Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
  • Jun 20 2016 02:11
    @alexvandesande banned @algotrader2013
  • Jun 05 2016 10:31
    @alexvandesande banned @adamskee
Rocky Fikki
@rfikki
Was the premise of Ethereum not always based on decentralization? Was not one of the features/design rational of Ethash not geared more towards the use of GPU's rather than ASIC's? Unless we are willing to abandon this premise, it would be wise to stay with the stated design rational of the algorithm and implement ProgPoW. The only question is whether indecision to implement ProgPoW will be usurped by the timing of PoS implementation. https://github.com/ethereum/wiki/wiki/Ethash-Design-Rationale
Jon Stevens
@lookfirst
PoW is years away from going anywhere. ASICs and in some ways, even worse, FGPAs, are here today. Ethash held its own for what... almost 4 years now. Seriously, good job on that one. But it is time to at least start to think about what needs to be done to hold down the fort for the next few years. If it isn't ProgPow, then figure out something else. Which is fine... I don't think anyone here is absolutely married to ProgPoW... it is just the only one on the horizen that has any sort of network effect right now. Most likely, it buys some time. That seems like a good enough reason to at least back it enough to answer that question with some analysis.
Micah Zoltu
@MicahZoltu
Man, I detest the EU so much... are they really wanting to go after car manufacturers because they put the year on the advertisement?
Andrea Lanfranchi
@AndreaLanfranchi
And what about the "Click here to accept cookies" ?
Micah Zoltu
@MicahZoltu
:rage:
Andrea Lanfranchi
@AndreaLanfranchi

are they really wanting to go after car manufacturers because they put the year on the advertisement?

That's not the reason. The reason is that "model year" causes excessive depreciation on used cars on those cases where, say, only turn lights were changed.

Jean Cyr
@jean-m-cyr
The excessive depreciation is not caused by the manufacturer. Rather, it caused by the silly consumers that accord excessive value to owning the latest model year. Can't blame the manufacturer for exploiting that.
Andrea Lanfranchi
@AndreaLanfranchi
That's why EU is looking after that. To protect users who are not "silly" (and have made an investment on the car) not to be damaged by silly users and by manufacturers who have exploited this behavior
Jean Cyr
@jean-m-cyr
Ah well, let's not get into the merits of the 'nanny state'.
Andrea Lanfranchi
@AndreaLanfranchi
There's a completely different mentality in the EU from US or Asia.
Jean Cyr
@jean-m-cyr
That's unfortunate.
Andrea Lanfranchi
@AndreaLanfranchi
Well yes and no. On some aspects the free market is too much regulated.
On other aspects there are way more protections on end-users
Micah Zoltu
@MicahZoltu
EU is definitely the epitome of a nanny state. Telling someone, "you can't put the year the car was made on the label because we have decided that people over-value that information compared to how they should value it."
I'm looking forward to the EU banning trademarks because people over-value them. No more Prada bags because putting the Prada trademark on a bag results in the bag being priced well over its functional value.
Jean Cyr
@jean-m-cyr
Andrea, to your earlier point about the un-silly consumer being hurt by the silly ones: The smart consumer will purchase a heavily depreciated used vehicle and get a much better deal.
Andrea Lanfranchi
@AndreaLanfranchi
Not that easy. Anyway ... come to Europe a little bit and you'll understand how things go here.
There is so much our cultures don't understand about each other.
Greg Colvin
@gcolvin

@MicahZoltu

We should just define Ethereum governance to be a coin toss and call it a day. We can start with ProgPoW: We'll toss a coin, if it is heads then we move forward with ProgPoW, if it is tails we don't.

The core developers discussed these issues in the open for months. We talked with GPU miners, ASIC manufactures, and many more. We considered every angle we could, including any you brought to our attention. We decided to move forward unless technical defects are found in progPow by full consensus--there were no blocking objections. To call our process a coin toss is franky insulting.

The only flaw I see in our process here is that Hudson @Souptacular did not make as clear a statement of that consensus as he could have.

Boris Mann
@bmann
I think Alexey was pointing out that he had misgivings / doesn’t agree BUT isn’t “over my dead body” against it
Rocky Fikki
@rfikki
@gcolvin that was my understanding as well...
Greg Colvin
@gcolvin
Yes, @bmann @rfikki The lack of blocking objections is full consensus. Moving forward despite blocking objections is rough consensus; we try to avoid that.
How the IETF reaches consensus
Micah Zoltu
@MicahZoltu
@gcolvin I wasn't calling the current process a coin toss, I was suggesting that we should replace it with a coin toss.
A coin toss would be a clear governance process that people can understand. The current processs is one where no one really understands how decisions are made. There is a lot of discussion, but it is not clear how decisions are made. The IETF process is a clear process because it is clear how decisions are made (the chair asserts outcome).
Greg Colvin
@gcolvin

A coin toss would be far inferior. The calls get messy sometimes, but the process is that issues are first placed on the agenda for discussion -- see Project Management: Meeting notes and agenda items. Issues are resolved by deliberating until we reach consensus, which might take more than one call, or never happen at all. Hudson the @Souptacular usually calls the consensus. I think we do need to make the consensus more explicit after the fact, as the notes are generally not an accurate transcription.

And once again: Consensus (noun, con-sen'-sus) -- a collective state of mind in which two or more people either agree or consider their disagreements minor enough to live with.

Greg Colvin
@gcolvin
Does that answer your question, @MicahZoltu?
Micah Zoltu
@MicahZoltu

deliberating until we reach consensus

As I have discussed in the past, being able to identify that concensus is reached first depends on defining the set of people over whom consensus is being achieved, and a metric for determining when consensus has been reached.

I would be content with:
  • The set of people over whom consensus is reached are the core developers.
  • The mechanism for identifying consensus being reached is Hudson asserting what the consensus is.
However, that is not the actual set of people that most think we are striving for consensus over, nor does Hudson actually make a call on all issues.
And when Hudson does make a call, many people do not believe that indicates consensus has been reached.
Rocky Fikki
@rfikki
So, it seems there is an outline of the process towards ProgPow here: https://medium.com/ethereum-cat-herders/progpow-audit-goals-expectations-75bb902a1f01
John Tromp
@tromp
odd that the audit will be performed by predominantly software engineers. i would have expected ASIC design expertise to be of prime importance
Micah Zoltu
@MicahZoltu
I think the purpose of the audit isn't to verify that it is "ASIC resistant" but rather to verify that it is cryptographically secure.
I feel like the saying about "don't roll your own crypto" applies here, albeit a little indirectly.
John Tromp
@tromp
it is specifically identified as the 2nd area for review
"ProgPoW meeting the goal of ASIC resistance: Known methods to speed up the calculation of the hash function, length of time it would it take to create a ProgPoW ASIC (if R&D begins immediately), and expected efficiency gains from the first generation of said ASICs."
Micah Zoltu
@MicahZoltu
Ah. Yeah, that certainly seems like it should be done by an expert in the field of ASIC design.
Peter Salanki
@salanki
Tromp: Yeah
John Tromp
@tromp
there would be little point in adopting progpow unless it severely reduces the economic feasibility of ASIC development
so i consider this the most essential area of review
John Tromp
@tromp
i wouldn't mind seeing a reasoned discussion between team ifdefelse and the person who wrote https://ether4life.tumblr.com/post/183604547634/progpow-ic-asic-design-engineer-view, to figure out exactly on what different assumptions their contradictory estimates of gpu-ASIC efficiency gap are based
Jon Stevens
@lookfirst
ifdefelse are working on one.
John Tromp
@tromp
that is excellent news
Jon Stevens
@lookfirst
It has been debunked in a few places. Let alone it is a anonymous tumblr.
Who uses tumblr anymore. ;-)
John Tromp
@tromp
debunking is good
Jon Stevens
@lookfirst
https://www.reddit.com/r/ethereum/comments/b3ogkw/progpow_from_an_ic_design_engineers_view/
John Tromp
@tromp
that is mostly attacking the messenger rather than addressing the arguments
Jon Stevens
@lookfirst
@salanki points are pretty spot on.
John Tromp
@tromp
i agree the author should identify the discrepancies with ifdefelse's reasoning as well, and point out where he thinks their math or assumptions goes astray