So you move to the cloud and fire your IT department, but then who do you trust? If you have an in-house IT department you have a local experts on your payroll. Whether they are trustworthy or not is another matter. If you outsource all your IT functions you have no local experts. At least with in-house IT you have fiscal leverage over the talent.
On top of that, outsourcing generally costs more. For nearly every project I have looked at, it costs more to move it to the cloud than to do it in house. It makes sense if you look at the numbers. You pay an outside firm 100 to 150 US dollars per hour for a technician when you can get an employee for half that price. If problems arise that that exceed your contracted support hours costs quickly skyrocket. The only way this makes sense is if you are small enough to not need full-time tech support people, or for short-term work getting big projects off the ground.
Once you do outsource or move to the cloud, how do you guarantee good support? The one thing we outsourced that is saving us some money is Email. Overall it's saving us a couple thousand a year from our previous solution, but the support is outright shit. Because it is outsourced, I have no leverage besides moving by business elsewhere, but migrating all that data will either take me a lot of time or cost a lot of money for consultants. I was much happier running our old email system in-house.
Why steal plates? A a color laser printer and some weatherproof, polyester labels should be able to print a reasonable facsimile of a license plate . It may be obvious close up, but I bet it would pass muster from 20 feet.
And the Whole Language Theory roars back to life. I hope that in your daughter's elementary school they are not rigidly fixated on kids reading by 'sight-words". In a nutshell, researchers discovered decades ago that proficient readers recognize thousands of words by shape. So the idea came about to skip all the messy phonetics and just teach kids how to read by intuitively recognizing word shapes. This was an extremely popular theory in the 70's and 80's. The Dick and Jane readers are an example of whole language theory used to teach reading. The problem is that it doesn't work for a lot of kids. The theory was generally discredited in the 90's as a teaching methodology. I believe the whole language theory is to blame for functionally illiterate adults who can not read words they were not taught in elementary school. My wife learned to read from dick and jane and is totally incapable of sounding a word out that she has not seen before.
There is a lot of truth in Whole Language theory. But if you never teach kids to sound out words, they will never know how to read a word they didn't learn from an adult. It's far better to teach kids to phonetics and let them progress naturally.
If something is advertised as "Unlimited" and then a limit is applied, it is, arguably, false advertising. That opens the ISP to civil suits. Class action, anyone?
Voice is a priority over FTP because it matters not one whit if your download takes 30 seconds longer. It does matter if a packet of voice data takes 100 ms longer. Setting QoS options on your router makes little difference if the ISP's upstream connection is saturated, so QoS only works right when it is done in the middle. QoS is a good thing that I want to see more of. Net Neutrality is NOT about QoS. Net neutrality is about ISPs' prejudicial use of traffic shaping to selectively promote or degrade services. Net neutrality is about not allowing ISP's to leverage their local monopolies to extort more money from customers and content providers.
I don't recall Microsoft's MDI, so I can't comment on it. Look, you aren't going to convince me, and I'm not going to convince you of which interface is better. And I don't want to. Truthfully, both have major issues and we could both go on all day about bad UI design in any platform. The interface you see as elegant and simple I see as over simplified and over reliant on cutesy icons. I need readable text and a program menu, dammit!
I agree with you that so many things in modern interfaces is based on earlier bad decisions. Which is why I can't use KDE. Every time I look at it I get depressed. I much preferred Gnome 2. I also like Mac System's interface a better than early versions of windows. But as much as it is a mistake to continually recommit the same bad decisions, it's worse to change things for the sake of change, breaking all the learned behaviors that users have been trained to expect for the past 20+ years. If you are going to do that, it had better be a significant improvement, and not some inconsequential improvement. I don't see OS X as a big enough improvement. It is an improvement over it's predecessors in many respects, but it also has some regressions, which are inevitable when you try to reinvent the wheel.
Mac applications act this way due to legacy decisions made for the original circa 1984 Mac, not because it's the right way to do things. At the time, it took the Mac a long time to start applications. Apple decided on this behavior to make the computer more responsive when opening new documents. Now days, document open much more quickly and this behavior is no longer required.
Fine if you have a brand new high spec Mac. But I've got a 2008 MacBook, upgraded to Lion. I've just timed reopening xcode (with a single project and the organizer open). It took 9 seconds. Pages took 7 seconds.
And of course if you don't have it in the dock, you need to add the time taken to find the app in to apps folder or Launchpad.
And what't the benefit to closing when the last window closes? You potentially save a a space in the dock. and... No, that's it. Vitual memory means that memory consumption isn't really an issue.
Mac OS does it be right way.
Note that new OS designs on phones and tablets tend to not close apps either. They only get closed when lack of resources demand it. And then the OS does it's best to hide the fact that it was closed by restarting in the exact same state.
Windows (and the way Linux GUIs copy it) are a suboptimal way. It always was, and it still is. It's just that because of market share most people are used to that way and consider it normal.
9 and 7 seconds are not a long time to start a program and load a file. I remember apps on old Mac System taking upwards of 30 seconds to start and load a file.
And since you brought up phones, I have a Nexus One. When I got it it would go about 1.5 days on a charge. Then I installed an application to kill background apps. Now the phone goes easily 2.5 days and is generally more responsive.
If a program has something useful to do in the background than it should be implemented as a light-weight daemon, rather than a full blown app like iTunes.
A demon, by definition, is a background task with no UI. Yet a music player needs a UI. That isn't the same thing as having a window. OSX makes that distinction at the fundamental level. Windows doesn't, and Linux UIs typically follow Windows even when it's broken.
OK, you are right about Daemons not having UI's. But windows does make a distinction. There are services that run in the background. There are apps that run in the system tray and provide a light-weight UI. You can always just minimize your music player when you are listening to music. Also, windows apps can be programmed to not shut down when the last window is closed and instead take residence in the system tray.
My question is, why does a music player have to run all the time even when y aren't playing music? Is it to watch for system events or to download updates in the background? Those tasks are better suited for a daemon or a scheduled event.
Fixed, no, but did you every use an original mac? They were woefully underpowered for the graphical UI. Amazing machines for their day, but they were slow.
when I click on a running app in the doc that has no open windows, the program doesn't do anything. It should, at that point, actually respond; open a new project, give me a file-open dialog box, anything but sit there looking pretty.
Many applications do just what you're asking. Click on the Finder and if there aren't any open windows it will pop one open with your default directory. Click on Mail and if you don't have a mail browser window open then it will open one.
It's an application-defined behavior and most of Apple's applications do something when they don't have an open window and their dock icon is clicked. Some 3rd party applications don't follow this UI principle and, yes, it's a shame when they don't.
The other issue with this behavior is that it is not easy to tell at a glance to tell what programs are running.
It is? All you have to do is glance over at the dock, everything that's running has a dot next to its icon. If that's not clear enough for you then you can always just clear out the non-running apps in your dock and only let the running apps show. Now EVERY app in the dock is a running app.
If we're talking about applications with open windows (which is what would show up in the Windows' Taskbar) then just use Apple's Mission Control. Every running application will have its own group of windows, grouped with the icon for the application. It's quick and easy to see what's running and pick a window.
Also remember that under Mac OS a running program that doesn't have any open windows doesn't need to use that many resources. It will stay in memory and possibly do some processing but if its resources are needed they can be reclaimed by the system and the application will be put in a suspended state, its memory will be paged out to disk, its running threads will be suspended or given a lower run priority, etc. Thus you can keep a lot of applications running in the background without windows and not worry that they'll slow you down much.
Yes, the dots are a challenge for me. I have to scan the bar and count dots. I find that annoying. I much prefer an OS that clearly groups icons. I tend to hyper-focus on what I'm doing and get annoyed by distractions like having to examine icons for dots when I realize I have too many apps open and have to shut them down. There is also that fact that as I get older and more nearsighted I am finding it harder and harder to pick out these details quickly on a computer screen.
The fact that the X button sometimes closes the application, and sometimes leaves the application running without a UI is also bad.
Why is it bad? It's a developer choice do do whichever is more appropriate for the app. On Windows an app MUST close when its last window closes, unless the developer puts it into the system tray.
The reasoning for leaving the app open when a multiple document app has it's last window closed is straight forward. It's a common usage pattern to finish working on one document and then start working on another. If apps quit when the last window closes, then this happens:
The user closes the first document, and the UI to open the next document (File/Open) disappears. They then have to restart the app, which involves waiting, before they ca open their next document.
But for apps which are not document based, that argument doesn't apply. Closing the window on a single window app really does mean you've finished working with that app for the time being.
Then there are other reasons for choosing one behaviour or another. If an app does useful work even when there are no Windows, then of course it makes sense to keep it open. iTunes is an obvious example.
There's a reason why Mac developers have this choice and Windows developers don't get it (apart from the system tray utility option). Because with Windows, the disappearance of the last window means that access to the menu has also disappeared. That's not the case with Mac.
Mac applications act this way due to legacy decisions made for the original circa 1984 Mac, not because it's the right way to do things. At the time, it took the Mac a long time to start applications. Apple decided on this behavior to make the computer more responsive when opening new documents. Now days, document open much more quickly and this behavior is no longer required. Personally, the behavior drives me nuts because when I click on a running app in the doc that has no open windows, the program doesn't do anything. It should, at that point, actually respond; open a new project, give me a file-open dialog box, anything but sit there looking pretty. Programs that do something useful in the background with no open files are few and far between. If a program has something useful to do in the background than it should be implemented as a light-weight daemon, rather than a full blown app like iTunes.
The other issue with this behavior is that it is not easy to tell at a glance to tell what programs are running. The strength of the Windows Task Bar is that it clearly separates running programs from application launch icons. Certainly, this is a matter of what people are accustomed to, but for myself and I think many people who are accustomed to windows, this is infuriating.
I'm willing to pay $2.99 for a good book. I might be willing to pay 99 cents for a piece of trash. What I'm not willing to do is to pay the same price for an e-book as I would for the currently in print version, especially when the in print version is a $25 hard cover. Understandably, there are greater costs when it comes to technical works. But, people do not and will not care about the author's labor. They care about the enjoyment they get out of it. People generally enjoy print books more and therefor value them more and are willing to pay more for them. Just like everything else, a book is worth what someone is willing to pay for it.
The essential problem with selling downloadable content, which all consumers grasp intuitively but that publishers seem foggy on, is that the supply is functionally infinite. Econ 101 teaches the supply vs demand pressures on pricing. In the case of printed books, the publishers have to predict demand, then they print the amount of books they expect to be able to sell, and they mostly sell then at a fixed-ish price. Around 20 to 30 bucks for a hardback, discounted to about 18 if they don't sell. Paperbacks sell for around 6 to 10. That artificially limits supply and helps them keep prices stable. After all, if you see a book on the shelf that interests you, you know that it may not be available next month or next year. But with ebooks, there is no limit to supply. Buyers know they can get a copy, unless the publisher stops selling, and they know that it costs the the publishers virtually nothing to keep a copy on their servers for anyone to download 2, 5, 10 years down the line, and they know that it costs the publisher virtually nothing to sell them a copy. Publishers have been very good at predicting demand for printed books. What publishers have been very bad at is estimating how much people are willing to pay for ebooks. Instead, the large publishing houses seem to have been operating in a way to protect their print book sales from erosion by ebooks.
This is too bad, because what major publishers bring to the table is the expertise to help an author refine their work and market it effectively. They also have reputations to protect, so what they print will not totally suck, and consumers know that. But, if these publishers don't adapt to the times, they will go out of business. What publishers need to do is to realize that X number of people are willing to buy at 5.99, y number of people will buy at 2.99, and z number of people will buy at 0.99. They should be pricing ebooks dynamically. Introduce them high to get the established fans and let the price drop accordingly as demand lessens. If demand spikes as word spreads, then let the price rise as well.
US corporations have largely lost the will to innovate. Just look at the decline of in house R&D divisions. They used to be commonplace in US firms, but now they are hard to find firms. These days, SOP for large corps is to let the start-ups do the innovating and try to buy up the ones that are succeeding. That makes sense in many ways. Why risk the business on developing a new research that may never succeed or even come to market? If you do internal research, why look into radical new technologies when it is easier and more reliable to incrementally improve your existing products?
However, American businesses have missed out on a lot of opportunities or been late to market because of this mentality. The US auto industry is the poster child for missed opportunities. In all likelihood, Detroit will only be interested if the prototype has been developed and proven to work, Toyota buys up the patents, successfully scales up production, brings them to market, and they sell like hotcakes.
I use an external dock for SATA drives or a USB to IDE adapter and plug it into a spare windows PC in the office. A quick format and then "cipher/W" does the trick.
Oh, I almost forgot the most important step.... I teach an intern how to do it so that I'm not wasting my own time.
I agree. I have done this trick several times to recover data at work. Usually we have a second identical drive on premises, and sometimes I have to use Ebay.
It's nickel-and-dime'ing. It's fine to pay for what you use, if you are using it for a small amount of data. But how much do you want to store? Lets say 100GB. It seems very reasonable to me for someone to keep that many pictures. It would cost you $168 per year, but your cost will soon balloon as you collect more pictures. There are on-line backup services that charge a lot less than that per year with no caps. And I did not lie about the transfer fees. They charge you to download your data. If your drive crashes you have to pay them to access your own data. S3 may be a terrific service for storing web site data or whatever, but it's not a service for the average user looking for on-line backup.
My problem with Amazon S3 is the pricing structure. They charge for storage per month, plus transfer fees, plus system requests, and they want you to move large quantities of data using portable storage. Their pricing page says it all.
Why not roll your own SAN? Build a small server out of cast-off parts. Stuff it with whatever size hard drives you want. Install your choice of OS and run the software to make it an iSCSI target. The beautiful part to it is that it is upgradable and repairable.
Depending on how much data there is, this is the cheapest solution for the long haul. Startech SATA docs are under 50 bucks. SATA drives start under 30 bucks. There is no need to buy any software, scripts can be scheduled to copy your data to external drives. This is not as safe as an online backup, but after a couple years of service, an on-line backup solution will cost more.
I don't disagree with you. But the same logic applies to the cable operators. They sell their own advertising. Depending on your local population size, they could make cable TV free and ad supported. Broadcast networks have done this for decades. Sure, they should charge the customer for installation and equipment, but the monthly service should be free.
Slashdot Mirror
Surely you mean, "they WILL be using neutrinos."
So you move to the cloud and fire your IT department, but then who do you trust? If you have an in-house IT department you have a local experts on your payroll. Whether they are trustworthy or not is another matter. If you outsource all your IT functions you have no local experts. At least with in-house IT you have fiscal leverage over the talent.
On top of that, outsourcing generally costs more. For nearly every project I have looked at, it costs more to move it to the cloud than to do it in house. It makes sense if you look at the numbers. You pay an outside firm 100 to 150 US dollars per hour for a technician when you can get an employee for half that price. If problems arise that that exceed your contracted support hours costs quickly skyrocket. The only way this makes sense is if you are small enough to not need full-time tech support people, or for short-term work getting big projects off the ground.
Once you do outsource or move to the cloud, how do you guarantee good support? The one thing we outsourced that is saving us some money is Email. Overall it's saving us a couple thousand a year from our previous solution, but the support is outright shit. Because it is outsourced, I have no leverage besides moving by business elsewhere, but migrating all that data will either take me a lot of time or cost a lot of money for consultants. I was much happier running our old email system in-house.
Why steal plates? A a color laser printer and some weatherproof, polyester labels should be able to print a reasonable facsimile of a license plate . It may be obvious close up, but I bet it would pass muster from 20 feet.
And the Whole Language Theory roars back to life. I hope that in your daughter's elementary school they are not rigidly fixated on kids reading by 'sight-words". In a nutshell, researchers discovered decades ago that proficient readers recognize thousands of words by shape. So the idea came about to skip all the messy phonetics and just teach kids how to read by intuitively recognizing word shapes. This was an extremely popular theory in the 70's and 80's. The Dick and Jane readers are an example of whole language theory used to teach reading. The problem is that it doesn't work for a lot of kids. The theory was generally discredited in the 90's as a teaching methodology. I believe the whole language theory is to blame for functionally illiterate adults who can not read words they were not taught in elementary school. My wife learned to read from dick and jane and is totally incapable of sounding a word out that she has not seen before. There is a lot of truth in Whole Language theory. But if you never teach kids to sound out words, they will never know how to read a word they didn't learn from an adult. It's far better to teach kids to phonetics and let them progress naturally.
If something is advertised as "Unlimited" and then a limit is applied, it is, arguably, false advertising. That opens the ISP to civil suits. Class action, anyone?
Voice is a priority over FTP because it matters not one whit if your download takes 30 seconds longer. It does matter if a packet of voice data takes 100 ms longer. Setting QoS options on your router makes little difference if the ISP's upstream connection is saturated, so QoS only works right when it is done in the middle. QoS is a good thing that I want to see more of. Net Neutrality is NOT about QoS. Net neutrality is about ISPs' prejudicial use of traffic shaping to selectively promote or degrade services. Net neutrality is about not allowing ISP's to leverage their local monopolies to extort more money from customers and content providers.
I don't recall Microsoft's MDI, so I can't comment on it. Look, you aren't going to convince me, and I'm not going to convince you of which interface is better. And I don't want to. Truthfully, both have major issues and we could both go on all day about bad UI design in any platform. The interface you see as elegant and simple I see as over simplified and over reliant on cutesy icons. I need readable text and a program menu, dammit!
I agree with you that so many things in modern interfaces is based on earlier bad decisions. Which is why I can't use KDE. Every time I look at it I get depressed. I much preferred Gnome 2. I also like Mac System's interface a better than early versions of windows. But as much as it is a mistake to continually recommit the same bad decisions, it's worse to change things for the sake of change, breaking all the learned behaviors that users have been trained to expect for the past 20+ years. If you are going to do that, it had better be a significant improvement, and not some inconsequential improvement. I don't see OS X as a big enough improvement. It is an improvement over it's predecessors in many respects, but it also has some regressions, which are inevitable when you try to reinvent the wheel.
Mac applications act this way due to legacy decisions made for the original circa 1984 Mac, not because it's the right way to do things. At the time, it took the Mac a long time to start applications. Apple decided on this behavior to make the computer more responsive when opening new documents. Now days, document open much more quickly and this behavior is no longer required.
Fine if you have a brand new high spec Mac. But I've got a 2008 MacBook, upgraded to Lion. I've just timed reopening xcode (with a single project and the organizer open). It took 9 seconds. Pages took 7 seconds.
And of course if you don't have it in the dock, you need to add the time taken to find the app in to apps folder or Launchpad.
And what't the benefit to closing when the last window closes? You potentially save a a space in the dock. and... No, that's it. Vitual memory means that memory consumption isn't really an issue.
Mac OS does it be right way.
Note that new OS designs on phones and tablets tend to not close apps either. They only get closed when lack of resources demand it. And then the OS does it's best to hide the fact that it was closed by restarting in the exact same state.
Windows (and the way Linux GUIs copy it) are a suboptimal way. It always was, and it still is. It's just that because of market share most people are used to that way and consider it normal.
9 and 7 seconds are not a long time to start a program and load a file. I remember apps on old Mac System taking upwards of 30 seconds to start and load a file.
And since you brought up phones, I have a Nexus One. When I got it it would go about 1.5 days on a charge. Then I installed an application to kill background apps. Now the phone goes easily 2.5 days and is generally more responsive.
If a program has something useful to do in the background than it should be implemented as a light-weight daemon, rather than a full blown app like iTunes.
A demon, by definition, is a background task with no UI. Yet a music player needs a UI. That isn't the same thing as having a window. OSX makes that distinction at the fundamental level. Windows doesn't, and Linux UIs typically follow Windows even when it's broken.
OK, you are right about Daemons not having UI's. But windows does make a distinction. There are services that run in the background. There are apps that run in the system tray and provide a light-weight UI. You can always just minimize your music player when you are listening to music. Also, windows apps can be programmed to not shut down when the last window is closed and instead take residence in the system tray. My question is, why does a music player have to run all the time even when y aren't playing music? Is it to watch for system events or to download updates in the background? Those tasks are better suited for a daemon or a scheduled event.
Fixed, no, but did you every use an original mac? They were woefully underpowered for the graphical UI. Amazing machines for their day, but they were slow.
when I click on a running app in the doc that has no open windows, the program doesn't do anything. It should, at that point, actually respond; open a new project, give me a file-open dialog box, anything but sit there looking pretty.
Many applications do just what you're asking. Click on the Finder and if there aren't any open windows it will pop one open with your default directory. Click on Mail and if you don't have a mail browser window open then it will open one.
It's an application-defined behavior and most of Apple's applications do something when they don't have an open window and their dock icon is clicked. Some 3rd party applications don't follow this UI principle and, yes, it's a shame when they don't.
The other issue with this behavior is that it is not easy to tell at a glance to tell what programs are running.
It is? All you have to do is glance over at the dock, everything that's running has a dot next to its icon. If that's not clear enough for you then you can always just clear out the non-running apps in your dock and only let the running apps show. Now EVERY app in the dock is a running app.
If we're talking about applications with open windows (which is what would show up in the Windows' Taskbar) then just use Apple's Mission Control. Every running application will have its own group of windows, grouped with the icon for the application. It's quick and easy to see what's running and pick a window.
Also remember that under Mac OS a running program that doesn't have any open windows doesn't need to use that many resources. It will stay in memory and possibly do some processing but if its resources are needed they can be reclaimed by the system and the application will be put in a suspended state, its memory will be paged out to disk, its running threads will be suspended or given a lower run priority, etc. Thus you can keep a lot of applications running in the background without windows and not worry that they'll slow you down much.
Yes, the dots are a challenge for me. I have to scan the bar and count dots. I find that annoying. I much prefer an OS that clearly groups icons. I tend to hyper-focus on what I'm doing and get annoyed by distractions like having to examine icons for dots when I realize I have too many apps open and have to shut them down. There is also that fact that as I get older and more nearsighted I am finding it harder and harder to pick out these details quickly on a computer screen.
It's a joke because there is no server-grade Apple hardware. Apple has allowed OSX VM's as guests on OSX host machines for many many years
The fact that the X button sometimes closes the application, and sometimes leaves the application running without a UI is also bad.
Why is it bad? It's a developer choice do do whichever is more appropriate for the app. On Windows an app MUST close when its last window closes, unless the developer puts it into the system tray.
The reasoning for leaving the app open when a multiple document app has it's last window closed is straight forward. It's a common usage pattern to finish working on one document and then start working on another. If apps quit when the last window closes, then this happens: The user closes the first document, and the UI to open the next document (File/Open) disappears. They then have to restart the app, which involves waiting, before they ca open their next document.
But for apps which are not document based, that argument doesn't apply. Closing the window on a single window app really does mean you've finished working with that app for the time being.
Then there are other reasons for choosing one behaviour or another. If an app does useful work even when there are no Windows, then of course it makes sense to keep it open. iTunes is an obvious example.
There's a reason why Mac developers have this choice and Windows developers don't get it (apart from the system tray utility option). Because with Windows, the disappearance of the last window means that access to the menu has also disappeared. That's not the case with Mac.
Mac applications act this way due to legacy decisions made for the original circa 1984 Mac, not because it's the right way to do things. At the time, it took the Mac a long time to start applications. Apple decided on this behavior to make the computer more responsive when opening new documents. Now days, document open much more quickly and this behavior is no longer required. Personally, the behavior drives me nuts because when I click on a running app in the doc that has no open windows, the program doesn't do anything. It should, at that point, actually respond; open a new project, give me a file-open dialog box, anything but sit there looking pretty. Programs that do something useful in the background with no open files are few and far between. If a program has something useful to do in the background than it should be implemented as a light-weight daemon, rather than a full blown app like iTunes.
The other issue with this behavior is that it is not easy to tell at a glance to tell what programs are running. The strength of the Windows Task Bar is that it clearly separates running programs from application launch icons. Certainly, this is a matter of what people are accustomed to, but for myself and I think many people who are accustomed to windows, this is infuriating.
I'm willing to pay $2.99 for a good book. I might be willing to pay 99 cents for a piece of trash. What I'm not willing to do is to pay the same price for an e-book as I would for the currently in print version, especially when the in print version is a $25 hard cover. Understandably, there are greater costs when it comes to technical works. But, people do not and will not care about the author's labor. They care about the enjoyment they get out of it. People generally enjoy print books more and therefor value them more and are willing to pay more for them. Just like everything else, a book is worth what someone is willing to pay for it.
The essential problem with selling downloadable content, which all consumers grasp intuitively but that publishers seem foggy on, is that the supply is functionally infinite. Econ 101 teaches the supply vs demand pressures on pricing. In the case of printed books, the publishers have to predict demand, then they print the amount of books they expect to be able to sell, and they mostly sell then at a fixed-ish price. Around 20 to 30 bucks for a hardback, discounted to about 18 if they don't sell. Paperbacks sell for around 6 to 10. That artificially limits supply and helps them keep prices stable. After all, if you see a book on the shelf that interests you, you know that it may not be available next month or next year. But with ebooks, there is no limit to supply. Buyers know they can get a copy, unless the publisher stops selling, and they know that it costs the the publishers virtually nothing to keep a copy on their servers for anyone to download 2, 5, 10 years down the line, and they know that it costs the publisher virtually nothing to sell them a copy. Publishers have been very good at predicting demand for printed books. What publishers have been very bad at is estimating how much people are willing to pay for ebooks. Instead, the large publishing houses seem to have been operating in a way to protect their print book sales from erosion by ebooks.
This is too bad, because what major publishers bring to the table is the expertise to help an author refine their work and market it effectively. They also have reputations to protect, so what they print will not totally suck, and consumers know that. But, if these publishers don't adapt to the times, they will go out of business. What publishers need to do is to realize that X number of people are willing to buy at 5.99, y number of people will buy at 2.99, and z number of people will buy at 0.99. They should be pricing ebooks dynamically. Introduce them high to get the established fans and let the price drop accordingly as demand lessens. If demand spikes as word spreads, then let the price rise as well.
US corporations have largely lost the will to innovate. Just look at the decline of in house R&D divisions. They used to be commonplace in US firms, but now they are hard to find firms. These days, SOP for large corps is to let the start-ups do the innovating and try to buy up the ones that are succeeding. That makes sense in many ways. Why risk the business on developing a new research that may never succeed or even come to market? If you do internal research, why look into radical new technologies when it is easier and more reliable to incrementally improve your existing products?
However, American businesses have missed out on a lot of opportunities or been late to market because of this mentality. The US auto industry is the poster child for missed opportunities. In all likelihood, Detroit will only be interested if the prototype has been developed and proven to work, Toyota buys up the patents, successfully scales up production, brings them to market, and they sell like hotcakes.
Does the duck weigh the same as witch, and is it therefore made of wood?
Even with strawberries?
I use an external dock for SATA drives or a USB to IDE adapter and plug it into a spare windows PC in the office. A quick format and then "cipher /W" does the trick.
Oh, I almost forgot the most important step.... I teach an intern how to do it so that I'm not wasting my own time.
I agree. I have done this trick several times to recover data at work. Usually we have a second identical drive on premises, and sometimes I have to use Ebay.
But not as much fun....
It's nickel-and-dime'ing. It's fine to pay for what you use, if you are using it for a small amount of data. But how much do you want to store? Lets say 100GB. It seems very reasonable to me for someone to keep that many pictures. It would cost you $168 per year, but your cost will soon balloon as you collect more pictures. There are on-line backup services that charge a lot less than that per year with no caps. And I did not lie about the transfer fees. They charge you to download your data. If your drive crashes you have to pay them to access your own data. S3 may be a terrific service for storing web site data or whatever, but it's not a service for the average user looking for on-line backup.
My problem with Amazon S3 is the pricing structure. They charge for storage per month, plus transfer fees, plus system requests, and they want you to move large quantities of data using portable storage. Their pricing page says it all.
Why not roll your own SAN? Build a small server out of cast-off parts. Stuff it with whatever size hard drives you want. Install your choice of OS and run the software to make it an iSCSI target. The beautiful part to it is that it is upgradable and repairable.
Depending on how much data there is, this is the cheapest solution for the long haul. Startech SATA docs are under 50 bucks. SATA drives start under 30 bucks. There is no need to buy any software, scripts can be scheduled to copy your data to external drives. This is not as safe as an online backup, but after a couple years of service, an on-line backup solution will cost more.
I don't disagree with you. But the same logic applies to the cable operators. They sell their own advertising. Depending on your local population size, they could make cable TV free and ad supported. Broadcast networks have done this for decades. Sure, they should charge the customer for installation and equipment, but the monthly service should be free.