Spambayes works pretty good too-- I get no more than a couple of messages a day at the most, often none at all. Now if only I could convince my ISP to use it instead of the RBL and other such crap they're using now and I'd get all of the email from people I want to get it from...
EXACTLY. Very unreliable, not due to spam but to spam filtering. Frankly I prefer reliability with 90% spam to what we've got now. At least I'd get to choose my own filtering and have noone to blame but myself for choosing it, if it's lousy. But then, I'm not an ISP who cares more about his bandwidth costs than he does his customer's email reliability...
What I like most about this scheme is that in order for it to work, it puts a huge onus on the copyright owners to proactively register their works with Google now, to an extent they probably didn't even have to with the US Copyright office. They can no longer complain to Google for lack of protection if they are too lazy to upload everything they own (and for most of the big majors, that's a lot of stuff). This lets Google off the hook and simplifies the takedown process for them, while giving the big media companies essentially what they asked for:
Google as new Copyright registrar. I love it!
With regards to matching-- I agree that accurate matching is probably not all that hard, you could probably reduce everything to 32x32 64 color pixels sampled @ 1 fps and fuzzy match the results and identify the vast majority of stuff-- and let humans check for false positives if necessary-- you could immediately have the system put a vid on "suspension" if it looks to be infringing until an eyeball gets the chance to look at it to confirm. Though I would guess that false positives could be pretty darn rare so that may not even be necessary. Things like uploading copyrighted vids backwards won't be useful to most people but could be included in the match anyway if it was a problem. Even completely shuffling the frame positions of the entire video may not be a good workaround, as you could sort the frames in some way and then compare them, then try them inverted or flipped or shifted-- the worst these operations might do in the long run is cause the matching algorithm to run longer as it tries more and more permutations that one might apply to the content...
On the other hand, Google may have to periodically rerun all the original content through their fingerprinting algorithm to regenerate the database as they tune up its accuracy, which given a huge database of registered works could end up being a resource consuming enterprise...
If you want an open phone, there are several on the market or very close to market that will work MUCH better and the companies will support you in creating the apps. There's no need to hack the iPhone and Apple has cheated no one.
There's no need to climb Mount Everest either, but lots of people do it just the same...
The thing is, even if the RIAA win every case they bring, it isn't going to fix their problem, at least in the long term. It may justify their existence to their bosses for a little longer, but that's about it. As long as the RIAA think that this sort of thing is the answer to their problem it continues to underscore just how completely out of touch they and their bosses are...
The digital horses are out of the barn, and neither litigation or stringent DRM is going to put them back in. And, both of those things make one of their other problems much worse-- that of PR. Odd that, as didn't these guys start out in PR? Isn't PR essentially what the record companies were mostly about? Watching them completely trashing their raison d'etre seems rather like watching someone commit seppuku who thinks it a good way to fix their stomach ache...
These guys are working in RETAIL-- Hel-lo! Anyone with any real tech creds wouldn't be caught dead in retail...
I suppose the question remains-- where *does* one go to get their computer fixed? I can't help much with the answer to that though, as I've always fixed my own. But then, I hand soldered every last darn solder joint on the first "home computer" I ever owned (IMSAI) which dates me as a geezer...
If you really want your model to look good at higher resolutions, you can always subdivide your triangle meshes, which is robust (unlike triangle reduction). Up-resing meshes is easy. Down-resing takes artist work.
Subdividing triangles and smoothing the result is not really "up-resing." You get more polygons but not more detail. Adding fine detail to a subdivided mesh still requires artist work.
Linux is FREE, you don't have to shell out your life savings in order to get it and use it effectively.
This is a really tired and lame argument in favor of Linux (FREE as in BEER). Only a moron would choose an operating system because it's free (as in beer). If Linux cost as much as Vista did and Vista was free, I'd still use Linux. For a lot of reasons, but for one thing, it runs on EVERYTHING-- a 15 year old PC, Mac hardware, the Power PC chip in my networked drive (Kurobox). When I need to automate on one I can write a script that will run everywhere-- write once run everwhere works for me and I don't even need Java. Then again, I like the command line as it allows me to do a lot of stuff in parallel. And for another thing, it's pretty darn reliable and I can easily make it conform to my workflow rather than me having to work around bogus assumptions (as is often true on the Mac), or buggy & inconsistent design (as is often true on Windows).
Choose Linux when it's BETTER, not because it's FREE (as in beer)...
Acutally, I have some rare CDs that have actually *increased* in value since I bought them. But with Amazon it's not all that hard to sell CDs-- you don't even have to scan them as someone else has already done it for you in most cases. And I'd rather trade-in a handful of old CDs that I don't want at a store in exchange for a couple of different ones than just piss the $$$ down the drain...
Even without DRM, I have a BIG problem with downloadable music. What's that you ask? You can't RESELL it. I buy a CD, even a used CD often for less than the $0.99/song that iTunes is charging, and if I later tire of it I can legally sell it to someone else. If I tire of a digital download, that's it. Just try selling your DRM-free bit collections on Amazon or eBay and see how far you get...
While I wouldn't buy a non-open device like this either, the problem is most customers don't know how to utilize openness so they don't care. As long as that is the case, there will be sufficient market-- in fact a LARGER market for dumbed-down devices like the iPhone than there will be for a more open device.
In fact, the very reason that the Mac is as successful as it is with it's dumbed-down interface is that a large number of people are more comfortable with a dumbed-down interface than they are with a power interface that expects them to make intelligent choices. Then again, the most popular OS is one designed by a guy named Gates who loves complex puzzles and seems to think we do too-- so there's hope at least...
I dunno, am I the ONLY one that did not become overly paranoid about terrorists coming to blow me up?
No, but like me, you probably don't work for the government. There's been several stories of late that recount that Bush & Cheney were absolutely convinced that additional terrorist attacks were likely after 9/11. And of course, that places like Washington DC were high on the target list. And initially, much of the US Congress thought that too. So here we're talking significant institutional paranoia in the institutions that actually make and enforce laws, and are supposed to deal with defending against attacks. This paranoia has turned out to have been a significant overreaction. That the resultant actions were based on imagination and ignorance and fraught with significant bureaucratic incompetence should be no surprise, as that's simply business as usual...
Well, I'm still waiting for a computer keyboard who's symbol set is not based on repurposed 50-year-old teletype equipment-- and has real multiply, divide and negative symbols (among other things) instead of co-opting what was available at the time for them-- given that basic math is a lot of what we do on computers these days, we really ought to have those symbols on our keyboards by now. Actually the first computer keyboard I ever used *did* have those symbols-- an IBM 2741 with an APL set-- ever since then it's been a step backward and there's been little advancement since.
Recall that when Color TV came out it was important for the transmissions to be viewed on the existing B&W sets-- so the color signal was encoded in a way that was designed to be compatible, rather than precise (NTSC often referred to as Never Twice the Same Color because of it). My family didn't get their first color TV 'til the Joneses on the block were buying their first replacement when their first color TV's tube started fading out-- in the latter half of the 1960s. In addition, I don't recall being all that impressed by Color TV when visiting friends houses that had them-- certainly not enough to go home and clamor that we should get one too. My first "personal" TV (that I got for my room) was B&W when our family TV was color, and I preferred to watch the B&W set because a) it was mine and 2) I could choose what to watch on it-- far more important than color. And I find HDTVs in Best Buy seem to flicker like strobe lights when things move-- now referred to as "judder" I think-- and it doesn't impress me. Given that, and the fact that many people have their kids growing up on VHS tape home-movies, I expect it will be some time before the "masses" have sufficiently moved to digital sets. Sure, those who like to watch sports games (which doesn't include me) and the latest TV shows (also, me not included) may jump on as early adopters, but that's somewhat less than everyone...
Analog medium has long ago reached market saturation, and may corporations see $$$$$ in a switchover, but that doesn't mean we will all just act like sheep and drink their kool-aid...
Have them enter their reg info on your website and get their license key, where you insert the users info encrypted into the software via the key such that it identifies the legal owner on startup. That can discourage legal owners from letting copies get out, without hampering the actual software operation. While pirates can patch it out, and it may be possible to register with bogus info, this is a pretty easy thing to do and does not impair users as the code does not otherwise restrict the usage of the product.
Users who do not have web access can get a license code over the phone which will input the same info--- it is often done with a combination of a license key and name information which are tied together so the license info they have will only work if they enter their name correctly into the products licensing screen. For example, the license code can contain a CRC of the customer info for example-- the customer calls in wanting to turn on the software, you have them fill out the reg info in the product itself and give you the key that *it* generates based on the entry of their name info. You then input that key to your system, validate their info and give them a license key that only works on the system that has the user info entered as they did.
I think you get the general idea. Check the web for similar schemes and fine tune it to your product, but this is probably the best way to discourage the use of casual illegal copies. The crack will likely end up being to patch the startup screen to say "Licensed to TheCrackCrew" which will advertise that it's a cracked copy on startup...
Uhm... I am the copyright owner of some music that I make available for free. If any of this happened to be the music in question the RIAA cannot prosecute her because of it, simply because it's "obvious" she doesn't own the copyright herself. The court has no way to know, just because she does not "own" the copyright, that she nevertheless has or does not have the right to hold or distribute a copy...
Since you brought up cars...Imagine what would happen if I "bought" a brand new Ford and Ford then told me that I couldn't give rides to my family, friends, co-workers, etc. because I didn't have a multi-passenger license.
Good thing he didn't bring up airplanes. Pilot licenses (different sort of license, I suppose), do not always include the right to carry passengers, and the rationale is not exactly "brain-dead." The license here is to operate the vehicle, not a licence connected with the vehicle itself regarding what you can do with it. Consequently, neither cars nor airplanes make a particularly good analogy to copyrights...
Obviously you don't run a mail server, and have no idea how much spam is being blocked for you. If your e-mail wasn't being filtered, you would either stop using it, or desperately try to find a way to filter it.
I have an unfiltered account that I manage Spambayes to filter it. Using that, I have full access to its results, can tune it as I like and can recover any messages that have been misfiltered. So far it has not yet misidentified legit email as spam, and I only get a couple of spam messages a day that slip through. I've had the email account for years, and it's apparently made it to plenty of spam mailing lists. Before I went to Spambayes I had a complex procmail filter that I finally got tired of adjusting-- but the point is I have *control* over it and I can check and recover when legit emails are miscategorized.
What would you suggest - that mail administrators employ a team of trained experts to sift through thousands upon thousands of messages every day, manually identifying what is spam and what is not?
No, simply that they leave it in the control of their users-- provide a choice of filtering options but do not impose any. If a users chooses to use Spamhaus, then great, but I've seen many services that a) don't give users a choice, and 2) don't provide access to the filtered messages. I've also seen accounts that attempt to use draconian measures of eliminating all mails that come from certain types of sending systems-- deemed to be "rogue" because they may not be configured to follow the latest "rules" of sending mail (and thereby eliminate a lot of legit emails from many systems in the process). The accounts quickly become unusable when about 1/4 to 1/2 of their incoming emails are bounced or filtered and no indication of this action is given to the mailbox owner that incoming messages have been rejected-- typically it is discovered when senders complain about no response or bounce messages.
The point is filtering must be a feature that can be completely monitored by the users-- an optional feature and a record of every filtered message must be available. I don't find filtering "for the users own good," or whining about the huge volume of SPAM in the world, to be particularly compelling excuses for introducing additional unreliability in email transmission.
I believe, and sincerely hope that it is merely a matter of time until a critically important email gets misrecognized as SPAM or as originating from a "rogue" service by an ISP who did NOT provide the feature optionally to their users, and is eventually discovered by a user who then files a lawsuit over it. Common carriers should not have the ability to invisibly and/or arbitrarily filter communications to its customers, especially when they do a lousy job of it (as IMHO, automated blacklists & greylists generally do).
The suit was mistargeted. Spamhaus doesn't force anyone to use it. It is the ISPs that impose it on email accounts, not Spamhaus, and consequently, THEY should be liable if they do not allow their users to disable such blocking. Use of Spamhaus contributes to email unreliability and should not be imposed by ISP services. An email account carries with it some expectation of usability, which IMHO cannot be simply TOS'ed away in the fine print. Email is unreliable enough without blacklist (or for that matter, even greylist) techniques being applied by lazy ISPs who are looking for a brainless way to reduce their email traffic load. Either ISPs are a common carrier or they are not, the imposition of blocking techniques should carry along with it some responsibilities for its failures.
I'd like to see the ability to completely turn off tabs in Firefox-- especially on Windows where tabbing is redundant (you already have tabs of all your available pages in the taskbar, or at least you *would* if half of them didn't end up on a Firefox tab). Tabbing means that you have to look more than one place to find all the pages you have open, and the inconsistency confuses users.
Slashdot Mirror
Spambayes works pretty good too-- I get no more than a couple of messages a day at the most, often none at all. Now if only I could convince my ISP to use it instead of the RBL and other such crap they're using now and I'd get all of the email from people I want to get it from...
EXACTLY. Very unreliable, not due to spam but to spam filtering. Frankly I prefer reliability with 90% spam to what we've got now. At least I'd get to choose my own filtering and have noone to blame but myself for choosing it, if it's lousy. But then, I'm not an ISP who cares more about his bandwidth costs than he does his customer's email reliability...
They're probably trying to retrain the spam filters, in preparation for their next volley...
What I like most about this scheme is that in order for it to work, it puts a huge onus on the copyright owners to proactively register their works with Google now, to an extent they probably didn't even have to with the US Copyright office. They can no longer complain to Google for lack of protection if they are too lazy to upload everything they own (and for most of the big majors, that's a lot of stuff). This lets Google off the hook and simplifies the takedown process for them, while giving the big media companies essentially what they asked for:
Google as new Copyright registrar. I love it!
With regards to matching-- I agree that accurate matching is probably not all that hard, you could probably reduce everything to 32x32 64 color pixels sampled @ 1 fps and fuzzy match the results and identify the vast majority of stuff-- and let humans check for false positives if necessary-- you could immediately have the system put a vid on "suspension" if it looks to be infringing until an eyeball gets the chance to look at it to confirm. Though I would guess that false positives could be pretty darn rare so that may not even be necessary. Things like uploading copyrighted vids backwards won't be useful to most people but could be included in the match anyway if it was a problem. Even completely shuffling the frame positions of the entire video may not be a good workaround, as you could sort the frames in some way and then compare them, then try them inverted or flipped or shifted-- the worst these operations might do in the long run is cause the matching algorithm to run longer as it tries more and more permutations that one might apply to the content...
On the other hand, Google may have to periodically rerun all the original content through their fingerprinting algorithm to regenerate the database as they tune up its accuracy, which given a huge database of registered works could end up being a resource consuming enterprise...
Butterfly nets...
If you want an open phone, there are several on the market or very close to market that will work MUCH better and the companies will support you in creating the apps. There's no need to hack the iPhone and Apple has cheated no one.
There's no need to climb Mount Everest either, but lots of people do it just the same...
The thing is, even if the RIAA win every case they bring, it isn't going to fix their problem, at least in the long term. It may justify their existence to their bosses for a little longer, but that's about it. As long as the RIAA think that this sort of thing is the answer to their problem it continues to underscore just how completely out of touch they and their bosses are...
The digital horses are out of the barn, and neither litigation or stringent DRM is going to put them back in. And, both of those things make one of their other problems much worse-- that of PR. Odd that, as didn't these guys start out in PR? Isn't PR essentially what the record companies were mostly about? Watching them completely trashing their raison d'etre seems rather like watching someone commit seppuku who thinks it a good way to fix their stomach ache...
These guys are working in RETAIL-- Hel-lo! Anyone with any real tech creds wouldn't be caught dead in retail...
I suppose the question remains-- where *does* one go to get their computer fixed? I can't help much with the answer to that though, as I've always fixed my own. But then, I hand soldered every last darn solder joint on the first "home computer" I ever owned (IMSAI) which dates me as a geezer...
If you really want your model to look good at higher resolutions, you can always subdivide your triangle meshes, which is robust (unlike triangle reduction). Up-resing meshes is easy. Down-resing takes artist work.
Subdividing triangles and smoothing the result is not really "up-resing." You get more polygons but not more detail. Adding fine detail to a subdivided mesh still requires artist work.
Either that or they subcontracted finding their targets to morons like those at Media-Defender...
Linux is FREE, you don't have to shell out your life savings in order to get it and use it effectively.
This is a really tired and lame argument in favor of Linux (FREE as in BEER). Only a moron would choose an operating system because it's free (as in beer). If Linux cost as much as Vista did and Vista was free, I'd still use Linux. For a lot of reasons, but for one thing, it runs on EVERYTHING-- a 15 year old PC, Mac hardware, the Power PC chip in my networked drive (Kurobox). When I need to automate on one I can write a script that will run everywhere-- write once run everwhere works for me and I don't even need Java. Then again, I like the command line as it allows me to do a lot of stuff in parallel. And for another thing, it's pretty darn reliable and I can easily make it conform to my workflow rather than me having to work around bogus assumptions (as is often true on the Mac), or buggy & inconsistent design (as is often true on Windows).
Choose Linux when it's BETTER, not because it's FREE (as in beer)...
Acutally, I have some rare CDs that have actually *increased* in value since I bought them. But with Amazon it's not all that hard to sell CDs-- you don't even have to scan them as someone else has already done it for you in most cases. And I'd rather trade-in a handful of old CDs that I don't want at a store in exchange for a couple of different ones than just piss the $$$ down the drain...
Even without DRM, I have a BIG problem with downloadable music. What's that you ask? You can't RESELL it. I buy a CD, even a used CD often for less than the $0.99/song that iTunes is charging, and if I later tire of it I can legally sell it to someone else. If I tire of a digital download, that's it. Just try selling your DRM-free bit collections on Amazon or eBay and see how far you get...
While I wouldn't buy a non-open device like this either, the problem is most customers don't know how to utilize openness so they don't care. As long as that is the case, there will be sufficient market-- in fact a LARGER market for dumbed-down devices like the iPhone than there will be for a more open device.
In fact, the very reason that the Mac is as successful as it is with it's dumbed-down interface is that a large number of people are more comfortable with a dumbed-down interface than they are with a power interface that expects them to make intelligent choices. Then again, the most popular OS is one designed by a guy named Gates who loves complex puzzles and seems to think we do too-- so there's hope at least...
DRM is like a lock on your door that someone ELSE owns the key to.
I dunno, am I the ONLY one that did not become overly paranoid about terrorists coming to blow me up?
No, but like me, you probably don't work for the government. There's been several stories of late that recount that Bush & Cheney were absolutely convinced that additional terrorist attacks were likely after 9/11. And of course, that places like Washington DC were high on the target list. And initially, much of the US Congress thought that too. So here we're talking significant institutional paranoia in the institutions that actually make and enforce laws, and are supposed to deal with defending against attacks. This paranoia has turned out to have been a significant overreaction. That the resultant actions were based on imagination and ignorance and fraught with significant bureaucratic incompetence should be no surprise, as that's simply business as usual...
Well, I'm still waiting for a computer keyboard who's symbol set is not based on repurposed 50-year-old teletype equipment-- and has real multiply, divide and negative symbols (among other things) instead of co-opting what was available at the time for them-- given that basic math is a lot of what we do on computers these days, we really ought to have those symbols on our keyboards by now. Actually the first computer keyboard I ever used *did* have those symbols-- an IBM 2741 with an APL set-- ever since then it's been a step backward and there's been little advancement since.
Recall that when Color TV came out it was important for the transmissions to be viewed on the existing B&W sets-- so the color signal was encoded in a way that was designed to be compatible, rather than precise (NTSC often referred to as Never Twice the Same Color because of it). My family didn't get their first color TV 'til the Joneses on the block were buying their first replacement when their first color TV's tube started fading out-- in the latter half of the 1960s. In addition, I don't recall being all that impressed by Color TV when visiting friends houses that had them-- certainly not enough to go home and clamor that we should get one too. My first "personal" TV (that I got for my room) was B&W when our family TV was color, and I preferred to watch the B&W set because a) it was mine and 2) I could choose what to watch on it-- far more important than color. And I find HDTVs in Best Buy seem to flicker like strobe lights when things move-- now referred to as "judder" I think-- and it doesn't impress me. Given that, and the fact that many people have their kids growing up on VHS tape home-movies, I expect it will be some time before the "masses" have sufficiently moved to digital sets. Sure, those who like to watch sports games (which doesn't include me) and the latest TV shows (also, me not included) may jump on as early adopters, but that's somewhat less than everyone...
Analog medium has long ago reached market saturation, and may corporations see $$$$$ in a switchover, but that doesn't mean we will all just act like sheep and drink their kool-aid...
Have them enter their reg info on your website and get their license key, where you insert the users info encrypted into the software via the key such that it identifies the legal owner on startup. That can discourage legal owners from letting copies get out, without hampering the actual software operation. While pirates can patch it out, and it may be possible to register with bogus info, this is a pretty easy thing to do and does not impair users as the code does not otherwise restrict the usage of the product.
Users who do not have web access can get a license code over the phone which will input the same info--- it is often done with a combination of a license key and name information which are tied together so the license info they have will only work if they enter their name correctly into the products licensing screen. For example, the license code can contain a CRC of the customer info for example-- the customer calls in wanting to turn on the software, you have them fill out the reg info in the product itself and give you the key that *it* generates based on the entry of their name info. You then input that key to your system, validate their info and give them a license key that only works on the system that has the user info entered as they did.
I think you get the general idea. Check the web for similar schemes and fine tune it to your product, but this is probably the best way to discourage the use of casual illegal copies. The crack will likely end up being to patch the startup screen to say "Licensed to TheCrackCrew" which will advertise that it's a cracked copy on startup...
Uhm... I am the copyright owner of some music that I make available for free. If any of this happened to be the music in question the RIAA cannot prosecute her because of it, simply because it's "obvious" she doesn't own the copyright herself. The court has no way to know, just because she does not "own" the copyright, that she nevertheless has or does not have the right to hold or distribute a copy...
"Not all conservatives are stupid, but it is true that most stupid people are conservative."
I guess that explains why the liberals are in favor of public education and the conservatives against.
Since you brought up cars...Imagine what would happen if I "bought" a brand new Ford and Ford then told me that I couldn't give rides to my family, friends, co-workers, etc. because I didn't have a multi-passenger license.
Good thing he didn't bring up airplanes. Pilot licenses (different sort of license, I suppose), do not always include the right to carry passengers, and the rationale is not exactly "brain-dead." The license here is to operate the vehicle, not a licence connected with the vehicle itself regarding what you can do with it. Consequently, neither cars nor airplanes make a particularly good analogy to copyrights...
Obviously you don't run a mail server, and have no idea how much spam is being blocked for you. If your e-mail wasn't being filtered, you would either stop using it, or desperately try to find a way to filter it.
I have an unfiltered account that I manage Spambayes to filter it. Using that, I have full access to its results, can tune it as I like and can recover any messages that have been misfiltered. So far it has not yet misidentified legit email as spam, and I only get a couple of spam messages a day that slip through. I've had the email account for years, and it's apparently made it to plenty of spam mailing lists. Before I went to Spambayes I had a complex procmail filter that I finally got tired of adjusting-- but the point is I have *control* over it and I can check and recover when legit emails are miscategorized.
What would you suggest - that mail administrators employ a team of trained experts to sift through thousands upon thousands of messages every day, manually identifying what is spam and what is not?
No, simply that they leave it in the control of their users-- provide a choice of filtering options but do not impose any. If a users chooses to use Spamhaus, then great, but I've seen many services that a) don't give users a choice, and 2) don't provide access to the filtered messages. I've also seen accounts that attempt to use draconian measures of eliminating all mails that come from certain types of sending systems-- deemed to be "rogue" because they may not be configured to follow the latest "rules" of sending mail (and thereby eliminate a lot of legit emails from many systems in the process). The accounts quickly become unusable when about 1/4 to 1/2 of their incoming emails are bounced or filtered and no indication of this action is given to the mailbox owner that incoming messages have been rejected-- typically it is discovered when senders complain about no response or bounce messages.
The point is filtering must be a feature that can be completely monitored by the users-- an optional feature and a record of every filtered message must be available. I don't find filtering "for the users own good," or whining about the huge volume of SPAM in the world, to be particularly compelling excuses for introducing additional unreliability in email transmission.
I believe, and sincerely hope that it is merely a matter of time until a critically important email gets misrecognized as SPAM or as originating from a "rogue" service by an ISP who did NOT provide the feature optionally to their users, and is eventually discovered by a user who then files a lawsuit over it. Common carriers should not have the ability to invisibly and/or arbitrarily filter communications to its customers, especially when they do a lousy job of it (as IMHO, automated blacklists & greylists generally do).
The suit was mistargeted. Spamhaus doesn't force anyone to use it. It is the ISPs that impose it on email accounts, not Spamhaus, and consequently, THEY should be liable if they do not allow their users to disable such blocking. Use of Spamhaus contributes to email unreliability and should not be imposed by ISP services. An email account carries with it some expectation of usability, which IMHO cannot be simply TOS'ed away in the fine print. Email is unreliable enough without blacklist (or for that matter, even greylist) techniques being applied by lazy ISPs who are looking for a brainless way to reduce their email traffic load. Either ISPs are a common carrier or they are not, the imposition of blocking techniques should carry along with it some responsibilities for its failures.
Perhaps it would be useful as a defense in an RIAA/MPAA lawsuit-- "It wasn't me! Comcast must have forged those packets!"
I'd like to see the ability to completely turn off tabs in Firefox-- especially on Windows where tabbing is redundant (you already have tabs of all your available pages in the taskbar, or at least you *would* if half of them didn't end up on a Firefox tab). Tabbing means that you have to look more than one place to find all the pages you have open, and the inconsistency confuses users.