EOF - If You Don't Believe in DRM, It Can't Hurt You

“Keep your management off my digital rights” isn't merely a slogan for freedom lovers. It's a smart IT decision.


Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

Details on DVD release dates

Anonymous's picture


Anonymous's picture

too bad I can't leave comments there. DVDs get copied really easy.

It gets pirated anyway.

Anonymous's picture

Everything gets pirated, DRM or not.

Absolutely. The fact is, if p

powdermonkey's picture

Absolutely. The fact is, if pirates (or even the merely curious and bored) want to break something, it will be broken and, probably, distributed in a broken form. People who are willing to circumvent the law will get the broken (ie, fully functional) version; people who are "honest" get the function-limited version. By paying for the genuine article they receive a product of lesser value. So where is the incentive to be honest?

James's Law?

James's picture

There needs to be a named Internet Law (like Godwin's, etc.) that states something along the lines of:

As a DRM scheme grows more complex, the ratio of illegitimate users affected to legitimate users affected approaches zero.

The only people DRM hurts are those who purchase the DRM'd product. Pirates will crack it, then distribute the product with DRM disabled. This is not hard to understand.

Don't forget artists

Anonymous's picture

DRM also hurts artists when it's easier to use an illegal copy than a paid-for but DRM-infected copy.

It's not about honesty.

Anonymous's picture

If it were about honesty, there would be some form of ethics involved.

One bit of information?

Greg's picture

This may be a little pedantic but ...

Bill Gates of Microsoft, in an interview with gizmodo.com, tried to pitch DRM using the example of an HIV test result, which is literally one bit of information.

According to Shannon's Information Theory, an HIV test result would only be "literally one bit of information" if the test result has a 50% chance of being positive and 50% chance of being negative.

Now, while these a priori probabilities may be the case for a health service primarily serving gay prostitutes in Thailand, for the general population something like 99%/1% would be closer.

This would require something like

H(T) = -0.01 log2 0.01 - 0.99 log2 0.99
= 0.0664 + 0.0144
= 0.0808 bits

or 8% of a bit!

NB: This makes the author's original argument even more compelling.

See: http://en.wikipedia.org/wiki/Information_theory



One Bit?

Anonymous's picture

I might be missing something but I cannot see how you could convey a test result (+ve/-ve) with less than one bit (regardless of how much "information" is in that bit.

You can convey the result in

Anonymous's picture

You can convey the result in less than one bit of actual data if you compress it along with other stuff. For example, assume 1 in 16 of 15000 independent tests are positive and you want to encode these 15000 bits. Without even getting into serious compression algorithms, let's pack results in groups of two:
0 both negative
100, 101 one or the other positive
11 both negative
This crappy code uses an average of 0.5*(225/256)+1.5*(15/256)*2+1*(1/256), that is (112.5+45+1)/256 =
0.619, bits per input bit.


Anonymous's picture

it takes more bits to explain the 8% bit.

leakable bits

Anonymous's picture

But we're only concerned with test results that are at risk of being leaked, not all the test results in the whole database. Untrustworthy employees are probably more likely to leak a + than a -.