FWIW, it's my understanding that Scala strongly emphasizes immutable data, so unless that's suitable to your use case it may be a poor choice. And if it *is* suitable, then you should give Erlang a look.
OTOH, it you want to interface with a lot of Java code, it may be a reasonable option.
The problem is allergies are problems with the fine-tuning of the immune system. Anything which massively inhibits this is going to cause massive problems...perhaps up to the "bubble-boy syndrome" level.
OTOH, I can't tell from the summary, but this might be something that eliminated immune response against some particular (set of?) proteins. That might be really impressive and useful. But it would also need to be tuned to particular allergens that were to be considered "not dangerous". So you'd need a large selection of targeted drugs. Still a great advance.
Otherwise this is a bit of highly important research, but it won't lead directly to anything usefully applicable (unless you want to do indirect high-tech assassination). But even so isolation of the mechanism should allow development of more nuanced approaches.
Until the manufacturers bear liability for the problems they enable.
The problem with that is, unless they also get the benefit of the benefits then enable, then they'll just stop selling things. Whoops! (For some of these manufactures I'd count that a win, e.g. MS, but mine is not the majority opinion.)
Why do you think popular opinion decides what is true?
I can see lots of reasons for being dubious about how accurately this tag will be applied, but disagreeing with popular opinion sure isn't one of them.
But that's exactly how it should be done. You post a label, and people can decide how trustworthy the label itself is. You don't keep people from seeing statements that you disagree with, you merely get out front and say "Well, I think these are lies, but here you go...".
IIRC, back when I bought Apple products I never even SAW the EULA until after I'd paid for it. And they altered the EULA in a "Security update" to something I found unacceptable.
Since then I haven't bought or recommended any Apple products. And I'm not likely to unless they make a legally binding promise to never alter the EULA of something that is purchased from them. Even then I probably wouldn't because enforcing the "legally binding promise" isn't cheap or easy, and I wouldn't trust them to keep the promise unless it was enforced frequently and expensively.
P.S.: This is *not* an argument in favor of Microsoft. I stopped trusting them before I stopped trusting Apple.
There is a case where an "undefined language condition" caused a C++ program to destroy a computer. (Well, such was reported in, I believe, "Effective Modern C++", though without sufficient data to track down the actual case.)
Given that, there's nothing unreasonable in saying your computer was destroyed by an MSWindows "upgrade". I don't know that it's true, but it isn't intrinsically unreasonable. More likely, the system could consume so much disk space that a complete version couldn't be written to disk, resulting in an unbootable system. In that case a geek could retrieve any data that wasn't overwritten by the installer. Now whether the user data *would* be overwritten by the installer I can't say. I expect you could probably recover any non-encrypted data by mounting the disk in another computer...but if the partition were encrypted (does mswindows do that?) then it might well be irrecoverable. Or if it were a RAID setup. (Again my ignorance of MSWindows is showing.)
They probably won't use Oreo, because that names a trademark. (You could argue that this is another area of business...but better to avoid legal wrangles.) Orangeade would probably be safe, but it's not similar to marshmallow or nougat. I suppose they could just use orange. (That would let them use quince for "q". But I still don't see how they'd handle 'x'.)
OK. I'll say color management was introduced on the Apple ][.
OTOH, this is a really stupid argument. Names are just names, and two guys named Harry can be quite different, yet both really named Harry. I don't know what the MSWind color management system is like, or what the Android color management system is like, but even if they're quite different they can both reasonably be called by the same name if they do *something* having to do with managing colors.
What you have said is, in theory, correct. There are, however, cases that cast a lot of doubt on how that theory is actually applied in practice. Naturally we probably only notice outliers, but those outliers *do* exist, so there is a definite chance that the judge could take a large number of years to decide that he's "unpersuadable".
Now the standard of "beyond a reasonable doubt" also has a large number of counter-examples, especially when the accused is being accused of something that most people consider horrendous. In fact, if the accusation is vile enough, many people won't even consider whether the evidence has any validity, or whether it could have been faked, or....well, much of any mitigating factor. He is being accused of obstruction of justice, which may or many not be true, and he will be punished extensively before there's ever a trial at which he, presumably, will have a fair defense. Many, however, never receive a fair defense, or even an only moderately poor defense. And even if he's found not guilty he will already have been punished extensively.
I don't know what an ideal way of handling things would be, but don't fantasize that we in the US have something even coming close to something fair to those who are poor or unpopular.
You are making the presumption that he is guilty. This may or may not be true. He might be innocent and actually have forgotten his password.
OTOH, just consider, if he gives them his password, they will be able to implant any evidence they choose onto his disks. Whoops! Forging dates isn't that hard.
Sorry, but that's not true. I know I've forgotten my own password a few times. I had to reinstall and recover from backups. (I really don't like sudo, and I also don't like logging in as root, and a decade or so ago I forgot the root password twice. It was really annoying, but not a real problem as one of the times I'd been thinking about switching distros anyway. The other time was more annoying, but I had the original CDs, and there hadn't been THAT many updates. [I said it was over a decade ago. I think it was while I was using Red Hat Professional edition, before I switched to KRUD Linux.])
That would be destruction of evidence...if done by the accused. After a warrant was served.
OTOH, if it's set to do this after n failed attempts, then it could quite likely be done by inept police, who didn't bother to image the disk before working on it. And then it's the police who destroyed the evidence...if such it was.
That said, I really doubt that applies in this case. But I know *I've* forgotten passwords, and had to reinstall, and recover from backups. So "I've forgotten the password" sounds possible, though unconvincing.
So did I. In fact I think I read about a new COBOL compiler just last year. That doesn't keep it from being a lousy language (with a few good features).
COBOL was a horrible language even for the day. Fortran was much better...but it was harder for the bosses to pretend to understand. (OTOH, COBOL was decent at some formatting tasks compared to Fortran, and I believe it included BCD numbers, where in Fortran you needed to use a library for that.)
It's true that there's a lot of ancient COBOL code still live, and I attribute it to the fact that nobody can really understand it, but, unlike assembler, it didn't automatically die when you change processors.
That's the theory, but if you follow the news stories there are repeated examples where the police act in ways that are grossly beyond their legal rights. Including confiscating and searching cameras (and phones).
Considering the story this page is attached to, your comment in indefensible stupidity. It probably would be possible to rephrase it into something that was defensible, but I can't do it for you because I don't understand what your point was.
IIRC Saddam Hussein *did* have weapons of mass destruction. Just not the ones they accused him of having. (I think he had lots of poison gas...but I don't really remember, that could have been some other bloke. But he certainly had napalm.)
A point, but copyrights don't protect implementations. They don't even protect you against significant rephrasing...or at least they didn't used to. Making an implementation isn't the same thing as publishing.
To me it seems the most applicable law is the one against unauthorized computer access, but that might not affect Uber, but only their employee, as it might not be possible to prove that he was acting as their agent in his actions. Of course, then you might go after Uber for receiving stolen merchandise....but copying isn't stealing, it's copyright violation, which is a separate crime. Etc.
So I think I see why Google is first trying the path they're trying.
I could come up with differences, but my first attempt didn't make Google look any better than Uber on this ground, so what's your selection of differences.
IIRC there once was a virus that included a copyright. And if I further recall correctly it even stated a particular open source license (GPL, I think) but it didn't include the text of the license. It wasn't a particularly destructive or successful virus, though.
P.S.: Don't go around believing news stories that don't have valid evidence. For all we know the code could be an officially ordered leak.
Slashdot Mirror
FWIW, it's my understanding that Scala strongly emphasizes immutable data, so unless that's suitable to your use case it may be a poor choice. And if it *is* suitable, then you should give Erlang a look.
OTOH, it you want to interface with a lot of Java code, it may be a reasonable option.
The problem is allergies are problems with the fine-tuning of the immune system. Anything which massively inhibits this is going to cause massive problems...perhaps up to the "bubble-boy syndrome" level.
OTOH, I can't tell from the summary, but this might be something that eliminated immune response against some particular (set of?) proteins. That might be really impressive and useful. But it would also need to be tuned to particular allergens that were to be considered "not dangerous". So you'd need a large selection of targeted drugs. Still a great advance.
Otherwise this is a bit of highly important research, but it won't lead directly to anything usefully applicable (unless you want to do indirect high-tech assassination). But even so isolation of the mechanism should allow development of more nuanced approaches.
Until the manufacturers bear liability for the problems they enable.
The problem with that is, unless they also get the benefit of the benefits then enable, then they'll just stop selling things. Whoops! (For some of these manufactures I'd count that a win, e.g. MS, but mine is not the majority opinion.)
Why do you think popular opinion decides what is true?
I can see lots of reasons for being dubious about how accurately this tag will be applied, but disagreeing with popular opinion sure isn't one of them.
But that's exactly how it should be done. You post a label, and people can decide how trustworthy the label itself is. You don't keep people from seeing statements that you disagree with, you merely get out front and say "Well, I think these are lies, but here you go...".
IIRC, back when I bought Apple products I never even SAW the EULA until after I'd paid for it. And they altered the EULA in a "Security update" to something I found unacceptable.
Since then I haven't bought or recommended any Apple products. And I'm not likely to unless they make a legally binding promise to never alter the EULA of something that is purchased from them. Even then I probably wouldn't because enforcing the "legally binding promise" isn't cheap or easy, and I wouldn't trust them to keep the promise unless it was enforced frequently and expensively.
P.S.: This is *not* an argument in favor of Microsoft. I stopped trusting them before I stopped trusting Apple.
To me it sounds like there wasn't enough spare space, and you ended up with an incomplete install that was unbootable *because* it was incomplete.
Linux and Mac users?
There is a case where an "undefined language condition" caused a C++ program to destroy a computer. (Well, such was reported in, I believe, "Effective Modern C++", though without sufficient data to track down the actual case.)
Given that, there's nothing unreasonable in saying your computer was destroyed by an MSWindows "upgrade". I don't know that it's true, but it isn't intrinsically unreasonable. More likely, the system could consume so much disk space that a complete version couldn't be written to disk, resulting in an unbootable system. In that case a geek could retrieve any data that wasn't overwritten by the installer. Now whether the user data *would* be overwritten by the installer I can't say. I expect you could probably recover any non-encrypted data by mounting the disk in another computer...but if the partition were encrypted (does mswindows do that?) then it might well be irrecoverable. Or if it were a RAID setup. (Again my ignorance of MSWindows is showing.)
They probably won't use Oreo, because that names a trademark. (You could argue that this is another area of business...but better to avoid legal wrangles.) Orangeade would probably be safe, but it's not similar to marshmallow or nougat. I suppose they could just use orange. (That would let them use quince for "q". But I still don't see how they'd handle 'x'.)
OK. I'll say color management was introduced on the Apple ][.
OTOH, this is a really stupid argument. Names are just names, and two guys named Harry can be quite different, yet both really named Harry. I don't know what the MSWind color management system is like, or what the Android color management system is like, but even if they're quite different they can both reasonably be called by the same name if they do *something* having to do with managing colors.
What you have said is, in theory, correct. There are, however, cases that cast a lot of doubt on how that theory is actually applied in practice. Naturally we probably only notice outliers, but those outliers *do* exist, so there is a definite chance that the judge could take a large number of years to decide that he's "unpersuadable".
Now the standard of "beyond a reasonable doubt" also has a large number of counter-examples, especially when the accused is being accused of something that most people consider horrendous. In fact, if the accusation is vile enough, many people won't even consider whether the evidence has any validity, or whether it could have been faked, or....well, much of any mitigating factor. He is being accused of obstruction of justice, which may or many not be true, and he will be punished extensively before there's ever a trial at which he, presumably, will have a fair defense. Many, however, never receive a fair defense, or even an only moderately poor defense. And even if he's found not guilty he will already have been punished extensively.
I don't know what an ideal way of handling things would be, but don't fantasize that we in the US have something even coming close to something fair to those who are poor or unpopular.
You are making the presumption that he is guilty. This may or may not be true. He might be innocent and actually have forgotten his password.
OTOH, just consider, if he gives them his password, they will be able to implant any evidence they choose onto his disks. Whoops! Forging dates isn't that hard.
Sorry, but that's not true. I know I've forgotten my own password a few times. I had to reinstall and recover from backups. (I really don't like sudo, and I also don't like logging in as root, and a decade or so ago I forgot the root password twice. It was really annoying, but not a real problem as one of the times I'd been thinking about switching distros anyway. The other time was more annoying, but I had the original CDs, and there hadn't been THAT many updates. [I said it was over a decade ago. I think it was while I was using Red Hat Professional edition, before I switched to KRUD Linux.])
That would be destruction of evidence...if done by the accused. After a warrant was served.
OTOH, if it's set to do this after n failed attempts, then it could quite likely be done by inept police, who didn't bother to image the disk before working on it. And then it's the police who destroyed the evidence...if such it was.
That said, I really doubt that applies in this case. But I know *I've* forgotten passwords, and had to reinstall, and recover from backups. So "I've forgotten the password" sounds possible, though unconvincing.
So did I. In fact I think I read about a new COBOL compiler just last year. That doesn't keep it from being a lousy language (with a few good features).
COBOL was a horrible language even for the day. Fortran was much better...but it was harder for the bosses to pretend to understand. (OTOH, COBOL was decent at some formatting tasks compared to Fortran, and I believe it included BCD numbers, where in Fortran you needed to use a library for that.)
It's true that there's a lot of ancient COBOL code still live, and I attribute it to the fact that nobody can really understand it, but, unlike assembler, it didn't automatically die when you change processors.
That's the theory, but if you follow the news stories there are repeated examples where the police act in ways that are grossly beyond their legal rights. Including confiscating and searching cameras (and phones).
Considering the story this page is attached to, your comment in indefensible stupidity. It probably would be possible to rephrase it into something that was defensible, but I can't do it for you because I don't understand what your point was.
IIRC Saddam Hussein *did* have weapons of mass destruction. Just not the ones they accused him of having. (I think he had lots of poison gas...but I don't really remember, that could have been some other bloke. But he certainly had napalm.)
Please do not so defame the Thuggees.
Sorry, he's already served two terms.
A point, but copyrights don't protect implementations. They don't even protect you against significant rephrasing...or at least they didn't used to. Making an implementation isn't the same thing as publishing.
To me it seems the most applicable law is the one against unauthorized computer access, but that might not affect Uber, but only their employee, as it might not be possible to prove that he was acting as their agent in his actions. Of course, then you might go after Uber for receiving stolen merchandise....but copying isn't stealing, it's copyright violation, which is a separate crime. Etc.
So I think I see why Google is first trying the path they're trying.
OK, I'll bite, why not?
I could come up with differences, but my first attempt didn't make Google look any better than Uber on this ground, so what's your selection of differences.
IIRC there once was a virus that included a copyright. And if I further recall correctly it even stated a particular open source license (GPL, I think) but it didn't include the text of the license. It wasn't a particularly destructive or successful virus, though.
P.S.: Don't go around believing news stories that don't have valid evidence. For all we know the code could be an officially ordered leak.