Liquid: the first commercial sidechain based on Strong Federation.

By Gabriele Domenichini

On November the 22nd 2016 in Odets room Marriott Marquis Hotel Nyc Blockstream presented Liquid Beta, a sidechain built in collaboration with some important exchanges. Liquid is based on the Strong Federation Sidechain concept where a group of “functionaries” produce blocks for the network.

In this kind of sidechain the participants are:

  • the users of the system
  • the block signers who validate transactions and sign the blocks
  • the watchmen who allow funds to be securely transferred to and
    from the chain by the pegging in and out

The last two roles will be played by exchanges who are signed up to
Liquid, using specialized HSM devices in protected environments,
communicating via the Tor network.

The components of Liquid and their interactions

We saw three operation during the demonstration:

  • a normal Liquid transaction
  • a Confidential Transaction
  • The verification of the details of the operations on the “Liquid explorer”

The demonstration was smooth and interesting. We first saw some
satoshis sent from one address to another other address and confirm in
less than a minute.

Following this we saw the same process with a Confidential
Transaction. The main visible difference is that the address is
slightly different in format and of course the result of what data is
visible on the network.

Finally the two transactions were shown on the Liquid explorer which
is pretty similar to a normal block explorer. On the screen we could
see the scrambled amounts of the Confidential Transaction maintaining
user privacy.

Many questions were left open, probably on purpose. One of these is
when we will be able to play with Confidential Assets that would allow
the user to send any kind of issued asset on Liquid, hiding the amount
and the asset type involved in the transaction. We did get to see a
demonstration of GreenBits — the GreenAddress Android wallet — showing
a currency asset and with confidential receiving addresses.

On the right the Android Emulator with the balance and the transactions in Pounds

The message that I got is: technology is almost ready. Time to figure out use cases and to experiment.

Much of the future of Liquid is bound to the kind of access the system
will allow to normal users. At the present moment, it seems that the
users will have to be exchange customers, so KYC will probably be
required to operate on the network. It is to be noted though that this
is not a requirement imposed by the technology at all.

To use the system, a user sends bitcoin with an enabled client to a
particular address. This address is prepared by the client with
another associated “Liquid address” which is also controlled by the
user. After the complete settlement of the transaction (some Bitcoin
network confirmations), the amount will appear on an output associated
to the Liquid address.

From that moment the user can send the money to any other Liquid
address with either a normal or a Confidential Transaction.

After the transaction has been broadcast, confirmation (settlement) is
pretty fast: The blocks are signed every minute and they can be
considered safely settled with one confirmation (This is because there
are no chain re-orgs in Liquid, unlike Bitcoin).

If Alice has money on Liquid she can send them to an address assigned to her on an exchange, pretty much like it happens today when we deposit bitcoins on an exchange. The difference is that on Liquid the transaction is considered settled in less than a minute.

This opens interesting opportunities for arbitrage (especially in periods of network congestion).

Moving funds among exchanges for arbitrage opportunities has never
been easy. Today arbitrageurs are compelled to keep a lot of money on
different exchanges to take advantage of the opportunities given by
price discrepancies. This represent a huge risk for exchanges and
traders. Bitcoin transaction are not so cheap, not so fast and not so
certain in confirmation times. Fiat transactions require days and are
not certain in settlement time either.

In this landscape Liquid represents a big innovation because with one
block every minute, no re-orgs and the certainty of having the
transaction confirmed in the next block, it is no longer necessary to
keep large stash of bitcoin on different exchanges.

As far as fiat money is concerned (EUR, USD, YEN etc.) Liquid could
become extremely useful also. The new Confidential Assets technology
brings the privacy of Confidential Transactions to any type of asset
issued on Liquid. Users from outside will see only that a transaction
has happened (and the addresses involved) but the amount and the type
of the assets will be hidden by cryptography while still being able to
be verified by the network.

Regulatory matters apart, it could be possible to transfer any kind of issued asset in a fast, private and secure way to any user of Liquid. Also it could be possible in the future to compose atomic transactions in which fiat money would be exchanged on the fly with bitcoins or maybe with other assets.

As it was implied during the presentation, it is possible to imagine a
future in which the activity of exchanges will be limited to the bid
and ask matching service that is their core competency and maybe to
the construction of the atomic swaps that would exchange fiat for
bitcoin. The custody of the funds has always been the source of their
biggest problems.

As Phil G Potter CSO of Bitfinex has stated in the meeting, level two solutions already exist. The most evident example are centralized wallet service providers like Coinbase and Xapo but also exchanges can be seen as level 2 bitcoin layers.

Liquid is a new form of second layer in which the user relies on multiple parties that are unlikely to collude and that maintain a network in which they don’t control the content because it is mainly hidden to them.

If the access of the system will not require too many bureaucratic hassles, Liquid will become a valid alternative in periods of network congestion for transactions of confidential amounts in an environment which is not trustless but which is neither completely centralized.

Thanks to Dr. Adam Back for reviewing and approval and to Jon Griffith for the huge effort of correcting the first version of this post.

Useful resources

http://www.coindesk.com/its-happening-blockstream-launches-liquid-sidechain-in-beta/

Disclaimer: The family office for which I work owns shares in Blockstream and this post doesn’t represent an official publication on the matter but just my takes as attendee at the demo.