Author

Topic: Non-bitcoin Cryptography questions (Read 610 times)

legendary
Activity: 3472
Merit: 4801
February 08, 2016, 09:46:20 PM
#13
I feel like the people who will be most interested in your question will be in Dev&Tech, so I moved it there. It is rather off-topic there, so if someone complains then maybe it should be moved to Off-topic.

Thanks. I appreciate the consideration.

If you end up needing to move it to off-topic, I fully understand and won't mind.
administrator
Activity: 5222
Merit: 13032
February 08, 2016, 09:27:33 PM
#12
I feel like the people who will be most interested in your question will be in Dev&Tech, so I moved it there. It is rather off-topic there, so if someone complains then maybe it should be moved to Off-topic.
legendary
Activity: 3472
Merit: 4801
February 08, 2016, 05:34:58 PM
#11
-snip-
be posting in Technical Support, i.e. https://bitcointalk.org/index.php?board=4.0
-snip-

No, thats mainly the section for bitcoin core. Im afraid offtopic is the correct section whether this gets you correct/good/helpfull answers or not.

That's what I was afraid of, and why I asked here first.

I suppose my post will either be deleted or moved to off-topic.
legendary
Activity: 3472
Merit: 4801
February 08, 2016, 05:33:50 PM
#10
- snip -
However, this does not prove that you were the author.
- snip -

I've provided much more detail here:
https://bitcointalksearch.org/topic/m.13819779

Note that I'm not trying to prove that I AM the author.  I think I'm actually trying to prove that I'm NOT the author, or more specifically, that a subset of data that I've provided comes from a larger set that was authored by someone else.

Assuming that I am "Barry" in the following example, here's essentially what I'm trying to do.  I'll lock this thread so any replies can go in my other thread that has all the details.

Alfred supplies large data set to Barry.
Alfred has bad data in his data set, and Barry doesn't know.

Barry breaks up the data into 4 subsets (C, D, E, and F).
He discards set F.
He sends set C to Charlie, D to Dennis, and E to Erin.

Dennis complains that Alfred supplied bad data, and decisions based on that data cost him a LOT of money.
Dennis attempts to recover his losses from Alfred.

Alfred doesn't want to pay, so he claims that the data he gave to Barry was correct.
Alfred accuses Barry of introducing bad data before sending it to Dennis.
Alfred manipulates his data to make it look like his data was correct all along.

Does Barry need to store ALL the data he ever receives from Alfred in perpetuity to prove that it came from Alfred, or is there some cryptographic solution that only requires Barry to store some hashes and signatures?
copper member
Activity: 1498
Merit: 1528
No I dont escrow anymore.
February 08, 2016, 05:28:44 PM
#9
-snip-
be posting in Technical Support, i.e. https://bitcointalk.org/index.php?board=4.0
-snip-

No, thats mainly the section for bitcoin core. Im afraid offtopic is the correct section whether this gets you correct/good/helpfull answers or not.
legendary
Activity: 1039
Merit: 1005
February 08, 2016, 04:58:46 PM
#8
Are you aware of https://www.proofofexistence.com/about?

They provide one half of the solution, which consists of a mechanism to state that a given document existed at a given time.
However, this does not prove that you were the author. In general, authorship is difficult to prove if you don't take care (if someone steals and copies your work and publishes it before you do, how do you prove that you wrote it originally?)
So to prevent that, you need to digitally sign your document and create a proof of existence for the signed document before making it available to anybody else.
Afterwards, someone else may take your document, replace your signature with theirs, and create a similar proof of existence, but they will not be able to create one with an earlier timestamp. When there is a dispute about authorship, you will be able to present your proof of existence.

I don't think I fully understand the bit about data changing hands - do you mean that the data should be passed unchanged between several persons? Then the original proof of existence mechanism works fine.
If you're thinking about collaborative work on the data, the signing and proof publishing steps needs to be performed after each modification, creating a chain of document versions.

Note that for all of this, the document itself does not need to be published, as the proof of existence only applies to a hash of the document, naturally.

Onkel Paul
legendary
Activity: 3472
Merit: 4801
February 08, 2016, 04:53:02 PM
#7
Cryptography being a technical pillar of bitcoin, you should most likely be posting in Technical Support, i.e. https://bitcointalk.org/index.php?board=4.0

Ok, I'll go ahead and post there.  My question isn't in any way related to Bitcoin, but I guess that's as good a place as any.

p.s. I miss D&T presence on this forum too. Feel free to PM me if you need his Linkedin.

I've got it, and I suppose I could try contacting him directly if I don't get the information I'm looking for here.
full member
Activity: 214
Merit: 278
February 08, 2016, 04:46:18 PM
#6
Cryptography being a technical pillar of bitcoin, you should most likely be posting in Technical Support, i.e. https://bitcointalk.org/index.php?board=4.0

p.s. I miss D&T presence on this forum too. Feel free to PM me if you need his Linkedin.
legendary
Activity: 3472
Merit: 4801
February 08, 2016, 04:37:42 PM
#5
DannyHamilton; could you please post a bit more details on what you are thinking? There are probably a lot of people here that could point you in the right direction.

Yes, I absolutely intend to post a LOT more details about what I'm thinking.  I just want to make sure I post it in the correct sub-forum to both get it noticed by people with the right knowledge (I suspect that "Meta" and "Off-topic" aren't the right places for that) AND avoid it being removed for being in the wrong place (I suspect that "Development & Technical Discussion" isn't the right place for that)

I suspect that what I want to do isn't even possible, but since there's a lot about cryptography that I don't fully understand I figured I should check with people more knowledgeable than me before I give up on the idea.

I'm sort of wishing that Tangible Cryptography (a.k.a. Death&Taxes) was still active on the forum, or that G. Maxwel had more time available for casual conversations.
full member
Activity: 172
Merit: 100
February 08, 2016, 03:05:30 PM
#4
No idea about the related forums.But I can certainly can help you with all the material I used for my State Level Cryptography Examination.It includes all the Ciphers [Encryption and Decryption] and MD5/SHA etc in detail.I have a nicely organised set of ebooks/Slides which has all the relevant information on Modern Cryptography .I don't mind sending the folder to you.

No offence, but MD5 and modern don't mix. Wink

DannyHamilton; could you please post a bit more details on what you are thinking? There are probably a lot of people here that could point you in the right direction.
legendary
Activity: 1988
Merit: 1317
Get your game girl
February 08, 2016, 02:47:07 PM
#3
No idea about the related forums.But I can certainly can help you with all the material I used for my State Level Cryptography Examination.It includes all the Ciphers [Encryption and Decryption] and MD5/SHA etc in detail.I have a nicely organised set of ebooks/Slides which has all the relevant information on Modern Cryptography .I don't mind sending the folder to you.
hero member
Activity: 812
Merit: 1000
Act #Neutral,Think y'self as a citizen of Universe
February 08, 2016, 02:38:17 PM
#2
someone point me to a great, helpful, crypto forum where I can post my questions?
There are many great, helpful, crypto forum but that doesn't means they contain skilled/talented programmers or cryptography experts.If you want to seek some really deep info then forums are not that helpful.

However, https://forum.hashkiller.co.uk should answer something & this guy is pretty skilled https://forum.insidepro.com/profile.php?mode=viewprofile&u=11412
legendary
Activity: 3472
Merit: 4801
February 08, 2016, 01:13:35 PM
#1
I've got some questions about how to prove authorship of data that might change hands a few times.  My hope is that there are some crypto solutions for my needs, but I'm not enough of a crypto expert to know what the possibilities are.

If I want to post my questions, is there a "proper" place in bitcointalk.org to ask detailed technical questions about cryptography in general (un-realted to bitcoin), or can someone point me to a great, helpful, crypto forum where I can post my questions?
Jump to: