Social engineering does more damage than you can undo with whatever vista+IEwhatever can undo. My lawyer is going to click yes if the bait looks good enough to her.
The only way to get in the way of that is to get in the way of that: Special purpose browsers that don't have a place to plug in a URL. And even that is not good enough, but it's better than trying to use ACLs to build walled gardens like this "integrity levels" thing Vista has.
I just calculated the event horizon for a micro black hole the mass of a U238 atom. (Got the formulae and stuff from wikipedia.)
If I got the math right, it's 5.87e-46 m.
A neutron is around 1.6e-15 m.
What are the odds of such a small hole actually coming into contact with anything? Even if it does, how does it "swallow" anything? At best, it might make a quark evaporate into its component strings, minus a string, maybe, every now and then.
Even a black hole of a full gram is going to have an event horizon around 1.49e-24 m, which is still a billionth of the size of a neutron.
If jupiter suddenly turned into a black hole, it would be about 3 km in diameter (if I calculated correctly). That's an awfully small thing to try to hit the sun with, and it has orbital velocity and all. The only effect on its moons would be that the explosions of the gasses that weren't pulled under the event horizon would likely push them away from Jupiter. If the moons's orbits turned elliptical enough to bring them under the event horizon, yeah, they would fall in. And the gasses released in the explosions would alter the orbit of Jupiter. But would it hit any other planets? Would it hit the sun? Massive, yes, but volume-wise, smaller than the moons of Mars. Well, yeah, if it got close to the earth, even as close as the moon, it could cause earthquakes and such. Just like a micro-black hole might hit another electron and cause it to evaporate into quarks and strings.
But in order for the Jupiter black hole to absorb the sun, it has to hit the sun, and that's fairly long odds. If it does, in a million years or so, eat the sun, the sun is still just a small star, and the resulting black hole about the radius of the earth. Back off to the scope of the solar system, and that's just one sun dying a rather unusual death. Maybe it accelerates in a different direction than it should have because of some jet stream produced by the collapse. But the overall effect on the galaxy is not significantly different than the sun dying as a brown dwarf in a few billion years. Shoot, if Jupiter suddenly turned into a black hole today, the odds of it coming close enough to cause earthquakes and such are still not such that we would be worried for several centuries.
Anyway, to put this into perspective, a few really tiny black holes running loose around the earth is like a few suns the size of jupiter running around loose in the galaxy. And we, compared to the micro-black holes, are bigger than the largest structures observed in the known universe. Or, let's say, what if there were millions of earth-sized black holes loose in the galaxy already? Would we even notice on our scale, much less would it make any difference in the large scope of the cluster of galaxies our galaxy is part of?
I mean, for all we could care, an electron might suddenly turn into a micro black hole, and it still has to get out of orbit and way down to the surface of the nucleus to do any damage, and when it gets there it's way, way, way, less than a billionth the size of the neutron. Quite likely, even moving slow, it would drift right through the neutron without touching any of the component quarks in any way. I'm not talking about even roulette table odds, and not even lottery odds. And it has to eat how many of those quarks before it amasses even a gram?
I just went over to wikipedia and looked around for formulae and masses and radii.
2Gm/C^2 or something like that.
(HEY! plug "proton mass" into google and see what comes out!)
Exercised my rusty mental processes and, if we can trust wikipedia and my math:
The limit horizon of a U238 atom is 5.86875420167021e-46 m.
This compares to its radius of 15 femtometers, or 1.5e-14 m.
This also compares to the radius of a proton, about 1.6e-15 m, and the radius of an electron, roughly 2.8e-15 m.
which makes me wonder what happens when a black hole the mass of a U238 atom comes in contact with an atomic particle. What are the odds against it actually coming close enough to any of the quarks to eat one, and can it eat one? Would it eat, maybe, one of the component strings, and let the rest of the strings loose?
How much mass do they assume these micro-black holes are going to develop?
Hmm. Could dark matter be clouds of free micro black holes?
Kind of what I'm thinking. Even if Hawking radiation doesn't, even a single U238 atom is a really small target. I mean, part of the reason for the size of the high-energy research tools is due to the trouble we have hitting such small targets. How much mass is a black hole with an event horizon even as large as a single proton going to have, and how often is such a black hole, drifting loose, going to bring it's event horizon into contact with particles it can "swallow"?
I suppose I could search the web for it, but if the event horizon is smaller than a proton, what happens when a proton comes in contact with it? Does it absorb some of the proton's strings, and let the rest loose?
As long as the browser has the ability to be re-directed to any site but the site it was defined for, you're going to have spoofing.
As long as you have spoofing, you're going to be losing your tokens.
Yeah, I know that having multiple single-purpose browsers that a general-purpose browser can invoke opens loopholes, but that's also part of what running as a separate user is for.
sudo isn't a sandbox, but it can put some walls up between a browser user and the log-in user.
Either Microsoft is playing with the English language again, or I really don't like the security model they've slapped together here.
Uhm, that's not a separate user. Thats a separate access mode for the logged in user. User Access Controls. Not the same thing.
It's also not a single-purpose browser. Close to a parameterized browser, but still not even that. Well, maybe they can achieve the equivalent of a parameterized browser within the login user context. Maybe. (But then you wouldn't hear anything at all about the toolbars that try to "tell" you when the web site you're visiting is trusted.)
And there's part of the reason why MSIE under Vista has given us a number of admin-level vulnerabilities, in spite of this security model.
There is no way for a general purpose browser to be secure. There's a semantic conflict. Oxymoron, in more human terms.
And access controls are not a substitute for actually putting a different user out on the web.
No, the Unix model doesn't get you there, either, although it could get you a lot closer if you could sudo your browser to another user in either X-11 or Aqua. (And, of course, then we'd have Microsoft forcing us to take them to court to show why their attempt to patent sudo is egregious.)
So the Vista box is the cheap one. But it's still small and lightweight, so a worthwhile prize even if not the top prize.
If it were in my neighborhood, I might go by and pick one or the other up (if no one beat me to it). I want a lightweight portable to take on the train.
Encouraging that the Ubuntu box survived the second day (Sony VAIO VGN-TZ37CN), surprising that the Vista box did, as well. (Fujitsu U810, 800 MHz iNTEL A110, but it does have 1G RAM. 40G HD isn't all that interesting.)
I really think sony doesn't want to sell laptops to people who know anything about them. Finding information on that VAIO on sonystyle.com is like pulling teeth.
sudo (especially, M$'s patented snake-oil version of sudo) all by itself isn't enough.
You have to have single-purpose browsers, and they can't be just parameterized instances of the general purpose browser (and, no, the current MSIE is not even such a parameterizable browser).
(as more than one person mentions above,)... that the attack on the mac was the first attempted hack under the relaxed rules. I think it's clear that the hacker wanted the mac, especially since there are known open vulnerabilities that could have been used on MSIE, and some highly probable directions fairly well known on Firefox.
We know that the browser is vulnerable. Anyone who thinks general purpose browsers are invincible is living in a dream world.
Miller, best known as one of the researchers who first hacked Apple's iPhone last year, didn't take much time. Within 2 minutes, he directed the contest's organizers to visit a Web site that contained his exploit code, which then allowed him to seize control of the computer, as about 20 onlookers cheered him on.
He was the first contestant to attempt an attack on any of the systems.
But the issue is really not which is more vulnerable, it is that you can't run a secure browser and a convenient browser unless they are two separate browsers.
It's time to abandon the general purpose browser. It's also time to quit surfing as your log-in user. You need a browser for surfing that you run (sudo or something) as a strictly limited privilege user without log-in capabilities.
I can't tell from the article, but the technical tarpit we're stuck in, here, is that we must have two channels to have privacy -- one channel must be open, and Google would be no more threat to that than the unencoded internet is already.
The other must be an encoded channel. Or, at bare minimum, a private channel not routed outside the school.
Even if the school does not use Google, it needs the two channels.
The question is whether the school is providing the private channel, and whether the staff is willing to use it.
Not to make bigger chips, but to solve the interconnect problem when you use a lot of small chips in a big package.
Although, even on-chip, at 1 cm^2 and above, optical conversion might beat be able to beat the reactance+buffering on a channel that crosses the whole chip, especially when a single physical channel might be able to carry 64 logical channels.
It's not a new idea, it's just one that needs to be revisited from time to time, to see if the optical tech is up to the job yet.
Can you name one viable standard where the sponsoring organization has been as blatant as Microsoft?
(Strong thrust, if a bit wild. Too bad the foil got under your mask. Suggest we take a break while you run down to the medic and get your eye attended?)
But, anyway, Apple has bought specs from Microsoft. Apple's software can open some of Microsoft's documents with reasonable results, but that is because they bought the information from Microsoft.
As far as NeoOffice, that's inheriting from the reverse engineering done by the openoffice group. Inheriting directly, I might add. And anything that requires reverse engineering to be done without paying a fee is not free in either of the senses usually argued here.
You do bring up some valid points before you stand on your head to look at them.
ODF is bloated. But, given the context, where Microsoft has been hiding the reality of how functional plain text is for as long as they've existed, and where Microsoft has been trying their hardest to distract people from the power of plaintext + tags ever since XML saw the light of day, ODF had to have the bloat to get the consideration.
Customers are beginning to demand documents they can use. They want their information back. Microsoft is not going to give it to them willingly.
If the early ODF apps are really slipshod, customers are going to find themselves face to face with the conflict between form and substance, and most of them are going to accept that, at least sometimes, they want more substance than Microsoft can give them. In fact, not want, but need.
Ignoring ODF will kill Microsoft, unless they can make the leap from form to substance and ignore it for the right reasons.
If they could make that leap, however, I think that would kill Microsoft in a different sense. Microsoft is too big to properly deal with the semantics of ordinary end-users.
If Microsoft really wants to answer the supposedly invalid arguments, take it off fast-track. In fact, withdraw it. Strip it down. Rework it. Make it a real standard. (Three or four separate standards, actually, at minimum.)
Then re-submit on a normal track.
They don't have to have the standard to build their software. (Although we could wish they had waited.) They can compete fairly in the present without having this non-standard be made a standard. They could behave responsibly, like the rest of the industry tries to from time to time.
Until they do, even the not-very-valid arguments are plenty valid.
Microsoft can wave their arms and scream loudly, "We'll fix it! WE'LL FIX IT!!!!!!"
"JUST VOTE FOR IT NOW AND WE'LL FIX IT!!!!!"
Yes, they can do that all they want. But if they really meant it, they would have seen the thousands of problems that there is no time to deal with now and said, "Oh.!" and withdrawn the thing, taken it back home, stripped it down, made a real standard out of it, and submitted it to a _normal_ process.
If I were trying to build such a standard all by myself, I might need the political clout of standardization to fund the process of getting it right. (And fat chance of actually getting access to a fast-track process for any of the standards I would like to present to the world.)
Microsoft has lots of money, lots of engineers, lots of time. The only possible reason for fast-tracking this is to shore up their monopoly in the face of ODF. Therefore, all the thousands of technical problems which have not even been considered, as compared to the hundreds that did get some sort of attention, are evidence of Microsoft's refusal to address the issues before the vote.
And, yes, they have to address the issues before the vote. No weasel-worded promises.
If they really mean to fix the rest of the problems, they must at bare minimum present the ISO with an iron-clad covenant to not only fix them in a timely manner, but to make the process and technology (including all patents) open enough to be confirmed in public process. But they are unwilling to even do that.
Not hundreds, but thousands of issues, that they have refused to address.
Slashdot Mirror
Social engineering does more damage than you can undo with whatever vista+IEwhatever can undo. My lawyer is going to click yes if the bait looks good enough to her.
The only way to get in the way of that is to get in the way of that: Special purpose browsers that don't have a place to plug in a URL. And even that is not good enough, but it's better than trying to use ACLs to build walled gardens like this "integrity levels" thing Vista has.
I just calculated the event horizon for a micro black hole the mass of a U238 atom. (Got the formulae and stuff from wikipedia.)
If I got the math right, it's 5.87e-46 m.
A neutron is around 1.6e-15 m.
What are the odds of such a small hole actually coming into contact with anything? Even if it does, how does it "swallow" anything? At best, it might make a quark evaporate into its component strings, minus a string, maybe, every now and then.
Even a black hole of a full gram is going to have an event horizon around 1.49e-24 m, which is still a billionth of the size of a neutron.
If jupiter suddenly turned into a black hole, it would be about 3 km in diameter (if I calculated correctly). That's an awfully small thing to try to hit the sun with, and it has orbital velocity and all. The only effect on its moons would be that the explosions of the gasses that weren't pulled under the event horizon would likely push them away from Jupiter. If the moons's orbits turned elliptical enough to bring them under the event horizon, yeah, they would fall in. And the gasses released in the explosions would alter the orbit of Jupiter. But would it hit any other planets? Would it hit the sun? Massive, yes, but volume-wise, smaller than the moons of Mars. Well, yeah, if it got close to the earth, even as close as the moon, it could cause earthquakes and such. Just like a micro-black hole might hit another electron and cause it to evaporate into quarks and strings.
But in order for the Jupiter black hole to absorb the sun, it has to hit the sun, and that's fairly long odds. If it does, in a million years or so, eat the sun, the sun is still just a small star, and the resulting black hole about the radius of the earth. Back off to the scope of the solar system, and that's just one sun dying a rather unusual death. Maybe it accelerates in a different direction than it should have because of some jet stream produced by the collapse. But the overall effect on the galaxy is not significantly different than the sun dying as a brown dwarf in a few billion years. Shoot, if Jupiter suddenly turned into a black hole today, the odds of it coming close enough to cause earthquakes and such are still not such that we would be worried for several centuries.
Anyway, to put this into perspective, a few really tiny black holes running loose around the earth is like a few suns the size of jupiter running around loose in the galaxy. And we, compared to the micro-black holes, are bigger than the largest structures observed in the known universe. Or, let's say, what if there were millions of earth-sized black holes loose in the galaxy already? Would we even notice on our scale, much less would it make any difference in the large scope of the cluster of galaxies our galaxy is part of?
I mean, for all we could care, an electron might suddenly turn into a micro black hole, and it still has to get out of orbit and way down to the surface of the nucleus to do any damage, and when it gets there it's way, way, way, less than a billionth the size of the neutron. Quite likely, even moving slow, it would drift right through the neutron without touching any of the component quarks in any way. I'm not talking about even roulette table odds, and not even lottery odds. And it has to eat how many of those quarks before it amasses even a gram?
I'm ranting. I need to put the kids to bed.
I just went over to wikipedia and looked around for formulae and masses and radii.
2Gm/C^2 or something like that.
(HEY! plug "proton mass" into google and see what comes out!)
Exercised my rusty mental processes and, if we can trust wikipedia and my math:
The limit horizon of a U238 atom is 5.86875420167021e-46 m.
This compares to its radius of 15 femtometers, or 1.5e-14 m.
This also compares to the radius of a proton, about 1.6e-15 m, and the radius of an electron, roughly 2.8e-15 m.
which makes me wonder what happens when a black hole the mass of a U238 atom comes in contact with an atomic particle. What are the odds against it actually coming close enough to any of the quarks to eat one, and can it eat one? Would it eat, maybe, one of the component strings, and let the rest of the strings loose?
How much mass do they assume these micro-black holes are going to develop?
Hmm. Could dark matter be clouds of free micro black holes?
Kind of what I'm thinking. Even if Hawking radiation doesn't, even a single U238 atom is a really small target. I mean, part of the reason for the size of the high-energy research tools is due to the trouble we have hitting such small targets. How much mass is a black hole with an event horizon even as large as a single proton going to have, and how often is such a black hole, drifting loose, going to bring it's event horizon into contact with particles it can "swallow"?
I suppose I could search the web for it, but if the event horizon is smaller than a proton, what happens when a proton comes in contact with it? Does it absorb some of the proton's strings, and let the rest loose?
Guess I should go look something up, here.
But I am wondering what the difference between the CNB and CNP is. Color?
That's basically the point.
As long as the browser has the ability to be re-directed to any site but the site it was defined for, you're going to have spoofing.
As long as you have spoofing, you're going to be losing your tokens.
Yeah, I know that having multiple single-purpose browsers that a general-purpose browser can invoke opens loopholes, but that's also part of what running as a separate user is for.
sudo isn't a sandbox, but it can put some walls up between a browser user and the log-in user.
Low integrity?
Either Microsoft is playing with the English language again, or I really don't like the security model they've slapped together here.
Uhm, that's not a separate user. Thats a separate access mode for the logged in user. User Access Controls. Not the same thing.
It's also not a single-purpose browser. Close to a parameterized browser, but still not even that. Well, maybe they can achieve the equivalent of a parameterized browser within the login user context. Maybe. (But then you wouldn't hear anything at all about the toolbars that try to "tell" you when the web site you're visiting is trusted.)
And there's part of the reason why MSIE under Vista has given us a number of admin-level vulnerabilities, in spite of this security model.
There is no way for a general purpose browser to be secure. There's a semantic conflict. Oxymoron, in more human terms.
And access controls are not a substitute for actually putting a different user out on the web.
No, the Unix model doesn't get you there, either, although it could get you a lot closer if you could sudo your browser to another user in either X-11 or Aqua. (And, of course, then we'd have Microsoft forcing us to take them to court to show why their attempt to patent sudo is egregious.)
1.2 GHz, but Core 2 duo, 2G RAM, 100G HD.
So the Vista box is the cheap one. But it's still small and lightweight, so a worthwhile prize even if not the top prize.
If it were in my neighborhood, I might go by and pick one or the other up (if no one beat me to it). I want a lightweight portable to take on the train.
as it says in the article.
2nd day was default Apple apps.
Encouraging that the Ubuntu box survived the second day (Sony VAIO VGN-TZ37CN), surprising that the Vista box did, as well. (Fujitsu U810, 800 MHz iNTEL A110, but it does have 1G RAM. 40G HD isn't all that interesting.)
I really think sony doesn't want to sell laptops to people who know anything about them. Finding information on that VAIO on sonystyle.com is like pulling teeth.
sudo (especially, M$'s patented snake-oil version of sudo) all by itself isn't enough.
You have to have single-purpose browsers, and they can't be just parameterized instances of the general purpose browser (and, no, the current MSIE is not even such a parameterizable browser).
(as more than one person mentions above,) ... that the attack on the mac was the first attempted hack under the relaxed rules. I think it's clear that the hacker wanted the mac, especially since there are known open vulnerabilities that could have been used on MSIE, and some highly probable directions fairly well known on Firefox.
We know that the browser is vulnerable. Anyone who thinks general purpose browsers are invincible is living in a dream world.
But the issue is really not which is more vulnerable, it is that you can't run a secure browser and a convenient browser unless they are two separate browsers.
It's time to abandon the general purpose browser. It's also time to quit surfing as your log-in user. You need a browser for surfing that you run (sudo or something) as a strictly limited privilege user without log-in capabilities.
I think the oblique open source joke is the real point.
Motorola keeps having to abandon its products because they keep having narrow brushes with the open and free world.
And the one is different from the other, how?
Well, let's see. Google, at any rate, can't yet sick cops with guns on you, I think.
You and I never had any affection for Gates, but the suits definitely loved him. (And in their world, anyone who is anybody is a suit.)
PHBs/suits loved Gates because he was one of them, and was still sort of successful at pretending to be one of us.
(Not that we are the homogenous group of geeks they perceive us to be. But somehow we are "them" to the suits.)
I can't tell from the article, but the technical tarpit we're stuck in, here, is that we must have two channels to have privacy -- one channel must be open, and Google would be no more threat to that than the unencoded internet is already.
The other must be an encoded channel. Or, at bare minimum, a private channel not routed outside the school.
Even if the school does not use Google, it needs the two channels.
The question is whether the school is providing the private channel, and whether the staff is willing to use it.
Can I quote that in my sig sometimes, instead?
I'm looking at that question, and thinking, uhm, you know, lobbies?
Government is watched by whom?
Private, profit-seeking corporation is watched by whom?
I don't think there is a good alternative.
Not to make bigger chips, but to solve the interconnect problem when you use a lot of small chips in a big package.
Although, even on-chip, at 1 cm^2 and above, optical conversion might beat be able to beat the reactance+buffering on a channel that crosses the whole chip, especially when a single physical channel might be able to carry 64 logical channels.
It's not a new idea, it's just one that needs to be revisited from time to time, to see if the optical tech is up to the job yet.
Can you name one viable standard where the sponsoring organization has been as blatant as Microsoft?
(Strong thrust, if a bit wild. Too bad the foil got under your mask. Suggest we take a break while you run down to the medic and get your eye attended?)
But, anyway, Apple has bought specs from Microsoft. Apple's software can open some of Microsoft's documents with reasonable results, but that is because they bought the information from Microsoft.
As far as NeoOffice, that's inheriting from the reverse engineering done by the openoffice group. Inheriting directly, I might add. And anything that requires reverse engineering to be done without paying a fee is not free in either of the senses usually argued here.
You do bring up some valid points before you stand on your head to look at them.
ODF is bloated. But, given the context, where Microsoft has been hiding the reality of how functional plain text is for as long as they've existed, and where Microsoft has been trying their hardest to distract people from the power of plaintext + tags ever since XML saw the light of day, ODF had to have the bloat to get the consideration.
Customers are beginning to demand documents they can use. They want their information back. Microsoft is not going to give it to them willingly.
If the early ODF apps are really slipshod, customers are going to find themselves face to face with the conflict between form and substance, and most of them are going to accept that, at least sometimes, they want more substance than Microsoft can give them. In fact, not want, but need.
Ignoring ODF will kill Microsoft, unless they can make the leap from form to substance and ignore it for the right reasons.
If they could make that leap, however, I think that would kill Microsoft in a different sense. Microsoft is too big to properly deal with the semantics of ordinary end-users.
If Microsoft really wants to answer the supposedly invalid arguments, take it off fast-track. In fact, withdraw it. Strip it down. Rework it. Make it a real standard. (Three or four separate standards, actually, at minimum.)
Then re-submit on a normal track.
They don't have to have the standard to build their software. (Although we could wish they had waited.) They can compete fairly in the present without having this non-standard be made a standard. They could behave responsibly, like the rest of the industry tries to from time to time.
Until they do, even the not-very-valid arguments are plenty valid.
If they were really willing to listen, they would take the thing off fast-track.
Fear of ODF would be no excuse for real engineers.
Get it?
Microsoft can wave their arms and scream loudly, "We'll fix it! WE'LL FIX IT!!!!!!"
"JUST VOTE FOR IT NOW AND WE'LL FIX IT!!!!!"
Yes, they can do that all they want. But if they really meant it, they would have seen the thousands of problems that there is no time to deal with now and said, "Oh.!" and withdrawn the thing, taken it back home, stripped it down, made a real standard out of it, and submitted it to a _normal_ process.
If I were trying to build such a standard all by myself, I might need the political clout of standardization to fund the process of getting it right. (And fat chance of actually getting access to a fast-track process for any of the standards I would like to present to the world.)
Microsoft has lots of money, lots of engineers, lots of time. The only possible reason for fast-tracking this is to shore up their monopoly in the face of ODF. Therefore, all the thousands of technical problems which have not even been considered, as compared to the hundreds that did get some sort of attention, are evidence of Microsoft's refusal to address the issues before the vote.
And, yes, they have to address the issues before the vote. No weasel-worded promises.
If they really mean to fix the rest of the problems, they must at bare minimum present the ISO with an iron-clad covenant to not only fix them in a timely manner, but to make the process and technology (including all patents) open enough to be confirmed in public process. But they are unwilling to even do that.
Not hundreds, but thousands of issues, that they have refused to address.