Return-Path: Received: from smtp1.linuxfoundation.org (smtp1.linux-foundation.org [172.17.192.35]) by mail.linuxfoundation.org (Postfix) with ESMTPS id B8B3CAD2 for ; Fri, 13 Oct 2017 13:57:37 +0000 (UTC) X-Greylist: whitelisted by SQLgrey-1.7.6 Received: from mail-wm0-f68.google.com (mail-wm0-f68.google.com [74.125.82.68]) by smtp1.linuxfoundation.org (Postfix) with ESMTPS id 087B54C7 for ; Fri, 13 Oct 2017 13:57:35 +0000 (UTC) Received: by mail-wm0-f68.google.com with SMTP id b189so21558591wmd.4 for ; Fri, 13 Oct 2017 06:57:35 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :content-transfer-encoding; bh=M9JYxHUROVTTtjyv9ZQ8N5qrqVn2DJNHxyZtWkyGKMM=; b=d9meGJuKHUrhS8HQaL5maXIM8KIvXknEzRBLjy/UIH0d9Pc6D8iiWm5TzJPZFjwjBT LEGtm2VpPmOGaTrlvtKUxuiiSCKadDPZbrMXniHF43EDZw2yeUbXh2iHBEc0UEweFdLV qkP/jSRhDdt8O2Sr2cyzGrzwoWAguu215GLTmfSSwODlrKntt3cKmLEd6X4IKJ0AMkNT 6XkgHuX+bgYb4GWTQajDcqVRPhXJBOD+6qv3k/JZW92KYgqQStUdqk8gvCA0MGLGgvTV rhZSi6WhKw8Juh0I0faCtN1i0RZEZjG6T1DQ+/zpzBR2Dw5upP5npwItUgP3hq4lq1zs IPHQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:in-reply-to:references:from:date :message-id:subject:to:content-transfer-encoding; bh=M9JYxHUROVTTtjyv9ZQ8N5qrqVn2DJNHxyZtWkyGKMM=; b=Scxx3vOyyNpMaweZJnluAz9kN81Qvf8RilTwKwGeq7P30+q4zs0qU3DfNlhLScFpgz ubHxgu0hD+n3HxZKST4KNQ5dK92Bc5wYQGbACEzVFwSUDFBXwCJdvY9rJnX7EUtoifnB ka4q54LbA8wqaS5DskWKf2EozkejfHoMEbHk3IwADYrbLKsVKaGtdxO3JCBP86sQ1nd+ IvfisAGze3Lr2EB/FPOzTBR5+jsWES+Fb4iqBAM0dITgpdg0M0Mk2rMrcBdQddcAO4Vj amz2X7JLEtH9AsoeegnAtkbgN/Rltc6RiaRZLJJlqIt1tfscHgsSGW4DREWllvL8Y//7 /4Qw== X-Gm-Message-State: AMCzsaVz6tu+Ttxj39Unwv/o82tRWHaUy0Ckfad2dWctY7Yp2roxvQXv Q5W0OVtjZKEMAiO8o2smJ99U1SAH2Hb5dI0qiJNfoA== X-Google-Smtp-Source: AOwi7QBBbnwN4xhC7GQnI47CcfmYGYM6eLwE11ymjjoFKhVHtUVkINDVdFupu2jXK4i/JhcJBAIvdoi/6NfEFfiLl+Q= X-Received: by 10.80.153.77 with SMTP id l13mr2276543edb.120.1507903054326; Fri, 13 Oct 2017 06:57:34 -0700 (PDT) MIME-Version: 1.0 Received: by 10.80.173.10 with HTTP; Fri, 13 Oct 2017 06:57:33 -0700 (PDT) In-Reply-To: References: From: Scott Roberts Date: Fri, 13 Oct 2017 09:57:33 -0400 Message-ID: To: Ilan Oh , Bitcoin Protocol Discussion Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Spam-Status: No, score=0.4 required=5.0 tests=DKIM_SIGNED,DKIM_VALID, DKIM_VALID_AU,FREEMAIL_FROM,LOTS_OF_MONEY,RCVD_IN_DNSWL_NONE, RCVD_IN_SORBS_SPAM,T_MONEY_PERCENT autolearn=disabled version=3.3.1 X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on smtp1.linux-foundation.org X-Mailman-Approved-At: Fri, 13 Oct 2017 13:58:10 +0000 Subject: Re: [bitcoin-dev] bitcoin-dev Digest, Vol 29, Issue 21 X-BeenThere: bitcoin-dev@lists.linuxfoundation.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: Bitcoin Protocol Discussion List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 13 Oct 2017 13:57:37 -0000 > ZmnSCPxj wrote >> Mining infrastructure follows price. Ilansky wrote: > In the case of bitcoin, it is the price that follows mining infrastructur= es. I generally agree with ZmnSCPxj that good ideas =3D> good devs =3D> hodlers =3D> price =3D> mining Except that each step is not an absolute, and can be biased by things like miners who seek profit via fees and other means that are not good for everyone else. Llansky's belief itself influences price away from the ideal. Marketing "easy profits for hodlers!" and first-to-market monopoly are other elements that influence price and thereby guide mining away from good ideas (like a constant value currency). Then price pulls in good devs that pulls in more mining. So it can snowball into a monster. We need not debate cause and effect since it's distant from the list's goals. The relevance to me is that the biases away from ZmnSCPxj's ideal are a reason a more responsive difficulty is needed. Mining is for determining truth of the blockchain, not to make sure there is only 1 blockchain. ZmnSCPxj indicates we should not do anything that has more precision or speed in determining the correct difficulty if it reduces Bitcoin's ability to be a monopoly. Not coincidentally, the monopoly helps ensure hodlers become the new 1%. A fork clone that uses the faster difficulty would attack BTC's slow difficulty if it achieves a comparable price. All other things being equal, it would lower BTC's value until it forks to fix the difficulty. On Fri, Oct 13, 2017 at 8:27 AM, Ilan Oh via bitcoin-dev wrote: >> Mining infrastructure follows price. If bitcoins were still trading at = 1 >> USD per coin, nobody will build mining infrastructure to the same level = as >> today, with 5000 USD per coin. > > In the case of bitcoin, it is the price that follows mining infrastructur= es. > The price is at 5000 because it is difficult to mine bitcoin not the othe= r > way around, like you mention it. Even with a fixed demand, price would go= up > as difficulty grow, the supply guide the market. There is a strong incent= ive > to mine blindly as it is difficult to estimate for a miner where is the > actual demand, with a start up currency without actual economic support. > Indeed at the genesis of this "mining-price" cycle the incentive was to > contribute to a network and create ones own supply, and not respond to a > demand. > > Ilansky > > Le 13 oct. 2017 13:55, a > =C3=A9crit : >> >> Send bitcoin-dev mailing list submissions to >> bitcoin-dev@lists.linuxfoundation.org >> >> To subscribe or unsubscribe via the World Wide Web, visit >> https://lists.linuxfoundation.org/mailman/listinfo/bitcoin-dev >> or, via email, send a message with subject or body 'help' to >> bitcoin-dev-request@lists.linuxfoundation.org >> >> You can reach the person managing the list at >> bitcoin-dev-owner@lists.linuxfoundation.org >> >> When replying, please edit your Subject line so it is more specific >> than "Re: Contents of bitcoin-dev digest..." >> >> >> Today's Topics: >> >> 1. Re: New difficulty algorithm part 2 (ZmnSCPxj) >> 2. Re: New difficulty algorithm part 2 (Scott Roberts) >> >> >> ---------------------------------------------------------------------- >> >> Message: 1 >> Date: Fri, 13 Oct 2017 00:45:33 -0400 >> From: ZmnSCPxj >> To: Scott Roberts >> Cc: "bitcoin-dev@lists.linuxfoundation.org" >> >> Subject: Re: [bitcoin-dev] New difficulty algorithm part 2 >> Message-ID: >> >> >> >> Content-Type: text/plain; charset=3D"utf-8" >> >> Good morning, >> >> >ZmnSCPxj wrote: >> >> Thus even if the unwanted chain provides 2 tokens as fee per block, >> >> whereas the wanted chain provides 1 token as fee per block, if the >> >> unwanted chain tokens are valued at 1/4 the wanted chain tokens, mine= rs >> >> will still prefer the wanted chain regardless. >> > >> >This is a good point I was not thinking about, but your math assumes >> >1/2 price for a coin that can do 2x more transactions. Holders like >> >Roger Ver have an interest in low price and more transactions. A coin >> >with 2x more transactions, 22% lower price, and 22% lower fees per >> >coin transferred will attract more merchants, customers, and miners >> >(they get 50% more total fees) and this will in turn attract more >> >hodlers and devs. This assumes it outweighs hodler security concerns. >> >Merchants and customers, to the extent they are not long term hodlers, >> >are not interested in price as much as stability, so they are somewhat >> >at odds with hodlers. >> >> As of this moment, BT1 / BT2 price ratio in BitFinex is slightly higher >> than 7 : 1. Twice the transaction rate cannot overcome this price ratio >> difference. Even if you were to claim that the BitFinex data is off by = a >> factor of 3, twice the transaction rate still cannot overcome the price >> ratio difference. Do you have stronger data than what is available on >> BitFinex? If not, your assumptions are incorrect and all conclusions >> suspect. >> >> >Bitcoin consensus truth is based on "might is right". Buyers and >> >sellers of goods and services ("users") can shift some might to miners >> >via fees, to the chagrin of hodlers who have more interest in security >> >and price increases. Some hodlers think meeting user needs is the >> >source of long term value. Others think mining infrastructure is. >> >> Mining infrastructure follows price. If bitcoins were still trading at = 1 >> USD per coin, nobody will build mining infrastructure to the same level = as >> today, with 5000 USD per coin. >> >> Price will follow user needs, i.e. demand. >> >> >You >> >seem to require hodlers to correctly identify and rely solely on good >> >developers. >> >> For the very specific case of 2X, it is very easy to make this >> identification. Even without understanding the work being done, one can >> reasonably say that it is far more likely that a loose group of 100 or m= ore >> developers will contain a few good or excellent developers, than a group= of >> a few developers containing a similar number of good or excellent >> developers. >> >> User needs will get met only on the chain that good developers work on. >> Bitcoin today has too many limitations: viruses on Windows can steal all >> your money, fee estimates consistently overestimate, fees rise during >> spamming attacks, easy to lose psuedonymity, tiny UTXOs are infeasible t= o >> spend, cannot support dozens of thousands of transactions per second. >> Rationally, long-term hodlers will select a chain with better developers= who >> are more likely to discover or innovate methods to reduce, eliminate, or >> sidestep those limitations. Perhaps the balance will change in the futu= re, >> but it is certainly not the balance now, and thus any difficulty algorit= hm >> change in response to the current situation will be premature, and far m= ore >> likely to cause disaster than avert one. >> >> >Whatever combination of these is the case, bad money can >> >still drive out good, especially if the market determination is not >> >efficient. >> > >> >A faster measurement of hashrate for difficulty enables the economic >> >determination to be more efficient and correct. It prevents the >> >biggest coin from bullying forks that have better ideas. Conversely, >> >it prevents miners from switching to an inferior coin simply because >> >it provides them with more "protection money" from fees that enables >> >them to bully Bitcoin Core out of existence, even in the presence of a >> >slightly larger hodler support. >> >> This requires that all chains follow the same difficulty adjustment: aft= er >> all, it is also entirely the possibility that 2X will be the lower-hashr= ate >> coin in a few months, with the Core chain bullying them out of existence= . >> Perhaps you should cross-post your analysis to bitcoin-segwit2x also. A= fter >> all, the 2X developers should also want to have faster price discovery o= f >> the true price of 2X, away from the unfavorable (incorrect?) pricing on >> BitFinex. >> >> >Devs are a governing authority under the influence of users, hodlers, >> >and miners. Miners are like banks lobbying government for higher total >> >fees. Hodlers are the new 1%, holding 90% of the coin, lobbying both >> >devs and users for security, but equally interested in price >> >increases. Users are "the people" that devs need to protect against >> >both hodlers and miners. They do not care about price as long as it is >> >stable. They do not want to become the 99% owning 10% of the coin or >> >have to pay unecessary fees merely for their coin to be the biggest >> >bully on the block. A faster responding difficulty will take a lot of >> >hot air out of the bully. It prevents miners from being able to >> >dictate that only coins with high fees are allowed. They are less >> >able to destroy small coins that have a fast defense. >> > >> >The 1% and banks would starve the people that feed them to death if >> >they were allowed complete control of the government. Are hodlers and >> >miners any wiser? >> >> Are developers any wiser, either? >> >> Then consider this wisdom: The fewer back-incompatible changes to a coin= , >> the better. Hardforks of any kind are an invitation to disaster and, at >> this point, require massive coordination effort which cannot be feasibly >> done within a month. Fast market determination can be done using off-ch= ain >> methods (such as on-exchange trades), and are generally robust against >> temporary problems on-chain, although admittedly there is a counterparty >> risk involved. The coin works, and in general there is usually very lit= tle >> need to fix it, especially using dangerous hardforks. >> >> >Devs need to strive for an expansion of the coin >> >quantity to keep value constant which is the foundation of the 5 >> >characteristics of an ideal currency. >> >> Is that your goal? This is a massive departure from the conception of >> Bitcoin as having a fixed limit and effectively becoming deflationary. = It >> will also lead to massive economic distortions in favor of those who rec= eive >> newly-minted coins. I doubt any developer would want to have this prope= rty. >> >> Regards, >> ZmnSCPxj >> -------------- next part -------------- >> An HTML attachment was scrubbed... >> URL: >> >> >> ------------------------------ >> >> Message: 2 >> Date: Fri, 13 Oct 2017 07:35:09 -0400 >> From: Scott Roberts >> To: ZmnSCPxj >> Cc: "bitcoin-dev@lists.linuxfoundation.org" >> >> Subject: Re: [bitcoin-dev] New difficulty algorithm part 2 >> Message-ID: >> >> >> Content-Type: text/plain; charset=3D"UTF-8" >> >> Yes, the current price ratio indicates there is no need for a new >> difficulty algorithm. I do not desire to fork before a disaster, or to >> otherwise employ a new difficulty before a fork is otherwise needed. >> >> A 2-week delay in difficulty response is a 2 week error in >> measurement. Slow response generally means less intelligence. >> >> My goal is not to have a bunch of BTC clones that merchants and buyers >> use equally, but to have a better difficulty algorithm in place to be >> used in the next BTC "Core" fork. If not for the current situation, >> then for future security. >> >> > This is a massive departure from the conception of Bitcoin as having = a >> > fixed limit and effectively becoming deflationary. >> >> You mean multiple forks is inflationary. The current limit in quantity >> is deflationary because the use of the coin is rising faster than its >> mining is producing (see velocity of money). Constant value is defined >> as being neither. Bitcoin's deflationary quality created a massive >> marketing advantage as well as paid the creator about million dollars >> an hour. If it suddenly were able to be a constant value coin, its use >> in the marketplace and as a real store of value would skyrocket and >> the cries of "Ponzi scheme" would stop. The trick is in determining >> constant value without a 3rd party such as an index of a basket of >> commodities (which both Keynes and von Mises wanted, but was scuttled >> by the U.S. at Bretton Woods). >> >> On Fri, Oct 13, 2017 at 12:45 AM, ZmnSCPxj >> wrote: >> > Good morning, >> > >> > >> >>ZmnSCPxj wrote: >> >>> Thus even if the unwanted chain provides 2 tokens as fee per block, >> >>> whereas the wanted chain provides 1 token as fee per block, if the >> >>> unwanted chain tokens are valued at 1/4 the wanted chain tokens, >> >>> miners >> >>> will still prefer the wanted chain regardless. >> >> >> >>This is a good point I was not thinking about, but your math assumes >> >>1/2 price for a coin that can do 2x more transactions. Holders like >> >>Roger Ver have an interest in low price and more transactions. A coin >> >>with 2x more transactions, 22% lower price, and 22% lower fees per >> >>coin transferred will attract more merchants, customers, and miners >> >>(they get 50% more total fees) and this will in turn attract more >> >>hodlers and devs. This assumes it outweighs hodler security concerns. >> >>Merchants and customers, to the extent they are not long term hodlers, >> >>are not interested in price as much as stability, so they are somewhat >> >>at odds with hodlers. >> > >> > As of this moment, BT1 / BT2 price ratio in BitFinex is slightly highe= r >> > than >> > 7 : 1. Twice the transaction rate cannot overcome this price ratio >> > difference. Even if you were to claim that the BitFinex data is off b= y >> > a >> > factor of 3, twice the transaction rate still cannot overcome the pric= e >> > ratio difference. Do you have stronger data than what is available on >> > BitFinex? If not, your assumptions are incorrect and all conclusions >> > suspect. >> > >> > >> >>Bitcoin consensus truth is based on "might is right". Buyers and >> >>sellers of goods and services ("users") can shift some might to miners >> >>via fees, to the chagrin of hodlers who have more interest in security >> >>and price increases. Some hodlers think meeting user needs is the >> >>source of long term value. Others think mining infrastructure is. >> > >> > Mining infrastructure follows price. If bitcoins were still trading a= t >> > 1 >> > USD per coin, nobody will build mining infrastructure to the same leve= l >> > as >> > today, with 5000 USD per coin. >> > >> > Price will follow user needs, i.e. demand. >> > >> >>You >> >>seem to require hodlers to correctly identify and rely solely on good >> >>developers. >> > >> > For the very specific case of 2X, it is very easy to make this >> > identification. Even without understanding the work being done, one c= an >> > reasonably say that it is far more likely that a loose group of 100 or >> > more >> > developers will contain a few good or excellent developers, than a gro= up >> > of >> > a few developers containing a similar number of good or excellent >> > developers. >> > >> > User needs will get met only on the chain that good developers work on= . >> > Bitcoin today has too many limitations: viruses on Windows can steal a= ll >> > your money, fee estimates consistently overestimate, fees rise during >> > spamming attacks, easy to lose psuedonymity, tiny UTXOs are infeasible >> > to >> > spend, cannot support dozens of thousands of transactions per second. >> > Rationally, long-term hodlers will select a chain with better develope= rs >> > who >> > are more likely to discover or innovate methods to reduce, eliminate, = or >> > sidestep those limitations. Perhaps the balance will change in the >> > future, >> > but it is certainly not the balance now, and thus any difficulty >> > algorithm >> > change in response to the current situation will be premature, and far >> > more >> > likely to cause disaster than avert one. >> > >> >>Whatever combination of these is the case, bad money can >> >>still drive out good, especially if the market determination is not >> >>efficient. >> >> >> >>A faster measurement of hashrate for difficulty enables the economic >> >>determination to be more efficient and correct. It prevents the >> >>biggest coin from bullying forks that have better ideas. Conversely, >> >>it prevents miners from switching to an inferior coin simply because >> >>it provides them with more "protection money" from fees that enables >> >>them to bully Bitcoin Core out of existence, even in the presence of a >> >>slightly larger hodler support. >> > >> > This requires that all chains follow the same difficulty adjustment: >> > after >> > all, it is also entirely the possibility that 2X will be the >> > lower-hashrate >> > coin in a few months, with the Core chain bullying them out of >> > existence. >> > Perhaps you should cross-post your analysis to bitcoin-segwit2x also. >> > After >> > all, the 2X developers should also want to have faster price discovery >> > of >> > the true price of 2X, away from the unfavorable (incorrect?) pricing o= n >> > BitFinex. >> > >> >>Devs are a governing authority under the influence of users, hodlers, >> >>and miners. Miners are like banks lobbying government for higher total >> >>fees. Hodlers are the new 1%, holding 90% of the coin, lobbying both >> >>devs and users for security, but equally interested in price >> >>increases. Users are "the people" that devs need to protect against >> >>both hodlers and miners. They do not care about price as long as it is >> >>stable. They do not want to become the 99% owning 10% of the coin or >> >>have to pay unecessary fees merely for their coin to be the biggest >> >>bully on the block. A faster responding difficulty will take a lot of >> >>hot air out of the bully. It prevents miners from being able to >> >>dictate that only coins with high fees are allowed. They are less >> >>able to destroy small coins that have a fast defense. >> >> >> >>The 1% and banks would starve the people that feed them to death if >> >>they were allowed complete control of the government. Are hodlers and >> >>miners any wiser? >> > >> > Are developers any wiser, either? >> > >> > Then consider this wisdom: The fewer back-incompatible changes to a >> > coin, >> > the better. Hardforks of any kind are an invitation to disaster and, = at >> > this point, require massive coordination effort which cannot be feasib= ly >> > done within a month. Fast market determination can be done using >> > off-chain >> > methods (such as on-exchange trades), and are generally robust against >> > temporary problems on-chain, although admittedly there is a counterpar= ty >> > risk involved. The coin works, and in general there is usually very >> > little >> > need to fix it, especially using dangerous hardforks. >> > >> >>Devs need to strive for an expansion of the coin >> >>quantity to keep value constant which is the foundation of the 5 >> >>characteristics of an ideal currency. >> > >> > Is that your goal? This is a massive departure from the conception of >> > Bitcoin as having a fixed limit and effectively becoming deflationary. >> > It >> > will also lead to massive economic distortions in favor of those who >> > receive >> > newly-minted coins. I doubt any developer would want to have this >> > property. >> > >> > Regards, >> > ZmnSCPxj >> >> >> ------------------------------ >> >> _______________________________________________ >> bitcoin-dev mailing list >> bitcoin-dev@lists.linuxfoundation.org >> https://lists.linuxfoundation.org/mailman/listinfo/bitcoin-dev >> >> >> End of bitcoin-dev Digest, Vol 29, Issue 21 >> ******************************************* > > > _______________________________________________ > bitcoin-dev mailing list > bitcoin-dev@lists.linuxfoundation.org > https://lists.linuxfoundation.org/mailman/listinfo/bitcoin-dev >