The article is quite clearly mistaken as regards peer-to-peer networks. The copying happens at the source peer, not the target peer. There has been caselaw about copying a program into RAM (e.g. running it) and onto a network (e.g. sharing an NFS installation of commercial software). That law would all come into play here.
Also, hard disks aren't taxed yet, as has been pointed out. Do the figures, per useful-megabyte of storage. Of course they might even fiddle it by saying that the normal hard disk recording mode is mp3, and compute that per hour against casette tape, in which case an extra 540 dollars goes onto the price of a 120gb hard disk. ($.27 per hour, 60 megs per hour as mp3,...) If they really wanted to try it on, they'd compute using 32K mp3 too.
This is clearly a journalistic article and while it raises some points of merit, the respose here has been utterly disproportionate.
That is one of the most contentless articles I have seen in a long time.
A regex is a type 3 grammar. Type 3 grammars haven't really changed since Chomsky's time.
The smartarses will now proceed to point out that a) Perl is actually limited type 2 b) Some change noone knows or cares about was made to some definition of the Chomsky hierarchy in ninteen dumdy-dum.
I have been using Mason and related packages for a long time. I agree strongly with the people who say that mixing code and markup is a bad idea. I would go even further from my own experience, and say that mixing execution and display in any way is a bad idea.
On receiving a web request, the first thing to do is process the input, perform any operations, and then set up necessary information for doing a page display, using a tool such as Mason, with the least possible amount of code. My Mason pages rarely contain more than a <% $object->get('foo') %>.
In this way one has the versatility to redirect internally if an operation fails, and one doesn't have to cope with a page crashing halfway through display.
Mason is an excellent tool for controlling a structured web site. Mixing markup and code, execution and display is a separate, general topic, and should be considered apart from any specific tool.
Slashdot Mirror
The article is quite clearly mistaken as regards peer-to-peer networks. The copying happens at the source peer, not the target peer. There has been caselaw about copying a program into RAM (e.g. running it) and onto a network (e.g. sharing an NFS installation of commercial software). That law would all come into play here.
...) If they really wanted to try it on, they'd compute using 32K mp3 too.
Also, hard disks aren't taxed yet, as has been pointed out. Do the figures, per useful-megabyte of storage. Of course they might even fiddle it by saying that the normal hard disk recording mode is mp3, and compute that per hour against casette tape, in which case an extra 540 dollars goes onto the price of a 120gb hard disk. ($.27 per hour, 60 megs per hour as mp3,
This is clearly a journalistic article and while it raises some points of merit, the respose here has been utterly disproportionate.
I bought my HP laserjet (4 with PostScript and 10baseT) for 10 quid. (Actually I bought three.) I pay 60 quid a time for a 98X cartridge. So...
That is one of the most contentless articles I have seen in a long time.
A regex is a type 3 grammar. Type 3 grammars haven't really changed since Chomsky's time.
The smartarses will now proceed to point out that
a) Perl is actually limited type 2
b) Some change noone knows or cares about was made to some definition of the Chomsky hierarchy in ninteen dumdy-dum.
Foo.
I have been using Mason and related packages for a long time. I agree strongly with the people who say that mixing code and markup is a bad idea. I would go even further from my own experience, and say that mixing execution and display in any way is a bad idea.
On receiving a web request, the first thing to do is process the input, perform any operations, and then set up necessary information for doing a page display, using a tool such as Mason, with the least possible amount of code. My Mason pages rarely contain more than a <% $object->get('foo') %>.
In this way one has the versatility to redirect internally if an operation fails, and one doesn't have to cope with a page crashing halfway through display.
Mason is an excellent tool for controlling a structured web site. Mixing markup and code, execution and display is a separate, general topic, and should be considered apart from any specific tool.