9 ms·
Select a document and have it certified in the Bitcoin blockchain
- cmrx64 10y agoNeat. Trusted timestamping (which this provides) is a useful primitive.
- Bartweiss 10y agoIt's actually surprising to me that there isn't already some established service for doing this. Situationally useful, but still a pretty cool way to handle things like dating a trade secret without revealing it.
- ryan-c 10y agoThere actually are several time stamping services based on RFC 3161 - many CAs offer this service, and Microsoft uses it as part of driver signing. https://en.wikipedia.org/wiki/Trusted_timestamping https://en.wikipedia.org/wiki/Trusted_timestamping
- ikeboy 10y agoPreviously: https://news.ycombinator.com/item?id=6819367 https://news.ycombinator.com/item?id=6819367 https://news.ycombinator.com/item?id=6809929 https://news.ycombinator.com/item?id=6809929
- strictnein 10y agoReading the https://proofofexistence.com/about https://proofofexistence.com/about is really helpful, because I really didn't understand this at all. While this is interesting, is there any concern about over use of the bitcoin blockchain? It's currently 65GB in size[0], which means it's fairly usable, and with a big enough system you could still store the entire thing in memory. What happens when it's 65TB? 65PB? Won't using it for lots and lots of things cause issues long term? Or am I missing something here? [0] https://blockchain.info/charts/blocks-size?timespan=all https://blockchain.info/charts/blocks-size?timespan=all
- ikeboy 10y agoEventually you'll be able to trim some parts, and only full nodes will need to store everything. The code is not quite ready yet, though. I believe there were plans to include a hash of the unspent transactions set in each block, so you could have effectively the same security by only verifying the block headers up to the latest block, and the unspent transactions set. Not sure if they ended up including that, though. There were a couple other interesting proposals along the same lines as well.
- thescriptkiddie 10y agoYou can already tell the bitcoin client to prune old blocks, because theoretically you don't need to keep any blocks more than a few hours old as long as you maintain a database with the last known ownership of every coin. In the future you could expect a setting to have the client ignore coins you don't control and crunch the entire db down to less than 1 GiB.
- ikeboy 10y agoYeah, but you still need to download all of them at least once to ensure there aren't missing transactions. The proposal with the UTXO set would allow you to only download headers+UTXO.
- thescriptkiddie 10y agoYes, you do currently have to download the whole blockchain at least once. I can't remember if you have to save the whole thing before pruning begins or if it can prune as it goes along. Hopefully the headers+UTXO proposal will be implemented soon, as the total size of the headers is only 32 MiB (80 bytes per block) and the total size of the UTXO set is only 1.2 GiB. We've already dramatically reduced the bandwidth requirements with Xtreme Thinblocks, so storage cost is the major limiting factor for running a node.
- TylerE 10y agoOf every address. There is no such thing as a bitcoin.
- overcast 10y agoI like this. Would have been amazing if this existed before the social explosion of the web. Where everyone is constantly stealing everyone elses work.
- brighton36 10y agoWhy is this better than archive.org?
- ikeboy 10y agoIt's computationally difficult for anyone to forge it. While the operators of archive.org could easily forge an old timestamp.
- hybridsole 10y agoComputationally difficult is a bit of an understatement. It is not possible to alter the bitcoin blockchain once a block is added and confirmed. With limitless resources, you could perform a withholding attack to modify the most recent half-hour to one hour's worth of transactions, but that would take tens of millions of dollars in hardware and only be temporarily successful.
- ikeboy 10y agoYou could go back in time indefinitely with unlimited resources. Would require a big reorg, but still possible.
- avar 10y agoFor the purposes of proving that a document exists no, you really couldn't go back in time, if this were to happen it would be a monumental time for bitcoin, and the old blockchain would be well known & referenced. So for the purposes of proving that you wrote that document it really doesn't matter if that were to happen.
- ikeboy 10y agoDepends who you're proving it to; if you can isolate them, it still works.
- slig 10y ago
- arnarbi 10y agoSo do I understand right that whoever hosts this service is paying 0.0001 BTC (4 US cents) for each document as the transaction fee?
- throwanem 10y agoNope. You have to pay .005 BTC to register a document.
- arnarbi 10y agoOh, I see. Seems .0049 goes to the maintainers then. https://www.blocktrail.com/BTC/tx/521f89f9f6980e57c7f3de60728080a245f62214cf5e69778a0fde36cedba487 https://www.blocktrail.com/BTC/tx/521f89f9f6980e57c7f3de6072...
- foobarbecue 10y agoI don't know for sure, but they are charging 5 mBTC. So if you're right, they could be earning quite the margin!
- Lxr 10y agoThis is really cool. I wonder if this would hold up in court in copyright or patent related disputes.
- chrisseaton 10y agoI think that old mail-it-to-yourself trick was a myth, so probably not.
- striking 10y agohttp://www.snopes.com/legal/postmark.asp http://www.snopes.com/legal/postmark.asp Tl;dr: myth.
- foobarbecue 10y agoI wonder what Descartes would say to this.
- mungoid 10y agoKinda cool, but how is this any better than just gpg signing? Is it because you could fake a timestamp with gpg but not with this?
- ryan-c 10y agoRight, the GPG timestamp is trivial to fake. There is also RFC 3151 timestamp protocol, but it uses a trusted third party rather than distributed trust.
- somecoder 10y agoI feel like I used this years ago... Oh, I did. https://news.ycombinator.com/item?id=6809929 https://news.ycombinator.com/item?id=6809929
- wslh 10y agoIt is good to compare with the legibility of a proof of existence Ethereum contract: https://github.com/maran/notareth/blob/master/contract.sol https://github.com/maran/notareth/blob/master/contract.sol
- TD-Linux 10y agoTo be clear, you're comparing this to the Bitcoin script used by this website? OP_RETURN <hash of document>
- wslh 10y agoThat is not the only part of the story and probably you know this and you are hiding it to the general audience that doesn't know about Bitcoin internals. You don't have a global getter on bitcoin transactions to get the hash of every document, so querying all the documents persisted on the Bitcoin blockchain you should build a database outside the Bitcoin blockchain to obtain them. And don't forget the strict buffer limitations in the OP_RETURN that makes difficult to add more data and require to span the buffer within multiple transactions.
- woodcut 10y agoThey should have called this Satoshi As A Service (SAAS)
- btcprint 10y agoYou win the internets today.
- aminok 10y agoBitcoin was originally described as a "peer-to-peer distributed timestamp server"[1] so this is an ideal use case. 1. https://bitcoin.org/bitcoin.pdf https://bitcoin.org/bitcoin.pdf
- angilly 10y agoI let out an audible "wow" when I read the about page. So freaking cool. The implications and uses for this are incredible.
- bobwaycott 10y agoLet me see if I'm thinking on the right track about potential applications here. First thing I thought of was proactive defense against someone patenting an idea by proving your prior art without necessarily divulging the information (or divulging it, while proving its chronologically prior existence). After that, my mind jumped to media, such as the ability to be provably certain a given piece of media hasn't been altered after the fact (docs, images, contracts, video, audio, etc.), and that it provably existed when a party claims it did. A bit on the nefarious side, I thought of being able to prove the existence of certain media a party wishes to use for blackmail or some other purpose. Maybe applicable to the HN/YC crowd—especially given a certain case that recently made news here about a YC alum—a means of proving the existence of, say, founder agreements when establishing a company, so that protracted legal disputes over ownership could be aided and possibly shortened by irrefutable proof those agreements existed at a particular point in time. Disclaimer: not a Bitcoin user, and know practically nothing substantial about it.
- nickpsecurity 10y agoryan-c beat me to it. Trusted time-stamping is probably better given there's reputable businesses that have existed for a long time and probably court precedents for their models. Their model is also way more efficient than Bitcoin as alternatives often are. Link here: https://en.wikipedia.org/wiki/Trusted_timestamping https://en.wikipedia.org/wiki/Trusted_timestamping Here's a few companies specializing in it with various tradeoffs: http://www.surety.com/digital-copyright-protection/prove-ownership.aspx http://www.surety.com/digital-copyright-protection/prove-own... http://www.proofspace.com/timestamping/ http://www.proofspace.com/timestamping/ https://www.digistamp.com/ https://www.digistamp.com/ Here's a paper on fast, decentralized security via witnesses that has timestamping at 120,000 requests a second with 4,000 witnesses: http://arxiv.org/pdf/1503.08768.pdf http://arxiv.org/pdf/1503.08768.pdf
- Natanael_L 10y agoUntil those companies tanks. Hopefully it wouldn't happen too fast, but the blockchain isn't dependent on a gatekeeper.
- nickpsecurity 10y agoIt can't be sabotaged by the developers or big mining pools? I keep seeing articles worrying about both.
- retube 10y ago> but the blockchain isn't dependent on a gatekeeper Apart from a tiny group of devs and mining pools engaged in non-stop bickering and infighting
- nickpsecurity 10y agoWhereas the companies I mentioned are about and currently practice long-term stability to provide a return on their investment. The infrastructure they need is also already built. So, their main incentive is to keep letting the money flow in doing something useful. Whereas, with issues you cited, I have no clue what the status or goals of Bitcoin will be in even 5 years.
- gwern 10y agoSimilar is OriginStamp http://www.originstamp.org/ http://www.originstamp.org/ which is free; it does this by batching up a day of hashes into a single master hash which is then timestamped by Bitcoin. I've written up some shell and Git scripts you can use to very easily timestamp stuff: http://www.gwern.net/Timestamping http://www.gwern.net/Timestamping Timestamping has come in handy for me in the past; you rarely need a cryptographically strong timestamp but when you do, it's important.
- exabrial 10y agoNice! I'm bookmarking this
- scottcanoni 10y agoWas I really the first to upload a TXT file that only had the words "hello world"? https://proofofexistence.com/detail/b94d27b9934d3e08a52e52d7da7dabfac484efe37a5380ee9088f7ace2efcde9 https://proofofexistence.com/detail/b94d27b9934d3e08a52e52d7... I was expecting to see someone else's document that had that text.
- xchaotic 10y agoI suspect they are salting and using timestamp so that two similar documents don't get the same hash?
- Noctem 10y agoI've seen people post SHA-1s on Twitter for things (like proof of an iOS jailbreak) that they aren't ready to release yet. I suppose this is a more reliable method for that.
- chungy 10y agoIt only takes one person at Twitter the corporation to edit their tweet database and change the sha1. The Bitcoin blockchain is nearly impossible to tamper.