One of my peeves with IPv6 is that in v4 I had over 16 million legal loopback addresses out of only 4 billion addresses; now in v6 I have exactly one out of a much larger pool. It is not often useful, but it isn't always useless to use more than one of the loopback addresses on a host.
I would have preferred loopback to be a/64 rather than a/128 in IPv6: it's not like the address-space is too small to afford it.
You are conflating ignorance of the law with ignorance of the facts (and you appear to have copious amounts of both).
A pre-existing patent is not law, it is fact. IBM advises its employees to remain ignorant of other patents while working on products because if those products are later found infringing then their liability is reduced.
I was a Master Inventor at IBM before I quit in 2010 and we absolutely did check for pre-existing patents before filing anything new. Occasionally this would lead to wasted effort within IBM because a proposal would get further through the process than it would if the inventors had looked for relevant patents before starting work.
It will probably stream OpenGL commands, not rendered images.
Doubtful.
If you can stream one rendered image every 16ms you can display 60fps. If you send OpenGL commands you will be better-off on a lot of frames but it will stutter every time you need to send a texture larger than a rendered frame (and most textures are larger than a rendered frame).
The game is to make the peak frame latency 16ms (or whatever your target is), not to reduce the average.
the GSM spec contains a special call type 'emergency' which is meant to be triggered when you press that, or dial 112 or 999 (or presumably 911) so the digits really don't matter to the network. The idea was that those calls could kick someone else off the network if it was congested, for an emergency. I don't believe it was ever implemented though.
It was implemented. In the UK at least if you call an emergency number when you have a weak signal it will dramatically improve for the duration of the call as the cell tower reconfigures itself to use up to its maximum power and, as you say, drops any other call that was interfering with the call placed by your handset.
The towers are smart enough not to drop any calls or boost the power unless it will help.
If being an atheist, Christian or Jedi ever becomes a matter that gets you brought to the attention of the Secret Service in the UK, we're beyond fucked already.
The forms linked from this page give you a good idea of what will bring you to the attention of SIS and friends.
b) When a company sells digital software themselves, they don't get to keep 100% of the sale price. They have to pay for hosting, bandwidth, marketing, sales processing, manhours involved in all of this, etc., etc., etc.
I own a small business selling apps on the iOS store and I agree that 30% is a reasonable price for what they provide.
Apple do nothing to market or promote your apps, so you should not have included that. You missed a relatively important factor in that they test and validate the apps they sell which gives consumers more confidence to purchase than they have on competitor's app stores.
This sounds reasonable to me - the telephone company has no business filtering phone calls, so it should not filter text messages either. Subscribers may choose to employ a spam-blocking service, which could be provided by other people than the phone company.
I agree with this in principle, but I would add that it is a matter between the subscribers and the telephone company; I do not see that any third party has grounds to interfere in that relationship.
If the telephone company's scheme is optional, subscribers are given a clear choice, and subscribers may change their choice at any time without penalty then I would say it's fine. Failing that, if subscribers have a choice of telcos and at least one of them offers an unfiltered option there is no need for any regulator to get involved.
Undoing moderation; please ignore this post
on
Running Netflix On Linux
·
· Score: 2, Insightful
It should be relatively easy to bust the myth: what are the capacities of the USB drives? If they are all >4Gb I don't buy that they are from 3 years ago. Drive make and model would also give you a fairly firm point for the earliest point in time at which they could have been purchased.
The only trouble with the America Invents Act of 2011 is it is not an Amendment to the Constitution, which it needs to be in order to change the original text of the Constitution which clearly specifies (re: "Inventors") the "first to invent" system over the "first to file" system.
No it doesn't. It specifies that the rights over their discoveries shall go to the Inventors but it nowhere specifies the process that inventors shall be required to undergo to secure those rights.
Under first-to-file as it is practiced elsewhere it is still illegal for anyone other than the inventor to secure a patent on a discovery. It is really a very minor change which only makes a difference in the case of near-contemporaneous discoveries. It mostly benefits accidental inventors who are less likely to be able to provide any evidence for the date of invention than the industrial-scale patent-generators who receive most patents.
A secondary effect which may turn out to be rather more important is that the current US system requires that you file within 1 year of the discovery or never file for patent protection; so an invention cannot be kept a trade-secret and used for 40 years before being patented by that same person/corporation when they feel there is a risk of someone else working it out, effectively securing them a 65-year monopoly. First-to-file elsewhere generally allows this; I am not familiar with the US statute in question.
You did your testing on a pre-production test domain, and you have a managed code base that allows for easy transition to other domain names?
Why add in an unnecessary risk? If third-party code (unbeknownst to you) names its tables with the domain name you can test it all you like on a pre-production name but as soon as you flip to the real name it will stop working. And the first time you will know about it is after you flip the switch and your customers complain.
Or are you seriously suggesting you create an entire new server environment to make modifications to your existing web sites?
Pretty-much everyone I know does this; you make a new VM in EC2 or Peer1 or Rackspace or whatever, install and configure everything under its final name, test it, then switch the DNS entries.
Once you are certain the DNS changes have propagated everywhere they're going to get (wait TTL * 2 after the change) you stop and delete the previous VMs.
If you are identifying people from a population of 30 million you need ceil(log2(30 000 000)) bits for your person identifier; which is 25 bits in this case. However you are likely to need to identify corporations as distinct from persons, which will probably take another bit or so. 26 bits per trading entity into a 24-byte (192-bit) TAC goes 7.4 times.
No matter how you put those IDs into the TAC you can never fit more than 7.4 at a time. So if you are a criminal (or privacy nut) who wants to use this system, make sure there are 8 trades between you and any other party you interact with if you want deniability if someone has access only to the TAC used for the final transaction to you. This is not a very plausible tracking scheme because for practical reasons you will need a timestamp and other gubbins to be encoded in the TAC.
Of course, if you have access to all the TACs you only need to fit two IDs in there at a time to build a chain. This is IMO very plausible.
You are probably right; however it does depend on how rapidly it is broken down and excreted by the body. If it is never got rid of, then this test is using a massive under-dose (from my point of view), since I have certainly swallowed 2 tubes-worth of Colgate Total over the course of my life.
If it is eliminated in 12 hours it is a huge over-dose as you say. Somewhere in between these points is an elimination-rate which makes this number entirely appropriate.
Slashdot Mirror
I'm not OP but I've bought a secondhand car at £14500 on a Chip & PIN card here in the UK. Wasn't even called to confirm.
So you run cc -E first then
Illegal means "in violation of the law".
All crime is illegal, not all illegality is crime.
+1, accurate.
One of my peeves with IPv6 is that in v4 I had over 16 million legal loopback addresses out of only 4 billion addresses; now in v6 I have exactly one out of a much larger pool. It is not often useful, but it isn't always useless to use more than one of the loopback addresses on a host.
I would have preferred loopback to be a /64 rather than a /128 in IPv6: it's not like the address-space is too small to afford it.
I have never seen a complete and unambiguous requirement for any real piece of software, and I would wager no other software engineer has either.
You are conflating ignorance of the law with ignorance of the facts (and you appear to have copious amounts of both).
A pre-existing patent is not law, it is fact. IBM advises its employees to remain ignorant of other patents while working on products because if those products are later found infringing then their liability is reduced.
I was a Master Inventor at IBM before I quit in 2010 and we absolutely did check for pre-existing patents before filing anything new. Occasionally this would lead to wasted effort within IBM because a proposal would get further through the process than it would if the inventors had looked for relevant patents before starting work.
It will probably stream OpenGL commands, not rendered images.
Doubtful.
If you can stream one rendered image every 16ms you can display 60fps. If you send OpenGL commands you will be better-off on a lot of frames but it will stutter every time you need to send a texture larger than a rendered frame (and most textures are larger than a rendered frame).
The game is to make the peak frame latency 16ms (or whatever your target is), not to reduce the average.
Actually the old-skool unit of mass is the slug, not the gram.
What I find odd is this: "stifle" is a relatively obscure word to use and yet they can't spell "dissent".
A very reliable rule of thumb is that if there are N possible hash values, then you should expect 0.5 hash collisions after hashing N items.
No you shouldn't. You should expect 0.5 collisions after hashing N/2 items. Trivially you are guaranteed at least M collisions when hashing N+M items.
I don't actually know off the top of my head how many collisions you should expect if you hashed N items.
It might even be a foreign network if you're near a boarder.
If you're going to be connected to the emergency services of the wrong country anyway, I would say you have bigger problems than which number to dial.
the GSM spec contains a special call type 'emergency' which is meant to be triggered when you press that, or dial 112 or 999 (or presumably 911) so the digits really don't matter to the network. The idea was that those calls could kick someone else off the network if it was congested, for an emergency. I don't believe it was ever implemented though.
It was implemented. In the UK at least if you call an emergency number when you have a weak signal it will dramatically improve for the duration of the call as the cell tower reconfigures itself to use up to its maximum power and, as you say, drops any other call that was interfering with the call placed by your handset.
The towers are smart enough not to drop any calls or boost the power unless it will help.
If being an atheist, Christian or Jedi ever becomes a matter that gets you brought to the attention of the Secret Service in the UK, we're beyond fucked already.
The forms linked from this page give you a good idea of what will bring you to the attention of SIS and friends.
I agree that a 30% cut is a bit too much...
b) When a company sells digital software themselves, they don't get to keep 100% of the sale price. They have to pay for hosting, bandwidth, marketing, sales processing, manhours involved in all of this, etc., etc., etc.
I own a small business selling apps on the iOS store and I agree that 30% is a reasonable price for what they provide.
Apple do nothing to market or promote your apps, so you should not have included that. You missed a relatively important factor in that they test and validate the apps they sell which gives consumers more confidence to purchase than they have on competitor's app stores.
This sounds reasonable to me - the telephone company has no business filtering phone calls, so it should not filter text messages either. Subscribers may choose to employ a spam-blocking service, which could be provided by other people than the phone company.
I agree with this in principle, but I would add that it is a matter between the subscribers and the telephone company; I do not see that any third party has grounds to interfere in that relationship.
If the telephone company's scheme is optional, subscribers are given a clear choice, and subscribers may change their choice at any time without penalty then I would say it's fine. Failing that, if subscribers have a choice of telcos and at least one of them offers an unfiltered option there is no need for any regulator to get involved.
n/t
It should be relatively easy to bust the myth: what are the capacities of the USB drives? If they are all >4Gb I don't buy that they are from 3 years ago. Drive make and model would also give you a fairly firm point for the earliest point in time at which they could have been purchased.
It's worth more than all the computers and related hardware in the office combined.
Debatable. It cost more than all the computers and related hardware in the office combined for sure.
The only trouble with the America Invents Act of 2011 is it is not an Amendment to the Constitution, which it needs to be in order to change the original text of the Constitution which clearly specifies (re: "Inventors") the "first to invent" system over the "first to file" system.
No it doesn't. It specifies that the rights over their discoveries shall go to the Inventors but it nowhere specifies the process that inventors shall be required to undergo to secure those rights.
Under first-to-file as it is practiced elsewhere it is still illegal for anyone other than the inventor to secure a patent on a discovery. It is really a very minor change which only makes a difference in the case of near-contemporaneous discoveries. It mostly benefits accidental inventors who are less likely to be able to provide any evidence for the date of invention than the industrial-scale patent-generators who receive most patents.
A secondary effect which may turn out to be rather more important is that the current US system requires that you file within 1 year of the discovery or never file for patent protection; so an invention cannot be kept a trade-secret and used for 40 years before being patented by that same person/corporation when they feel there is a risk of someone else working it out, effectively securing them a 65-year monopoly. First-to-file elsewhere generally allows this; I am not familiar with the US statute in question.
However, SHA-2 could be broken tomorrow, and this time we won't have a decade's wait while a suitable replacement is designed.
As all road cyclists should know:
Mountain bike shoes and pedals have their place.
On a mountain bike.
http://www.velominati.com/the-rules/#34
Not quite sure what that has to do with Islam, but it's always useful to be reminded of the sacred texts.
Most of these whippersnappers have never even heard of the unsigned keyword.
You did your testing on a pre-production test domain, and you have a managed code base that allows for easy transition to other domain names?
Why add in an unnecessary risk? If third-party code (unbeknownst to you) names its tables with the domain name you can test it all you like on a pre-production name but as soon as you flip to the real name it will stop working. And the first time you will know about it is after you flip the switch and your customers complain.
Or are you seriously suggesting you create an entire new server environment to make modifications to your existing web sites?
Pretty-much everyone I know does this; you make a new VM in EC2 or Peer1 or Rackspace or whatever, install and configure everything under its final name, test it, then switch the DNS entries.
Once you are certain the DNS changes have propagated everywhere they're going to get (wait TTL * 2 after the change) you stop and delete the previous VMs.
If you are identifying people from a population of 30 million you need ceil(log2(30 000 000)) bits for your person identifier; which is 25 bits in this case. However you are likely to need to identify corporations as distinct from persons, which will probably take another bit or so. 26 bits per trading entity into a 24-byte (192-bit) TAC goes 7.4 times.
No matter how you put those IDs into the TAC you can never fit more than 7.4 at a time. So if you are a criminal (or privacy nut) who wants to use this system, make sure there are 8 trades between you and any other party you interact with if you want deniability if someone has access only to the TAC used for the final transaction to you. This is not a very plausible tracking scheme because for practical reasons you will need a timestamp and other gubbins to be encoded in the TAC.
Of course, if you have access to all the TACs you only need to fit two IDs in there at a time to build a chain. This is IMO very plausible.
You are probably right; however it does depend on how rapidly it is broken down and excreted by the body. If it is never got rid of, then this test is using a massive under-dose (from my point of view), since I have certainly swallowed 2 tubes-worth of Colgate Total over the course of my life.
If it is eliminated in 12 hours it is a huge over-dose as you say. Somewhere in between these points is an elimination-rate which makes this number entirely appropriate.