BitShares Forum
Main => General Discussion => Topic started by: gamey on April 27, 2014, 08:11:40 pm
-
I don't really get this. I see Peter Todd talking about it on twitter pointing to this posting.
https://bitcointalk.org/index.php?topic=586832.0
(https://bitcointalk.org/index.php?topic=586832.0)
If anyone else has other links on the discussion of this approach, please post them here. Thoughts ?
-
Ok it hurts my ego to get no responses. I find this interesting but do not quite fully grasp it. The developer seems to know what he is talking about here.
Lets Talk Bitcoin ep 104.
http://letstalkbitcoin.com/ltb104-tree-chains-with-peter-todd/ (http://letstalkbitcoin.com/ltb104-tree-chains-with-peter-todd/)
Starting minute 20 you get to hear basically the same argument as bytemaster's on why sidechains aren't so great.
At about minute 40 (Part 2) they explain treechains. It is pretty complicated. I'd find something written preferable as I learn better from reading, but this is the best explanation I've found so far. If anyone finds a better resource please post it. I would like to hear bytemaster's thoughts on this. I myself need to relisten to parts of the podcast to take it in.
-
Thanks for sharing. I just go through the primary summary, true that it is complicated and need more time to understand...
http://www.mail-archive.com/bitcoin-development@lists.sourceforge.net/msg04388.html (http://www.mail-archive.com/bitcoin-development@lists.sourceforge.net/msg04388.html) .
But the rest posts and debate in this mail list thread is , eh.., embarrassed?
-
Peter Todd responded once on my comment on the LTB-podcast, but I've not gotten a single reaction from him since. He did have something to say about PoS and that he thought that could never work. I've tried asking him to elaborate on why his concept is less centralized than standard merge mining or side-chains in general. I even tried challenging other commenters to try and eek out some more insights, but most of them seem to just echo what they heard in the interview and take those remarks at face value, without being able to explain the details as to why those statements would be valid.
My brain is not firing on all cylinders yet due to a tenacious flue so I've not been able to work through the mailing-list article yet.
But the rest posts and debate in this mail list thread is , eh.., embarrassed?
Indeed. I was trying to get more insight from people reacting to his proposal, but instead the discussion seemed to focus around Peter Todds personality and how he was scaring away would be developers and new ideas.
-
from bitcointalk:
With tree-chains we create a global, yet scalable, blockchain publishing service that everyone contributes to and everyone uses. Everyone works together to keep that layer secure, and anyone can write code to make use of it. The disadvantage is that you can't rely on miners to verify for you - you have to verify blocks client-side - but that also means those same miners can't attack you without attacking all blockchains at once.
search "herding the child-chains" - it appears that "global and scalable" means "still sha256 ASIC dominated", whereas sidechains have no blockchain security constraints
http://www.mail-archive.com/bitcoin-development@lists.sourceforge.net/msg04388.html
Otherwise very interesting and this research will likely benefit everyone. I still need to digest it before I can talk any more
-
That sounds like a forced merge-mine network that only does POW time-stamping and no validation of transactions. I did not know validation was not a required component before doing the PoW on top, up till now I figured that the whole point of why bitcoin needed PoW was to secure the validation. Now I'm really curious how P. Todd figures that this would solve mining centralization and how miner-rewards work in this system.
Also if validation is not done by the miners, what happens to the invalid transactions? Are those simply included and hashed into the blockchain as invalid/dead-ends and ignored/skipped over by all participants? Wouldn't there be some serious issues with scale, complexity and bandwidth in that case?