Getting your own pilot's license is a bit of work but easily do-able on your average geek's salary. Then go in on a Cessna with a few friends or join a flying rental club and you've got something that can do the shorter hops easily. It won't be cheaper, but it's not as insanely expensive as most think, and no one will search you or even ask you where you're going (unless you fly through class B or C airspace, and then only in general terms).
Alternately, in a couple years the Very Light Jet (VLJ) market is supposed to take off and offer the kind of services you suggest on a level that an upper-middle-class American can afford, but not yet. Watch Eclipse, Honda, and the others roll out their aircraft and look for the small carriers to use'em.
That's because you don't give a shit about the long term. Other people do. The comment was directed at them. Deal with it.
Stop flaming and trolling. Of course I care about the long term. This code sharing has still not been proven to be the best long-term solution. If it can't produce better products, in which "better" isn't a self-referential value assigned to "open source", then caring about the long term has nothing to do with it. At that point it just becomes a religious argument, analogous to brand loyalty or buying things because you like the color of the box.
In my opinion, the long-term is best served by putting money on the technologies that produce the best products. Generally, that's easy-- buy the better products. But if the word "better" gets co-opted to mean something other than direct, measurable improvements, then you lose all qualitative measure of what's leading to the best solution.
In short, I'm not going to buy a worse solution just because it's open source. If open source produces a better product, it will get my money and will flourish. If not, it won't. Either way, the long term is best served.
it's also better because unlike Transgaming, Codeweavers contributes back to WINE.
This must be some metric of "better" that I, as a software consumer, am unfamiliar. I've heard of better performance, better user experience, better return on investment... but "better because some developer I don't know helps out some other developer I don't know" does not trump the others in my book. As long as everyone's following the rules and licenses they acquired their code under, which they are, this really won't be an issue to Mac consumers. So really, I see this as one case in which open source is being judged along with the products... if sharing with WINE makes Codeweavers a technically better product, then your point may be valid-- if not, it's evidence that open source is not working in this case. I prefer to follow the evidence, not the religion, and not pre-judge the situation.
Besides, I *VASTLY* prefer Transgaming's approach to the market. I want to see Mac ports of games, with Mac installers, Mac support, and possibly even a mention of minimum Mac requirements, all of which Transgaming is implying they'll provide. I do NOT want some compatibility box I can buy that lets me run Windows retail boxed games if I can get them installed and translate their requirements to Mac models and operating systems. Considering how easy Transgaming claims they're going to make things, if a company is not willing to put in the little extra expense, I'll find one who is.
Then why are there two separate versions of XP, one for 64-bit and one for 32-bit? Why couldn't Microsoft do what Leopard has apparently done in that respect?
So... 64-bits. All the other x86 operating systems out there seemed to have a large hiccup during the x86-64 transition. Apple claims Leopard will run 32 and 64 bits side-by-side, top-to-bottom in one OS that supports everything. Maybe I missed something. Why was this so hard to do with XP and Linux? Did Apple do something exceptionally clever? Is it their lack of required legacy support? How did they pull this off? Or was this not as surprising or significant an announcement as it seemed to me?
It's just a "delete doesn't really mean delete" feature. Everybody's done that. I think email was the first thing to do it, but I'm not really sure.
Actually, the Trash Can in the Lisa/Mac Finder in 1983/4 was the first time I'd seen an easily recoverable ubiquitous delete feature built into the UI (along with ubiquitous "undo" in all applications on the platform). But that's a very, very simplistic view of what Time Machine is... Time Machine is a historical record that can be directly integrated. So you can do queries in time as well as space, and scroll back to arbitrary points in time in the past within each application's UI. Go back in time in iPhoto, Address Book, etc. And have the application actually operating on the historical data, so the user interface is active on the past data WITHOUT restoring it... just to browse. Name a UI on another platform that can do that.
Anyway, I'm going to stop going on about it. Just please, don't reply to this thread or claim to understand the feature until you've actually looked at the video of it in operation. I think the hardest thing for Apple here is trying to explain what Time Machine is to people with preconceived notions about what the state-of-the-art is in this field.
But I don't know why everyone's so focused on Spaces. Yes, it's a great implementation of an old concept, but it's hardly the most significant feature announced in 10.5. That would have to go to the insanely innovative Time Machine.
Most Slashdot posters completely missed the point with Time Machine. Watch the video on Apple's site (or the WWDC keynote) to see... but a basic use case of what's cool: 1. Open Address Book and search for a person 2. Note that the person doesn't exist, but you knew you had them around at some point 3. Click the "Time Machine" icon... 4. Now Address Book appears in the Time Machine view, with the query still live 5. Click the "Back" arrow... and Time Machine zips back in time to a point at which the query returns something 6. Click on the record then the Restore button, and everything snaps back to the current, with the record now appearing in Address Book. No file system, calendars, or even leaving the current app involved, and the data was still directly selectable from within the current app's UI in the historical version.
This is something that hasn't been done by anyone, and isn't really comparable to Windows' new restore feature. Doing live queries through time? All while staying in your currently open app's UI? And having the historical data directly manipulable in the application's UI? This is really innovative stuff, and I don't think it got enough love in the Slashdot forums yesterday.
The most interesting of these to me is the iCal server. This looks like it could become the best open-source competition to Microsoft Exchange. The Leopard version detailed here looks like a pretty compelling product to have as an Apache-licensed piece of code, and I could see the code getting merged into a lot of other products.
outside the USA, the future will be Linux - China, India, Brazil, Eastern Europe, and other places with low costs of living and an educated population are going to power the world's economy, and I don't see the rest of the world paying the Microsoft tax.
I agree for the near-term, but probably not for the reason you think. I think areas where people's time aren't as valuable and there are more unemployed people around to do "grunt" technical work, Linux makes a huge amount of sense. In countries where people's time are more valuable, easier systems like Windows and MacOS make a lot more sense. (My old motto was "Linux is only 'free' if your time is worthless.") This isn't meant as a flame, but it's hard to argue that Linux is simpler or more productive than MacOS for most people. So it ends up being a return-on-investment proposition, with all time spent from installation to compatibility resolutions to upkeep and updates on the cost side of the equation. In the countries you mention, it's almost certainly better to hire someone cheap to do the legwork and save the money you'd have spent on a commercial license. In the long run I don't think that will be true, and it's almost certainly not true now in the United States.
Yes, and the ability to copy a file to another HD every time you change it, name it something distinctive, then go look for it later when you need it has been around since the beginning.
The point is making it REALLY usable. Being able to visually scroll back in time through your auto-configuring backup system is a really innovative implementation of an old feature.
Certainly there's room for abuse here, but Apple has more latitude here, not being a heavily business-oriented machine. For home use it's fantastic. And tying it to the online/live backup system is nice. (Hook up a new external/internal drive, tell MacOS X to use it for backups, and that's it. Now you're backing up all your data. AND you have a snazzy GUI to slide back in time, grab something/everything from the past, and bring it to the main HD.)
They still have time to add a way to turn this off for a given domain.
I could also see Spotlight's newfound ability to search other machines on your local subnet (if the user has turned Spotlight sharing on) as easily abusable in a corporate/school environment but indispensible in a home environment.
All actual evidence appears to contradict you. Do a survey of the countries with the best IP laws (the ones that protect the inventor's ability to profit from his/her invention) and compare with the countries that have the most relaxed IP laws. I think you'll find that economies that reward innovation (ie. have a strong patent system) tend to do much, much better economically. If someone has a novel invention and the means to patent and license it, I think the system should reward that. In addition, I fear a situation with no patents in which all innovation is bottled in silos with no one having any incentive to publish their invention for public use.
My biggest problem with this change is the trolls who are going to sit around looking for other people to invent things and not have the resources to patent it, then get in there first with a patent application. Isn't this going to mean that all open source inventions are going to have to immediately start patenting any innovation so someone else doesn't take it and patent it first?
The only thing I dislike about Cedega is the licensing issue - ultimately, that's why I stopped using it.
It sounds like Cedega would do well to just drop linux support where people feel somehow "cheated" because a company took a BSD-licensed product, followed the licensing terms, and produced a great product (oh, the humanity!). Mac users seem to be more worried about just paying a little more for something that works, so they'll probably do all right in that arena.
So score one for HD DVD's VC1 compression codec over the MPEG2/AVC scheme used for Blu-ray
Both Blu-Ray and HD DVD support both VC1 and MPEG2/AVC, if I'm not mistaken. They are comparing the encoding on a specific movie, but imply that it's inherent to the format.
Unfortunately, due to disc space limitations, Warner has elected to drop the [TrueHD Dolby Digital] track altogether on the Blu-ray release. [...] But more troubling is that Warner has also dropped the Dolby Digital-Plus track off of this Blu-ray release
Disc space limitations on Blu-Ray? 25GB on a single layer is really not enough (compared to 30 on a dual-layer with HD DVD) that two audio tracks had to be dropped? Something is fishy here.
Despite this, it has become the most popular FOSS license by far. Think about it.
Dear Anonymous Coward,
GPLv2 also struck a good balance between restricting a developer's rights to distribute software how they saw fit and use it in commercially viable devices, and the requirement to always reciprocate the usage of the source code. GPLv3 makes no such balance-- it is an all-out war against corporate software. Thus my comment that GPLv3 will become, rather than a successor to GPLv2, "just another open source license" of which we have over a dozen already. At least it will separate out the socialist/communists (GPLv3) from the drive-out-extra-costs-from-the-supply-chain capitalists (GPLv2).
They do it all the time, and always have (e.g. the Clinton haircut at LAX uproar).
I think the difference here is that the Nashville event actually happened. The Clinton haircut "delays" were pretty much an invention of the right-wing media.
Hezbollah, which is part of the Lebanese government, launched crossborder raids and fairly advanced rockets on civilian targets many miles inside Israel.
And Israel has done the same to Lebanon. Remember, Hezbollah only EXISTS because of Israel's invasion of Lebanon decades ago. They reap what they sow.
Except for the fact that the air and water are cleaner than they were 50 years ago, and keep getting cleaner. Older, less "horrible" animal farming techniques required cooking meat nearly crunchy just to make sure you didn't get trichinosis and other diseases that more "humanely" treated animals always got.
Firstly, I said water and soil. The air is definitely cleaner because we replaced a lot of coal with natural gas, cleaned up a lot of power plants, and destroyed our steel industry and sent it overseas. The water, though, is getting some pretty strange and complex dissolved chemicals these days, from mercury to MBTE to newer fertilizers, etc. A lot of that is getting into the fish, which used to be one of the healthiest things you could possibly eat. Now if you eat more than a couple servings a week you risk all sorts of poisonings. And the soil outside the Superfund sites is probably a lot worse off than it was 50 years ago as the runoff from new mining techniques and the aforementioned chemicals permeate it.
And over the last 50 years the United States made the transition from small cattle ranches to corporate ranches, and most cattle moved from grazing to corn-feed with growth hormones. Thus, the beef got a lot cheaper and a lot worse for you.
50 years ago puts you at 1956, baby boomers have been born and already half-way thru grade school. Hell, in another 13 years man walks on the moon.
You'd be surprised how many people didn't have electric refrigerators until after this date, and how salted the meats tended to still be. Many, many baby boomers that lived outside the cities got their own phone lines and electric refrigeration in the late 50's and 60's as greater government involvement in the process generally raised the country's base standard of living. Add to this the momentum of what people tend to eat, and the salt content didn't really start decreasing until much more recently.
Also: salt. Despite what you hear about salted fries and such, the average human eats VASTLY less salt today than they did 50 or 100 years ago, when salting meat was the primary means of preservation. Today virtually every house has great refrigeration, the foods have better preservatives, and people have an awareness of the dangers of salt on the cardiovascular system.
That being said, the water and soil pollution, horrible animal farming techniques, and a lack of any new antibiotics or other non-deathbed "wellness" medicine over the past 50 years probably argues in the grandparent-poster's favor.
Voice recognition requires some training regardless of who provides it. We're not Star Trek here...
Not necessarily. Even in the early 90's Carnegie Mellon's sphinx2 project did a fairly reasonable job on a completely random sample of people, untrained. With modern semantic parsers and such, it should be even better. This isn't an insurmountable problem, it's just a matter of how much RAM and CPU you want to throw at the problem. In that respect, Dragon probably has an advantage, as a built-in OS feature probably can't suck up 100+MB off the top in an always-resident manner. (Incidentally, no computer or human will ever be 100% accurate at speech recognition, but the "edit" commands should probably be a LOT higher on the priority list for recognition than they appear to be in this demo.)
Incidentally, the sphinx2 project tended to use early Alphas (with the beta of OSF/1) as servers, and NeXT workstations as clients. I think a lot of those folks went to either Apple or IBM, then later Dragon.
Mathematics doesn't really exist, though. There is no such thing as "proof" in science, just hypotheses, theories, observations, conclusions, and consensuses (what's the plural of consensus, anyway?). This observation may well have muddied the consensus that had previously grown around the black hole theory, but I'm sure it hasn't convinced most yet.
Slashdot Mirror
Getting your own pilot's license is a bit of work but easily do-able on your average geek's salary. Then go in on a Cessna with a few friends or join a flying rental club and you've got something that can do the shorter hops easily. It won't be cheaper, but it's not as insanely expensive as most think, and no one will search you or even ask you where you're going (unless you fly through class B or C airspace, and then only in general terms).
Alternately, in a couple years the Very Light Jet (VLJ) market is supposed to take off and offer the kind of services you suggest on a level that an upper-middle-class American can afford, but not yet. Watch Eclipse, Honda, and the others roll out their aircraft and look for the small carriers to use'em.
That's because you don't give a shit about the long term. Other people do. The comment was directed at them. Deal with it.
Stop flaming and trolling. Of course I care about the long term. This code sharing has still not been proven to be the best long-term solution. If it can't produce better products, in which "better" isn't a self-referential value assigned to "open source", then caring about the long term has nothing to do with it. At that point it just becomes a religious argument, analogous to brand loyalty or buying things because you like the color of the box.
In my opinion, the long-term is best served by putting money on the technologies that produce the best products. Generally, that's easy-- buy the better products. But if the word "better" gets co-opted to mean something other than direct, measurable improvements, then you lose all qualitative measure of what's leading to the best solution.
In short, I'm not going to buy a worse solution just because it's open source. If open source produces a better product, it will get my money and will flourish. If not, it won't. Either way, the long term is best served.
it's also better because unlike Transgaming, Codeweavers contributes back to WINE.
This must be some metric of "better" that I, as a software consumer, am unfamiliar. I've heard of better performance, better user experience, better return on investment... but "better because some developer I don't know helps out some other developer I don't know" does not trump the others in my book. As long as everyone's following the rules and licenses they acquired their code under, which they are, this really won't be an issue to Mac consumers. So really, I see this as one case in which open source is being judged along with the products... if sharing with WINE makes Codeweavers a technically better product, then your point may be valid-- if not, it's evidence that open source is not working in this case. I prefer to follow the evidence, not the religion, and not pre-judge the situation.
Besides, I *VASTLY* prefer Transgaming's approach to the market. I want to see Mac ports of games, with Mac installers, Mac support, and possibly even a mention of minimum Mac requirements, all of which Transgaming is implying they'll provide. I do NOT want some compatibility box I can buy that lets me run Windows retail boxed games if I can get them installed and translate their requirements to Mac models and operating systems. Considering how easy Transgaming claims they're going to make things, if a company is not willing to put in the little extra expense, I'll find one who is.
I meant domain in the more network/Windows sense. ie. An administrator able to disable it for everyone and enforce the policy.
Then why are there two separate versions of XP, one for 64-bit and one for 32-bit? Why couldn't Microsoft do what Leopard has apparently done in that respect?
So... 64-bits. All the other x86 operating systems out there seemed to have a large hiccup during the x86-64 transition. Apple claims Leopard will run 32 and 64 bits side-by-side, top-to-bottom in one OS that supports everything. Maybe I missed something. Why was this so hard to do with XP and Linux? Did Apple do something exceptionally clever? Is it their lack of required legacy support? How did they pull this off? Or was this not as surprising or significant an announcement as it seemed to me?
It's just a "delete doesn't really mean delete" feature. Everybody's done that. I think email was the first thing to do it, but I'm not really sure.
Actually, the Trash Can in the Lisa/Mac Finder in 1983/4 was the first time I'd seen an easily recoverable ubiquitous delete feature built into the UI (along with ubiquitous "undo" in all applications on the platform). But that's a very, very simplistic view of what Time Machine is... Time Machine is a historical record that can be directly integrated. So you can do queries in time as well as space, and scroll back to arbitrary points in time in the past within each application's UI. Go back in time in iPhoto, Address Book, etc. And have the application actually operating on the historical data, so the user interface is active on the past data WITHOUT restoring it... just to browse. Name a UI on another platform that can do that.
Anyway, I'm going to stop going on about it. Just please, don't reply to this thread or claim to understand the feature until you've actually looked at the video of it in operation. I think the hardest thing for Apple here is trying to explain what Time Machine is to people with preconceived notions about what the state-of-the-art is in this field.
Firstly, a huge Leopard preview site is up at Apple's "sneak preview" site
But I don't know why everyone's so focused on Spaces. Yes, it's a great implementation of an old concept, but it's hardly the most significant feature announced in 10.5. That would have to go to the insanely innovative Time Machine.
Most Slashdot posters completely missed the point with Time Machine. Watch the video on Apple's site (or the WWDC keynote) to see... but a basic use case of what's cool:
1. Open Address Book and search for a person
2. Note that the person doesn't exist, but you knew you had them around at some point
3. Click the "Time Machine" icon...
4. Now Address Book appears in the Time Machine view, with the query still live
5. Click the "Back" arrow... and Time Machine zips back in time to a point at which the query returns something
6. Click on the record then the Restore button, and everything snaps back to the current, with the record now appearing in Address Book. No file system, calendars, or even leaving the current app involved, and the data was still directly selectable from within the current app's UI in the historical version.
This is something that hasn't been done by anyone, and isn't really comparable to Windows' new restore feature. Doing live queries through time? All while staying in your currently open app's UI? And having the historical data directly manipulable in the application's UI? This is really innovative stuff, and I don't think it got enough love in the Slashdot forums yesterday.
The most interesting of these to me is the iCal server. This looks like it could become the best open-source competition to Microsoft Exchange. The Leopard version detailed here looks like a pretty compelling product to have as an Apache-licensed piece of code, and I could see the code getting merged into a lot of other products.
outside the USA, the future will be Linux - China, India, Brazil, Eastern Europe, and other places with low costs of living and an educated population are going to power the world's economy, and I don't see the rest of the world paying the Microsoft tax.
I agree for the near-term, but probably not for the reason you think. I think areas where people's time aren't as valuable and there are more unemployed people around to do "grunt" technical work, Linux makes a huge amount of sense. In countries where people's time are more valuable, easier systems like Windows and MacOS make a lot more sense. (My old motto was "Linux is only 'free' if your time is worthless.") This isn't meant as a flame, but it's hard to argue that Linux is simpler or more productive than MacOS for most people. So it ends up being a return-on-investment proposition, with all time spent from installation to compatibility resolutions to upkeep and updates on the cost side of the equation. In the countries you mention, it's almost certainly better to hire someone cheap to do the legwork and save the money you'd have spent on a commercial license. In the long run I don't think that will be true, and it's almost certainly not true now in the United States.
Yes, and the ability to copy a file to another HD every time you change it, name it something distinctive, then go look for it later when you need it has been around since the beginning.
The point is making it REALLY usable. Being able to visually scroll back in time through your auto-configuring backup system is a really innovative implementation of an old feature.
You got a link for that particular feature
Link: http://www.apple.com/macosx/leopard/index.html
(It's advertised with a graphic on Apple.com's main page)
All the features mentioned in the keynote are described there, along with little animations.
The new text-to-speech voice also has a playable sound file there, and yes, it's very impressive.
Certainly there's room for abuse here, but Apple has more latitude here, not being a heavily business-oriented machine. For home use it's fantastic. And tying it to the online/live backup system is nice. (Hook up a new external/internal drive, tell MacOS X to use it for backups, and that's it. Now you're backing up all your data. AND you have a snazzy GUI to slide back in time, grab something/everything from the past, and bring it to the main HD.)
They still have time to add a way to turn this off for a given domain.
I could also see Spotlight's newfound ability to search other machines on your local subnet (if the user has turned Spotlight sharing on) as easily abusable in a corporate/school environment but indispensible in a home environment.
Patents don't work. It's that simple.
All actual evidence appears to contradict you. Do a survey of the countries with the best IP laws (the ones that protect the inventor's ability to profit from his/her invention) and compare with the countries that have the most relaxed IP laws. I think you'll find that economies that reward innovation (ie. have a strong patent system) tend to do much, much better economically. If someone has a novel invention and the means to patent and license it, I think the system should reward that. In addition, I fear a situation with no patents in which all innovation is bottled in silos with no one having any incentive to publish their invention for public use.
My biggest problem with this change is the trolls who are going to sit around looking for other people to invent things and not have the resources to patent it, then get in there first with a patent application. Isn't this going to mean that all open source inventions are going to have to immediately start patenting any innovation so someone else doesn't take it and patent it first?
The only thing I dislike about Cedega is the licensing issue - ultimately, that's why I stopped using it.
It sounds like Cedega would do well to just drop linux support where people feel somehow "cheated" because a company took a BSD-licensed product, followed the licensing terms, and produced a great product (oh, the humanity!). Mac users seem to be more worried about just paying a little more for something that works, so they'll probably do all right in that arena.
So score one for HD DVD's VC1 compression codec over the MPEG2/AVC scheme used for Blu-ray
Both Blu-Ray and HD DVD support both VC1 and MPEG2/AVC, if I'm not mistaken. They are comparing the encoding on a specific movie, but imply that it's inherent to the format.
Unfortunately, due to disc space limitations, Warner has elected to drop the [TrueHD Dolby Digital] track altogether on the Blu-ray release. [...] But more troubling is that Warner has also dropped the Dolby Digital-Plus track off of this Blu-ray release
Disc space limitations on Blu-Ray? 25GB on a single layer is really not enough (compared to 30 on a dual-layer with HD DVD) that two audio tracks had to be dropped? Something is fishy here.
Despite this, it has become the most popular FOSS license by far. Think about it.
Dear Anonymous Coward,
GPLv2 also struck a good balance between restricting a developer's rights to distribute software how they saw fit and use it in commercially viable devices, and the requirement to always reciprocate the usage of the source code. GPLv3 makes no such balance-- it is an all-out war against corporate software. Thus my comment that GPLv3 will become, rather than a successor to GPLv2, "just another open source license" of which we have over a dozen already. At least it will separate out the socialist/communists (GPLv3) from the drive-out-extra-costs-from-the-supply-chain capitalists (GPLv2).
They do it all the time, and always have (e.g. the Clinton haircut at LAX uproar).
I think the difference here is that the Nashville event actually happened. The Clinton haircut "delays" were pretty much an invention of the right-wing media.
Hezbollah, which is part of the Lebanese government, launched crossborder raids and fairly advanced rockets on civilian targets many miles inside Israel.
And Israel has done the same to Lebanon. Remember, Hezbollah only EXISTS because of Israel's invasion of Lebanon decades ago. They reap what they sow.
So you base purchasing decisions on marketing campaigns instead of your estimation of how useful/fun/productive something will be?
So Apple marketing has a few putzes. Their computers and software are still pretty sweet.
Except for the fact that the air and water are cleaner than they were 50 years ago, and keep getting cleaner. Older, less "horrible" animal farming techniques required cooking meat nearly crunchy just to make sure you didn't get trichinosis and other diseases that more "humanely" treated animals always got.
Firstly, I said water and soil. The air is definitely cleaner because we replaced a lot of coal with natural gas, cleaned up a lot of power plants, and destroyed our steel industry and sent it overseas. The water, though, is getting some pretty strange and complex dissolved chemicals these days, from mercury to MBTE to newer fertilizers, etc. A lot of that is getting into the fish, which used to be one of the healthiest things you could possibly eat. Now if you eat more than a couple servings a week you risk all sorts of poisonings. And the soil outside the Superfund sites is probably a lot worse off than it was 50 years ago as the runoff from new mining techniques and the aforementioned chemicals permeate it.
And over the last 50 years the United States made the transition from small cattle ranches to corporate ranches, and most cattle moved from grazing to corn-feed with growth hormones. Thus, the beef got a lot cheaper and a lot worse for you.
50 years ago puts you at 1956, baby boomers have been born and already half-way thru grade school. Hell, in another 13 years man walks on the moon.
You'd be surprised how many people didn't have electric refrigerators until after this date, and how salted the meats tended to still be. Many, many baby boomers that lived outside the cities got their own phone lines and electric refrigeration in the late 50's and 60's as greater government involvement in the process generally raised the country's base standard of living. Add to this the momentum of what people tend to eat, and the salt content didn't really start decreasing until much more recently.
Also: salt. Despite what you hear about salted fries and such, the average human eats VASTLY less salt today than they did 50 or 100 years ago, when salting meat was the primary means of preservation. Today virtually every house has great refrigeration, the foods have better preservatives, and people have an awareness of the dangers of salt on the cardiovascular system.
That being said, the water and soil pollution, horrible animal farming techniques, and a lack of any new antibiotics or other non-deathbed "wellness" medicine over the past 50 years probably argues in the grandparent-poster's favor.
Voice recognition requires some training regardless of who provides it. We're not Star Trek here...
Not necessarily. Even in the early 90's Carnegie Mellon's sphinx2 project did a fairly reasonable job on a completely random sample of people, untrained. With modern semantic parsers and such, it should be even better. This isn't an insurmountable problem, it's just a matter of how much RAM and CPU you want to throw at the problem. In that respect, Dragon probably has an advantage, as a built-in OS feature probably can't suck up 100+MB off the top in an always-resident manner. (Incidentally, no computer or human will ever be 100% accurate at speech recognition, but the "edit" commands should probably be a LOT higher on the priority list for recognition than they appear to be in this demo.)
Incidentally, the sphinx2 project tended to use early Alphas (with the beta of OSF/1) as servers, and NeXT workstations as clients. I think a lot of those folks went to either Apple or IBM, then later Dragon.
Mathematics doesn't really exist, though. There is no such thing as "proof" in science, just hypotheses, theories, observations, conclusions, and consensuses (what's the plural of consensus, anyway?). This observation may well have muddied the consensus that had previously grown around the black hole theory, but I'm sure it hasn't convinced most yet.