As I mentioned in another comment - if you jump right to OO programming, you're essentially teaching the kid how to hurdle in the Olympics without first teaching him "This is a hurdle. This is how you approach a hurdle. This is how you land after a hurdle. This is how you jump a hurdle. This is how you run between hurdles."
OO is useful, and important, but it's hardly baby steps. YOu're talking about taking a bunch of kids, whose likely maximum experience to "computer programming" is playing their XBox or PS/2, and trying to teach them how Sprites work.
Too far, too fast. They don't even correctly understand what a "program" is.
That's so far past what most of these kids should be learning...
When you're first learning how to drive a car, you don't say, "We should go to NYC, because driving in heavy traffic is a good way to learn avoidance and awareness techniques", when you don't even know how to make smooth turns and brake without throwing your passengers through the windshield, and still get the windshield wipers and turn signals confused.
These kids aren't learning how to program games, or webpages, or databases....they're learning what programs are. They're learning to think in the paradigm of "This problem requires these steps to solve, and those steps require these tasks be done". That's not programming in the sense that any of us who "program" think - it's so basic that the concepts involved are second nature for us. This class is targeted at a group of kids with the intent of developing that second nature - of teaching the kids how to jump hurdles without killing themselves, not how to make them into Olympic hurdlers.
You're teaching grade school kids how to program. Not how to make websites, or how to beat Flash into submission, or how to deal with cookies. You need to teach them about programming, and how these words make a computer do what you intend. They have decades ahead of them to argue the intricacies of VI vs. Emacs, or Perl vs Python vs PHP, or C+ vs C vs Pascal. When they get into later Computer classes, they'll be exposed to Basic, or Java, or.NET, or C. But the best thing to do at this age is teach them how to program - the steps required to go from "I want my computer to do this" to "X requires steps A, B, and C" to "A means tell the computer (foo(bar));".
Most of us on here are so far past the point of where we first picked up the keyboard, and wrote, "10 PRINT 'Hello world'; 20 GOTO 10;" that we miss the important fact that people do not think like a computer program We infer, we use outside sources of information, we have a lifetime of learning to filter our decisions on. Training the mind to see a problem as discrete steps which can be represented by even more discrete commands to a computer is not natural.
My advisor in college was an AI fanatic who did a whole pile of research into computer education and learning. Scheme is not super-powerful, all-encompasing, and world-ending. It will not be the next-generation solution to world peace. But it simply and elegantly introduces people using it to most, if not all, of the cores of good programming and computer programming concepts - stacks and queues, functions, recursion, all the core "how to program" type of stuff.
Scheme (and LISP as a superset), as a language for getting things done (the Mancala program I wrote in my AI class notwithstanding), is not a first choice for much of anything most people want to do. However, as a programming fundamentals teaching language, it's a well documented basic teaching language.
...I'd be willing to bet that most people run their computes with Admin accounts.
It's too much fo a hassle to deal with the "You can't do that, log out, log in as admin, do that, log out, log back in as yourself" for most people. Hell, I KNOW what the hazards are, but I sitll do it.
Saying "It's only insecure when you run as administrator" is like saying "It's only dangerous when you smoke the cigarettes". Of course it's only dangerous that way, but that's not stopping thousands of people from doing it.
Call bullshit if you like, but Cingular sells 7 of them, in increasing frequencies, on their CingularExtras site.
We run that site at work, so of course, we've tested them all. Most of us back in Operations can hear the first 4, with the exception of 3 guys - 2 in their late 30's, and one in his mid-20's. Those of us that can hear it (8 of us) range from early 20's to late 30's. About half of us can hear the 5th one. A couple of us can at least tell that the 6th one is doing something (including me at 32), and only the mid-20 year old guys can hear the 7th one.
Everyone buys "cars" that are "advertised" (via the speedometer) to obtain "speeds" of upward of 100 "mph".
Everyone knows that your actual speeds may vary, in that during "rush hour", your "throughput" may only be 20 or 30 "mph", and in normal traffic, you'll go 60-80 "mph", and if you get out into the middle of Nowhere, Montana, you might actually be able to go 100-110 "mph" or more.
No one is suing Toyota or Ford or Kia because they don't get their "advertised speeds", because with regards to cars, everyone knows that your actual speeds may vary based on where you're driving and what the rest of the traffic is like. The internet, really, is no different. Under optimal conditions, with the right settings and lack of anything in your way, you can obtain the advertised speeds. In practicality, these speeds are likely only obtained between your computer and the first gateway you hit, because after that point, there's too many other variables in play to be able to make any guarantee on service times.
So someone is looking for hosting. Go to Google. Find lots of hosting sites. Some featured advertisers, some just search results. In among all of the results, find sites that let the public review hosting providers.
You look over a few providers, pick some to look into in more detail, google for reviews, see they're all mostly good, and settle on one. Then you find _another_ site, where all the reviews for a hosting provider you picked are completely the opposite of everything else you've researched on them. You start to wonder which review sites are valid, and which ones are being astroturfed.
I don't really, outside of knowing people who host with X company and asking them, know of a way to find out if a site is good or not. I just recently had to find a new hosting provider (my free provider was closing up his space), and searched around for about 3 months before settling on a company, only to find an entire site complaining about all the horrible service and technical people they have.
When I moved, I started searching around some more for local hosting providers (I mived to Raleigh, Research Triangle Park is right here with piles of tech companies). I found a couple I liked, then googled their addresses to see where they were. The ones whose location didn't look like a place of business, I wrote to and asked what their hosting was like and where it was physically located. Asked a few technical questions (I actually had real questions, not just test questions, but if you don't have real questions, just make up some that you know the answer to and play dumb). I settled on one, and he's been pretty decent for me over the last few months, working through some wierd bugs that have cropped up on occassion.
So to answer your question - be very careful. When you google for reviews of the companies, don't stop at page 3 - go to page 10 or 15. Call and talk to someone there and tell them you're looking at someplace to move your hosting to. Send email with technical questions and see what answers you get back. If this is for business-class hosting, go to their hosting facility, they should be able to arrange a tour, if not of the raised floor itself, then the facilities and monitoring center and such. Basically, don't let other people's opinions be your sole guide, feel them out yourself and see what kind of impression they make.
The first line of the article begins with a false premise, and incorrectly reflects the amended law as of any time since 1978, claiming that what was written in 1968 was the correct interpretation of a document written 150 years, before communication could possibly involve anything more than direct than person-to-person or written correspondence.
The Omnibus... act was amended in 1978 to specifically remove the language which places the President's authority over all other concerns. Check out the current version of 18 USC 2511, and specifically the MISC2 section at the end, which outlines the changes to the statute through all amendments. The 1978 amendmeent, in fact, was the same one that overrode the portions of Omnibus... to reflect the details of the FISA legislation passed in the same year, which granted specific powers, to be exercised via specific procedures, with regards to electronic surveillance.
In particular, the "constitutional power" verbage was removed as overreaching, and 2 (e) and (f) were added to reflect the ability of the Federal government to conduct electronic and other surveillance of foreign communications on foreign communications networks granted by FISA. In no way, shape, or form does the collection of data regarding my phone usage fall under those terms, no matter how many degrees of separation from Al-Qaeda I am via Kevin Bacon.
If you want to be a strict Federalist interpretist of the Constitution, you better send your women back to the kitchen and keep your negroes in line...can't have them out, you know, voting and owning property.
"...Because XM makes available vast catalogues of music in every genre, XM subscribers will have little need ever again to buy legitimate copies of plaintiffs' sound recordings,"
Unless someone invented some sort of new way of compressing and storing broadcasts, by my quick math, figuring an average of 4 minutes per song, a user would need 214 Inno's to record the "vast catalog" and never have to buy music again. And this doesn't include any new music that comes out from this day forward.
Only in RIAA-world do the suits think the average consumer has $77,000+ (for 214 Inno's at $360 each) to plunk down right now, plus 63+ weeks to spend 24/7, recording entire catalogs of music.
It's a limited storage device with even more restrictions on moving content than cassette/CD have now, and they're already proven legal in piles of court cases. You almost have to wonder if RIAA has any income stream, given how hard they're trying to make money through the legal system.
You're developing. You're not doing graphics work, or anything that necessarily requires a local workhorse. The most intense part of the work that you might be required to run locally would be the IDE. In all the dev work I've done, none of it needed to be done locally on the system I was actually typing on, and in most cases, I avoided such situations because my remote system was more ingrained in the environment, and was bigger anyway.
Spend the money on a monster remote server with a massive disk. Go big...quad-core, 4G RAM, a gig or two of disk space. Set up the remote desktop of your choice (XCDMP, VNC, whatever). Set up a CVS system on the box. Set up your IDE on the box. Run it all over ssh tunnels or a secure VPN system or whatever. Your compile times will be better, the system specs will be bigger and better, and all you need is some sort fo large-screen notebook running whatever you want, as long as it supports your remote display method.
Do you know that they don't have a warrant? Are you somehow privy to classified investigations?
Because if they did have warrants, there would be a documented paper trail. In some court somewhere, the NSA would be generating 200 million sealed warrants and subpoenas. This kind of volume would be noticed.
Even if the administration were going through the FISA court, those warrants are only good for gathering enough evidence, on short notice and immediate need, to follow up and come along behind with real warrants and subpoenas.
The reports of what is going on is unwarranted, unsupported data collection on the phone habits of over 200 million US citizens. Not even in the Kevin Bacon game can you possibly come up with 200 million people who are seven degrees related to each other through identified terror suspects.
The output channel is the last place where people who want to break DRM can still manage to do so. No matter what DRM is in place on the computer to prevent copying/duplication/modification, once it's gone through all that, it goes so some sort of output device (speakers for sound, monitor for video). At that point, the DRM protecting the content is nullified, and whatever you wanted to do with the signal at that point you can.
DRM on the output channel prevents all of the "analog" (I know, most new video outputs are digital, but the same methods apply) hacks from extracting the content from the DRM.
That's still a third party - Google went to Firefox and pitched the deal, and Firefox, who had to pick some sort of default (there does have to be _a_ default), picked Google.
It's not the same thing as Megacompany putting their search engine in their browser that ships with their OS, along with their media player and their applications etc etc etc.
I notice that in my version of Firefox the search box defaults to Google, and that the pulldown menu of pre-entered options doesn't even include MSN Search, but Google seems to have been oddly quiet on that front for the many years prior to IE7 that Firefox has made this feature available.
MSN search (Windows Live) is "newer" than stable releases of Firefox, at least as far as publically available WinLive goes.
The problem here is that neither Firefox (or Safari, or Opera, or whatever) are sponsored by Google. Those browsers are the result of one third party choosing another third party for it's default.
Microsoft defaulting to WinLive is just like Windows coming with only IE, and WMP, and not giving the users choices for other third-party sources.
At every mid-to-large company I've worked for (the smallest being a 500 person company that runs some little websites like CingularExtras.com), the problem isn't the root access - they hired you, you work there, and you have the trust to do what you need to do in order to perform your assigned tasks. In the end, whether you do that as root, or as bsmith who runs "sudo su -" or even "su -" doesn't really matter.
What does matter, is accounting for which user ran that "rm -rf/mnt/financialReports" and isn't owning up to it. Or which user made system tuning changes that don't seem to be working right. Or any number of things which may or may not be malicious, but need questions answered.
None of the companies I worked for put sudo on servers to limit what their sys admins could do - they put it on the systems so that they could, if something were to happen, figure out which of their admins they needed to go ask questions of. Files are missing, what were you doing. Server isn't behaving right any more, what was the last thing you worked on.
You can't do any of this (easily) if 5 admins are all logging in as the same user. You can do it fairly trivially if you're logging in as your own id and switching to root.
X10 is limited to on/off/up/down. For example, you can't fade-up the lights on a home theater room if they were turned off. They first must "pop" to full brightness then fade down. There are complicated ways around this, but they really isn't worth it.
Untrue. There are now 2-way X10 devices. I use the lamp modules in my living room to do exactly what you complain doesn't work - going from 0% on to 15% on, primarally brought about by the birth of my daughters, who didn't take well the having bright lights popped in their eyes at 2 AM when we got up to feed them.
That's not the point. I'm sure they will, but that's still not the point.
For the market that NPR targets, I would bet those users are more likely to pay for them, in order to have them legally, than to get them elsewhere.
These are the same people who donate money to local public radio. These aren't scraping-by college kids, for the most part, they're people who are willing to pay for quality, and to keep the source of the entertainment in business.
(Why is it I never have mod points when I need them:) )
This is the obvious answer. In the old distribution model, local stations held membership drives to raise funds, which were used to "purchase" distribution rights to the national shows, and pay the staff for the local shows. In the new distribution model, the national "station" would collect subscription fees, which are used to pay for distribution rights of the national shows, and a portion would/could (maybe opt-in your local PubRadio station) be diverted to the local station to pay the staff for local shows.
Most PBS stations are already set up to do 12-part (monthly) draft payments for donations, as well as one-time collections, so set the membership fee to, I dunno, $5 per month or $50 for the year. Change the Podcast availability to such that you need to have an account to be able to download. Free/non-donating accounts can only download 2 'casts a week, donating accounts get unlimited access.
I think that most people who listen to NPR feel that they get more than $5 worth of information out of it a month, especially if they listen to more than one show...hell, I'd probably pay $5 a month just for "Prairie Home Companion", let alone "Car Talk", "Marketplace", "All Things Considered", and "Talk of the Nation", just to name a few off the top of my head that I listen to regularly on the radio.
The problem in this case isn't collaboration, it's the sharing of large files needed in the collaboration.
In this case, a simple repository is all that's needed to take care of this problem. Have a large file? Open the repository site, drop in the file, it returns a link to you when it's uploaded. There are commercial document repositories, and there are Open Source repositories, either of which solves the problem of putting documents in a central place for groups to utilize.
The real problem for Collaboration Software Vendors is that, for a group using a CS system, there is usually one and only one way to do things; one way to share documents, one way to organize projects, one way to build plans. All of the members of the group have to adapt to the CS. Groups who use email and calenders for their collaboration don't necessarily get all the features of a CS system, but the "system" they use is maximally flexible. User A can organize and manage things how they are comfortable with them, and user B can do something completely different with the same information.
The problem isn't inertia of users getting out of email and into CS, but a lack of flexibility on the part of the CS systems that users aren't willing to give up.
...but Slashdot won't let me dump all 5 of them to score this story off of the front page.
Ooh, look, a 12 hour video of someone playing a video game and starting to babble incoherently come hour 10.5. The fascination level of spectator video games for 12 hours ranks slightly above my fascination level of watching paint dry. And not even good mind-altering chemical fume paint, just low-odor latex paint.
ActiveX is a mechanism that allows compiled code delivered via the web to run on the client. This feature is an absolute must-have for many corporate environments.
Must-have? For what?
What do corporate environments do that require code to be downloaded willy-nilly and run on the client, that an actual installer, that you intentionally go and run, can't be used for instead?
Especially in a corporate environment, there should be no random installation of code on computers. Anything a user needs to run should be made available in prechecked, verified installer packages that can be installed in whatever maintenance cycle said corporation runs with - automated overnight installs, periodic patches, etc.
The only place I can see ActiveX being "necessary" would be for home users. And with as insufficient as the framework is, as buggy as the implementation is, as lax as the security is, the entire thing should just be nuked from orbit. Take it out of IE7, publish an aftermarket patch through the KB that will allow users to re-enable ActiveX functionality, and make it so that only people who specifically want ActiveX have it.
There's no reason for your browser to have the ability to physically install code that outlives the life of your user session. If there is, said code should be a thin application, not a browser-based "program".
IANAE (Engineer), so I don't know, just pointing out that the topology they're talking about has been around for decades. Maybe it won't really be 10B5 to the home, but something similar, since technically, 10B5 is supposed to run over 50ohm coaxial cable, and most home coax is 75ohm, IIRC.
Slashdot Mirror
As I mentioned in another comment - if you jump right to OO programming, you're essentially teaching the kid how to hurdle in the Olympics without first teaching him "This is a hurdle. This is how you approach a hurdle. This is how you land after a hurdle. This is how you jump a hurdle. This is how you run between hurdles."
OO is useful, and important, but it's hardly baby steps. YOu're talking about taking a bunch of kids, whose likely maximum experience to "computer programming" is playing their XBox or PS/2, and trying to teach them how Sprites work.
Too far, too fast. They don't even correctly understand what a "program" is.
That's so far past what most of these kids should be learning...
When you're first learning how to drive a car, you don't say, "We should go to NYC, because driving in heavy traffic is a good way to learn avoidance and awareness techniques", when you don't even know how to make smooth turns and brake without throwing your passengers through the windshield, and still get the windshield wipers and turn signals confused.
These kids aren't learning how to program games, or webpages, or databases....they're learning what programs are. They're learning to think in the paradigm of "This problem requires these steps to solve, and those steps require these tasks be done". That's not programming in the sense that any of us who "program" think - it's so basic that the concepts involved are second nature for us. This class is targeted at a group of kids with the intent of developing that second nature - of teaching the kids how to jump hurdles without killing themselves, not how to make them into Olympic hurdlers.
Seriously....mod this parent up.
You're teaching grade school kids how to program. Not how to make websites, or how to beat Flash into submission, or how to deal with cookies. You need to teach them about programming, and how these words make a computer do what you intend. They have decades ahead of them to argue the intricacies of VI vs. Emacs, or Perl vs Python vs PHP, or C+ vs C vs Pascal. When they get into later Computer classes, they'll be exposed to Basic, or Java, or .NET, or C. But the best thing to do at this age is teach them how to program - the steps required to go from "I want my computer to do this" to "X requires steps A, B, and C" to "A means tell the computer (foo(bar));".
Most of us on here are so far past the point of where we first picked up the keyboard, and wrote, "10 PRINT 'Hello world'; 20 GOTO 10;" that we miss the important fact that people do not think like a computer program We infer, we use outside sources of information, we have a lifetime of learning to filter our decisions on. Training the mind to see a problem as discrete steps which can be represented by even more discrete commands to a computer is not natural.
My advisor in college was an AI fanatic who did a whole pile of research into computer education and learning. Scheme is not super-powerful, all-encompasing, and world-ending. It will not be the next-generation solution to world peace. But it simply and elegantly introduces people using it to most, if not all, of the cores of good programming and computer programming concepts - stacks and queues, functions, recursion, all the core "how to program" type of stuff.
Scheme (and LISP as a superset), as a language for getting things done (the Mancala program I wrote in my AI class notwithstanding), is not a first choice for much of anything most people want to do. However, as a programming fundamentals teaching language, it's a well documented basic teaching language.
The default accounts in XP didn't have Admin rights either, but it didn't stop me from granting them to the accounts I created.
...I'd be willing to bet that most people run their computes with Admin accounts.
It's too much fo a hassle to deal with the "You can't do that, log out, log in as admin, do that, log out, log back in as yourself" for most people. Hell, I KNOW what the hazards are, but I sitll do it.
Saying "It's only insecure when you run as administrator" is like saying "It's only dangerous when you smoke the cigarettes". Of course it's only dangerous that way, but that's not stopping thousands of people from doing it.
Call bullshit if you like, but Cingular sells 7 of them, in increasing frequencies, on their CingularExtras site.
We run that site at work, so of course, we've tested them all. Most of us back in Operations can hear the first 4, with the exception of 3 guys - 2 in their late 30's, and one in his mid-20's. Those of us that can hear it (8 of us) range from early 20's to late 30's. About half of us can hear the 5th one. A couple of us can at least tell that the 6th one is doing something (including me at 32), and only the mid-20 year old guys can hear the 7th one.
Everyone buys "cars" that are "advertised" (via the speedometer) to obtain "speeds" of upward of 100 "mph".
Everyone knows that your actual speeds may vary, in that during "rush hour", your "throughput" may only be 20 or 30 "mph", and in normal traffic, you'll go 60-80 "mph", and if you get out into the middle of Nowhere, Montana, you might actually be able to go 100-110 "mph" or more.
No one is suing Toyota or Ford or Kia because they don't get their "advertised speeds", because with regards to cars, everyone knows that your actual speeds may vary based on where you're driving and what the rest of the traffic is like. The internet, really, is no different. Under optimal conditions, with the right settings and lack of anything in your way, you can obtain the advertised speeds. In practicality, these speeds are likely only obtained between your computer and the first gateway you hit, because after that point, there's too many other variables in play to be able to make any guarantee on service times.
So someone is looking for hosting. Go to Google. Find lots of hosting sites. Some featured advertisers, some just search results. In among all of the results, find sites that let the public review hosting providers.
You look over a few providers, pick some to look into in more detail, google for reviews, see they're all mostly good, and settle on one. Then you find _another_ site, where all the reviews for a hosting provider you picked are completely the opposite of everything else you've researched on them. You start to wonder which review sites are valid, and which ones are being astroturfed.
I don't really, outside of knowing people who host with X company and asking them, know of a way to find out if a site is good or not. I just recently had to find a new hosting provider (my free provider was closing up his space), and searched around for about 3 months before settling on a company, only to find an entire site complaining about all the horrible service and technical people they have.
When I moved, I started searching around some more for local hosting providers (I mived to Raleigh, Research Triangle Park is right here with piles of tech companies). I found a couple I liked, then googled their addresses to see where they were. The ones whose location didn't look like a place of business, I wrote to and asked what their hosting was like and where it was physically located. Asked a few technical questions (I actually had real questions, not just test questions, but if you don't have real questions, just make up some that you know the answer to and play dumb). I settled on one, and he's been pretty decent for me over the last few months, working through some wierd bugs that have cropped up on occassion.
So to answer your question - be very careful. When you google for reviews of the companies, don't stop at page 3 - go to page 10 or 15. Call and talk to someone there and tell them you're looking at someplace to move your hosting to. Send email with technical questions and see what answers you get back. If this is for business-class hosting, go to their hosting facility, they should be able to arrange a tour, if not of the raised floor itself, then the facilities and monitoring center and such. Basically, don't let other people's opinions be your sole guide, feel them out yourself and see what kind of impression they make.
The first line of the article begins with a false premise, and incorrectly reflects the amended law as of any time since 1978, claiming that what was written in 1968 was the correct interpretation of a document written 150 years, before communication could possibly involve anything more than direct than person-to-person or written correspondence.
The Omnibus... act was amended in 1978 to specifically remove the language which places the President's authority over all other concerns. Check out the current version of 18 USC 2511, and specifically the MISC2 section at the end, which outlines the changes to the statute through all amendments. The 1978 amendmeent, in fact, was the same one that overrode the portions of Omnibus... to reflect the details of the FISA legislation passed in the same year, which granted specific powers, to be exercised via specific procedures, with regards to electronic surveillance.
In particular, the "constitutional power" verbage was removed as overreaching, and 2 (e) and (f) were added to reflect the ability of the Federal government to conduct electronic and other surveillance of foreign communications on foreign communications networks granted by FISA. In no way, shape, or form does the collection of data regarding my phone usage fall under those terms, no matter how many degrees of separation from Al-Qaeda I am via Kevin Bacon.
If you want to be a strict Federalist interpretist of the Constitution, you better send your women back to the kitchen and keep your negroes in line...can't have them out, you know, voting and owning property.
Unless someone invented some sort of new way of compressing and storing broadcasts, by my quick math, figuring an average of 4 minutes per song, a user would need 214 Inno's to record the "vast catalog" and never have to buy music again. And this doesn't include any new music that comes out from this day forward.
Only in RIAA-world do the suits think the average consumer has $77,000+ (for 214 Inno's at $360 each) to plunk down right now, plus 63+ weeks to spend 24/7, recording entire catalogs of music.
It's a limited storage device with even more restrictions on moving content than cassette/CD have now, and they're already proven legal in piles of court cases. You almost have to wonder if RIAA has any income stream, given how hard they're trying to make money through the legal system.
You're developing. You're not doing graphics work, or anything that necessarily requires a local workhorse. The most intense part of the work that you might be required to run locally would be the IDE. In all the dev work I've done, none of it needed to be done locally on the system I was actually typing on, and in most cases, I avoided such situations because my remote system was more ingrained in the environment, and was bigger anyway.
Spend the money on a monster remote server with a massive disk. Go big...quad-core, 4G RAM, a gig or two of disk space. Set up the remote desktop of your choice (XCDMP, VNC, whatever). Set up a CVS system on the box. Set up your IDE on the box. Run it all over ssh tunnels or a secure VPN system or whatever. Your compile times will be better, the system specs will be bigger and better, and all you need is some sort fo large-screen notebook running whatever you want, as long as it supports your remote display method.
Because if they did have warrants, there would be a documented paper trail. In some court somewhere, the NSA would be generating 200 million sealed warrants and subpoenas. This kind of volume would be noticed.
Even if the administration were going through the FISA court, those warrants are only good for gathering enough evidence, on short notice and immediate need, to follow up and come along behind with real warrants and subpoenas.
The reports of what is going on is unwarranted, unsupported data collection on the phone habits of over 200 million US citizens. Not even in the Kevin Bacon game can you possibly come up with 200 million people who are seven degrees related to each other through identified terror suspects.
Not pointless at all.
The output channel is the last place where people who want to break DRM can still manage to do so. No matter what DRM is in place on the computer to prevent copying/duplication/modification, once it's gone through all that, it goes so some sort of output device (speakers for sound, monitor for video). At that point, the DRM protecting the content is nullified, and whatever you wanted to do with the signal at that point you can.
DRM on the output channel prevents all of the "analog" (I know, most new video outputs are digital, but the same methods apply) hacks from extracting the content from the DRM.
That's still a third party - Google went to Firefox and pitched the deal, and Firefox, who had to pick some sort of default (there does have to be _a_ default), picked Google.
It's not the same thing as Megacompany putting their search engine in their browser that ships with their OS, along with their media player and their applications etc etc etc.
MSN search (Windows Live) is "newer" than stable releases of Firefox, at least as far as publically available WinLive goes.
The problem here is that neither Firefox (or Safari, or Opera, or whatever) are sponsored by Google. Those browsers are the result of one third party choosing another third party for it's default.
Microsoft defaulting to WinLive is just like Windows coming with only IE, and WMP, and not giving the users choices for other third-party sources.
What does matter, is accounting for which user ran that "rm -rf /mnt/financialReports" and isn't owning up to it. Or which user made system tuning changes that don't seem to be working right. Or any number of things which may or may not be malicious, but need questions answered.
None of the companies I worked for put sudo on servers to limit what their sys admins could do - they put it on the systems so that they could, if something were to happen, figure out which of their admins they needed to go ask questions of. Files are missing, what were you doing. Server isn't behaving right any more, what was the last thing you worked on.
You can't do any of this (easily) if 5 admins are all logging in as the same user. You can do it fairly trivially if you're logging in as your own id and switching to root.
That's not the point. I'm sure they will, but that's still not the point.
For the market that NPR targets, I would bet those users are more likely to pay for them, in order to have them legally, than to get them elsewhere.
These are the same people who donate money to local public radio. These aren't scraping-by college kids, for the most part, they're people who are willing to pay for quality, and to keep the source of the entertainment in business.
(Why is it I never have mod points when I need them :) )
This is the obvious answer. In the old distribution model, local stations held membership drives to raise funds, which were used to "purchase" distribution rights to the national shows, and pay the staff for the local shows. In the new distribution model, the national "station" would collect subscription fees, which are used to pay for distribution rights of the national shows, and a portion would/could (maybe opt-in your local PubRadio station) be diverted to the local station to pay the staff for local shows.
Most PBS stations are already set up to do 12-part (monthly) draft payments for donations, as well as one-time collections, so set the membership fee to, I dunno, $5 per month or $50 for the year. Change the Podcast availability to such that you need to have an account to be able to download. Free/non-donating accounts can only download 2 'casts a week, donating accounts get unlimited access.
I think that most people who listen to NPR feel that they get more than $5 worth of information out of it a month, especially if they listen to more than one show...hell, I'd probably pay $5 a month just for "Prairie Home Companion", let alone "Car Talk", "Marketplace", "All Things Considered", and "Talk of the Nation", just to name a few off the top of my head that I listen to regularly on the radio.
The problem in this case isn't collaboration, it's the sharing of large files needed in the collaboration.
In this case, a simple repository is all that's needed to take care of this problem. Have a large file? Open the repository site, drop in the file, it returns a link to you when it's uploaded. There are commercial document repositories, and there are Open Source repositories, either of which solves the problem of putting documents in a central place for groups to utilize.
The real problem for Collaboration Software Vendors is that, for a group using a CS system, there is usually one and only one way to do things; one way to share documents, one way to organize projects, one way to build plans. All of the members of the group have to adapt to the CS. Groups who use email and calenders for their collaboration don't necessarily get all the features of a CS system, but the "system" they use is maximally flexible. User A can organize and manage things how they are comfortable with them, and user B can do something completely different with the same information.
The problem isn't inertia of users getting out of email and into CS, but a lack of flexibility on the part of the CS systems that users aren't willing to give up.
...but Slashdot won't let me dump all 5 of them to score this story off of the front page.
Ooh, look, a 12 hour video of someone playing a video game and starting to babble incoherently come hour 10.5. The fascination level of spectator video games for 12 hours ranks slightly above my fascination level of watching paint dry. And not even good mind-altering chemical fume paint, just low-odor latex paint.
Must-have? For what?
What do corporate environments do that require code to be downloaded willy-nilly and run on the client, that an actual installer, that you intentionally go and run, can't be used for instead?
Especially in a corporate environment, there should be no random installation of code on computers. Anything a user needs to run should be made available in prechecked, verified installer packages that can be installed in whatever maintenance cycle said corporation runs with - automated overnight installs, periodic patches, etc.
The only place I can see ActiveX being "necessary" would be for home users. And with as insufficient as the framework is, as buggy as the implementation is, as lax as the security is, the entire thing should just be nuked from orbit. Take it out of IE7, publish an aftermarket patch through the KB that will allow users to re-enable ActiveX functionality, and make it so that only people who specifically want ActiveX have it.
There's no reason for your browser to have the ability to physically install code that outlives the life of your user session. If there is, said code should be a thin application, not a browser-based "program".
IANAE (Engineer), so I don't know, just pointing out that the topology they're talking about has been around for decades. Maybe it won't really be 10B5 to the home, but something similar, since technically, 10B5 is supposed to run over 50ohm coaxial cable, and most home coax is 75ohm, IIRC.