After a hard fork, as long as both branches survive, there are twice as many bitcoins around.  If you had 100 BTC before the fork, then after the fork you have 100 "Series A" BTC and 100 "Series B" BTC. 

As I clearly said every time, whether people bother to mine and use each Series is a political and marketing qestion.  There is no technical obstacle to both series of bitcoins surviving and retaining some value (but quite likely their combined total value will be no greater than what the "Series A" were worth before the fork).

At the last hard fork, everyone agreed to abandon the "Series A" bitcoins and use only the "Series B" ones.   If we have another hard fork to increase the block size, some people seem to be willing stay on the "Series A" branch.  I don't know what will happen, but, agan, it will be decided by political processes, not by technical mechanisms.

Exactly


The bitcoins are effectively duplicated because they can be manipulated independently in the two blockchains.

This is just a particular  instance of the observation that, while there is a finite supply of "the" bitcoins, there is no limit for the number of cryptocoins.  Random altcoins are often dismissed because they lack bitcoin's history; but a bitcoin clone created by such a hard fork would share bitcoins history up to the "schism".  Which branch would be the "legitimate" bitcoin is not a technical question, just a political and marketing one.


The possibility of a hard-fork clone does not expose any security risk, it is directed against the "scarcity" claim.

In 2011 (or 2010?) someone noticed a bug in the implementation of the multiply instruction, that did not properly guard against overflow.   Exploiting that bug, the guy managed to creat an unspent transaction output with a humongous amount of bitcoins.  In order to fix that bug, Gavin and other developers had to perform a hard fork, starting from some block before that buggy transaction. (You can easily find this story in the internet.)

So, for a while, there were two clones of bitcoin, starting off from that same block: one compliant to the old software, with the buggy OP_MUL and the gazillion BTC output, and one compliant with the new version, where the OP_MUL instruction was disabled.  Most of the transactions that had been processed in the old chain after the fork were re-issued on the new chain, but some (inclunding that one with the gazillion output) had to be discarded.  So the two chains were effectively two similar, but not identical, versions of bitcoin.

By common agreement, the old chain was abandoned and everybody switched to use the new chain, by upgrading their software.  But anyone could have continued to mine the old branch (except that the difficulty would have to be lowered artificially, for a while, to allow a decent block rate).  Any bitcoiner could have continued to move, spend, sell his old bitcoins, using the old software and the same private keys, independently of what he did with the new bitcoins in the new branch.   Of course, the OP_MUL bug would have quickly made a big mess of that old branch.

But the point is that the the old chain was not "killed" by some mechanism in the protocol, it was just abandoned because no one had interest in it.  If the bug had not been not so severe, perhaps some people would have continued to use it, for fun or even for profit.  Anyway, while the two branches ran in parallel, the number of existing bitcoins was effectively doubled.

I posted a recipe for creating such a clone.  It is not original, just the recipe for a hard fork.  Do you think that there is something wrong with the recipe?

I just posted it, again:  https://bitcointalk.org/index.php?topic=178336.msg10811221#msg10811221

I consider part of my job to understand computing subjects that have impact on society and offer advice on them. (For example, for the last 10 years I have been writing and speaking out about the risks of the Brazilian voting system.)

I started looking into bitcoin in late 2013 when it was starting to be sold here as a fabulous investment or hedge against inflation.  Fortunately, my fellow Brazilians have been largely skeptical of those claims, perhaps because they had just witnessed the collapse of a huge tech-based MLM scam, that stole a billion dollars or more from millions of "investors".  So my advice has been unnecessary so far.

As for this thread specifically, I have more than 50 bitcointalk threads on my watchlist, but this is the only one that has significant traffic and shows up every time I refresh the list.  It is also sort of the "local pub" where all news and topics get mentioned, from Bloom filters to beers.  And, with few exceptions, posters here are basically gamblers, which are mostly OK people; while some of the people who post on some of the other threads are defintely not OK...


Well, I have the same impression of bitcoin believers: they won't admit that bitcoin has serious problems, no matter how obvious...

No, rather accomodation of the main tool of accounting and of other important software tools.


Good programmers know how to use FP without any loss of precision.  Indeed, they often use FP for integer computatons because on some processor chips the FP multiply and divide units are often faster than the 64-bit integer counterparts.  (That is how the infamous "Pentium division bug" was found, by the way.)


That recommendation is relevant only for amounts larger than 10^15 cents = 10 trillion dollars (or 10^15 dollars if all amounts are whole dollars).   Most accounting is done in Excel-like spreadsheetsthat use FP doubles for all quantities (even though users are rarely aware of it).

However, the thieves may be mobsters themselves.  They could refund damages of "colleagues", but may just show the Varoufakis Finger to rivals and small crooks.

I read a write-up of Gavin's proposal, some months ago, but it did not seem to have been thoroughly thought of yet at the time.   From what I understood, I thought that a malicious miner could broadcast non-invertible filters and make honest miners waste time waiting for the actual data to show up, and thus gain some advantage.  I don't know whether all potential attacks, like this one, were considered and excluded.

(You surely sound like a Brazilian, and an anti-Dilma one specifically.  At very least, you must be from some Latin American country?)

A 70% loss of value in 20 years, and a predicted 8% inflation rate per year are nothing to boast about, sure.  But they are still better than a 8% loss in 24 hours, or  78% loss in 14 months, both totally unpredicted and unpredictable. 

Several bitcoin companies have learned the hard way that bitcoin cannot be used as unit of accounting or as liquid capital storage, not so much because of its (past and possible) loss of value, but because of its unpredictability.  (IIRC, SatoshiLabs, the makers of Trezor, were one such example.)

For contracts that span 2-3 months, an inflation rate of 1%/year can be ignored.  An inflation of 10%/year cannot be ignored; but, if it is roughly predictable, it can be accounted for in the contract; either by adjusting the amounts to be paid, or pegging them to some mutually trusted inflation index.   On the other hand, a sudden and unpredicted inflation of 10% over one week can turn a good profit into a significant loss, even if some payments are pegged to some index.

By the way, the BRL:USD exchange rate was 2.74 in 2005, 2.71 last December, and below 2.70 at all times between those dates. Not that bad, eh?   

Finally, and, most importantly, I am not advising anyone to invest in BRL, or dollars, or euros.  No one with a bit of brain invests in a currency (except for very short term speculation, and even then it is a very risky gamble).

As of last September there were less than 700'000 blockchain addresses with 0.1 BTC or more.  One would think that each bitcoin believer owns at least one such address.  While there are many exceptions, there are also many bitcoiners who own two or more such addresses; so, "one million" seems to be a generous estimate for the number of bitcoin believers.

I have no source for the number of people in the world who know enough about computing to understand Satoshi's paper or any of the alternative technical explanations about bitcoin, but they must be several million.  That includes any computer professional who understands what public key cryptography is supposed to do (even if he does not know the details of the algorithms).   Do you dispute that number?

As for another bit of evidence, I know at several dozen such people personally, including a couple of cryptography experts; and I have still to meet one who "believes in bitcoin".


I explained how in that post; it is nothing new or complicated, it has happened a couple of times already, and is often proposed (incorrectly) as a defense of last resort against an "evil" mining majority.

The kid gets a copy of the software, makes some cosmetic change that clearly distinguishes messages of his clone from those of the original version after a certain block N, and lowers the difficulty by fiat in the clone after that block so that he can start mining  on his laptop.   After that block, presto, every bitcoin in existence gets a clone in the kid's chain, that can be accessed with the same keys and moved (spent, sold, etc.) independently of the original bitcoin. 

Whether bitcoiners will be aware of the existence of those "kidcoins", and will choose to use and/or mine them, is only a marketing problem, not a technical one...


A miner now earns ~7000 dollars for solving an empty block, and ~7025 dollars for packing as many transactions as will fit in it.  Larger blocks take slightly longer to propagate through the network, so the miner must consider whether those 25 dollars are worth the risk of losing the race just because of that handicap.

AFAIK, right now one can see many blocks being mined which are only partially full, while many transactions are pending in the queue.  Clearly, the current transaction fees are not enough of an incentive.


Indeed, and those historical examples are a strong argument for centralization of mining being inevitable.  But while centralization is bad in general, it is a fundamental problem for bitcoin, because "absence of a trusted third party or central controlling authority" was supposed to be the goal of the project.  As bitcoin gurus themselves concede, a mining company or cartel who has a comfortable majority of the hashpower can control the system, e.g. by starving miners who are not part of the cartel, freezing any account that the cartel feels like freezing, etc..  Right now, one must trust that the 4-5 largest miners will not use their power for "bad things"... 

The security of the bitcoin system is supposed t come not just from the technical features of the protocol, but mainly from the incentives that are assumed to convince the majority of the players to cooperate and protect it, rather than sabotage it.   That assumption is far from certain if a handful of companies has the majority of the hashpower.

These are serious problems of the current implementation and of the community, indeed.  However, the low transation rate may be fixable, with larger blocks or some other technical machinery. 

The second problem -- disincentive for miners to include as many transactions as possible -- may be more severe.  Has anyone proposed a solution?  It is a consequence of the block reward being so high compared to the transaction fees.

Last times I checked the distribution was variable but generally concentrated, 4-5 companies owning 51% of the hashpower.  And  they are now all Chinese.

That is not a "mechanism" but only a "hope".  Currently the price is almost entirely (90% or more) set by speculation, ad that is the mais reason for the high volatility.  If the price increases to 1000 $/BTC and beyond, that situation would only get worse.  


Is there any example of a deflationary currency that has succeeded?  


You miss the point.  The money that new investors are putting into the system goes to the miners' pockets and to pay their bills.  Like if a 100 dollar bill cost 100 dollars to print. Or if all the money that investors put into Apple stocks were spent to print super-unforgeable stock certificates, instead of factories and stores.

Moreover the people who use the system (by issuing transactions) have the illusion that the service is free, because the real costs are being paid by those new investors.  That is not a sane free market.


Note, "for most people".  

(And cash and gold are not as irreversible as bitcoin. They require physical presence, so the police has a better chance at recovering stolen cash or gold than of recovering stolen bitcoins.)


We'll see.  Bitstamp was recently hacked, BCI relased broken code to their clients...


It is not just hand-waving.  There must be millions of computer scientists who can read  Satoshi's whitepaper with no difficulty, and can understand the essential priciples of the protocol, even without reading the source code.  However, not all computer scientists believe in bitcoin.  In fact, based on the limited sample I have, there must be far more non-believers than believers among the computer scientists.

Some problems:

* Centralization of mining seems inevitable.

* There is no mechanism to stabilize the value.

* Limited supply leads to expectations of high value which leads to hoarding instead of use.

* Block reward leads to a hyperdeveloped mining industry supported by investors rather than by users.

* For most people, irreversibility is a serious defect, not a feature.

* Risk of theft is too high for people who are not computer experts.

And maybe more.


Perhaps a million people have a reasonable understanding of bitcoin and believe in its eventual succeess.  Probably millions of people also have a reasonable understanding of bitcoin but do not believe in its eventual success.  The other 7000 million either haven't heard of it, or have heard of it but don't care to know more, or know just enough to dismiss it.

I don't quite understand your remark.  I don't see "Wall Street" rushing to buy bitcoins.  I see a lot of venture capital going into companies that plan to make a profit from others using and trading bitcoin.

E.g.,  the bitcoin exchanges make money whether the bitcoin price is rising or falling. Ditto for bitcoin payment processors, managers of bitcoin funds,  bitcoin storage companies, makers of hardware wallets, etc..  Those ventures are "safe bets" because they may well return the invested capital, with profit, even if the BTC price were to crash and they had to close. 

I see bitcoin like the Wright Brothers' Flyer One model:  a notable experiment, but which still needs many improvements and bug fixes (some of them with no solution in sight) before it can become a viable commercial service.

Unfortunately that experiment has been appropriated by several groups for purposes that it was not intended for, and was not ready for; the latest one being as an investment/speculation instrument.   It is as if some entrepreneurs in 1908 had tried to set up airlines with fleets of Flyer Ones, or sell Flyer Ones to the general public by claiming that one day they would be in high demand and therefore extremely valuable.

I missed the news about her, indeed.  But the list is very long -- of people who obviously do not believe in bitcoin, but still hope to profit from the Bitcoin phenomenon...

Credulous bitcoin investors are still pouring more than 1 million of fresh sweat-earned dollars into the bitcoin system, every day.  (How much more, no one knows.)  All sort of smart guys, from "Wall Street" or not, are very much interested in pocketing some of that money.  That is the only reason why Andreessen, Sielbert, the Winkles, and many, many other professional money skimmers have been investing in bitcoin enterprises...

It seems that many of his aliases have been around all this time, no?

There could have been no rational justification for the choice of the initial block reward and its evolution over time, nor for the maximum number of coins -- although these parameters are obviously linked.  Arbitrary for arbitrary,  halving the rewards at fixed intervals was chosen because it was the simplest schedule to explain and implement.   

The next decision was to limit the number of atomic units (satoshis) to about 2^51 - 1 = 2'251'799'813'685'247, which is the maximum range of integers that can be safely manipulated in IEEE double-precision floating point without rouding.  That would allow any bitcoin accounting to be done on an Excel spreadsheet without losing a satoshi.  (Excel, and many other programs, use double-precsion floating point for all numbers,  Most computer nerds would never have tought of that.  They generally avoid floating-point because of its complicated rounding, and would have picked 2^63 or 2^62 instead.  But Satoshi obviously was not a typical computer nerd, knew how Excel works, and knew how to use FP without rounding.)

Once fixed 2^51 a limit for the max number of satoshis in the system, Satoshi had to choose the initial block reward and the halving period.  50 BTC is not a round binary number satoshis, but it is a nice round number of BTCs.  WIth that initial reward, the halving had to be every ~210'000 blocks, to keep the total number of satoshis below 2^51.  It was rounded to 210'000, meaning that the total will be a little less than 21 million.   If the initial reward had been set at 100 BTC/block, for example, the halving period would have had to be ~105'000 blocks, or about 2 years.