Does anyone have any experience with the searching and indexing capabilities in these devices?
I want to store my technical library in one of these, but have been resistant to switch, because I want searching and indexing that allows me to use this device as a decent replacement for that 'Technical Library Search' usecase, where someone asks some questions, and I can use this device to do one search and have the results, across all the documents that i have stored, displayed in a results list. for easy access.
Does anyone have any experience with these functions in a variety of readers?
This strikes me as time consuming to have to think the letters to type a word. I want to be able to think the word and have it appear. When do we get a semantic, bi-directional neural interface?
Think about this: When a person starts to think about a document I bet there is a planning part of my brain that is forming an outline of the document, before I even start to come up with the actual content in it. I'd LOVE to tap into that planning, and be able to lay out an outline, just by thinking about it, and then be able to fill that outline with content through thought alone.
I have this software running on my phone, and it does work. What stuns me is that while this thing is in 'beta' and returning poor search results, they have the opportunity to 'train up' the AI, while also keeping hold of a bevy of images that they collect from a few thousand (or hundred thousand) phones that geeks like us were willing to install it on...
I bet that the corpus of images they collect during the next 4 years - the beta period - will be pretty impressive, and kind of scary. I bet that they claim rights on all of them. I guess we need to start watermarking the photos from our phonecams.
It really is about time that AT&T had motivation to actually upgrade their network so that it is usable. Considering the disparity between our infrastructure, and most of the rest of the world, I think this is progress that is along time coming.
I also think that it is very similar to the responses that AT&T had when DSL became a reality. Here in the Midwest, Southwestern Bell sat on DSL for YEARS before actually building it into their network. And they made close to the same excuse that AT&T is making about this.
Silverlight and flash are two evil things on the web. Time after time, I download the software, and run the installer, only to have NOTHING HAPPEN.
Either that, or the plugin does not install into all my browsers, just the dominant one for that OS - it installs into IE, and Safari, but not Firefox on either platform.
What good is this type of thing anyway? Sure it provides a framework for fewer roundtrips to the server, but if it doesn't work, you're right back where you started.
The BIOS is meant to handle the bare metal parts of the IO of the machine. There are now efforts to put applications in the BIOS - MCE machines, faststart, etc.
But to launch an application, the faststart (or whatever) needs to know what filesystem is in use, and what libraries the application relies upon... Oh yeah, theres not even the video subsystem - lets assume that we can load that part, then you've got GUI presentation layer to deal with... The App does not draw its own screens, I hope...
I just see this as being fraught with problems.
At the same time, what if you did incrementally put services in the BIOS, like what if you put 3 different schedulers in the BIOS, and gave the user the choice which one to use. Then the OS would have to detect whether or not the BIOS implemented this, and use it if there, and if not, fallback to whatever is configured in the OS.
The more I think about doing this, the more frought with problems it seems to be. I'm not so sure that a bare metal scheduler is what you want in the OS, you might want to register software services to use it, and perhaps a BIOS scheduler would not be as flexible as the one in the OS...
I applaud the IDEA of taking more functions from the OS. But the how is not so clear to me yet.
The idea itself is not bad - distill down just the most recent changes to a privacy policy, then show these changes to the end user, and ask for their consent.
If they dont consent, then either cancel them then and there, or give them time to migrate there data. Also notice that the patent says that if you dont agree, the application will not be able to use that information that is part of the privacy policy change.
I also agree that is not a patent. It's far too obvious.
People dont know the difference between the location input control, and the search input control. More than half the users I've seen, will type www.whatever.com in the search input, not understanding what they are doing.
I'm not sure what the remedy is here. There's an obvious difference between submitting a query, and typing a location, but the difference is clouded for the end user, because the search engines still display a result.
Perhaps that engines should stop responding to searches that contain fqdns? Like that's going to work... Perhaps the location bar should just merge with the search bar, and if the whatever you type is a valid web page, then the browser puts a link at the very tippy top of the list of results, and also displays the results from your favorite engine.
I find it really disturbing that this article got printed. It does not bode well for the art, or the technology of robotics.
This article is basically the precursor to Bill's Letter to Robot Hobbyists about how they should stop stealing his software, and actually pay for something, goddamnit.
Bill wants us all to know that he's working on this, and has the patents to prove it...
When you think about it, the two pieces of software that he talks about come down to not much more than decent message passing. Maybe there's something to be said for running a driver in its own wrapper, and passing events to it, but I bet someone has thought of that before, and done it too.
I think I do agree with him that the industry is a fragmented, but I dont really know much about it at all, and so I just have a perception that each manufacturer, or even each of their customers has their own toolset, and they use those tools because they are used to them.
Good things would happen if there were several competing toolsets that each worked with all the different hardware, so that customers had some choice...
Maybe someone in the slashdot crowd would care to comment on the state of robotics, control, and software development.
Myself, this article does not bode well for me, I think Bill will use this to spread more FUD, and eventually send out the dreaded letter to the hobbyists...
This doesn't matter. What it does do is allow others to play in the iTMS space. I think that's a good thing.
It's just like someone will crack the Zune, if they have not already. Any standalone device with a form of encryption, DRM, or security will eventually become crack'd.
I've been predicting this for some time, except that It wont be eye implants, it'll be a chip in your visual cortex - The Microsoft Vision Subsystem Cortex Intellectual Property Analyzer.
It will be able to blue-screen, or blur (ala Cops) any content that you have not purchased a license to see.
Thanks to all the genome research that's being done, the public key will be your DNA. If you lose your license key... Well, you get the idea.
I am a Mac User, but I have a long history of UNIX behind me.
When confronted with a modern mac user, I am amazed at the level of complacency they display. They literally do not understand the responsibility they bear.
Hey, at least Windows users understand that they have insecure systems. They might not be able to get at the magnitude, but they'll tell you: "Yes, that thing pops up all the time, I always just cancel it..." -- sigh...
But Mac users -- "Why would i need a firewall?"
-- All this said, I'm a die hard Mac user. I know where the vulnerabilities are. But that's because I have REAL understanding of the systems that I have to use.
Apple has had support for extended meta data for years. It's not stored in any SQL database, but you can create arbitrary attribute value pairs for any file. Right now, you can do this from the command line.
I think that Apple has chosen, wisely, NOT to do anything with this. They have a really great R&D lab there, there must be a reason that they've never exposed this functionality for an end user. I bet it's just too complex for a user. Who wants to tag the files we create? So you only get the benefit if YOU ACTIVELY do it. What if you just dont understand it?
I understand the power of having fully user editable meta data, but there are just some times when you dont want an end user messing with things like that.
I think that's why apple lets you tag files with a label. It's just simpler, and users can understand it.
There was a 95 car pile up on the 101 this morning, as EM emmision from a sunspot caused a problem in the cars driving there.
All the drivers were killed by the Blue Windscreen of Death, as the 'Microsoft Collision Avoidance Service' stopped functioning (crashed). MCAS is a new component in the 'Windows 2005 for Car' edition.
Bill Gates was asked if there are any problems with the new system. He is quoted on the event... "All these problems happen because of third parrty software that is downloaded to our systems. Windows Car is a flagship of innovation that gives our users real choice, a great user experience during the driving time in their day."
You'd have to boil some water in a pan before some Linux or ReiserFS programmer comes along and makes this filesystem a moot point.
This article doesn't say anything about the things that most users and OS/applications care about. What about extended meta-data? What about better indices?
Let me guess, the 128bit fs takes up so much room, that I cant fit extra meta-data on the filesystem.
I've wanted to say this about Linux vs. Sun for a while: The main strength that Linux has is that UFS is a dog, and developers are writing filesystems for Linux that do AMAZING things - libferris anyone? ReiserFS is still my favorite, I use it all over the place.
With Sun, you dont get the innovation that you get from Linux, because they dont let at the innards of their filesystem interface. Sure veritas is cool - but that's ONE out of say 5 or 10 available for Linux.
The one thing that is cool in the article is the storage reservation capability. That would be a nice feature to have.
I wonder when someone will patent procedures involved in bringing suit against someone?
Microsfots lawyers will patent their procedure for responding to complaints, and then bring suit against other law firms that match that...
sigh...
Records in databases are not facts
on
Who Owns The Facts?
·
· Score: 2, Interesting
I've read the bill, and I agree with those that dont see the harm here. It seems to me that what this bill is trying to protect is the investment in the collection, storage and maintenance of data.
The fact that some database contains my home address is not related to the fact that my home address is correct in that database. I guess what I'm saying is that my address never will be a primary key for whatever. It is not an absolute. But it is a fact. The owner of that data does not own that fact, they just own a character string. They cannot claim to own my address. If they do, I can easily move.
The fact is that every Atom has an atomic weight. That is a fact. Do we know those atomic weights with any great precision? No - we can make approximations. But we dont have those facts. So, noone can own them.
The fact is that most people have a home address, but just because I think I know what it is at a given moment does not make it a fact, and publishing that information in a database does not mean that I own it.
I guess I just dont get how this is a question of who owns the facts.
Someday, after the grand unification theory has been discovered, invented, whatever you want to call it - I bet someone will try to publish a database of real, actual facts - immutable properties of the universe. They may try to claim ownership of them. I hope they get the shit knocked out of them when they try.
The first thing I did was polish up the resume. and start sending it around.
I decided that the shotgun approach was best, so I posted it to monster, and then contacted all my friends looking for good recruiters.
I decided to pull no punches. And I got confronted with alot of people who would rather place me than apy attention to what my qulaifications were, and how right I was for the job.
Ultimately, none of them got me a job, it took an old friend to do that, but they did get me some good interviews.
Plus, I learned a good lesson - it's not on them to be selective - that part is all you. So if they hit you with a position you dont like, just ignore it.
It is a Market, you know...
It took me three months to find a job, by the way, alot longer than I wanted, but still seemed like record time WRT others that had gotten the same deal.
A programmer is someone who is able to solve a given business problem, using creativity, and their knowledge of algorithms and data structures.
An Engineer designs solutions that, within a range of certainty will not fail. Note that this range is never achieving 100% Most engineers that have to solve problems such as bridge construction can vouch for this, as can software engineers.
A Scientist searches for and evaluates new solutions to problems in the hopes of find the one true solution to a problem. This is why they use the scientific method.
Engineers dont use the scientific method, it takes too much time. But they dont have a set of rules and strict limits that they work with either. What they primarily work with is methmatical models which indicate the chance of success/failure of any given design. They are able to be creative, adn suggest approaches which might improve those chances.
Really to be an engineer, I think that what is required mostly is real world experience. The only way to learn how to evaluate a problem is sit down and confront it. That doesn't take book learnin' It takes experience.
Slashdot Mirror
I was in the AI lab at MIT, testing my wits against LISP. In walks Marvin Minsky.
I asked him if he could give me a tip or two about atoms.
His response to me was: "Well, why dont you wait until the computer speaks your language... Then program it in that?"
That was alot longer ago than 15 years...
Does anyone have any experience with the searching and indexing capabilities in these devices?
I want to store my technical library in one of these, but have been resistant to switch, because I want searching and indexing that allows me to use this device as a decent replacement for that 'Technical Library Search' usecase, where someone asks some questions, and I can use this device to do one search and have the results, across all the documents that i have stored, displayed in a results list. for easy access.
Does anyone have any experience with these functions in a variety of readers?
Thanks.
Well,
This strikes me as time consuming to have to think the letters to type a word. I want to be able to think the word and have it appear. When do we get a semantic, bi-directional neural interface?
Think about this: When a person starts to think about a document I bet there is a planning part of my brain that is forming an outline of the document, before I even start to come up with the actual content in it. I'd LOVE to tap into that planning, and be able to lay out an outline, just by thinking about it, and then be able to fill that outline with content through thought alone.
Imagine applying this to code generation!
I have this software running on my phone, and it does work.
What stuns me is that while this thing is in 'beta' and returning poor search results, they have the opportunity to 'train up' the AI, while also keeping hold of a bevy of images that they collect from a few thousand (or hundred thousand) phones that geeks like us were willing to install it on...
I bet that the corpus of images they collect during the next 4 years - the beta period - will be pretty impressive, and kind of scary. I bet that they claim rights on all of them. I guess we need to start watermarking the photos from our phonecams.
Just my thoughts...
It really is about time that AT&T had motivation to actually upgrade their network so that it is usable. Considering the disparity between our infrastructure, and most of the rest of the world, I think this is progress that is along time coming.
I also think that it is very similar to the responses that AT&T had when DSL became a reality. Here in the Midwest, Southwestern Bell sat on DSL for YEARS before actually building it into their network. And they made close to the same excuse that AT&T is making about this.
Not amazed, not amused, just waiting...
Oh yeah, an Android phone would help to AT&T...
Silverlight and flash are two evil things on the web.
Time after time, I download the software, and run the installer, only to have NOTHING HAPPEN.
Either that, or the plugin does not install into all my browsers, just the dominant one for that OS - it installs into IE, and Safari, but not Firefox on either platform.
What good is this type of thing anyway? Sure it provides a framework for fewer roundtrips to the server, but if it doesn't work, you're right back where you started.
Frustrating.
Thank god for a really good argument, and comment on this post.
Thank you!
The BIOS is meant to handle the bare metal parts of the IO of the machine.
There are now efforts to put applications in the BIOS - MCE machines, faststart, etc.
But to launch an application, the faststart (or whatever) needs to know what filesystem is in use, and what libraries the application relies upon... Oh yeah, theres not even the video subsystem - lets assume that we can load that part, then you've got GUI presentation layer to deal with... The App does not draw its own screens, I hope...
I just see this as being fraught with problems.
At the same time, what if you did incrementally put services in the BIOS, like what if you put 3 different schedulers in the BIOS, and gave the user the choice which one to use. Then the OS would have to detect whether or not the BIOS implemented this, and use it if there, and if not, fallback to whatever is configured in the OS.
The more I think about doing this, the more frought with problems it seems to be. I'm not so sure that a bare metal scheduler is what you want in the OS, you might want to register software services to use it, and perhaps a BIOS scheduler would not be as flexible as the one in the OS...
I applaud the IDEA of taking more functions from the OS. But the how is not so clear to me yet.
I agree with this post.
The idea itself is not bad - distill down just the most recent changes to a privacy policy, then show these changes to the end user, and ask for their consent.
If they dont consent, then either cancel them then and there, or give them time to migrate there data. Also notice that the patent says that if you dont agree, the application will not be able to use that information that is part of the privacy policy change.
I also agree that is not a patent. It's far too obvious.
People dont know the difference between the location input control, and the search input control. More than half the users I've seen, will type www.whatever.com in the search input, not understanding what they are doing.
I'm not sure what the remedy is here. There's an obvious difference between submitting a query, and typing a location, but the difference is clouded for the end user, because the search engines still display a result.
Perhaps that engines should stop responding to searches that contain fqdns? Like that's going to work... Perhaps the location bar should just merge with the search bar, and if the whatever you type is a valid web page, then the browser puts a link at the very tippy top of the list of results, and also displays the results from your favorite engine.
I dunno...
I find it really disturbing that this article got printed. It does not bode well for the art, or the technology of robotics.
This article is basically the precursor to Bill's Letter to Robot Hobbyists about how they should stop stealing his software, and actually pay for something, goddamnit.
Bill wants us all to know that he's working on this, and has the patents to prove it...
When you think about it, the two pieces of software that he talks about come down to not much more than decent message passing. Maybe there's something to be said for running a driver in its own wrapper, and passing events to it, but I bet someone has thought of that before, and done it too.
I think I do agree with him that the industry is a fragmented, but I dont really know much about it at all, and so I just have a perception that each manufacturer, or even each of their customers has their own toolset, and they use those tools because they are used to them.
Good things would happen if there were several competing toolsets that each worked with all the different hardware, so that customers had some choice...
Maybe someone in the slashdot crowd would care to comment on the state of robotics, control, and software development.
Myself, this article does not bode well for me, I think Bill will use this to spread more FUD, and eventually send out the dreaded letter to the hobbyists...
The headline has it right.
This doesn't matter. What it does do is allow others to play in the iTMS space. I think that's a good thing.
It's just like someone will crack the Zune, if they have not already. Any standalone device with a form of encryption, DRM, or security will eventually become crack'd.
I've been predicting this for some time, except that It wont be eye implants, it'll be a chip in your visual cortex - The Microsoft Vision Subsystem Cortex Intellectual Property Analyzer.
It will be able to blue-screen, or blur (ala Cops) any content that you have not purchased a license to see.
Thanks to all the genome research that's being done, the public key will be your DNA. If you lose your license key... Well, you get the idea.
Come on, why hasn't someone made Duke Nukem Forever into a movie?
Could that ever suck?!
I dont think so.
In the data centers I've seen, The only things you could install were Sun servers with DC power supplies. I was shocekd, because I'd never seen one...
That was about 5 years ago. So I think this has been a thing for while now.
dave
I have to agree here.
I am a Mac User, but I have a long history of UNIX behind me.
When confronted with a modern mac user, I am amazed at the level of complacency they display. They literally do not understand the responsibility they bear.
Hey, at least Windows users understand that they have insecure systems. They might not be able to get at the magnitude, but they'll tell you: "Yes, that thing pops up all the time, I always just cancel it..." -- sigh...
But Mac users -- "Why would i need a firewall?"
-- All this said, I'm a die hard Mac user. I know where the vulnerabilities are. But that's because I have REAL understanding of the systems that I have to use.
Thanks.
Dave
Apple has had support for extended meta data for years. It's not stored in any SQL database, but you can create arbitrary attribute value pairs for any file. Right now, you can do this from the command line.
I think that Apple has chosen, wisely, NOT to do anything with this. They have a really great R&D lab there, there must be a reason that they've never exposed this functionality for an end user. I bet it's just too complex for a user. Who wants to tag the files we create? So you only get the benefit if YOU ACTIVELY do it. What if you just dont understand it?
I understand the power of having fully user editable meta data, but there are just some times when you dont want an end user messing with things like that.
I think that's why apple lets you tag files with a label. It's just simpler, and users can understand it.
I'm surprised that SNMP, and support for the toaster MIB is not included.
Wouldn't this be an essential part of functionality...
There was a 95 car pile up on the 101 this morning, as EM emmision from a sunspot caused a problem in the cars driving there.
All the drivers were killed by the Blue Windscreen of Death, as the 'Microsoft Collision Avoidance Service' stopped functioning (crashed). MCAS is a new component in the 'Windows 2005 for Car' edition.
Bill Gates was asked if there are any problems with the new system. He is quoted on the event... "All these problems happen because of third parrty software that is downloaded to our systems. Windows Car is a flagship of innovation that gives our users real choice, a great user experience during the driving time in their day."
You'd have to boil some water in a pan before some Linux or ReiserFS programmer comes along and makes this filesystem a moot point.
This article doesn't say anything about the things that most users and OS/applications care about. What about extended meta-data? What about better indices?
Let me guess, the 128bit fs takes up so much room, that I cant fit extra meta-data on the filesystem.
I've wanted to say this about Linux vs. Sun for a while: The main strength that Linux has is that UFS is a dog, and developers are writing filesystems for Linux that do AMAZING things - libferris anyone? ReiserFS is still my favorite, I use it all over the place.
With Sun, you dont get the innovation that you get from Linux, because they dont let at the innards of their filesystem interface. Sure veritas is cool - but that's ONE out of say 5 or 10 available for Linux.
The one thing that is cool in the article is the storage reservation capability. That would be a nice feature to have.
I wonder when someone will patent procedures involved in bringing suit against someone?
Microsfots lawyers will patent their procedure for responding to complaints, and then bring suit against other law firms that match that...
sigh...
I've read the bill, and I agree with those that dont see the harm here. It seems to me that what this bill is trying to protect is the investment in the collection, storage and maintenance of data.
The fact that some database contains my home address is not related to the fact that my home address is correct in that database. I guess what I'm saying is that my address never will be a primary key for whatever. It is not an absolute. But it is a fact. The owner of that data does not own that fact, they just own a character string. They cannot claim to own my address. If they do, I can easily move.
The fact is that every Atom has an atomic weight. That is a fact. Do we know those atomic weights with any great precision? No - we can make approximations. But we dont have those facts. So, noone can own them.
The fact is that most people have a home address, but just because I think I know what it is at a given moment does not make it a fact, and publishing that information in a database does not mean that I own it.
I guess I just dont get how this is a question of who owns the facts.
Someday, after the grand unification theory has been discovered, invented, whatever you want to call it - I bet someone will try to publish a database of real, actual facts - immutable properties of the universe. They may try to claim ownership of them. I hope they get the shit knocked out of them when they try.
After applying this update, all my MP3's are now Quicktime Movie files...
When I try to open one, instead of invoking iTunes, my system opens macam - software to drive my webcam.
Very confusing. I need to dig some more...
This is not as bad as my trial with Windows XP this morning... Roaming Profiles SUCK...
I got laid off last year in December.
The first thing I did was polish up the resume. and start sending it around.
I decided that the shotgun approach was best, so I posted it to monster, and then contacted all my friends looking for good recruiters.
I decided to pull no punches. And I got confronted with alot of people who would rather place me than apy attention to what my qulaifications were, and how right I was for the job.
Ultimately, none of them got me a job, it took an old friend to do that, but they did get me some good interviews.
Plus, I learned a good lesson - it's not on them to be selective - that part is all you. So if they hit you with a position you dont like, just ignore it.
It is a Market, you know...
It took me three months to find a job, by the way, alot longer than I wanted, but still seemed like record time WRT others that had gotten the same deal.
A programmer is someone who is able to solve a given business problem, using creativity, and their knowledge of algorithms and data structures.
An Engineer designs solutions that, within a range of certainty will not fail. Note that this range is never achieving 100% Most engineers that have to solve problems such as bridge construction can vouch for this, as can software engineers.
A Scientist searches for and evaluates new solutions to problems in the hopes of find the one true solution to a problem. This is why they use the scientific method.
Engineers dont use the scientific method, it takes too much time. But they dont have a set of rules and strict limits that they work with either. What they primarily work with is methmatical models which indicate the chance of success/failure of any given design. They are able to be creative, adn suggest approaches which might improve those chances.
Really to be an engineer, I think that what is required mostly is real world experience. The only way to learn how to evaluate a problem is sit down and confront it. That doesn't take book learnin' It takes experience.