What percentage of the Darwin Award winners were high on drugs?
People are perfectly capable of doing stupid things on their own, without external stimuli. The only stupid things that people on pot do frequently are 1) drive under the influence (see alcohol) and 2) eat excessively.
The driving *is* a problem, but legalization might reduce it. Might not, too. Hard to predict. Anyway, reckless driving is already a crime, and you don't need any special laws. I suppose that DUI is reasonable...but I've seen lots of drivers that weren't drunk that *I* thought should have been pulled over, and possibly had their license lifted. Probably what's needed is actual enforcement of the laws against reckless driving.
...Microsoft has been punished already. Time to move on....
???
Microsoft is currently up in front of courts in Europe on several different charges. It has been ordered to change it's behavior in certain specific respects, and has continued to refuse to do so. Not verbally, but via it's actions.
In addition the recent plausibly illegal vote manipulation in favor of OOXML has yet to be investigated, though there are reports of several different countries starting investigations. We probably won't hear the results of that for years.
"Time to move on" is either the last ditch appeal of the cornered con-man, or the cry of someone who has been conditioned by TV to short sound bytes and factoids. (Is it 15 minutes between advertisements, or 15 minutes between the start of one series of advertisements and the start of the next? I've stopped watching, so I no longer know.)
I'll agree that it's probably political grandstanding...but it's also a reasonable question. I'm not certain that it's legal for a government to buy from a country in MS' legal position. Naturally this will vary from country to country. That any such law will be selective in it's enforcement is an unfortunate truth.
From what I remember from an article (decades ago!), the code released in such a manner isn't the complete code, and isn't enough to compile. As such, you can use it as example cases, or to find bugs or fixes, but you can't test that it's the actual code used by any particular MS release.
Implied by this is that you couldn't use said code to create an alternative to MS OS. Not "you couldn't legally do it" but "you can't do it for any reason".
Well, scanning the article, it still seems to be a promise rather than an actual delivery. AFAICT they haven't even announced the license that they're going to use for the final release. (I'll make a wild guess that it's that abortion that MS recently got the OSI to approve.)
It all depends on how closely you define "earthlike". If you just mean something nearly the right size in close to the right position, then I expect that you're right. They're common. If, OTOH, you mean a place where we could live without space-suits...then it's a lot less likely.
What most people don't internalize is just how special the moon has made Earth. That was a humongous collision back then, any it probably stripped off a large amount of atmosphere. Without that, it would require a much lighter world to be "earthlike". (Venus is too heavy.) But Venus is already light enough that it's techtonic plates have locked (though that might be due to dessication).
Small worlds mean the core cools quickly. Large worlds mean deep atmospheres.
Worlds suitable for SOME kind of life may be common. Worlds suitable for us are probably an extreme rarity.
And I haven't even considered the allergenic possibilities of alien life. Plan on living in domes, space-suits, and space ships when you're off-planet.
This would be more like distributing a diff file that corrected typos. Still legal under US copyright law, I believe. Other countries might have differing rules. (E.g., a diff file requires short quotes before and after the change, and Australian copyright law reportedly has no "fair use" provision.
Personally, while I ceased buying CDs, I did not then begin downloading music or videos. Actually, I haven't even bothered to figure out how to play a MP3, though I'm rather certain that it would be easy.
I also stopped going to movies. (TV I gave up years ago, because the ads were too irritating.)
And yes, I still consider those who commit copyright infringement on the member companies of the RIAA or MPAA to be relatively innocent. It's not something I would do, but it a much less significant crime than bribing congress. If I thought that it actually injured the aforementioned companies, then I'd not only support it, but actually participate. Unfortunately, it does them no harm, so I'm just boycotting (which doesn't do them much harm, but it's legal, to the extent that that matters).
Of course, possibly I'm giving you too much credit. You may be an astroturfer or a troll. But some people do seem to actually believe as you do, and really do justify themselves by saying "everybody does it", irrational as that may be. Unethical action doesn't become ethical just because "everybody does it". OTOH, if, indeed, "everybody does it", it might be wise to examine carefully the grounds upon which you have decided that it's unethical. Usually, however, the ground that really need questioning are those that cause you to conclude that "everybody does it". And occasionally you may find that your ethics are flawed. Also frequently, unfortunately, you find that people just do what's convenient for them, even if it really is unethical. In such cases for YOU to participate remains unethical.
The courts have, in their "wisdom", decided that militia means "run by the government". That this isn't what the word meant at the time has long been known, but the courts decided what was politically desirable.
Why was the vote count for 2004 sealed? Because of political expediency. The courts rate political expedience over truth, justice, tradition, and the plain meaning of the words. If you're involved in a case, you'd best pray that it's either of negligible importance, or that it offers the court a chance to, by asserting your position, make a political point that it desires to make.
I figured that I'd start where I might have more effect. I helped campaign for a local city council candidate. A week or two after she took office she traded the area I was in to another councilman in exchange for a "more secure" area.
My new council member is probably corrupt and definitely powerful. Also definitely deaf to complaints... actually that's not true. He just does what he wants, and doesn't mention it until later. Like chopping down the trees in front of people's houses. Afterwards he will issue a formal apology, so he does listen...or his aides do. But it doesn't change his actions. *I* think he gets a rake off from contractors hired by the city, but I have no evidence. OTOH, he also definitely doesn't like trees...so he could just be acting out of prejudice.
Nancy Pelosi is the real villain here. She got elected promising that "impeachment is not on the table", and she's lived up to her campaign promise. That one, at least.
Since she's the Speaker, everybody knows that impeachment isn't going anywhere until she changes her mind. She can prevent it from even getting on the floor. So why bother (except for publicity).
The problem is, there's very little we can do. We can hope that most of the damage he's done is already known, and that we can recover. Beyond that???
I've voted in every election, even though with Diebold counting the votes I'm not at all convinced it matters. My other suggestions would be violently inflammatory, and also wouldn't do any good. Not while Cheney is VP. Campaigning for impeachement runs up against the stone wall of Nancy Pelosi. My she die of an incurable and disgusting disease. And news media are censoring public demonstrations. Even ones organized by congressional members.
So why not distract myself and think about something less painful.
My mistake. It thought it was earlier. Certainly it was before software patents had become as abusive as they now are (or at least before they were as blatantly abusive).
Still, this only increases my admiration for the foresight of RMS.
I'm not certain that this is a bad thing. Not certain. It might be a good thing. Perhaps.
OTOH, it makes me more interested in OpenSolaris, and it makes me wish the Hurd people would stop starting over from scratch. It also makes me more interested in BSD, even though I prefer the GPL.
Sun has talked about releasing OpenSolaris under GPL3. If they do, I'm going to be VERY interested.
If Linux were under GPL3, I wouldn't be worried about Novell. It isn't. The language of the GPL is equivocal in terms of what it means WRT patents. I think it means that if you can't distribute something legally and allow those who receive it to also distribute it, then you don't have the right to distribute it. Unfortunately, it's not totally clear about this. It was written before software could be patented, and it certainly didn't contemplate patents like a patent on adding 2 + 2 in Basic. (That particular one is harmless...but it's a magnificent example of the kind of foolish stuff that's allowed to be patented.) So the writing of the GPL frequently used general terms. Terms which apply with equal force to trademarks, copyrights, and patents, even though all are very distinct in their limitations and powers. As a result, it allows patents to be used in most places that it allows trademarks to be used. UGH!!! A sensible interpretation of the GPL would, indeed, mean that the GPL3 was unnecessary. I don't feel like I can count on the courts coming to a sensible interpretation in any reasonable amount of time.
So I trust GPL3 code coming from Novell. Other code...leaves me hesitant.
This is sort of like how I feel about Mono. I'm not certain it's booby-trapped, but I can't tell, so I'd rather avoid it. I'm risk averse. I know it. I've always been risk averse. To me, trusting Novell looks like excessive risk. I *hope* their code is being thoroughly vetted by those who know better than I do what's dangerous. I fear it isn't.
Somehow I'm not surprised. This *is* Adobe we're talking about.
What I'm not clear on is "Who is this Linux Foundation?" Google seems to indicate that it's a recent creation of a bunch of companies, and that they claim that they're organized to promote FOSS, but I didn't follow things any further. I'm not at all certain that they should be trusted. I'm not sure they shouldn't be, but allowing Adobe to join seems to indicate that they probably shouldn't.
When I think of Adobe, I think of how they sic'ed the Feds on Dimitri Skylarov, and they stood back and protested that they weren't involved. I wouldn't trust them one inch. If things started to get built in Air, they'd probably pull a Microsoft and change the specs, auto-upgrading the installed base to the new version.
I didn't trust Java until it was GPL'd, even though there were competing implementations. I don't think I plan on trusting an application with a sole-source implementation under a closed license from a company I think of as villainous.
Try it this way: Most projects produce terrible code. That'ld be correct. The license is an insignificant contribution...and I can't even tell in which direction, so it may not matter at all.
I think that I tend to produce very good code...but it takes me a long time compared to many others. Possibly there's a tradeoff here.
The nice thing about FOSS projects is that it lets you get SOMETHING out there fast, and then if there's a need, it will be continually improved. I do believe that commercially sponsored projects can develop large applications to a decent level of reliability more quickly. For this purpose ANY FOSS license will work....any that's accepted by the community, and I believe that GPL3 readily qualifies here. It may well be, however, that GPL3 projects are less likely to attract commercial sponsors. OK. They aren't allowed to fork the code into a closed tree. That makes it less attractive to them. But more attractive to me.
Over time I believe that the FOSS projects will provide the higher quality code that fit better with the need of both the developers and the end-users. This doesn't mean that it will happen as quickly with FOSS code as it would with commercial code. But it will be more usable, and more tailored to the "customers" desires. (And, naturally, also more tailored to the desires of the developers.) Over time.
Anecdotally the use of MOST FOSS licenses is increasing. (Not all. Artistic seems on the decline, as does the GNAT variant of the GPL. I'd say the same about the Eiffel variant, but it never was popular enough to estimate. And about most I don't have even enough anecdotal evidence to make a guess.)
P.S.: I *do* count gcc. That may have been a hostile fork, as you say, but it was one specifically allowed by the license. I also count X Window, for the same reason. That's a part of how FOSS operates.
Right. They should be nailed for the Sony/BMG rootkit, and the bastardly response they made when it came to public attention.
I won't deal with Sony because of that rootkit, but I feel sort of sorry about it. If Sony/BMG were driven into bankruptcy I'd... well, I wouldn't rejoice, because I don't care quite enough. But my emotions would tend in that direction.
Slashdot is the kind of place that lets anonymous people post their opinions without any evidence to back them up. You don't find credibility at a place like that. What you might find...
What you might find is an argument that you have an informed opinion on. Or a place where you can just mouth off. And if you're anonymous, who's going to care.
I'm not certain that you're a troll. You seem too confused.
Interesting. I'd never even *thought* of ssh'ing to myself.
OTOH, I'm not certain that's the real answer. I can su to another user, but when as the other user I try to execute firefox it says:
(firefox-bin:21096): Gtk-WARNING **: cannot open display:
Since I'm at that point, as far as I can tell, logged in as that user, and since that user can open GUI's when logged in as himself...It looks more as if X Window won't start up twice (reasonable).
OK, Just tried: ssh user@localhost firefox
after establishing that localhost was, indeed, a valid host I got (firefox-bin:21136): Gtk-WARNING **: cannot open display:
I eventually found it after I looked up setuid, and found that *it* didn't have a man page either, so I knew I needed to look elsewhere. I'd just never heard of seteuid before. I'm still not exactly sure what it's purpose is. setuid makes sense. I guess I need to study a bit more, and either I'll eventually figure out why it's needed, or I'll decide it doesn't have anything to do with the kind of things I do.
The problem with setuid/seteuid is that they need to be issued at program edit time rather than at run time.
What I want is to change the user id (and options) of an application at invocation time to be something that was not predictable at an earlier time. And isn't predictably the same on all systems. Also I want all files created by the application to be created with the specified UID and in the home directory of the specified user. And for the application to only be able to read files in that directory.
I think that something like this could be managed with SELinux, but that's really heavy-weight overkill for what I'm after (so I haven't investigated).
I can already do this with non-gui applications, so that's the main part of the work already done. But I don't know *why* I can't do this with GUI applications. And there may be a good reason, so I don't want to push.
Read the earlier thread about Flash and broker processes. Apparently the security system was circumvented by Adobe, so there was no reason for the browser not to be MSIE. Or not to be Firefox. It was an external "broker process" that was running as a service for Flash that was the probable culprit.
And apparently that same broker process runs on the Mac and Linux versions of Flash. But the contest rules only allowed a contestant to claim one prise. He said it would be "a couple of hours work to get it running on Linux", and it would probably be wise to take him at his word. Flash is a trojan. It installs a service and blocks future warnings when the service is activated. That it's normally "used as intended" doesn't mitigate this.
Odd...I kept using MSWind. MSWind95. I still use it. It's on an isolated machine not connected to the net. But some programs won't run on anything else, and I STILL need those programs.
Of course I also have Linux and Mac...but Mac changed their EULA last year (or I noticed a change last year) and I won't be buying any more of them. I'm currently campaigning to get them disconnected from the net. (They are, all but one. Unfortunately, that one is the most important one.)
Take these problems (and retained old systems) as evidence that open file formats are MANDATORY. Don't buy any software that won't export files into open formats without loss.
I think that's actually a pretty good approach, and one that I have long supported.
Well, what I want is a slight variation. I want to be able to do things like:
su network -c "firefox" where network is the user name of the account. Unfortunately, if I try that I get:
Xlib: connection to ":0.0" refused by server
Xlib: No protocol specified
(firefox-bin:20368): Gtk-WARNING **: cannot open display::0.0
I realize that this only makes sense on a machine that's essentially used by only one person, but that describes my computer, and also *most* Linux systems these days.
Anyway, if this worked, then a local browser(or other isolated application) exploit couldn't damage anything but the particular files used by the application that had THAT user id. The problem appears to be in XWindow, but my attempts to configure XHost didn't allow an altered user ID to open a GUI window. (They can do most anything else, though. Perhaps there's a good reason why this isn't allowed.)
That was modded funny, and it is...but it's also true if you parse it correctly.
"If you're a lawyer, never tell this to a doctor." should be parsed as "If you tell a doctor you're a lawyer, you'll never be able to get an appointment or medical service."
That's still slightly an exaggeration...but only slightly.
Slashdot Mirror
What percentage of the Darwin Award winners were high on drugs?
People are perfectly capable of doing stupid things on their own, without external stimuli. The only stupid things that people on pot do frequently are 1) drive under the influence (see alcohol) and 2) eat excessively.
The driving *is* a problem, but legalization might reduce it. Might not, too. Hard to predict. Anyway, reckless driving is already a crime, and you don't need any special laws. I suppose that DUI is reasonable...but I've seen lots of drivers that weren't drunk that *I* thought should have been pulled over, and possibly had their license lifted. Probably what's needed is actual enforcement of the laws against reckless driving.
...Microsoft has been punished already. Time to move on. ...
???
Microsoft is currently up in front of courts in Europe on several different charges. It has been ordered to change it's behavior in certain specific respects, and has continued to refuse to do so. Not verbally, but via it's actions.
In addition the recent plausibly illegal vote manipulation in favor of OOXML has yet to be investigated, though there are reports of several different countries starting investigations. We probably won't hear the results of that for years.
"Time to move on" is either the last ditch appeal of the cornered con-man, or the cry of someone who has been conditioned by TV to short sound bytes and factoids. (Is it 15 minutes between advertisements, or 15 minutes between the start of one series of advertisements and the start of the next? I've stopped watching, so I no longer know.)
I'll agree that it's probably political grandstanding...but it's also a reasonable question. I'm not certain that it's legal for a government to buy from a country in MS' legal position. Naturally this will vary from country to country. That any such law will be selective in it's enforcement is an unfortunate truth.
From what I remember from an article (decades ago!), the code released in such a manner isn't the complete code, and isn't enough to compile. As such, you can use it as example cases, or to find bugs or fixes, but you can't test that it's the actual code used by any particular MS release.
Implied by this is that you couldn't use said code to create an alternative to MS OS. Not "you couldn't legally do it" but "you can't do it for any reason".
Well, scanning the article, it still seems to be a promise rather than an actual delivery. AFAICT they haven't even announced the license that they're going to use for the final release. (I'll make a wild guess that it's that abortion that MS recently got the OSI to approve.)
It all depends on how closely you define "earthlike". If you just mean something nearly the right size in close to the right position, then I expect that you're right. They're common. If, OTOH, you mean a place where we could live without space-suits...then it's a lot less likely.
What most people don't internalize is just how special the moon has made Earth. That was a humongous collision back then, any it probably stripped off a large amount of atmosphere. Without that, it would require a much lighter world to be "earthlike". (Venus is too heavy.) But Venus is already light enough that it's techtonic plates have locked (though that might be due to dessication).
Small worlds mean the core cools quickly. Large worlds mean deep atmospheres.
Worlds suitable for SOME kind of life may be common. Worlds suitable for us are probably an extreme rarity.
And I haven't even considered the allergenic possibilities of alien life. Plan on living in domes, space-suits, and space ships when you're off-planet.
This would be more like distributing a diff file that corrected typos. Still legal under US copyright law, I believe. Other countries might have differing rules. (E.g., a diff file requires short quotes before and after the change, and Australian copyright law reportedly has no "fair use" provision.
You are, I believe, describing yourself.
Personally, while I ceased buying CDs, I did not then begin downloading music or videos. Actually, I haven't even bothered to figure out how to play a MP3, though I'm rather certain that it would be easy.
I also stopped going to movies. (TV I gave up years ago, because the ads were too irritating.)
And yes, I still consider those who commit copyright infringement on the member companies of the RIAA or MPAA to be relatively innocent. It's not something I would do, but it a much less significant crime than bribing congress. If I thought that it actually injured the aforementioned companies, then I'd not only support it, but actually participate. Unfortunately, it does them no harm, so I'm just boycotting (which doesn't do them much harm, but it's legal, to the extent that that matters).
Of course, possibly I'm giving you too much credit. You may be an astroturfer or a troll. But some people do seem to actually believe as you do, and really do justify themselves by saying "everybody does it", irrational as that may be. Unethical action doesn't become ethical just because "everybody does it". OTOH, if, indeed, "everybody does it", it might be wise to examine carefully the grounds upon which you have decided that it's unethical. Usually, however, the ground that really need questioning are those that cause you to conclude that "everybody does it". And occasionally you may find that your ethics are flawed. Also frequently, unfortunately, you find that people just do what's convenient for them, even if it really is unethical. In such cases for YOU to participate remains unethical.
The courts have, in their "wisdom", decided that militia means "run by the government". That this isn't what the word meant at the time has long been known, but the courts decided what was politically desirable.
Why was the vote count for 2004 sealed? Because of political expediency. The courts rate political expedience over truth, justice, tradition, and the plain meaning of the words. If you're involved in a case, you'd best pray that it's either of negligible importance, or that it offers the court a chance to, by asserting your position, make a political point that it desires to make.
I figured that I'd start where I might have more effect. I helped campaign for a local city council candidate. A week or two after she took office she traded the area I was in to another councilman in exchange for a "more secure" area.
My new council member is probably corrupt and definitely powerful. Also definitely deaf to complaints... actually that's not true. He just does what he wants, and doesn't mention it until later. Like chopping down the trees in front of people's houses. Afterwards he will issue a formal apology, so he does listen...or his aides do. But it doesn't change his actions. *I* think he gets a rake off from contractors hired by the city, but I have no evidence. OTOH, he also definitely doesn't like trees...so he could just be acting out of prejudice.
Nancy Pelosi is the real villain here. She got elected promising that "impeachment is not on the table", and she's lived up to her campaign promise. That one, at least.
Since she's the Speaker, everybody knows that impeachment isn't going anywhere until she changes her mind. She can prevent it from even getting on the floor. So why bother (except for publicity).
May she die of a disgusting disease.
The problem is, there's very little we can do. We can hope that most of the damage he's done is already known, and that we can recover. Beyond that???
I've voted in every election, even though with Diebold counting the votes I'm not at all convinced it matters. My other suggestions would be violently inflammatory, and also wouldn't do any good. Not while Cheney is VP. Campaigning for impeachement runs up against the stone wall of Nancy Pelosi. My she die of an incurable and disgusting disease. And news media are censoring public demonstrations. Even ones organized by congressional members.
So why not distract myself and think about something less painful.
My mistake. It thought it was earlier. Certainly it was before software patents had become as abusive as they now are (or at least before they were as blatantly abusive).
Still, this only increases my admiration for the foresight of RMS.
I'm not certain that this is a bad thing. Not certain. It might be a good thing. Perhaps.
OTOH, it makes me more interested in OpenSolaris, and it makes me wish the Hurd people would stop starting over from scratch. It also makes me more interested in BSD, even though I prefer the GPL.
Sun has talked about releasing OpenSolaris under GPL3. If they do, I'm going to be VERY interested.
If Linux were under GPL3, I wouldn't be worried about Novell. It isn't. The language of the GPL is equivocal in terms of what it means WRT patents. I think it means that if you can't distribute something legally and allow those who receive it to also distribute it, then you don't have the right to distribute it. Unfortunately, it's not totally clear about this. It was written before software could be patented, and it certainly didn't contemplate patents like a patent on adding 2 + 2 in Basic. (That particular one is harmless...but it's a magnificent example of the kind of foolish stuff that's allowed to be patented.) So the writing of the GPL frequently used general terms. Terms which apply with equal force to trademarks, copyrights, and patents, even though all are very distinct in their limitations and powers. As a result, it allows patents to be used in most places that it allows trademarks to be used. UGH!!! A sensible interpretation of the GPL would, indeed, mean that the GPL3 was unnecessary. I don't feel like I can count on the courts coming to a sensible interpretation in any reasonable amount of time.
So I trust GPL3 code coming from Novell. Other code...leaves me hesitant.
This is sort of like how I feel about Mono. I'm not certain it's booby-trapped, but I can't tell, so I'd rather avoid it. I'm risk averse. I know it. I've always been risk averse. To me, trusting Novell looks like excessive risk. I *hope* their code is being thoroughly vetted by those who know better than I do what's dangerous. I fear it isn't.
Somehow I'm not surprised. This *is* Adobe we're talking about.
What I'm not clear on is "Who is this Linux Foundation?" Google seems to indicate that it's a recent creation of a bunch of companies, and that they claim that they're organized to promote FOSS, but I didn't follow things any further. I'm not at all certain that they should be trusted. I'm not sure they shouldn't be, but allowing Adobe to join seems to indicate that they probably shouldn't.
When I think of Adobe, I think of how they sic'ed the Feds on Dimitri Skylarov, and they stood back and protested that they weren't involved. I wouldn't trust them one inch. If things started to get built in Air, they'd probably pull a Microsoft and change the specs, auto-upgrading the installed base to the new version.
I didn't trust Java until it was GPL'd, even though there were competing implementations. I don't think I plan on trusting an application with a sole-source implementation under a closed license from a company I think of as villainous.
Try it this way: Most projects produce terrible code. That'ld be correct. The license is an insignificant contribution...and I can't even tell in which direction, so it may not matter at all.
I think that I tend to produce very good code...but it takes me a long time compared to many others. Possibly there's a tradeoff here.
The nice thing about FOSS projects is that it lets you get SOMETHING out there fast, and then if there's a need, it will be continually improved. I do believe that commercially sponsored projects can develop large applications to a decent level of reliability more quickly. For this purpose ANY FOSS license will work....any that's accepted by the community, and I believe that GPL3 readily qualifies here. It may well be, however, that GPL3 projects are less likely to attract commercial sponsors. OK. They aren't allowed to fork the code into a closed tree. That makes it less attractive to them. But more attractive to me.
Over time I believe that the FOSS projects will provide the higher quality code that fit better with the need of both the developers and the end-users. This doesn't mean that it will happen as quickly with FOSS code as it would with commercial code. But it will be more usable, and more tailored to the "customers" desires. (And, naturally, also more tailored to the desires of the developers.) Over time.
Anecdotally the use of MOST FOSS licenses is increasing. (Not all. Artistic seems on the decline, as does the GNAT variant of the GPL. I'd say the same about the Eiffel variant, but it never was popular enough to estimate. And about most I don't have even enough anecdotal evidence to make a guess.)
P.S.: I *do* count gcc. That may have been a hostile fork, as you say, but it was one specifically allowed by the license. I also count X Window, for the same reason. That's a part of how FOSS operates.
Right. They should be nailed for the Sony/BMG rootkit, and the bastardly response they made when it came to public attention.
... well, I wouldn't rejoice, because I don't care quite enough. But my emotions would tend in that direction.
I won't deal with Sony because of that rootkit, but I feel sort of sorry about it. If Sony/BMG were driven into bankruptcy I'd
Slashdot had credibility???
...
Slashdot is the kind of place that lets anonymous people post their opinions without any evidence to back them up. You don't find credibility at a place like that. What you might find
What you might find is an argument that you have an informed opinion on. Or a place where you can just mouth off. And if you're anonymous, who's going to care.
I'm not certain that you're a troll. You seem too confused.
Interesting. I'd never even *thought* of ssh'ing to myself.
OTOH, I'm not certain that's the real answer. I can su to another user, but when as the other user I try to execute firefox it says:
(firefox-bin:21096): Gtk-WARNING **: cannot open display:
Since I'm at that point, as far as I can tell, logged in as that user, and since that user can open GUI's when logged in as himself...It looks more as if X Window won't start up twice (reasonable).
OK, Just tried:
ssh user@localhost firefox
after establishing that localhost was, indeed, a valid host I got
(firefox-bin:21136): Gtk-WARNING **: cannot open display:
So that's not the answer.
Right on the first bounce. Debian.
I eventually found it after I looked up setuid, and found that *it* didn't have a man page either, so I knew I needed to look elsewhere. I'd just never heard of seteuid before. I'm still not exactly sure what it's purpose is. setuid makes sense. I guess I need to study a bit more, and either I'll eventually figure out why it's needed, or I'll decide it doesn't have anything to do with the kind of things I do.
The problem with setuid/seteuid is that they need to be issued at program edit time rather than at run time.
What I want is to change the user id (and options) of an application at invocation time to be something that was not predictable at an earlier time. And isn't predictably the same on all systems. Also I want all files created by the application to be created with the specified UID and in the home directory of the specified user. And for the application to only be able to read files in that directory.
I think that something like this could be managed with SELinux, but that's really heavy-weight overkill for what I'm after (so I haven't investigated).
I can already do this with non-gui applications, so that's the main part of the work already done. But I don't know *why* I can't do this with GUI applications. And there may be a good reason, so I don't want to push.
Read the earlier thread about Flash and broker processes. Apparently the security system was circumvented by Adobe, so there was no reason for the browser not to be MSIE. Or not to be Firefox. It was an external "broker process" that was running as a service for Flash that was the probable culprit.
And apparently that same broker process runs on the Mac and Linux versions of Flash. But the contest rules only allowed a contestant to claim one prise. He said it would be "a couple of hours work to get it running on Linux", and it would probably be wise to take him at his word. Flash is a trojan. It installs a service and blocks future warnings when the service is activated. That it's normally "used as intended" doesn't mitigate this.
Odd...I kept using MSWind. MSWind95. I still use it. It's on an isolated machine not connected to the net. But some programs won't run on anything else, and I STILL need those programs.
Of course I also have Linux and Mac...but Mac changed their EULA last year (or I noticed a change last year) and I won't be buying any more of them. I'm currently campaigning to get them disconnected from the net. (They are, all but one. Unfortunately, that one is the most important one.)
Take these problems (and retained old systems) as evidence that open file formats are MANDATORY. Don't buy any software that won't export files into open formats without loss.
dala1:~/downloads/D$ man seteuid
No manual entry for seteuid
Is seteuid some MSWind specific call? Or is it a typo for setuid? Or is it something else?
I'm not now and never have been an MSWind system programmer. (I'll admit to a bit of MSAccess, though...but that was long ago.)
I think that's actually a pretty good approach, and one that I have long supported.
:0.0
Well, what I want is a slight variation. I want to be able to do things like:
su network -c "firefox"
where network is the user name of the account. Unfortunately, if I try that I get:
Xlib: connection to ":0.0" refused by server
Xlib: No protocol specified
(firefox-bin:20368): Gtk-WARNING **: cannot open display:
I realize that this only makes sense on a machine that's essentially used by only one person, but that describes my computer, and also *most* Linux systems these days.
Anyway, if this worked, then a local browser(or other isolated application) exploit couldn't damage anything but the particular files used by the application that had THAT user id. The problem appears to be in XWindow, but my attempts to configure XHost didn't allow an altered user ID to open a GUI window. (They can do most anything else, though. Perhaps there's a good reason why this isn't allowed.)
That was modded funny, and it is...but it's also true if you parse it correctly.
"If you're a lawyer, never tell this to a doctor." should be parsed as "If you tell a doctor you're a lawyer, you'll never be able to get an appointment or medical service."
That's still slightly an exaggeration...but only slightly.