You want to be only as loyal as makes sense to you. For every story of how workers being loyal has been good I can relate stories of how businesses made decisions that amounted to "sorry, things are bad, had to cut you, and can't afford a dime of severance, wish we could, but bummer" at 4:48pm on a Tuesday.
Depending upon how close you are to finishing a project, you could ask for a touch more time from your new job and then tell your current employer "I can stay 6 weeks if you'd like to finish all this up, if you are willing to pay me at my new pay rate" Don't give them an extra money and give up 800 pounds!
But in the end, companies that wish to staff so thin that they get single points of failure are no different that hosting companies that don't do backups to save money and then expect expect everyone to work crazy hours when their primary storage fails after several years -- they all want everything but don't want to pay for it.
There's an acid test here: was this framework developed for use by the company, or did the company just decide to grab a copy of the code from Sourceforge/etc and go down this path after you left?
If the former, where you said "hey, I wasn't asked to develop this code, but in my own time I thought it would solve an issue at the company" then you most likely don't have a case. You wrote code for the company. They can do what they want with it.
If the latter, then you can argue the case of whether or not they have the rights to everything you developed. If you were working in an "exempt" position then in the same way that they can't count your hours you can't automatically have "your own time."
But the first case is going to be virtually impossible to defend unless you have one heck of a paper trail, and even then, they can argue you represented the issue in bad faith knowing it was something critical to the company yet you portrayed it as "not a big deal."
Actually, airlines pay substantial fuel taxes which not only pay for ATC but get siphoned off into the general fund. Then they pay landing fees at many airports as well.
No major airport in the US is run at a loss. Some of the smaller airports may be, where the city feels that the benefit of the airport outweighs the cost, but all major airports pay their own way.
The AF payment is being given with no waivers or settlements agreed to in order to help the families with their direct, immediate costs such as funerals and travel. The true settlements will take a while but will (based on precedent) be in excess of a million dollars per victim.
But isn't just as valid to say...charge her $1 for each time her "source" was shared?
And it's trivial with p2p networks for you to share it 20 times to people that shared it 20 times to people that shared it 20 times? That (or however you divvy up the expansion) is only a net total of 8,000 times per song...and that's not wildly unreasonable.
As much as the question of 'at cost' matters, you also have to look at the way it was funded -- via bonds.
If we decide to offer "Slashlight Highspeed Internet for Towns" we have to raise money for the business venture. We will be faced with a much higher bar if we want to try and raise venture capital -- questions about profit, business plans, value, etc.
If we go to a bank to borrow the money...yeah right. Banks might lend you some money against equipment (that is, the things they could take back) but not against operational expenses like payroll and power bills.
And in the end, with bonds, the city IS on the hook for the money. If the venture doesn't work, then those bonds have to be paid somehow.
Now, if the cities want to use bond money to loan money to private companies, then that's cool. I can think of a lot of businesses I would like to get into but can't because I just can't raise the money...and I know I'm not alone.
But in reality, when public entities get into business, because they simply have a different risk model (in other words...well, if the business fails, then they'll just have to raise taxes to pay the bonds) they can fundamentally take on more risk and lower profit margins than private businesses, who have to answer for their use of money.
This is exactly on point. Getting Cogent style (which means, not directly peered to everyone) bandwidth costs $10/mbit, assuming you are buying at least 100mbit. That's raw bandwdth, assuming zero infrastructure; in other words, you as the customer own the cost of connecting to Cogent's network.
Connecting to a carrier like Level(3), even when buying in 1000mbit quantities, is nowhere near $10/mbit; it's more like $20/mbit. In gigabit quantities. Where you the customer have to provide the connection to them; that price just gets you a router port on their gateway router.
So clearly, delivering things like 20mbit downstream to consumers means overselling, and managing overselling.
With video coming along, etc., it's simply time to start establishing policies that reflect that overselling ratios are going to drop considerably.
It's called, welcome to the real world. Providing bandwidth costs money.
If everyone was honest and paid for it (with the 'criminal' being the exception) we wouldn't have DRM in the first place.
You would have to come up with some heavy statistics before most people (myself included) would believe that the majority of people running modified DLL's actually owned legitimate copies of CS4.
As someone who has "purchased" literally millions of dollars of consulting working from both large firms and single individuals over the past decade, I heartily recommend this post.
The major thing I look for is an estimate, and then constant, regular tracking against that estimate. If something starts to unwind or get out of scope, I want to know early and often because that allows me to make my decisions.
My bosses -- in fact, anyone's bosses -- don't like surprises. So we work together so there aren't any.
Unfortunately, your premise is incorrect in a reality where an ISP has more than one router. The average home user is not connecting to a device that's directly peered to other providers. There's usually a 3 or 4 level tree there.
So your statement is true within a provider's local "branch" (for example, a small town) but it's not true between towns, or states, and most certainly not true between providers.
The entire reason that home bandwidth is priced so cheap is that providers have counted on being able to oversubscribe and therefore build out their tier 1 (global interconnect) to a less than 1:1 ratio of their tier 3/4 local networks.
That doesn't mean they should continue to market things as "unlimited" but it's always tough to market difficult metrics to the average consumer.
In other news today, German prosecutors have announced that they only pursue companines that infringe on more than 3000 pieces of open source software, as they have decided they don't with to be the F/OSS's enforcers of the GPL.
As applications arise that use more and more bandwidth, this kind of thing is inevitable.
Real bandwidth costs have never sync'd up to the costs charged to residential users, on the assumption that residential utilization would always be low.
As residential utilization goes up, you have to either raise the cost to better reflect the real costs of bandwidth, or you have to limit usage, especially amongst the outliers.
How much is 'real' bandwidth? Well, consider that companies buying a gig of bandwidth from a mainline provider are paying roughly 30 dollars a meg. Get down into the weeds of of someone who wants 5 meg, and it's easily 100 dollars a meg -- just for the bandwidth itself. Plus the costs of the the connection infrastructure.
Maybe you can get a great deal on business grade, no limit 5 meg circuits for $289 (and that would be a HECK of a deal).
That's a big difference from $42.95 that Comcast charges out by me. And right now I get way more than 5 meg on my downloads for all the various content I grab, so in reality, my experience is MUCH better than it would be with a 5 meg circuit.
BTW, for those who claim they can't get bandwidth; T1's work just about everywhere. And you can get a T1 with truly unlimited service for about $249.
The thing is, the eee pc doesn't do one major thing this device does -- work on data on your smartphone (that is, you don't have yet another device with data that has to be transfered). And with the phones it works on, you don't have to set up yet another email client (which, depending upon how your company runs exchange, might not even be available).
The idea, remember, is just a remote console for your smartphone, its apps, and its data. So a lot different than the eee pc.
Mind you, I'm not saying I see the value in it for other reasons -- mainly, I feel it's too expensive for what it provides -- but I don't completely believe it's the eee pc that's the redfly killer...rather, that the value proposition for the redfly doesn't make sense.
Microsoft only made IE free (they had been charging) once it was clear that Netscape (which had been releasing it's product for free as a 'beta' and for 'trial' which was unlimited for practical purposes) was not really intending to may the browser a piece of 'pay' software. Sure, they went through some motions, and you could buy a paid version in stores, even, but if you were online you could download and use the browser for free. Netscape's numbers showed that well over 90 percent of all browser users never bought the software. What choice did that leave Microsoft?
So you'd prefer the model where you come up with a cool idea, develop it, go to sell it, and then a big company can just hire 200 people to replicate it and sell it for near peanuts and you end up getting nothing for your work?
Sorry. Patents are good things. Without patents, he who has money rules the world forever.
I agree with your points. If VMware has made changes to their 2.4 kernel, those changes and source need to be released.
The definition of derivative should not be "it requires the Linux kernel to load" but rather "the Linux kernel requires it to load." THEN it's a mod to the kernel. but if the kernel can boot otherwise, sorry, game over, not derivative.
More importantly, GPLv3 serves as a reminder that choosing Open Source with the idea that "you'll never get left behind" is not a true position -- that all any given Open Source project really represents is the right to have the source code AS IT EXISTS NOW. In the future more restrictive licensing could be released that could impact you.
In short, GPLv3 really made "Open Source" more like "Closed Source" by clearly pointing out that what you may be allowed to do now you may not be allowed to do later (unless you fork and thus lose the community aspect that made it interesting in the first place).
After all, who's to say GPLv4 won't say "you must release any changes back to the community whether you distribute or not" ?
GPLv3 is the best possible thing that could have happened to Closed Source vendors because it just kicked the chair out from under most of the arguments in favor of Open Source.
Actually GSM phones in Japan will only work if they support UMTS2100 -- which is a Japan (and Korea?)-only UMTS frequency range -- the rest of the world is using UMTS at 1900Mhz.
While some phones support both frequencies for UMTS, most don't. The Nokia N73 and the Treo 750 are two phones that I know support both ranges, but a Samsung Blackjack, for example, does not.
There's been no sign that these patents are so sweeping that Verizon has patented VOIP-to-POTS. If they had they'd be going after Skype which has much deeper pockets.
I think Vonage is simply tossing out a smokescreen here to try and get a ruling that forces a license fee they can live with.
If Google makes a mistake and makes your personal data available to everyone on the internet, they aren't liable at all. Think of the consequences if people's personal financial information was inadvertently exposed. You as a company would have massive liability. Google would have none.
And indeed, even if they just lose all your data or decide to turn off your service, you don't have recourse.
No sane company hosts critical data with services that requires a full release of liability to use it.
Too many companies abuse "exempt" status, in my opinion. In fact, it's really the standard. Make everyone exempt and then never hire enough people. Work 60 hours last week to get the job done? Sorry, you're "exempt." But the same companies will track a Friday you don't come in if things are slow -- when you shouldn't be tracked on hours at all! In short, it's used as a way of making you at LEAST a 40 hour employee, with all the penalties for working less and none of the benefits from working more. Explain to me how this is good for the worker?
I believe we'd be better off as a whole if "exempt" status was banished entirely.
You may well be right. In my mind, this (YouTube) has always been a bad thing from a copyright perspective.
There's an excellent chance that this situation will be judged NOT to be safe harbor.
Short explanation:
The current safe harbor (where the infringed party (Viacom) has to advise the infringer (YouTube)) is based on the telco/ISP protection model where what the infringer is selling is the bandwidth/hosting.
And that made perfect sense that AOL (which sells hosting) shouldn't have to police content. Broadly speaking, AOL makes its money from the hosting and providing internet access -- not from the content hosted.
But YouTube makes money from the advertising displayed and keyed on the content. They don't charge the people posting content to host it at all -- not for the number of downloads, etc. So they are, quite realisitically, involved in the content.
Google certainly has the money to start producing its own TV shows and content that people will want to watch. Let them do that and prove their distribution model based on their own risk, rather than taking someone else's content, potentially impacting THEIR money, and saying "here, be happy with this new business model"
Slashdot Mirror
You want to be only as loyal as makes sense to you. For every story of how workers being loyal has been good I can relate stories of how businesses made decisions that amounted to "sorry, things are bad, had to cut you, and can't afford a dime of severance, wish we could, but bummer" at 4:48pm on a Tuesday.
Depending upon how close you are to finishing a project, you could ask for a touch more time from your new job and then tell your current employer "I can stay 6 weeks if you'd like to finish all this up, if you are willing to pay me at my new pay rate" Don't give them an extra money and give up 800 pounds!
But in the end, companies that wish to staff so thin that they get single points of failure are no different that hosting companies that don't do backups to save money and then expect expect everyone to work crazy hours when their primary storage fails after several years -- they all want everything but don't want to pay for it.
There's an acid test here: was this framework developed for use by the company, or did the company just decide to grab a copy of the code from Sourceforge/etc and go down this path after you left?
If the former, where you said "hey, I wasn't asked to develop this code, but in my own time I thought it would solve an issue at the company" then you most likely don't have a case. You wrote code for the company. They can do what they want with it.
If the latter, then you can argue the case of whether or not they have the rights to everything you developed. If you were working in an "exempt" position then in the same way that they can't count your hours you can't automatically have "your own time."
But the first case is going to be virtually impossible to defend unless you have one heck of a paper trail, and even then, they can argue you represented the issue in bad faith knowing it was something critical to the company yet you portrayed it as "not a big deal."
Actually, airlines pay substantial fuel taxes which not only pay for ATC but get siphoned off into the general fund. Then they pay landing fees at many airports as well.
No major airport in the US is run at a loss. Some of the smaller airports may be, where the city feels that the benefit of the airport outweighs the cost, but all major airports pay their own way.
The AF payment is being given with no waivers or settlements agreed to in order to help the families with their direct, immediate costs such as funerals and travel. The true settlements will take a while but will (based on precedent) be in excess of a million dollars per victim.
But isn't just as valid to say...charge her $1 for each time her "source" was shared?
And it's trivial with p2p networks for you to share it 20 times to people that shared it 20 times to people that shared it 20 times? That (or however you divvy up the expansion) is only a net total of 8,000 times per song...and that's not wildly unreasonable.
Then you'd be at $576,000.00
As much as the question of 'at cost' matters, you also have to look at the way it was funded -- via bonds.
If we decide to offer "Slashlight Highspeed Internet for Towns" we have to raise money for the business venture. We will be faced with a much higher bar if we want to try and raise venture capital -- questions about profit, business plans, value, etc.
If we go to a bank to borrow the money...yeah right. Banks might lend you some money against equipment (that is, the things they could take back) but not against operational expenses like payroll and power bills.
And in the end, with bonds, the city IS on the hook for the money. If the venture doesn't work, then those bonds have to be paid somehow.
Now, if the cities want to use bond money to loan money to private companies, then that's cool. I can think of a lot of businesses I would like to get into but can't because I just can't raise the money...and I know I'm not alone.
But in reality, when public entities get into business, because they simply have a different risk model (in other words...well, if the business fails, then they'll just have to raise taxes to pay the bonds) they can fundamentally take on more risk and lower profit margins than private businesses, who have to answer for their use of money.
Strenuously argue that the RIAA is evil and that copying music is not an issue...through a myriad of excuses.
Why is this any different? Music, news...one can make all the same arguments.
And honestly...talking about "how to tinker with your feeds to make it tougher..." is just another form of DRM, isn't it?
Something to think about.
This is exactly on point. Getting Cogent style (which means, not directly peered to everyone) bandwidth costs $10/mbit, assuming you are buying at least 100mbit. That's raw bandwdth, assuming zero infrastructure; in other words, you as the customer own the cost of connecting to Cogent's network.
Connecting to a carrier like Level(3), even when buying in 1000mbit quantities, is nowhere near $10/mbit; it's more like $20/mbit. In gigabit quantities. Where you the customer have to provide the connection to them; that price just gets you a router port on their gateway router.
So clearly, delivering things like 20mbit downstream to consumers means overselling, and managing overselling.
With video coming along, etc., it's simply time to start establishing policies that reflect that overselling ratios are going to drop considerably.
It's called, welcome to the real world. Providing bandwidth costs money.
If everyone was honest and paid for it (with the 'criminal' being the exception) we wouldn't have DRM in the first place.
You would have to come up with some heavy statistics before most people (myself included) would believe that the majority of people running modified DLL's actually owned legitimate copies of CS4.
As someone who has "purchased" literally millions of dollars of consulting working from both large firms and single individuals over the past decade, I heartily recommend this post.
The major thing I look for is an estimate, and then constant, regular tracking against that estimate. If something starts to unwind or get out of scope, I want to know early and often because that allows me to make my decisions.
My bosses -- in fact, anyone's bosses -- don't like surprises. So we work together so there aren't any.
Unfortunately, your premise is incorrect in a reality where an ISP has more than one router. The average home user is not connecting to a device that's directly peered to other providers. There's usually a 3 or 4 level tree there.
So your statement is true within a provider's local "branch" (for example, a small town) but it's not true between towns, or states, and most certainly not true between providers.
The entire reason that home bandwidth is priced so cheap is that providers have counted on being able to oversubscribe and therefore build out their tier 1 (global interconnect) to a less than 1:1 ratio of their tier 3/4 local networks.
That doesn't mean they should continue to market things as "unlimited" but it's always tough to market difficult metrics to the average consumer.
In other news today, German prosecutors have announced that they only pursue companines that infringe on more than 3000 pieces of open source software, as they have decided they don't with to be the F/OSS's enforcers of the GPL.
As applications arise that use more and more bandwidth, this kind of thing is inevitable.
Real bandwidth costs have never sync'd up to the costs charged to residential users, on the assumption that residential utilization would always be low.
As residential utilization goes up, you have to either raise the cost to better reflect the real costs of bandwidth, or you have to limit usage, especially amongst the outliers.
How much is 'real' bandwidth? Well, consider that companies buying a gig of bandwidth from a mainline provider are paying roughly 30 dollars a meg. Get down into the weeds of of someone who wants 5 meg, and it's easily 100 dollars a meg -- just for the bandwidth itself. Plus the costs of the the connection infrastructure.
Maybe you can get a great deal on business grade, no limit 5 meg circuits for $289 (and that would be a HECK of a deal).
That's a big difference from $42.95 that Comcast charges out by me. And right now I get way more than 5 meg on my downloads for all the various content I grab, so in reality, my experience is MUCH better than it would be with a 5 meg circuit.
BTW, for those who claim they can't get bandwidth; T1's work just about everywhere. And you can get a T1 with truly unlimited service for about $249.
The thing is, the eee pc doesn't do one major thing this device does -- work on data on your smartphone (that is, you don't have yet another device with data that has to be transfered). And with the phones it works on, you don't have to set up yet another email client (which, depending upon how your company runs exchange, might not even be available).
The idea, remember, is just a remote console for your smartphone, its apps, and its data. So a lot different than the eee pc.
Mind you, I'm not saying I see the value in it for other reasons -- mainly, I feel it's too expensive for what it provides -- but I don't completely believe it's the eee pc that's the redfly killer...rather, that the value proposition for the redfly doesn't make sense.
Microsoft only made IE free (they had been charging) once it was clear that Netscape (which had been releasing it's product for free as a 'beta' and for 'trial' which was unlimited for practical purposes) was not really intending to may the browser a piece of 'pay' software. Sure, they went through some motions, and you could buy a paid version in stores, even, but if you were online you could download and use the browser for free. Netscape's numbers showed that well over 90 percent of all browser users never bought the software. What choice did that leave Microsoft?
So you'd prefer the model where you come up with a cool idea, develop it, go to sell it, and then a big company can just hire 200 people to replicate it and sell it for near peanuts and you end up getting nothing for your work?
Sorry. Patents are good things. Without patents, he who has money rules the world forever.
What the judge is saying is that just because you keep a file in RAM and not on on disk, you can't claim you aren't "storing" that data.
I agree with your points. If VMware has made changes to their 2.4 kernel, those changes and source need to be released.
The definition of derivative should not be "it requires the Linux kernel to load" but rather "the Linux kernel requires it to load." THEN it's a mod to the kernel. but if the kernel can boot otherwise, sorry, game over, not derivative.
Steve
More importantly, GPLv3 serves as a reminder that choosing Open Source with the idea that "you'll never get left behind" is not a true position -- that all any given Open Source project really represents is the right to have the source code AS IT EXISTS NOW. In the future more restrictive licensing could be released that could impact you.
In short, GPLv3 really made "Open Source" more like "Closed Source" by clearly pointing out that what you may be allowed to do now you may not be allowed to do later (unless you fork and thus lose the community aspect that made it interesting in the first place).
After all, who's to say GPLv4 won't say "you must release any changes back to the community whether you distribute or not" ?
GPLv3 is the best possible thing that could have happened to Closed Source vendors because it just kicked the chair out from under most of the arguments in favor of Open Source.
Actually GSM phones in Japan will only work if they support UMTS2100 -- which is a Japan (and Korea?)-only UMTS frequency range -- the rest of the world is using UMTS at 1900Mhz.
While some phones support both frequencies for UMTS, most don't. The Nokia N73 and the Treo 750 are two phones that I know support both ranges, but a Samsung Blackjack, for example, does not.
There's been no sign that these patents are so sweeping that Verizon has patented VOIP-to-POTS. If they had they'd be going after Skype which has much deeper pockets.
I think Vonage is simply tossing out a smokescreen here to try and get a ruling that forces a license fee they can live with.
Not to mention again that this solution, while cool, still doesn't address mobile devices. So it's again being steps behind.
Oh, and it's not collaborative for resources like meeting rooms, etc.
Nothing wrong with the google solutions, but honestly, it's still a long way from Exchange.
It's not stupid.
If Google makes a mistake and makes your personal data available to everyone on the internet, they aren't liable at all. Think of the consequences if people's personal financial information was inadvertently exposed. You as a company would have massive liability. Google would have none.
And indeed, even if they just lose all your data or decide to turn off your service, you don't have recourse.
No sane company hosts critical data with services that requires a full release of liability to use it.
Too many companies abuse "exempt" status, in my opinion. In fact, it's really the standard. Make everyone exempt and then never hire enough people. Work 60 hours last week to get the job done? Sorry, you're "exempt." But the same companies will track a Friday you don't come in if things are slow -- when you shouldn't be tracked on hours at all! In short, it's used as a way of making you at LEAST a 40 hour employee, with all the penalties for working less and none of the benefits from working more. Explain to me how this is good for the worker?
I believe we'd be better off as a whole if "exempt" status was banished entirely.
You may well be right. In my mind, this (YouTube) has always been a bad thing from a copyright perspective.
There's an excellent chance that this situation will be judged NOT to be safe harbor.
Short explanation:
The current safe harbor (where the infringed party (Viacom) has to advise the infringer (YouTube)) is based on the telco/ISP protection model where what the infringer is selling is the bandwidth/hosting.
And that made perfect sense that AOL (which sells hosting) shouldn't have to police content. Broadly speaking, AOL makes its money from the hosting and providing internet access -- not from the content hosted.
But YouTube makes money from the advertising displayed and keyed on the content. They don't charge the people posting content to host it at all -- not for the number of downloads, etc. So they are, quite realisitically, involved in the content.
Google certainly has the money to start producing its own TV shows and content that people will want to watch. Let them do that and prove their distribution model based on their own risk, rather than taking someone else's content, potentially impacting THEIR money, and saying "here, be happy with this new business model"