Encrypt your data using a key large enough in proportion to the length of time it will take to brute force if someone started today with a supercomputer. Essentially it takes X amount of time to check if a key is valid multiply that by number of combinations and you have a rough guess. So if you want something to be safe for a longer period of time (assuming no fundamental weakness is found in the algorithm), then encrypt it with a larger key - every bit doubles the probable time to break it at current cpu speeds. Of course you have to factor in the approximate doubling of cpu speeds every 18 months... but all that really means is that if we add a bit to the key length every 18 months going forward it will continue to take just as long to break into newly encrypted data.
The fact remains that most people don't have anything that needs to be kept 'secret' for a long time anyway. Credit card numbers for online purchases? Those expire after a couple years and the amount of financial gain is not worth the time/cost to break the code. Given that you still need supercomputer equivalents to brute force this encryption it's unlikely that your neighbor is going to be reading your email anytime soon. Even at 109-bit.
XDocs is just a modern clone of Lotus Notes
on
Microsoft takes on PDF
·
· Score: 5, Insightful
It's a just an editor and toolkit to put forms on your company intranet (and probably later the Internet) to gather data. That's all it is - not a PDF killer - not even a PDF competitor. From Microsoft's XDocs web site:
"XDocs," a code name for the newest member of the Microsoft Office family, streamlines the process of gathering information by enabling teams and organizations to easily create and work with rich, dynamic forms. The information collected can be integrated with a broad range of business processes because XDocs supports any customer-defined XML schema and integrates with XML Web services. As a result, XDocs helps to connect information workers directly to organizational information and gives them the ability to act on it, which leads to greater business impact.
Does that sound like a pdf killer to you? Does it even sound like they're after the same market? Sure they're using XML and they're making "documents" - still sounds more like Lotus Notes than Acrobat. But who uses Acrobat/PDF to collect data? Yes, there are forms in PDF, but the implementation is not nearly flexible enough to build a data collection application, nor can you build decent data collection apps around MS Word.
XDocs is designed to work with any customer-defined XML schema. Where's the proprietary nature there? You give it your proprietary schema and then you use it to build forms to collect data into that schema. All Microsoft is doing is implementing a framework to easilly collect and present information. This is exactly what Lotus Notes was doing more than 5 years ago, only with XDocs the collected data is stored using your XML DTD instead of Lotus's proprietary NSF format. I'm sure Microsoft will extend it to the web - just using an XSL transform to change the XDoc into HTML and collect your data that way.
None of this prevents you from using a PDF to archive resulting documents. To be sure, you can probably embed an XDoc form into an XML dataset and view the resulting file with an XDoc viewer - but that's still one more app that everyone needs, and PDF is still the best portable format for archiving all sorts of documents and images. XDoc just collects information. Yes... all very insidious of Microsoft. A PDF killer.. I don't think so. I don't even see it as a PDF competitor.
Because others have addressed this in a general sense, I will only ask for a specific example of PHP failing this. I agree that many programmers over-tie them, but don't blame the language for things that stupid programmers do.
PHP - unlike mod_perl w/Template Toolkit or HTML::Template, or the.NET ASPx model, or JSP with custom taglibs - does not give you a method to cleanly separate your dynamic html from the code that creates the data that drives it. Sure, you can write your own, or write 3x the code (or more) to get XSLT to render XML but why not just use something that gives you a decent easy to use system out of the box. The point of separating code, content, and presentation logic is that you can have developers writing code, editors developing content and designers work on the presentation. And *none* of them need to know how to do the others job. Some people are good at all three, most aren't - that's just a simple fact. The Template Toolkit presentation language is very simple. It does just enough to make HTML dynamic without the burdon of the underlying programming logic. Likewise with ASPx or custom taglibs.
I challenge you to show everyone just what the best way with PHP to accomplish a clean separation of presentation logic from code. And then we can count lines of code and judge the complexity.
Note that HTML itself is a protocol. Thus wrapping a protocol with yet another protocol may be a bit too wrap-happy in some cases.
I beg to differ... HTML is a mark up language - not a protocol. HTTP is protocol.
Re:Maintence must be easier
on
Yahoo Moving to PHP
·
· Score: 3, Informative
My understanding was that MySQL got popular because it had a more useful 'text' column type, which is all messageboards, CMS's etc really need, and admin is fairly simple.
Yep, that's pretty much it. If PostgreSQL didn't have the 8k row limit and the need for periodic 'vacuums' at the time when the web was starting to boom, it would have been the db of choice. Those problems are fixed now but it doesn't matter. It supported transactions, views, sub-selects, and most other SQL features years before MySQL did. The fact that PostgreSQL is now a better database in nearly all respects (including speed) doesn't matter because MySQL is entrenched.
Postulate: If you build a system that any idiot can use, only an idiot will want to use it.
I'd have to disagree. First off Windows is not that easy that an idiot can use it. Try putting a person who's never touched a computer in front of Windows or X Windows(running KDE/Gnome/etc). Either one would take time and patience - but the X Windows solution would take *more* time and patience to get to the same level of comfort with the operating system and say for example a browser and office suite. And it's not because people are idiots. 90% of what you do with a computer is auto-pilot. You probably don't think about using the mouse, typing, scrolling a window --- completely basic operations of a GUI interface that you don't ever think about. All you think about is the result you want, not how to accomplish it. With X Windows and the various window managers and gui toolkits you end up with a vast array of applications that all behave differently. Some are so different they actually require thinking about just how to use the interface itself.
Microsoft and Apple have spent billions developing a user interface that is clean, consistent, and easy to use. And unless OSS clones those interfaces down to the pixel - or spends the billions themselves - the interfaces are going to continue to suck. And that doesn't even count all the applications that need to be altered just to get a semblance of order. A standard user interface is a good thing! It helps a user immediately sit down with a new app and be as productive as possible. If you not only have to learn a new application but also a new interface you're going to be slowed down.
Anyway... my point is that anyone can use Windows, MacOS, or X Windows with enough time and practice. But, at least with Windows and MacOS you have a head start as soon as you learn the interface.
AFAIK, the problem you suggest has been corrected in later versions of emerge. Unless you remove a package forcefully you will be warned about the dependancy issues.
Why is that everytime someone mentions lfs, someone has to say, "Why not just use gentoo?"
Because it's a troll. It gets discussion started...
It makes us (the users) look like the next generation zealots. I have a better idea - learn what distros do what things and at what difficulty and then choose for yourself. Suit your own needs, dammit.
Exactly! Over the past 8 or so years I've used Redhat, Debian, Mandrake, a couple of BSD's, and LFS. Now I use Gentoo because it suits me - and I think it would suit nearly everyone who has an interest in LFS. I can't see why most people, even those who want the flexiblity of a source based system, would spend the time to maintain an LFS based system unless they had nothing on a computer except learn about how the computer works. You have no time left over to take advantage of what the computer can actually do for you -- save you time. How much different are your compile time choices going to be from the ebuild's defaults? And if they are different, then edit the ebuild file.
LFS is just tedious to maintain. Which is part of the reason why it's perfect for an embedded system. You get exactly what you need, nothing more, and you never change it.
As others have said, lfs is great for getting your hands dirty and learning some stuff. Gentoo is for after your hands are dirty and you want to clean them up...
LFS is a wonderful experience to install. I'm not discouraging anyone from going out and installing LFS. I just believe that after you've done it once, you don't need to do it again - and that's where Gentoo comes in. Gentoo essentially is what Automated LFS aims to be.
Re:Gentoo config files
on
LFS 4.0 Released
·
· Score: 2, Informative
The gentoo config files are not all that bad - they bring order to the chaos that is a source based distribution. With LFS you either remember what you have installed (which may be easy since you tend not to install very much when you have to do it from scratch) or most likely keep track if it in a file (or on paper). Either way you need to know what you have installed so you have that info available to pass as configure options. With gentoo you keep track of that in one place. Then for every package that *can* use, for example, OpenLDAP it will automatically be configured to use it. It's so much easier than LFS - and yet I'm failing to see what you lose with Gentoo...
Why not just use Gentoo? You get all of the benefits of a fully customized and compiled distribution when you want it. Yet it's completely automated for when you don't want to be bothered with every little package that goes into a fully functional system.
Maybe in a million years there is a breed of post-humans that are REALLY tall (with other differences) and can't interbreed with the 'short ones'... those that didn't experience these changes.
Could it be possible that this is happening already? Look at all the infertile couples in the world... it may be possible that some of them are having trouble because they are in fact different (of course, still closely related) species that cannot interbreed and/or only certain combinations of DNA in their egg and sperm will produce a viable embryo. Perhaps we're becoming different species aleady and we just don't know it.
First off, you can't deny that mutation happens. We see it all the time in the real world (not some lab), and we even see it in modern humans. People born with tails, extra digits on their hands or feet, more or less body hair than thier parents, etc... the list goes on. It's mutation, it really happens - and people who have these mutated genes pass them on to their offspring. As soon as those genes are passed on that's evolution in its simplest form. Evolution is not a theory, it is a fact. It is undeniable for it can be demonstrated in the real world for all to see. Whether you believe that it happens through pure chaos or if God caused those genes to mutate - or even some combination of the two - that's up to you. Darwin's Origin of Species describes in great detail and lays out all the facts that were known at the time. No scientist, whether they believe in creation or not, can deny that evolution occurred and is occuring.
Now, lets move on to the broader theory of evolution which essentially postulates that if you could look back in time you would see that all life on Earth decended from from single cell organisms and ultimately strands of dna. (Which isn't to say that parallel evolution didn't happen in the early stages and it's entirely possible that the different kingdoms of biology each came from their own pools of goo. But, it's entirely beside the point.) Look at the timeline.
3.5 billion years... an incomprehensible period of time - 500 thousand times the length of recorded history, 50 million times the length of a human life. Is it any wonder why we have missing links in the geological record? No, and it's impressive that we have found what we have. We don't need the missing links to see that evolution happened. And we certainly don't need the missing links of the last 15 or so changes in a beetle which went through millions of changes just to become a beetle. The Bombardier Beetle is an odd occurance to be sure. It is not, however, a gaping hole in the theory of evolution. It is not usual or expected - but what is? We're lucky to be here debating this at all.
Being able to drop a column without recreating the table is a nice enhancement --- but Prepared Queries, Table Functions, Priviledges, and Schemas are the things that people should really be cheering about. Table functions especially - they give managable stored procedures to postgresql. No more hacks with views or temp tables, just the ability to return rows of tabulated data from functions and use them in select statements. This will really be an excellent release.
The idea is perfectly sound. It's no different than charging a watch battery or spring with the energy involved in walking, or a windup radio. It's called energy recycling.
It's very different... not only is the energy required to power a watch just a small fraction of what is being put into it by being worn while walking, but you're forgetting that the energy from walking comes from the person wearing the watch -- not the watch itself! What you're describing is not energy recycling, it's energy transference. So, only if the car were powered by its driver, would we have a similar scenario.
Why the hell would using MySQL (you meant SQL Server right?) and ASP cost you anything more?
It's likely that he would have save huge amounts of time by using the immense open source codebase available for a MySQL/PHP solution. If all he wants is a hobbled version of Slashcode, then he would only have had to take PHPNuke and strip it down.
SQL Server/ASP would mean writing everything from scratch, like he said. So it's probably not the cost of the tools -- just the time to implement.
and what's with the link to the california soccar association?
ROFL - that's way too funny! I really should have checked that link - CSAN should have been CSAM.
But yes, I don't think I'd hesitate to use them in a production environment -- that's what testing is for. But really, there's no reason why those libraries shouldn't be trusted - mostly what they do is P/Invoke the functions from the native library code.
The point is moot though... there's no production quality version of mono to run these libraries on.
If you think C# is great then good for you. I for one don't care for a language that lacks the ability to write gui apps or database apps. Both of those are absent unless you are using windows.
What about GTKSharp, Npgsql, MySQL.Net and the list goes on.... it seems to me like you could build a gui database app sans ADO.NET and Windows forms.
The problem with C# is that all the open source code is scattered all over net instead of being in a community repository. There is one at CSAN, but last I checked it had a total of 3 modules. We need a central repository with a consistant namespace. I, for one, am tired of using something like JoesAssemblies.SomeNetworkProtocol when I could be using CSAN.Net.SomeProtocol.
I'm a Perl guy by nature - but I've embraced C# because it's the closest thing to Perl 6 we're going to see for a long time. The only disadvantage with C# is the lack of a huge code library to build your applications on.
or... Option 4: heads on own arm/servo and on opposite ends of the platter.
Why not just have the heads on the left and right side of the platter. it might make for a slightly larger drive, but you can absolutely have several arms/servos at different points on the disk. Based on the size of arms I'd say you could have up to 6 with quite a bit of breathing room. It's not hard to imagine what a nightmare the controller design would be though...
I wasn't commenting about the difference between desktop and server hard drives. The day is here now when we need high throughput at the desktop -- think video editing. Many desktop users don't need the low latency that servers require, but they do need the throughput.
That's not entirely true -- a 10k drive will pump data at a faster rate than a 7.2k drive. But the fact remains that ATA100 and ATA133 are much faster than a 10k drive can output - even ATA66 is barely going to be saturated by the data throughput of a 10k drive.
Re:Wondering why NPR might do this?
on
Blogspace vs. NPR
·
· Score: 2
Yes, I know what it says... What they're trying to do is prevent people from embedding material in their web sites -- not prevent people from linking to npr.org. I get my implication from the context. I agree with you though, as written, the policy is offensive. They might not get it - it's pretty clear that the people who wrote the policy are not web developers -- but that's why I'm sure the policy will be rephrased to get at the actual intent.
Re:Wondering why NPR might do this?
on
Blogspace vs. NPR
·
· Score: 5, Insightful
Yes - exactly. If anyone actually took the time to look at the Link Permission Request page they might see that what they are really referring to when they say links is links to the audio content. They're example text: (e.g., "Listen to NPR's David Kestenbaum's report on the Space Shuttle, originally broadcast on NPR's All Things Considered® April 4, 2002").
They did go about this all wrong by using very broad wording. I can't imagine that they don't want people linking to their html pages freely (e.g. http://news.npr.org/). It seems like everybody here is flying off the handle over what really is nothing. The linking policy has an intent, and I'm certain that the wording of it will be changed - within a week at most - to match that intent.
But it's not heard as it's being played - merely recorded onto disk. You have to take the disk out of the piano, send the files on it over the internet and then play it back -- synced up to the video that was transmitted live via satelite and recorded at the destination. This all takes about 30 minutes according to the article. That's not live... just a 'recording of a live performance'. Maybe later when they have it streaming from the piano telesynced to the video it can be called a 'televised live performance'. But it's still not live.
Slashdot Mirror
Encrypt your data using a key large enough in proportion to the length of time it will take to brute force if someone started today with a supercomputer. Essentially it takes X amount of time to check if a key is valid multiply that by number of combinations and you have a rough guess. So if you want something to be safe for a longer period of time (assuming no fundamental weakness is found in the algorithm), then encrypt it with a larger key - every bit doubles the probable time to break it at current cpu speeds. Of course you have to factor in the approximate doubling of cpu speeds every 18 months... but all that really means is that if we add a bit to the key length every 18 months going forward it will continue to take just as long to break into newly encrypted data.
The fact remains that most people don't have anything that needs to be kept 'secret' for a long time anyway. Credit card numbers for online purchases? Those expire after a couple years and the amount of financial gain is not worth the time/cost to break the code. Given that you still need supercomputer equivalents to brute force this encryption it's unlikely that your neighbor is going to be reading your email anytime soon. Even at 109-bit.
"XDocs," a code name for the newest member of the Microsoft Office family, streamlines the process of gathering information by enabling teams and organizations to easily create and work with rich, dynamic forms. The information collected can be integrated with a broad range of business processes because XDocs supports any customer-defined XML schema and integrates with XML Web services. As a result, XDocs helps to connect information workers directly to organizational information and gives them the ability to act on it, which leads to greater business impact.
Does that sound like a pdf killer to you? Does it even sound like they're after the same market? Sure they're using XML and they're making "documents" - still sounds more like Lotus Notes than Acrobat. But who uses Acrobat/PDF to collect data? Yes, there are forms in PDF, but the implementation is not nearly flexible enough to build a data collection application, nor can you build decent data collection apps around MS Word.
XDocs is designed to work with any customer-defined XML schema. Where's the proprietary nature there? You give it your proprietary schema and then you use it to build forms to collect data into that schema. All Microsoft is doing is implementing a framework to easilly collect and present information. This is exactly what Lotus Notes was doing more than 5 years ago, only with XDocs the collected data is stored using your XML DTD instead of Lotus's proprietary NSF format. I'm sure Microsoft will extend it to the web - just using an XSL transform to change the XDoc into HTML and collect your data that way.
None of this prevents you from using a PDF to archive resulting documents. To be sure, you can probably embed an XDoc form into an XML dataset and view the resulting file with an XDoc viewer - but that's still one more app that everyone needs, and PDF is still the best portable format for archiving all sorts of documents and images. XDoc just collects information. Yes... all very insidious of Microsoft. A PDF killer.. I don't think so. I don't even see it as a PDF competitor.
Because others have addressed this in a general sense, I will only ask for a specific example of PHP failing this. I agree that many programmers over-tie them, but don't blame the language for things that stupid programmers do.
PHP - unlike mod_perl w/Template Toolkit or HTML::Template, or the .NET ASPx model, or JSP with custom taglibs - does not give you a method to cleanly separate your dynamic html from the code that creates the data that drives it. Sure, you can write your own, or write 3x the code (or more) to get XSLT to render XML but why not just use something that gives you a decent easy to use system out of the box. The point of separating code, content, and presentation logic is that you can have developers writing code, editors developing content and designers work on the presentation. And *none* of them need to know how to do the others job. Some people are good at all three, most aren't - that's just a simple fact. The Template Toolkit presentation language is very simple. It does just enough to make HTML dynamic without the burdon of the underlying programming logic. Likewise with ASPx or custom taglibs.
I challenge you to show everyone just what the best way with PHP to accomplish a clean separation of presentation logic from code. And then we can count lines of code and judge the complexity.
Note that HTML itself is a protocol. Thus wrapping a protocol with yet another protocol may be a bit too wrap-happy in some cases.
I beg to differ... HTML is a mark up language - not a protocol. HTTP is protocol.
Yep, that's pretty much it. If PostgreSQL didn't have the 8k row limit and the need for periodic 'vacuums' at the time when the web was starting to boom, it would have been the db of choice. Those problems are fixed now but it doesn't matter. It supported transactions, views, sub-selects, and most other SQL features years before MySQL did. The fact that PostgreSQL is now a better database in nearly all respects (including speed) doesn't matter because MySQL is entrenched.
The Athlon XP 2800 runs at 2.25 GHz. It's not shipping yet, but they are producing it and it's priced 10% less than the comparable Intel P4 2.8GHz.
I'd have to disagree. First off Windows is not that easy that an idiot can use it. Try putting a person who's never touched a computer in front of Windows or X Windows(running KDE/Gnome/etc). Either one would take time and patience - but the X Windows solution would take *more* time and patience to get to the same level of comfort with the operating system and say for example a browser and office suite. And it's not because people are idiots. 90% of what you do with a computer is auto-pilot. You probably don't think about using the mouse, typing, scrolling a window --- completely basic operations of a GUI interface that you don't ever think about. All you think about is the result you want, not how to accomplish it. With X Windows and the various window managers and gui toolkits you end up with a vast array of applications that all behave differently. Some are so different they actually require thinking about just how to use the interface itself.
Microsoft and Apple have spent billions developing a user interface that is clean, consistent, and easy to use. And unless OSS clones those interfaces down to the pixel - or spends the billions themselves - the interfaces are going to continue to suck. And that doesn't even count all the applications that need to be altered just to get a semblance of order. A standard user interface is a good thing! It helps a user immediately sit down with a new app and be as productive as possible. If you not only have to learn a new application but also a new interface you're going to be slowed down.
Anyway... my point is that anyone can use Windows, MacOS, or X Windows with enough time and practice. But, at least with Windows and MacOS you have a head start as soon as you learn the interface.
AFAIK, the problem you suggest has been corrected in later versions of emerge. Unless you remove a package forcefully you will be warned about the dependancy issues.
Because it's a troll. It gets discussion started...
It makes us (the users) look like the next generation zealots. I have a better idea - learn what distros do what things and at what difficulty and then choose for yourself. Suit your own needs, dammit.
Exactly! Over the past 8 or so years I've used Redhat, Debian, Mandrake, a couple of BSD's, and LFS. Now I use Gentoo because it suits me - and I think it would suit nearly everyone who has an interest in LFS. I can't see why most people, even those who want the flexiblity of a source based system, would spend the time to maintain an LFS based system unless they had nothing on a computer except learn about how the computer works. You have no time left over to take advantage of what the computer can actually do for you -- save you time. How much different are your compile time choices going to be from the ebuild's defaults? And if they are different, then edit the ebuild file.
LFS is just tedious to maintain. Which is part of the reason why it's perfect for an embedded system. You get exactly what you need, nothing more, and you never change it.
As others have said, lfs is great for getting your hands dirty and learning some stuff. Gentoo is for after your hands are dirty and you want to clean them up...
LFS is a wonderful experience to install. I'm not discouraging anyone from going out and installing LFS. I just believe that after you've done it once, you don't need to do it again - and that's where Gentoo comes in. Gentoo essentially is what Automated LFS aims to be.
The gentoo config files are not all that bad - they bring order to the chaos that is a source based distribution. With LFS you either remember what you have installed (which may be easy since you tend not to install very much when you have to do it from scratch) or most likely keep track if it in a file (or on paper). Either way you need to know what you have installed so you have that info available to pass as configure options. With gentoo you keep track of that in one place. Then for every package that *can* use, for example, OpenLDAP it will automatically be configured to use it. It's so much easier than LFS - and yet I'm failing to see what you lose with Gentoo...
Why not just use Gentoo? You get all of the benefits of a fully customized and compiled distribution when you want it. Yet it's completely automated for when you don't want to be bothered with every little package that goes into a fully functional system.
Could it be possible that this is happening already? Look at all the infertile couples in the world... it may be possible that some of them are having trouble because they are in fact different (of course, still closely related) species that cannot interbreed and/or only certain combinations of DNA in their egg and sperm will produce a viable embryo. Perhaps we're becoming different species aleady and we just don't know it.
Now, lets move on to the broader theory of evolution which essentially postulates that if you could look back in time you would see that all life on Earth decended from from single cell organisms and ultimately strands of dna. (Which isn't to say that parallel evolution didn't happen in the early stages and it's entirely possible that the different kingdoms of biology each came from their own pools of goo. But, it's entirely beside the point.) Look at the timeline. 3.5 billion years... an incomprehensible period of time - 500 thousand times the length of recorded history, 50 million times the length of a human life. Is it any wonder why we have missing links in the geological record? No, and it's impressive that we have found what we have. We don't need the missing links to see that evolution happened. And we certainly don't need the missing links of the last 15 or so changes in a beetle which went through millions of changes just to become a beetle. The Bombardier Beetle is an odd occurance to be sure. It is not, however, a gaping hole in the theory of evolution. It is not usual or expected - but what is? We're lucky to be here debating this at all.
Being able to drop a column without recreating the table is a nice enhancement --- but Prepared Queries, Table Functions, Priviledges, and Schemas are the things that people should really be cheering about. Table functions especially - they give managable stored procedures to postgresql. No more hacks with views or temp tables, just the ability to return rows of tabulated data from functions and use them in select statements. This will really be an excellent release.
It's very different... not only is the energy required to power a watch just a small fraction of what is being put into it by being worn while walking, but you're forgetting that the energy from walking comes from the person wearing the watch -- not the watch itself! What you're describing is not energy recycling, it's energy transference. So, only if the car were powered by its driver, would we have a similar scenario.
It's likely that he would have save huge amounts of time by using the immense open source codebase available for a MySQL/PHP solution. If all he wants is a hobbled version of Slashcode, then he would only have had to take PHPNuke and strip it down.
SQL Server/ASP would mean writing everything from scratch, like he said. So it's probably not the cost of the tools -- just the time to implement.
ROFL - that's way too funny! I really should have checked that link - CSAN should have been CSAM.
But yes, I don't think I'd hesitate to use them in a production environment -- that's what testing is for. But really, there's no reason why those libraries shouldn't be trusted - mostly what they do is P/Invoke the functions from the native library code.
The point is moot though... there's no production quality version of mono to run these libraries on.
If you think C# is great then good for you. I for one don't care for a language that lacks the ability to write gui apps or database apps. Both of those are absent unless you are using windows.
What about GTKSharp, Npgsql, MySQL.Net and the list goes on.... it seems to me like you could build a gui database app sans ADO.NET and Windows forms.
The problem with C# is that all the open source code is scattered all over net instead of being in a community repository. There is one at CSAN, but last I checked it had a total of 3 modules. We need a central repository with a consistant namespace. I, for one, am tired of using something like JoesAssemblies.SomeNetworkProtocol when I could be using CSAN.Net.SomeProtocol.
I'm a Perl guy by nature - but I've embraced C# because it's the closest thing to Perl 6 we're going to see for a long time. The only disadvantage with C# is the lack of a huge code library to build your applications on.
Try these: Comment 1 regarding multi-armed hard drives [slashdot.org]
:-(
and
Comment 2 further explaining how they would work [slashdot.org]
Those comments must have been archived already -- "Nothing for you to see here. Please move along." is all I see
or... Option 4: heads on own arm/servo and on opposite ends of the platter.
Why not just have the heads on the left and right side of the platter. it might make for a slightly larger drive, but you can absolutely have several arms/servos at different points on the disk. Based on the size of arms I'd say you could have up to 6 with quite a bit of breathing room.
It's not hard to imagine what a nightmare the controller design would be though...
I wasn't commenting about the difference between desktop and server hard drives. The day is here now when we need high throughput at the desktop -- think video editing. Many desktop users don't need the low latency that servers require, but they do need the throughput.
That's not entirely true -- a 10k drive will pump data at a faster rate than a 7.2k drive. But the fact remains that ATA100 and ATA133 are much faster than a 10k drive can output - even ATA66 is barely going to be saturated by the data throughput of a 10k drive.
By throwing them in the snow of course.
Yes, I know what it says... What they're trying to do is prevent people from embedding material in their web sites -- not prevent people from linking to npr.org. I get my implication from the context. I agree with you though, as written, the policy is offensive. They might not get it - it's pretty clear that the people who wrote the policy are not web developers -- but that's why I'm sure the policy will be rephrased to get at the actual intent.
They did go about this all wrong by using very broad wording. I can't imagine that they don't want people linking to their html pages freely (e.g. http://news.npr.org/). It seems like everybody here is flying off the handle over what really is nothing. The linking policy has an intent, and I'm certain that the wording of it will be changed - within a week at most - to match that intent.
But it's not heard as it's being played - merely recorded onto disk. You have to take the disk out of the piano, send the files on it over the internet and then play it back -- synced up to the video that was transmitted live via satelite and recorded at the destination. This all takes about 30 minutes according to the article. That's not live... just a 'recording of a live performance'. Maybe later when they have it streaming from the piano telesynced to the video it can be called a 'televised live performance'. But it's still not live.