In the interview, Scott exhibits amazing ignorance for an author of his supposed stature. As someone else pointed out, use of interfaces has been common in the Win32 world, with COM, for at least about eight years.
That's exactly correct, but make it ten years, not eight. Or perhaps even longer: COM was first given to developers with the OLE2 beta in October 1992, but the object model (without the central class registry) was introduced with MAPI, a little earlier than that.
There has to be some balance. If a vulnerability has existed for months or years without known exploits, the discoverer must consider that there is a high likelihood that even the slowest vendor could fix it before any black-hat re-discovers and exploits it. If that's the case, it is irresponsible to disclose it without giving the vendor -- even slow, crap, poor-at-bugfixing vendors -- a reasonable window to fix it.
I'm thinking particularly of the GreyMagic disclosures of cached object/XSS vulnerabilities here: As far as I know, they existed for around 18 months without anyone of any hat colour knowing, then GreyMagic unilaterally decided that 93% of internet users deserve to be rooted.
The answer is no, as suggested by the Second Law. To see why, you need an account of how solar cells actually work.
Infalling light is absorbed by the material, by dislodging electrons either from a bound orbit, or a semi-free state like a metallic subtrate or semiconductor. The electron absorbs the photon, re-emitting some of the light as another, lower frequency photon, and taking some of the the momentum and energy with it. The new energy and momentum is sufficient to transport it over a potential barrier to the other side of a layered semiconductor or similar. (All semiconductors are layered -- typically they upper few micrometers are doped, wheras lower down it has a different composition).
The solar cell's composition is such that the electron moves up a potential hill, over the brow, and sits in a dip at the top (on the other side of the solar cell). From there, it can get down the hill again (lose it's energy) either by getting over the brow again, or by travelling down the handy attached wires and charging a battery (say).
Imagine if the back of the cell was transparent, and also exposed to sunlight: Then the sunlight falling on that side would knock electrons down the hill as well! This would actually happen EVEN MORE because the work needed to get over the brow out of the little dip at the top is much smaller than the work needed to get all the way up the hill. The difference is that you can't make the electron do any useful work if it is already at the bottom of the hill.
Now if the solar cell is at the same temperature as the black body radiation, the (usually metal or glass) substrate on which the cell is mounted will emit black body radiation too. By the same argument, the equal amount of infalling light from the back of the cell will result in at least as many electrons being knocked down the hill as are being knocked up the hill by the infalling light you want to do the work.
Upshot: The back of the solar cell must be cooler than the temperature of the infalling radiation for it to work.
To clear up a last couple of points: If the back of the cell is not a black-body emitter, it will either be partially transparent or partially reflective or both. There are no other alternatives (because of the quantum symmetries involved). If it is partially transparent, then the temperature of the objects behind the solar cell becomes relevant. If it is partially reflective, the difference is made up by the reflected heat not absorbed by the solar cell -- including the black body radiation of the cell semiconductor if it is not transparent, or the infalling light if it is.
Of course, that's technically not an error message. Messenger Service messages are sent by users to users or machines. The sysadmin probably sent this one.
Really? Well I missed that. Perhaps you can provide a reference?
Because if you are talking about the Antitrust trial, they were not charged with a crime because it is a civil trial not a criminal one.
Since the complaint (not charge) is a civil matter, they can't be convicted, only found liable.
They haven't been convicted because it is a civil trial.
They haven't been found guilty, because it is a civil complaint. (The Sherman Act also creates a criminal offence, but microsoft weren't charged with it.)
They haven't been shown to be criminals, because this is a civil trial
They can't be punished because the DoJ didn't ask for punitive damages.
The point is, that even though it can be a crime, Microsoft have not been charged with that offence. Instead civil proceedings were brought, for a civil offence also provided by the Sherman Act.
The differences? The civil offence is less serious, and doesn't have to be proven beyond a reasonable doubt.
Now you may wish that they had been charged with a criminal offence, but they weren't so it is not correct to call them criminals.
They were not charged with a criminal offence in the matter. Abusing a monopoly position can be a criminal offence, but in this case they were not charged with one, a civil complaint was brought instead.
The practical differences are these:
They can't be punished, unless the complainant asks for punitive damages, which they didn't in this case. The court has the power only to impose a remedy which will right the wrong (usually compensation and/or an injunction). Anything which goes beyond this into punishment is illegal.
The complainant has to meet a much lower burden of proof, that of "the balance of probabilities" not "beyond reasonable doubt" as in a criminal trial. Thus reasonable people may doubt the verdict.
The offences committed are much less serious than those required for a criminal antitrust offence. It's therefore questionable whether what MS has done would amount to a criminal offence. If it would, why weren't they charged with one? You may wish the law was more severe, but it is what it is.
They aren't convicted of anything, they haven't been proven guilty, or proven to be criminals. Calling them such is wrong.
You may wish they had been convicted, you may wish that it was a crime, you may not think that the law is tough enough, you may want them to be punished, you may hate Microsoft with your very heart and soul, but that doesn't change the fact that they haven't been found guilty, convicted or even charged with any crime.
It was a civil trial. What is so difficult to understand?
Unless said company is a
convicted abusive monopolist who has voluntarily agreed to release said secrets to avoid a more severe penalty.
But we aren't talking about Microsoft here are we? Because Microsoft haven't been charged with any crime have they? Because you know that the whole antitrust trial is a civil matter don't you? And you know the court doesn't have the power to impose a punishment, only a remedy, don't you?
Microsoft has managed to survive many tough legal blows to date and, despite a criminal conviction, seemed like they would survive with a slap on the wrist. (...)
What criminal conviction would that be then? The whole antitrust trial is a civil matter, and doesn't involve any criminal charges at all.
MS haven't been found guilty of anything. They have been found, on the balance of probability, to have a monopoly in intel-compatable operating systems. Merely being a monopoly is not a crime, what matters is what you do with it.
Possibly this has changed, but disregarding the old Schedule D Notices was not automatically an offence. Rather the notice served to tell the media organisation concerned that the information was likely to compromise national security. Once in receipt of a D Notice, the people concerned are forwarned, and if the media organisation still reports the information, and if a court agrees that national security was in fact compromised then the organisation is in trouble.
Because an offence is not automatically committed, and the matter still has to go before a court and the government prove that national security was in fact compromised to get a conviction, the D Notices seem not to be abused. And because they are not abused, they are widely heeded: What newspaper really wants to compromise national security?
-- aspect of it (the lack of understanding of which I am becoming quite frustrated with) my reading is the same as that of the parent post.
With regard to the "mingling" charge, the appeals court rejected MS appeal on that because, if I remember correctly, MS failed to introduce any evidence to rebut the evidence offered by the plaintiffs. In other words, MS fumbled the ball, and the same facts could go a different way if MS's lawyers paid more attention. So I wouldn't be too surprised by that.
Given also that there was no finding that the mingling had actually aided MS, any remedy would be unlikely to address the issue with more than an injunction requiring them not to do it again (remember that a remedy is not a punishment -- it is only intended in law to put right the wrong). Given that the DoJ still haven't defined the difference between Application and OS, that could be quite difficult to enforce.
As a matter of fact they have been convited of anti-trust violations in more than one trial, including the current trial.
Microsoft have not been convicted of anything because this is a civil trial not a criminal trial.
It's true that Microsoft have been found liable on some of the points, but:
They haven't been convicted because it is a civil trial.
They haven't been found guilty, because it is a civil offence.
They haven't been shown to be criminals, because this is a civil trial
They can't be punished because the DoJ didn't ask for punitive damages.
Now, you may wish they had been convicted, you may wish that it was a crime, you may not think that the law is tough enough, you may want them to be punished, you may hate Microsoft with your very heart and soul, but that doesn't change the fact that they haven't been found guilty, convicted or even charged with any crime.
It was a civil trial. What is so difficult to understand?
It's worth noting that not everything was remanded back on the facts, so we can be sure MS will have to do something, and won't get off with nothing at all.
Actually the outcome of the case is assured. Microsoft
has been found guilty.
No, they haven't. They have been found liable, which is not the same at all.
the judge can still take the other points into consideration when handing down a sentence
There is a reason it is called a remedy not a sentence. The sole purpose of the remedy is to remedy the wrong done, by compensating the injured parties and maybe preventing further wrongdoing. A sentence is a punishment, and since this is a civil trial, and punitive damages were not asked for, the court is not empowered to punish Microsoft at all, only to remedy the wrong found to have been done.
(...) splitting the company is the only way to reliably keep MS from repeating the illegal actions that they were convicted of (...)
Microsoft have not been convicted of anything. They haven't even been charged with anything. They have been sued, and it is civil trial.
Until the judge issues her decision, anything that MS is doing is just posturing...
You are absolutely right about that. It's a gesture, nothing more. If it is not done properly it's an empty gesture, but either way this is a storm in a teacup, and has very little to do with the eventual outcome.
They were 'convicted' and went to appeal (...) essentially there's little chance that Microsoft will actually have been convicted of anything (...)
Not least, of course, because they haven't been charged with anything. Your scare quotes above suggest that you know this, but it does no harm to get it right for the benefit of readers who don't. Microsoft can't be punished in this matter, because it is a civil trial, and punitive damages weren't asked for.
They can do whatever they want to Microsoft for all I care, but they'd damned well better do it right. Judges screwing with the process are just going to get their rulings overturned or remanded, which wastes everyone's time and money.
Slashdot Mirror
That's exactly correct, but make it ten years, not eight. Or perhaps even longer: COM was first given to developers with the OLE2 beta in October 1992, but the object model (without the central class registry) was introduced with MAPI, a little earlier than that.
I bet you didn't know you can still format XML tags with CSS regardless of whether they are valid HTML tags?
How to Write a CSS Style Sheet for Browsing XML
The parent post is correct. J2EE is a mess, and anyone who has used both will know which is better.
This is just a matter of the two platforms leapfrogging each other, rather than either being ahead consistently.
Shurely shome mishtake?
Surely you mean:
Even though they can get it all in return for hosting viruses and trojans on Kazaa?
Even thought they can get it in exchange for their integrity on Kazaa?
Even though they can get it in return for sitting online for hours looking for a decent quality recording of a track?
Hear hear.
There has to be some balance. If a vulnerability has existed for months or years without known exploits, the discoverer must consider that there is a high likelihood that even the slowest vendor could fix it before any black-hat re-discovers and exploits it. If that's the case, it is irresponsible to disclose it without giving the vendor -- even slow, crap, poor-at-bugfixing vendors -- a reasonable window to fix it.
I'm thinking particularly of the GreyMagic disclosures of cached object/XSS vulnerabilities here: As far as I know, they existed for around 18 months without anyone of any hat colour knowing, then GreyMagic unilaterally decided that 93% of internet users deserve to be rooted.
This is an interesting twist on the subject!
The answer is no, as suggested by the Second Law. To see why, you need an account of how solar cells actually work.
Infalling light is absorbed by the material, by dislodging electrons either from a bound orbit, or a semi-free state like a metallic subtrate or semiconductor. The electron absorbs the photon, re-emitting some of the light as another, lower frequency photon, and taking some of the the momentum and energy with it. The new energy and momentum is sufficient to transport it over a potential barrier to the other side of a layered semiconductor or similar. (All semiconductors are layered -- typically they upper few micrometers are doped, wheras lower down it has a different composition).
The solar cell's composition is such that the electron moves up a potential hill, over the brow, and sits in a dip at the top (on the other side of the solar cell). From there, it can get down the hill again (lose it's energy) either by getting over the brow again, or by travelling down the handy attached wires and charging a battery (say).
Imagine if the back of the cell was transparent, and also exposed to sunlight: Then the sunlight falling on that side would knock electrons down the hill as well! This would actually happen EVEN MORE because the work needed to get over the brow out of the little dip at the top is much smaller than the work needed to get all the way up the hill. The difference is that you can't make the electron do any useful work if it is already at the bottom of the hill.
Now if the solar cell is at the same temperature as the black body radiation, the (usually metal or glass) substrate on which the cell is mounted will emit black body radiation too. By the same argument, the equal amount of infalling light from the back of the cell will result in at least as many electrons being knocked down the hill as are being knocked up the hill by the infalling light you want to do the work.
Upshot: The back of the solar cell must be cooler than the temperature of the infalling radiation for it to work.
To clear up a last couple of points: If the back of the cell is not a black-body emitter, it will either be partially transparent or partially reflective or both. There are no other alternatives (because of the quantum symmetries involved). If it is partially transparent, then the temperature of the objects behind the solar cell becomes relevant. If it is partially reflective, the difference is made up by the reflected heat not absorbed by the solar cell -- including the black body radiation of the cell semiconductor if it is not transparent, or the infalling light if it is.
While it's true that book-learning is no substitute for experience, the converse also holds.
I can only suggest that you get a clue. Here's one, courtesy of google. To save you the effort, here's some of the best:
+1 Funny!
Of course, that's technically not an error message. Messenger Service messages are sent by users to users or machines. The sysadmin probably sent this one.
This has nothing to do with what I posted. I've said nothing to justify Microsoft's actions, I'm just correcting your errors.
You said they were convicted criminals, I said that they haven't been charged with any criminal offence.
If you are happy to retract, then fine.
Really? Well I missed that. Perhaps you can provide a reference?
Because if you are talking about the Antitrust trial, they were not charged with a crime because it is a civil trial not a criminal one.
Since the complaint (not charge) is a civil matter, they can't be convicted, only found liable.
Thanks for the handy, irrellevant references.
The point is, that even though it can be a crime, Microsoft have not been charged with that offence. Instead civil proceedings were brought, for a civil offence also provided by the Sherman Act.
The differences? The civil offence is less serious, and doesn't have to be proven beyond a reasonable doubt.
Now you may wish that they had been charged with a criminal offence, but they weren't so it is not correct to call them criminals.
They were not charged with a criminal offence in the matter. Abusing a monopoly position can be a criminal offence, but in this case they were not charged with one, a civil complaint was brought instead.
The practical differences are these:
You may wish they had been convicted, you may wish that it was a crime, you may not think that the law is tough enough, you may want them to be punished, you may hate Microsoft with your very heart and soul, but that doesn't change the fact that they haven't been found guilty, convicted or even charged with any crime.
It was a civil trial. What is so difficult to understand?
But we aren't talking about Microsoft here are we? Because Microsoft haven't been charged with any crime have they? Because you know that the whole antitrust trial is a civil matter don't you? And you know the court doesn't have the power to impose a punishment, only a remedy, don't you?
What criminal conviction would that be then? The whole antitrust trial is a civil matter, and doesn't involve any criminal charges at all.
MS haven't been found guilty of anything. They have been found, on the balance of probability, to have a monopoly in intel-compatable operating systems. Merely being a monopoly is not a crime, what matters is what you do with it.
Actually no they have not. It was a civil case. They won't be convicted, and they shan't be sentenced either since it is a civil matter. That means
Why not get the facts right? It's not hard, try it. You might like it!
Possibly this has changed, but disregarding the old Schedule D Notices was not automatically an offence. Rather the notice served to tell the media organisation concerned that the information was likely to compromise national security. Once in receipt of a D Notice, the people concerned are forwarned, and if the media organisation still reports the information, and if a court agrees that national security was in fact compromised then the organisation is in trouble.
Because an offence is not automatically committed, and the matter still has to go before a court and the government prove that national security was in fact compromised to get a conviction, the D Notices seem not to be abused. And because they are not abused, they are widely heeded: What newspaper really wants to compromise national security?
Leaving aside the whole --
- it was a civil trial not a criminal trial so
- they aren't "charges" but complaints and
- MS aren't "guilty" but liable
-- aspect of it (the lack of understanding of which I am becoming quite frustrated with) my reading is the same as that of the parent post.With regard to the "mingling" charge, the appeals court rejected MS appeal on that because, if I remember correctly, MS failed to introduce any evidence to rebut the evidence offered by the plaintiffs. In other words, MS fumbled the ball, and the same facts could go a different way if MS's lawyers paid more attention. So I wouldn't be too surprised by that.
Given also that there was no finding that the mingling had actually aided MS, any remedy would be unlikely to address the issue with more than an injunction requiring them not to do it again (remember that a remedy is not a punishment -- it is only intended in law to put right the wrong). Given that the DoJ still haven't defined the difference between Application and OS, that could be quite difficult to enforce.
Microsoft have not been convicted of anything because this is a civil trial not a criminal trial.
It's true that Microsoft have been found liable on some of the points, but:
Now, you may wish they had been convicted, you may wish that it was a crime, you may not think that the law is tough enough, you may want them to be punished, you may hate Microsoft with your very heart and soul, but that doesn't change the fact that they haven't been found guilty, convicted or even charged with any crime.
It was a civil trial. What is so difficult to understand?
It's worth noting that not everything was remanded back on the facts, so we can be sure MS will have to do something, and won't get off with nothing at all.
No, they haven't. They have been found liable, which is not the same at all.
There is a reason it is called a remedy not a sentence. The sole purpose of the remedy is to remedy the wrong done, by compensating the injured parties and maybe preventing further wrongdoing. A sentence is a punishment, and since this is a civil trial, and punitive damages were not asked for, the court is not empowered to punish Microsoft at all, only to remedy the wrong found to have been done.
Microsoft have not been convicted of anything. They haven't even been charged with anything. They have been sued, and it is civil trial.
You are absolutely right about that. It's a gesture, nothing more. If it is not done properly it's an empty gesture, but either way this is a storm in a teacup, and has very little to do with the eventual outcome.
Not least, of course, because they haven't been charged with anything. Your scare quotes above suggest that you know this, but it does no harm to get it right for the benefit of readers who don't. Microsoft can't be punished in this matter, because it is a civil trial, and punitive damages weren't asked for.
Hear Hear!