I would guess the number would be more like a few thousand - but fundamentally I'm in agreement with you.
Harnessing the power of the atom certainly gives us enough power to sterilize the land (I don't think anything we do short of boiling them right off would sterilize the ocean)
Then again the same thing could be done more cheaply with a few hundred thousand samarium bombs - except the cockroaches might survive that.
My theory is they chose a non-standard port specifically to limit their usability - by using a typically blocked port, they've stopped all the slashdotters who surf from work from overloading their service, for instance...
Actually, when I say it's vetted, I was referring to the Java sandbox model - it's a security model that has worked for sometime before Microsoft ever thought to include it in their runtime.
I don't know how buggy it is or not. I don't know how secure it is or not. I know it's cryptographically secure, in that there's a lot of math being done to verify that the permission set granted is the one intended by the owner of the computer, and that the code being run is the code that the owner believes he has acquired. I know it's very atomic - permissions can be granted or restricted to very atomic operations. I know it's supported by the Common Language Runtime and enforced on a per-method-call basis.
And I know that it was released around the same time during which Microsoft halted all new code development for four weeks to give developers time to address security concerns in every single piece of software they make.
That's the kind of effort I tend to trust. They've been stressing security ever since, and I doubt they just kind of forgot to check this.
And as I said, the security model isn't new. It's been in production use since 2002, and I saw CTP's of it back in 2000. AFAIK, there are no changes planned in it before Vista is released.
Not only that, but VB6 at least stored part of its design-time properties seperately from the form itself and there was no way to really tell what it would be doing.
I was commenting that this is no longer the case - VB has become a world class language now, and coding in it actually feels like you're writing code now.
Even DirectX is accesible to the VB developer now - at performances that are within a few percentage points of C#.
But that doesn't really address the underlying question - which is, has it ever been, previous to the current VB.NET environment, advantageous to write code in VB?
I think so. Say what you will about performance and sloppy code - I would rather take 15 minutes to draw a form in a designer and another 30 to wire up its events than take 4 hours to hand code a bunch of windows API stuff just to draw my form and get everything just how I want it and another 2 or 3 hours wiring up the events by hand - a common design experience in C++ in the early days.
VB revolutionized the world by making it easy and cheap to write software. No mind that the software didn't perform as well as it's C-bequeathed brethren. No matter that there were things you simply couldn't do. The fact is, it has spent decades meeting the needs of enterprise development more economically than any of its competitors - and this one fact has enabled computers to become the financial powerhouses they are today.
A computer without software to run is virtually indistinguishable from high-silicon-content sand. And not quite as useful. VB made it so anyone with a half-baked idea and a little money could afford to have their own apps custom written for them.
I've got a better idea - every election the feds generate a public/private keypair. Load up all the machines with the public key. Each voting machine has open code that takes a voter record (in a government-defined format), encrypts it, stores it, and clears its system memory. At the end of the day each machine spits out whatever flash memory format is being used to store the data - including a tamper-proof hash. These cards are carried to the vote tallying machine - built by a third party - which loads them into memory and physically locks access. Once every single chip is accounted for, the government releases the private key, everything gets decrypted and tallied, and everyone goes home satisfied.
Within the election machine itself no candidate is identified by party - have them all identified by unique ID, translated via a code table to the names shown on screen. The unique ID should be a random string of bits - as long as its unique, there was no way to predict it, and noone had any control over how it was chosen - you'll be secure. The voting machines cannot purposely skew voting results one way or the other if they don't know who they're voting for
IANAS but I'll answer your questions as best I can...
Use of a little unsharp masking: this is the digital equivalent of tweaking the focus of the enlarger in a chemical darkroom-- a practice that I believe was very common in the days of chemical microphotography
It is the digital equivalent - but unfortunately, in this case, the digital equivalent of an analogue process introduces severe aliasing artifacts, and can actually create or destroy spurious signals with repeated use - most scientists frown heavily on the use of sharp or unsharp masks for this reason. Paranormal researchers of integrity insist that no such mask ever be used to analyse images in their fields, in fact.
Use of the histogram controls and gamma adjustments: the digital equivalent of choosing films, papers, chemistries, and processing times and temperatures to expand or compress the tones, or shift the tonal balance of the final print (except with digital photography, we can now do this kind of stuff after the image is acquired rather than guessing beforehand)
This is mostly okay - as long as they're used responsibly. The trick is to avoid clipping artifacts - use these controls to normalise the data. Gamma is especially useful as it preserves the digital signal while allowing you to tweak sensitivity settings to bring out details in the image.
Use of hue and saturation controls: the digital equivalent of choosing films, papers, and filters with different color sensitivities
These types of techniques are widely used in astronomy (or equivalent techniques) - quite useful when part of your data metric actually lies outside the human spectrum of vision. Other fields use them to varying degrees - forensical analysis in particular is quite fond of them. Again, be careful not to clip your saturation when you're adjusting it, and you'll be okay.
Use of different kinds of noise suppression: in some respects, similar to controlling the grain of chemical images, but also needed to manage artifact specific to digital cameras
That depends - if you're using a standard Filter window or fourier transform, absolutely not... those techniques always insert spurious signals into data. Gaussian anti-noise techniques should be ok as a final pass but not before and not repeatedly. Most other anti-noise techniques I'd be very skeptical of.
To check if something's ok, generate a screen of static - just random noise. Then apply your technique to it over and over and over. If the static flattens out to a solid color, or clumps together becoming more defined, then you've shown that your technique can mask or amplify signals in your image.
BTW, if you want to do the kinds of things you mention above in a scientifically respectable way, you might want to look into high dynamic range imagery, which solves some of the clipping problems I mentioned above.
Heh sidegrade is more like it! Why the hell would Microsoft work so hard on a kernel just to throw it away in their next os drop? And load it with such useless crap while they're doing it!
But from Microsoft's point of view, ME was available as an upgrade to 98 - at least from a pricing standport - and is still supported officially...
The rich design features available will be run in a sandboxed mode just like Java - its a vetted security model that's extremely low risk. It'll be either version 2 or 3 of the technology, too, so it's not a virgin product (the same sandbox model has been support by the.NET CLR since version 1.0)
It does do it better and with better design time support.
It can target XHTML/CSS3/ECMASCRIPT browsers (I've heard it's cross browser ECMASCRIPT. I'm expecting some features, at least, to only be available in IE however)... a key of Microsoft web design technology since.NET has been that it "degrade gracefully"... at PDC 2000, when they first announced.NET CLR and ASP.NET, practically every speaker used that phrase when they'd explain how web controls work.
Microsoft understands that ActiveX was a shitty technology because of its platform dependence. The marketplace of rich UI controls they were expecting never really materialized. Now, ActiveX was the best possible at the time... JavaScript was hardly a gleam in Sun's eye, and it was a logical extension of the Netscape plugin module, which was itself a non-standard part of HTML at the time, but the point is that it's technological limitations severely hindered it.
Microsoft gets this. They've been working hard not to make the same mistakes again. Yes, some features of Quartz depend on the Windows Presentation Foundation (I believe they're calling it Avalon???) to be present - and if you've seen a WPF demo you'll understand why... javascript will never be able to present the type of rich UI I saw.
It may not be as open as SVG or canvas... but that doesn't really matter for people writing extranet sites, which is a large portion of active web development - in my seven years as a web technology consultant I have only been asked to work on a consumer-facing website 3 or 4 times, out of a total of approximately 50 I've worked on.
What matters to me is my clients for a majority have a fixed install base. Getting a component that I can redistribute or get my clients to install via automatic update is a huge boon in this situation - and provides an incredibly rich design time experience that allows me to deliver responsive software widely.
I don't care if that's SVG, canvas, WPF, or some bits picked off a deep-space radar dish in the New Mexico desert - as long as I can use it, that's good enough for me. The lifetime of corporate applications varies between 3 and 15 years. I have yet to see a single solution I've developed go 5 years without a retrofit - as businesses grow, their needs change and technology evolves to better fill them. Once you've ROIed you can usually build something better, cheaper, that'll help your business grow faster... and many of my clients choose to take advantage of that option shortly after that point in time.
I'm not worried that my applications will outlive Microsoft's support of WPF, nor am I worried that Microsoft will suddenly choose to limit my use of it. It's in their best interest for WPF to be as widely disseminated as possible. I don't need WPF to be open - that's why I buy software, so someone else can worry about the bugs while I go get my job done. Closed or open, I have no intention of ever cracking open the source files.
My only concern is building my apps better and cheaper than anyone has ever built apps before. Quartz will help me do that, and I can't wait to get my hands on it. Windows was fundamentally succesful in the operating system market back when they had something that could be called competition because their programming language, Visual Basic, was far cheaper to write rich, useable applications in than any other product on the market. More, better apps naturally led to a larger client base.
ASP.NET was supposed to deliver the same kind of experience to the web. It made things MUCH better. But the bottom line is that the web is far more complicated and limited than the Windows API - and the same kind of designer just isn't suited for it. With Quartz, I anticipate being able to bring the same kinds of efficiency and economic power to my clients that VB brought to Windows clients in the early to mid 90s.
Sorry for the ramble heh I kept thinking up new points and persepectives I thought the fellow slashdotters might enjoy reading. I'm not trying to flame anyone with the above - personally I wish that SVG had taken off, I was looking forward to developing with it...
I've seen quartz run. I've even had lunch with some of the product team. The same event I saw a demo of WPF (I forget what Microsoft is calling it now - most everything I saw at the conference was identified by codename - but WPF stands for Windows Presentation Foundation... I think it might be "Avalon" now)
I think you have some misguided concepts about how Quartz works. Quartz is just a web designer - with support for rich UI features. It has compatibility target levels - if you want to run on NN 4.0 it supports that with a reduced feature set. If you're interested, my favorite feature of Quartz I saw was actually the XML/XSLT WYSIWYG support (a close runner up is the AJAX RAD)... and that feature is compatible with no-CSS HTML 4.0! I know becuase there's an app I helped write (in 2002 without WYSIWYG) using the exact technology.
If you are targeting the WPF that still doesn't mean that your users will need Windows Vista to run it. Hell, the first demo of WPF I saw was on a Windows XP box. You don't really think the guys developing WPF have been doing it on Vista? Vista isn't even alpha yet... so WPF has been running on XP for some time already. The release of WPF will be back ported to at least XP, and I've heard ME and 2000 server are distinct possibilities. 98 won't be supported, but then 98 has been EOLed for a while. Besides... if you can't be bothered to upgrade your operating system once in eight years, then you obviously aren't interested in taking advantage of the latest technology anyways, right?
Others have made good replies to this but thought I'd add my $0.02
Generally, image manipulation will leave a signature of some sort on the file - do a fourier transform (view the image as frequency data as opposed to spatial) and you can see some of them pretty clearly. They generally show up as very low or very high frequency noise distributed more or less uniformly around the origin. Then there's edge detection; most computer-based photo manipulation creates or erodes edges and a basic edge detection algorithm will show the problem to most human observers.
As mentioned by others, a low quality original can make it much harder to detect manipulation.
Again, don't tell me what's wrong with patents. You say it doesn't succeed on 3 out of 4 of my wishlist items for an ideal intellectual property protection system. I say that it succeeds but that there is room for improvement - and neither of us really have any facts to back up our opinions. That's not the point. The point is neither you nor anyone else is suggesting a system that will better achieve those goals...
Re:Patnets brought to their logical conclusion
on
Supreme Court spurns RIM
·
· Score: 3, Insightful
1. Insert standard slashdot spelling nazi bitchfest here
2. I hear lots of people say that patents are 'filed' (perhaps you mean flawed???) and need to be done away with. But I never hear anyone make a recommendation for a replacement. Patents serve an incredibly useful purpose in our economy - they are there to protect the investment required to develop intellectual property while still allowing its wide dissemination, at the same time making a provision for late-term royalty free use.
I agree that there are problems with the way patents are being handled. I do not agree that they are useless.
I challenge you and anyone else who says that patents should be abolished to suggest an alternative solution that would:
- Protect the investment of innovators, allowing them to market a product at a markup required to recoup their initial investment before cheap 'me-too' alternatives are released on the market
- Protect the interests of other businesses by providing a legal framework for licensing ideas to other entities
- Protect the interests of the public by ensuring that free market competition would eventually take place
- Protect the interests of humanity and the sciences by ensuring that the details of ideas are stored in a large public archive
Unless your solution solves all those needs, it's worse shit than the system we currently have. The problem with our current system is that there's no patent-challenge built-in; if you think that your product is a non-infringing invention, there's no way to vet that out other than build the damn thing and wait for your court summons.
Amazon's one-click patent is an excellent example. It was clearly a non-novel use with publicly available prior art, but they were granted the patent anyways. Noone can take them to court to challenge the patent's legitimacy - they have to take you to court. And until then the sword of damoscles hangs over the heads of all e-commerce sites that utilize streamlined checkout processes.
Make it. Please. MMORPG-style (maybe FPS)... the show even has some nice mechanics that can translate directly into good gameplay mechanics - for intance the way that the gate's controller can deny entry to their world by continually dialing out. The stargate in general makes a great gameplay mechanic by seperating the universe into discrete atomic units that are easily managed.
You can even mirror the shows' plotline - several distinct races with shifting alliances try to carve out a safe place in their world.
Hell, even include Go'auld recruitment - when a player gets infected their character gets a massive boost in stats, can raise an army of (loyal or disloyal) soldiers, access to advanced technology, and a new reward system to encourage them to throw old loyalties aside and become the brutal dictator inside.
Unpopulated worlds get their own dungeons teeming with artifacts. For either single play or squad play with lots of fun puzzles. Some dungeons should have an immovable artifact of great power - get the artifact, it should be easy to get control of the gate and set yourself up a little empire. Or, if you're part of a clan, take the world for your clan and get promoted...
I would play. And I've avoided MMORPG succesfully for a decade. But I would still play.
And that's why our robots have to actually fight - to prove that. Of course, team warfare is different than singleton combat... and I would imagine that every now and then you'd have to actually fire up the factory and make a few million kill-bots... point being that people STILL AREN'T DYING
So each nation develops prototype robot warriors - when in a dispute each nation sends one and lets them duke it out. The idea behind war is not to kill your enemy - it is to convine them that you are right, or at least that you can back up your argument with more force than they can muster.
If after a number of fights between my robot and yours we can look at the fight statistics and conclude that I can build enough robots to soundly beat the number of robots you build, then haven't I accomplished my objective?
Yeah I say get rid of cross country lines; all the lines should be about the same as the tax-payer subsidized commuter flights.
I used to pay $80 to take a flight from Greensboro, NC to Pittsburgh, PA once a week! The plane was never full, either. No way they were making money on that flight...
Get rid of commuter flights. Give me mid-range high speed train lines any day of the week! From my local airport, we offer commuter flights to Charlotte (18 minutes airborne time!!!), NYC, Washington DC, Atlanta, and Pittsburgh. If they just provided express service from Charlotte to those cities I'd be happier... if they threw in Orlando or Miami I'd be ecstatic.
Hell while I'm making a wish list, New Orleans, Detroit, and St Louis wouldn't be bad either.
It takes me $230 round trip to drive to Fort Myers, FL to visit my family. I'd be willing to pay a little MORE than that to ride a train - both to reduce wear and tear on my car, the reduced transit time, and of course the increase in productive use of my time (i.e. gaming/coding/studying/writing/sleeping instead of driving)
The big problem is the BIOS - apples (even the new intel laptops) use proprietary BIOS code, and you can't acquire the chip without a license. Not sure if Apple sells upgrade chips, but if they don't, then there's just no legal way to build a clone that will actually run OS X.
If you can buy the operating system seperately (should be soon) then you can put together all the same hardware... except for that chip. FYI, that bios should also prevent Windows from installing on Apple hardware for now, as well.
On the flip side, you can install a BIOS in an EEPROM pretty easily, and if you're willing to pirate it out of a legit apple, it'd probably be pretty easy to get OS X installed and running once you get that out of the way.
Cheap Dell Apple clones? No way. Cheap Chinese pirate Apple clones? Absolutely.
Be glad you've never heard of it - it's a piece of crap. It worked even less well than MythTV for me (see my other post in this thread for an account of that experience)
First MythTV refused to use an NTSC-compatible timing. So I hacked it up and took out all the timings and added one that worked.
It never actually let me watch a show though. I probably spent 20 hours installing and configuring, reinstalling and reconfiguring before I gave up and installed windows.
MCE is nice but I like showshifter better. Regardless, I haven't looked back since.
Slashdot Mirror
I would guess the number would be more like a few thousand - but fundamentally I'm in agreement with you.
Harnessing the power of the atom certainly gives us enough power to sterilize the land (I don't think anything we do short of boiling them right off would sterilize the ocean)
Then again the same thing could be done more cheaply with a few hundred thousand samarium bombs - except the cockroaches might survive that.
Well... flash is several orders of magnitude slower than DDRAM so I would say it probably compares quite well!
The sustained transfer rate is 150 MB/s... when was the last time you wrote your whole 128 MB USB flash whatever whatever in under a second?
Anywhere close to a second?
Ten seconds?
My theory is they chose a non-standard port specifically to limit their usability - by using a typically blocked port, they've stopped all the slashdotters who surf from work from overloading their service, for instance...
Gigabyte is the name of the company, not the size of the product.
Actually, when I say it's vetted, I was referring to the Java sandbox model - it's a security model that has worked for sometime before Microsoft ever thought to include it in their runtime.
I don't know how buggy it is or not. I don't know how secure it is or not. I know it's cryptographically secure, in that there's a lot of math being done to verify that the permission set granted is the one intended by the owner of the computer, and that the code being run is the code that the owner believes he has acquired. I know it's very atomic - permissions can be granted or restricted to very atomic operations. I know it's supported by the Common Language Runtime and enforced on a per-method-call basis.
And I know that it was released around the same time during which Microsoft halted all new code development for four weeks to give developers time to address security concerns in every single piece of software they make.
That's the kind of effort I tend to trust. They've been stressing security ever since, and I doubt they just kind of forgot to check this.
And as I said, the security model isn't new. It's been in production use since 2002, and I saw CTP's of it back in 2000. AFAIK, there are no changes planned in it before Vista is released.
Not only that, but VB6 at least stored part of its design-time properties seperately from the form itself and there was no way to really tell what it would be doing.
I was commenting that this is no longer the case - VB has become a world class language now, and coding in it actually feels like you're writing code now.
Even DirectX is accesible to the VB developer now - at performances that are within a few percentage points of C#.
But that doesn't really address the underlying question - which is, has it ever been, previous to the current VB.NET environment, advantageous to write code in VB?
I think so. Say what you will about performance and sloppy code - I would rather take 15 minutes to draw a form in a designer and another 30 to wire up its events than take 4 hours to hand code a bunch of windows API stuff just to draw my form and get everything just how I want it and another 2 or 3 hours wiring up the events by hand - a common design experience in C++ in the early days.
VB revolutionized the world by making it easy and cheap to write software. No mind that the software didn't perform as well as it's C-bequeathed brethren. No matter that there were things you simply couldn't do. The fact is, it has spent decades meeting the needs of enterprise development more economically than any of its competitors - and this one fact has enabled computers to become the financial powerhouses they are today.
A computer without software to run is virtually indistinguishable from high-silicon-content sand. And not quite as useful. VB made it so anyone with a half-baked idea and a little money could afford to have their own apps custom written for them.
I've got a better idea - every election the feds generate a public/private keypair. Load up all the machines with the public key. Each voting machine has open code that takes a voter record (in a government-defined format), encrypts it, stores it, and clears its system memory. At the end of the day each machine spits out whatever flash memory format is being used to store the data - including a tamper-proof hash. These cards are carried to the vote tallying machine - built by a third party - which loads them into memory and physically locks access. Once every single chip is accounted for, the government releases the private key, everything gets decrypted and tallied, and everyone goes home satisfied.
Within the election machine itself no candidate is identified by party - have them all identified by unique ID, translated via a code table to the names shown on screen. The unique ID should be a random string of bits - as long as its unique, there was no way to predict it, and noone had any control over how it was chosen - you'll be secure. The voting machines cannot purposely skew voting results one way or the other if they don't know who they're voting for
There's no way for you to "override" the machine and do the braking yourself, Fred Flinstone style.
Gears, handbrake, ignition switch.
Especially that reverse gear. Let me tell you, nothing stops a car faster than that reverse gear...
Sure, it's great for learning to program, but it's not good for much beyond that
VB.NET compiles almost identically to C# now. In the bad old days what you say might be true but not any more.
"Visual" anything is a BAD way to go because it generates a bunch of code for you, and half the time you don't even know what the hell it does.
VB.NET generates all its code into your file. You edit the code it generates, the designer picks up the changes and visa versa.
IANAS but I'll answer your questions as best I can...
Use of a little unsharp masking: this is the digital equivalent of tweaking the focus of the enlarger in a chemical darkroom-- a practice that I believe was very common in the days of chemical microphotography
It is the digital equivalent - but unfortunately, in this case, the digital equivalent of an analogue process introduces severe aliasing artifacts, and can actually create or destroy spurious signals with repeated use - most scientists frown heavily on the use of sharp or unsharp masks for this reason. Paranormal researchers of integrity insist that no such mask ever be used to analyse images in their fields, in fact.
Use of the histogram controls and gamma adjustments: the digital equivalent of choosing films, papers, chemistries, and processing times and temperatures to expand or compress the tones, or shift the tonal balance of the final print (except with digital photography, we can now do this kind of stuff after the image is acquired rather than guessing beforehand)
This is mostly okay - as long as they're used responsibly. The trick is to avoid clipping artifacts - use these controls to normalise the data. Gamma is especially useful as it preserves the digital signal while allowing you to tweak sensitivity settings to bring out details in the image.
Use of hue and saturation controls: the digital equivalent of choosing films, papers, and filters with different color sensitivities
These types of techniques are widely used in astronomy (or equivalent techniques) - quite useful when part of your data metric actually lies outside the human spectrum of vision. Other fields use them to varying degrees - forensical analysis in particular is quite fond of them. Again, be careful not to clip your saturation when you're adjusting it, and you'll be okay.
Use of different kinds of noise suppression: in some respects, similar to controlling the grain of chemical images, but also needed to manage artifact specific to digital cameras
That depends - if you're using a standard Filter window or fourier transform, absolutely not... those techniques always insert spurious signals into data. Gaussian anti-noise techniques should be ok as a final pass but not before and not repeatedly. Most other anti-noise techniques I'd be very skeptical of.
To check if something's ok, generate a screen of static - just random noise. Then apply your technique to it over and over and over. If the static flattens out to a solid color, or clumps together becoming more defined, then you've shown that your technique can mask or amplify signals in your image.
BTW, if you want to do the kinds of things you mention above in a scientifically respectable way, you might want to look into high dynamic range imagery, which solves some of the clipping problems I mentioned above.
Heh sidegrade is more like it! Why the hell would Microsoft work so hard on a kernel just to throw it away in their next os drop? And load it with such useless crap while they're doing it!
But from Microsoft's point of view, ME was available as an upgrade to 98 - at least from a pricing standport - and is still supported officially...
The rich design features available will be run in a sandboxed mode just like Java - its a vetted security model that's extremely low risk. It'll be either version 2 or 3 of the technology, too, so it's not a virgin product (the same sandbox model has been support by the .NET CLR since version 1.0)
It doesn't do anything you can't already do.
.NET has been that it "degrade gracefully"... at PDC 2000, when they first announced .NET CLR and ASP.NET, practically every speaker used that phrase when they'd explain how web controls work.
It does do it better and with better design time support.
It can target XHTML/CSS3/ECMASCRIPT browsers (I've heard it's cross browser ECMASCRIPT. I'm expecting some features, at least, to only be available in IE however)... a key of Microsoft web design technology since
Microsoft understands that ActiveX was a shitty technology because of its platform dependence. The marketplace of rich UI controls they were expecting never really materialized. Now, ActiveX was the best possible at the time... JavaScript was hardly a gleam in Sun's eye, and it was a logical extension of the Netscape plugin module, which was itself a non-standard part of HTML at the time, but the point is that it's technological limitations severely hindered it.
Microsoft gets this. They've been working hard not to make the same mistakes again. Yes, some features of Quartz depend on the Windows Presentation Foundation (I believe they're calling it Avalon???) to be present - and if you've seen a WPF demo you'll understand why... javascript will never be able to present the type of rich UI I saw.
It may not be as open as SVG or canvas... but that doesn't really matter for people writing extranet sites, which is a large portion of active web development - in my seven years as a web technology consultant I have only been asked to work on a consumer-facing website 3 or 4 times, out of a total of approximately 50 I've worked on.
What matters to me is my clients for a majority have a fixed install base. Getting a component that I can redistribute or get my clients to install via automatic update is a huge boon in this situation - and provides an incredibly rich design time experience that allows me to deliver responsive software widely.
I don't care if that's SVG, canvas, WPF, or some bits picked off a deep-space radar dish in the New Mexico desert - as long as I can use it, that's good enough for me. The lifetime of corporate applications varies between 3 and 15 years. I have yet to see a single solution I've developed go 5 years without a retrofit - as businesses grow, their needs change and technology evolves to better fill them. Once you've ROIed you can usually build something better, cheaper, that'll help your business grow faster... and many of my clients choose to take advantage of that option shortly after that point in time.
I'm not worried that my applications will outlive Microsoft's support of WPF, nor am I worried that Microsoft will suddenly choose to limit my use of it. It's in their best interest for WPF to be as widely disseminated as possible. I don't need WPF to be open - that's why I buy software, so someone else can worry about the bugs while I go get my job done. Closed or open, I have no intention of ever cracking open the source files.
My only concern is building my apps better and cheaper than anyone has ever built apps before. Quartz will help me do that, and I can't wait to get my hands on it. Windows was fundamentally succesful in the operating system market back when they had something that could be called competition because their programming language, Visual Basic, was far cheaper to write rich, useable applications in than any other product on the market. More, better apps naturally led to a larger client base.
ASP.NET was supposed to deliver the same kind of experience to the web. It made things MUCH better. But the bottom line is that the web is far more complicated and limited than the Windows API - and the same kind of designer just isn't suited for it. With Quartz, I anticipate being able to bring the same kinds of efficiency and economic power to my clients that VB brought to Windows clients in the early to mid 90s.
Sorry for the ramble heh I kept thinking up new points and persepectives I thought the fellow slashdotters might enjoy reading. I'm not trying to flame anyone with the above - personally I wish that SVG had taken off, I was looking forward to developing with it...
I've seen quartz run. I've even had lunch with some of the product team. The same event I saw a demo of WPF (I forget what Microsoft is calling it now - most everything I saw at the conference was identified by codename - but WPF stands for Windows Presentation Foundation... I think it might be "Avalon" now)
I think you have some misguided concepts about how Quartz works. Quartz is just a web designer - with support for rich UI features. It has compatibility target levels - if you want to run on NN 4.0 it supports that with a reduced feature set. If you're interested, my favorite feature of Quartz I saw was actually the XML/XSLT WYSIWYG support (a close runner up is the AJAX RAD)... and that feature is compatible with no-CSS HTML 4.0! I know becuase there's an app I helped write (in 2002 without WYSIWYG) using the exact technology.
If you are targeting the WPF that still doesn't mean that your users will need Windows Vista to run it. Hell, the first demo of WPF I saw was on a Windows XP box. You don't really think the guys developing WPF have been doing it on Vista? Vista isn't even alpha yet... so WPF has been running on XP for some time already. The release of WPF will be back ported to at least XP, and I've heard ME and 2000 server are distinct possibilities. 98 won't be supported, but then 98 has been EOLed for a while. Besides... if you can't be bothered to upgrade your operating system once in eight years, then you obviously aren't interested in taking advantage of the latest technology anyways, right?
Others have made good replies to this but thought I'd add my $0.02
Generally, image manipulation will leave a signature of some sort on the file - do a fourier transform (view the image as frequency data as opposed to spatial) and you can see some of them pretty clearly. They generally show up as very low or very high frequency noise distributed more or less uniformly around the origin. Then there's edge detection; most computer-based photo manipulation creates or erodes edges and a basic edge detection algorithm will show the problem to most human observers.
As mentioned by others, a low quality original can make it much harder to detect manipulation.
Again, don't tell me what's wrong with patents. You say it doesn't succeed on 3 out of 4 of my wishlist items for an ideal intellectual property protection system. I say that it succeeds but that there is room for improvement - and neither of us really have any facts to back up our opinions. That's not the point. The point is neither you nor anyone else is suggesting a system that will better achieve those goals...
1. Insert standard slashdot spelling nazi bitchfest here
2. I hear lots of people say that patents are 'filed' (perhaps you mean flawed???) and need to be done away with. But I never hear anyone make a recommendation for a replacement. Patents serve an incredibly useful purpose in our economy - they are there to protect the investment required to develop intellectual property while still allowing its wide dissemination, at the same time making a provision for late-term royalty free use.
I agree that there are problems with the way patents are being handled. I do not agree that they are useless.
I challenge you and anyone else who says that patents should be abolished to suggest an alternative solution that would:
- Protect the investment of innovators, allowing them to market a product at a markup required to recoup their initial investment before cheap 'me-too' alternatives are released on the market
- Protect the interests of other businesses by providing a legal framework for licensing ideas to other entities
- Protect the interests of the public by ensuring that free market competition would eventually take place
- Protect the interests of humanity and the sciences by ensuring that the details of ideas are stored in a large public archive
Unless your solution solves all those needs, it's worse shit than the system we currently have. The problem with our current system is that there's no patent-challenge built-in; if you think that your product is a non-infringing invention, there's no way to vet that out other than build the damn thing and wait for your court summons.
Amazon's one-click patent is an excellent example. It was clearly a non-novel use with publicly available prior art, but they were granted the patent anyways. Noone can take them to court to challenge the patent's legitimacy - they have to take you to court. And until then the sword of damoscles hangs over the heads of all e-commerce sites that utilize streamlined checkout processes.
Make it. Please. MMORPG-style (maybe FPS)... the show even has some nice mechanics that can translate directly into good gameplay mechanics - for intance the way that the gate's controller can deny entry to their world by continually dialing out. The stargate in general makes a great gameplay mechanic by seperating the universe into discrete atomic units that are easily managed.
You can even mirror the shows' plotline - several distinct races with shifting alliances try to carve out a safe place in their world.
Hell, even include Go'auld recruitment - when a player gets infected their character gets a massive boost in stats, can raise an army of (loyal or disloyal) soldiers, access to advanced technology, and a new reward system to encourage them to throw old loyalties aside and become the brutal dictator inside.
Unpopulated worlds get their own dungeons teeming with artifacts. For either single play or squad play with lots of fun puzzles. Some dungeons should have an immovable artifact of great power - get the artifact, it should be easy to get control of the gate and set yourself up a little empire. Or, if you're part of a clan, take the world for your clan and get promoted...
I would play. And I've avoided MMORPG succesfully for a decade. But I would still play.
And that's why our robots have to actually fight - to prove that. Of course, team warfare is different than singleton combat... and I would imagine that every now and then you'd have to actually fire up the factory and make a few million kill-bots... point being that people STILL AREN'T DYING
So each nation develops prototype robot warriors - when in a dispute each nation sends one and lets them duke it out. The idea behind war is not to kill your enemy - it is to convine them that you are right, or at least that you can back up your argument with more force than they can muster.
If after a number of fights between my robot and yours we can look at the fight statistics and conclude that I can build enough robots to soundly beat the number of robots you build, then haven't I accomplished my objective?
Yeah I say get rid of cross country lines; all the lines should be about the same as the tax-payer subsidized commuter flights.
I used to pay $80 to take a flight from Greensboro, NC to Pittsburgh, PA once a week! The plane was never full, either. No way they were making money on that flight...
Get rid of commuter flights. Give me mid-range high speed train lines any day of the week! From my local airport, we offer commuter flights to Charlotte (18 minutes airborne time!!!), NYC, Washington DC, Atlanta, and Pittsburgh. If they just provided express service from Charlotte to those cities I'd be happier... if they threw in Orlando or Miami I'd be ecstatic.
Hell while I'm making a wish list, New Orleans, Detroit, and St Louis wouldn't be bad either.
It takes me $230 round trip to drive to Fort Myers, FL to visit my family. I'd be willing to pay a little MORE than that to ride a train - both to reduce wear and tear on my car, the reduced transit time, and of course the increase in productive use of my time (i.e. gaming/coding/studying/writing/sleeping instead of driving)
Are you reading, America?
280 km/h!!!!
Somewhere around 150 mph???
Yeah. I'd love to drive a couple hours to a major metropolitan city, take a train down to Florida, and have my dad pick me up at the station.
I'd go to Florida 2-3 times a year if that were the case; assuming of course that train tickets would cost substantially less than plane tickets...
The big problem is the BIOS - apples (even the new intel laptops) use proprietary BIOS code, and you can't acquire the chip without a license. Not sure if Apple sells upgrade chips, but if they don't, then there's just no legal way to build a clone that will actually run OS X.
If you can buy the operating system seperately (should be soon) then you can put together all the same hardware... except for that chip. FYI, that bios should also prevent Windows from installing on Apple hardware for now, as well.
On the flip side, you can install a BIOS in an EEPROM pretty easily, and if you're willing to pirate it out of a legit apple, it'd probably be pretty easy to get OS X installed and running once you get that out of the way.
Cheap Dell Apple clones? No way. Cheap Chinese pirate Apple clones? Absolutely.
Be glad you've never heard of it - it's a piece of crap. It worked even less well than MythTV for me (see my other post in this thread for an account of that experience)
I own the PVR250.
First MythTV refused to use an NTSC-compatible timing. So I hacked it up and took out all the timings and added one that worked.
It never actually let me watch a show though. I probably spent 20 hours installing and configuring, reinstalling and reconfiguring before I gave up and installed windows.
MCE is nice but I like showshifter better. Regardless, I haven't looked back since.