# Conseil de lecture
**URL:** https://forum.duniter.org/t/conseil-de-lecture/136
**Category:** Français
**Created:** [15 January 2015 08:32 UTC](https://forum.duniter.org/t/conseil-de-lecture/136 "2015-01-15T08:32:50Z")
**Posts on this page:** 17
**Page:** 7
### Author:  [@Inso](https://forum.duniter.org/u/Inso)
#### Post date: [18 April 2016 09:13 UTC](https://forum.duniter.org/t/conseil-de-lecture/136/121 "2016-04-18T09:13:09Z")
The process remains the same. But if you want to hold previous generated pubkeys in the WoT, they will have to sign the new public keys of the second assembly, for the WoT to have a link between the first assembly and the second assembly.
---
### Author:  [@demo](https://forum.duniter.org/u/demo)
#### Post date: [18 April 2016 09:20 UTC](https://forum.duniter.org/t/conseil-de-lecture/136/122 "2016-04-18T09:20:19Z")
> [@Inso](#):
>
> The process remains the same. But if you want to hold previous generated pubkeys in the WoT, they will have to sign the new public keys of the second assembly, for the WoT to have a link between the first assembly and the second assembly.
The second assembly is not the same that repeats itself. Its another assembly in another place.
There is a link of just one observer that he/she was elected (or selected randomly) in order to attend the procedures of the second assembly and verify that it was not a fake one.
It is a WoT of assemblies, not a WoT individuals. We have to consider the whole assembly as an individual in the WoT, but in the same time we have to give Universal Dividant to all members of the assemblies (that of course have not equal number of participants).
So do we need 4 observers to sign all the keys of the second assembly?
Is this a solution?
It is…ok then…thank you
---
### Author:  [@demo](https://forum.duniter.org/u/demo)
#### Post date: [18 April 2016 09:32 UTC](https://forum.duniter.org/t/conseil-de-lecture/136/123 "2016-04-18T09:32:17Z")
> [@Inso](#):
>
> The process remains the same. But if you want to hold previous generated pubkeys in the WoT, they will have to sign the new public keys of the second assembly, for the WoT to have a link between the first assembly and the second assembly.
On the other hand, what you said (signing old public keys when an assembly repeats itself every 4 years) could be a problem when we want anonymity.
---
### Author:  [@Inso](https://forum.duniter.org/u/Inso)
#### Post date: [18 April 2016 09:40 UTC](https://forum.duniter.org/t/conseil-de-lecture/136/124 "2016-04-18T09:40:40Z")
Public keys can expire every 4 years. You can generate new public keys, signed by a more recent assembly.
For example, with **-\>** the process of elected leaders of first assembly checking the generation of the second assembly.
Year 1 : Assembly A (WoT = A)
Year 2 : Assembly A -\> Assembly B (Wot = A+B)
Year 3 : Assembly B -\> Assembly C (Wot = A+B+C)
Year 4 : Assembly C -\> Assembly A’ (Wot = B+C+A’)
---
### Author:  [@demo](https://forum.duniter.org/u/demo)
#### Post date: [19 April 2016 08:55 UTC](https://forum.duniter.org/t/conseil-de-lecture/136/125 "2016-04-19T08:55:45Z")
Yet another question (and I am sorry for asking again questions)
What about anonymity in ucoin?
Are public keys directly associated with the amount of ucoin eachone owns?
Do you take any precautions in favor of anonymity?
There is a protocol in cryptography where you can prove that X is greater than Y, whithout revealing neither X nor Y. Do you know that? Are you using this or something similar, in order to protect the anonymity of rich ucoin owners?
---
### Author:  [@cgeek](https://forum.duniter.org/u/cgeek)
#### Post date: [19 April 2016 09:09 UTC](https://forum.duniter.org/t/conseil-de-lecture/136/126 "2016-04-19T09:09:52Z")
No.
Also, you can use pseudonyms.
---
### Author:  [@demo](https://forum.duniter.org/u/demo)
#### Post date: [19 April 2016 09:17 UTC](https://forum.duniter.org/t/conseil-de-lecture/136/127 "2016-04-19T09:17:55Z")
> [@cgeek](#):
>
> No.
>
> Also, you can use pseudonyms.
So you take no precaution for anonymity?
Pseudonym is not enough.
We have to find a way to encrypt somehow the amount of ucoin each public key owns.
There are special protocols for that (all or nothing disclosure of secrets(ANDOS) e.t.c)
---
### Author:  [@cgeek](https://forum.duniter.org/u/cgeek)
#### Post date: [19 April 2016 09:18 UTC](https://forum.duniter.org/t/conseil-de-lecture/136/128 "2016-04-19T09:18:37Z")
Yes, you will do it. In your own project.
---
### Author:  [@demo](https://forum.duniter.org/u/demo)
#### Post date: [19 April 2016 09:19 UTC](https://forum.duniter.org/t/conseil-de-lecture/136/129 "2016-04-19T09:19:12Z")
> [@cgeek](#):
>
> Yes, you will do it. In your own project.
I dont think I am good enough understanding these protocols.
🙂
But they do exist, so if you can undertand them you can give them a try.
Unless you hate anonymity, and you want to create a ucoin facebook.
---
### Author:  [@cgeek](https://forum.duniter.org/u/cgeek)
#### Post date: [19 April 2016 09:20 UTC](https://forum.duniter.org/t/conseil-de-lecture/136/130 "2016-04-19T09:20:41Z")
What I don’t want is answering trolls.
---
### Author:  [@demo](https://forum.duniter.org/u/demo)
#### Post date: [19 April 2016 09:21 UTC](https://forum.duniter.org/t/conseil-de-lecture/136/131 "2016-04-19T09:21:22Z")
> [@cgeek](#):
>
> What I don’t want is answering trolls.
Its not trolling.
Do you want me to search for references on these protocols?
If you are interested I will.
> **[All-or-Nothing Disclosure of Secrets](https://link.springer.com/chapter/10.1007/3-540-47721-7_17)**
>
> 'All-or-Nothing Disclosure of Secrets' published in 'Advances in Cryptology — CRYPTO’ 86'
I will search also the one where you can prove that X is greater than Y, whithout revealing the numbers.
---
### Author:  [@cgeek](https://forum.duniter.org/u/cgeek)
#### Post date: [19 April 2016 09:25 UTC](https://forum.duniter.org/t/conseil-de-lecture/136/132 "2016-04-19T09:25:42Z")
No I am not interested in such things, because I don’t have the time/energy/skills to make a plane of today. Also, pretending to make a plane of today is for sure the best way to not produce any plane at all.
I am just trying to make something that flies. And if it does not, people interested in making something that flies or flies better will involve themselve to produce such a thing if they want to.
edit: you can still share your links, other people might want them
---
### Author:  [@demo](https://forum.duniter.org/u/demo)
#### Post date: [19 April 2016 09:27 UTC](https://forum.duniter.org/t/conseil-de-lecture/136/133 "2016-04-19T09:27:39Z")
> [@cgeek](#):
>
> edit: you can still share your links, other people might want them
Ok I will search the references and post them here.
I am a big fun of anonymity, because I own a vast amount of ucoins, and I am afraid people will rob me.
😛
---
### Author:  [@cgeek](https://forum.duniter.org/u/cgeek)
#### Post date: [19 April 2016 09:29 UTC](https://forum.duniter.org/t/conseil-de-lecture/136/134 "2016-04-19T09:29:59Z")
It exists 2 types of accounts:
- member accounts
- non-members accounts
Both can handle & transfer money, but ony member accounts can produce it. So if you want anonymity, just use non-member accounts to anonymize your money. You can create them on will.
---
### Author:  [@demo](https://forum.duniter.org/u/demo)
#### Post date: [19 April 2016 11:27 UTC](https://forum.duniter.org/t/conseil-de-lecture/136/135 "2016-04-19T11:27:04Z")
> [@cgeek](#):
>
> It exists 2 types of accounts:
>
> - member accounts
> - non-members accounts
>
> Both can handle & transfer money, but ony member accounts can produce it. So if you want anonymity, just use non-member accounts to anonymize your money. You can create them on will.
All transactions are kept in the blockchain, so I can track down the non-members and find out the member hiding behind.
This is not exactly anonymity. It is obfuscation. It is a good thing of course, but it is not ideal.
---
### Author:  [@cgeek](https://forum.duniter.org/u/cgeek)
#### Post date: [19 April 2016 11:35 UTC](https://forum.duniter.org/t/conseil-de-lecture/136/136 "2016-04-19T11:35:47Z")
Well obfuscation gives a level of anonymity.
You could also pass by an anonymizer (a third party), but this supposes the third party won’t share the information with the one you want to be anonymized _against_.
So the level of obfuscation you will choose depends who you want to be hidden from.
---
### Author:  [@demo](https://forum.duniter.org/u/demo)
#### Post date: [19 April 2016 23:04 UTC](https://forum.duniter.org/t/conseil-de-lecture/136/138 "2016-04-19T23:04:33Z")
> [@demo](#):
>
> > [@cgeek](#):
> >
> > What I don’t want is answering trolls.
>
> Its not trolling.
> Do you want me to search for references on these protocols?
> If you are interested I will.
>
> [All-or-Nothing Disclosure of Secrets | SpringerLink](http://link.springer.com/chapter/10.1007%2F3-540-47721-7_17)
>
> I will search also the one where you can prove that X is greater than Y, whithout revealing the numbers.
ok…as I promise, I put the link here.
the protocol where we can prove X\>Y without revealing neither X nor Y is called **Yao’s Millionaires’ Problem**. Another similar prototcol is when two ucoin millionaires want to determine if their wealth is equal without disclosing any information about their riches and it is named **socialist millionaire problem**.
And a lot of other cryptographic protocols derive from them. Just search about them in a search engine.
From starters, look [Yao's Millionaires' problem - Wikipedia](https://en.wikipedia.org/wiki/Yao's_Millionaires'_Problem) [https://scholar.google.gr/scholar?q=Yao’s+millionaire+problem](https://scholar.google.gr/scholar?q=Yao%27s+millionaire+problem) e.t.c.
And of course there is dashcoin cryptocurrency’s **Darksend** anonymization method, or **CoinJoin** anonymization method that are trying to protect anonymity (but unfortunately they often destroy the history of money transactions, and this is bad for society if society wants to trace illegal actions).
SImilar methods may be used in Ucoin. But of course Ucoin’s job is much more difficult when dealing with anonymity, because Ucoin has also a WoT (expecially if you insist of using a WoT of individuals instead of a WoT of assemblies, anonymity problem becomes harder). Someone, someday, may combine something like the above, somehow, in order to protect both anonymity and the history of money transactions, in Ucoin.
🙂
[Previous page](https://forum.duniter.org/t/conseil-de-lecture/136.md?page=6)