This just shows how devoid of ideas the music business has become.
DVD-video discs have been used as an audio-only medium literally for years, albeit as a niche audiophile product. The DVD spec allows DVD-video discs to contain uncompressed PCM digital audio at 24 bits and 96KHz, a big step up from red-book CD audio (16 bits and 44KHz), and therefore (theoretically) more accurate. The small enthusiast-oriented record labels that produced such titles called them DADs, presumeably for Digital Audio Disc.
At this point, we've had three separate attempts to exploit the DVD's physical format for audio purposes: SACD, DVD-A, and DAD. All of them were more expensive and less convenient than CDs have become. All of them had built-in copy protection. None of them have exactly taken the world by storm. Why should something like this be any different?
Every DVD player, every CD-ROM/DVD-ROM drive, and every purpose-built CD player for that matter, can play CDs. Portable MP3 players all include systems for moving music from CDs to the player. That means the number of devices in the world designed with CDs in mind (either by working with them natively or by working with them smoothly) is skyrocketing, maybe faster than ever before in history. The CD is probably the closest thing we've ever had to a universal, standard digital media format. It has proven to be astonishingly flexible and (for the fast-paced world of computers and gadgets) shockingly long-lived.
Can it really be smart to turn one's back on all that?
I think Microsoft's "Start Anything" ad campaign was good. It just does not get as much hype as the new "Hi, I'm a Mac. And I'm a PC" ads because they don't try to be put people down. Apple has a superiority complex. When was the last time you saw a Microsoft ad that was like that?
The conventional marketing wisdom is: When You're Number One, You Don't Talk About Number Two. Market-dominating brands don't do marketing comparisons, because they don't want to remind anyone that any competition even exists.
Hertz never compares itself to Avis, but Avis can tell you that because they're number two, they try harder. Coke ads never mention Pepsi, but Pepsi ads sometimes do show people choosing Pepsi over Coke. Dell tells you how great Dell servers are, but Sun tells you that Sun servers are better than Dell servers. Norelco tells you about the shave; Remington tells you how much cheaper they are than Norelco. And so on, ad nauseam.
So I don't see Microsoft's restraint as any sort of nobility or decency on their part. It's just standard marketing practice. Everybody who uses a computer has heard of Windows; Microsoft can only shoot themselves in the foot by reminding the public that there are alternatives. If you saw some shrill (but earnest) ad comparing Windows to Jim-Bob's Pocket Fisherman And Computer Operating System, wouldn't you suddenly be curious about Jim-Bob's product, even if it looked pretty bad next to Windows in the comparison?
You might say, "Well, gee, you should have put 'French' in your search" but is this really necessary? So there is some money to be made in "learning" search engines that tailor themselves to the user or perhaps the results could be displayed intuitively in domains of knowledge (a la Clusty [clusty.com]).
I suspect that whatever search engine innovations come along, the use of a little smarts and specificity on the user's part will always pay off.
If a search engine returns results in clusters and hierarchies, then some familiarity with that engine's clusters and hierarchies will make you a better searcher, and ultimately will make that engine a more effective and precise tool for you.
A user who has a very specific thing in mind (like a French horn) yet uses a search engine to look for the vaguest, most context-free possible word that might be related (like horn) is always going to get more noise than signal. Clusty's approach is to 1) organize the noise and 2) compress the organized noise into an outline style hierarchy so that more noise fits on a single screen. That's not really improving precision, it's really a tacit admission that your search string sucks. It may in fact be a good idea to do 1) and 2), but neither addresses the fundamental difficulty - that you refused to provide information that would have dramatically improved the precision of the results.
In part, this is just the natural result when highly technical products and services are sold to a relatively ignorant public in a competitive system.
When the users aren't clued-in enough to appreciate real differences between service/product A and service/product B, claimed differences become more important, from an economic point of view.
If provider A claims N Mbps, provider B better counter with similar speeds or lower prices. If the users, by and large, wouldn't actually know a Mbps if it hit them on the head, then the easiest and most profitable way to compete is claim to provide N+1 Mbps. After all, for most light web browsing / chat-room / e-mail users, 1Mbps and 10Mbps connections provide similar experiences. What the service really is capable of is less important than the way the users feel about it.
The same circumstances drove claimed CD-ROM drive speeds into meaningless exaggeration in the late 90s. The same circumstances drove Intel to chase gigahertz rather than real-world performance in the Pentium IV line. The same circumstances cause Wi-Fi equipment vendors to make wild claims of 100+ Mbps speeds, when users will be lucky to see a tenth of that.
The phenomenon applies to other fields as well. Digital cameras make a big deal about megapixels, because that's easy to measure and compare, even though image quality is about more than megapixels, even though other, non-image-quality issues may be of far more importance. Plenty of owners of status-symbol watches have no idea what "jewel" means in that context, but are confident that more is better. Few owners of cars with badges like "DOHC" or "VTEC" can give a coherent explanation of what those badges mean, but the badged cars sell for a premium anyway.
...if you're going to claim he outright lied to the American people about the programs, this is where you need a source that doesn't lie by exclusion. Left out of the incriminating Bush quotes on wiretapping was this vital tidbit: "The Patriot Act authorizes what are called delayed notification search warrants." Without that, one could easily surmise that he was, in fact, directly contradicting himself.
Is this the section you're referring to?
Bush:...there's something called delayed notification warrants. Those are very important. I see some people, first responders nodding their heads about what they mean. These are a common tool used to catch mobsters. In other words, it allows people to collect data before everybody is aware of what's going on. It requires a court order. It requires protection under the law. We couldn't use these against terrorists, but we could use against gangs.
In other words, the "delayed notification" nature of a search may change the timeline of the legal proceedings that go on, but it doesn't change the fundemental requirement for a warrant, which Bush has deliberately avoided. "Delayed notification" means the authorities are permitted to notify the subject of a search after the search is over, instead of at the outset. How that applies to wiretaps isn't clear, since the subject of a wiretap is never notified while the wiretap is happening.
I don't see how this lets Bush off the hook. He assured us that he was pursuing terror suspects in a legal manner with judicial oversight. It wasn't true, and he knew it wasn't true. He said it anyway.
Yet many people (myself excluded) DO still trust him. Your assessment of his trustworthiness, while shared by fewer and fewer people every day, is still opinion, not fact.
I hope you meant to say that my assessment of his trustworthiness is shared by more and more people every day.
I agree that this comes closer to being an opinion than anything else I said. But to dismiss it as strictly an opinion is to say that "trust" and "trustworthiness" are strictly subjective concepts, and that every statement about trust is automatically an opinion. Do you really believe that? Is it never possible to confidently say "So-and-so cannot be trusted" knowing that the statement is a true fact?
I wish people such as yourself would stop confusing "strong opinion" for "fact".
Could you be more specific? I don't think I'm doing that here.
Speak softly and carry a big stick; which in this case would be credible (read: not left-leaning blogs) citations.
You're implying that my sources are no good without actually making a specific accusation.
My three links were:
1) A video and transcript of the opening of the Senate Judiciary Committee session. A factual record, not an opinion piece.
2) A Washington Post news story. Not an opinion piece. Not a left-leaning blog.
3) Remarks concerning the PATRIOT act by President George W. Bush, which included links to the official transcripts.
None of these items are matters of "strong opinion," and I frankly think that would be clear from even a cursory examination of the actual linked pages rather than just the domain names. The sources in these cases are ultimately 1) The Senate 2) The Attorney General 3) The President.
Fact is, the NSA program still is for US to offshore calls.
The fact is that when Gonzales told us that the NSA only listens to international calls he didn't take an oath, and that he later rescinded or re-qualified much of his testimony, in particular making the point that although one particlar intelligence program involved listening to international calls, a certain gigantic multi-billion dollar signals-intelligence agency might actually be running more than one signals-gathering program. (Like, whoa, seriously?)
To attach the name "fact" to information obtained under such circumstances is, I think, very, very optimistic, in a sense.
The fact is that Bush has told us bald-faced lies about domestic spying activities, and at this point it would require a hearty steaming ladle-full of naivete to imagine that the general public now knows the truth, the whole truth, and nothing but the truth.
The fact is that the Bush administration has aptly demonstrated that it simply cannot be trusted, and the sorry fact is that we can be certain of precious little beyond that.
The fact is, recognizing these facts does not constitute "fanaticism." I believe there's a saying down in Texas: "Fool me once, shame on you. Fool me -- you can't get fooled again."
Given the extremely vague nature of what Cory has posted I'm not surprised they are suing. Is it right for them to sue? Probably not...
They're not suing, at least, not yet.
It's not just a question of "Is it right to [threaten to] sue," it's also a question of "Is it smart to [threaten to] sue?" When you stand accused of something, threatening your accuser makes you look guilty as hell.
The PR guy could have taken a friendly approach, and tried to convince BoingBoing that StarForce was the victim of erroneous rumours. That would put the megaphone of BoingBoing on his side correcting those rumours.
Instead, he threatened them, and now he's now got BoingBoing very loudly proclaiming that his company - the one he's supposed to be doing PR for - is so incredibly evil, even their PR department does a convincing "Mwa-ha-ha!"
There are a few ways to explain this that I can think of: 1) The StarForce software is every bit as dangerous as claimed, 2) the PR guy is a huge moron, 3) Or possibly both.
...it's very reasonable for it to create a separate thread...
The impression I got from the article was that Gibson thought an extra thread was created specifically when a magic number was found in the WMF file, and not just to deal with WMF parsing up-front. If that is the case, it's extremely damning.
Looking at the article again, though, Gibson isn't that explicit. It doesn't make it clear when that thread is created. It doesn't even make it clear whether or not Gibson knows when that thread is created. Your interpretation is also entirely reasonable.
Even without access to the most sophisticated debugging tools, one would imagine this could be discovered by counting how many threads are running when parsing a normal WMF vs. a backdoored WMF.
Gibson is an extremely talented guy, and I'm surprised he hasn't nailed this down more firmly. Perhaps he's better in text than in extemporaneous interviews.
It's also undeniable that Gibson has some paranoid-fruitcake tendencies, although these days I think we need all the paranoid fruitcakes we can get our hands on.
For me, that length==1 trigger is the most convincing evidence.
I don't think it's surprising that a piece of code might behave in an odd way if it's given invalid input, i.e., if a buffer length is wrong.
I think the real giveaway here is that Windows creates a new thread when presented with this magic length. That's like rolling out the red carpet for the attacking Huns. I don't think the average buffer overflow type exploit gets it's own thread or process.
And of course it's still possible that it was all a mistake. The C language can be used to write some extremely tangled code, if one is so inclined. Something like an incorrectly used setjmp/longjmp could have effects like this.
Caller ID is not free, it is an optional pay service on most carriers, and not everyone has it, so not everyone can just sit there and screen the calls.... I guess you could block the caller... but that too incurs a fee.
Well, heck, the phone line itself isn't free either. I don't see how "X is not free" is a refutation of "X works extremely well."
There are phones and add-on boxes with built-in caller-ID based filtering, meaning its not necessary to pay a recurring fee to block certain phone numbers.
I don't think it's out of line for the government to outlaw harassment.
Perhaps not, but it seems clear that technical/practical solutions (like caller-ID based blocks, e-mail filtering, etc) to the problem of unwanted communications are likely to be more effective and more enforceable than laws will be.
I've got a slightly different version of one of these lamps on top of my CRT monitor. It's so cheaply made that I feel like I'm going to break it every time I touch it. Yet there's no denying that it puts light in all the right places - on the wall behind the monitor, on the cieling, on the keyboard - but not on the screen itself, and not on anything in front of the screen that might cause reflections.
An ordinary articulated desk lamp is easily maneuvered to put light on the wall behind your monitor. However it's done, I think that arrangement is notably easier on the eyes than anything else I've tried.
There is not an 'EXEC' segement type in the metadata specification itself, if you will.
In the internet age, it's hard to believe, but in fact, yes, there is. This isn't a buffer overflow exploit; this is actually the way metafiles were intended to work. AC makes the same point a bit more rudely.
Supposing there exists a "much more robust security infrastructure" - how is it going to be improved by the addition of a Play-Doh, uh, I mean a fingerprint scanner? Why not just stick with the robust stuff, and forget the shiny newfangled contraptions?
This isn't the first demonstration that fingerprint scanners are useless. A few years ago, a Japanese university professor showed that it was possible to make a gelatin mold from a latent print (i.e., without direct access to the authorized finger in question) that would fool the readers most of the time! What is a fingerprint scanner adding but a false sense of security?
However, the management interface of that image gallery could be [a web application]... There is absolutely no reason to bookmark in that, either, or be able to go 'back'. (Back where?)
Sure, if you build the app so that all the management happens on one page, that could be OK. If you've got some page reloads in there (and a lot of web applications can't seem to avoid them) then you've got a place to go back to, all right. And if there is page navigation involved, then bookmarking (whether you see a reason for it or not) isn't inherently nonsensical, either.
And in my view, it doesn't matter whether going "back" makes sense; it shouldn't be possible for the user to break the app by pressing the wrong button. You wouldn't accept that kind of nonsense from a desktop app.
I guess Sygate is no more - they've been borgified by Symantec, and Sygate products are being discontinued. That's a shame - the Sygate Personal Firewall was easy to set up and use, but it offered a lot more technical options and information than the average consumer security app, too.
I'm currently using a software firewall for this, however one thing I don't like about it is that it doesn't tell me which ports an application is using.
I use Sygate Personal Firewall, which at least used to be free for personal use. It can display a list of all open port numbers, indicating the responsible exe, the listen/connect status, and the blocked/allowed status for each. I'm not sure if the "Allow this program to access the network?" dialog includes the port in question or not, though.
The article is about using AJAX on a webpage, but the biggest use of AJAX is on a web application... I think the article writer was focusing mostly on webpages where AJAX is clearly geared towards the web application developer.
The way I read it, the article is making the point that most "web applications," including the newfangled Ajax sort, break the web-page paradigm. Objections along the lines of "but - but - but - it's different for web applications!" is to miss his point entirely. He's saying things damn well shouldn't be different for "web applications." From the article:
The fundamental design of the Web is based on having the page as the atomic unit of information, and the notion of the page permeates all aspects of the Web... Ajax breaks the unified model of the Web and introduce a new way of looking at data that has not been well integrated into the other aspects of the Web.
Don't get me wrong, I loves me my del.icio.us and my Flickr, but I can see that Nielsen's got "Web 2.0" dead to rights on this point anyway.
Commercial projectors can pump out a lot more light than a diy pj
I'm no expert - but are you sure? HID bulbs produce more lumens per watt than incandescent/halogen/xenon bulbs. HID vehicle headlamps manage about 3000 lumens from just 35W. A 400W HID bulb should be an absolute monster. And the relatively huge LCD could very well have a higher proportion of transmissive area than a smaller one.
That sounds to me like a recipie for putting out something like a cubic furlong of photons.
Actually, this struct doesn't appear in the Win32 or the MFC API anywhere (nor does anything that looks significantly like it)...
I beg to differ. MFC may not contain this sort of thing, but Win32 and the system API behind it absolutely, positively include lots of structs like that. Check out the serial port DCB struct, or many of the associated serial-communications related structs, for example. Check out almost any TAPI-related struct. Many other subsystems are the same, I'm sure.
Usually, the length is actually used as a version code, not a buffer limit. OS code and user code can both check the length to see which version of the struct they're dealing with. As long as it's really used that way, it's not a problem.
this kind of struct will *never* be a problem. Let's consider all of the cases:
Allocating the struct isn't the main problem. The structs Win32 hands back can be downright baroque in their complexity, including variable length data objects and pointers to those objects. An application program written with the assumption that those data objects will not exceed some documented maximum length could easily wind up with a buffer overflow on the stack when interpreting, parsing, or otherwise manipulating a maliciously constructed struct.
Let's assume for a second though that someone gives me the buffer pointer...
Aren't you hosed right there? If the pointer points to your own stack, and you write through it, then bye-bye process. If what you write is some data chunk also provided by the same malicious someone, then you could very well be dumping exploit code right into your own stack.
In this case, thats ok if that someone else was responsible for allocating & owns the struct.
Some Windows APIs work that way - the API allocates, inits, and owns the memory the struct is in, requiring the client program to make a special type of free() call when it's done using the struct.
Other Windows APIs (I personally had experience with the TAPI system) didn't work that way. Client/application code would call an API which would either return some structured data or inform you that your buffer was too small. These things were designed to be called twice - they worked something like this:
result = GiveMeSomeDataPlease(NULL, 0, &sizeRequired); if ( result == BUFFER_TOO_SMALL )/* and it always will be */ { myBuffer = malloc(sizeRequired); result = GiveMeSomeDataPlase(myBuffer, sizeRequired, &sizeActuallyUsed); }
Or sometimes, they even worked like this:
do { sizeBuffer *= 2; myBuffer = realloc(myBuffer, sizeBuffer); result = GiveMeSomeDataPlease(myBuffer, sizeBuffer); } while ( result == BUFFER_TOO_SMALL )
In effect, the API told your code how much memory to allocate, and the API dumped a bunch of data into your process's memory space. In the case of the TAPI system, that data often had a fairly complex structure, including nested, versioned structs, offset references, pointers, variable-length buffers, etc. Checking the integrity of such a thing is not trivial or practical. The application basically has to trust that all that stuff is valid and correct.
It's not hard to see that an infected API pushing a corrupt struct of this kind could at least crash another process. It's not a stretch to imagine some crazy structure data leading to a stack overflow, opening the process up to just about anything. The seriousness of that possibility may depend on which Windows subsystems use that type of interface; if there's malicious code running in your TAPI system, your PC is already completely pwn3d anyway, and there would be no reason for the malicious code to inject itself this way. But this style of interface in some internet-related system could be problematic.
Slashdot Mirror
This just shows how devoid of ideas the music business has become.
DVD-video discs have been used as an audio-only medium literally for years, albeit as a niche audiophile product. The DVD spec allows DVD-video discs to contain uncompressed PCM digital audio at 24 bits and 96KHz, a big step up from red-book CD audio (16 bits and 44KHz), and therefore (theoretically) more accurate. The small enthusiast-oriented record labels that produced such titles called them DADs, presumeably for Digital Audio Disc.
At this point, we've had three separate attempts to exploit the DVD's physical format for audio purposes: SACD, DVD-A, and DAD. All of them were more expensive and less convenient than CDs have become. All of them had built-in copy protection. None of them have exactly taken the world by storm. Why should something like this be any different?
Every DVD player, every CD-ROM/DVD-ROM drive, and every purpose-built CD player for that matter, can play CDs. Portable MP3 players all include systems for moving music from CDs to the player. That means the number of devices in the world designed with CDs in mind (either by working with them natively or by working with them smoothly) is skyrocketing, maybe faster than ever before in history. The CD is probably the closest thing we've ever had to a universal, standard digital media format. It has proven to be astonishingly flexible and (for the fast-paced world of computers and gadgets) shockingly long-lived.
Can it really be smart to turn one's back on all that?
The excellent and GPL'ed 7-Zip can unpack RAR files. Is that better?
I think Microsoft's "Start Anything" ad campaign was good. It just does not get as much hype as the new "Hi, I'm a Mac. And I'm a PC" ads because they don't try to be put people down. Apple has a superiority complex. When was the last time you saw a Microsoft ad that was like that?
The conventional marketing wisdom is: When You're Number One, You Don't Talk About Number Two. Market-dominating brands don't do marketing comparisons, because they don't want to remind anyone that any competition even exists.
Hertz never compares itself to Avis, but Avis can tell you that because they're number two, they try harder. Coke ads never mention Pepsi, but Pepsi ads sometimes do show people choosing Pepsi over Coke. Dell tells you how great Dell servers are, but Sun tells you that Sun servers are better than Dell servers. Norelco tells you about the shave; Remington tells you how much cheaper they are than Norelco. And so on, ad nauseam.
So I don't see Microsoft's restraint as any sort of nobility or decency on their part. It's just standard marketing practice. Everybody who uses a computer has heard of Windows; Microsoft can only shoot themselves in the foot by reminding the public that there are alternatives. If you saw some shrill (but earnest) ad comparing Windows to Jim-Bob's Pocket Fisherman And Computer Operating System, wouldn't you suddenly be curious about Jim-Bob's product, even if it looked pretty bad next to Windows in the comparison?
You might say, "Well, gee, you should have put 'French' in your search" but is this really necessary? So there is some money to be made in "learning" search engines that tailor themselves to the user or perhaps the results could be displayed intuitively in domains of knowledge (a la Clusty [clusty.com]).
I suspect that whatever search engine innovations come along, the use of a little smarts and specificity on the user's part will always pay off.
If a search engine returns results in clusters and hierarchies, then some familiarity with that engine's clusters and hierarchies will make you a better searcher, and ultimately will make that engine a more effective and precise tool for you.
A user who has a very specific thing in mind (like a French horn) yet uses a search engine to look for the vaguest, most context-free possible word that might be related (like horn) is always going to get more noise than signal. Clusty's approach is to 1) organize the noise and 2) compress the organized noise into an outline style hierarchy so that more noise fits on a single screen. That's not really improving precision, it's really a tacit admission that your search string sucks. It may in fact be a good idea to do 1) and 2), but neither addresses the fundamental difficulty - that you refused to provide information that would have dramatically improved the precision of the results.
Carter does not pull his punches about Gore's activism...
That's Bob Carter, the Aussie researcher who says that global warming would be a good thing, and not Jimmy Carter, the former US President.
Sure, if you read the article carefully enough to remember the names exactly, you wouldn't get confused. But look where we are.
In part, this is just the natural result when highly technical products and services are sold to a relatively ignorant public in a competitive system.
When the users aren't clued-in enough to appreciate real differences between service/product A and service/product B, claimed differences become more important, from an economic point of view.
If provider A claims N Mbps, provider B better counter with similar speeds or lower prices. If the users, by and large, wouldn't actually know a Mbps if it hit them on the head, then the easiest and most profitable way to compete is claim to provide N+1 Mbps. After all, for most light web browsing / chat-room / e-mail users, 1Mbps and 10Mbps connections provide similar experiences. What the service really is capable of is less important than the way the users feel about it.
The same circumstances drove claimed CD-ROM drive speeds into meaningless exaggeration in the late 90s. The same circumstances drove Intel to chase gigahertz rather than real-world performance in the Pentium IV line. The same circumstances cause Wi-Fi equipment vendors to make wild claims of 100+ Mbps speeds, when users will be lucky to see a tenth of that.
The phenomenon applies to other fields as well. Digital cameras make a big deal about megapixels, because that's easy to measure and compare, even though image quality is about more than megapixels, even though other, non-image-quality issues may be of far more importance. Plenty of owners of status-symbol watches have no idea what "jewel" means in that context, but are confident that more is better. Few owners of cars with badges like "DOHC" or "VTEC" can give a coherent explanation of what those badges mean, but the badged cars sell for a premium anyway.
Is this the section you're referring to?
Bush: ...there's something called delayed notification warrants. Those are very important. I see some people, first responders nodding their heads about what they mean. These are a common tool used to catch mobsters. In other words, it allows people to collect data before everybody is aware of what's going on. It requires a court order. It requires protection under the law. We couldn't use these against terrorists, but we could use against gangs.
In other words, the "delayed notification" nature of a search may change the timeline of the legal proceedings that go on, but it doesn't change the fundemental requirement for a warrant, which Bush has deliberately avoided. "Delayed notification" means the authorities are permitted to notify the subject of a search after the search is over, instead of at the outset. How that applies to wiretaps isn't clear, since the subject of a wiretap is never notified while the wiretap is happening.
I don't see how this lets Bush off the hook. He assured us that he was pursuing terror suspects in a legal manner with judicial oversight. It wasn't true, and he knew it wasn't true. He said it anyway.
Yet many people (myself excluded) DO still trust him. Your assessment of his trustworthiness, while shared by fewer and fewer people every day, is still opinion, not fact.
I hope you meant to say that my assessment of his trustworthiness is shared by more and more people every day.
I agree that this comes closer to being an opinion than anything else I said. But to dismiss it as strictly an opinion is to say that "trust" and "trustworthiness" are strictly subjective concepts, and that every statement about trust is automatically an opinion. Do you really believe that? Is it never possible to confidently say "So-and-so cannot be trusted" knowing that the statement is a true fact?
It would be more helpful to my understanding if you could be specific. Specifically what sentence or point seems like interpretation to you?
Could you be more specific? I don't think I'm doing that here.
Speak softly and carry a big stick; which in this case would be credible (read: not left-leaning blogs) citations.
You're implying that my sources are no good without actually making a specific accusation.
My three links were:
1) A video and transcript of the opening of the Senate Judiciary Committee session. A factual record, not an opinion piece.
2) A Washington Post news story. Not an opinion piece. Not a left-leaning blog.
3) Remarks concerning the PATRIOT act by President George W. Bush, which included links to the official transcripts.
None of these items are matters of "strong opinion," and I frankly think that would be clear from even a cursory examination of the actual linked pages rather than just the domain names. The sources in these cases are ultimately 1) The Senate 2) The Attorney General 3) The President.
The fact is that when Gonzales told us that the NSA only listens to international calls he didn't take an oath, and that he later rescinded or re-qualified much of his testimony, in particular making the point that although one particlar intelligence program involved listening to international calls, a certain gigantic multi-billion dollar signals-intelligence agency might actually be running more than one signals-gathering program. (Like, whoa, seriously?)
To attach the name "fact" to information obtained under such circumstances is, I think, very, very optimistic, in a sense.
The fact is that Bush has told us bald-faced lies about domestic spying activities, and at this point it would require a hearty steaming ladle-full of naivete to imagine that the general public now knows the truth, the whole truth, and nothing but the truth.
The fact is that the Bush administration has aptly demonstrated that it simply cannot be trusted, and the sorry fact is that we can be certain of precious little beyond that.
The fact is, recognizing these facts does not constitute "fanaticism." I believe there's a saying down in Texas: "Fool me once, shame on you. Fool me -- you can't get fooled again."
Given the extremely vague nature of what Cory has posted I'm not surprised they are suing. Is it right for them to sue? Probably not...
They're not suing, at least, not yet.
It's not just a question of "Is it right to [threaten to] sue," it's also a question of "Is it smart to [threaten to] sue?" When you stand accused of something, threatening your accuser makes you look guilty as hell.
The PR guy could have taken a friendly approach, and tried to convince BoingBoing that StarForce was the victim of erroneous rumours. That would put the megaphone of BoingBoing on his side correcting those rumours.
Instead, he threatened them, and now he's now got BoingBoing very loudly proclaiming that his company - the one he's supposed to be doing PR for - is so incredibly evil, even their PR department does a convincing "Mwa-ha-ha!"
There are a few ways to explain this that I can think of: 1) The StarForce software is every bit as dangerous as claimed, 2) the PR guy is a huge moron, 3) Or possibly both.
...it's very reasonable for it to create a separate thread...
The impression I got from the article was that Gibson thought an extra thread was created specifically when a magic number was found in the WMF file, and not just to deal with WMF parsing up-front. If that is the case, it's extremely damning.
Looking at the article again, though, Gibson isn't that explicit. It doesn't make it clear when that thread is created. It doesn't even make it clear whether or not Gibson knows when that thread is created. Your interpretation is also entirely reasonable.
Even without access to the most sophisticated debugging tools, one would imagine this could be discovered by counting how many threads are running when parsing a normal WMF vs. a backdoored WMF.
Gibson is an extremely talented guy, and I'm surprised he hasn't nailed this down more firmly. Perhaps he's better in text than in extemporaneous interviews.
It's also undeniable that Gibson has some paranoid-fruitcake tendencies, although these days I think we need all the paranoid fruitcakes we can get our hands on.
For me, that length==1 trigger is the most convincing evidence.
I don't think it's surprising that a piece of code might behave in an odd way if it's given invalid input, i.e., if a buffer length is wrong.
I think the real giveaway here is that Windows creates a new thread when presented with this magic length. That's like rolling out the red carpet for the attacking Huns. I don't think the average buffer overflow type exploit gets it's own thread or process.
And of course it's still possible that it was all a mistake. The C language can be used to write some extremely tangled code, if one is so inclined. Something like an incorrectly used setjmp/longjmp could have effects like this.
Caller ID is not free, it is an optional pay service on most carriers, and not everyone has it, so not everyone can just sit there and screen the calls. ... I guess you could block the caller... but that too incurs a fee.
Well, heck, the phone line itself isn't free either. I don't see how "X is not free" is a refutation of "X works extremely well."
There are phones and add-on boxes with built-in caller-ID based filtering, meaning its not necessary to pay a recurring fee to block certain phone numbers.
I don't think it's out of line for the government to outlaw harassment.
Perhaps not, but it seems clear that technical/practical solutions (like caller-ID based blocks, e-mail filtering, etc) to the problem of unwanted communications are likely to be more effective and more enforceable than laws will be.
I've got a slightly different version of one of these lamps on top of my CRT monitor. It's so cheaply made that I feel like I'm going to break it every time I touch it. Yet there's no denying that it puts light in all the right places - on the wall behind the monitor, on the cieling, on the keyboard - but not on the screen itself, and not on anything in front of the screen that might cause reflections.
An ordinary articulated desk lamp is easily maneuvered to put light on the wall behind your monitor. However it's done, I think that arrangement is notably easier on the eyes than anything else I've tried.
There is not an 'EXEC' segement type in the metadata specification itself, if you will.
In the internet age, it's hard to believe, but in fact, yes, there is. This isn't a buffer overflow exploit; this is actually the way metafiles were intended to work. AC makes the same point a bit more rudely.
Supposing there exists a "much more robust security infrastructure" - how is it going to be improved by the addition of a Play-Doh, uh, I mean a fingerprint scanner? Why not just stick with the robust stuff, and forget the shiny newfangled contraptions?
This isn't the first demonstration that fingerprint scanners are useless. A few years ago, a Japanese university professor showed that it was possible to make a gelatin mold from a latent print (i.e., without direct access to the authorized finger in question) that would fool the readers most of the time! What is a fingerprint scanner adding but a false sense of security?
A lot of movies get made and you wonder "WTF? How did this piece of drek ever scare up the capital?"
It's all the Germans' fault.
However, the management interface of that image gallery could be [a web application]... There is absolutely no reason to bookmark in that, either, or be able to go 'back'. (Back where?)
Sure, if you build the app so that all the management happens on one page, that could be OK. If you've got some page reloads in there (and a lot of web applications can't seem to avoid them) then you've got a place to go back to, all right. And if there is page navigation involved, then bookmarking (whether you see a reason for it or not) isn't inherently nonsensical, either.
And in my view, it doesn't matter whether going "back" makes sense; it shouldn't be possible for the user to break the app by pressing the wrong button. You wouldn't accept that kind of nonsense from a desktop app.
I guess Sygate is no more - they've been borgified by Symantec, and Sygate products are being discontinued. That's a shame - the Sygate Personal Firewall was easy to set up and use, but it offered a lot more technical options and information than the average consumer security app, too.
I'm currently using a software firewall for this, however one thing I don't like about it is that it doesn't tell me which ports an application is using.
I use Sygate Personal Firewall, which at least used to be free for personal use. It can display a list of all open port numbers, indicating the responsible exe, the listen/connect status, and the blocked/allowed status for each. I'm not sure if the "Allow this program to access the network?" dialog includes the port in question or not, though.
The way I read it, the article is making the point that most "web applications," including the newfangled Ajax sort, break the web-page paradigm. Objections along the lines of "but - but - but - it's different for web applications!" is to miss his point entirely. He's saying things damn well shouldn't be different for "web applications." From the article:
Don't get me wrong, I loves me my del.icio.us and my Flickr, but I can see that Nielsen's got "Web 2.0" dead to rights on this point anyway.
Commercial projectors can pump out a lot more light than a diy pj
I'm no expert - but are you sure? HID bulbs produce more lumens per watt than incandescent/halogen/xenon bulbs. HID vehicle headlamps manage about 3000 lumens from just 35W. A 400W HID bulb should be an absolute monster. And the relatively huge LCD could very well have a higher proportion of transmissive area than a smaller one.
That sounds to me like a recipie for putting out something like a cubic furlong of photons.
Actually, this struct doesn't appear in the Win32 or the MFC API anywhere (nor does anything that looks significantly like it)...
I beg to differ. MFC may not contain this sort of thing, but Win32 and the system API behind it absolutely, positively include lots of structs like that. Check out the serial port DCB struct, or many of the associated serial-communications related structs, for example. Check out almost any TAPI-related struct. Many other subsystems are the same, I'm sure.
Usually, the length is actually used as a version code, not a buffer limit. OS code and user code can both check the length to see which version of the struct they're dealing with. As long as it's really used that way, it's not a problem.
this kind of struct will *never* be a problem. Let's consider all of the cases:
Allocating the struct isn't the main problem. The structs Win32 hands back can be downright baroque in their complexity, including variable length data objects and pointers to those objects. An application program written with the assumption that those data objects will not exceed some documented maximum length could easily wind up with a buffer overflow on the stack when interpreting, parsing, or otherwise manipulating a maliciously constructed struct.
Let's assume for a second though that someone gives me the buffer pointer...
Aren't you hosed right there? If the pointer points to your own stack, and you write through it, then bye-bye process. If what you write is some data chunk also provided by the same malicious someone, then you could very well be dumping exploit code right into your own stack.
In this case, thats ok if that someone else was responsible for allocating & owns the struct.
/* and it always will be */
Some Windows APIs work that way - the API allocates, inits, and owns the memory the struct is in, requiring the client program to make a special type of free() call when it's done using the struct.
Other Windows APIs (I personally had experience with the TAPI system) didn't work that way. Client/application code would call an API which would either return some structured data or inform you that your buffer was too small. These things were designed to be called twice - they worked something like this:
result = GiveMeSomeDataPlease(NULL, 0, &sizeRequired);
if ( result == BUFFER_TOO_SMALL )
{
myBuffer = malloc(sizeRequired);
result = GiveMeSomeDataPlase(myBuffer, sizeRequired, &sizeActuallyUsed);
}
Or sometimes, they even worked like this:
do
{
sizeBuffer *= 2;
myBuffer = realloc(myBuffer, sizeBuffer);
result = GiveMeSomeDataPlease(myBuffer, sizeBuffer);
}
while ( result == BUFFER_TOO_SMALL )
In effect, the API told your code how much memory to allocate, and the API dumped a bunch of data into your process's memory space. In the case of the TAPI system, that data often had a fairly complex structure, including nested, versioned structs, offset references, pointers, variable-length buffers, etc. Checking the integrity of such a thing is not trivial or practical. The application basically has to trust that all that stuff is valid and correct.
It's not hard to see that an infected API pushing a corrupt struct of this kind could at least crash another process. It's not a stretch to imagine some crazy structure data leading to a stack overflow, opening the process up to just about anything. The seriousness of that possibility may depend on which Windows subsystems use that type of interface; if there's malicious code running in your TAPI system, your PC is already completely pwn3d anyway, and there would be no reason for the malicious code to inject itself this way. But this style of interface in some internet-related system could be problematic.