There is a new nxt-1.2.8-merge branch.
Please test the new client at testnet:
http://woll-e.net:9876/You can also download, compile and run the client yourself:
git clone
https://bitbucket.org/NFDcoin/nfdcoin.gitcd nfdcoin
git checkout nxt-1.2.8-merge
./compile.sh
echo "nxt.isTestnet=true" > conf/nxt.properties
./run.sh
For the moment please make sure that it will run only on testnet!
Awesome work! Your really doing a great job with the NFD software. Some of the best I've seen. Do you have any innovative features on your roadmap? I think a CFD or betting interface would be fantastic and quite strategic.
Can you please explain in more detail what do you mean with CFD and betting interface?
CFD = contract for difference. wiki it.
If you are familiar with Counterparty, then CFD are essentially features (not present in NXT, but easy to implement) that allow users to publish feeds (for a tx fee), for example, the price of Gold, and for other NFD users to place bets or wagers on the values of those feeds at predetermined dates. Example 1: someone might place a bet that the price of Gold will be below some value on November 9th. Others might wager against that bet. Example 2: users might bet on the outcome of a world-cup game.
Counterparty lacks a decent UI and reasonable block time, so users currently have to visit centalized websites to see and place bets. Totally lame!
Also, Counterparty requires feeds to be trusted. A user publishing a feed could bet on his own feed and publish fraudulent results that make him win the bet. NFD could eliminate this issue by making feeds trustless.
That is, the NFD feed is implemented as a module, which the feed issuer must upload. The module provides two methods: 1) consume() -- fetches some piece of live online data (e.g. curl a URL and extract a price) and records the feed value and feed timestamp in the block chain. 2) verify() -- fetches some historic data (at the feed value's timestamp recorded in NFD blockchain) for that same datasource (the feed designer must make sure the datasource is verifiable and at an official URL that the feed issuer does not control, e.g. yahoo finance). Block verification then requires validating the feed transaction by calling its validate() method. Every participant of the NFD network verifies each feed data item. Once issued the feed's methods cannot be changed. The feed is reviewed and if people consider the code trustworthy, then they can bet on it. If the remote data source changes (yahoo changes their URLS), then the feed will be broken. The NFD feed code should detect this and bets should be cancelled without harm. If a feed becomes broken, someone can write a new (updated) one and announce it. If the code is good and trustless (as can be), then people will bet on it.
I can see NFD easily implementing something like this, which would not be competing with NXT, but tapping into a new market and getting the NFD network very active.
This will surely put a huge spice in the NFD transaction volume and do great things for holders of NFD.
That's very interesting!
A feed module must be included in the bockchain, because NFD NRS is only one possible client. A feed module could be a kind of configuration.
The feed configuration needs at least data URL, which is the point of trust and maybe a regular expression to extract the data.
Historical data must be written to blockchain. This is an additional transaction for every datagram.
Questions to solve:
Who pays the fee for the historical data transactions?
How do the issuer participate? Does he/she get fees from CFDs?
If a feed is temporarly unavailable, then the CFD is closed?
The most problem I see are DOS/DDOS against the feed source.
