When I've completed ACME's “sea trials”, I'll have pretty much all I need, I reckon. I'm following a couple of approaches towards integrating metadata: Slimcoin's OP_RETURN approach and, just for completeness, Datacoin's 128Kb-per-tx “data” field.

ACME instances:

http://tessier.bel-epa.com   port 5064 for Slimcoin and port 5059 for Datacoin.

I haven't yet enabled caching so you'll need some patience for the occasional sloooow page load. Not everything works but the Slimcoin adaptation is further along than the Datacoin adaptation. The key work is taking place in “Publications”. In Slimcoin, txs with OP_RETURN data are retrieved and the data displayed verbatim (lacking a “show me only OP_RETURN values for tx made by addresses I choose to specify” ). In Datacoin, I have identified the tx with data fields (4300-odd) and am trying to MIME-typify the stored binary (could be a JPEG, DOC, ZIP, *anything*) to present a paginated list (as well as an “only these addresses” filter). The Datacoin dev devised a solution using Google's protocol buffer library but I speed-read one post that concluded the approach wasn't powerful enough. I'm just verifying for myself that is the case before settling on the RDF-based approach.

Datacoin is a bit of a puzzle and potentially a cautionary tale from my perspective. There's a thread from three years ago which pretty much covers it, provides answers, functionality, 3rd-party support services, everything you'd think was necessary for this to get off the ground, yet only 0.2% of the not-quite 2m blocks+tx have non-empty data fields and a significant percentage of those are noise, i.e. hold no discernible value (although the value may just be in the timestamp). So why didn't it take off? I need an answer to that question before going any further in terms of “people using the blockchain as a publishing device”.

(the gory details are in my posts passim to the Slimcoin and Datacoin bct threads, not written it up yet, sorry).

Cheers

Graham

I'm working out the details of a metadata strategy with help from the Slimcoin and Datacoin communities.

Cheers

Graham

The first one is running off've our rented Hetzner box, should be good for a few weeks yet (until we upgrade), the rest are from “getpeerinfo” on that box ..


addnode=144.76.64.49

addnode=73.211.15.158:1108
addnode=5.189.154.154:1108
addnode=86.88.39.133:49534
addnode=31.130.253.12:53214
addnode=51.6.166.115:1108
addnode=24.11.237.176:53281
addnode=96.246.48.110:50519


btw, just to avoid misperceptions ... not a “dev”, just a contributor.

Cheers

Graham

It's a long shot but I've been caught out before with self-edited Makefiles fumbling the removal of binaries from the build directory, being skipped by make and, if I'm wrestling with cross-compilation, ultimately causing the linker to complain about unrecognised format. I also used to see that after the laptop crashed, leaving truncated .o files in the build directory. What you're describing suggests that the secp256k1 code is being natively compiled when it should be cross-compiled. I've also encountered crypto files that need to be compiled with plain C, requiring ${CC} to be properly bound to the corresponding MXE binary.

Cheers

Graham

That entails a long running jump and grab to an 0.14.2 Core clone. And a hard fork. That's quite a big “ask”.

Right now, I'm trying to get a sense of the role that Datacoin's unique strorage feature plays in the overall functioning of the coin. It's PoW, so people can readily acquire coins by mining, to pay the fees to store data. It's all nicely self-contained and unchanged from the day it was launched. There's nothing to stop people from making full use of the feature.

There have been a total of 4,327 data tx made since Nov 2013 during which time (/me checks) 1985959 blocks have been chained. A significant number of tx can only be classified as “noise” (dozens and dozens of entries 'W10=' and 'Broughs=').  Even if this “noise” is included, I make that around 0.2% (of blocks).

I'm sorting out some public filesharing space so I can publish the raw lists should anyone wish to conduct their own analysis.

Cheers

Graham

1. Noted.
2. Good questions. As Chicago has pointed out, there is a twisty-turny path to upgrading Datacoin to a clone of 0.10 Core. Dash is a clone of 0.10, as are its klonekinder. Monero uses the Cryptonote protocol and would be an unsuitable target. I've not investigated Zcash in any detail, so I don't have much sense of the degree of complexity of the migration. So right now, any “anonymity” feature is restricted to coin-mixing-via-overlay-network, a la Dash, i.e. not actually anonymous.

On other matters, the entire(*) Datacoin blockchain has now been mapped to RDF, so addresses now resolve. It's a single-threaded server, straight Python (waitress) and I'm also not at all sure that I've configured Fuseki with sufficiently generous resources and because I'm still working on it, nothing is being cached, so response alacrity is “best effort” atm.

(*) I accidentally-on-purpose omitted to copy the contents of the tx data field over to the RDF graph as the stored data isn't strictly metadata. Instead, in the next workspace over, as I write this, a small Python script is working its way methodically through the blockchain, recording the hash of each block that has one or more data txs, the hashes of those txs and the size of the stored data for the tx.

Sample output, copynpasta'd:

790000 <- height achieved thus far
["2be1fffdb16c1163ece07029862f58fcc90f55e91dba42bd1f17a8f5a7187366", "1ec6a6370f603be614cc29338aeb2a9abaf55bb9a41012fb80e38bffd3c4b9de", 136],
["2be1fffdb16c1163ece07029862f58fcc90f55e91dba42bd1f17a8f5a7187366", "3b500c4bf2841c80bd3ff3ebe7090865e521bbc55ab753d0ba1cf16466595a5f", 140],
["2be1fffdb16c1163ece07029862f58fcc90f55e91dba42bd1f17a8f5a7187366", "1a3966eb8ac879c5de7d19675f46b7ffbe1074e57ab0b6c35c3287991b3b438b", 136],
["9da914b8e0236dda7e87c55e3079a0825c3be173119384ac636d732f9d77d536", "4ca286534961fc8ff10dd7e4ef3d42ccba9a9c3efac2b5aa23476f09cc6b6b2e", 136],

you can try them out:

http:// tessier.bel-epa.com port 5059  /main/tx/1ec6a6370f603be614cc29338aeb2a9abaf55bb9a41012fb80e38bffd3c4b9de

I'll then add those facts to a separate RDF graph (to keep the blockchain mapping untainted) and re-do the exercise, this time using a routine that attempts to determine the MIME type of the stored data, which when confirmed, can be added to the graph.

(If the way that ACME presents the metadata makes little sense to you, check out the accordion drop-down “Unprocessed JSON response”, that's what we have to work with - if anyone has suggestions for making a clearer presentation, please feel free to pitch in.)

And before anyone jumps in to  point out the obvious, a Python web app that marshals and presents data retrieved over the wire can be rewritten using a Javascript framework, browserified and embedded in the client, courtesy of Qt5's webengine implementation of v8. Windows binaries would have to be compiled natively because (aiui) there is no MXE cross-compiler implementation of qtwebengine because of, well v8 and Microsoft but I have proof-of-concept browserified js apps running out of a Linux build of a 0.14 Core clone.

The possibilities are entertaining: https://doc.qt.io/qt-5/qtwebengine-features.html -- and, because it's Qt and restricted to the GUI part of the wallet, adding user-appealing functionality doesn't entail any fiddling with the blockchain engine and is basically just a bolt-on.

Cheers

Graham

Use your computer's search facility to find where the boost libraries were installed.

Cheers

Graham

Sterlingcoin is being delisted by Cryptopia and they paid for their listing.


Inverse ordering by "Wallet Status" on https://www.cryptopia.co.nz/CoinInfo/ brings “Delisting” to the top and filtering out those with 0 connections or POW, gives an interesting list of candidates.

Cheers

Graham

See the code example upthread:

https://bitcointalk.org/index.php?topic=1080289.msg17336514#msg17336514

Cheers

Graham

Installed or not, the libraries aren't being found by the build process, looks like you need to tend to your environment variable bindings.

Cheers

Graham

Sure ...

https://stackoverflow.com/questions/2404115/is-auto-ptr-deprecated/27113612#27113612

Replace the references to auto_ptr with unique_ptr.

Cheers

Graham

I have no knowledge of KMD BarterDEX, I tried reading their white paper but it made little sense to me. At least I tried.

Slimcoin's JSON-RPC API does not include a gettxout call.

That's the lacuna that ACME addresses.

Cheers

Graham

Ah, thanks for that clarification, I've not had the time to pin down the crossover details precisely. Thinking back, the coin I was upgrading was a straight clone, just param changes, so when they introduced 0.9, I just re-rebranded Bitcoin and migrated the coin's params to chainparams.

Co-incidentally, the refactoring has given me an opportunity to lend a bit more depth to an existing piece of work ...

More than one person has failed to detect the tongue-in-cheek nature of Minki’s Foundry: https://minkiz.co/foundry.

Originally the advertised “latest Bitcoin Core release” was 0.9 because that's when I published the work. Riffing off've the re-rebranding and the distillation of the parameter differences into a single location, I created for myself a “template branch” of Bitcoin Core with all the re-rebranding and parameter distillation abstracted out and replaced by Jinja2 template statements. I had been playing around with the notion of distilling a coin's profile (hence the DOACC metadata collection http://github.com/DOACC/individuals) and for a while I toyed with the idea of trawling Github-hosted chainparams.cpp files and extracting the specific technical metadata for adding to DOACC. In the end, I brought DOACC to a close, it had served its purpose, and I decided to go in the opposite direction - use the metadata + templates to implement the Foundry offers. Now that the pow functionality has been handily abstracted out to pow.[cpp|h] it's just too tempting to resist. I'm too involved atm but I have a compiled version of FerrariBoat to return to at some point in the future. (The naming convention is homage to a classic Snickers ad: https://www.youtube.com/watch?v=anm2SWU_BHo)

This is the YAML description of Chaincoin from a while ago, with Bitcoin values as end-of-line comments:


CHC:
  checkpointestimatedtransactions: 572007 # 60000.0
  checkpointtimestamp: 1434866355 # 1397080064
  checkpointtotaltransactions: 960 # 36544669
  checkpoints: [
    [0, "0x00000f639db5734b2b861ef8dbccc33aebd7de44d13de000a12d093bcc866c64"],
    [6143, "0x0000000026fb51f5bc9943ed69d9ff7697ecf7fed419d88b417655f93a487ce1"],
    [12797, "0x000000002c29644e179baa188fa6b9b9454721f1f21f2b9f31eebe9acc1a31db"],
    [30092, "0x0000000098a23e1c503f71a6d61c333c5abaabb4c5fa1b474012e004db4bfbbe"],
    [80998, "0x000000010ebcfe9a00a99f2b61104f4a141555a707f1c007aba8a978f6030cfb"],
    [144759, "0x000000047e7b7bfd63b4f019a0a24c8d65b10afa6eb80721e10fa7c49ce6fb6e"],
    [189046, "0x00000000bd507c435b46ee8a13b25b85ec38fdb0eb5b00faeaa0611cd6a483d3"],
    [277316, "0x00000016a20503fe496e79d34fb85c33f633059315c046ffa1b4826d08a1e856"],
    [483849, "0x000001eb7f8124282ab62296e63d3145ff6c84cf18afae4d4b8e02cd3182b6a8"]
  ] # [['   0', "0x0"]]
  clientname: "Satoshi" # "Satoshi"
  coinbasematurity: 100 # 100
  coinlcname: "chaincoin" # "bitcoin"
  coinbsname: "chaincoin" # "bitcoin"
  coinname: "Chaincoin" # "Bitcoin"
  coindisplayname: ""Chaincoin" " # "Bitcoin"
  coinucname: "CHAINCOIN" # "BITCOIN"
  coinsymbol: "CHC" # "BTC"
  coinunit: "Chaincoin" # "Bitcoin"
  coinlcsymbol: "chc" # "btc"
  cointype: "0x80000005" # "0x80000005"
  proofofworklimit: "CBigNum(~uint256(0) >> 20)" # CBigNum(~uint256(0) >> 20)
  psztimestamp: "18-01-14 - Anti-fracking campaigners chain themselves to petrol pumps" # "The Times 03/Jan/2009 Chancellor on brink of second bailout for banks"
  defaultminrelaytxfee: 1000 # 1000
  genesisblockhash: "0x00000f639db5734b2b861ef8dbccc33aebd7de44d13de000a12d093bcc866c64" # "0x000000000019d6689c085ae165831e934ff763ae46a2a6c172b3f1b60a8ce26f"
  genesisoutputscript: "04becedf6ebadd4596964d890f677f8d2e74fdcc313c6416434384a66d6d8758d1c92de272dc671 3e4a81d98841dfdfdc95e204ba915447d2fe9313435c78af3e8" # ""
  hashmerkleroot: "0xfa6ef9872494fa9662cf0fecf8c0135a6932e76d7a8764e1155207f3205c7c88" # "0x4a5e1e4baab89f3a32518a88c31bc87f618f76673e2cc77ab2127b7afdeda33b"
  npowhumantargetspacing: "90 seconds" # "10 minutes"
  npowhumantargettimespan: "90 seconds" # "2 weeks"
  maxmoney: 30000000 # 21000000
  mintxfee: 10000 # 10000
  msgstrts: ["0xa3", "0xd2", "0x7a", "0x03"] # ["0xf9", "0xbe", "0xb4", "0xd9"]
  nbits: "0x1e0fffff" # "0x1d00ffff"
  nnonce: 2099366979 # 2083236893
  nsubsidy: 16 # 50
  npowtargetspacing: 90 # "10 * 60"
  npowtargettimespan: 90 # "14 * 24 * 60 * 60"
  ntime: 1390078220 # 1231006505
  nvalue: 16 # 50
  nversion: 1 # 1
  port: 11994 # 8333
  rpcport: 11995 # 8332
  prefixnum: 28 # 0
  scriptaddress: 4 # 5
  scriptpubkeyhex: "0x" # "0x"
  secretkey: 156 # 128
  subsidyhalvinginterval: 700800 # 210000
  vseeds: [
    ["seed1.chaincoin.org", "seed1.chaincoin.org"],
    ["seed2.chaincoin.org", "seed2.chaincoin.org"],
    ["seed3.chaincoin.org", "seed3.chaincoin.org"],
    ["seed4.chaincoin.org", "seed4.chaincoin.org"]
  ] # []
  xpub: "(0x02)(0xFE)(0x52)(0xF8)" # "(0x04)(0x88)(0xB2)(0x1E)"
  xprv: "(0x02)(0xFE)(0x52)(0xCC)" # "(0x04)(0x88)(0xAD)(0xE4)"
  regtestcheckpointestimatedtransactions: 0 # 0
  regtestcheckpoints: [] # [['   0', "0x0"]]
  regtestcheckpointtimestamp: 0 # 0
  regtestcheckpointtotaltransactions: 0 # 0
  regtestnpowhumantargetspacing: "90 seconds" # "10 minutes"
  regtestnpowhumantargettimespan: "90 seconds" # "two weeks"
  regtestmsgstrts: ["0xfc", "0x1f", "0xc3", "0x56"] # ["0xfa", "0xbf", "0xb5", "0xda"]
  regtestproofofworklimit: "CBigNum(~uint256(0) >> 1)"
  regtestngenesisblockhash: "0x" # "0x0f9188f13cb7b2c71f2a335e3a4fc328bf5beb436012afca590b1a11466e2206"
  regtestmerkelroothash: "0x" # "0x4a5e1e4baab89f3a32518a88c31bc87f618f76673e2cc77ab2127b7afdeda33b"
  regtestnbits: "0x207fffff" # "0x207fffff"
  regtestnnonce: 3 # 2
  regtestntime: 1296688602 # 1296688602
  regtestnvalue: 16 # 50
  regtestnversion: 1 # 1
  regtestpubkeyaddress: 80 # 111
  regtestscriptaddress: 44 # 196
  regtestsecretkey: 208 # 239
  regtestsubsidyhalvinginterval: 150 # 150
  regtestnpowtargetspacing: 90 # "10 * 60"
  regtestnpowtargettimespan: 90 # "14 * 24 * 60 * 60"
  regtestport: 18444 # 18444
  regtestrpcport: 18445 # 18332
  regtestpub: "(0x04)(0x35)(0x87)(0xCF)" # "(0x04)(0x35)(0x87)(0xCF)"
  regtestprv: "(0x04)(0x35)(0x83)(0x94)" # "(0x04)(0x35)(0x83)(0x94)"
  testnetcheckpointestimatedtransactions: 960 # 300
  testnetcheckpoints: [
    [0, "0x0000082f5939c2154dbcba35f784530d12e9d72472fcfaf29674ea312cdf4c83"]
  ] # [['   0', "0x0"]]
  testnetcheckpointtimestamp: 1388868139 # 1337966069
  testnetcheckpointtotaltransactions: 0 # 1488
  testnetgenesisblockhash: "0x0000082f5939c2154dbcba35f784530d12e9d72472fcfaf29674ea312cdf4c83" # "0x000000000933ea01ad0ee984209779baaec3ced90fa3f408719526f8d77f4943"
  testnetmerkleroothash: "0x" # "0x4a5e1e4baab89f3a32518a88c31bc87f618f76673e2cc77ab2127b7afdeda33b"
  testnetnpowhumantargetspacing: "90 seconds" # "10 minutes"
  testnetnpowhumantargettimespan: "90 seconds" # "two weeks"
  testnetmsgstrts: ["0xfb", "0xc2", "0x11", "0x02"] # ["0x0b", "0x11", "0x09", "0x07"]
  testnetnbits: "0x1e0fffff" # "0x207fffff"
  testnetnnonce: 423087994 # 1211565
  testnetnpowtargetspacing: 90 # "10 * 60"
  testnetnpowtargettimespan: 90 # "14 * 24 * 60 * 60"
  testnetntime: 1388868139 # 1374901773
  testnetnvalue: 16 # 50
  testnetnversion: 1 # 1
  testnetport: 21994 # 18333
  testnetrpcport: 21995 # 18332
  testnetpubkeyaddress: 80 # 111
  testnetscriptaddress: 44 # 196
  testnetsecretkey: 216 # 239
  testnetsubsidyhalvinginterval: 150 # 210000
  testnetvseeds: [] # []
  testnetpub: "(0x3a)(0x80)(0x61)(0xa0)" # "(0x04)(0x35)(0x87)(0xCF)"
  testnetprv: "(0x3a)(0x80)(0x58)(0x37)" # "(0x04)(0x35)(0x83)(0x94)"
  testnetcointype: "0x80000001"


That pretty much does the main job. Not sure whether I can get enough abstraction to do the hash algo swapping by template, perhaps just a dynamically-computed patch for now.

I s'pose I could ultimately publish FerrariBoat on the Datacoin blockchain in the form of a patch against a clone of a given Bitcoin Core version.

Cheers

Graham
 

I wouldn't know. Their text reads like it's been generated by a Markov chain:

“Money should be exchanged freely and safely from person to person, and currently the most practical method for it is a central exchange. Such centralized solution requires the funds to be exchanged into IOU’s that are backed by the service provider. Thus the clients are under a constant risk of their assets being stolen either by an inside theft or an outside hack. In order to remove these risks, a decentralized alternative must be established.”

Cheers

Graham

Alas no, Slimcoin doesn't accept batch RPC commands.

Cheers

Graham

I'm still providing some pro tem technical support to the Helium team, if that counts.

Cheers

Graham

Maybe Sunny has relaxed his grip, maybe he's suffering from acute NIH. Someone did contribute a strict-DER fix:
https://github.com/coinkeeper/2015-06-22_18-45_primecoin

Perhaps it was just inadequately advertised. It does seem odd that it wasn't integrated into the “main” development effort, perhaps it fell short of Sunny King's standards.


ISTM that there were some quite important changes, especially between the 0.9 and 0.10 versions, which basically render the latter as the minimum acceptable version according to contemporary perceptions of robustness and vulnerability. AIUI, there was general developer relief at getting 0.10 safely out of the door as it addressed various previously-unmentionable issues, some of which might not unreasonably be perceived as vulnerabilities to exploitation.

Cheers

Graham

Looks more like someone at U of Georgia is finding their way around git.


Dunno, it's in the code.

Added to issues list: https://github.com/gjhiggins/datacoin/issues

Cheers

Graham

From the Slimcoin white paper:


The Dcrypt algorithm is an algorithm that uses SHA256 internally. What makes this algorithm FPGA/ASIC resistant is they way it was designed.

It is unknown beforehand how many internal SHA256 hashes will be calculated
It is unknown beforehand how much system memory will be required
Large amounts of memory IO is required
Each next step of the Dcrypt algorithm requires the result of the previous step, making parallelization difficult/impossible
The final step of Dctypt is to SHA256 hash a large generated piece of data of variable size


There are references around to a PDF white paper on Dcrypt itself but not even archive.org has a copy :-(

Cheers

Graham

Thanks for the positive response. It's not a wallet, it's a web app talking to the wallet. The visual aspects are courtesy of the Semantic-UI CSS framework which I omitted to credit. You can toggle between viewing metadata for mainnet and testnet by clicking on the coinlogo+name in the navbar (an easter egg).

Cheers

Graham

(edit: added URL, clarified)