Hacker News Reader

Comment by leejo | Hacker News Reader

Something like this already exists: https://www.ascribe.io/ # I had to use it to submit some photos to a call for work a couple of years ago. After the call I wanted to delete the work, because I didn't want it hanging around in public view. Oh look! You can't! Because It's blockchain!

Immutable history is not always a compelling requirement, in fact sometimes it is at odds with many a use case.

View on HN

This is what my criticism is, as well. A few days ago I had been pondering a similar concept of a blockchain-based platform for tracking image ownership. The main blocker I encountered while thinking about it was the problem of immutability. Specifically, what happens if something is wrong?

Like, the first problem I see here is a troll just importing images from all over the Internet and claiming ownership. If something like this happens, would it be possible for the actual artist to ever regain "ownership"?

You really think someone would do that? Just go on the blockchain and tell lies?

If the photo was already on the internet in a server, there are already records (like the Internet Archive) to verify that it had to have been created prior. The blockchain only adds an additional proof, but can't revert some past history that some other entity can verify actually happened.

Then why add the extra complexity of the blockchain at all?

You may be the only one who has the original raw file and publish a scaled down or compressed jpeg version. You should still have the negative.

Wouldn't it be enough to submit a hash of the photo? Then sure, that hash is around forever in the blockchain, but without the original photo that hash is useless. It can only be used to prove that the photo existed at the time the hash was added, and with a digital signature, who added it.

You could steal a photo by tweaking a single pixel or EXIF field, and suddenly it's an original image with no registered hash.

Yeah, any system like this has to deal with the analog hole somehow, and there really are no perfect solutions. Reminds me of the post the other day about a white noise video on Youtube getting a bunch of copyright infringement notices. You can take a signal out of the analog, but you can't take the analog out of a signal!

You can do a lot to mitigate that http://www.hackerfactor.com/blog/?/archives/529-Kind-of-Like...

I think the difference is that Ascribe doesn't have a token, so payment doesn't take place on the platform. Ascribe is merely a registry & they are building tech that will scan the internet checking for unauthorized copies of work. I think the scanning ability is very powerful, but that's totally outside of blockchain! I think the value of KodakCoin is the payment mechanism, which in theory will be decentralized i.e. replace Paypal. I'm still trying to wrap my head around the benefits. Bottom line though, everyone knows Kodak and no one knows Ascribe, so I can see how Ascribe would be a good acquisition target for Kodak down the line! Please challenge me!

> I think the difference is that Ascribe doesn't have a token, so payment doesn't take place on the platform.

Who cares? More concretely, how many professional or semi-professional photographers care?

> Bottom line though, everyone knows Kodak and no one knows Ascribe

But Ascribe is building useful technology while Kodak throws around buzzwords.

> Immutable history is not always a compelling requirement, in fact sometimes it is at odds with many a use case.

Good point. Perhaps the technology will account for this somehow, and thus be distinguishable in some way from other blockchain technologies . . . ?

Reply on news.ycombinator.com