For me, it makes data semantic-free so that it can be passed between systems, and changes on the transmitting end don't fuck up the receiving end.
XML itself doesn't provide this for free. But what it does do is make the schema a first-class entity against which documents can be validated. This change, and the toolchains that followed, made the programmatic interface presented by the document format much more evident to onlookers. Put another way, the schema and the serialization format were no longer entangled. That, in turn, made it a whole heck of a lot easier for developers (and even some technical business types) to comprehend the patterns that allow backwards-compatible format changes.
Plus, it's readable.
Before someone flames the parent, XML being notoriously human-unfriendly in some circles, compare XML's readability to that of a random poorly-documented (endian-specific!!) binary format spun off by some dude who left the organization six years ago.
XML gets people who have no business doing serialization and parsing out of that business entirely. Instead, we have schemas (either purely ad-hoc, or formalized) and common libraries to express those needs. Like any computing tool, XML has been used for good, bad, and ugly ends... but it brought the idea of schema-based document format and consistent Document Object Model (DOM) APIs to the forefront of data/document handling. In essence, this is a two-level interface: the document schema (a first-class entity, unlike an ad-hoc data format) and the programming APIs.
Even if you're no big fan of XML, the good ideas, patterns, anti-patterns, and mindset changes that spread along with it have positively influenced many other useful document technologies: JSON, YAML, the newer binary serialization formats (such as Protocol Buffers, Thrift, and Avro), and more.
Devil's advocate: how is it DirecTV's fault that she unilaterally did something that made her unable to receive the service she'd paid for?
It's DirecTV's fault because to do otherwise makes them asshats. Life happens, in ways that people often can't control. Any responsible business allows for such in their policies and contracts. For example, a local gym chain that I attended for a time had not only reasonable and responsible account termination rules and practices, but they even let you out of a contract no-fuss if you moved and were no longer within some very modest distance of a franchise location. They got my business over another local gym precisely because of a much better reputation for having an honest and well-defined relationship with their customers.
It's not just photographs. This is a problem when generally when a medium is applied to a primarily technical aim (e.g. breaking a record) vs. an aesthetic one. The best example of this I've witnessed was during my freshman year of college, when a music department Prof. had the class listen to the first public recording of tape loop reverb. IIRC, it came out of MIT. The recording was performed on the recorder (the woodwind instrument) by the then-current department chair.
Now try to imagine sounds that would make Vogons would tremble in simultaneous delight and terror at this, and admit in defeat that their poetry is no equal. I can't recall hearing a brilliant rendition of anything on the recorder. Now combine that lackluster sound, with a/cough/ less than virtuoso performance and a good mangling by those first doozy steps into studio-created reverberation.... bleaaargh!
You have explicit rights (in most cases) to privacy and use of the property that you actually own. That's it, the line is drawn there. I can't believe there is any controversy over this.
Of course there's controversy! In case you haven't noticed, a vast majority of our "personal" data has wandered out onto networks and servers that none of us control. This has been a gradual process going on for years, but the very attributes of modern networked computer systems make the real-world impact of these changes much greater now. This has radically changed the landscape under which the 4th amendment (and a lot of other law) was originally conceived. I'd say that review of the applicability of constitutional protections in modern contexts could be viewed as one of the most important roles of the SCOTUS.
The people who are using nvidia's driver obviously care about 3d performance otherwise they'd already be using the open source driver with 2d support.
Not so much. I have an Ubuntu desktop box with an NVidia card (work system; I didn't spec it), and the open source driver seemed to have a number of issues when running dual head. I flipped to the binary driver, it just worked, and haven't looked back. It sounds like the Nouveau driver will bring some substantial forward progress to open source NVidia drivers. ( No thanks to NVidia themselves.:-P )
If you consider something at "University" a Youth Mistake. Most people are generally at the age of adulthood since then.
If you think someone at University at a typical post-high school age is an "adult", then practical experience, cognitive science, and auto insurer's actuarial statistics have something quite different to say. Even ignoring brain maturation issues, in today's society that's the time when most folks are away from home and on their own for the first time, and are really just starting to figure out Which End Is Up.
Or perhaps MM simply doesn't want to get the plug pulled by a conventional cloud compute provider, due to the questionable PR (and possibly other attention) that this service may
One could view this as an alternative to the old "publish the exploit as a goad to the provider" tactic. Previously, some cryptographic weaknesses required someone to have the resources to obtain a compute cluster large enough to deal with some specific cracking problem. With this approach, it isn't even necessary to be able to set up an EC2 job -- just shell out a few bucks and away-you-go. It'll definitely be a wake-up call for some folks that big compute clusters are trivially available to anyone.
Piper and his colleagues don’t know what the correct amino acid balance might be for humans, and he says it would be a nearly impossible feat to adjust people’s diets to get just the right mix. Instead, the team is investigating how tweaking amino acid content in the diet affects cells. If the researchers can identify pathways affected by amino acid imbalances, they might be able to design drugs or other therapies that could give the benefits of caloric restriction without cutting calories.
Part of the problem was also part of Salon's strength: they were started and run by writers. Old-school, ink-and-paper writers.
And that's a much, much broader problem. Traditional print publishing has been a cutthroat, generally low-margin business for decades. This created other friction, such as top web development and design talent being able to command salaries quite a bit above that of the traditional senior staff. This is one reason why the websites for publishing houses were absolute crap for a long long time -- it was too bitter a pill to hire a bright young twenty something at double the executive salaries.
In fact it can be said about every living human being who choose to do something different from what you choose to do with your time. "They should get a life."
I don't find that most folks are really that myopic. Those who work towards goals understandable to others don't tend to receive the "getalife" tag. Athleticism is widely valued and understood. This is especially so for Olympic athletes, who tend to receive admiration even from those who eschew mainstream sports. Likewise, many people appreciate those who show skill in building tangible things (e.g. hotrods, etc.), even if those things are somewhat esoteric. It's not the time invested, but rather our value judgment of the activity weighed against the invested time. Someone who games moderately doesn't necessarily get tagged; we all have our ways of kicking back and relaxing. Someone who obsessively games probably gets a measure of "get a life" from folks. Activities that aren't perceived as building "real-world" skill or producing anything are generally looked down on (ala ye olde TV-era couch potato).
Um, if it is not profitable, why buy the company and shut it down?
Corporate acquisitions are often done for reasons other than to bring in a new profitable business unit. In some cases, it's to acquire the IP (patents, working proprietary code), the personnel resources (brain-trust, often skilled in a particular domain), or to kill a nascent competitor before it can germinate.
On that theme, it's easy to calculate some reasonable bounds, based off of actual cat metabolism. Small cats, around 7 lbs., will require ~125 kcal/day to maintain body weight. We can use that kcal/day value as a rough bound, which results in a mighty 6W. For the whole cat. Granted, that includes a lot of nap time, but it also includes all other metabolic functions.
Obviously, I have no trouble whatsoever believing that it's possible to do better than 320,000 W in simulating a cat brain. Even padding for sleep, we've clearly got a long, long way to go.
Though you might agree with this particular case, it opens up the door for more intrusive precedents.
This is insane; like arguing that prosecuting graffiti as vandalism will suddenly descend into house painters doing hard time. Exactly the same kind of "I'm too stupid to distinguish good from bad" as school officials occasionally get laughed at when some draconian rule (say, meant to address gang activity) is applied mercilessly to an obviously good student.
As to DDoS attacks, how many times are you allowed to knock on a church's door before it becomes illegal?
You've been struck dumb by your paranoia. There's a qualitative difference between a knocking on a church's door (accessing a website) and trying to knock down a church's door (DDoS'ing it). No legitimate user, even those employing web crawlers, will generate within orders of magnitude of the traffic as a DDoS attack. Even if a non-malicious user somehow "accidentally" DDoS' a site, that person would still be liable for that negligent behavior.
It may be, if it's your employees rallying in complaint with potential backing of legal action. I've had relatives years back have to deal with health issues from smoke in the workplace. This was before the current shift to a predominantly anti-smoking mindset in work environments. This was incredibly stressful for them and took years of pressure from many affected workers to get the problem recognized and to get effective relief in place.
The outside peanut gallery in these situations is amazingly stupid: this isn't "Apple" making this decision as if it were some single vast entity out of a comic book. In the real world, employees dealing with this likely started to complain, bringing up worker safety concerns and uttering phrases containing "OHSA". Execs at the right levels finally decided to back their employees. The execs may have had other motives, but if it were as simple as not wanting to fix smoke-damaged products, they could just as well argue that these products failed due to the owner's negligence and therefore the warranty was void.
This is not to say that this claim is or isn't a legally sustainable one, but its genesis is almost certainly not some deliberately sinister plot to a) not fix things or b) give smoking customers a hard time.
It's pretty obvious that the open source project enables (encourages, even) folks to do whatever they please. The distinction will be similar to the official Chrome browser vs. the Chromium open source project (and forks/descendents). However, the summary misses the context of Google's direction here, primarily that all of the data on a ChromeOS system is intended to be in the cloud:
"If I lose my Chrome OS machine, I should be able to buy a new one, log in, and within a few seconds get all my applications back," Pichai claimed.
It's interesting is to compare this strategy with Apple's recent history. Google will be targeting only specific hardware with ChromeOS, selling an integrated hardware/software appliance instead of just a CD with some bits on it. The OSS project takes up the slack by enabling hackers/homebrewers to do whatever they want, as well as letting other hardware makers to customize and ship ChromeOS to their own platform. Google benefits from all of these angles, since their interest is getting users into their part of the cloud. Since Apple outright sells the hardware/software integration, splitting off the software clearly hasn't presented obvious business value to them.
A law that explicitly disallows merchants to give credit card information to another party would fix this problem.
Here's a way to hack that. First issue: merchants like to process these credit card transactions, too. That means providing this info to another party, the payment processor. I don't mean to Visa, Mastercard, etc. I mean to the companies that comprise the various tiers of third-party payment processor services. This includes some banks, Paypal, and a whole industry of service providers that nearly everyone (even big retailers like Amazon) have to go through in order to process payment transactions (and not just for credit cards). So it turns out that third-parties MUST get this info. So you say, only "payment processors" (insert handwavy legal definition here) can receive such info.
Fine, then what happens when one of these scammer companies becomes (or is bought by, etc.) a payment processor? Back to square one, eh? Think like someone trying to hack/game the system.
If it's possible to craft appropriate legislation that limits the transfer of payment information, that might (or might not) be a good idea. But I'd argue that it's not so much the improper passing of information that's the issue here, it's the (fraudulent, or nearly so) piggybacking of an unauthorized transaction (or worse, subscription) onto an authorized transaction. In the current system, passing of credentials is equivalent to "authorizing" which confuses these two issues. On this basis, it might be more fruitful to apply and/or strengthen fraud laws as these scams are all dependent on confounding the customer to get money instead of transacting legitimate, well-informed business.
Erk. I now know how Star Trek characters feel upon finding themselves in one of those episodes that makes them into old codgers/crones for the next 45 minutes.
Actually, if you'd bothered to RTFA (I know, I know) the author gives concrete examples of where the SWF payload can be embedded in a swath of other media types and still successfully be used as an attack vector. It's a more complex problem than just serving user-generated SWF files.
Counterpoint: They sell Unix-based systems with UI infrastructure and frameworks that kick the tar out of what's available on Windows or Linux, which have actual third-party support for a breadth of (gasp) commercial software, and which has a large, healthy native FOSS community. Oh, and which can run virtually all *nix-based FOSS software I've cared to look as well.
And just for a data point, look up the stats on a Dell Studio 13 compared to the current Macbook. They're nearly identical. Price, disk, RAM are all identical. The Macbook has a.13 GHz "edge" on its clock speed; yawn. So there's no "ego placation" going on by way of price status here.
Aside from the fact that you're a troll, there's a deeper meme here worth debunking: that accessibility features are just for the "impaired".
Gregg Vanderheiden gave the closing plenary talk at the SIGCHI 2001 conference. The subject was how creative integration of accessibility features can greatly improve functionality for all users, including examples of products originally designed for people with impairments which went on to wider commercial success. As an example of this kind of thinking, with portable devices (mobile phones, music players, PDAs) we're all "blind" at some time or another -- we cannot or do not want to redirect our visual attention to the device. So what happens when the normal function of the device includes cues to operation that don't require vision (via audio, haptics, etc.)? The device becomes more useful to everyone, including those with visual impairment. Likewise, by including design elements that work when users can't hear a device that device is more useful to both the hearing impaired and to users in loud environments.
There's a summary of this presentation with more details here: http://www.ariadne.ac.uk/issue28/chi/ Scroll down past the stuff about Bill Gates' opening keynote (which was utterly lame in comparison to Vanderheiden's talk, IMO).
Slashdot Mirror
For me, it makes data semantic-free so that it can be passed between systems, and changes on the transmitting end don't fuck up the receiving end.
XML itself doesn't provide this for free. But what it does do is make the schema a first-class entity against which documents can be validated. This change, and the toolchains that followed, made the programmatic interface presented by the document format much more evident to onlookers. Put another way, the schema and the serialization format were no longer entangled. That, in turn, made it a whole heck of a lot easier for developers (and even some technical business types) to comprehend the patterns that allow backwards-compatible format changes.
Plus, it's readable.
Before someone flames the parent, XML being notoriously human-unfriendly in some circles, compare XML's readability to that of a random poorly-documented (endian-specific!!) binary format spun off by some dude who left the organization six years ago.
XML gets people who have no business doing serialization and parsing out of that business entirely. Instead, we have schemas (either purely ad-hoc, or formalized) and common libraries to express those needs. Like any computing tool, XML has been used for good, bad, and ugly ends... but it brought the idea of schema-based document format and consistent Document Object Model (DOM) APIs to the forefront of data/document handling. In essence, this is a two-level interface: the document schema (a first-class entity, unlike an ad-hoc data format) and the programming APIs.
Even if you're no big fan of XML, the good ideas, patterns, anti-patterns, and mindset changes that spread along with it have positively influenced many other useful document technologies: JSON, YAML, the newer binary serialization formats (such as Protocol Buffers, Thrift, and Avro), and more.
Devil's advocate: how is it DirecTV's fault that she unilaterally did something that made her unable to receive the service she'd paid for?
It's DirecTV's fault because to do otherwise makes them asshats. Life happens, in ways that people often can't control. Any responsible business allows for such in their policies and contracts. For example, a local gym chain that I attended for a time had not only reasonable and responsible account termination rules and practices, but they even let you out of a contract no-fuss if you moved and were no longer within some very modest distance of a franchise location. They got my business over another local gym precisely because of a much better reputation for having an honest and well-defined relationship with their customers.
"You are no longer an optimum energy producing unit. You are being ejected from the Matrix. Goodbye!"
<schlorp, spitooie!>
It's not just photographs. This is a problem when generally when a medium is applied to a primarily technical aim (e.g. breaking a record) vs. an aesthetic one. The best example of this I've witnessed was during my freshman year of college, when a music department Prof. had the class listen to the first public recording of tape loop reverb. IIRC, it came out of MIT. The recording was performed on the recorder (the woodwind instrument) by the then-current department chair.
Now try to imagine sounds that would make Vogons would tremble in simultaneous delight and terror at this, and admit in defeat that their poetry is no equal. I can't recall hearing a brilliant rendition of anything on the recorder. Now combine that lackluster sound, with a /cough/ less than virtuoso performance and a good mangling by those first doozy steps into studio-created reverberation.... bleaaargh!
You have explicit rights (in most cases) to privacy and use of the property that you actually own. That's it, the line is drawn there. I can't believe there is any controversy over this.
Of course there's controversy! In case you haven't noticed, a vast majority of our "personal" data has wandered out onto networks and servers that none of us control. This has been a gradual process going on for years, but the very attributes of modern networked computer systems make the real-world impact of these changes much greater now. This has radically changed the landscape under which the 4th amendment (and a lot of other law) was originally conceived. I'd say that review of the applicability of constitutional protections in modern contexts could be viewed as one of the most important roles of the SCOTUS.
The people who are using nvidia's driver obviously care about 3d performance otherwise they'd already be using the open source driver with 2d support.
Not so much. I have an Ubuntu desktop box with an NVidia card (work system; I didn't spec it), and the open source driver seemed to have a number of issues when running dual head. I flipped to the binary driver, it just worked, and haven't looked back. It sounds like the Nouveau driver will bring some substantial forward progress to open source NVidia drivers. ( No thanks to NVidia themselves. :-P )
If you consider something at "University" a Youth Mistake. Most people are generally at the age of adulthood since then.
If you think someone at University at a typical post-high school age is an "adult", then practical experience, cognitive science, and auto insurer's actuarial statistics have something quite different to say. Even ignoring brain maturation issues, in today's society that's the time when most folks are away from home and on their own for the first time, and are really just starting to figure out Which End Is Up.
Or perhaps MM simply doesn't want to get the plug pulled by a conventional cloud compute provider, due to the questionable PR (and possibly other attention) that this service may
One could view this as an alternative to the old "publish the exploit as a goad to the provider" tactic. Previously, some cryptographic weaknesses required someone to have the resources to obtain a compute cluster large enough to deal with some specific cracking problem. With this approach, it isn't even necessary to be able to set up an EC2 job -- just shell out a few bucks and away-you-go. It'll definitely be a wake-up call for some folks that big compute clusters are trivially available to anyone.
TFA directly addresses that point:
Piper and his colleagues don’t know what the correct amino acid balance might be for humans, and he says it would be a nearly impossible feat to adjust people’s diets to get just the right mix. Instead, the team is investigating how tweaking amino acid content in the diet affects cells. If the researchers can identify pathways affected by amino acid imbalances, they might be able to design drugs or other therapies that could give the benefits of caloric restriction without cutting calories.
Part of the problem was also part of Salon's strength: they were started and run by writers. Old-school, ink-and-paper writers.
And that's a much, much broader problem. Traditional print publishing has been a cutthroat, generally low-margin business for decades. This created other friction, such as top web development and design talent being able to command salaries quite a bit above that of the traditional senior staff. This is one reason why the websites for publishing houses were absolute crap for a long long time -- it was too bitter a pill to hire a bright young twenty something at double the executive salaries.
http://balloon.media.mit.edu/evilwombat/ ;-)
I'm reminded of this quote from the Max Headroom series:
Credit fraud? My God, that's worse than murder!
Reference
Note: if copyright durations weren't so ridiculously long, the above reference link would probably be directly to streaming video of that scene.
In fact it can be said about every living human being who choose to do something different from what you choose to do with your time. "They should get a life."
I don't find that most folks are really that myopic. Those who work towards goals understandable to others don't tend to receive the "getalife" tag. Athleticism is widely valued and understood. This is especially so for Olympic athletes, who tend to receive admiration even from those who eschew mainstream sports. Likewise, many people appreciate those who show skill in building tangible things (e.g. hotrods, etc.), even if those things are somewhat esoteric. It's not the time invested, but rather our value judgment of the activity weighed against the invested time. Someone who games moderately doesn't necessarily get tagged; we all have our ways of kicking back and relaxing. Someone who obsessively games probably gets a measure of "get a life" from folks. Activities that aren't perceived as building "real-world" skill or producing anything are generally looked down on (ala ye olde TV-era couch potato).
Um, if it is not profitable, why buy the company and shut it down?
Corporate acquisitions are often done for reasons other than to bring in a new profitable business unit. In some cases, it's to acquire the IP (patents, working proprietary code), the personnel resources (brain-trust, often skilled in a particular domain), or to kill a nascent competitor before it can germinate.
On that theme, it's easy to calculate some reasonable bounds, based off of actual cat metabolism. Small cats, around 7 lbs., will require ~125 kcal/day to maintain body weight. We can use that kcal/day value as a rough bound, which results in a mighty 6W. For the whole cat. Granted, that includes a lot of nap time, but it also includes all other metabolic functions.
Obviously, I have no trouble whatsoever believing that it's possible to do better than 320,000 W in simulating a cat brain. Even padding for sleep, we've clearly got a long, long way to go.
Though you might agree with this particular case, it opens up the door for more intrusive precedents.
This is insane; like arguing that prosecuting graffiti as vandalism will suddenly descend into house painters doing hard time. Exactly the same kind of "I'm too stupid to distinguish good from bad" as school officials occasionally get laughed at when some draconian rule (say, meant to address gang activity) is applied mercilessly to an obviously good student.
As to DDoS attacks, how many times are you allowed to knock on a church's door before it becomes illegal?
You've been struck dumb by your paranoia. There's a qualitative difference between a knocking on a church's door (accessing a website) and trying to knock down a church's door (DDoS'ing it). No legitimate user, even those employing web crawlers, will generate within orders of magnitude of the traffic as a DDoS attack. Even if a non-malicious user somehow "accidentally" DDoS' a site, that person would still be liable for that negligent behavior.
It may be, if it's your employees rallying in complaint with potential backing of legal action. I've had relatives years back have to deal with health issues from smoke in the workplace. This was before the current shift to a predominantly anti-smoking mindset in work environments. This was incredibly stressful for them and took years of pressure from many affected workers to get the problem recognized and to get effective relief in place.
The outside peanut gallery in these situations is amazingly stupid: this isn't "Apple" making this decision as if it were some single vast entity out of a comic book. In the real world, employees dealing with this likely started to complain, bringing up worker safety concerns and uttering phrases containing "OHSA". Execs at the right levels finally decided to back their employees. The execs may have had other motives, but if it were as simple as not wanting to fix smoke-damaged products, they could just as well argue that these products failed due to the owner's negligence and therefore the warranty was void.
This is not to say that this claim is or isn't a legally sustainable one, but its genesis is almost certainly not some deliberately sinister plot to a) not fix things or b) give smoking customers a hard time.
Don't worry. When disaster strikes, at least you'll be in contention for a Darwin Award. ;-)
It's pretty obvious that the open source project enables (encourages, even) folks to do whatever they please. The distinction will be similar to the official Chrome browser vs. the Chromium open source project (and forks/descendents). However, the summary misses the context of Google's direction here, primarily that all of the data on a ChromeOS system is intended to be in the cloud:
"If I lose my Chrome OS machine, I should be able to buy a new one, log in, and within a few seconds get all my applications back," Pichai claimed.
It's interesting is to compare this strategy with Apple's recent history. Google will be targeting only specific hardware with ChromeOS, selling an integrated hardware/software appliance instead of just a CD with some bits on it. The OSS project takes up the slack by enabling hackers/homebrewers to do whatever they want, as well as letting other hardware makers to customize and ship ChromeOS to their own platform. Google benefits from all of these angles, since their interest is getting users into their part of the cloud. Since Apple outright sells the hardware/software integration, splitting off the software clearly hasn't presented obvious business value to them.
A law that explicitly disallows merchants to give credit card information to another party would fix this problem.
Here's a way to hack that. First issue: merchants like to process these credit card transactions, too. That means providing this info to another party, the payment processor. I don't mean to Visa, Mastercard, etc. I mean to the companies that comprise the various tiers of third-party payment processor services. This includes some banks, Paypal, and a whole industry of service providers that nearly everyone (even big retailers like Amazon) have to go through in order to process payment transactions (and not just for credit cards). So it turns out that third-parties MUST get this info. So you say, only "payment processors" (insert handwavy legal definition here) can receive such info.
Fine, then what happens when one of these scammer companies becomes (or is bought by, etc.) a payment processor? Back to square one, eh? Think like someone trying to hack/game the system.
If it's possible to craft appropriate legislation that limits the transfer of payment information, that might (or might not) be a good idea. But I'd argue that it's not so much the improper passing of information that's the issue here, it's the (fraudulent, or nearly so) piggybacking of an unauthorized transaction (or worse, subscription) onto an authorized transaction. In the current system, passing of credentials is equivalent to "authorizing" which confuses these two issues. On this basis, it might be more fruitful to apply and/or strengthen fraud laws as these scams are all dependent on confounding the customer to get money instead of transacting legitimate, well-informed business.
Erk. I now know how Star Trek characters feel upon finding themselves in one of those episodes that makes them into old codgers/crones for the next 45 minutes.
Actually, if you'd bothered to RTFA (I know, I know) the author gives concrete examples of where the SWF payload can be embedded in a swath of other media types and still successfully be used as an attack vector. It's a more complex problem than just serving user-generated SWF files.
They sell placation of the buyers ego.
Counterpoint: They sell Unix-based systems with UI infrastructure and frameworks that kick the tar out of what's available on Windows or Linux, which have actual third-party support for a breadth of (gasp) commercial software, and which has a large, healthy native FOSS community. Oh, and which can run virtually all *nix-based FOSS software I've cared to look as well.
And just for a data point, look up the stats on a Dell Studio 13 compared to the current Macbook. They're nearly identical. Price, disk, RAM are all identical. The Macbook has a .13 GHz "edge" on its clock speed; yawn. So there's no "ego placation" going on by way of price status here.
Aside from the fact that you're a troll, there's a deeper meme here worth debunking: that accessibility features are just for the "impaired".
Gregg Vanderheiden gave the closing plenary talk at the SIGCHI 2001 conference. The subject was how creative integration of accessibility features can greatly improve functionality for all users, including examples of products originally designed for people with impairments which went on to wider commercial success. As an example of this kind of thinking, with portable devices (mobile phones, music players, PDAs) we're all "blind" at some time or another -- we cannot or do not want to redirect our visual attention to the device. So what happens when the normal function of the device includes cues to operation that don't require vision (via audio, haptics, etc.)? The device becomes more useful to everyone, including those with visual impairment. Likewise, by including design elements that work when users can't hear a device that device is more useful to both the hearing impaired and to users in loud environments.
There's a summary of this presentation with more details here: http://www.ariadne.ac.uk/issue28/chi/ Scroll down past the stuff about Bill Gates' opening keynote (which was utterly lame in comparison to Vanderheiden's talk, IMO).