Do they ban google as well? How about about.com, or lycos, or dmoz?
I guess if the only place kids can use the Internet for research is at home or the public library, that'll save money on that pesky school library and Internet connection.
Apple could get into the position Sony finds itself with the PS3 - last to market at the highest price.
You can already get smartphones with media players and good web browsers for less than the iPhone *today*, and a 4GB secure digital card for your smartphone will set you back less than the price of an iPod Shuffle.
If you don't mind watching your programs in "squashed mode". I guess you could pretend you were watching movies on TV back in the '60s, when they used to broadcast them in anamorphic format. Way retro, dude... cue up Barbarella and have a toke...
It is the same price as a $299 smart phone plus $199 iPod nano.
For less than that I can get a brand new unlocked Treo or Pocket PC smart phone with a real keypad I can use with any GSM carrier. I can get 4GB of flash for about $50.
The service will also be cheaper than other phones because there is no hardware discount as with other phones.
Outside the US that's true, but it's not going to be available outside the US. Inside the US, well, lock-in is the name of the game. If you get an iPhone you're locked to one carrier, you've got no leverage to work for discounts.
The thing that people keep skipping over is the Web browser.
This might sound strange from me, given how much I love to trash Microsoft, but Pocket Internet Explorer is a killer application. It's a well designed handheld browser, and the biggest difference between it and the desktop IE is that it works better with good CSS-compliant websites and doesn't use Microsoft's "Typhoid Mary" security model.
I haven't used a recent handheld version of Opera, but it's another option.
To get that kind of Web browsing you have to go to a MacBook at $1100.
I've got a Macbook Pro. Safari doesn't thrill me... I use Camino.
But if that's what flips your switches, Nokia uses Webkit too.
The biggest thing is the software, though.
Yep, either Windows CE or Palm OS are years ahead of the iPhone here. Even Nokia's oldschool Symbian phones have a decent software ecosystem.
And there's also a robust accessory ecosystem for them, like there is for the iPod.
Running OS X? It's running a Darwin-based OS that's locked so you can't add software to it. If you want a phone where you have to depend on a single vendor for your software you can get one a LOT cheaper than an iPhone. Hell I think it's kind of cheating to even call it a smartphone until they fix that.
Everybody needs a new TV.
*snort*
If you don't mind paying twice as much for less phone, you're not even in the market I'm talking about.
These are "DVD era" video outputs, and they work on any TV that has component inputs, which is most everything from the 21st century.
So long as it's widescreen. Most non-digital TVs in the US aren't.
That be the general definition of CSS (for better or worse).
CSS separates style from content (that's right there in the name, cascading style sheets), it doesn't address layout at all, which is why people using it for layout have to come up with horrible hacks with floats and the like. They're no better than tables, and I'm glad -moz-column-width is ugly and prefixed and not a standard, because it's too damn specialised. Reminds me of... too bad that wasn't <-mosaic-image...> so there was a reason to switch to <fig>...</fig>
You REALLY want to use CSS for layout, you need a new container type.
Call it box or flow or slot, what it does is define the layout of the places where text WILL go. The text is in a separate div with a target that specifies the box to flow it into.
That way you can define a set of boxes around a picture, all with the same name, and then just feed text into it. The text will start filling the first box, then overflow into the second, and so on. When the text reaches the end of the last box the boxes start expanding to accomodate the text, unless they have the an attribute that defines them as fixed-size. The boxes define the layout, the divs define the text to go into it, and spans define the style of the text.
Columns shouldn't require CSS, or table fakery. They need to REALLY separate layout from content.
You should be able to define the flow of running text independently of the text itself, so you neither have to guess where to put the column breaks or use floats to fake columns.
The iPhone is about $350 too much to be a killer product, even if it was a killer product otherwise. And I don't see it being a killer product to begin with. Touch screens on phones suck dirty swamp water through clogged pores.
AppleTV is priced to sell a lot of units, but there's a hidden cost to it - most people will need to buy a new TV for it. It needs to work well on regular screens without a hack to really take off.
You can netboot OS X and use the same commands. You can even make it automatically do the install from launchd, either unattended or after login, and there's GUI tools for setting that part up.
Here's the requirements for the Linux client:
Minimum requirements: * Internet Connection: Cable or DSL * Computer Processor: 800MHz Pentium III or Athlon, or better * Computer Memory: 256MB or better (strongly recommend more!) * Linux Operating System: A reasonably modern 32-bit Linux environment is required. If you are running a 64-bit Linux distribution then you will need its 32-bit compatibility environment installed. * Video/Graphics Card: o nVidia GeForce 2, GeForce 4mx, or better o OR ATI Radeon 8500, 9250, or better
**NOTE**: Second Life absolutely requires you to have recent, correctly-configured OpenGL 3D drivers for your hardware - the graphics drivers that came with your operating system may not be good enough! See the TROUBLESHOOTING section if you encounter problems starting Second Life.
For a more comfortable experience, the RECOMMENDED hardware for the Second Life Linux client is very similar to that for Windows, as detailed at:.../sysreqs.php
Instead of trying to turn the Palm platform into a laptop replacement, they should have concentrated on its strengths (small size, low overhead, low processor requirements, SUPER-reliable syncing) and ridden the wave of Moore's Law down to Walmart. How many Palms do you think they'd sell if you could buy one for $29.95 at the checkout counter when you're picking up your school supplies... bundled with a program that emulates the school's required Ti-83 that you're having to fork out just as much money for?
Yes, Palm could *easily* have done this, without sacrificing profits. Instead, they overreacted to the Pocket PC and pulled one of the three classic blunders (never get involved in a land war in Asia, never go up against a Sicilian when death is on the line, and never go after Microsoft on their playing field)... and now they've basically got nothing left but their name.
There's nothing more you can do to strip away anonymity than to track the IP address of posters, because there's nothing more that gmail and hotmail and yahoo will do for you.
They can't... they have no more idea who randomstalker@freemailservice.example.com is than you, unless you think the IP address of the public terminal they used to set up the account is useful.
The short version of the reason is "scripts versus compiled code".
The long version:
* Web applications are developed interactively. You fix a typo, you hit reload, you see the results immediately. The interactive-versus-batch debate should have been over by 1980, but we still see dektop apps written almost completely in compiled languages that require a huge clumsy IDE.
* The part that runs in the browser is visible to the users, so when they are technically competant they can give you detailed feedback.
* It's automatically cross-platform.
* The problem space is reduced because web applications just can't do a lot of things, so people don't expect you to do them, so you don't have to implement them or explain why you didn't.
Any scripting language can give you the first two advantages... you don't need to write web applications to write in an interactive language that gives you immediate and early feedback, or that your users can dig into and improve.
Using a portable GUI toolkit like Tk lets you write cross-platform scripts as well.
And, ironically, using a toolkit that limits what you can do (like, say, Dashboard or Konfabulator widgets) can actually make programming more enjoyable because it reduces the problem space.
Which is why there are so many people writing little scripted plugins for *everything*. Firefox extensions. Applescript. Shell scripts. CGI scripts. PHP. Tcl/Tk. Perl. Widgets.
It's like the impact UNIX had on command lines in the '70s. Your mainframe programming environments were, like GUI environments now, very complex. You could specify everything, down to things like the virtual block size of your virtual disk. UNIX reduced the kinds of things you could do in your program... and it turned out that being able to specify the block size of your virtual disk wasn't something people really cared about.
I've been waiting for a similar simplification to take over the GUI world for years, something that dumped the irrelevant mainframe-style super-detailed API and concentrated on the essentials. I had hopes for Plan 9's window system, but Bell Labs played that too close to their chest. Tcl/Tk is wonderful, but if you don't "get" lispy reflective programming you're not going to like it, and Perl/Tk hasn't taken off either.
The browser has moved into this space by default, because it's solved the distribution problem, but it's not a simple programming model. It's not the answer, and neither is any proprietary platform built on top of it. It's been over 20 years now that I've been waiting for something to really fill in the 2d desktop... and I'm no longer expecting that anything will.
Something like an open source re-implementation of Konfabulator, maybe... it could be built on a standalone Firefox XUL engine with no browser component dragging in the horrifying security issues that trying to run trusted and untrusted applets in the same engine bring.
But... if you want desktop programming to be as fun as web programming, find a scripting language with a GUI toolkit and start playing...
You had me until you said C++ was an oddball language that was hard to work with.
Just because something is popular doesn't mean that it's well designed, or even mediocre, or even following the usual design principles of things in the same class.
Consider the current US system of measurement, Microsoft Windows, or classic Mac OS. These are all oddball designs in a world of better systems, and cause (or caused) all kinds of problems for people who had to use them.
C++ is partly a throwback to the very earliest models of OO languages (like Simula), and partly a language crippled by the early implementations that used C as a back end. It's very difficult to interoperate with C++ from other languages, the early binding of method names makes it a poor choice for an OS API, and the mixture of objects and primitive C types leaves it open to all the type security problems of C... with few of the benefits of C's low-level design to compensate.
we're certainly approaching the line where the home internet user will only have a NATed address.
I agree that this is true.
I agree that this is a problem.
If I were the God of the Interweb, we'd have been on IPv6 years ago.
But practically speaking, the people who are running the Interweb don't care about this, in fact they would rather have users on NATted addresses, so the address space problem is not going to lead to IPv6, it's going to lead to more NATting.
The original comment was about provisioning 80 million customers, though, and how you'd do that without IPv6... and the followup implied that you'd need 80 million static IPs to do it. All I'm saying is that ain't so... the v4 address space won't run out as long as ISPs can continue to extend NAT and PAT.
I would love to see IPv6 deployed, I don't want to lose my routable address even if I don't run a server, and I expect that's going to happen sooner or later. But it's not up to me, and address space pressure isn't going to make it happen.
Theo and friends got shown off as what they are; petty thieves
I think that's too strong. Theo's response was classic displacement and he really needs to get over that... but it sure looks like the original commits of this code were a mistake rather than a deliberate attempt to break the GPL, and that Theo was not aware that GPL code had been committed to the repository until after the fact.
Except that an awful lot of 3G (UMTS) services operate via IP, and are supposed to continue to do so (and be directly addressable from the home network) whilst roaming.
That's a nice feature, one that seems to be designed to promote IPv6, but there are techniques to implement that with NAT/PAT using asymmetrical routing.
I'm not entirely sure how giving every phone the same IP and NATing achieves anything beyond confusion
It's a thought experiment that demonstrates that you don't need 80 million IP addresses to provide services to 80 million cell phones. Obviously you would use a less parsimonious distribution of NATted addresses, and use asymmetrical routing to implement roaming.
or what 'NAT them all by PSTN at your border router' means
Doesn't matter if you use PSTN, MSIN, or anything else, just so long as it's a unique ID per phone that can be used to associate any open TCP sessions the phone has with a matching internet-routable IP and port at the border between the cell net and the public internet.
Slashdot Mirror
Do they ban google as well? How about about.com, or lycos, or dmoz?
I guess if the only place kids can use the Internet for research is at home or the public library, that'll save money on that pesky school library and Internet connection.
Apple could get into the position Sony finds itself with the PS3 - last to market at the highest price.
You can already get smartphones with media players and good web browsers for less than the iPhone *today*, and a 4GB secure digital card for your smartphone will set you back less than the price of an iPod Shuffle.
AppleTV has a 480i mode.
If you don't mind watching your programs in "squashed mode". I guess you could pretend you were watching movies on TV back in the '60s, when they used to broadcast them in anamorphic format. Way retro, dude... cue up Barbarella and have a toke...
It is the same price as a $299 smart phone plus $199 iPod nano.
For less than that I can get a brand new unlocked Treo or Pocket PC smart phone with a real keypad I can use with any GSM carrier. I can get 4GB of flash for about $50.
The service will also be cheaper than other phones because there is no hardware discount as with other phones.
Outside the US that's true, but it's not going to be available outside the US. Inside the US, well, lock-in is the name of the game. If you get an iPhone you're locked to one carrier, you've got no leverage to work for discounts.
The thing that people keep skipping over is the Web browser.
This might sound strange from me, given how much I love to trash Microsoft, but Pocket Internet Explorer is a killer application. It's a well designed handheld browser, and the biggest difference between it and the desktop IE is that it works better with good CSS-compliant websites and doesn't use Microsoft's "Typhoid Mary" security model.
I haven't used a recent handheld version of Opera, but it's another option.
To get that kind of Web browsing you have to go to a MacBook at $1100.
I've got a Macbook Pro. Safari doesn't thrill me... I use Camino.
But if that's what flips your switches, Nokia uses Webkit too.
The biggest thing is the software, though.
Yep, either Windows CE or Palm OS are years ahead of the iPhone here. Even Nokia's oldschool Symbian phones have a decent software ecosystem.
And there's also a robust accessory ecosystem for them, like there is for the iPod.
Running OS X? It's running a Darwin-based OS that's locked so you can't add software to it. If you want a phone where you have to depend on a single vendor for your software you can get one a LOT cheaper than an iPhone. Hell I think it's kind of cheating to even call it a smartphone until they fix that.
Everybody needs a new TV.
*snort*
If you don't mind paying twice as much for less phone, you're not even in the market I'm talking about.
These are "DVD era" video outputs, and they work on any TV that has component inputs, which is most everything from the 21st century.
So long as it's widescreen. Most non-digital TVs in the US aren't.
I'd rather they removed the tag.
That be the general definition of CSS (for better or worse).
... too bad that wasn't <-mosaic-image ...> so there was a reason to switch to <fig>...</fig>
CSS separates style from content (that's right there in the name, cascading style sheets), it doesn't address layout at all, which is why people using it for layout have to come up with horrible hacks with floats and the like. They're no better than tables, and I'm glad -moz-column-width is ugly and prefixed and not a standard, because it's too damn specialised. Reminds me of
You REALLY want to use CSS for layout, you need a new container type.
Call it box or flow or slot, what it does is define the layout of the places where text WILL go. The text is in a separate div with a target that specifies the box to flow it into.
That way you can define a set of boxes around a picture, all with the same name, and then just feed text into it. The text will start filling the first box, then overflow into the second, and so on. When the text reaches the end of the last box the boxes start expanding to accomodate the text, unless they have the an attribute that defines them as fixed-size. The boxes define the layout, the divs define the text to go into it, and spans define the style of the text.
Columns shouldn't require CSS, or table fakery. They need to REALLY separate layout from content.
You should be able to define the flow of running text independently of the text itself, so you neither have to guess where to put the column breaks or use floats to fake columns.
Pity Babbage's customers. He delayed the Difference Engine to get the Analytical Engine out, and it never shipped!
The iPhone is about $350 too much to be a killer product, even if it was a killer product otherwise. And I don't see it being a killer product to begin with. Touch screens on phones suck dirty swamp water through clogged pores.
AppleTV is priced to sell a lot of units, but there's a hidden cost to it - most people will need to buy a new TV for it. It needs to work well on regular screens without a hack to really take off.
Either they're using that as an excuse, or they've been huffing the Reality Distortion Field through used crack pipes.
I mean... sheesh. That'd be like is Microsoft delayed Windows 2000 for Microsoft Bob.
Parallels cheaped out and use a local disk share with full local user rights for drag-and-drop.
If you want to use Parallels as a sandbox, make sure you have "Enable sharing for drag-and-drop" disabled.
You can netboot OS X and use the same commands. You can even make it automatically do the install from launchd, either unattended or after login, and there's GUI tools for setting that part up.
Here's the requirements for the Linux client: Minimum requirements:
* Internet Connection: Cable or DSL
* Computer Processor: 800MHz Pentium III or Athlon, or better
* Computer Memory: 256MB or better (strongly recommend more!)
* Linux Operating System: A reasonably modern 32-bit Linux environment is required. If you are running a 64-bit Linux distribution then you will need its 32-bit compatibility environment installed.
* Video/Graphics Card:
o nVidia GeForce 2, GeForce 4mx, or better
o OR ATI Radeon 8500, 9250, or better
**NOTE**: Second Life absolutely requires you to have recent, correctly-configured OpenGL 3D drivers for your hardware - the graphics drivers that came with your operating system may not be good enough! See the TROUBLESHOOTING section if you encounter problems starting Second Life.
For a more comfortable experience, the RECOMMENDED hardware for the Second Life Linux client is very similar to that for Windows, as detailed at:
Wait for Microsoft to claim they were already talking to the labels about DRM-free music when Apple stole their idea.
Instead of trying to turn the Palm platform into a laptop replacement, they should have concentrated on its strengths (small size, low overhead, low processor requirements, SUPER-reliable syncing) and ridden the wave of Moore's Law down to Walmart. How many Palms do you think they'd sell if you could buy one for $29.95 at the checkout counter when you're picking up your school supplies... bundled with a program that emulates the school's required Ti-83 that you're having to fork out just as much money for?
Yes, Palm could *easily* have done this, without sacrificing profits. Instead, they overreacted to the Pocket PC and pulled one of the three classic blunders (never get involved in a land war in Asia, never go up against a Sicilian when death is on the line, and never go after Microsoft on their playing field)... and now they've basically got nothing left but their name.
There's nothing more you can do to strip away anonymity than to track the IP address of posters, because there's nothing more that gmail and hotmail and yahoo will do for you.
They can't... they have no more idea who randomstalker@freemailservice.example.com is than you, unless you think the IP address of the public terminal they used to set up the account is useful.
The Code of Hammurabi already had sections regulating brewers and taverns. Anyone got an earlier precedent than blurred cuneiform?
The short version of the reason is "scripts versus compiled code".
The long version:
* Web applications are developed interactively. You fix a typo, you hit reload, you see the results immediately. The interactive-versus-batch debate should have been over by 1980, but we still see dektop apps written almost completely in compiled languages that require a huge clumsy IDE.
* The part that runs in the browser is visible to the users, so when they are technically competant they can give you detailed feedback.
* It's automatically cross-platform.
* The problem space is reduced because web applications just can't do a lot of things, so people don't expect you to do them, so you don't have to implement them or explain why you didn't.
Any scripting language can give you the first two advantages... you don't need to write web applications to write in an interactive language that gives you immediate and early feedback, or that your users can dig into and improve.
Using a portable GUI toolkit like Tk lets you write cross-platform scripts as well.
And, ironically, using a toolkit that limits what you can do (like, say, Dashboard or Konfabulator widgets) can actually make programming more enjoyable because it reduces the problem space.
Which is why there are so many people writing little scripted plugins for *everything*. Firefox extensions. Applescript. Shell scripts. CGI scripts. PHP. Tcl/Tk. Perl. Widgets.
It's like the impact UNIX had on command lines in the '70s. Your mainframe programming environments were, like GUI environments now, very complex. You could specify everything, down to things like the virtual block size of your virtual disk. UNIX reduced the kinds of things you could do in your program... and it turned out that being able to specify the block size of your virtual disk wasn't something people really cared about.
I've been waiting for a similar simplification to take over the GUI world for years, something that dumped the irrelevant mainframe-style super-detailed API and concentrated on the essentials. I had hopes for Plan 9's window system, but Bell Labs played that too close to their chest. Tcl/Tk is wonderful, but if you don't "get" lispy reflective programming you're not going to like it, and Perl/Tk hasn't taken off either.
The browser has moved into this space by default, because it's solved the distribution problem, but it's not a simple programming model. It's not the answer, and neither is any proprietary platform built on top of it. It's been over 20 years now that I've been waiting for something to really fill in the 2d desktop... and I'm no longer expecting that anything will.
Something like an open source re-implementation of Konfabulator, maybe... it could be built on a standalone Firefox XUL engine with no browser component dragging in the horrifying security issues that trying to run trusted and untrusted applets in the same engine bring.
But... if you want desktop programming to be as fun as web programming, find a scripting language with a GUI toolkit and start playing...
Going by the poll on that page, anyway.
You had me until you said C++ was an oddball language that was hard to work with.
Just because something is popular doesn't mean that it's well designed, or even mediocre, or even following the usual design principles of things in the same class.
Consider the current US system of measurement, Microsoft Windows, or classic Mac OS. These are all oddball designs in a world of better systems, and cause (or caused) all kinds of problems for people who had to use them.
C++ is partly a throwback to the very earliest models of OO languages (like Simula), and partly a language crippled by the early implementations that used C as a back end. It's very difficult to interoperate with C++ from other languages, the early binding of method names makes it a poor choice for an OS API, and the mixture of objects and primitive C types leaves it open to all the type security problems of C... with few of the benefits of C's low-level design to compensate.
we're certainly approaching the line where the home internet user will only have a NATed address.
I agree that this is true.
I agree that this is a problem.
If I were the God of the Interweb, we'd have been on IPv6 years ago.
But practically speaking, the people who are running the Interweb don't care about this, in fact they would rather have users on NATted addresses, so the address space problem is not going to lead to IPv6, it's going to lead to more NATting.
The original comment was about provisioning 80 million customers, though, and how you'd do that without IPv6... and the followup implied that you'd need 80 million static IPs to do it. All I'm saying is that ain't so... the v4 address space won't run out as long as ISPs can continue to extend NAT and PAT.
I would love to see IPv6 deployed, I don't want to lose my routable address even if I don't run a server, and I expect that's going to happen sooner or later. But it's not up to me, and address space pressure isn't going to make it happen.
Theo and friends got shown off as what they are; petty thieves
I think that's too strong. Theo's response was classic displacement and he really needs to get over that... but it sure looks like the original commits of this code were a mistake rather than a deliberate attempt to break the GPL, and that Theo was not aware that GPL code had been committed to the repository until after the fact.
Except that an awful lot of 3G (UMTS) services operate via IP, and are supposed to continue to do so (and be directly addressable from the home network) whilst roaming.
That's a nice feature, one that seems to be designed to promote IPv6, but there are techniques to implement that with NAT/PAT using asymmetrical routing.
I'm not entirely sure how giving every phone the same IP and NATing achieves anything beyond confusion
It's a thought experiment that demonstrates that you don't need 80 million IP addresses to provide services to 80 million cell phones. Obviously you would use a less parsimonious distribution of NATted addresses, and use asymmetrical routing to implement roaming.
or what 'NAT them all by PSTN at your border router' means
Doesn't matter if you use PSTN, MSIN, or anything else, just so long as it's a unique ID per phone that can be used to associate any open TCP sessions the phone has with a matching internet-routable IP and port at the border between the cell net and the public internet.