Web Client

Not a lot of people know but Bitcoin for most nodes is not the graphical Qt version showing your wallet balances and transactions. Instead most nodes run the daemon or server version. This version runs in the background and is communicated by using RPC commands on the commandline or through an application.

With MicroCash I want to have the server actually be connectable through a graphical web client. This way I can drop support for the MicroCash C# client which I think just added a lot of time to Realsolids development schedule. He told me the reason for that client was that others could make it while he made the MicroCash server but instead he was left with both projects.

To allow web browsers to connect to a MicroCash server requires a bit more code but not too much. After all, the RPC code already is pretty much like a webserver. I think it will actually be fairly efficient to convert Realsolid’s server code to support this because he already had a wallet system in Microcash server that was accessible over the command line or RPC. All I need to do is hook that up to a graphical web UI and call the same functions. However since I haven’t yet migrated his wallet code into my server this is still something I have to do.

With this web browser support it means MicroCash will nearly be platform independant even if the server is bound to a few platforms (Windowns or Linux servers). I still need to have a talk with Realsolid about this but I believe I may be able to do a brain wallet that would allow a web browser to use an untrusted node as a buffer to the MicroCash network securely. This would mean all clients would be so called “light clients” only requiring a user to remember a password to access his funds from anywhere in the world. There are a lot of ideas about this that I still have to put together into an actual framework but it is exciting when it comes to actual use cases for MicroCash.

Advertisements

2 thoughts on “Web Client

  1. Cluck says:

    hi michael,

    good to see that finally someone picked up the development of MC again!
    after reading thru all your blog posts, I now have some hope that the MC project could eventually get finished…

    back in november 2013, I created some mockups for a possible Microcash Web Client. At that time I realized that a lightweight, javascript based web client would make more sense than to begin with a fat C# based client. please find a pdf document here: http://bit.ly/HJrSmI

    the web client was planned to be based on ExtJS (sencha.com/products/extjs). today I would rather go with meteor.js (meteor.com).

    to securely store client-side data I had plans to use: remoteStorage (https://remotestorage.io) => so only the user decides where he wants to store sensitive information like private keys, wallet addresses, etc.

    at that time I created multiple small prototypes and proof-of-concepts (based on extjs 4), plus also native prototypes for iOS / Android / Windows Phone devices, using pure C# and Xamarin tools. The prototypes were far from being complete – I just wanted to test some mission critical aspects in isolation, like: communication with the MC server (RPC), client side crypto-stuff, client-side storage, etc.

    unfortunately, as always, RS was very offending, aggressive and bigheaded – so I decided to drop everything and stopped my commitment to MC…

    hope it helps
    cheers

    cluck (IRC: @thecluck)

    Like

  2. Thanks for the documents and information thecluck. I also think the C# client is not needed at this stage but I also think it would be pretty easy to develop clients in any language once MicroCash itself is finalized. My web client idea is actually about remotely connecting to a node and controlling it that way. A litejavacript client in the future would be very nice though!!

    Programmers can often be offensive so don’t try to take it too seriously, thanks again!

    Like

Comments are closed.

%d bloggers like this: