"Instant-on" PCs aren't new. Take a look at the Soyo SY-P4VAL version M (I think this was on Slashdot before but I don't have a link). The built-in BIOS "media center" software lets you play MP3 CDs, audio CDs, VCDs, DVDs, and watch TV. It's only ~$130, as opposed to $2500 - $2600.
Granted, this article is about a laptop with instant-on capabilities, which is of course cooler and more expensive.
The physical limits of CD burning (the speeds that often cause shattering) aren't because CDs are easily breakable. It's because CDs are imperfectly manufactured, and therefore imbalanced - a CD spinning at 52x that isn't perfectly round will be wobbling with an incredible amount of force.
So... for DVDs to be able to spin at faster speeds, the discs (and drives) will have to be manufactured to very high specs. Very slight variations in the roundness of the disc would cause enough vibration to break the disc. A non-round or off-center hole in the middle would also cause this problem.
The browser wars of the 90's are over. Nobody is "selling" their browsers for their proprietary features. This is why you don't see many (well, not that many) IE-only pages any more - people want to be compliant.
Microsoft's Internet Explorer is too old. Features that almost every other browser has, like tabbed browsing, skins, etc. are not included, and there are so many holes it's like Swiss cheese.
Microsoft isn't pursuing it because there's no money in the browser market. As the article says, Apache is free, HTTP is free, most browsers are free, PHP, Perl, HTML, MySQL, and almost everything Internet-related is completly free (not always as in speech, but free nonetheless). Microsoft has no motivation to make an amazing browser, because it doesn't get them anything but a name (which they already have).
Over the next few years, the only good browsers will be coming from groups like Mozilla who aren't in a money-making business at all and only want to have a great, stable, secure, fast, and standards-compliant browser. They don't want to necessarily dominate the browser market (though I'm sure they'd love that) - they just want to make a good product.
That is why the browser wars are over. The good browsers will rise, the bad ones will fall - and the good browsers will only come from developers who are in it for "the cause" and not the money.
...was filling his car with gasoline... Which is why NJ is the safest place to buy gasoline. In NJ, it's illegal to pump your own gas, so the attendant has to do it for you. That way, when his cell phone rings, you'll be safely inside your car when it explodes:)
Exposé is exactly what I think we need. Sphere XP is doing the same general thing... you see all the windows at a glance, click the one you want, and it's brought to the foreground. The problem is that a 3D world is too hard to navigate efficiently.
Enter... iEx - an Exposé clone for Windows. It doesn't appear to have very active development but it works pretty well (I've only been using it for the past five minutes). It only runs on Windows XP.
Just downloaded Sphere XP from the PlanetMirror site mentioned in this post. It looks pretty cool, but there are a couple features it needs to become something useful rather than just a neat demo.
First, there needs to be some way of changing the windows' distance from the camera. With the current design, the windows are really small until you click on them, and then - WHOOSH (that's annoying, anyone who's used it knows what I mean) - they're huge and 2D again.
Second, we need a way to use the windows when they're "floating". Currently, you have to restore a window to its original 2D state if you want to be able to actually use it.
An interesting idea just popped into my head - since desktop computing is [still] meant to be 2D, why not use some of this guy's ideas in a 2D manner? Specifically, I'd like to see window scaling - you could resize a window so that the entire interface was scaled down: smaller text, smaller widgets, everything. It'd be just like if you took a screenshot of the window and scaled it down in Photoshop. Maybe this could be operated by using the right mouse button on the window's standard resize handles? Who knows. Does anyone know of anything like this that exists?
I don't understand why nobody gets this - Gmail is a service, not a requirement! It is not mandatory that everyone in the world signs up for Gmail. For crying out loud, it's free! If you like it, use it; if you don't, then nothing is stopping you from not signing up.
All spam filters "read" your email. AOL, Hotmail, anything with SpamAssasin, any service with spam protection needs to "read" messages to analyze them.
Oh, and about this: ..."residual copies of email may remain on our systems for some time"...
They use computers with hard drives! They can't guarantee that data is completely shredded. I'm sure they're not performing a secure wipe of every sector containing portions of an email once it's deleted.
If you started looking, most of the privacy "concerns" with Google's service apply to almost any email service. It's a huge fuss over nothing.
Think that circuit boards can only be bent by water if they're the cheap cardboard kind? Think again. Any cheap electronic toy's circuit board can be "bent" (in the musical context) by placing drops of water on the board in strategic locations.
Electronic toys have also been known to melt (well, 'bend') when the batteries start running low.
For all you Windows people who can't run a C/C++/Perl/PHP/whatever program, I wrote a nice little binary-to-text converter with a [gasp] GUI written in [gasp] Visual Basic.NET . Needs the.NET framework installed (stupid.NET exe...).
You won't be able to just click the link with IE - you'll have to right-click the link, "save target as..." and put it on your desktop or something.
Not like I really needed a binary-to-text converter, especially a.NET one, but I was bored.
This is kind of long, but everyone I've told it to thought it was pretty funny so I thought I'd share it here.
I was coding a (somewhat simple) PHP application to read data from a textfile and stick it into an HTML table, along with a header and footer to make the page look pretty. It also had an "add" function where a user could add entries to the textfile which would then add a row to the table. It was a program for updating a web page with the latest scores from a sports league. Pretty simple, but it was my first PHP program I'd written from scratch so it took me all day to code, debug, test, code, debug, test, etc.
I finally got it working exactly the way I liked it, so I uploaded the "final" version to the server and ran it one last time. Then I noticed a typo on the submit button. Simple enough, I said. I had been using a Cygwin version of vi to write the script (you'll see why this is important later), but I had already closed all the windows I had open with the exception of my browser. So, I opened up FileZilla (FTP client), connected to the server, and double-clicked the file. Here's where the trouble starts.
When you double-click a file in FileZilla (at least with my configuration) it downloads the file to whatever directory is open in the file browser pane. Normally, if a file with the same name already exists, it'll prompt before overwriting the file. But I had just restored a site from a local backup so I had FileZilla set to automatically overwrite without prompting. The bottom line is that it downloaded the PHP file off the server onto my machine, erasing my local copy of the script. In a normal situation this would have been fine. BUT...
I had FileZilla set up to open PHP files in Notepad - yes, Windows Notepad. You probably know what happens when a *nix textfile is opened in Notepad - it displays nice little "blocks" wherever a line return should be. I didn't feel like getting vi back out, so I used Ctrl+F to search for the typo and correct it. Done. I closed Notepad (after saving the file) and uploaded it back to the server.
Everything should have been fine at this point. However, when I refreshed the page in my browser, I got a syntax error. Argh! I quickly opened the file in Notepad again - and later vi - only to discover that Notepad or FileZilla or something had stripped all the line returns out of my script! It seems to me that the script should have run after this, but for some reason it didn't, and FileZilla had overwritten my local backup.
So, I spent the next hour or so going through my code and pushing Enter wherever it was necessary until I got the code running again. The best part?
I showed the finished program to the client later on and they told me they had decided against using a computer to keep track of scores. Thanks for your effort but we decided we don't want to use your program.
I think a step in this direction has been taken with Google's Personalized Search (see the FAQ for details). You create a profile describing your interests (either in general, or what type of thing you're interested in at the moment, just for this search). You can then adjust via a slider how closely the search results should conform to this profile.
So if you wanted to search for an article on how to weave baskets (I know, weird example) that you had seen in the past, you could tell Google that you're interested in crafts and such but not shopping or historical research. Google would then filter out sites about the history of basket weaving and sites selling woven baskets. That doesn't leave a whole lot of room for anything but articles on how to weave baskets yourself.
I'm sure that'd get you there [almost] as fast as a Google search on your history. You'd probably come across even more relevant information than if you only searched your history. It's just another technique for Google to deliver the most relevant results in the search engine industry.
Sounds like you're referring to a hardware watchdog. This would only reboot the box if the machine hard locked- you wouldn't want that to happen for any other reason.
I'm talking about a software "watchdog" - some sort of daemon that monitors mythfrontend and mythbackend to make sure they're healthy and responsive. If one of them dies or freezes, the watchdog would have to kill one or both of them and then restart them. It should keep a log of its actions and have a limit on how many times it can kill/restart Myth befroe it gives up. If Myth won't restart after, say, 3 tries, it needs to:
(1) figure out why it won't start or freezes. If it's because of some service that has crashed, i.e. sound card driver, capture system, etc., then it needs to be able to identify what the service does and tie it into the error message.
(2) then restart those services and try again. If all else fails, restart the machine. Remember, if someone had a show set up for recording, the point is to get the box back up as soon as possible - without user intervention. It doesn't matter if somebody sitting in front of the box with a keyboard attached could make it work again - it needs to get up and running on its own so that it can continue recording where it left off.
Rememer, the average user isn't going to read log files or anything. If it freezes or dies, it needs to correct the problems on its own and be back up ASAP.
Since MythTV is still in heavy development (currently version 0.14), how are they planning to keep users up to date? I use MythTV on a dedicated box at home, and while it's reasonably solid, every now and then Myth segfaults dropping me to a console or KDE desktop. The average PVR user isn't going to like that.
Maybe there's a "watchdog" on the Myth system that reboots the system if Myth dies. The user won't want to wait for the system to reboot every time that they fast forward too quickly (happens on mine) or 'activate' some other bug.
Take the MediaMVP by Hauppauge. It's a Linux-based device designed for use by the "average user". How do they keep it up to date? Well, very little OS or software is on the unit itself. It netboots off a server you run on one of your PCs. The server (Windows app) checks for updates from Hauppauge's site, and when it gets one, it just loads it to the MediaMVP the next time it boots.
Obviously, you wouldn't want to require every user to have a LAN connection to an always-on home computer for net booting (or would you?). Besides, the whole Linux/Myth suite (remember, we need X too) would be a pretty hefty network transfer. With a "closed" device like this one, it'll be hard for users to install bugfixes, patch security holes, etc. etc. without being experienced in Linux (think power user instead of grandma).
I think the Home Media Center/re is going to need some sort of update method for both Myth and Linux. One little annoying bug would drive me crazy if there was nothing I could do about it.
Add thumbprint scanner on the right-hand side of a certain page in a bundle of pages (a eInk book). Maybe make a cutout at the bottom-right of every page except the middle one, so you can thumb to it quickly. (Your thumb would catch the page without the cutout).
Then all you have to do is pick up the book, and it knows who you are. You could then have it call up your "saved session" with your location in a book, your menu layout, your preferences, etc. Cool!
While e-ink is amazing, it'll never (never say never) replace a paper book. It has lots of wonderful uses, and I have nothing against it. It'll just never quite replace paper.
Take catalogs for example. Sure, you could look up what you want online. I do that almost all the time, and I'd be one of the first to buy a Web-connected e-ink catalog. But when you just want to browse, I can't stand having to click-tap-scroll-push-wheel through the pages. I need to be able to put my thumb on the side of a book, flip through and feel the breeze in my face while watching the pages go by...
Seriously though, until they can think of a better input device to navigate an eBook, they'll never replace paper. I'll get an eBook. You'll get an eBook. We'll all accept whatever's out there, eventually (we'll probably even begin to like DRM!). But IMHO, paper is here to stay.
Slashdot Mirror
"Instant-on" PCs aren't new. Take a look at the Soyo SY-P4VAL version M (I think this was on Slashdot before but I don't have a link). The built-in BIOS "media center" software lets you play MP3 CDs, audio CDs, VCDs, DVDs, and watch TV. It's only ~$130, as opposed to $2500 - $2600.
Granted, this article is about a laptop with instant-on capabilities, which is of course cooler and more expensive.
(that's what the little character looks like, anyway.) Serisouly, it looks like a neat concept though.
So... for DVDs to be able to spin at faster speeds, the discs (and drives) will have to be manufactured to very high specs. Very slight variations in the roundness of the disc would cause enough vibration to break the disc. A non-round or off-center hole in the middle would also cause this problem.
Reminds me of propeller clocks (also here, here, here...)
...or the similar mechanically scanned displays.
Spacewriter sells some very cool full-color displays. Their iBall 3D display is also sold at AudioVisualizers - check their site out for more animated demos.
There's also the Virtual Game System (Google cache) which was amazing; unfortunately the site is down so you'll have to settle for text and no pictures.
The browser wars of the 90's are over. Nobody is "selling" their browsers for their proprietary features. This is why you don't see many (well, not that many) IE-only pages any more - people want to be compliant.
Microsoft's Internet Explorer is too old. Features that almost every other browser has, like tabbed browsing, skins, etc. are not included, and there are so many holes it's like Swiss cheese.
Microsoft isn't pursuing it because there's no money in the browser market. As the article says, Apache is free, HTTP is free, most browsers are free, PHP, Perl, HTML, MySQL, and almost everything Internet-related is completly free (not always as in speech, but free nonetheless). Microsoft has no motivation to make an amazing browser, because it doesn't get them anything but a name (which they already have).
Over the next few years, the only good browsers will be coming from groups like Mozilla who aren't in a money-making business at all and only want to have a great, stable, secure, fast, and standards-compliant browser. They don't want to necessarily dominate the browser market (though I'm sure they'd love that) - they just want to make a good product.
That is why the browser wars are over. The good browsers will rise, the bad ones will fall - and the good browsers will only come from developers who are in it for "the cause" and not the money.
...was filling his car with gasoline... :)
Which is why NJ is the safest place to buy gasoline. In NJ, it's illegal to pump your own gas, so the attendant has to do it for you. That way, when his cell phone rings, you'll be safely inside your car when it explodes
Enter... iEx - an Exposé clone for Windows. It doesn't appear to have very active development but it works pretty well (I've only been using it for the past five minutes). It only runs on Windows XP.
First, there needs to be some way of changing the windows' distance from the camera. With the current design, the windows are really small until you click on them, and then - WHOOSH (that's annoying, anyone who's used it knows what I mean) - they're huge and 2D again.
Second, we need a way to use the windows when they're "floating". Currently, you have to restore a window to its original 2D state if you want to be able to actually use it.
An interesting idea just popped into my head - since desktop computing is [still] meant to be 2D, why not use some of this guy's ideas in a 2D manner? Specifically, I'd like to see window scaling - you could resize a window so that the entire interface was scaled down: smaller text, smaller widgets, everything. It'd be just like if you took a screenshot of the window and scaled it down in Photoshop. Maybe this could be operated by using the right mouse button on the window's standard resize handles? Who knows. Does anyone know of anything like this that exists?
Dance, monkey boy ("Woohoo! I love this company!") (3MB)
and last but not least...
The 3-minute Developers music video (9.5MB)
Perhaps I missed something in the article... but why are the installation screenshots RedHat and the desktop screenshots SuSe?
All spam filters "read" your email. AOL, Hotmail, anything with SpamAssasin, any service with spam protection needs to "read" messages to analyze them.
Oh, and about this:
..."residual copies of email may remain on our systems for some time"...
They use computers with hard drives! They can't guarantee that data is completely shredded. I'm sure they're not performing a secure wipe of every sector containing portions of an email once it's deleted.
If you started looking, most of the privacy "concerns" with Google's service apply to almost any email service. It's a huge fuss over nothing.
Seems somebody rolled out of bed on the wrong side today
Great, let's make him feel better by slashdotting his server...
Electronic toys have also been known to melt (well, 'bend') when the batteries start running low.
You won't be able to just click the link with IE - you'll have to right-click the link, "save target as..." and put it on your desktop or something.
Not like I really needed a binary-to-text converter, especially a .NET one, but I was bored.
I was coding a (somewhat simple) PHP application to read data from a textfile and stick it into an HTML table, along with a header and footer to make the page look pretty. It also had an "add" function where a user could add entries to the textfile which would then add a row to the table. It was a program for updating a web page with the latest scores from a sports league. Pretty simple, but it was my first PHP program I'd written from scratch so it took me all day to code, debug, test, code, debug, test, etc.
I finally got it working exactly the way I liked it, so I uploaded the "final" version to the server and ran it one last time. Then I noticed a typo on the submit button. Simple enough, I said. I had been using a Cygwin version of vi to write the script (you'll see why this is important later), but I had already closed all the windows I had open with the exception of my browser. So, I opened up FileZilla (FTP client), connected to the server, and double-clicked the file. Here's where the trouble starts.
When you double-click a file in FileZilla (at least with my configuration) it downloads the file to whatever directory is open in the file browser pane. Normally, if a file with the same name already exists, it'll prompt before overwriting the file. But I had just restored a site from a local backup so I had FileZilla set to automatically overwrite without prompting. The bottom line is that it downloaded the PHP file off the server onto my machine, erasing my local copy of the script. In a normal situation this would have been fine. BUT...
I had FileZilla set up to open PHP files in Notepad - yes, Windows Notepad. You probably know what happens when a *nix textfile is opened in Notepad - it displays nice little "blocks" wherever a line return should be. I didn't feel like getting vi back out, so I used Ctrl+F to search for the typo and correct it. Done. I closed Notepad (after saving the file) and uploaded it back to the server.
Everything should have been fine at this point. However, when I refreshed the page in my browser, I got a syntax error. Argh! I quickly opened the file in Notepad again - and later vi - only to discover that Notepad or FileZilla or something had stripped all the line returns out of my script! It seems to me that the script should have run after this, but for some reason it didn't, and FileZilla had overwritten my local backup.
So, I spent the next hour or so going through my code and pushing Enter wherever it was necessary until I got the code running again. The best part?
I showed the finished program to the client later on and they told me they had decided against using a computer to keep track of scores. Thanks for your effort but we decided we don't want to use your program.
*sigh*...
So if you wanted to search for an article on how to weave baskets (I know, weird example) that you had seen in the past, you could tell Google that you're interested in crafts and such but not shopping or historical research. Google would then filter out sites about the history of basket weaving and sites selling woven baskets. That doesn't leave a whole lot of room for anything but articles on how to weave baskets yourself.
I'm sure that'd get you there [almost] as fast as a Google search on your history. You'd probably come across even more relevant information than if you only searched your history. It's just another technique for Google to deliver the most relevant results in the search engine industry.
I'm talking about a software "watchdog" - some sort of daemon that monitors mythfrontend and mythbackend to make sure they're healthy and responsive. If one of them dies or freezes, the watchdog would have to kill one or both of them and then restart them. It should keep a log of its actions and have a limit on how many times it can kill/restart Myth befroe it gives up. If Myth won't restart after, say, 3 tries, it needs to:
(1) figure out why it won't start or freezes. If it's because of some service that has crashed, i.e. sound card driver, capture system, etc., then it needs to be able to identify what the service does and tie it into the error message.
(2) then restart those services and try again. If all else fails, restart the machine. Remember, if someone had a show set up for recording, the point is to get the box back up as soon as possible - without user intervention. It doesn't matter if somebody sitting in front of the box with a keyboard attached could make it work again - it needs to get up and running on its own so that it can continue recording where it left off.
Rememer, the average user isn't going to read log files or anything. If it freezes or dies, it needs to correct the problems on its own and be back up ASAP.
Maybe there's a "watchdog" on the Myth system that reboots the system if Myth dies. The user won't want to wait for the system to reboot every time that they fast forward too quickly (happens on mine) or 'activate' some other bug.
Take the MediaMVP by Hauppauge. It's a Linux-based device designed for use by the "average user". How do they keep it up to date? Well, very little OS or software is on the unit itself. It netboots off a server you run on one of your PCs. The server (Windows app) checks for updates from Hauppauge's site, and when it gets one, it just loads it to the MediaMVP the next time it boots.
Obviously, you wouldn't want to require every user to have a LAN connection to an always-on home computer for net booting (or would you?). Besides, the whole Linux/Myth suite (remember, we need X too) would be a pretty hefty network transfer. With a "closed" device like this one, it'll be hard for users to install bugfixes, patch security holes, etc. etc. without being experienced in Linux (think power user instead of grandma).
I think the Home Media Center/re is going to need some sort of update method for both Myth and Linux. One little annoying bug would drive me crazy if there was nothing I could do about it.
Then all you have to do is pick up the book, and it knows who you are. You could then have it call up your "saved session" with your location in a book, your menu layout, your preferences, etc. Cool!
Take catalogs for example. Sure, you could look up what you want online. I do that almost all the time, and I'd be one of the first to buy a Web-connected e-ink catalog. But when you just want to browse, I can't stand having to click-tap-scroll-push-wheel through the pages. I need to be able to put my thumb on the side of a book, flip through and feel the breeze in my face while watching the pages go by...
Seriously though, until they can think of a better input device to navigate an eBook, they'll never replace paper. I'll get an eBook. You'll get an eBook. We'll all accept whatever's out there, eventually (we'll probably even begin to like DRM!). But IMHO, paper is here to stay.