Its very interesting all the different mice like systems out there.
We have various forms of the traditional mouse. We are all pretty familar with its capabilities.
The trackball has the same set of abilities, plus the spin ability that mice lack. I'm not sure of many uses for just spining the ball fast and letting it roll to a stop, but I;d be surprised if that is not useful in some manner. In theory, a trackball can also allow yaw control, although I've not seen it done, and it could be awkward to use. This is a stationary device.
We have the touch pad. Its only real use advantage over a mouse is the fact that it is stationary. It has a manufacturing benefit of taking up less room than most other stationary input schemes.
We have the track point. This has the fairly unique ability to allow one to apply a constant velocity to the mouse cursor. I know of one game in particular where that would be very useful, since a specific constant velocity would cancel out the gravity in the parts of the game that use it, returning normal control to those levels. It is also stationary, but is limited. There is no scroll wheel style support that the others devices have.
Next we have absoluter positioning devices.
One of the most familiar is the standard touch screen. This like other absolute positioning devices lets the cursor teleport from one location to another. This is rather nice for a few stupid flash games, like the get through the maze without touching the lines games. The downside of these is that they generally don't allow mouse positioning without simultaneously clicking.
Next we have the Wacom style digitizer. This not only allows for cursor movement without clicking, but also normally provides pressure data, and in some cases can provide angle relative to the "paper" . (Used when simulating the use of caligraphy writing instruments where line width is based on angle, among others). This is also the input device that best allows for writing and drawing. Anybody who has ever tried to write with a mouse or trackpad knows how terrible the results come out, while digitizers of sufficient resolution should result in writing almost instinguishable from that done on paper.
This is just as good a place as any to ask about the support of templates in this device. Important components of some articles are generate by templates. One example is the infamous [citation needed] text, which is generated by the "{{cn}}" template. Other times, important words in a sentence are used as a argument for a template, to produce some from of link automatically.
Some other mobile Wikipedia solutions, such as one I saw for the iPhone, just ignore templates. That means that important words in a sentence could potentially be omitted. In some cases, entire sections of an article may be omitted. I consider that extremely problematic.
Assuming they are properly supported, are references supported? In few articles I've seen the feature used for general footnotes in addition to references, and having those disappear could be problematic.
What about the LaTeX math equations? A lot of mathematical and Computer Science articles become completely worthless if those are omitted, but including them means included quite a few generated images for some of the more complex ones that cannot be rendered as html.
And what about the ez-timeline extention. Are the images that it generates included?
What about the hieroglyphics that articles may include by way of the wikihiro extention?
I would not be willing to use a static mobile Wikipedia that did not support templates, references, tables, external links, LaTeX equations.
Other people might insist that the categories pages be included, and that time lines and hieroglyphics be supported.
Correction only Windows Perfessional upgrade is available through the Campus agreement, not Ultimate. A bit surprising, since a campus would probably desire at least Windows Enterpise on the lab computers.
If they are giving students MSDNAA copies of Windows 7, they had better be alerting the students that they may not use it for any commerical endevor, even side jobs, If they are not in Science, Technology, Engineering, or Mathmatics department, or using the computer for "Visualization, Illustration, Design, [or] Art", they may not install the OS at all. Upon graduation, they must uninstall Windows 7.
I think it is more likely that they have a Microsoft Campus Agreement. The agreement's student option allows the school yo pay some amount per full time equivlent student, and allows the school to provide copies of the software for the student's personal machines. All graduating students get a perpetual license to use the software. Everybody else must remove it upon leaving the school. The license is still for non-commercial use only, but it does apply to all majors.
My school uses this program to provide copies of MS Office Enterpise for $50 to students. A permenate non-commecial MS Office enterprise license is still worth a hell of a lot more than $50, so this is a great deal for students. I'm not sure what they charge for Windows, but it will still not be a bad deal. I'm guessing it is the ultimate edition that is available. This program though only provides upgrade versions of Windows, not full versions, and I'm betting the discs provided have the retail upgrade-only image.
When I see a website that accepts no advertising, buys all it's products for cash, anonymously from retail stores and has a test suite that reflects what actual users actually do, then their reviews will have merit (although I can't see anyone anywhere paying the hue cover price for such a publication)
Except perhaps the part about how well the test suite reflects what users do, You have just described the periodical "Consumer Reports". No advertising, buys products anonymously at retail. People do pay cover price for it, although subscription price is far more common. The biggest problem is they cover everything, not just technology, so the level of technology coverage leaves a lot to be desired.
And reviewers almost always use systems that had specs well beyond the vista absolute minimum specs, while many users who had issues were using a one of the cheepest machines on the market at the time, which were not exceeding the absolute minimum specifications by much. Vista RTM definitely was far more responsive, etc on a better spec'ed machine.
Correct. C++0x will have native support for threads, based largely on Boost threading libraries, but with some differences, and the potential for better support of some features, by adding support to the language core.
Rapidshare would not the right right site for this. Legitimate content would be downloadbale only through things like the iTunes store.
The other alternative is streaming sites. Threre are three models for streaming sites.
Streaming with ads. Most of the legal online streaming is of this sort. Simply goto Hulu.com and search for the show you want. If hulu has it will show up. If hulu does not have it, but the network's own website has it, it will give a link. Some of the sites have amazing image quality.[1]
Pay per video. I've not actually seen many serious proposals for a site using this model.
Monthy fee for unlimited: Netflix's steaming service uses this model, although mostly in the movies catagory, which for some reason, people seem to treat very differently.
The problem with all models of streaming is that they don't work unless they have a sufficently wide selection. One of the biggest complaints about Netflix's streaming video service is that compared to Netflix's main business, the selection is pretty limited. If you feel like watching some specific movie right now (that has reached DVD), you can't really depend on it being available. On the other hand, with the main service, just about every movie out there is availble. People like to be able to know that virtually whatever you want is available. That is one of the great benefits of the iTunes store. Although there are a few exceptions, just about every song commerically released in the past 10 years or more is present. You hear a song on the radio, and you can be almost certain it is availble on iTunes. That is what the video sites are missing.
But back to the video sites. The ad supported ones not only don't have as much selection as they should, but the videos are often available only for a limited amount of time. So you can't just go back and watch the back episodes of a show you stumbled upon while flipping channels, and decided you liked. Most of this seems to be due to executive decisions based on not wanting to hurt DVD sales, or alienate TV stations that might want to pick the show up for syndication. (This would be syndication of back episodes while the series is still going, as opposed to rerun syndication of concluded shows, or syndication of gameshows, soap operas, talkshows, and other shows with potentialy unlimited number of seasons).
The best chance for a good legal streaming Television show site would probably be a monthly subscription site, that has deals in place for all the major networks, and on the day of launch has all primetime[2] shows currently airing on all networks with full back catalogs, and deals in the works for many concluded programs. That would be an enormous undertaking, as hundreds of contracts might need to be renegotiated to make room for such a site. So this is not likely to hapen for a while, but I'm still hopeful.
[1] Notable is abc.com, which uses a dynamic quality system, that adjusts the quality based on how far ahead you are with downloading, so you get the highest quality you can without having the video stop to buffer. The highest quality streams of those videos rival the quality of the 5 episodes 42 minutes long on 1 DVD format.
[2] Here I am excluding game shows, talk shows, game operas, news programs, and soaps, even if they air during primtime, and am including other first-run programming similar to most primetime shows, even if they are being aired outside of primetime. The excluded shows may or may not be present at all. (It is very unlikely anybody wants to watch yesterday's Good Morning America.) Those that are present may have only limited and potentially rolling back episodes. It might be nice to have a few back episodes of Price is Right, but there is little interest in going back much firther than 5 or so. I'd imagine there is more interest in old episodes of "Spiderman: The Animated Series" then there is in old Survivor Season 1 episodes, or old Season 30 episodes of "As the World Turns".
Nobody would ever call it GNU/X11/GNOME/Linux, since GNOME is part of the GNU project. rms officially has no issue with with it being called GNU/X11/Linux, although he might note that by either alphabetical order or code size, it might be better as GNU/Linux/X11.
AIUI, The universe is not expanding into anything. It is strictly equivalent to the universe remaining the same size, and everything in the universe shrinking. Since anything that we would use to measure would be shrinking too, it has the appearance of the universe expanding. Things close enough are held together by gravity, which is why the measuring stick does not stay the same size, but shrinks at the same rate as the individual particles. If this were not the case, the measuring stick would not be shrinking, only its particles would be, which would result in the universe appearing to remain the same size.
Neither the heat nor the energy can reduce to zero. The temperature of the universe will approach absolute zero, but it cannot be actually zero. What does happen is that the amount of energy that can be converted to useful work will reach zero. The amount of energy is constant (as long as mass is considered a form of energy, see: conservation of mass/energy).
You mean like all the many independent app stores that previously existed for the PalmOS, Windowm Mobile, and Symbian OS, but are being phased out in favor of Apple style APP stores?
Or how about independent app sales by the application developer, who you pay with a credit card, and they send you the app, or more commonly they send you an unlock code for the app, which is publicly available shareware style?
Those systems work well, except that the average moron that buys an Windows Mobile based phone from HTC have no idea they can just go online and download any Windows Mobile 6 compatible application they find. For whatever reason that concept just does not occur the the average idiot. I don't know why. Whenever I get a new device where I might be able to run software from the Internet, one of the first things I do is go looking around at what is available. I'm guessing most readers of this site are like this too.
Palm here though is hosting the apps on palm owned webspace.
I just don't get it. If palm is letting non-reviewed apps be made available (They will probably still review any placed in the store, but being in the store will not be a requirement for installation), then why require them to come from a special location?
Why not just create a standard archive format for app distribution, and let anybody make and distribute apps. Just like with the Windows Mobile platform, or for that matter the PalmOS platform?
The real reason that people run 32 bit windows on 64 bit machines is that many drivers are not available in 64 bit form, signed or not.
Then there are compatibilty issues in terms of things like browser plugins, and other areas where things should just work, but don't because not every single component is available in a 64 bit version. Since you can't mix 32 bit and 64 bit code in a single process (although with extreme difficulty you can simulate it) many problems can be caused where they simply don't belong.
Also please consider the lack of the NTVDM, and WOW, so my old 16 bit code will not run. To this day I still have a few old 16 bit applications on my system, and I fully expect them to run just fine, to the point where if I were running a 64 bit Windows, I would actually be surprised that they do not run.
Windows XP's backwards compatibility is not quite perfect, but the vast majority of old software runs just fine, and nearly all will run if one fiddles with the compatibility settings. I'm sure there are a lot of businesses out there that depend on that.
Indeed, both torrents and Usenet are secondary distribution systems. The original scene releases work through a completely different system of topsites and PREs, etc.
Of course, I've definately downloaded a torrent before consisting of many rar files, inside which was a zip file, inside which were the original scene rars, insider which was the content, plus some supplementary material in a zip file.
That means that some files have had 4 layers of compression. That drives me nuts personally. I far prefer that after the initial scene release, the files be extracted, and torrented as-is, or at worse as one large rar file.
The really screwed up thing is that it is very rare for people to differentiate between coders, decoders, codecs, and encoding schemes.
DivX is a codec. A codec is a specific piece of software for converting between video and a specific encoding scheme. It is actually a terrible term. There is no such thing as a codec. There are encoders and decoders. They are often distributed in pairs. When doing so, the decoder will always support the output of the encoder, but might also support video using encoding scheme features not used by the encoder. The only reasonable definition for codec is an encoder-decoder pair distributed together, but the term is very frequently used to refer to just a decoder or to an ecoding scheme.
"ISO/IEC 14496-2 ASP profile", sometimes shorted to "MPEG-4 Part 2 ASP" or similar, is a video encoding scheme. A subset of this format is supported by the DivX decoder. Any decoder that fully supports this encoding format can play any stream that conforms to it. So any decoder for this format can play any video encoded by the DivX encoder.
AVI is a container format. It happens to the container format traditionally used for MPEG-4 ASP videos encoded by DivX.
Unfortunately it is very common to use codec names to refer to encoding schemes, even among the people well aware of the codec/container-format separation. Part of the problem is that very little implements all of an encoding scheme. I mean there is no formal name for the MPEG-2 ASP subset that the DivX decoder supports. So people just call it DivX.
The whole situation would be much better if nobody ever used encoding scheme subsets, without formally defining them, and given them a specific name. Then we kill off the codec idea, and just have decoders which handle a specific encoding scheme (or multiple encoding schemes) fully, and thus any decoder that supports a specific scheme is fully interchangeable. Container formats would not have a codec field, but merely a feild that describes what scheme is used.
Enocders would be a completely seperate thing. They generally are today anyway. It is not terribly common to use the encoder portion of codecs, but rather specialized software for encoding into the format.
One would identify files by listing the container format, video encoding scheme, and audio encoding schemes. For example: an AVI-ASP-AAC movie. Common combinations might be given unique names as a shorthand.
The most important part of this idealized scenario is that all encoding schemes, combination shorthands, decoders, and encoders have seperate names. Having an encoding scheme, encoder, and decoder with the same name only leads to confusion.
Just look at the Wikipedia articles for codecs currently and you will see articles very confused because they fail to distinguish between a codec and an encoding scheme.
But you are NOT intended to be able to just read the article there. You are intended to be able to read the headline, see the images, and to see roughly how large the article is. (A few sentences, a few paragraphs, more than a page?).
If the story looks interesting, you click on it, and read it on the original web site. At said site, all those things should work fine.
Wow! Somebody did a wonder job with making that. It really captures the look and feel of the new Office interface. It would be awful to use, of course, but that image makes it pretty clear how broken the ribbon UI is.
The services question are 1-900-xxx-xxxx numbers here, and charge outrageous fees. The problem is that in some rural areas, apparently, the small teclos charge the long distance carriers (the phone network backbone) significantly more to connect each call then the connection would normally be worth. The per connection fee is charged to the long distance carrier regardless of what type of number is dialed.
These small telcos then agree to share revenue from connection costs with those 900 number services, in exchange for them using them for service. Those numbers are high volume, so they pull in many, many calls. If the revenue sharing is done right, by doing this the small telco can make more money then if they did not partner with these services, and instead charged a reasonable connection fee. Under such a system, both the small telco and the 900 number service benefit, at the expense of the long distance carrier.
The result of all this is that it cost more for Google (or AT&T, or any long distance carrier) to connect a call to the normal numbers served by that small Telco then it should, because those calls are subsidizing the 900 number services.
Since Google is providing a free service, the additional expense of those calls hurts them much more than a traditional long-distance carrier. For Google any connection costs are being payed for from revenue from other ventures, since they make no money on this. For AT&T they get payed each month by the customer. Now, AT&T really cant charge more for those calls, since they generally charge a flat rate per call charge, or more often these days, a flat rate per line (unlimited number of calls). But never the less, AT&T is making money from their long distance service, even with these extra charges, so it is not as big a deal to them.
I agree that short-distance public transportation[1] is largely an all or nothing proposition. It it does not have sufficient coverage, it is damn near worthless, except for a small number of people, who happen to frequently travel between two places served by the system. So if it is not fully implemented all at once, it ends up looking like it is a bad proposition.
For an example of a very nice transportation system consider London's Underground. No matter where you are, stops are only a few blocks away at most, The routes appear to have been planned for being relatively efficient in getting from any destination to another, requiring few train changes, and many major destinations require none. The lines are all identified by a unique name, and distinct color.[2] The system was damned enjoyable to use, and I know if I lived in London, I would either not own a car, or own one only for trips outside of London. [3] Further, that is all based solely on the subway system, not considering the other parts of the system, like buses.
Unfortunately there is another problem with public transportation in the US. The US often has fairly low population density. Public transportation tends to work better in high population density areas. One of the obvious issues is greater cost due to the longer distances. Even if we consider only the Metro areas of the county, the United States tend to have lower poulation densities then the city sizes would imply. We seem to have cultivated a suburb culture, in which significant commutes between the suburbs and city are common. The problem here is that suburbs low population denisty make it difficult to provide good coverage. In many suburbs it would just not be economical to have a subway stop every a few city-block-distances, for example. The further appart the stops are placed, the further people need to walk, and a distance of more than a few city-block-distances becomes immpractical to walk.
It is reasonable to have many bus stops in the suburbs, but it would not be economical to have buses arrive every 10 minutes or so. In many areas the frequency would be every hour or worse. That is also a problem. If the timing of the bus happens to be off a bit too much (taking it would cause you to be 15 minutes late to work, for example), you would need to take the previous bus, which could effectively add nearly an hour of wasted time (not traveling) to you commute. That makes it much less useful.
[1] Long distance public transportation, such as airlines can need less coverage to be useful. [2] All too many subway systems use numbers or letters, which are harder to remember, and have multiple lines re-use the same color for some weird reason. [3] Even if public transportation can get me to my destination, depending on that destination, there might not be sufficient public transportation at my destination to move around to where I want.
And looking closer at Honda's version, they use a special form of wheel that works side to side as well as front to back. There are several methods of doing that, but conceptually a large rubber (or other material) sphere can be used. Mount motors with wheels against the sphere at right angles to each other. Think like the sensors in old ball-based mice, which are at right angles to the ball, but rather than being sensors they are motors for spinning the ball as desired.
So now you have the ability to control forward and backwards motion using one motor, (along with the gyroscope). So you set this up to work just like a standard e-unicycle. It just tries to keep the seat straight up.
Now set up a similar system for balancing side to side. The result of this would ideally be a device that moves whatever way you lean, without ever turning. So when you sat on it you would always face the same direction, but could move in any direction.
In reality the feedback systems will always have some slop so moving forward and to the side will result in some turning. Even gyroscopes drift, and so would this system. By deliberately adding additional slop to the feedback system, a fairly natural lean-like turning can result, but with additional stability. When going forwards/back slowly enough leaning to the side will result in sideways movement with almost no turning.
The real answer is that side to side balance is maintained by precession, (like on a bicycle) combined with some additional balancing by shifting body weight (also critical on a bike). Unlike on a bike, steering based corrections to balance are not present. With sufficiently wide wheel unicycles, wheel geometry becomes the primary side to side stabilzing method.
Steering is completely based on leaning in normal unicycles. Normal bicycles also include an additional steering component (the additional wheel that turns).
Forward and backwards balance is maintained by a combination of of creating a mental feedback loop that causes one to vary cycling speed as necessary to keep the seat roughly upright, along with manually shifting weight forwards and backwards.
For electronic unicycles, steering and side to side balance generally are the same as with manual ones. However the forward and backwards stability does not rely on any weight shifting on the part of the rider, but solely on varying the motor speed as needed to keep the seat upright.
Using a feedback system for keeping the seat upright automatically gives the segway-style speed control on these devices. In order to do more traditional style speed controls requires a more complicated system that varies the angle of the seat that the system tries to maintain as necessary such that the average speed remains as desired. Far more complicated, and not needed, so I've not seen any e-unicycle that does not use segway-style speed control.
There are some tricks that allow steering not based on leaning, and some of the e-unicycle designs I have seen use those, but others use lean based steering which works fine, except for at near stationary speeds, but some of these other systems allow for a smaller turning radius.
What Firefox is apparently actually proposing, is not an office-like ribbon, but a UI that is very similar to that of chrome. (I base that on the image shown in the article). Compared to my current use which has a title bar, menu bar, main bar and tab-bar, the proposed interface will take up less room.
A ribbon does reasonably well in exposing users to useful features in the program that have been there for ages of which the user was unaware. The problem is it takes up a lot of space, and if the icon is not right where you would put it, it can be rather difficult to find. With menus, if the option is not in the top level menu you would put it in, it is still easy enough to look through the other top level menus to find it. With the ribbon, it can be a real pain.
Slashdot Mirror
Its very interesting all the different mice like systems out there.
We have various forms of the traditional mouse. We are all pretty familar with its capabilities.
The trackball has the same set of abilities, plus the spin ability that mice lack. I'm not sure of many uses for just spining the ball fast and letting it roll to a stop, but I;d be surprised if that is not useful in some manner.
In theory, a trackball can also allow yaw control, although I've not seen it done, and it could be awkward to use. This is a stationary device.
We have the touch pad. Its only real use advantage over a mouse is the fact that it is stationary. It has a manufacturing benefit of taking up less room than most other stationary input schemes.
We have the track point. This has the fairly unique ability to allow one to apply a constant velocity to the mouse cursor. I know of one game in particular where that would be very useful, since a specific constant velocity would cancel out the gravity in the parts of the game that use it, returning normal control to those levels. It is also stationary, but is limited. There is no scroll wheel style support that the others devices have.
Next we have absoluter positioning devices.
One of the most familiar is the standard touch screen. This like other absolute positioning devices lets the cursor teleport from one location to another. This is rather nice for a few stupid flash games, like the get through the maze without touching the lines games. The downside of these is that they generally don't allow mouse positioning without simultaneously clicking.
Next we have the Wacom style digitizer. This not only allows for cursor movement without clicking, but also normally provides pressure data, and in some cases can provide angle relative to the "paper" . (Used when simulating the use of caligraphy writing instruments where line width is based on angle, among others). This is also the input device that best allows for writing and drawing. Anybody who has ever tried to write with a mouse or trackpad knows how terrible the results come out, while digitizers of sufficient resolution should result in writing almost instinguishable from that done on paper.
This is just as good a place as any to ask about the support of templates in this device. Important components of some articles are generate by templates. One example is the infamous [citation needed] text, which is generated by the "{{cn}}" template. Other times, important words in a sentence are used as a argument for a template, to produce some from of link automatically.
Some other mobile Wikipedia solutions, such as one I saw for the iPhone, just ignore templates. That means that important words in a sentence could potentially be omitted. In some cases, entire sections of an article may be omitted. I consider that extremely problematic.
Assuming they are properly supported, are references supported? In few articles I've seen the feature used for general footnotes in addition to references, and having those disappear could be problematic.
What about the LaTeX math equations? A lot of mathematical and Computer Science articles become completely worthless if those are omitted, but including them means included quite a few generated images for some of the more complex ones that cannot be rendered as html.
And what about the ez-timeline extention. Are the images that it generates included?
What about the hieroglyphics that articles may include by way of the wikihiro extention?
I would not be willing to use a static mobile Wikipedia that did not support templates, references, tables, external links, LaTeX equations.
Other people might insist that the categories pages be included, and that time lines and hieroglyphics be supported.
"A bitching sailor is a happy sailor"
Is that bitching as in complaining, or as in "That's a bitchin' ride, bro."?
Correction only Windows Perfessional upgrade is available through the Campus agreement, not Ultimate. A bit surprising, since a campus would probably desire at least Windows Enterpise on the lab computers.
If they are giving students MSDNAA copies of Windows 7, they had better be alerting the students that they may not use it for any commerical endevor, even side jobs, If they are not in Science, Technology, Engineering, or Mathmatics department, or using the computer for "Visualization, Illustration, Design, [or] Art", they may not install the OS at all. Upon graduation, they must uninstall Windows 7.
I think it is more likely that they have a Microsoft Campus Agreement. The agreement's student option allows the school yo pay some amount per full time equivlent student, and allows the school to provide copies of the software for the student's personal machines. All graduating students get a perpetual license to use the software. Everybody else must remove it upon leaving the school. The license is still for non-commercial use only, but it does apply to all majors.
My school uses this program to provide copies of MS Office Enterpise for $50 to students. A permenate non-commecial MS Office enterprise license is still worth a hell of a lot more than $50, so this is a great deal for students. I'm not sure what they charge for Windows, but it will still not be a bad deal. I'm guessing it is the ultimate edition that is available. This program though only provides upgrade versions of Windows, not full versions, and I'm betting the discs provided have the retail upgrade-only image.
When I see a website that accepts no advertising, buys all it's products for cash, anonymously from retail stores and has a test suite that reflects what actual users actually do, then their reviews will have merit (although I can't see anyone anywhere paying the hue cover price for such a publication)
Except perhaps the part about how well the test suite reflects what users do, You have just described the periodical "Consumer Reports". No advertising, buys products anonymously at retail. People do pay cover price for it, although subscription price is far more common. The biggest problem is they cover everything, not just technology, so the level of technology coverage leaves a lot to be desired.
And reviewers almost always use systems that had specs well beyond the vista absolute minimum specs, while many users who had issues were using a one of the cheepest machines on the market at the time, which were not exceeding the absolute minimum specifications by much. Vista RTM definitely was far more responsive, etc on a better spec'ed machine.
Cosign.
Tan Gent.
Correct. C++0x will have native support for threads, based largely on Boost threading libraries, but with some differences, and the potential for better support of some features, by adding support to the language core.
Rapidshare would not the right right site for this. Legitimate content would be downloadbale only through things like the iTunes store.
The other alternative is streaming sites. Threre are three models for streaming sites.
Streaming with ads. Most of the legal online streaming is of this sort. Simply goto Hulu.com and search for the show you want. If hulu has it will show up. If hulu does not have it, but the network's own website has it, it will give a link. Some of the sites have amazing image quality.[1]
Pay per video. I've not actually seen many serious proposals for a site using this model.
Monthy fee for unlimited: Netflix's steaming service uses this model, although mostly in the movies catagory, which for some reason, people seem to treat very differently.
The problem with all models of streaming is that they don't work unless they have a sufficently wide selection. One of the biggest complaints about Netflix's streaming video service is that compared to Netflix's main business, the selection is pretty limited. If you feel like watching some specific movie right now (that has reached DVD), you can't really depend on it being available. On the other hand, with the main service, just about every movie out there is availble. People like to be able to know that virtually whatever you want is available. That is one of the great benefits of the iTunes store. Although there are a few exceptions, just about every song commerically released in the past 10 years or more is present. You hear a song on the radio, and you can be almost certain it is availble on iTunes. That is what the video sites are missing.
But back to the video sites. The ad supported ones not only don't have as much selection as they should, but the videos are often available only for a limited amount of time. So you can't just go back and watch the back episodes of a show you stumbled upon while flipping channels, and decided you liked. Most of this seems to be due to executive decisions based on not wanting to hurt DVD sales, or alienate TV stations that might want to pick the show up for syndication. (This would be syndication of back episodes while the series is still going, as opposed to rerun syndication of concluded shows, or syndication of gameshows, soap operas, talkshows, and other shows with potentialy unlimited number of seasons).
The best chance for a good legal streaming Television show site would probably be a monthly subscription site, that has deals in place for all the major networks, and on the day of launch has all primetime[2] shows currently airing on all networks with full back catalogs, and deals in the works for many concluded programs. That would be an enormous undertaking, as hundreds of contracts might need to be renegotiated to make room for such a site. So this is not likely to hapen for a while, but I'm still hopeful.
[1] Notable is abc.com, which uses a dynamic quality system, that adjusts the quality based on how far ahead you are with downloading, so you get the highest quality you can without having the video stop to buffer. The highest quality streams of those videos rival the quality of the 5 episodes 42 minutes long on 1 DVD format.
[2] Here I am excluding game shows, talk shows, game operas, news programs, and soaps, even if they air during primtime, and am including other first-run programming similar to most primetime shows, even if they are being aired outside of primetime. The excluded shows may or may not be present at all. (It is very unlikely anybody wants to watch yesterday's Good Morning America.) Those that are present may have only limited and potentially rolling back episodes. It might be nice to have a few back episodes of Price is Right, but there is little interest in going back much firther than 5 or so. I'd imagine there is more interest in old episodes of "Spiderman: The Animated Series" then there is in old Survivor Season 1 episodes, or old Season 30 episodes of "As the World Turns".
Nobody would ever call it GNU/X11/GNOME/Linux, since GNOME is part of the GNU project.
rms officially has no issue with with it being called GNU/X11/Linux, although he might note that by either alphabetical order or code size, it might be better as GNU/Linux/X11.
AIUI, The universe is not expanding into anything. It is strictly equivalent to the universe remaining the same size, and everything in the universe shrinking. Since anything that we would use to measure would be shrinking too, it has the appearance of the universe expanding. Things close enough are held together by gravity, which is why the measuring stick does not stay the same size, but shrinks at the same rate as the individual particles. If this were not the case, the measuring stick would not be shrinking, only its particles would be, which would result in the universe appearing to remain the same size.
Neither the heat nor the energy can reduce to zero. The temperature of the universe will approach absolute zero, but it cannot be actually zero. What does happen is that the amount of energy that can be converted to useful work will reach zero. The amount of energy is constant (as long as mass is considered a form of energy, see: conservation of mass/energy).
You mean like all the many independent app stores that previously existed for the PalmOS, Windowm Mobile, and Symbian OS, but are being phased out in favor of Apple style APP stores?
Or how about independent app sales by the application developer, who you pay with a credit card, and they send you the app, or more commonly they send you an unlock code for the app, which is publicly available shareware style?
Those systems work well, except that the average moron that buys an Windows Mobile based phone from HTC have no idea they can just go online and download any Windows Mobile 6 compatible application they find.
For whatever reason that concept just does not occur the the average idiot. I don't know why. Whenever I get a new device where I might be able to run software from the Internet, one of the first things I do is go looking around at what is available. I'm guessing most readers of this site are like this too.
Palm here though is hosting the apps on palm owned webspace.
I just don't get it. If palm is letting non-reviewed apps be made available (They will probably still review any placed in the store, but being in the store will not be a requirement for installation), then why require them to come from a special location?
Why not just create a standard archive format for app distribution, and let anybody make and distribute apps. Just like with the Windows Mobile platform, or for that matter the PalmOS platform?
The real reason that people run 32 bit windows on 64 bit machines is that many drivers are not available in 64 bit form, signed or not.
Then there are compatibilty issues in terms of things like browser plugins, and other areas where things should just work, but don't because not every single component is available in a 64 bit version. Since you can't mix 32 bit and 64 bit code in a single process (although with extreme difficulty you can simulate it) many problems can be caused where they simply don't belong.
Also please consider the lack of the NTVDM, and WOW, so my old 16 bit code will not run. To this day I still have a few old 16 bit applications on my system, and I fully expect them to run just fine, to the point where if I were running a 64 bit Windows, I would actually be surprised that they do not run.
Windows XP's backwards compatibility is not quite perfect, but the vast majority of old software runs just fine, and nearly all will run if one fiddles with the compatibility settings. I'm sure there are a lot of businesses out there that depend on that.
Indeed, both torrents and Usenet are secondary distribution systems. The original scene releases work through a completely different system of topsites and PREs, etc.
Of course, I've definately downloaded a torrent before consisting of many rar files, inside which was a zip file, inside which were the original scene rars, insider which was the content, plus some supplementary material in a zip file.
That means that some files have had 4 layers of compression. That drives me nuts personally. I far prefer that after the initial scene release, the files be extracted, and torrented as-is, or at worse as one large rar file.
The really screwed up thing is that it is very rare for people to differentiate between coders, decoders, codecs, and encoding schemes.
DivX is a codec. A codec is a specific piece of software for converting between video and a specific encoding scheme. It is actually a terrible term. There is no such thing as a codec. There are encoders and decoders. They are often distributed in pairs. When doing so, the decoder will always support the output of the encoder, but might also support video using encoding scheme features not used by the encoder. The only reasonable definition for codec is an encoder-decoder pair distributed together, but the term is very frequently used to refer to just a decoder or to an ecoding scheme.
"ISO/IEC 14496-2 ASP profile", sometimes shorted to "MPEG-4 Part 2 ASP" or similar, is a video encoding scheme. A subset of this format is supported by the DivX decoder. Any decoder that fully supports this encoding format can play any stream that conforms to it. So any decoder for this format can play any video encoded by the DivX encoder.
AVI is a container format. It happens to the container format traditionally used for MPEG-4 ASP videos encoded by DivX.
Unfortunately it is very common to use codec names to refer to encoding schemes, even among the people well aware of the codec/container-format separation. Part of the problem is that very little implements all of an encoding scheme. I mean there is no formal name for the MPEG-2 ASP subset that the DivX decoder supports. So people just call it DivX.
The whole situation would be much better if nobody ever used encoding scheme subsets, without formally defining them, and given them a specific name. Then we kill off the codec idea, and just have decoders which handle a specific encoding scheme (or multiple encoding schemes) fully, and thus any decoder that supports a specific scheme is fully interchangeable. Container formats would not have a codec field, but merely a feild that describes what scheme is used.
Enocders would be a completely seperate thing. They generally are today anyway. It is not terribly common to use the encoder portion of codecs, but rather specialized software for encoding into the format.
One would identify files by listing the container format, video encoding scheme, and audio encoding schemes. For example: an AVI-ASP-AAC movie. Common combinations might be given unique names as a shorthand.
The most important part of this idealized scenario is that all encoding schemes, combination shorthands, decoders, and encoders have seperate names. Having an encoding scheme, encoder, and decoder with the same name only leads to confusion.
Just look at the Wikipedia articles for codecs currently and you will see articles very confused because they fail to distinguish between a codec and an encoding scheme.
But you are NOT intended to be able to just read the article there. You are intended to be able to read the headline, see the images, and to see roughly how large the article is. (A few sentences, a few paragraphs, more than a page?).
If the story looks interesting, you click on it, and read it on the original web site. At said site, all those things should work fine.
Wow! Somebody did a wonder job with making that. It really captures the look and feel of the new Office interface. It would be awful to use, of course, but that image makes it pretty clear how broken the ribbon UI is.
The services question are 1-900-xxx-xxxx numbers here, and charge outrageous fees. The problem is that in some rural areas, apparently, the small teclos charge the long distance carriers (the phone network backbone) significantly more to connect each call then the connection would normally be worth. The per connection fee is charged to the long distance carrier regardless of what type of number is dialed.
These small telcos then agree to share revenue from connection costs with those 900 number services, in exchange for them using them for service. Those numbers are high volume, so they pull in many, many calls. If the revenue sharing is done right, by doing this the small telco can make more money then if they did not partner with these services, and instead charged a reasonable connection fee. Under such a system, both the small telco and the 900 number service benefit, at the expense of the long distance carrier.
The result of all this is that it cost more for Google (or AT&T, or any long distance carrier) to connect a call to the normal numbers served by that small Telco then it should, because those calls are subsidizing the 900 number services.
Since Google is providing a free service, the additional expense of those calls hurts them much more than a traditional long-distance carrier. For Google any connection costs are being payed for from revenue from other ventures, since they make no money on this. For AT&T they get payed each month by the customer. Now, AT&T really cant charge more for those calls, since they generally charge a flat rate per call charge, or more often these days, a flat rate per line (unlimited number of calls). But never the less, AT&T is making money from their long distance service, even with these extra charges, so it is not as big a deal to them.
I agree that short-distance public transportation[1] is largely an all or nothing proposition. It it does not have sufficient coverage, it is damn near worthless, except for a small number of people, who happen to frequently travel between two places served by the system. So if it is not fully implemented all at once, it ends up looking like it is a bad proposition.
For an example of a very nice transportation system consider London's Underground. No matter where you are, stops are only a few blocks away at most, The routes appear to have been planned for being relatively efficient in getting from any destination to another, requiring few train changes, and many major destinations require none. The lines are all identified by a unique name, and distinct color.[2]
The system was damned enjoyable to use, and I know if I lived in London, I would either not own a car, or own one only for trips outside of London. [3] Further, that is all based solely on the subway system, not considering the other parts of the system, like buses.
Unfortunately there is another problem with public transportation in the US. The US often has fairly low population density. Public transportation tends to work better in high population density areas. One of the obvious issues is greater cost due to the longer distances. Even if we consider only the Metro areas of the county, the United States tend to have lower poulation densities then the city sizes would imply. We seem to have cultivated a suburb culture, in which significant commutes between the suburbs and city are common. The problem here is that suburbs low population denisty make it difficult to provide good coverage. In many suburbs it would just not be economical to have a subway stop every a few city-block-distances, for example. The further appart the stops are placed, the further people need to walk, and a distance of more than a few city-block-distances becomes immpractical to walk.
It is reasonable to have many bus stops in the suburbs, but it would not be economical to have buses arrive every 10 minutes or so. In many areas the frequency would be every hour or worse. That is also a problem. If the timing of the bus happens to be off a bit too much (taking it would cause you to be 15 minutes late to work, for example), you would need to take the previous bus, which could effectively add nearly an hour of wasted time (not traveling) to you commute. That makes it much less useful.
[1] Long distance public transportation, such as airlines can need less coverage to be useful.
[2] All too many subway systems use numbers or letters, which are harder to remember, and have multiple lines re-use the same color for some weird reason.
[3] Even if public transportation can get me to my destination, depending on that destination, there might not be sufficient public transportation at my destination to move around to where I want.
And looking closer at Honda's version, they use a special form of wheel that works side to side as well as front to back. There are several methods of doing that, but conceptually a large rubber (or other material) sphere can be used. Mount motors with wheels against the sphere at right angles to each other. Think like the sensors in old ball-based mice, which are at right angles to the ball, but rather than being sensors they are motors for spinning the ball as desired.
So now you have the ability to control forward and backwards motion using one motor, (along with the gyroscope). So you set this up to work just like a standard e-unicycle. It just tries to keep the seat straight up.
Now set up a similar system for balancing side to side. The result of this would ideally be a device that moves whatever way you lean, without ever turning. So when you sat on it you would always face the same direction, but could move in any direction.
In reality the feedback systems will always have some slop so moving forward and to the side will result in some turning. Even gyroscopes drift, and so would this system. By deliberately adding additional slop to the feedback system, a fairly natural lean-like turning can result, but with additional stability. When going forwards/back slowly enough leaning to the side will result in sideways movement with almost no turning.
The real answer is that side to side balance is maintained by precession, (like on a bicycle) combined with some additional balancing by shifting body weight (also critical on a bike). Unlike on a bike, steering based corrections to balance are not present. With sufficiently wide wheel unicycles, wheel geometry becomes the primary side to side stabilzing method.
Steering is completely based on leaning in normal unicycles. Normal bicycles also include an additional steering component (the additional wheel that turns).
Forward and backwards balance is maintained by a combination of of creating a mental feedback loop that causes one to vary cycling speed as necessary to keep the seat roughly upright, along with manually shifting weight forwards and backwards.
For electronic unicycles, steering and side to side balance generally are the same as with manual ones. However the forward and backwards stability does not rely on any weight shifting on the part of the rider, but solely on varying the motor speed as needed to keep the seat upright.
Using a feedback system for keeping the seat upright automatically gives the segway-style speed control on these devices. In order to do more traditional style speed controls requires a more complicated system that varies the angle of the seat that the system tries to maintain as necessary such that the average speed remains as desired. Far more complicated, and not needed, so I've not seen any e-unicycle that does not use segway-style speed control.
There are some tricks that allow steering not based on leaning, and some of the e-unicycle designs I have seen use those, but others use lean based steering which works fine, except for at near stationary speeds, but some of these other systems allow for a smaller turning radius.
Very true.
What Firefox is apparently actually proposing, is not an office-like ribbon, but a UI that is very similar to that of chrome. (I base that on the image shown in the article). Compared to my current use which has a title bar, menu bar, main bar and tab-bar, the proposed interface will take up less room.
A ribbon does reasonably well in exposing users to useful features in the program that have been there for ages of which the user was unaware. The problem is it takes up a lot of space, and if the icon is not right where you would put it, it can be rather difficult to find. With menus, if the option is not in the top level menu you would put it in, it is still easy enough to look through the other top level menus to find it. With the ribbon, it can be a real pain.