The actual remarks by Commissioner Clyburn are a lot less technical than the SL summary implies. What she wants doesn't have so much to do with the physical deployment of broadband (the "last mile"), as it does with actually adopting the broadband that's available (what she called the last "half-mile".)
Apparently, 93 million Americans don't use broadband, even though they could. Why? The Commish says its because 1) broadband costs too much, 2) non-users lack "digital literacy", and 3) non-users just don't see the benefit. Her proposed fixes have a lot more in common with the Peace Corps than with the White Spaces Coalition: provide free broadband to the poor, and form a National Digital Literacy Corps to deliver a National Digital Literacy Program while going home-to-home to help set up broadband. Her model is the recent national switch-over to digital TV, where hundreds of thousands of volunteers went around and hooked people's TVs up.
I purchased four XO-1s when they originally came out a couple years ago. I gave away two, boxed one for posterity, and am still using one for browsing (Opera) and note-taking (Zim) when I'm at conferences. I still get heads-turns and kids inching over to take a look over my shoulder everywhere I go.
The XO-1 has an early version of the Pixel Qi screen, and it is extremely functional. I'm still amazed every time I'm reading an ebook on the subway, and walk from the deep darkness of the subway tunnel into blinding, direct sunlight, and the XO-1 display is still completely readable.
The XO-1's processor, however, is quite slow, and that becomes a pain in the neck for browsing. A decently-performing netbook doesn't cost very much these days, but the screens are a disappointment. I'm really looking forward to snagging a Pixel Qi DIY kit, buying a cheap netbook, and fixing up my ride.
One thing that Courier nails is the concept of using both fingers and pen. Go ahead, try it. There are two basic pen positions: a "writing" position that uses all five fingers, and a "resting" position where the pen is rotated 90 degrees, and held in place by a single finger, leaving the other fingers free. The writing position is vertical, resting on the edge of the hand. The resting position is horizontal, palm down.
The Courier UI mockup uses both of these hand configurations and orientations. Flat, horizontal motions such as flipping a page or image dragging are done in the resting position. Vertical motions such as drawing and writing are done in writing position. Switching between the two is very fast and natural-feeling.
Having a pen dispenses with the need for a QUERTY keyboard, but block-printing is not the solution either. For one thing, it's too slow: the average printing speed is about 15wpm. A better solution might be a stylus-based keyboard. Several years ago, IBM invented a shorthand named Shark (commercialized as ShapeWriter, I believe) that was extremely effective. After just a few minutes of practice with it, I was able to achieve 40wpm.
Amazon's DMCA procedures are very deliberative. If you don't like youtube's enforcement policies, shop around! Host your file on S3, and post a video preview on youtube with a link to S3.
For example, Amazon's procedure requires an electronic or physical signature on the takedown notice and a statement made under _penalty of perjury_ that the notice is accurate and that you are the copyright owner or authorized to act on the copyright owner's behalf. This won't slow down a legitimate owner or agent much, but it will give you and Amazon a basis to sue if the takedown notice doesn't come from an owner or authorized agent. (Personally, I suspect that a fair number of takedown notices are not coming from people who have been properly authorized. People claim copyright on all sorts of things that aren't theirs to claim at all -- even large publishing businesses who ought to know better.)
Just read the MS document, and concur with other comments here. Knowing the document's contents actually RAISES my level of trust, since it lays out precisely the conditions under which private information will be revealed (basically: go through the court system), and what information is available.
All companies who deliver cloud-based services should reveal this information to their customers. There is absolutely no reason to keep it secret. Microsoft, Google, Yahoo: man up, and make it official.
Seeing how Network Solutions is utterly mis-handling this situation is destroying my trust in them by the bucket-load. I am seriously considering moving my domains off of them. I certainly wouldn't want to be strong-armed just because I made some big company like MS mad!
A simple system can be specified with "rough" UML models, that's true. That is precisely one of the design goals of the UML language, in fact, to be fast and efficient replacement for human language.
However, an elaborate system needs much more elaborate specifications. It won't be enough to say, "I want a table with a couple of joins, here's a rough sketch", you will need to be much more explicit about fields, data types, keys, relations, triggers, and all the rest.
It won't be enough to say, "This system will call that service, here's a rough sketch", you will need to be much more explicit about data contracts, protocols, end points, delegations, object mapping, and all the rest.
UML is just as fast and efficient at capturing these kinds of details as it is at sketching out the big picture. However, many developers get stuck on the "rough sketch" level of UML, and never go beyond. They wind up sticking their specification details into some code editor, where no one else on the development team can see them except another developer.
However, a detailed spec is just as valuable to a tester, business analyst, security analyst, and database engineer as it is to the developer.
Don't know if they're still available, but when I went out on my own for a few years, I opted to pay most of my health care out-of-pocket. I bought a high-deductible, catastrophic insurance plan that would cover any medical expense that would have bankrupted my family.
It was a wonderful experience. The catastrophic insurance rates were very low. Over three years, the amount I spent out-of-pocket was about 70% of the amount I had been spending on an employer-supported plan. There were no forms to fill out. No paperwork, except at tax time, when I could claim a medical deduction for the only time in my life. We could visit any doctor we wished. Some doctors even gave us a small discount for paying up front.
In my experience, my family has never needed more than a tiny fraction of the health care services available in the employer plans I've been forced to participate in. I wouldn't mind moving permanently to a high-deductible, catastrophic system at all.
Programming is a young person's game. In the beginning, it's exciting to have this fantastic machine bow to your every command. After 20 years, however, I had done everything I'd ever wanted to do in programming. I was more than happy to let the next generation take over.
What I became interested in was moving up a level in abstraction, and I got into modeling and architecting designs rather than executing them. Seems like a fairly natural progression to me, so yeah, I'm not surprised that the number of developers 40+ is small.
What's best about these guys is that they often haven't produced even a line of code. They just spew out UML diagram after UML diagram. I look at the diagrams, talk to the users, and it becomes obvious what should be done.
Hmm. I'd like to call a mild foul on this statement.
Are you saying you look at the UML models, ignore them, and then just talk to your users, or that you look at the models, and based on what you've gleaned from them, talk to your users, get to the heart of the matter, and start coding?
If it's the former, then why bother even looking at the models in the first place? They're messed up, as you say, so ditch 'em.
But if it's the latter, then at least part of the reason why you're able to be so effective is because you're using somebody else's work as your starting-point. Hardly seems fair to say they're "fuckups", when you're basing your work on theirs.
Religion has a huge impact on many aspects of society: language, culture, politics -- even science. Religion could certainly be a legitimate topic of academic study, done properly. For example, I doubt it is possible to truly understand the history of the United States without understanding the role of religious belief. It's just too intertwined.
Your point about people trying to pass religion off as if it were science is well taken, however. Bugs me when people try to pass humanism off as science, too.
Interesting how so many ads were trying to put the viewer into a desired mood, and then associate it with a product (the TV ads, the car ads, Coke, etc.), while Google's ad was just a straight-forward story. The Google sequence of search terms were exactly like frames of a comic, where the viewer fills in the storyline inferred by the sequence. (Having just read Understanding Comics by Scott McCloud, I now know that technique is called "closure" -- thanks, Scott.)
Telling a story instead of setting a mood really highlights the Google esthetic: function, always function, before form.
This is surely one of the most frightening and excruciating conditions one can imagine. My guess is that one would compensate somehow, but the loneliness and helplessness must be crushing. If I ever become vegetative, I would certainly want to be given the kind of tests outlined here. If there is no response, then I would like to be taken home and cared for until nature takes its course. The only reason I would want to stay in a hospital or care facility is if there is a chance I can improve.
With one patient - a Belgian man injured in a traffic accident seven years ago - they asked a series of questions.
He was able to communicate "yes" and "no" using just his thoughts.
The team told him to use "motor" imagery like a tennis match to indicate "yes" and "spatial" imagery like thinking about roaming the streets for a "no".
The patient responded accurately to five out of six autobiographical questions posed by the scientists.
For example, he confirmed that his father's name was Alexander.
You know, a lot of the UI features of the iPad don't look that revolutionary. Looks almost like a variant of Moblin to me. How hard would it be to create a Linux desktop manager that duplicates the functionality of the iPad? Shoot, it wouldn't even be that hard to go several features better (multi-tasking, daylight-readable screen, video camera, etc.)
It's great that Apple has put their vision out there, but it looks like when all is said and done, they're betting on their media tie-ins to keep their dominant hipster status. The special sauce doesn't seem to be in the software or the hardware of the iPad.
So, while Apple is busy trying to wrangle exclusive deals with Big Content, other smartbook vendors and the FOSS community can be busy analyzing the design choices of the iPad, and dreaming up an even better vision.
I believe the authors' point is that even though the errors do get replicated as you say, the resulting proteins created by the code are often unaffected by them. It's the resistance to error that's hard for Darwinian evolution to explain.
Though [the genetic code] was discovered in the 1960s, no one had been able to explain how evolution could have made it so exquisitely tuned to resisting errors. Mutations happen in DNA coding all the time, and yet the proteins it produces often remain unaffected by these glitches. Darwinian evolution simply cannot explain how such a code could arise. But horizontal gene transfer can, say Woese and Goldenfeld.
The thought was that China would be replaced with another developing country hungry for trade with the US. Don't really know much about this, but India comes to mind, along with any of the newly-democratic countries in East Europe or South America. Obviously couldn't happen overnight...
Different cultures and people have different values. Just because you think something is more moral doesn't mean everyone does so.
Value systems are properties of personal conscience, not of culture. As individuals, we cannot be "forced" to believe something against our will. You can be forced to _act_ like you believe something you don't (as in the book, 1984), but that would just be an act. You wouldn't really believe it.
Since values belong to persons, and persons make up cultures, then values precede culture, and in many ways, even transcend them. For example, If I value free speech and live in the US, I don't think I would value it any less if I lived in China. I may let my culture influence me in deciding which values I hold, but in the end, they're my values, no one else's.
It isn't valid to charge someone with arrogance merely because they state what they believe. If you tell me what you believe, you aren't coercing me in any way. You're just expressing yourself.
You're probably right. Then again, we are talking about a country that feels the need to filter what every single person in the whole blessed country reads every moment of every day on the Internet...
The thought was, China will suffer a major loss of prestige if Google goes through with this. Other companies will follow suit, leading to loss of access, loss of influence, loss of opportunity, and ultimately, loss of business.
China will lash out with wounded national pride, as they seem wont to do. This will further alienate them from the international community, leading to further loss of status.
Loss of prestige will encourage civil unrest, something China dreads. In some places, China seems to be a tinderbox, just waiting to catch fire. They are suffering from the global recession just as much as anyone.
That is a whole lot of pain that China doesn't need right now.
One fascinating aspect of this story is how Google, just a private corporation, is able to credibly threaten an entire country -- and a near-superpower one, at that! That used to take the kind of might only a government could wield.
No longer.
The web levels everyone -- and I mean EVERYONE -- to one, lowest common denominator: access.
over at PC World. Actually, I like the idea of the XO-3. Sure, it's totally blue-sky, but it's great to have at least one outfit taking a completely clean-slate design approach to mobile computing.
I like the hinged-panel XO-2 and MS Courier better, however. I think it's just more practical to have one part of the screen that can tilt up into the light. That said, the ring thingy of the XO-3 is interesting, too. I hadn't really thought about the mechanics of trying to hold a panel with one hand while touching with the other.
Remember 10/GUI, Clayton Miller's 10-fingered touch screen interface? Imagine a flexible 10/GUI touch pad that could be pulled out from under the XO screen. That might be interesting.
In an urban setting, the Segway was, and still is, an obviously better way to move people than a car. The fact that it couldn't get permission to use the sidewalks is more a chicken-and-egg thing than it is a fault of the basic premise. If there were 1,000 Segway-like vehicles on the streets per day in a downtown area, you can be sure that the local laws would be modified to allow them.
One of the biggest problems with mass transit is it doesn't have a good feeder system. We go straight from an arterial system (mass transit) to a capillary system (foot traffic), with nothing in between. Someday someone will fill this gap. The Segway was at least a plausible candidate to do the job.
"A system allowing a user of a browser program on a computer connected to an open distributed hypermedia system to access and execute an embedded program object. The program object is embedded into a hypermedia document much like data objects. The user may select the program object from the screen. Once selected the program object executes on the user's (client) computer or may execute on a remote server or additional remote computers in a distributed processing arrangement. After launching the program object, the user is able to interact with the object as the invention provides for ongoing interprocess communication between the application object (program) and the browser program. One application of the embedded program object allows a user to view large and complex multi-dimensional objects from within the browser's window. The user can manipulate a control panel to change the viewpoint used to view the image. The invention allows a program to execute on a remote server or other computers to calculate the viewing transformations and send frame data to the client computer thus providing the user of the client computer with interactive features and allowing the user to have access to greater computing power than may be available at the user's client computer."
In other words, the patent is on the entire concept of embedding objects in a browser. I think this illustrates perfectly some of the faults of software patents: 1) It is a concept for an invention, not an actual invention; 2) It is a re-statement of general practices and patterns (remote procedure call; client/server; interactive user interface) that only looks new because it is being re-applied to another technology (browsers, in this case); 3) It is over-broad in scope, covering not a particular invention but an entire class of inventions; 4) It is general in execution, not requiring any specific device or implementation.
Slashdot Mirror
The actual remarks by Commissioner Clyburn are a lot less technical than the SL summary implies. What she wants doesn't have so much to do with the physical deployment of broadband (the "last mile"), as it does with actually adopting the broadband that's available (what she called the last "half-mile".)
Apparently, 93 million Americans don't use broadband, even though they could. Why? The Commish says its because 1) broadband costs too much, 2) non-users lack "digital literacy", and 3) non-users just don't see the benefit. Her proposed fixes have a lot more in common with the Peace Corps than with the White Spaces Coalition: provide free broadband to the poor, and form a National Digital Literacy Corps to deliver a National Digital Literacy Program while going home-to-home to help set up broadband. Her model is the recent national switch-over to digital TV, where hundreds of thousands of volunteers went around and hooked people's TVs up.
I purchased four XO-1s when they originally came out a couple years ago. I gave away two, boxed one for posterity, and am still using one for browsing (Opera) and note-taking (Zim) when I'm at conferences. I still get heads-turns and kids inching over to take a look over my shoulder everywhere I go.
The XO-1 has an early version of the Pixel Qi screen, and it is extremely functional. I'm still amazed every time I'm reading an ebook on the subway, and walk from the deep darkness of the subway tunnel into blinding, direct sunlight, and the XO-1 display is still completely readable.
The XO-1's processor, however, is quite slow, and that becomes a pain in the neck for browsing. A decently-performing netbook doesn't cost very much these days, but the screens are a disappointment. I'm really looking forward to snagging a Pixel Qi DIY kit, buying a cheap netbook, and fixing up my ride.
Bring it, Mary Lou!
One thing that Courier nails is the concept of using both fingers and pen. Go ahead, try it. There are two basic pen positions: a "writing" position that uses all five fingers, and a "resting" position where the pen is rotated 90 degrees, and held in place by a single finger, leaving the other fingers free. The writing position is vertical, resting on the edge of the hand. The resting position is horizontal, palm down.
The Courier UI mockup uses both of these hand configurations and orientations. Flat, horizontal motions such as flipping a page or image dragging are done in the resting position. Vertical motions such as drawing and writing are done in writing position. Switching between the two is very fast and natural-feeling.
Having a pen dispenses with the need for a QUERTY keyboard, but block-printing is not the solution either. For one thing, it's too slow: the average printing speed is about 15wpm. A better solution might be a stylus-based keyboard. Several years ago, IBM invented a shorthand named Shark (commercialized as ShapeWriter, I believe) that was extremely effective. After just a few minutes of practice with it, I was able to achieve 40wpm.
Amazon's DMCA procedures are very deliberative. If you don't like youtube's enforcement policies, shop around! Host your file on S3, and post a video preview on youtube with a link to S3.
For example, Amazon's procedure requires an electronic or physical signature on the takedown notice and a statement made under _penalty of perjury_ that the notice is accurate and that you are the copyright owner or authorized to act on the copyright owner's behalf. This won't slow down a legitimate owner or agent much, but it will give you and Amazon a basis to sue if the takedown notice doesn't come from an owner or authorized agent. (Personally, I suspect that a fair number of takedown notices are not coming from people who have been properly authorized. People claim copyright on all sorts of things that aren't theirs to claim at all -- even large publishing businesses who ought to know better.)
Just read the MS document, and concur with other comments here. Knowing the document's contents actually RAISES my level of trust, since it lays out precisely the conditions under which private information will be revealed (basically: go through the court system), and what information is available.
All companies who deliver cloud-based services should reveal this information to their customers. There is absolutely no reason to keep it secret. Microsoft, Google, Yahoo: man up, and make it official.
Seeing how Network Solutions is utterly mis-handling this situation is destroying my trust in them by the bucket-load. I am seriously considering moving my domains off of them. I certainly wouldn't want to be strong-armed just because I made some big company like MS mad!
A simple system can be specified with "rough" UML models, that's true. That is precisely one of the design goals of the UML language, in fact, to be fast and efficient replacement for human language.
However, an elaborate system needs much more elaborate specifications. It won't be enough to say, "I want a table with a couple of joins, here's a rough sketch", you will need to be much more explicit about fields, data types, keys, relations, triggers, and all the rest.
It won't be enough to say, "This system will call that service, here's a rough sketch", you will need to be much more explicit about data contracts, protocols, end points, delegations, object mapping, and all the rest.
UML is just as fast and efficient at capturing these kinds of details as it is at sketching out the big picture. However, many developers get stuck on the "rough sketch" level of UML, and never go beyond. They wind up sticking their specification details into some code editor, where no one else on the development team can see them except another developer.
However, a detailed spec is just as valuable to a tester, business analyst, security analyst, and database engineer as it is to the developer.
Share the wealth!
The Rock is trading at $62.95 this week. Hurry up and buy!
Don't know if they're still available, but when I went out on my own for a few years, I opted to pay most of my health care out-of-pocket. I bought a high-deductible, catastrophic insurance plan that would cover any medical expense that would have bankrupted my family.
It was a wonderful experience. The catastrophic insurance rates were very low. Over three years, the amount I spent out-of-pocket was about 70% of the amount I had been spending on an employer-supported plan. There were no forms to fill out. No paperwork, except at tax time, when I could claim a medical deduction for the only time in my life. We could visit any doctor we wished. Some doctors even gave us a small discount for paying up front.
In my experience, my family has never needed more than a tiny fraction of the health care services available in the employer plans I've been forced to participate in. I wouldn't mind moving permanently to a high-deductible, catastrophic system at all.
Programming is a young person's game. In the beginning, it's exciting to have this fantastic machine bow to your every command. After 20 years, however, I had done everything I'd ever wanted to do in programming. I was more than happy to let the next generation take over.
What I became interested in was moving up a level in abstraction, and I got into modeling and architecting designs rather than executing them. Seems like a fairly natural progression to me, so yeah, I'm not surprised that the number of developers 40+ is small.
Hmm. I'd like to call a mild foul on this statement.
Are you saying you look at the UML models, ignore them, and then just talk to your users, or that you look at the models, and based on what you've gleaned from them, talk to your users, get to the heart of the matter, and start coding?
If it's the former, then why bother even looking at the models in the first place? They're messed up, as you say, so ditch 'em.
But if it's the latter, then at least part of the reason why you're able to be so effective is because you're using somebody else's work as your starting-point. Hardly seems fair to say they're "fuckups", when you're basing your work on theirs.
Religion has a huge impact on many aspects of society: language, culture, politics -- even science. Religion could certainly be a legitimate topic of academic study, done properly. For example, I doubt it is possible to truly understand the history of the United States without understanding the role of religious belief. It's just too intertwined.
Your point about people trying to pass religion off as if it were science is well taken, however. Bugs me when people try to pass humanism off as science, too.
Interesting how so many ads were trying to put the viewer into a desired mood, and then associate it with a product (the TV ads, the car ads, Coke, etc.), while Google's ad was just a straight-forward story. The Google sequence of search terms were exactly like frames of a comic, where the viewer fills in the storyline inferred by the sequence. (Having just read Understanding Comics by Scott McCloud, I now know that technique is called "closure" -- thanks, Scott.)
Telling a story instead of setting a mood really highlights the Google esthetic: function, always function, before form.
This is surely one of the most frightening and excruciating conditions one can imagine. My guess is that one would compensate somehow, but the loneliness and helplessness must be crushing. If I ever become vegetative, I would certainly want to be given the kind of tests outlined here. If there is no response, then I would like to be taken home and cared for until nature takes its course. The only reason I would want to stay in a hospital or care facility is if there is a chance I can improve.
You know, a lot of the UI features of the iPad don't look that revolutionary. Looks almost like a variant of Moblin to me. How hard would it be to create a Linux desktop manager that duplicates the functionality of the iPad? Shoot, it wouldn't even be that hard to go several features better (multi-tasking, daylight-readable screen, video camera, etc.)
It's great that Apple has put their vision out there, but it looks like when all is said and done, they're betting on their media tie-ins to keep their dominant hipster status. The special sauce doesn't seem to be in the software or the hardware of the iPad.
So, while Apple is busy trying to wrangle exclusive deals with Big Content, other smartbook vendors and the FOSS community can be busy analyzing the design choices of the iPad, and dreaming up an even better vision.
I believe the authors' point is that even though the errors do get replicated as you say, the resulting proteins created by the code are often unaffected by them. It's the resistance to error that's hard for Darwinian evolution to explain.
From the article:
Interesting.
The thought was that China would be replaced with another developing country hungry for trade with the US. Don't really know much about this, but India comes to mind, along with any of the newly-democratic countries in East Europe or South America. Obviously couldn't happen overnight...
Value systems are properties of personal conscience, not of culture. As individuals, we cannot be "forced" to believe something against our will. You can be forced to _act_ like you believe something you don't (as in the book, 1984), but that would just be an act. You wouldn't really believe it.
Since values belong to persons, and persons make up cultures, then values precede culture, and in many ways, even transcend them. For example, If I value free speech and live in the US, I don't think I would value it any less if I lived in China. I may let my culture influence me in deciding which values I hold, but in the end, they're my values, no one else's.
It isn't valid to charge someone with arrogance merely because they state what they believe. If you tell me what you believe, you aren't coercing me in any way. You're just expressing yourself.
You're probably right. Then again, we are talking about a country that feels the need to filter what every single person in the whole blessed country reads every moment of every day on the Internet...
The thought was, China will suffer a major loss of prestige if Google goes through with this. Other companies will follow suit, leading to loss of access, loss of influence, loss of opportunity, and ultimately, loss of business.
China will lash out with wounded national pride, as they seem wont to do. This will further alienate them from the international community, leading to further loss of status.
Loss of prestige will encourage civil unrest, something China dreads. In some places, China seems to be a tinderbox, just waiting to catch fire. They are suffering from the global recession just as much as anyone.
That is a whole lot of pain that China doesn't need right now.
One fascinating aspect of this story is how Google, just a private corporation, is able to credibly threaten an entire country -- and a near-superpower one, at that! That used to take the kind of might only a government could wield.
No longer.
The web levels everyone -- and I mean EVERYONE -- to one, lowest common denominator: access.
over at PC World. Actually, I like the idea of the XO-3. Sure, it's totally blue-sky, but it's great to have at least one outfit taking a completely clean-slate design approach to mobile computing.
I like the hinged-panel XO-2 and MS Courier better, however. I think it's just more practical to have one part of the screen that can tilt up into the light. That said, the ring thingy of the XO-3 is interesting, too. I hadn't really thought about the mechanics of trying to hold a panel with one hand while touching with the other.
Remember 10/GUI, Clayton Miller's 10-fingered touch screen interface? Imagine a flexible 10/GUI touch pad that could be pulled out from under the XO screen. That might be interesting.
In an urban setting, the Segway was, and still is, an obviously better way to move people than a car. The fact that it couldn't get permission to use the sidewalks is more a chicken-and-egg thing than it is a fault of the basic premise. If there were 1,000 Segway-like vehicles on the streets per day in a downtown area, you can be sure that the local laws would be modified to allow them.
One of the biggest problems with mass transit is it doesn't have a good feeder system. We go straight from an arterial system (mass transit) to a capillary system (foot traffic), with nothing in between. Someday someone will fill this gap. The Segway was at least a plausible candidate to do the job.
Patent US5838906
Abstract:
In other words, the patent is on the entire concept of embedding objects in a browser. I think this illustrates perfectly some of the faults of software patents: 1) It is a concept for an invention, not an actual invention; 2) It is a re-statement of general practices and patterns (remote procedure call; client/server; interactive user interface) that only looks new because it is being re-applied to another technology (browsers, in this case); 3) It is over-broad in scope, covering not a particular invention but an entire class of inventions; 4) It is general in execution, not requiring any specific device or implementation.