News

A Telegram group and a Discord channel have been created by muf18. They are not currently bridged to the official IRC channel, though it is under consideration. As a reminder, the IRC channel is bridged with the Matrix network which has many clients.

The ercoin_abci module (over 2000 lines of code) has been split into smaller ones. This will hopefully make the code more accessible to new developers.

Validators now won’t get full rewards if an epoch lasted longer than expected (this is introduced to eliminate an incentive to delay blocks and make an epoch last longer, covering more transactions).

Calculation of genesis data has been documented and implemented. The browser wallet supports creating genesis transactions.

A new transaction type for burning coins has been introduced. Also balances of expired accounts are now destroyed instead of being put into long fee deposit. This makes Ercoin a slightly deflationary currency.

Tendermint version has been upgraded to 0.20.0.

ercoin.tech is now generated using Zodiac instead of Hakyll. The main reason for that change is that there was no caching of Hakyll compilation between CI builds and as a result they used to last very long. Zodiac is a much more lightweight solution, as it depends on Make, AWK and sh instead of 3.5 GB of Hakyll binaries.

The browser wallet has become an official one and it now supports lock transaction type.

Paspagon (file payment processor which had been supporting BlackCoin exclusively) has suspended services.

Development news:

The GUI wallet now supports persistence, transaction creation (for transfer and account transaction types) and communication with Tendermint (using RPC). It can therefore be used to play with an Ercoin node.

It is now easier to run a node for testing purposes, as IPFS is no longer used to bootstrap initial state and also there is Docker support. If Docker Compose is installed, then starting a node (consisting of ABCI server and Tendermint as ABCI client) is just a matter of executing a few commands.

Tendermint has been updated to 0.17.1.

Ercoin now supports persistence, hence restarting a node doesn’t mean replaying all transactions. State is dumped to disc at the beginning of each epoch.

Note that currently the GUI wallet encodes keys using Base58, while the command line wallet uses hex encoding.

Development news:

Fee deposit mechanism has been changed. Deposit has been divided into short (fees paid for current processing) and long (fees paid for long term storage). At the end of every epoch, short fee deposit and part of long fee deposit are divided between those validators that have less than 2/3 of absencies in block signatures, proportionally to their voting power. The sharp threshold intends to disincentivize censorship attacks. The whole mechanism allows to bypass/diminish issues with putting proposer into block header and injecting entropy into Tendermint. See comments in issue 952 in Tendermint for details.

Despite to what https://beacon.nist.gov/home says, version 2.0 of the NIST randomness beacon has already been relesased. I’ve submitted a merge request to Ercoin preparing ground for planting beacon-generated values.

There have been new versions of Tendermint (and ABCI) released. The newest one is 0.15.x (Ercoin currently runs on 0.14.x).

Ercoin now has continuous integration.

Issue related to Triq plugin in erlang.mk has been fixed (in Erlang; fix is not available in any of the released versions yet). I’m willing to pay part of the relevant bounty.

I’ve started writing a simple GUI wallet in ClojureScript. Source code is currently available at https://gitlab.com/KrzysiekJ/ercoin-wallet. Every commit is automatically built and deployed to https://krzysiekj.gitlab.io/ercoin-wallet/ (so there is already a working demo). For now, there is only account (keypair) management, without persistence, without transaction handling and without actual communication with the node (which will use WebSocket endpoint provided by Tendermint).

Some development-related news:

Tendermint has recently added a roadmap to its repository. The first two entries are the issues that Ercoin is particularly interested in:

• “Upgrade the header to support better proves on validtors, results, evidence, and possibly more”. Relevant TODO: https://gitlab.com/Ercoin/ercoin/blob/bbc1297188da73e95c0a46196ae1a4475c8877c7/src/ercoin_abci.erl#L1728
• “Better support for injecting randomness”. Relevant issue: https://github.com/tendermint/tendermint/issues/574

The issue with Triq plugin for erlang.mk has been diagnosed. The source of the problem occured to be in Erlang itself. Bounty has been extended until November 1st, giving precedence to the person that debugged the issue.

NIST beacon page still states that “Version 2.0 of the NIST beacon is planned to be deployed in the Summer of 2017.” Not much time left then. Relevant TODO: https://gitlab.com/Ercoin/ercoin/blob/bbc1297188da73e95c0a46196ae1a4475c8877c7/src/ercoin_abci.erl#L561

Yes, this is even tested.

A simple website is now live: http://ercoin.tech. It is crude on purpose, as its aim is to provide information and it’s targeted primarily for potential contributors, not for broad audience yet.

Look at the top of this topic.

I’ve created a non-binding poll about which domain should Ercoin use for its website (which doesn’t exist yet).

I’ve established two bounties for Ercoin-related development, paid in BlackCoin exclusively:

• 120 BLK for adding a LFE lexer to Rouge. This will add syntax highlighting to LFE modules in GitLab and GitHub Pages.
• 125 BLK for fixing a bug with Triq plugin in erlang.mk.

See the relevant issues for details.

I’ve established two bounties for Ercoin-related development:

• 120 BLK for adding a LFE lexer to Rouge. This will add syntax highlighting to LFE modules in GitLab.
• 125 BLK for fixing a bug with Triq plugin in erlang.mk. This will ease testing.

See the relevant issues for details.

BlackCoin is the first 100% proof of stake, free cryptocurrency with fair distribution (no entity privileged: no premine, no ICO).

First pull request to Ercoin was submitted (some minor commits have also been pushed directly to the master branch). It adds apportionment modules written in LFE (Lisp Flavoured Erlang), hence should be more-or-less understandable to people familiar with Lisp syntax.

BlackCoin is the first free, 100% proof of stake cryptocurrency with fair distribution. It is still easier to obtain BLK than to buy mining hardware needed to participate in Bitcoin distribution.


I don’t see how initial coin distribution via burning multiple coins could be reasonably done. Let’s say that one user burns 10 BTC and another burns 10,000 BLK. How do we compare these two values?


From a coarse perspective, there may be some similarities, but mechanisms are fundamentally different, starting from the fact that there will be no concept of unspent transaction outputs in Ercoin.


Erlang in itself may not justify making a new coin, but it has some features which make it an attractive language for such purposes. Among them are:

• Great support for concurrency. A cryptocurrency network is a set of many nodes that send themselves messages asynchronously. This can be naturally modelled in Erlang. (Actually in case of Ercoin Tendermint mostly takes care of concurrency).
• Libraries for creating state machines. A cryptocurrency is basically a decentralized state machine.
• Bit syntax, making it easy to handle protocol messages.

Erlang is a functional language and there are other languages compiled to Erlang’s virtual machine, like LFE and Elixir (LFE is actually used in Ercoin). It is also praised for its fault tolerance and ability to achieve high uptime (with hot code reloading), which may have some significance for maintaining network nodes.


Roughly speaking, to have a working network suitable for pre-production testing:

• a few missing things need to be implemented;
• there are some changes needed in Tendermint;
• NIST should publish a new API for their entropy beacon.

There will be a time window (for example, blocks 1,800,000 — 1,850,00) scheduled during which people will be able to destroy their blackcoins to obtain initially distributed ercoins, proportionally to the destroyed amount. Total coin supply is not chosen yet. Marketing is yet to be worked out, though I suppose it will concentrate on features, efficient economic model and simplicity.

By fair distribution I mean a distribution in which no entity is privileged. Quoting the README: “Initial amount of coins will be distributed to addresses proportionally to the associated amounts of BlackCoin burnt in a selected time window using OP_RETURN.”. BlackCoin already has a command for creating OP_RETURN outputs. I’ve created a simple program for listing such outputs in a selected block range.

For some background and further information, you can read a post to the Erlang mailing list. In case of questions, feel free to ask.

The name “Ercoin” is derived from Erlang in which it is written. I don’t recall Erlang being associated with errors due to its name (actually the language with its libraries is appreciated for fault tolerance), though I’m not a native speaker to judge it well. Picking another name can be considered.

I’ve started a separate thread for Ercoin. An IRC channel has also been created: #ercoin on irc.freenode.net.