They normally "don't work" only if they are incompatible with each other. If you have 3 programs designed to give you mouse gestures, I would expect issues as well honestly.
This suggests that firefox needs to take a less generic approach to addins. They could instead move toward a service provider type system. You'd be able to select from a list of "mouse gesture providers", "bookmark providers" or "auto-form fill providers".
They already do this in a way... what do you think themes are? Last I checked you could only use a single theme at a time, why can't they make it so you only use one "decorate each webpage with annoying icons and popups (cooliris)" addin?
Because if you are a web developer you need to test your sites on various browsers to check that they display properly.
And if you are a professional web developer who uses Linux (or Mac) for your platform, you will have purchased the proper tools. In this case, you'd own a copy of VMWare Workstation and run windows inside of it for testing.
Ideally, you should be able to just follow the web standards and be confident that there is no problem
Ideally there would be a reference implementation you could use. HTML has no such beast and even if you coded to "teh magic standards", it would look different across each browser. You call this a feature, I call it the biggest flaw in the whole damn web development universe. Despite the heated rhetoric of the semantic web peanut gallery, most web developers want their web applications to look and function the same no matter what the platform. Following the standards might ensure cross platform functionality, but it sure as hell doesn't ensure cross platform "lookability".
Curiously, an advantage of using Linux is that you can test various versions of IE at the same time, which you cannot do in Windows
Only if you are too cheap to purchase the right tools. Hell, you don't even need to purchase them... VMWare Server is free!
Which is the best part of BSD. I love unix and I love windows. Nobody cares if I submit new freebsd ports and happen to use a text editor in windows to create the Makefile and Outlook to mail the attachment to the port maintainers:-)
If you tried that on Linux, you'd probably be flamed to a crisp.
If one or more of your tabs do crash, your tabs are automatically reloaded and you are returned to whatever page you were on before the crash.
If they do crash recovery like the Office applications do, before loading up after the crash they'd probably pop up a dialog that says "Dear sir, your tab entitled 'Slashdot Horse Porn' has crashed, do you want me to try loading 'Slashdot Horse Porn' again?
I'll even take jab at Firefox and mention this is probably easier to do on IE7/8 because they run each tab in a separate thread.
VMWare is your friend. Yeah you gotta boot up the image first, but you can get a base image of windows installed and then take snapshot images of IE6, IE7, IE6 + XP SP2, etc. Once you are done for the day, you can roll the image back to the snapshot and in the morning boot into a virgin install of whatever OS/SP/Browser combo you choose.
And for that matter, I'd consider anybody insane to install a beta version of IE8 on anything *but* a virtual machine.
Buddy, I stopped caring about IE5.5 years ago and wish IE6 would finally die. If IE5.5 is still a target for whatever you are doing... I feel your pain and I'll buy you a shot of good iWhiskey.
Sadly, no. They decided to create some proprietary "GUI" thing so "lusers" could click and drool with a mouse. But I agree, if Microsoft really wanted to embrace standards, they'd do well to ditch the whole GUI thing and go back to the command line. Once typing "win" at the dos prompt became "legacy", they really jumped the shark, didn't they?
And for that matter, RMS needs to really hound the firefox guys to support Info documentation, doesn't he? Maybe then he could finally get firefox to run inside emacs where it belongs.
If you've been doing work in a real database like SQL server, you'll be much happier if you select another real database like PostgreSQL. MySQL is about the same as Access in terms of functionality, but with less reliability and interoperability. Sure MySQL claims to have stored procedures, views, triggers, foreign keys and other "enterprise database features that you dont really need", but every one of those "enterprise class features" has a set of disclaimers that would make Publishers Clearing House blush.
PgAdminIII is an excellent GUI that isn't as meaty as SQL Server Management Studio, but it gets most of your job done. It is still a bit more "gotta know SQL" centric though...
The big problem with PostgreSQL is that it isn't as well integrated into the.NET stuff as SQL Server (or Oracle). For example, if you like LINQ (and I like LINQ), there isn't much support for PostgreSQL. Lord knows about the ADO.NET entity framework. This is the most promising.NET stuff for now.
MySQL, as much as I hate it, has a bit better.NET integration. That is about the only thing it has going for it though... unless you consider pain and misery a feature.
We really like custom login forms and even in Web v0.2 (beta), people were using HTML and cookies to log into their CGI apps.
Both of the popular web servers (IIS & Apache) will let you hook their authentication/authorization phases. Both of them speak DIGEST and both will let you pass around blobs of goo so you can hook up your authentication and authorization handlers to your response handler.
You'd still have to come up with a way to customize the login form because nobody likes the browser ones. I'm not sure how this part would work because as you say, DIGEST authentication requires the browser popup up it's own dialogs.
I actually think there is one more catch. Who does the hashing of the password? Obviously you dont want the client to give you a plaintext password. I'll have to look into this as I recall there being a way to get it to do MD5. If so, unless you are starting fresh, what do you do with the existing 20,000 users passwords in your database who are hashed some other way?
When you look at it... I think the way we handle our authentication/authorization on the web is the perfect blend of built-in and roll-your-own. It is easy to understand, easy to customize and our languages and frameworks offer a fairly standard way to manage things.
The fun problem is the "web way" of doing authentication/authorization breaks down the second you toss "real" applications in the mix. Have fun getting your client-side "real" program.exe talking with any kind of web service that uses HTTP cookies for session and HTML forms to log in. They are easy(er) if you use DIGEST though:-)
While I dont thing the concept eminent domain should apply to granting right of way to private industry, if I was the wal-mart/city lawyer I'd argue:
- Perhaps some kind of state zoning pretty much made it feasable to build the mall in exactly one location. - The mall will create jobs and tourism (hah..)
Basically, you can probably recycle many of the arguments used to get state/city subsidies for stadiums and such.
But again, I don't agree with it, but I'm sure there is more to these stories then first meets the eye.
You can even create single purpose ssh keys which execute one command each
Close, but it is still not part of the specification itself.
Does javascript handle this? Can I use it in an AJAX call? Does it work out of the box from the CPAN libraries? Can you do it in PHP with a normal set of compile-time flags? Can you have anonymous clients authenticate themselves using a login/password (i.e. a flickr like web service?)
Want the ultimate proof that SOAP and XML-RPC is a failed specification? Every single javascript libraray (Prototype, jQuery, mooTools) doesn't support either one, yet every virtually every single XHttpRequest made is pretty much an RPC call.
Except WSDL is as evil to look at as any XSD file I've seen and last I checked (which admitidly was many moons ago), Perl will not generate one based on your code and maintain it for you automagically.
Its not a problem making it work with different scripting languages
Correction. Most of them have good support for *consuming* other peoples SOAP service. Try to *create* a SOAP service in, say, Perl. I've never tried in other languages, but that is only because after trying in Perl, I swore off SOAP forever.
SOAP probably kicks some kind of alien psuedo-ass in a strongly typed language like Java or C# where the compiler can easily sues out your object model. Even then, since there is no standard for authentication or session management, you are left with only a hollow shell of a solution.
First, figure out multiple ways to build your large scale project.
This is required as part of your environmental impact statement. It is done to figure out your next point:
For example, if you're building a road, maybe you can run it through points A, B, or C.
You make the assumption that A,B and C are all equal cost and just as good. I'll punch a whole in this easy:
- Mountain Passes. Sometimes there is only one route for 500 miles. - Transit Lines. To get any kind of ROI, you gotta have stops in pretty precise locations, often giving you only a few blocks of wiggle room. Lets not forget if you plan to tunnel, there might be geological issues that make one route much, much cheaper. - Utilities get in the way. What if your route crosses over some other dudes utility and they refuse to budge? What if that utility was 500 miles long?
in exchange for the right of Eminent Domain over your property for the next 3 years
What uninteresting thing are you building that will only be around for three years. Last I checked the interstate highway system is older then three years. Last I checked, the Hover Dam is older then three years.
In that case, I would argue that the people have spoken.
You know what? Fuck the people. Sometimes you have to cram stuff down their throats. That is what makes a good politician... people objected to most of the bridges that were built in New York. You think people have issue with them now? People object to every sewer treatment plant and every landfill created on our planet. Yet without them, civilization could not exist.
Being libertarian means you haven't grown up politically. You can still subscribe to free market ideals and understand that sometimes you need a little central planning. Look how much more productive and environmentally "clean" a well organized, properly zoned city like New York is compared to an un-zoned, unplanned suburb like Anytown USA. Sure those suburbs were cheap to build, but look how costly they are once gas prices have gone up? Look how costly they become when they get even the least bit dense (and I'm talking "cost" in the economic sense, so I'm including Time and Psychological costs as well).
Life gets really strange when you grow up and realize you can be a democrat and remain compatible with your ideas of a free market economy.
Why can't one of these new fanged RPC thingies HANDLE FUCKING AUTHENTICATION and AUTHORIZATION in a standard way? And if there *is* a standard way, I dont know about, WHY IS IT NOT OBVIOUS AND EASY!?
Why do I *have* to roll my own fucking way to hold a session using SOAP? Why couldn't they have some concept of a cookie or some easy way to maintain state that was buried deep within the protocol so it was transparent to me?
All these XML based RPC specs suck ass. Some architechure astronauts snorted a bunch of crack and decided to do crazy stuff like let you use SMTP for a transport. It would be much simpler to just pick a damn transport and stick with it so you can exploit the charctaristics of said transport (like cookies for HTTP or ??? for SMTP).
Unless you trust your IDE and dont need to get say, a C# client to talk to a mod_perl backend, you are in a world of hurt. For that matter, if you use any kind of dynamic language (PHP, Ruby, Perl), god help you with anything. You'll be writing a fuckton of XML that you'll have to keep in sync with your codebase. Forget any kind of automated tools.
Given the fact that the NIMBY factor for power lines,power plants, nuke power, roads, dams, whatever is so high, the odds of at least one person objecting is virtually 100%.
Therefore, if you would like to have nuke power, power lines, roads, high speed rail, whatever, you will *need* to force somebody to fucking move for the greater good. Otherwise, you will never get the right-of-way to make your project happen. We have granted our government the ability to force people to fucking move out of the way.
We call this Eminent Domain.
Why anyone wants Federal control of anything is beyond me
Given that large scale projects are impossible without forcing somebody to move, do you feel comfortable granting eminent domain to private industry?
If you say "make it all states rights" given that many of these large scale projects affect multiple states, you'll wind up with heavy federal oversight anyway. Let states do it all, and they'll sue eachother when the other guy builds a huge damn. They'll sue when their state law conflicts with the other state law. You either get federal agencies for interstate projects, or you get a metric assload of federal judicial "weight".
The main advantage I can see is that it allows coding in languages other than ActionScript -- theoretically, anything that runs on the CLR/DLR.
That is exactly the advantage. If you are already familiar with any of.NET or it's languages and you already own Visual Studio, you are good to go. Bonus points are assigned to the fact that since it is a little brother of WPF, it doesn't take much work to fold in your existing codebase into your silverlight codebase.
Flex requires learning a whole new product suite, a whole new language, a whole new IDE... everything you know.
Because 99% of all normal, non nerd internet users would rather watch the video within a website rather then choose between a dozen file formats that sometimes stream properly, sometimes confuse the browser "what kind of file is this, good sir?", sometimes dont stream but try to download all 1.2 gigs before loading the client.
Video on the inter-tubes exists because of the embedded video client. Take away the embedding, and we'll be back to square one.
But as long as it satisfies some Linux nerd (who choose their OS knowing full well they would not be able to enjoy various forms of media on the internet), the ends justify the means, right?
Not to mention most programmers are comfortable with Visual Studio/.NET/WPF and can take their existing skillsets and quickly crank out a pretty decent silverlight applet. If you go flex, you have to invest in a whole different product suite, a whole different language, whole different everything. Flex is great if your shop is mostly designers who are familiar with the Adobe suite. Silverlight is great for shops who are mostly developers.
PS: Mono/Moonlight will be the Linux silverlight client.
Or perhaps it just costs 10x more per mile to deploy any kind of utility/road/anything in a city. You've got to deal with permiting, zoning, mitigation, anti-development hippies, NIMN's (not in my neighborhood), parked cars to tow, drunk bums passed out on your construction site, utility relocation, you name it and you have to pay for it.
It is easy to do anything in the suburbs. No zoning, less red-tape, less density, less 100 year old wood stave pipes, whatever.
Poverty has nothing to do with it. If anything, there are probably more affluent 20-something condo owners living in a city who would jump all over FiOS then there are suburban house dwellers.
would they be able to demand access to the source code
I dunno, who's got the most nukes?
If weapons systems were GPLd
Then the military would have to way to keep the F-16's weapons control software inhouse. Why would you need to rely on a crashed jet when you could just carry a floppy disk out the front door with the same thing?
Or in GPL-only land do we grant our government the right to create closed-source software for national security purposes? What about civilian aircraft? If Boeing had to GPL all its software, wouldn't airbus be able to lift that work and kick our ass in the aerospace industry? Or is the idea the entire globe has to sign a "GPL only" treaty and make it a condition for joining things like the WTO?
And to reply to myself, once you start self-reflecting, you might start thinking "well, I like GPL but the government needs to provide a way for it to function". I promise whatever direction your line of thinking takes you, you'll get something that looks a bit like the skeleton form of copyright law.
While I dont have any proof and I dont know enough of it's history, I imagine copyright law existed in some form throughout history. I also suspect it was never under the name "copyright law" because most of it grew organically... court case by court case. Hell, even if we tossed out copyright as we know it, the we'd probably grow right back into it through decades of case law.
It would be as close to BSD as you could possibly get only there would be no way to enforce the clause about including the original authors. The liability clauses would probably be enforceable though. I suspect many would probably include the original author names at the top of their code just so other programmers who look at it know the lineage and history of the codebase. I also suspect many would give back patches and code changes because there is a strong benefit to keeping your codebase in sync with the mainline.
GPL would not exist at all because it uses copyright to "crack open" any codebase who incorporates the license. The "crack open" nature of GPL is what makes it a good fit for certian types of projects, but it also provides a strong dis-incentive toward "lifting" the code without giving back. Since people love to "lift" code (or music/video/whatever) because it is easy and free, the GPL needs teeth or it's main goals wouldn't pan out. For example without copyright, Apple could lift GPL code like Linux just as easy as they "lifted" the BSD codebase and the Linux crew would be powerless to stop them.
GPL is effective because you can sue people who do not follow it. The only reason you can sue people is GPL hinges on copyright law. Take away copyright law and GPL has no teeth.
Bottom line is GPL advocates *must* embrace, endorse and respect copyright law. GPL and the "community driven" software ecosystem that grew around it could not exist without copyright. Those who think their beloved GPL could exist without copyright need to do some serious self-reflection.
Slashdot Mirror
They normally "don't work" only if they are incompatible with each other. If you have 3 programs designed to give you mouse gestures, I would expect issues as well honestly.
This suggests that firefox needs to take a less generic approach to addins. They could instead move toward a service provider type system. You'd be able to select from a list of "mouse gesture providers", "bookmark providers" or "auto-form fill providers".
They already do this in a way... what do you think themes are? Last I checked you could only use a single theme at a time, why can't they make it so you only use one "decorate each webpage with annoying icons and popups (cooliris)" addin?
Because if you are a web developer you need to test your sites on various browsers to check that they display properly.
And if you are a professional web developer who uses Linux (or Mac) for your platform, you will have purchased the proper tools. In this case, you'd own a copy of VMWare Workstation and run windows inside of it for testing.
Ideally, you should be able to just follow the web standards and be confident that there is no problem
Ideally there would be a reference implementation you could use. HTML has no such beast and even if you coded to "teh magic standards", it would look different across each browser. You call this a feature, I call it the biggest flaw in the whole damn web development universe. Despite the heated rhetoric of the semantic web peanut gallery, most web developers want their web applications to look and function the same no matter what the platform. Following the standards might ensure cross platform functionality, but it sure as hell doesn't ensure cross platform "lookability".
Curiously, an advantage of using Linux is that you can test various versions of IE at the same time, which you cannot do in Windows
Only if you are too cheap to purchase the right tools. Hell, you don't even need to purchase them... VMWare Server is free!
BSD users use BSD because they love Unix.
Which is the best part of BSD. I love unix and I love windows. Nobody cares if I submit new freebsd ports and happen to use a text editor in windows to create the Makefile and Outlook to mail the attachment to the port maintainers :-)
If you tried that on Linux, you'd probably be flamed to a crisp.
If one or more of your tabs do crash, your tabs are automatically reloaded and you are returned to whatever page you were on before the crash.
If they do crash recovery like the Office applications do, before loading up after the crash they'd probably pop up a dialog that says "Dear sir, your tab entitled 'Slashdot Horse Porn' has crashed, do you want me to try loading 'Slashdot Horse Porn' again?
I'll even take jab at Firefox and mention this is probably easier to do on IE7/8 because they run each tab in a separate thread.
VMWare is your friend. Yeah you gotta boot up the image first, but you can get a base image of windows installed and then take snapshot images of IE6, IE7, IE6 + XP SP2, etc. Once you are done for the day, you can roll the image back to the snapshot and in the morning boot into a virgin install of whatever OS/SP/Browser combo you choose.
And for that matter, I'd consider anybody insane to install a beta version of IE8 on anything *but* a virtual machine.
Buddy, I stopped caring about IE5.5 years ago and wish IE6 would finally die. If IE5.5 is still a target for whatever you are doing... I feel your pain and I'll buy you a shot of good iWhiskey.
Sadly, no. They decided to create some proprietary "GUI" thing so "lusers" could click and drool with a mouse. But I agree, if Microsoft really wanted to embrace standards, they'd do well to ditch the whole GUI thing and go back to the command line. Once typing "win" at the dos prompt became "legacy", they really jumped the shark, didn't they?
And for that matter, RMS needs to really hound the firefox guys to support Info documentation, doesn't he? Maybe then he could finally get firefox to run inside emacs where it belongs.
Your post has a flaw.
What's the point if they all render the exact same?
if they all render the exact same?
render the exact same
Rule number one of web development is that no browser renders anything exactly the same. Not even the same version on the same platform.
If you've been doing work in a real database like SQL server, you'll be much happier if you select another real database like PostgreSQL. MySQL is about the same as Access in terms of functionality, but with less reliability and interoperability. Sure MySQL claims to have stored procedures, views, triggers, foreign keys and other "enterprise database features that you dont really need", but every one of those "enterprise class features" has a set of disclaimers that would make Publishers Clearing House blush.
PgAdminIII is an excellent GUI that isn't as meaty as SQL Server Management Studio, but it gets most of your job done. It is still a bit more "gotta know SQL" centric though...
The big problem with PostgreSQL is that it isn't as well integrated into the .NET stuff as SQL Server (or Oracle). For example, if you like LINQ (and I like LINQ), there isn't much support for PostgreSQL. Lord knows about the ADO.NET entity framework. This is the most promising .NET stuff for now.
MySQL, as much as I hate it, has a bit better .NET integration. That is about the only thing it has going for it though... unless you consider pain and misery a feature.
The problem is #1, not really #2.
We really like custom login forms and even in Web v0.2 (beta), people were using HTML and cookies to log into their CGI apps.
Both of the popular web servers (IIS & Apache) will let you hook their authentication/authorization phases. Both of them speak DIGEST and both will let you pass around blobs of goo so you can hook up your authentication and authorization handlers to your response handler.
You'd still have to come up with a way to customize the login form because nobody likes the browser ones. I'm not sure how this part would work because as you say, DIGEST authentication requires the browser popup up it's own dialogs.
I actually think there is one more catch. Who does the hashing of the password? Obviously you dont want the client to give you a plaintext password. I'll have to look into this as I recall there being a way to get it to do MD5. If so, unless you are starting fresh, what do you do with the existing 20,000 users passwords in your database who are hashed some other way?
When you look at it... I think the way we handle our authentication/authorization on the web is the perfect blend of built-in and roll-your-own. It is easy to understand, easy to customize and our languages and frameworks offer a fairly standard way to manage things.
The fun problem is the "web way" of doing authentication/authorization breaks down the second you toss "real" applications in the mix. Have fun getting your client-side "real" program.exe talking with any kind of web service that uses HTTP cookies for session and HTML forms to log in. They are easy(er) if you use DIGEST though :-)
While I dont thing the concept eminent domain should apply to granting right of way to private industry, if I was the wal-mart/city lawyer I'd argue:
- Perhaps some kind of state zoning pretty much made it feasable to build the mall in exactly one location.
- The mall will create jobs and tourism (hah..)
Basically, you can probably recycle many of the arguments used to get state/city subsidies for stadiums and such.
But again, I don't agree with it, but I'm sure there is more to these stories then first meets the eye.
I've wondered that as well. You probably could, but you'be be doing crap underneath the SOAP level and thus would prevent some clients from using it.
If they had a way to keep a sesssion (like a session id via cookies)... oh man would life be easy.
You can even create single purpose ssh keys which execute one command each
Close, but it is still not part of the specification itself.
Does javascript handle this? Can I use it in an AJAX call? Does it work out of the box from the CPAN libraries? Can you do it in PHP with a normal set of compile-time flags? Can you have anonymous clients authenticate themselves using a login/password (i.e. a flickr like web service?)
Want the ultimate proof that SOAP and XML-RPC is a failed specification? Every single javascript libraray (Prototype, jQuery, mooTools) doesn't support either one, yet every virtually every single XHttpRequest made is pretty much an RPC call.
Except WSDL is as evil to look at as any XSD file I've seen and last I checked (which admitidly was many moons ago), Perl will not generate one based on your code and maintain it for you automagically.
Its not a problem making it work with different scripting languages
Correction. Most of them have good support for *consuming* other peoples SOAP service. Try to *create* a SOAP service in, say, Perl. I've never tried in other languages, but that is only because after trying in Perl, I swore off SOAP forever.
SOAP probably kicks some kind of alien psuedo-ass in a strongly typed language like Java or C# where the compiler can easily sues out your object model. Even then, since there is no standard for authentication or session management, you are left with only a hollow shell of a solution.
First, figure out multiple ways to build your large scale project.
This is required as part of your environmental impact statement. It is done to figure out your next point:
For example, if you're building a road, maybe you can run it through points A, B, or C.
You make the assumption that A,B and C are all equal cost and just as good. I'll punch a whole in this easy:
- Mountain Passes. Sometimes there is only one route for 500 miles.
- Transit Lines. To get any kind of ROI, you gotta have stops in pretty precise locations, often giving you only a few blocks of wiggle room. Lets not forget if you plan to tunnel, there might be geological issues that make one route much, much cheaper.
- Utilities get in the way. What if your route crosses over some other dudes utility and they refuse to budge? What if that utility was 500 miles long?
in exchange for the right of Eminent Domain over your property for the next 3 years
What uninteresting thing are you building that will only be around for three years. Last I checked the interstate highway system is older then three years. Last I checked, the Hover Dam is older then three years.
In that case, I would argue that the people have spoken.
You know what? Fuck the people. Sometimes you have to cram stuff down their throats. That is what makes a good politician... people objected to most of the bridges that were built in New York. You think people have issue with them now? People object to every sewer treatment plant and every landfill created on our planet. Yet without them, civilization could not exist.
Being libertarian means you haven't grown up politically. You can still subscribe to free market ideals and understand that sometimes you need a little central planning. Look how much more productive and environmentally "clean" a well organized, properly zoned city like New York is compared to an un-zoned, unplanned suburb like Anytown USA. Sure those suburbs were cheap to build, but look how costly they are once gas prices have gone up? Look how costly they become when they get even the least bit dense (and I'm talking "cost" in the economic sense, so I'm including Time and Psychological costs as well).
Life gets really strange when you grow up and realize you can be a democrat and remain compatible with your ideas of a free market economy.
Why can't one of these new fanged RPC thingies HANDLE FUCKING AUTHENTICATION and AUTHORIZATION in a standard way? And if there *is* a standard way, I dont know about, WHY IS IT NOT OBVIOUS AND EASY!?
Why do I *have* to roll my own fucking way to hold a session using SOAP? Why couldn't they have some concept of a cookie or some easy way to maintain state that was buried deep within the protocol so it was transparent to me?
All these XML based RPC specs suck ass. Some architechure astronauts snorted a bunch of crack and decided to do crazy stuff like let you use SMTP for a transport. It would be much simpler to just pick a damn transport and stick with it so you can exploit the charctaristics of said transport (like cookies for HTTP or ??? for SMTP).
Unless you trust your IDE and dont need to get say, a C# client to talk to a mod_perl backend, you are in a world of hurt. For that matter, if you use any kind of dynamic language (PHP, Ruby, Perl), god help you with anything. You'll be writing a fuckton of XML that you'll have to keep in sync with your codebase. Forget any kind of automated tools.
In short, screw SOAP. Keep it fucking simple!!!
Given the fact that the NIMBY factor for power lines,power plants, nuke power, roads, dams, whatever is so high, the odds of at least one person objecting is virtually 100%.
Therefore, if you would like to have nuke power, power lines, roads, high speed rail, whatever, you will *need* to force somebody to fucking move for the greater good. Otherwise, you will never get the right-of-way to make your project happen. We have granted our government the ability to force people to fucking move out of the way.
We call this Eminent Domain.
Why anyone wants Federal control of anything is beyond me
Given that large scale projects are impossible without forcing somebody to move, do you feel comfortable granting eminent domain to private industry?
If you say "make it all states rights" given that many of these large scale projects affect multiple states, you'll wind up with heavy federal oversight anyway. Let states do it all, and they'll sue eachother when the other guy builds a huge damn. They'll sue when their state law conflicts with the other state law. You either get federal agencies for interstate projects, or you get a metric assload of federal judicial "weight".
Is this some kind of evolved version of Ron Paul spam? Help me out, I'm confused here.
The main advantage I can see is that it allows coding in languages other than ActionScript -- theoretically, anything that runs on the CLR/DLR.
That is exactly the advantage. If you are already familiar with any of .NET or it's languages and you already own Visual Studio, you are good to go. Bonus points are assigned to the fact that since it is a little brother of WPF, it doesn't take much work to fold in your existing codebase into your silverlight codebase.
Flex requires learning a whole new product suite, a whole new language, a whole new IDE... everything you know.
Because 99% of all normal, non nerd internet users would rather watch the video within a website rather then choose between a dozen file formats that sometimes stream properly, sometimes confuse the browser "what kind of file is this, good sir?", sometimes dont stream but try to download all 1.2 gigs before loading the client.
Video on the inter-tubes exists because of the embedded video client. Take away the embedding, and we'll be back to square one.
But as long as it satisfies some Linux nerd (who choose their OS knowing full well they would not be able to enjoy various forms of media on the internet), the ends justify the means, right?
Not to mention most programmers are comfortable with Visual Studio/.NET/WPF and can take their existing skillsets and quickly crank out a pretty decent silverlight applet. If you go flex, you have to invest in a whole different product suite, a whole different language, whole different everything. Flex is great if your shop is mostly designers who are familiar with the Adobe suite. Silverlight is great for shops who are mostly developers.
PS: Mono/Moonlight will be the Linux silverlight client.
Or perhaps it just costs 10x more per mile to deploy any kind of utility/road/anything in a city. You've got to deal with permiting, zoning, mitigation, anti-development hippies, NIMN's (not in my neighborhood), parked cars to tow, drunk bums passed out on your construction site, utility relocation, you name it and you have to pay for it.
It is easy to do anything in the suburbs. No zoning, less red-tape, less density, less 100 year old wood stave pipes, whatever.
Poverty has nothing to do with it. If anything, there are probably more affluent 20-something condo owners living in a city who would jump all over FiOS then there are suburban house dwellers.
would they be able to demand access to the source code
I dunno, who's got the most nukes?
If weapons systems were GPLd
Then the military would have to way to keep the F-16's weapons control software inhouse. Why would you need to rely on a crashed jet when you could just carry a floppy disk out the front door with the same thing?
Or in GPL-only land do we grant our government the right to create closed-source software for national security purposes? What about civilian aircraft? If Boeing had to GPL all its software, wouldn't airbus be able to lift that work and kick our ass in the aerospace industry? Or is the idea the entire globe has to sign a "GPL only" treaty and make it a condition for joining things like the WTO?
And to reply to myself, once you start self-reflecting, you might start thinking "well, I like GPL but the government needs to provide a way for it to function". I promise whatever direction your line of thinking takes you, you'll get something that looks a bit like the skeleton form of copyright law.
While I dont have any proof and I dont know enough of it's history, I imagine copyright law existed in some form throughout history. I also suspect it was never under the name "copyright law" because most of it grew organically... court case by court case. Hell, even if we tossed out copyright as we know it, the we'd probably grow right back into it through decades of case law.
Imagine a world without copyrights.
It would be as close to BSD as you could possibly get only there would be no way to enforce the clause about including the original authors. The liability clauses would probably be enforceable though. I suspect many would probably include the original author names at the top of their code just so other programmers who look at it know the lineage and history of the codebase. I also suspect many would give back patches and code changes because there is a strong benefit to keeping your codebase in sync with the mainline.
GPL would not exist at all because it uses copyright to "crack open" any codebase who incorporates the license. The "crack open" nature of GPL is what makes it a good fit for certian types of projects, but it also provides a strong dis-incentive toward "lifting" the code without giving back. Since people love to "lift" code (or music/video/whatever) because it is easy and free, the GPL needs teeth or it's main goals wouldn't pan out. For example without copyright, Apple could lift GPL code like Linux just as easy as they "lifted" the BSD codebase and the Linux crew would be powerless to stop them.
GPL is effective because you can sue people who do not follow it. The only reason you can sue people is GPL hinges on copyright law. Take away copyright law and GPL has no teeth.
Bottom line is GPL advocates *must* embrace, endorse and respect copyright law. GPL and the "community driven" software ecosystem that grew around it could not exist without copyright. Those who think their beloved GPL could exist without copyright need to do some serious self-reflection.