make a huge difference in eliminating the "no-money-down" type programs that really take advantage of people who believe everything they read in subway advertisements.
As you would expect, the for-profits are quicker to sell what they know people want to buy. And many people want cheap, easy degrees. Particularly in government service, it doesn't matter where you get your degree, as long as it's from an accredited institution, which almost all institutions aside from pure diploma mills (and a number of law schools) are. Like people said, I'm sure you could learn a lot online if you were really excited about the material. But most people aren't paying for the material; they are (or should be) paying for the structure and feedback that they need to help (force) themselves to learn the material, just like hiring a personal trainer.
The online degrees may be a great deal for the first people to get them, before employers get wise to the average level of learning completed. Then the backlash will come.
Go with debian and apt-get or dpkg your way back to sanity. Yes, you have to get the oo.org package from the vpnjunkies.de site, but if you are really, really paranoid (or have to be), then you should be compiling it yourself, anyway.
I started with SuSE, was mad I couldn't get openoffice or the wireless drivers or kernel I wanted, but then found debian. Debian's upgrade and package install, _especially_ the kernel, are just so good that I don't ever expect to switch. If the system setup wasn't so painful, everyone who doesn't need a Rhat service contract ought to use it.
Teach Yourself in 24 Hours: Recovering from Being Owned
Table of Contents Chp 1: Introduction Chp 2: Explaining why login is acting funny Chp 3: Recognizing and using the kill switch Chp 4: How to use the US Postal Service to notify users of service and email interruption Chp 5: Installing from Backups Chp 6: Special for "Teach yourself Unix System Administration in 24 Hours": Installing fresh from CD-ROM Chp 7: How to buy and use "Teach yourself Unix System Administration in 24 Hours" Chp 8: Go to Chapter 2
Save $4.95 and buy both books now! -
Yes, I had to change the title, but it would be tough to learn, in 24 hours, how to recover in 24 hours from being owned. -
The hard disks they used to put in Macs sounded like a little man inside shaking a gourd every time it spun up. I hope the new ones are better than that.
Slashdot Mirror
does the "normal" signature algorithm take into account the recipient? i'm guessing it doesn't. then the message digest pre-signature will be the same for each message the spammer sends, which would negate the real compute time problem.
finally, i thought that network capacity was the real constraint on spammers, not compute time. already they slightly vary each message to try to get around kill lists.
sorry to be a naysayer - i get as much spam as everyone else.
So let's say everyone signs.
option 1: just a signature.
that's the same problem as an email address. easy for a spammer to use someone else's.
option 2: what i think you intended: pr-key signed message digest.
who checks the signature before dropping it in the bit bucket? each end-user? could work, but how does the proggy know where to get the public key?
keygen for each spam doesn't really matter if you are a spammer. who cares if someone fakes your fake signature? make fake key pairs that aren't really sets of primes. or just gen 1415 real primes and use each x,y combo to generate 1,000,405 key pairs. sure, that would be cryptographic suicide, but again, why would a spammer care if they are just going to ignore the existing laws?
okay, so maybe you then slow down the servers that register pks so that people can't register keys so quickly. but just one "rogue" signature server and you are SOL. don't have all the signature servers in your list--well, then you start dropping legimate mail. and, you create a chokepoint like DNS for people to start holding you up for charges.
what's wrong with SPF? that seems like an idea that can work.
Only major brokers (i.e. members) have the direct access to make an error like this possible. It's not something a normal person using Etrade could accomplish, and not something Etrade should allow someone to do. Of course, things should be automated. But there are different levels of automation. Letting the computer trade by program without intervention should be at the top of the pyramid and should not be done lightly. And if you expect to reap the benefits of such high automation, then you should also be willing to pay the costs. Like your analogy, if you have a really important message, then maybe just sending an email without also calling to check it got there is not enough. That's all I'm saying.
Cancelling trades, even real errors, is the worst policy for NASDAQ. Here's why:
#1 It destroys faith in the markets. A lot of the people who bought stock to provide liquidity had their buys cancelled, but not their sales. Therefore they lost money. This is a rare, but not the first time, this has happened. All those liquidity providers will be a little slower to stabilize the markets in the future because of the risk someone comes and cancels their trades later. Than means when you go to sell your 1000 shares of SCOX when the market is down and SCOX is down more, but you need the money to buy your house, that bid you are counting on might not be there, or it might be a lower price that it would otherwise be. And this happens thousands of times compared to the occasional "real" error, so the cost of this "what if my trade gets cancelled worrying" is very high and very real.
#2 Negates personal (and corporate) responsibility by the people who caused the problem (which turned out to be a problem for a lot of people, not just them). People should think a bit before they hook up hundreds of millions of dollars to an automated machine, especially one that does things like "sell if I'm down 5%." Then they should think again. And if they handle that much money, we should be able to rely that they are sophisticated entities that should absorb their own errors. If not, someone needs to take their computers away and give them back their Nintendos to play with instead.
#3 Nasdaq only cancelled orders executed through certain systems under their direct purview, not all trades done during that time regardless of system. Half a solution really is worse than none at all. Pretending the rest of the market doesn't exist (it's not part of us, so it's not our problem) is not a high quality solution.
If you want serious reliability, it is possible. You have to think, and you have to be willing to pay for it, one way or another, if it's really that important.
I tried to find a good link to Jay Forrester's Reliability of Components article but the only thing I could find was the IEEE which wanted me to buy it (again--I can't find it). If anyone knows of a valid link, or has the pdf, please respond or email me.
You sound like you know what you are talking about, but it still doesn't make that much sense.
Was the client long one million shares of the stock, and they went to send 1000 orders to trade out of the stock, all at the same time, rather than 1000 orders for 1000 shares each spaced two minutes apart (would take days).
Or was the error that the client was only long 1000 shares, and then they autobailed their 1000 shares but the execution for some reason didn't feed back to their inventory, so it keep trying to sell the "same" 1000 shares over and over (and would have until the stock hit zero).
If it was more the former, that would fit what you explained. If it was more the latter, that wouldn't explain why the sales never flattened their inventory and thereby (mathematically) stopped the autobail process.
If the cursor moves randomly ... NOMAD ... SOBAD, MYBAD ... POPAD ... CANTADD ... OLDGRANDAD ... NORAD
If it is just horribly buggy
If every time you try to run the shell, it spams you with paid advertisements
If it has problems with for loops
If its syntax drives you to drink
And if hackers use it to root the system and trigger a catastrophe
Sticking a "hot" walkman in heavy water might not be the best idea. Maybe it could come with a nice lead gift box.
What's good for the goose is not always good for the gander. Taxing things at a price that doesn't precisely reflect its negative costs to society, if any, makes the pie smaller for everyone.
If you don't want email at zero marginal cost, then feel free to start your own "premium" service that does charge you, and everyone else, and runs on a private or pseudo-private network.
Taxing should be the last refuge because it is such a blunt instrument.
Exactly!
Maybe the interests of efficiency dictate that some smaller standard fossil-fuel power plants should be connected to the net. Some peakers may me turned on and off every hour or even more frequently as needed, and they tend to be smaller, too, thus making it even less economical to keep someone there all the time just to throw a manual switch.
But internal nuclear control systems open to net traffic, or secondarily vulernable in any way to DOSing? That's insanity.
Anyone reading this work at a nuke station or on their control systems? Please tell me the critical control and monitoring systems are on a sealed, unconnected network. I know that people are effectively still dependent on the public network for inter-plant communcation, but at least no one can "open the pod bay doors" on the nuke plant except the people in the control room. Also please tell me that they removed the floppy and CD-rom drives and locked down the units.
Think like the NSA does. Separate computers for the public and private network. I'd like my nuclear reactors to be at least that secure.
More properly, the cat may or may not have eaten the first one....
Does everyone else know this acronym?
Thanks.
Not a lawyer. Now then...
When there is no clear market price for what you are donating, then original cost, or if ascertainable, replacement cost, would be strong guide in determining the value of the goods provided. Depreciated original cost (whatever the book value would have been) might also be a good starting estimate.
Donating is sometimes a great idea. The problem is that since the company went out of business, it's almost certain they lost money, thus they have no taxes to offset. If was a C corp, then unless that business revives itself without a change of control within the statutory period for reclaiming tax losses, the "losses" will never offset any taxes inside the corp.
It would be very interesting to hear someone who knows says who "owns" the code if the corporation was formally wound-up without some of the intangible assets having been sold. Theoretically the liquidators should have gotten some value for it. It may have even been sold to the law firm or other insider for $0.01.
I would suggest taking the "I'm poor but would like to work on this / see the code live" approach. That will lead well into either the "why doesn't the current holder give it away to something like the FSF and take the deduction" or the "what's the lowest price I can pay for it" discussion or the licensing/royalty discussion, which also sounds like a good idea.
Feynman explains the "why" and in such a natural way that you're much more likely to remember it later, especially if the original poster is learning it by himself. (BTW, the exercises reprints accompanying the lectures are available -- look for scsi-guy's ebay store, for one.) The lectures wouldn't work that well to teach yourself during a semester; you can't juggle four or five real classes, all the homework, and still internalize Feyman's lectures, which were supposed to be presented over two years.
It raises the costs a lot (too much unless you are treating yourself or better yet, can get them from the library), but I found that listening to the audio tapes while reading the lectures to be a very powerful experience. There's no substitute for the subtlety, emphasis, and cadence that a great teacher brings to a subject he has mastered.
The beauty of the Feynman lectures is that they are very accessible. You don't have to "get deep" to get understand the main points.
As far as the depth goes--either Feynman himself or the foreword mentioned that they were a bit much for many of the students, but that as the course progressed, many of the seats vacated by the students were taken over by the other Caltech faculty.
One extremely well-written textbook is Calculus in Context. It is modestly below the level of my high school and college math texts, but years later, when I want to remind myself how to do something and why it's the right thing to do, it's a great reference.
If anyone reads this and can give a lead on where to buy copies of the calculus text that Feynman himself mentioned in "Surely You're Joking..." I'd really appreciate it.
topic
IANAL:
Common sense here. When you send a normal, unsolicited message to someone, you no longer own the message or have any rights to it. Think about about the volumes of published letters of historical figures. That's why you are supposed to think twice before hitting the send button, the modern equivalent of having Mary Lincoln let the President's scathing letters season overnight.
Suppose I send you an email calling you a bastard, and put some legalese at the end. It just doesn't matter. The fact is that I called you a bastard, and you can tell anyone you want about it, and show them the email to prove it.
Obviously the situation is different if there is a formal relationship of trust or duty between the parties. Likewise, if the government wants you not to disclose something (e.g., a letter authorizing a wiretap order you as an SP are required to perform)--that's another story.
That's a fair comment. It also tells you how bad the problem is that NYS has had to this step.
Maybe it would make more sense if NYS at least kept the old policy with respect to its own state schools (presumably most likely to be serving the public as intended), but it might be against the law. (IANAL.)
The big public (i.e. they have listed stocks) companies are DeVry (c'mon, techies should know this one), Strayer, Corinthian College, Career Education Corp, Education Management Corp, and Apollo Group (owner of UoP) and parent of separately-listed UoP Online.
UoP is the "gold standard" because they only do degree programs. The rest have greater or lesser participation in "diploma" programs, which could be anything from art school to diesel mechanics. (Think Sally Struthers, and I'm not talking about hungry kids.)
Two-thirds of the all the for-profit enrollment goes to these institutions. The rest mostly go to numerous privately-owned for-profit colleges.
The big guys all have online programs to some extent, while the little guys are also developing them thanks to online service providers like microcap EVCI, which used to be a videoconference company but now licenses software and acts a service provider for online education to many colleges, including some of the big ones.
All the big colleges are expanding by buying up the smaller institutions. However, already owning 2/3 of the space, they are now finding it tougher to expand profitably and have started buying things like Caribbean medical schools (Ross U.). Because of the way Title IV federal funding for education works, it is much more favorable to by a branch already in operation that to open a new one. To continue to expand, they have to gain students from the non-profit colleges, namely the community colleges.
At quick glance one can't tell a for-profit from a non-for-profit unless you check it out. And it's not clear that you should care too much--many non-profits are run basically for the benefit of administrators and faculty--that's who gets the economic profit!
The big difference used to be the aggressive recruiting by the for-profits, which has since been disallowed because the institutions would price whatever program (degree or diploma) at the level of the government loans and just sing people up, telling them that they didn't have to front any money. Then the poor bastards would graduate (or more often, not) 18 or 24 months later none the wiser, default on their loans, and the institution would still get paid, because the loans are government guaranteed (besides which, they already collected their money). New York state is now changing the law to at least withhold 1/3 of funds until the student actually graduates; it's a small hardship for students to raise the cash ( a few thousand) but will make a huge difference in eliminating the "no-money-down" type programs that really take advantage of people who believe everything they read in subway advertisements.
As you would expect, the for-profits are quicker to sell what they know people want to buy. And many people want cheap, easy degrees. Particularly in government service, it doesn't matter where you get your degree, as long as it's from an accredited institution, which almost all institutions aside from pure diploma mills (and a number of law schools) are. Like people said, I'm sure you could learn a lot online if you were really excited about the material. But most people aren't paying for the material; they are (or should be) paying for the structure and feedback that they need to help (force) themselves to learn the material, just like hiring a personal trainer.
The online degrees may be a great deal for the first people to get them, before employers get wise to the average level of learning completed. Then the backlash will come.
A little less talk; a little more action....
(why i remember this, we'll never know)
Go with debian and apt-get or dpkg your way back to sanity. Yes, you have to get the oo.org package from the vpnjunkies.de site, but if you are really, really paranoid (or have to be), then you should be compiling it yourself, anyway.
I started with SuSE, was mad I couldn't get openoffice or the wireless drivers or kernel I wanted, but then found debian. Debian's upgrade and package install, _especially_ the kernel, are just so good that I don't ever expect to switch. If the system setup wasn't so painful, everyone who doesn't need a Rhat service contract ought to use it.
Damn glass gets hot, don't it!
Teach Yourself in 24 Hours: Recovering from Being Owned
Table of Contents
Chp 1: Introduction
Chp 2: Explaining why login is acting funny
Chp 3: Recognizing and using the kill switch
Chp 4: How to use the US Postal Service to notify users of service and email interruption
Chp 5: Installing from Backups
Chp 6: Special for "Teach yourself Unix System Administration in 24 Hours": Installing fresh from CD-ROM
Chp 7: How to buy and use "Teach yourself Unix System Administration in 24 Hours"
Chp 8: Go to Chapter 2
Save $4.95 and buy both books now!
-
Yes, I had to change the title, but it would be tough to learn, in 24 hours, how to recover in 24 hours from being owned.
-
And the WROX Linux programming book features...a dvd store database. Hmmm...
The hard disks they used to put in Macs sounded like a little man inside shaking a gourd every time it spun up. I hope the new ones are better than that.
It can also toast four regular bagels or eight slices of Wonder bread. No extra charge.