Modelling is great but at some point you have to validate the model or you risk compounding problems.
Aircraft manufacturers use modelling all the time to design aircraft to regulatory requirements, but they still have to validate that model with a wing bend and a fatigue test airframe before the FAA or EASA will certify the aircraft. A failed test will still validate the model, just adjusted for te real world results (the Airbus A380 failed its wing bend at something like 147%, 3% down on requirements, but they were allowed to validate the fix based on the proven adjusted model).
And of course models are only valid while you are dealing with a bomb design you already understand - what about we bomb designs? New implosion patterns? New materials? Models only go so far.
This is an attempt to stop ongoing production of newer designs of weapon.
Oh, it's much worse than just getting away with an article - we (the western world) allowed a terrorist organisation to take control of a country because we felt bad for what the previous regime had done to blacks in that country.
You misunderstand - the license of the content Boundless use is not at issue here, its the (for want of a better way of putting it) creation of a "derivative work" of a third parties textbook by using that content in the exact same manner as the third party does in its text.
Boundless are doing it on demand, and using the third party text as a reference for creating their own version - that is what the third party is taking issue with.
I was going to post something similar to Kalriath, but my iPad ate my Safari instance last night, so here we go again...
They didn't validate the entire string, because thats not possible with the check numbers involved - only the card issuer or the payment networks can actually tell you if its a valid card number and they wouldn't tell you unless you tried to present the card. You can have a card number which validates as a card type, and internally validates as a correctly formed number, but those checks are actually fairly loose and its not necessarily a matter of fact that the number is a valid live number.
In any large raw dataset, its trivial to find one or two number strings which validate to a card issuer.
The problem is, they haven't actually verified that what they have is an actual credit card number, they've just pulled a number out that happens to validate and have the same starting digits as a card type but there is no related information - so why would the credit card number on its own find it's way into these streams and not the other details off the card.
At the moment, they found a number, that's it. What would be an actual test is to use an Xbox, use a card on that Xbox, and then see if you can recover that card from that Xbox - that's not what they did, so the results can't be validated.
The XBox just throws an error to the user if it loses connection to Live while you are doing something - thats why I had to add my card details via the website (XBox lost connection to the local wifi for a bit).
The way I see this statement from Microsoft is "well, if all the processes are followed correctly by our developers, we don't see this happening, so its unlikely. However, there is a chance that a developer may have used the wrong caching or serialisation library for this routine which may have inadvertently left traces on the XBoxes hard disk, so we are going to look into it."
I entered my card details on the XBox Live website directly, not via my Xbox - I don't see why Microsoft would deliberately store the card details in two places if you entered it on an XBox, when the card authorisation has to be done by the remote servers anyway, so thats why I'm personally leaning to the above understanding.
Also, it was noted in the last story about this that the example credit card number given as "successfully retrieved" was not of a type accepted by XBox Live as a payment source...
I agree that it does change the light the poster is in, but the "mistake" is in no way understandable - Learning Together should never be able to transfer a domain out of someone elses account with anything less than either explicit permission from that account holder, or a court order.
Its that simple - its not their account, and any disputes or ownership issues reside with the account owner, Rackspace have absolutely no standing here to be doing anything on that account for anyone other than the account owner. That should have been checked first and foremost - it does not matter what documentation Learning Together have, if it doesn't authenticate them as the account owner then they don't get to do anything.
Thats the mistake Rackspace made, and its not a small one and nor is it an understandable one - they allowed a non-account-owner to make changes to the account. Thats wrong on so many levels.
Whether he purchased the domain for them or not, thats an ownership dispute between him and Learning Together, not something for Rackspace to resolve arbitrarily.
I do wonder just how bad the US cable market really is - I have Sky in the UK, and I pay roughly $75 a month for TV, Broadband (unlimited 20Mbit ADSL) and voice with free evening and weekend calls.
The Sky subscription gets me Anytime+, which gives me access to tonnes of tv and movies for viewing at any time. It also gives me SkyGo, which is essentially the same as Anytime+ but available on computer, tablets and phones - I have no need for Netflix or Lovefilm. Sky also have a suite of apps to interact with my dvr and do other things.
This really doesn't have anything to do with "the cloud" - its plain old hosting, and everyone needs a domain registrar in order to hold a domain. In this case the registrar fucked up and allowed access to an account they shouldn't.
You don't know the business relationship there - regardless, under no circumstance short of court order does Rackspace have the right to arbitrarily grant Learning Together access to an account they do not have authorisation to access. Rackspace did bad here, however you look at it.
... And the N900 lovers come out of the woodwork - nice modding there guys, just because i consider your pet hardware to be shite.
I can post photos later on of the second N900 she used, with most of the keyboards plastic covering rubbed off through normal usage. She's not abusive to her phones (her phones prior to and post the N900 haven't suffered in any way) and yet she's had major issues with two of them including a broken slide action on the first device. She does use it a lot tho.
The preference between a physical and onscreen keyboard is just that, a preference, most certainly not "matter of fact" one way or the other. I don't prefer one over the other.
I just really didn't get on with the N900, despite sticking with it for ages before I switched to someone else, and I can't see what others see in it.
You would still have to "seed" the search area with identifiable items so that your test is proper, and clear the search area for the "no results" outcome, rather than simply relying on whatever is drifting around when you got there.
You know, actually engineer a proper test environment...
He's 'revered' (your choice of words, not mine) because of *how* he went on to make all that money and drag Apple from the brink of bankruptcy and being a failed company to one of the largest and most successful companies in the world in just over a decade and a half.
It takes a die hard hater to not see the brilliance in his management of Apple from 1996 onward. He controlled the company a lot tigher than most CEOs would but made it work, and put in place a workplace environment that allowed other very significant minds to flourish (Ives is a heck of a good example - basically untouchable within Apple, reported to Jobs and had as much freedom as Jobs).
Slashdot Mirror
Once SpaceX actually put someone into orbit, then lets talk - but trumpeting them now is just getting ahead of yourself.
Thats because in a class action lawsuit, the lawyers are the ones doing the work for you...
Why shouldn't they get paid? You can always bring your own suit against the plaintiffs rather than take part in their suit....
Just a note, but Russia has the largest stockpile, not the US.
Modelling is great but at some point you have to validate the model or you risk compounding problems.
Aircraft manufacturers use modelling all the time to design aircraft to regulatory requirements, but they still have to validate that model with a wing bend and a fatigue test airframe before the FAA or EASA will certify the aircraft. A failed test will still validate the model, just adjusted for te real world results (the Airbus A380 failed its wing bend at something like 147%, 3% down on requirements, but they were allowed to validate the fix based on the proven adjusted model).
And of course models are only valid while you are dealing with a bomb design you already understand - what about we bomb designs? New implosion patterns? New materials? Models only go so far.
This is an attempt to stop ongoing production of newer designs of weapon.
Heh, you know what? I'm sat here now thinking "I shouldn't have posted that because now I'm going to get labeled as a racist..."
Funny eh?
Oh, it's much worse than just getting away with an article - we (the western world) allowed a terrorist organisation to take control of a country because we felt bad for what the previous regime had done to blacks in that country.
Yes, I'm talking about the ANC.
You misunderstand - the license of the content Boundless use is not at issue here, its the (for want of a better way of putting it) creation of a "derivative work" of a third parties textbook by using that content in the exact same manner as the third party does in its text.
Boundless are doing it on demand, and using the third party text as a reference for creating their own version - that is what the third party is taking issue with.
And what if they own one of the large CAs?
It's as much Linux as an Xbox is a Windows platform.
You overlook "at the base"... The USN has lost more than 25 planes in total over the past 5 years alone.
At the moment (iOS5.1) I cannot highlight anything on the page outside of a text area so I can't copy and paste...
I was going to post something similar to Kalriath, but my iPad ate my Safari instance last night, so here we go again...
They didn't validate the entire string, because thats not possible with the check numbers involved - only the card issuer or the payment networks can actually tell you if its a valid card number and they wouldn't tell you unless you tried to present the card. You can have a card number which validates as a card type, and internally validates as a correctly formed number, but those checks are actually fairly loose and its not necessarily a matter of fact that the number is a valid live number.
In any large raw dataset, its trivial to find one or two number strings which validate to a card issuer.
The problem is, they haven't actually verified that what they have is an actual credit card number, they've just pulled a number out that happens to validate and have the same starting digits as a card type but there is no related information - so why would the credit card number on its own find it's way into these streams and not the other details off the card.
At the moment, they found a number, that's it. What would be an actual test is to use an Xbox, use a card on that Xbox, and then see if you can recover that card from that Xbox - that's not what they did, so the results can't be validated.
The XBox just throws an error to the user if it loses connection to Live while you are doing something - thats why I had to add my card details via the website (XBox lost connection to the local wifi for a bit).
The way I see this statement from Microsoft is "well, if all the processes are followed correctly by our developers, we don't see this happening, so its unlikely. However, there is a chance that a developer may have used the wrong caching or serialisation library for this routine which may have inadvertently left traces on the XBoxes hard disk, so we are going to look into it."
I entered my card details on the XBox Live website directly, not via my Xbox - I don't see why Microsoft would deliberately store the card details in two places if you entered it on an XBox, when the card authorisation has to be done by the remote servers anyway, so thats why I'm personally leaning to the above understanding.
Also, it was noted in the last story about this that the example credit card number given as "successfully retrieved" was not of a type accepted by XBox Live as a payment source...
I agree that it does change the light the poster is in, but the "mistake" is in no way understandable - Learning Together should never be able to transfer a domain out of someone elses account with anything less than either explicit permission from that account holder, or a court order.
Its that simple - its not their account, and any disputes or ownership issues reside with the account owner, Rackspace have absolutely no standing here to be doing anything on that account for anyone other than the account owner. That should have been checked first and foremost - it does not matter what documentation Learning Together have, if it doesn't authenticate them as the account owner then they don't get to do anything.
Thats the mistake Rackspace made, and its not a small one and nor is it an understandable one - they allowed a non-account-owner to make changes to the account. Thats wrong on so many levels.
Whether he purchased the domain for them or not, thats an ownership dispute between him and Learning Together, not something for Rackspace to resolve arbitrarily.
I do wonder just how bad the US cable market really is - I have Sky in the UK, and I pay roughly $75 a month for TV, Broadband (unlimited 20Mbit ADSL) and voice with free evening and weekend calls.
The Sky subscription gets me Anytime+, which gives me access to tonnes of tv and movies for viewing at any time. It also gives me SkyGo, which is essentially the same as Anytime+ but available on computer, tablets and phones - I have no need for Netflix or Lovefilm. Sky also have a suite of apps to interact with my dvr and do other things.
How does that compare to the US?
Domain names are *always* going to be in the hands of someone else...
This really doesn't have anything to do with "the cloud" - its plain old hosting, and everyone needs a domain registrar in order to hold a domain. In this case the registrar fucked up and allowed access to an account they shouldn't.
You don't know the business relationship there - regardless, under no circumstance short of court order does Rackspace have the right to arbitrarily grant Learning Together access to an account they do not have authorisation to access. Rackspace did bad here, however you look at it.
... And the N900 lovers come out of the woodwork - nice modding there guys, just because i consider your pet hardware to be shite.
I can post photos later on of the second N900 she used, with most of the keyboards plastic covering rubbed off through normal usage. She's not abusive to her phones (her phones prior to and post the N900 haven't suffered in any way) and yet she's had major issues with two of them including a broken slide action on the first device. She does use it a lot tho.
The preference between a physical and onscreen keyboard is just that, a preference, most certainly not "matter of fact" one way or the other. I don't prefer one over the other.
I just really didn't get on with the N900, despite sticking with it for ages before I switched to someone else, and I can't see what others see in it.
I had the N900, used it for several months, I don't consider it worth owning at all - and I can't understand why people are so enthusiastic about it.
Terrible interface, terrible screen, terrible keyboard, hardware wasnt at all resilient (the wifes one died physically after 6 months of usage).
You would still have to "seed" the search area with identifiable items so that your test is proper, and clear the search area for the "no results" outcome, rather than simply relying on whatever is drifting around when you got there.
You know, actually engineer a proper test environment...
He's 'revered' (your choice of words, not mine) because of *how* he went on to make all that money and drag Apple from the brink of bankruptcy and being a failed company to one of the largest and most successful companies in the world in just over a decade and a half.
It takes a die hard hater to not see the brilliance in his management of Apple from 1996 onward. He controlled the company a lot tigher than most CEOs would but made it work, and put in place a workplace environment that allowed other very significant minds to flourish (Ives is a heck of a good example - basically untouchable within Apple, reported to Jobs and had as much freedom as Jobs).
Cuba have asked the US to leave on several occasions, but the US refuse to comply.