Author Topic: Siren Music  (Read 11117 times)

0 Members and 1 Guest are viewing this topic.

Offline toast

  • Hero Member
  • *****
  • Posts: 4001
    • View Profile
  • BitShares: nikolai
Quote
You're peaking my curiosity now. The obvious thing is that they collude with someone buying the lottery ticket and tell them the secret. Why wouldn't they do this? Is each delegate responsible for the funds of his personal lotto? That would be the obvious re-alignment, but you'd have to assume all delegates were interested in running lottos. If delegates have different ownership from shareholders there would be obvious agency problems.

Multiple delegates (say, 51% or even 100%) put in secrets towards a shared hash, so even one honest delegate ensures you cannot predict the result. The worst you can do is refuse to reveal in response a particular winner, but the clients can auto-fire if they detect that.
Do not use this post as information for making any important decisions. The only agreements I ever make are informal and non-binding. Take the same precautions as when dealing with a compromised account, scammer, sockpuppet, etc.

Offline MolonLabe

  • Full Member
  • ***
  • Posts: 58
    • View Profile
Quote from: Bytemaster
There has been significant progress and developments over the past several days.   One of the most significant is the introduction of a random number generation scheme and random delegate selection today.   This was done for two reasons: it resolves a security vulnerability discovered with sequential DPOS and it facilitates one of the more challenging parts of the Lotto DAC.
This is not an explanation. I hope you already know that....

MolonLabe,

Thanks for your input!  I think you bring up some real risks and this project is not without risk, at least from my level of knowledge of it.  I also think you are trying to ask important questions and I appreciate your focus on the tough technical obstacles. 
You're welcome.

That said, to me it seems bytemaster is working on the right problems and I think solving these problems could have a big impact on the world and be VERY valuable if they are solved, don't you?
Of course. The question is "what approach is most likely to actually get the problems solved" hence my concern about the simultaneous risks.

I also think bytemaster might be in a tough position in that as a leader for bitshares he must weigh projecting some confidence rather than saying "hey guys I'm stuck here and have hit a roadblock, is there anyone who knows enough to help me think through this without wasting my time making me explain a million things from scratch that they don't understand."
Sounds awful. I wouldn't do it. I'd say "you invested, go away and let me do my work." I'm so crazy I might even dock people their investment if they continue distracting me.

Have you taken time to read the source code and try to have a deep understanding of implementation?  It's not a criticism either way, I am just curious if you are a programmer etc.
Yes, but one thing that frustrates me, as I mentioned, is that the rules appear to change frequently. It takes a lot of effort to refocus on everything, and I think it indicates some lurking instability in the idea-generating process. Not that I'm a fan of peer review, but usually ideas are passed around to a few people before the work/investment is made, and therefore last minute changes are less frequent.

Well Bitcoin right now is widely deployed, so changing the core block validation algorithm would inevitably cause massive disruption to Bitcoin services and therefore the price as well.
No it wouldn't. Do you really believe that? A carefully staged, thoughtfully tested improvement? No one would care.

Therefore core Bitcoin developers are dis-incentivized from acknowledging that anything else in the cryto space could possibly provide a better solution than what Bitcoin offers. So Im not particularly interested in what core Bitcoin developers have to say.
This sounds like a rationalization. If gmaxwell endorsed dpos as a work of unique genius, you're saying you "wouldn't care"? Instead you'd probably be posting about it everywhere. There is constant disagreement from the Bitcoin devs, they are not a unified anti-Altcoin conspiracy. We already know that Bitcoin works, but feel free to suggest your own experts.

The fact is that mining leads to centralization - everyone agrees that something needs to be done about it.
No they don't (I don't, for example). Even if it did there is P2Pool. There are like 12 mining groups in perfect competition selling totally homogenous hashes. Even if there were only 1 mining pool there wouldn't be centralization. Google may not have competitors, but it faces competition.

On the issue of whether BitAssets will track properly - the underlying theory is that when a BTSX holder sees that each share is trading for $100 on the centralized exchanges, he will act in his own self interest and never demand any less than 100 BitUSD for his shares. He cant ask for any more than that because he knows that other buyers will bid at 100 BitUSD per share. On the short side the exact same psychology happens. Therefore the buyers and shorts will reach consensus at the real price of each share, and the correct amount of BitUSD will be created.

So yes BitUSD is going to track the price of "real" USD.
I am familiar with the arguments against my position (that BitUSD will not track). You, however, are not.

Well here is a reason to care - are you happy with the status quo...only banks are allowed to create dollars...corrupt opaque central bank backed by violence...use the blockchain to give that power back to the people...issue dollars backed by a real trustworthy asset...government lies.
I'm sorry but this is clearly political babble. I was not asking for reasons to care, I was asking if anyone would care. Surely you hold the opinion that people should care more about central banking, but they don't so care. Completely fatuous and embarrassing remark on your part, you've played into AdamBLevine's hand like the fool he predicted you'd be.

It's the delegates that provide the secret.
You're peaking my curiosity now. The obvious thing is that they collude with someone buying the lottery ticket and tell them the secret. Why wouldn't they do this? Is each delegate responsible for the funds of his personal lotto? That would be the obvious re-alignment, but you'd have to assume all delegates were interested in running lottos. If delegates have different ownership from shareholders there would be obvious agency problems.

For convenience sake would it be possible to split your points up in separate threads, because you've listed quite a large number of different topics that might make the discussion more confusing rather than enlightening. Maybe one for each number or main point of critique?
I think that, as you say, the posts have been made. I suppose the thesis of this post is that dpos should be tested separately from BitsharesX, which should run purely on proof of work until the novelties of BitsharesX can be tested. My comments about BitsharesX were just to reinforce how serious the simultaneous-risk problem is.

I might be even more critical about the chances of success for the entire blockchain ecosystem than you are MoLoLabe.
If you can't criticize, you can't optimize. :D

I've run out of time and have to run, so I hope you don't take my not responding to all your points as a token of ill will or any of the other original and very creative interpretations and assumptions about all members of the forums.
Nope. Again I would prefer the problems be avoided rather than addressed, through smaller projects.

Offline JoeyD

MoLonLabe not only am I worried about the points you listed, but I could even add a whole lot more, so I personally welcome the discussion of your points. Quite a few of your points have come up several times, but often on unrelated topics and have gotten buried since.

For convenience sake would it be possible to split your points up in separate threads, because you've listed quite a large number of different topics that might make the discussion more confusing rather than enlightening. Maybe one for each number or main point of critique?

First off, I have a bit of a different take on point 3. I do not agree that bitcoin was written by a genius programmer (doesn't mean he wasn't a genius in many other areas) and fixing that apparently takes quite the army of genius coders and a lot of time and effort. This sort of makes the case for the need to at least try to do the coding right from the start. As for big name supervision, I was under the impression that the person who has discovered most of bitcoins fundamental bugs was also asked to vet the bitshares code.

One of the biggest issues in my point of view (second only to the quality of the code produced by the invictus-team) is that the community- and informationmanagement has not been and still isn't at the level it should be, neither on this forum nor on the website. I don't know all the details why that is, but I do know that it isn't being done as well as other open-dev communities are doing it, like for example blender.org, raspberrypi.org etcetera.

Instead of trying to sell a consumer-product, I think the focus should be on the community first and foremost and I do believe a part of that responsibility lies with Invictus. I personally don´t care about any fancy website, but would like to see something like a blog regularly updated with the more elaborate updates and explanations, a central stickied post on the forum with a todolist, listing what has been done and what needs to be done and things that would be nice to be able to do. That thread with the todolist could maybe also list why some choices had to be scrapped or changed and why delays have occured.

Just those two central points of reference would make finding correct information a whole lot easier imho and maybe help foster more discussion like this thread. Having no real central easy to browse point of information and history also leads to confusion and rumors like the one about the "choice for dpos", because now it looks like Invictus just randomly choose to go with dpos over pow which is not what happened.

Btw I myself don't see pow as any less experimental as (D)PoS. To me pow is not a proven concept at all and also not the guaranteed solution to all things as I have seen some people suggest. Hell I doubt even Dr. A. Back is that much of a believer in it. But then again I might be even more critical about the chances of success for the entire blockchain ecosystem than you are MoLoLabe.

I've run out of time and have to run, so I hope you don't take my not responding to all your points as a token of ill will or any of the other original and very creative interpretations and assumptions about all members of the forums. I'm just one individual and I do not represent the entirety of the bitshares-community if such a thing is even possible.
« Last Edit: May 23, 2014, 06:07:37 pm by JoeyD »

Offline Overthetop

MolonLabe,

Thanks for your input!  I think you bring up some real risks and this project is not without risk, at least from my level of knowledge of it.  I also think you are trying to ask important questions and I appreciate your focus on the tough technical obstacles. 

I think the more people we have that are thinking deeply about these things the better. The more people that take the time learn enough to really get on the same page and see the obstacles clearly the better (this is very important, but very difficult for many).  Perhaps a lot of people take a leap of faith and trust bytemaster rather than asking tons of questions to try to get up to speed, because they know he only has so many hours in a day.

That said, to me it seems bytemaster is working on the right problems and I think solving these problems could have a big impact on the world and be VERY valuable if they are solved, don't you?

I also think bytemaster might be in a tough position in that as a leader for bitshares he must weigh projecting some confidence rather than saying "hey guys I'm stuck here and have hit a roadblock, is there anyone who knows enough to help me think through this without wasting my time making me explain a million things from scratch that they don't understand."

Have you taken time to read the source code and try to have a deep understanding of implementation?  It's not a criticism either way, I am just curious if you are a programmer etc.

I don't think people on the forum hate criticism in principle, it's just people don't always agree on what is a valid criticism; not all criticism is well thought out.  Also, a lot of people don't feel like they have a good enough understanding to fully evaluate all the more technical criticism.
+5%
个人微博账号: Overthetop_万里晴空
“块链创新与创业”交流群: 330378613

Offline toast

  • Hero Member
  • *****
  • Posts: 4001
    • View Profile
  • BitShares: nikolai
I think you'll find critique is not welcome from the community here, and providing any will make you the enemy.

So enjoy that.

Come on man, not following your advice is not the same as ignoring it. I for one always stop to read what you say when I'm skimming any thread. Nobody perceives you as the enemy.

I wish I could say I were more familiar. When I see, for example, "Charity" I usually just don't even open the link.

p2p appears to actually solve a problem related to distributed trust, and already exists, so sure. Wouldn't lotto require random number generation? Is there a reader's digest version of who checked the technical details and incentive structure of that? I'm afraid I just decided it wasn't worth looking into. Can you convince me that it is?

Moreover, and to the point, will they use proof of work or dpos?

I think the lotto random number works something like:
You buy a lottery ticket and submit with your lottery ticket the hash of a secret, then when drawing the lottery everyone reveals their secret and a hash is taken of all of them and that hash is the random number that determines who wins your lottery.  No one else knows your secret and no one can lie about what their secret was after the fact.


It's the delegates that provide the secret.

I'm afraid I just decided it wasn't worth looking into. Can you convince me that it is?
No, tl;dr is "delegate secret/reveal".
Quote
Moreover, and to the point, will they use proof of work or dpos?
DPOS. I understand your anxiety here.

Do not use this post as information for making any important decisions. The only agreements I ever make are informal and non-binding. Take the same precautions as when dealing with a compromised account, scammer, sockpuppet, etc.

Offline AdamBLevine

  • Sr. Member
  • ****
  • Posts: 492
    • View Profile
    • Let's Talk Bitcoin!
I think you'll find critique is not welcome from the community here, and providing any will make you the enemy.

So enjoy that.

Polite/technical critiques are highly valued.
Crude/ad-hominum critiques, not so much.

 :)

Is the OP Crude/ad-hominum?
Email me at adam@letstalkbitcoin.com

Offline Agent86

  • Sr. Member
  • ****
  • Posts: 471
  • BTSX: agent86
    • View Profile
I wish I could say I were more familiar. When I see, for example, "Charity" I usually just don't even open the link.

p2p appears to actually solve a problem related to distributed trust, and already exists, so sure. Wouldn't lotto require random number generation? Is there a reader's digest version of who checked the technical details and incentive structure of that? I'm afraid I just decided it wasn't worth looking into. Can you convince me that it is?

Moreover, and to the point, will they use proof of work or dpos?

I think the lotto random number works something like:
You buy a lottery ticket and submit with your lottery ticket the hash of a secret, then when drawing the lottery everyone reveals their secret and a hash is taken of all of them and that hash is the random number that determines who wins your lottery.  No one else knows your secret and no one can lie about what their secret was after the fact.

Offline Agent86

  • Sr. Member
  • ****
  • Posts: 471
  • BTSX: agent86
    • View Profile
MolonLabe,

Thanks for your input!  I think you bring up some real risks and this project is not without risk, at least from my level of knowledge of it.  I also think you are trying to ask important questions and I appreciate your focus on the tough technical obstacles. 

I think the more people we have that are thinking deeply about these things the better. The more people that take the time learn enough to really get on the same page and see the obstacles clearly the better (this is very important, but very difficult for many).  Perhaps a lot of people take a leap of faith and trust bytemaster rather than asking tons of questions to try to get up to speed, because they know he only has so many hours in a day.

That said, to me it seems bytemaster is working on the right problems and I think solving these problems could have a big impact on the world and be VERY valuable if they are solved, don't you?

I also think bytemaster might be in a tough position in that as a leader for bitshares he must weigh projecting some confidence rather than saying "hey guys I'm stuck here and have hit a roadblock, is there anyone who knows enough to help me think through this without wasting my time making me explain a million things from scratch that they don't understand."

Have you taken time to read the source code and try to have a deep understanding of implementation?  It's not a criticism either way, I am just curious if you are a programmer etc.

I don't think people on the forum hate criticism in principle, it's just people don't always agree on what is a valid criticism; not all criticism is well thought out.  Also, a lot of people don't feel like they have a good enough understanding to fully evaluate all the more technical criticism.

Offline Stan

  • Hero Member
  • *****
  • Posts: 2908
  • You need to think BIGGER, Pinky...
    • View Profile
    • Cryptonomex
  • BitShares: Stan
I think you'll find critique is not welcome from the community here, and providing any will make you the enemy.

So enjoy that.

Polite/technical critiques are highly valued.
Crude/ad-hominum critiques, not so much.

 :)
Anything said on these forums does not constitute an intent to create a legal obligation or contract of any kind.   These are merely my opinions which I reserve the right to change at any time.

Offline liondani

  • Hero Member
  • *****
  • Posts: 3737
  • Inch by inch, play by play
    • View Profile
    • My detailed info
  • BitShares: liondani
  • GitHub: liondani
Wouldn't lotto require random number generation? Is there a reader's digest version of who checked the technical details and incentive structure of that? I'm afraid I just decided it wasn't worth looking into. Can you convince me that it is?
Moreover, and to the point, will they use proof of work or dpos?

Quote from: Bytemaster
There has been significant progress and developments over the past several days.   One of the most significant is the introduction of a random number generation scheme and random delegate selection today.   This was done for two reasons: it resolves a security vulnerability discovered with sequential DPOS and it facilitates one of the more challenging parts of the Lotto DAC.

We have implemented UPNP networking.


Offline MolonLabe

  • Full Member
  • ***
  • Posts: 58
    • View Profile
I wish I could say I were more familiar. When I see, for example, "Charity" I usually just don't even open the link.

p2p appears to actually solve a problem related to distributed trust, and already exists, so sure. Wouldn't lotto require random number generation? Is there a reader's digest version of who checked the technical details and incentive structure of that? I'm afraid I just decided it wasn't worth looking into. Can you convince me that it is?

Moreover, and to the point, will they use proof of work or dpos?
« Last Edit: May 23, 2014, 02:11:08 pm by MolonLabe »

Offline toast

  • Hero Member
  • *****
  • Posts: 4001
    • View Profile
  • BitShares: nikolai
A smaller project, like .p2p or lotto?

Sent from my SCH-I535 using Tapatalk

Do not use this post as information for making any important decisions. The only agreements I ever make are informal and non-binding. Take the same precautions as when dealing with a compromised account, scammer, sockpuppet, etc.

Offline MolonLabe

  • Full Member
  • ***
  • Posts: 58
    • View Profile
 >)

I thrive on enemies.

Offline AdamBLevine

  • Sr. Member
  • ****
  • Posts: 492
    • View Profile
    • Let's Talk Bitcoin!
I think you'll find critique is not welcome from the community here, and providing any will make you the enemy.

So enjoy that.
Email me at adam@letstalkbitcoin.com

Offline MolonLabe

  • Full Member
  • ***
  • Posts: 58
    • View Profile
This project is already asking people to accept many separate significant sources of uncertainty at once:

1] The "economic experiment" (that BitAssets will track real assets, which I personally don't believe in at all).
2] The economic model of this experiment (which I also don't believe in, as shorts have asymmetric fragility and inferior optionality).
3] Software/technical bugs (which have interfered even with Bitcoin [written by many programming geniuses] several times over its history).
4] Market response (Will a critical mass of merchants care bout this? Will a critical mass of traders use this?).
5] Other (legal response, competition, forking the project, theft/poaching/organizational decay).

However, on top of that, you are adding [6] a completely new consensus mechanism, "delegated proof of stake", which is less than 2 months old? Have any big names in cryptocoin security, eg Gmaxwell or andytoshi, endorsed this mechanism? Have any third-parties (ie, not positive bias, cognitive-dissonance fueled I3 investors) endorsed the mechanism? Have any notable people commented on it at all?

Worse, still, will there be Something New two months from now? Something that is even less understood and tested?

Why doesn't anyone else care about this?

I am not an investor in Bitshares, but I would like to see something in Bitcoin 2.0 space succeed. Surely it would be wiser to stick to an idea which already exists, yet is struggling to maintain trust or vitality (like e-gold), and start small by blockchaining it in as similar a way as possible. Take an existing service, like Dropbox (not like insurance or lending, which require complex and novel customer service and human input, nor something like Music-distribution which really isn't an industry at all anymore), and just copy it by changing as few things about Bitcoin as possible. Even this is likely to be impossibly risky and technical.

I'm sorry to say that it is my opinion that a great deal of money and talent will go to waste here. I can't imagine this project succeeding (ie running for 1 year without significant economic or technical problems), without more than a year of testing.

Pick a smaller project!