The overhead of compositing even ten or twenty such views is negligible. Really. A decent PC from 2002 could do that much. If it's got the computing resources to render the image in the first place it's not going to be overloaded with any credible number of viewers.
Multiple locations is irrelevant, since you will have a complete PC at each location anyway.
I'm not sure that servers are as great a business model as he thinks. There's a lot more competition for servers, and the desktop is Ubuntu's strength compared to other Linux distros.
But I may be wrong... what's the big attraction for Ubuntu on a server as compared to other Linux servers?
The key difference with how they are doing it is it scales.....no matter how many people join in, performance doesn't take a hit. two images are always being projected at all times and you can have two people or hundreds moving their 'windows' around and the setup doesn't see any difference.
Even over a large table, I don't see how having more than three or four people sweeping sheets of paper around to view the secondary image is going to be in any way practical. What kind of application are you thinking of where you would have enough people using it to make an extra compositing step noticable, considering that the load from compositing a dozen translucent windows on my pitiful G4 Mac mini was negligible?
This is cool technology, but if it can sense the location of IR-reflective objects on the table it doesn't need to actually project anything onto the paper. You could simply lay a frame on the table so it could sense the corners of the frame, then composite the image onto the display as if the frame was a sheet of paper. Then the transparency of the paper can be handled in software, you don't need the special surface, and you can have as many "sheets of paper" as you want.
Projecting onto objects above the table is cool, but not super practical. The "IR Mouse" is really more interesting.
The OS obviously does matter to the application, or else I would still be using FreeBSD on a Thinkpad instead of OS X on a Macbook. And I'd be happier. Certainly my wrists would be happier: Apple's hardware looks pretty but it's an ergonomic disaster.
The hardware certainly matters. Apple's restricted hardware kept me from getting a new Mac until the mac mini came out and I could get a desktop Mac that was actually an upgrade over my beige G3.
Applications matter, or (as noted) I'd be sticking with a free OS regardless of the available applications.
Someone who says "X doesn't matter" is trying to sell you something that doesn't do X. Well, except Chumbawamba. But they're artists.
It's all about compromise. I will never compromise with terrorist operating systems.
Used to be tinc.org, back when we set this up because Network Solutions was playing sillybuggers with ".com". I think we were the second alternative TLD after the ".cool" people. We might have been the first, we didn't find out about each other until we'd already set up our zones and nameservers and the like...
The menu system that many MS apps employed previously were definitely not the best option.
That's not a problem with menus, that's a problem with those particular applications. Removing the menu bar because some applications had let theirs get disorganized and were due for a refresh is like replacing the steering wheel with a joystick because putting the radio controls in the steering wheel on one car model turned out to be a bad idea.
Except Microsoft went further and put the gas cap on the dashboard and moved the speedometer to the glove compartment while they were about it.
When did they ever say that Windows 7 would be anything more than Vista SP3?
Windows 7 takes a different approach to the componentization and backwards compatibility issues; in short, it doesn't think about them at all. Windows 7 will be a from-the-ground-up packaging of the Windows codebase; partially source, but not binary compatible with previous versions of Windows. Making the break from backwards compatibility is a dangerous proposal but a dream for software developers. Performance of native applications can be increased, distribution sizes can be cut down, functionality can be added without the worry of breaking old applications, and the overall end-user experience can be significantly improved.
In Windows 7, Microsoft will break from the Windows' norm by breaking previous API compatibility, offering new API frameworks as a native solution, and providing support for legacy frameworks (COM, ATL,.NET Framework, etc) through monolithic libraries designed to provide the functionality of all previous revisions of the modules in question. This extends/replaces the WinSxS philosophy, providing every single function, past and present, in fully comprehensive libraries. This should allow the majority of legacy applications to run perfectly, while still retaining native performance for applications compiled specifically with the Windows 7 platform in mind. It should also be possible for applications produced with previous versions of Visual Studio to be directly recompiled into native code using the new API frameworks.
This is what they were allegedly telling beta developers back in April.
I have literally dozens of workarounds for non-existent or buggy API calls in Windows 2000, that are available in XP and work.
XP has a whole load of additional bundled applications, many of which, such as Windows Media Player 9 and the crippled Citrix Terminal Server, include new kernel components.
Now compare that to the differences between Windows 2000 and Windows NT4, or between XP and Vista. The differences between 2000 and XP are small compared to the differences between actual new OS releases.
I'm still running Windows 2000, and apart from applications that depend on new bundled components, just about everything I've found that doesn't work under 2000 works fine once you update 2000 with all the support packs and hotfixes, and disable the version checking that says they require XP. About the only thing I haven't been able to do is jam the Microsoft bluetooth stack into 2000... and most BT apps don't work with the older 3rd party stacks any more.
What new users? Where does Microsoft find computer users who have NEVER seen a menu?
This is probably going to turn out like Apple's bogus studies on the one-button mouse. I've been asking for details of THAT study for 20 years, and the best I've found have been third-hand reports that indicate they did one study, one that was fundamentally flawed, and never published it for review anywhere.
History is full of bad user interface ideas pushed by software companies with a goal of differentiating themselves from the competition instead of attempting to design the best interface possible. The companies that have any market success carry along with that success a "halo" that attaches itself to every decision they made, good or bad, until those decisions become lore.
How exactly is the menu-system "convoluted"? Well, Microsoft added a bunch of complications to THEIR menus over the years, but that's not an inherent part of the menu interface. Comparing Microsoft's menus against the ribbon is like comparing a sick racehorse against a sloth. The sloth may win the race, but that doesn't mean you should go out and harness one up to your buggy.
So...what's wrong with the ribbon?
It's an awkward compromise between Xerox' context-sensitive menus and Apple's menu bar.
It abandons the tight state-sensitive behavior of contextual menus because it's continually displayed and so can't restrict itself to only providing options for specific objects, but retains much of the clutter of menus because it has to display actions associated with multiple objects.
It abandons the scannability and location-sensitive behavior of menus because you only see actions related to the high-level of the window. You can't scan it to learn the range of actions available from the program.
Since you have to have Office installed locally I suspect that it involves an ActiveX control with whatever plugins are necessary to wedge it into Safari and Firefox on Windows.
Most of the original scope of Windows 7 has been abandoned. The new cleaned-up native API? Not a word about that. The Classic-like sandboxes for legacy APIs? Gone. What we have is more like a Plus Pack for Windows Vista, the same way Windows XP was a Plus Pack for Windows 2000.
So I don't think there's any reason to treat it as a joke. Windows 7 really is Mojave. It's Vista with some new bundled apps and gratuitous user interface changes (who came up with the ribbon? What was he on? Does the DEA know about it?), and a fresh new name to try and dump the bad PR from the botched release. It worked in the Mojave Experiment, so they see no reason not to go ahead and expand its scope.
Slashdot Mirror
The overhead of compositing even ten or twenty such views is negligible. Really. A decent PC from 2002 could do that much. If it's got the computing resources to render the image in the first place it's not going to be overloaded with any credible number of viewers.
Multiple locations is irrelevant, since you will have a complete PC at each location anyway.
And I got one of the early ones that could be turned into regular scanners by cutting a trace.
I'm not sure that servers are as great a business model as he thinks. There's a lot more competition for servers, and the desktop is Ubuntu's strength compared to other Linux distros.
But I may be wrong... what's the big attraction for Ubuntu on a server as compared to other Linux servers?
The key difference with how they are doing it is it scales.....no matter how many people join in, performance doesn't take a hit. two images are always being projected at all times and you can have two people or hundreds moving their 'windows' around and the setup doesn't see any difference.
Even over a large table, I don't see how having more than three or four people sweeping sheets of paper around to view the secondary image is going to be in any way practical. What kind of application are you thinking of where you would have enough people using it to make an extra compositing step noticable, considering that the load from compositing a dozen translucent windows on my pitiful G4 Mac mini was negligible?
This is cool technology, but if it can sense the location of IR-reflective objects on the table it doesn't need to actually project anything onto the paper. You could simply lay a frame on the table so it could sense the corners of the frame, then composite the image onto the display as if the frame was a sheet of paper. Then the transparency of the paper can be handled in software, you don't need the special surface, and you can have as many "sheets of paper" as you want.
Projecting onto objects above the table is cool, but not super practical. The "IR Mouse" is really more interesting.
Once they implement the Freedom Proof Fence.
*snort*
The OS obviously does matter to the application, or else I would still be using FreeBSD on a Thinkpad instead of OS X on a Macbook. And I'd be happier. Certainly my wrists would be happier: Apple's hardware looks pretty but it's an ergonomic disaster.
The hardware certainly matters. Apple's restricted hardware kept me from getting a new Mac until the mac mini came out and I could get a desktop Mac that was actually an upgrade over my beige G3.
Applications matter, or (as noted) I'd be sticking with a free OS regardless of the available applications.
Someone who says "X doesn't matter" is trying to sell you something that doesn't do X. Well, except Chumbawamba. But they're artists.
It's all about compromise. I will never compromise with terrorist operating systems.
I'm not really addressing your conclusions here, I'm just wondering about one of your assumptions...
Think about it: the OS doesn't *really* matter (if it did OS X and Linux and all the rest would never have any users).
If the OS didn't matter I'd be using Windows. It's because the OS matters that there's more than one OS out there.
Can you explain what you mean here?
Hey, the ".dot" domain belongs to The Internet Namespace Cooperative.
http://tinc.taronga.com/
Used to be tinc.org, back when we set this up because Network Solutions was playing sillybuggers with ".com". I think we were the second alternative TLD after the ".cool" people. We might have been the first, we didn't find out about each other until we'd already set up our zones and nameservers and the like...
For all those wanting to go back to a paper-based system...
It's not "wanting to go back".
It's "not wanting to lose the paper".
You can have an electronic system that produces a paper ballot.
The menu system that many MS apps employed previously were definitely not the best option.
That's not a problem with menus, that's a problem with those particular applications. Removing the menu bar because some applications had let theirs get disorganized and were due for a refresh is like replacing the steering wheel with a joystick because putting the radio controls in the steering wheel on one car model turned out to be a bad idea.
Except Microsoft went further and put the gas cap on the dashboard and moved the speedometer to the glove compartment while they were about it.
When did they ever say that Windows 7 would be anything more than Vista SP3?
This is what they were allegedly telling beta developers back in April.
I have literally dozens of workarounds for non-existent or buggy API calls in Windows 2000, that are available in XP and work.
XP has a whole load of additional bundled applications, many of which, such as Windows Media Player 9 and the crippled Citrix Terminal Server, include new kernel components.
Now compare that to the differences between Windows 2000 and Windows NT4, or between XP and Vista. The differences between 2000 and XP are small compared to the differences between actual new OS releases.
I'm still running Windows 2000, and apart from applications that depend on new bundled components, just about everything I've found that doesn't work under 2000 works fine once you update 2000 with all the support packs and hotfixes, and disable the version checking that says they require XP. About the only thing I haven't been able to do is jam the Microsoft bluetooth stack into 2000... and most BT apps don't work with the older 3rd party stacks any more.
What new users? Where does Microsoft find computer users who have NEVER seen a menu?
This is probably going to turn out like Apple's bogus studies on the one-button mouse. I've been asking for details of THAT study for 20 years, and the best I've found have been third-hand reports that indicate they did one study, one that was fundamentally flawed, and never published it for review anywhere.
History is full of bad user interface ideas pushed by software companies with a goal of differentiating themselves from the competition instead of attempting to design the best interface possible. The companies that have any market success carry along with that success a "halo" that attaches itself to every decision they made, good or bad, until those decisions become lore.
Microsoft is embracing the cloud. I'm worrying about the weather.
I'm worried about the poor cloud.
You can do full web-based previews and rudimentary editing without Office installed locally.
Gee, I don't have a Microsoft Magic Password on my Mac, so I can't thrill myself with the rudimentary editing!
Guess I'll just have to put up with Office X.
The hell were their UI people thinking?
Objection! Counsel is assuming facts not in evidence!
the convoluted menu system
How exactly is the menu-system "convoluted"? Well, Microsoft added a bunch of complications to THEIR menus over the years, but that's not an inherent part of the menu interface. Comparing Microsoft's menus against the ribbon is like comparing a sick racehorse against a sloth. The sloth may win the race, but that doesn't mean you should go out and harness one up to your buggy.
So...what's wrong with the ribbon?
It's an awkward compromise between Xerox' context-sensitive menus and Apple's menu bar.
It abandons the tight state-sensitive behavior of contextual menus because it's continually displayed and so can't restrict itself to only providing options for specific objects, but retains much of the clutter of menus because it has to display actions associated with multiple objects.
It abandons the scannability and location-sensitive behavior of menus because you only see actions related to the high-level of the window. You can't scan it to learn the range of actions available from the program.
Since you have to have Office installed locally I suspect that it involves an ActiveX control with whatever plugins are necessary to wedge it into Safari and Firefox on Windows.
"Nichtlachen-Keinwortz"? I think they agreed to be taken over last month after getting into trouble with too many bad debts.
Be careful, young man. Mr Bent has no sense of humor about such allegations.
I guess when they finally release a version of WinFS it will be bundled with Duke Nukem Forever?
And the object-oriented GUI. We're still waiting for Cairo, let alone Longhorn.
Microsoft is making Copland actually look like a product. At least Apple shipped some pre-alpha code to a few developers, even if it didn't work.
Vista XP, Mojave, whatever they call it, it's just Vista with a makeover. It's too serious to be funny.
Most of the original scope of Windows 7 has been abandoned. The new cleaned-up native API? Not a word about that. The Classic-like sandboxes for legacy APIs? Gone. What we have is more like a Plus Pack for Windows Vista, the same way Windows XP was a Plus Pack for Windows 2000.
So I don't think there's any reason to treat it as a joke. Windows 7 really is Mojave. It's Vista with some new bundled apps and gratuitous user interface changes (who came up with the ribbon? What was he on? Does the DEA know about it?), and a fresh new name to try and dump the bad PR from the botched release. It worked in the Mojave Experiment, so they see no reason not to go ahead and expand its scope.
You've got Nichtlachen-Keinwortz Syndrome, right?
in London, 20% of all business access points continue to be completely unprotected.
So the title should read "London is still world's Wi-Fi Wardriving Capital", yesno?