Comments on: RIAA Releases Watermarking Payload Specification

By: popicok

popicok — Sun, 09 May 2010 15:06:50 +0000

Whats to stop people using Audacity set to record “Stereo Output Bus” which nearly all even basic ‘on board’ sound chips allow these days. People are using the likes of we7 and Spotify with Audacity in this way. Careful line up and tweaked level control gives you a very good quality copy.

By: Tomasz

Tomasz — Sat, 27 Feb 2010 14:32:17 +0000

Yes, this is the right answer. Thus, if we can’t share the new ideas of digital watermarking algorithms implemented in DRM systems we will never choose the one, optimal method. Furthermore, every DRM system developer will claim that his watermarking algorithm is the best solution. Till now, the well known trade-off between robustness, data payload and transparency seems to be unsolved task.
Does we really want to have? The winner is the only one and probably there is a unnamed winner right now, but what a pity we will never know his/her name. Any solutions? Top 10 list, international digital watermarking championships, global rank of the algorithm popularity, Internet watermarking platform driven by users and developers? etc.

By: Miss Cucumber

Miss Cucumber — Mon, 22 Feb 2010 09:50:24 +0000

“Because digital watermarking is not a cryptography (who did write?, any hint? paper?)”

It is there:

http://www.irisa.fr/temics/publis/2006/42830001.pdf

By: Tomasz

Tomasz — Tue, 16 Feb 2010 22:11:14 +0000

Thus, is this case to be more precise should we still write “watermark payload”? A lot of people would consider that it is possible to embed such high data payload (108 bits of data).

May I digress for a moment? I have notice there is a new technology named Hidden Link Technology (HLT) based on digital watermarking (https://mydrmspace.com/hlt/)

Is it new quality in the Internet advertisement or hidden message distribution?

By: Bill Rosenblatt

Bill Rosenblatt — Tue, 16 Feb 2010 21:49:29 +0000

Tomasz, Those of us who have worked on content identifier standards tend to view IDs as tantamount to metadata. The usual practice is to use an ID number as an index to a database that contains metadata. The music industry has identifier standards like ISRC that facilitate this. It would be impossible to fit useful metadata in a typical audio watermark payload size, such as the 108 bytes used in the RIAA standard. ID standards like ISRC are designed to be used as watermark payloads for just these reasons.

By: Tomasz

Tomasz — Tue, 16 Feb 2010 21:39:19 +0000

Bill,

I think that digital watermark ID is not the same as metadata content.
In particular, the ID payload and the metadata payload can be different. Of course, it is possible to allocate ID to the metadata content. In this case the can observe the huge “gain” of the data payload.

By: Bill Rosenblatt

Bill Rosenblatt — Tue, 16 Feb 2010 21:13:37 +0000

Tomasz,

Actually this is exactly what the RIAA is trying to do – to standardize on the metadata and to leave it to watermarking vendors to implement their own proprietary algorithms. You are correct, the vendors don’t reveal details of their algorithms.

By: Tomasz

Tomasz — Tue, 16 Feb 2010 21:03:07 +0000

There is another solution: let’s standardise metadata for digital watermarking e.g. like Dublin Core standard for CD. It seems to be much simpler to embed unique ID (as a not standard and free for watermarking vendors) and assign ID to metadata (http://www.youtube.com/watch?v=Jv1LpAKXIO4).
Because digital watermarking is not a cryptography (who did write?, any hint? paper?)
I strongly do not recommend revealing any digital watermarking algorithm details.
Of course in my opinion, Kerckhoff’s Law should be taking into consideration during digital watermarking algorithm developing process but it is not an explanation I should reveal any algorithms details.
I think that is still possible to choose of the one, mature digital watermarking standard without revealing embedder and extractor details. Just let’s share algorithms (executable files) with users let’s give them possibility to choose the best one.

By: Bill Rosenblatt

Bill Rosenblatt — Wed, 19 Aug 2009 12:50:05 +0000

Thanks for the explanation. This all hinges on the degree to which the record labels are actually going to use transactional watermarking as an anti-piracy tactic, as opposed to using it for other purposes (e.g. marketing and contextual advertising) and as opposed to not using transactional watermarking at all. It’s certainly a factor that they ought to be taking into consideration (if they aren’t already) when comparing the efficacy of transactional watermarking for anti-piracy with that of DRM.

By: Miss cucumber

Miss cucumber — Wed, 19 Aug 2009 12:41:28 +0000

Collusion happens where several pirates buy the same content, and mix the versions they get to forge a pirated copy which can’t be traced back to one of them.
There are usually three possibilities:
1) If the embedded bits coding the transaction ID are hidden sequentially (for instance a bit per second), the attackers forge a pirated copy picking chunks from the different versions they received.
2) If these bits are spread all over the content, they use signal processing algorithms to mix the versions (like averaging sample by sample)
3) and of course, they can degrade the content adding noise, compressing it at a lower bit-rate, etc.

Of course the watermarking is robut so that 3) alone is useless, but I am not sure that this is the case with 1), 2) or any combination of the three kinds. Conclusion: the number of ways for making a collusion is almost infinite.