It’s like this: If you send data to someone over the Net, you can’t control what they do with it. At least cost-effectively. Or, if you want a good outcome.
This applies to Internet Standards. Any standard that tries to constrain the way in which data, once received, is processed, is broken.
Similarly, to business. Any use of legislation or technology that tries to control what people can do with digital media objects, once they’ve been transmitted, is broken. Also any business model that relies on such control.
For supporting arguments, check the last five years of this blog. I offer as further supporting evidence the fact that the Internet works and that it’s a good place to learn, have fun, and make money.
This is true irrespective of value judgments. If it were a bad thing, we’d just have to live with it. Fortunately, this is a very good thing.
Comment feed for ongoing:
From: scott lewis (Feb 19 2010, at 11:09)
So, XML is very, very broken, then?
[link]
From: stand (Feb 19 2010, at 11:34)
Here's a point of discussion then. Where does this leave the prospects for sharing electronic medical records?
[link]
From: JulesLt (Feb 19 2010, at 12:11)
Oh, I don't know, there's plenty you can do with the law . . . the GPL for example.
[link]
From: John Cowan (Feb 19 2010, at 12:14)
"Any use of legislation or technology that tries to control what people can do with digital media objects, once they’ve been transmitted, is broken."
That's like saying that any use of legislation or technology that tries to control what people can do with knives, once they've been sold, is broken. And yet knife crime remains at tolerably low levels, even in the U.K.
[link]
From: brad emerson (Feb 19 2010, at 12:23)
yes.
[link]
From: Gregory (Feb 19 2010, at 12:52)
SSL is pretty darn cheap.
[link]
From: Steve Loughran (Feb 19 2010, at 13:20)
This is tricky. A lot of people criticised the MS OOXML spec because it didn't specify anything about how "conformant" apps rendered data. HTML5 does appear to be trying to lock down some of the stuff, so that people coding for it can be somewhat confident that it will be rendered similarly. But will it work? I doubt it. What if the far end is only a black and white laser printer? Can people mandate the colour to B/W transforms? It's those unexpected uses of your data that are both interesting and hard to control.
[link]
From: Mr Universe (Feb 19 2010, at 13:54)
You can't stop the signal Mal.
[link]
From: William (Feb 19 2010, at 14:05)
The corollary of this "let your data go" perspective is: don't let your data go (if you don't want it to be reused in ways you don't control).
So rather than "let your data go", I think of this as "realize that when your data is gone, it's gone". And as you say, it's (on balance) a good thing.
[link]
From: Dave Walker (Feb 19 2010, at 15:42)
Depends on what you've done to the data, beforehand, based on the threat model you associate with it.
If you've encrypted it before you sent it, then you have a pretty good handle on the net not accessing its cleartext content; checksum it, and you can show they've not corrupted it, either.
They can delete it, of course, so the sensible way to defend against that threat is to scatter-gun it out there, with different copies encrypted with different keys (provided you have the infrastructure on-hand locally, to manage all these keys and handle re-mirroring when you want to modify something).
It's possible, but nobody ever said it would be easy...
[link]
From: len (Feb 19 2010, at 20:10)
Well said, Tim. And right.
[link]
From: Rick Jelliffe (Feb 19 2010, at 21:11)
John: I don't know that there are UK laws on what you can do with knifes are there? There are laws on not hurting people, and laws on not carrying around dangerous weapons. Is it your theory that if there were no laws, crime rates would stay the same? :-)
[link]
From: Eddie (Feb 20 2010, at 15:48)
Here is a topical link. At least as this discussion pertains to Canada.
http://arstechnica.com/tech-policy/news/2010/02/canada-no-country-is-farther-behind-on-online-copyright.ars
Regards.
[link]
From: Erik Neu (Feb 26 2010, at 13:24)
Seems at least 90% accurate. I am sure if you think hard enough you can find hypothetical exceptions. Just the fact that you have to think so hard to find them shows the general accuracy of this statement.
As to the knife analogy, I think it doesn't work so well. The laws that prohibit doing certain things that potentially might involve knives are generally focused on the thing not done (attacking people, perhaps carrying any dangerous weapon in pulbic), and not on the specific not-permitted activities of the knife object. The laws don't go into not using hunting knives to open envelopes, pry open stuck drawers, etc.
[link]