I don't understand what you mean by censoring code. There is a virtual machine and the virtual machine instructions are all deterministic. There is no instruction that could produce a random value.
+ No, because you can't check time either, so you can't write that program!
Hmmm... a) That sounds extremely crippled b) That sounds extremely broken if the result is ...forking (I'm not sure that it can keep things deterministic when there are tens if not hundreds of possible triggers for indirectly randomizing things). |
|
|
|
@AlexGR, everything in Ethereum is deterministic. There are no operations that have different results on different nodes. At least if everything works properly. If not then it can fork the chain.
Aha, that's closer to the answer I was hoping for. Thanks. Ok, so is it censoring code that it doesn't like or something? Let's say randomizing numbers is out of the question. Now let's do something else. I'm adding a constant number (not random), let's say the number 42, for 1 entire second. Everything is given / predetermined: a) The number to add (42) and b) how long I want it to perform what I want (1000 msecs). The result though is different because one pc will have made 500 million additions, another will have made 10 billion additions, depending their cpu power. Can I fork the network now?  |
|
|
|
|
Let's say I have a very simple program to run on the "distributed computing platform" of ethereum.
The program does one thing.
Provide an output of a random number (from 1 to 100)
How can that work on different nodes, and have them ....validating the same thing?
Unless it doesn't allow that type of functionality.
|
|
|
|
|
It's just an example, trying to understand how a random function would work.
|
|
|
|
if Ethereum is a distributed computing platform
It isn't; it's a replicated/redundant computing platform. If I make a simple program to get a random number from 1 to 100, or base my program on a randomizer, how can that be repeated in every single node? Shouldn't they give different outputs? How can that be validated? Let's say I want to make a betting app, based on the random rolling of dice. How can that work? |
|
|
|
Hmm, I was wondering about that too... if the "fuel" gets too expensive, then how can you run apps? It's like an economic disincentive to go well (price-wise) because increased "fuel price" will hamper use and growth... Despite being involved with altcoins, I haven't really looked into ethereum, primarily because I consider it a rule* that when a coin is too complex to be understood by most, it stands little chance in the marketplace. That principle has served me well over the years in that most cryptocurrencies that claimed to be 2.0 or something, tanked. The second reason was that when I was trying to find monetary information (amount of coins, futureproofing, inflation model etc etc) these were buried or lacking - and I was like "wtf? how can I even touch this?". It was also heavily promoted and I found this "fishy". Anyway, what I'd like to know, is, if Ethereum is a distributed computing platform, then what prevents someone from using Ethereum (as a distributed computing platform) to ...mine Ethereum or cpu-based altcoins? I'm reading about all nodes wastefully repeating the same computation to validate it but, as far as I understand, that could only be applied to something predetermined like 1+1=2. If I have a simple program that asks, say, a random number from 1 to 1000, then all nodes executing and verifying this program would give me a different value. The computation would be the same (in terms of cost) but it would allow me to perform work like random-brute force on a hash, or something. * https://bitcointalk.org/index.php?topic=637645.msg7114149#msg7114149 & https://bitcointalk.org/index.php?topic=637645.msg7116563#msg7116563 |
|
|
|
|
Bitcoin is, and will be, different things, for different people, in different time/space co-ordinates.
It cannot be defined as "e-cash", "decentralized paypal", "settlement network", "digital gold", "platform for blockchain-based apps" etc etc, because all these depend heavily on things that are dependent on time and space conditions.
At 3tx/s, one would say bitcoin can only be a settlement network or digital gold. However, due to low adoption right now, and for the previous 7 years, it can ...buy your coffee (ignoring double spending for a practical 0-conf payment). Fees like 2-3 cents don't register and are certainly not "settlement-network" level.
Could this change in 2-3-5 years if, say, we have a technological bottleneck in terms of scaling? yes it could.
Could this change AGAIN in 10-20 years, as tx/s capacity is in the tens of thousands/sec, allowing much more things than previously possible? yes it could
|
|
|
|
|
Not everyday seeing +/-300mn marketcap ups and downs in altcoins...
|
|
|
|
It took me 2 minutes to understand what you were saying, then realized poor AlexGR who is not that Alex Green, but instead ended up with the worst possible BCT name for 2015  Yah, our AlexGR is a long time Dash man, and OK, so no worries there I know, he's ranted about it before. Just couldn't resist yanking his chain. In any case, if anyone has stuck btc or dash maybe he might get it back eventually (?)... |
|
|
|
I guess it makes sense to support Core if you don't understand how Bitcoin works.
What are you referring to? |
|
|
|
|
Αυτος ειναι ο ογκος συναλλαγων στα χρηματιστηρια, οχι ο ογκος συναλλαγων των πληρωμων.
Δλδ μετραει ποσα btc μπορει να γινονται buy/sell σε ενα exchange αλλα οχι ποσα btc μετακινει ο κοσμος μεταξυ του, για να αγοραζει προϊοντα και υπηρεσιες, κτλ κτλ.
|
|
|
|
Από της 29 Φεβρουαρίου μέχρι της 5 Μαρτίου είχαμε την πολύ μεγάλη αύξηση του αριθμού και του όγκου συναλλαγών, την οποία ερμηνεύσαμε ως επίθεση DDoS. Έκτοτε (6/3 έως 12/3)συνέβη το εξής:
α) Ο αριθμός των συναλλαγών έχει πέσει σε χαμηλά επίπεδα (μ.ο 185.000)
β) Ο όγκος των συναλλαγών παραμένει στα ίδια υψηλά επίπεδα με εκείνα της "επίθεσης" (μ.ο 5εκ)
Υπήρχε κάποια συμφωνία, τροποποίηση, αλλαγή στον κώδικα ή οτιδήποτε άλλο που να εξηγεί τη νέα πραγματικότητα;
Ο ογκος δεν ειναι το https://blockchain.info/charts/trade-volumeαλλα το https://blockchain.info/charts/estimated-transaction-volume-usdΤο πρωτο αφορα exchanges και trading, το δευτερο τη συνολικη αξια των συναλλαγων που χρησιμοποιουν bitcoin. Οσο για το τι εγινε, ειναι απλο. Ο spammer εκοψε ρυθμο. |
|
|
|
1.1 to 1.5gb blocks are not feasible right now in a decentralized manner.
no one is asking for 1.5gb blocks today!! thats a futile argument to meander away from the real debate that bitcoiners are talking about today(2mb+segwit before 2018(preferably under 16 months), and then the POSSIBILITIES over the next TWENTY+ YEARS Although it's you I quote, it's for the OP to read. |
|
|
|
The willingness of people to save money by doing something in a more economical way can never die .
True, but you can try to stop them from trying to find certain types of more economical methods by using a position of power to dictate what terms the market must operate under. IT services cost, which are related to IT capabilities, will tend to follow IT capabilities cost. This means that the IT market (hardware, network etc) dictates, in large, the cost as capabilities cap the max amount of tx/s in a given setup (p2p / decentralized). In 1995, what we do right now at 3tx/s, with a p2p network that anyone can run on their home PC, would be impossible. Bitcoin would be considered a fancy idea that can't work due to extreme resource requirements. There were no 100gb home disks. There was no 4gb ram to run. There was no 8-threaded cpu, running at 4ghz. If you take the opposite time direction, in 2035 the cost of running at 3000tx/s or 10.000 tx/s may be the same as it is to run today the 3tx/s network. More technological capacity => more network capacity => cheaper txs due to better and more affordable technological means. |
|
|
|
At the same time we must make sure that measures to counter spam doesn't get in the way of other users. An open distributed decentralized ledger without spam or other unsavory things is a dead ledger.
The willingness of people to save money by doing something in a more economical way can never die . Even at a theoretical 1$ per tx (which would need BTC to be at 8000$ - 16.000$), that 1$ in fees is quite smaller compared to bank wire fees of tens of dollars / western union fees and still cheaper than paypal payments above 20-30$ (because paypal goes with a fixed fee PLUS a 2-3% of the sale). Even at 1$ fees it is still more profitable to use BTC than legacy systems for hundreds of transactions per second that are currently conducted and pay A LOT more. At that point you've indirectly evicted cheap spam (no "banning") AND you are also useful to everyday people who are saving money thanks to BTC. He set the block size limit to orders of magnitude larger than the then average block size.
He just gave attackers 1mb room to play with. To make extrapolations of use in 2010 compared to max block size (and perhaps imply that we should make like 50-100x larger blocks than what we need) is ...well. |
|
|
|
also to note. Visa's 42,000-45,000 tx/s is not based on real data.. its based on small group of computers in a lab, pushed to their limit with fake data.
I think the target is definitely not in the 40k tx/s range. https://en.wikipedia.org/wiki/Visa_Inc. In 2009, Visa’s global network (known as VisaNet) processed 62 billion transactions with a total volume of $4.4 trillion.[6][7] /365 = 170mn txs per day /86400 secs per day = 1966 tx/s on avg. Right now it should be close to 3k on avg due to growth in electronic payments. So the target is around 3-4.000 tx/s. This is 1100 to 1500 times more than the 2.7tx/s.... so we are looking at 1.1 to 1.5gb blocks just to handle an average annualized load (no peaks). 1.1 to 1.5gb blocks are not feasible right now in a decentralized manner. |
|
|
|
AlexGR, do you allow for the possibility that maybe small blocks isn't a good idea?
Timing is crucial. Even 1TB per year blocks (20mb/block) will have its time when 20mb/block will be "alright". Upgrade too soon, you'll have 10gb txs and 990gb spam. Upgrade on time, you'll get 800-950gb txs and 50-200gb spam. I don't understand why it's spam when it pays a fee. Miners are free to exclude tx if the fee is too small. Do you get spam in your ordinary physical mailbox? Do you think this spam is free? Isn't someone paying for the design, the paper, the kids who deliver it, etc etc? Does this alter what it really is? It's still spam. Who cares if its paid or not. As for miners, they will tend to follow the network guidelines, usually set by devs suggestions on default values. (However devs can't suggest higher fees due to the political climate) I sort of like his moral approach to txs. Maybe we could introduce KYC on protocol level and purge the system for drug and CP payments as well. The latter seems even worse than spam txs.
There is no moral approach. It's just common sense to protect the system from abuse. Who told you that the system can't be attacked or abused? Certainly not Satoshi. My reasoning goes like this: What would Satoshi do if he had to deal with a situation like this? a) Give spammers as much space as they want - ever increasing capacity to meed their ever-increasing demands for cheap space to bloat, at near-zero costs for them, and significant costs for the network, increased centralization and problems to btc's adoption b) Do something about it (blocksize limits / fees directly / fees as a consequence of a fee market, etc) Well, the only thing that makes sense is B. That's why not only bitcoin devs are going with (b) but ALSO altcoin devs that faced some kind of attack of this nature and patched their systems to protect them. But the armchair experts all "know better", pretending spam isn't spam, and that spam growth equals growth. Newsflash: The actual btc growth, that needs to be overlapped at the metcalfe chart with the marketcap, is not spam but legit txs - whether one has a good system to evaluate that, or by using statistics like excluding long chains (of spam). If we can pump the marketcap by saying "see? wow, we have such a network effect" by simply spamming 10mb blocks and actually harming bitcoin in the process => then the market is broken because it is run by total idiots and bitcoin's price mechanism would be even more broken if its price was in direct relation to the spamming (rather than actual use). Likewise, if someone says that ...1mb prevents growth because txs are hitting the limit, he's got his fundamentals waaaay wrong. Legit growth is ongoing, despite the 1mb limit - and still way below 250-300k txs per day: Excluding chains >100: https://blockchain.info/charts/n-transactions-excluding-chains-longer-than-100?timespan=2year&showDataPoints=false&daysAverageString=1&show_header=true&scale=0&address=Excluding chains >10: https://blockchain.info/charts/n-transactions-excluding-chains-longer-than-10?timespan=2year&showDataPoints=false&daysAverageString=1&show_header=true&scale=0&address= |
|
|
|
Can someone please explain to me how we have these Satoshi quotes and yet there is still any discussion / debate / disagreement on it? The quote says one thing, reality says another. Reality = right now you can't get to a blocksize where you can do visa level tx capacity. And this is similarly true, not only for bitcoin, but bitcoin-based clones and other blockchain-based systems. |
|
|
|
By the time of the next 2-3-4 halvings, our blocksize limits could be in the hundreds of megabytes.
given that phones will have terabytes of non-volatile nanosecond latency memory by then, it shouldn't be a problem. but, competing with low-latency centralized settlement while surviving kyc/aml will be a challenge before 2020. I'm pretty sure there'll be pleeeenty of challenges till 2020 |
|
|
|
|
Show Posts
