Daniel Lepage on Tue, 23 Jan 2007 07:30:40 -0700 (MST) |
[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]
Re: [s-d] proposal parser |
Here's a thought. What if rather than have a bunch of email-based tools, we bundled them together as plugins to a common interface? I'm envisioning something like this: B Nomic Standard Email Interface 1. People Any person can sign up for the service by choosing a username and a password, and specifying at least one email address. A person can add multiple email addresses if e wants to. When the interface receives an email, it will determine who sent it based on the sending address; if the address doesn't match any known address, then it looks for a password within the message to identify the user. 2. Security Each person can set their own security settings. Maybe e only wants it to accept email that comes from one of eir registered addresses AND contains eir password, or maybe e even wants it to rely on PGP encryption. By letting the user choose, you ensure that anyone who wants advanced measures like PGP can have them, but anyone who doesn't care as much needn't bother. 3. Modules There are a number of modules that the service can use. The chosen module can be specified within the subject of the message, so that "[Proposal]" in the subject line selects the proposal handler, "[Escrow]" selects the escrow agent, etc. Some modules would be restricted to certain users - for example, only Peter could start/ stop the clock through this system. 4. Headers The system looks for the delimiter "------" in the body of the message; everything before that line is parsed as parameters to the module, and everything after is treated as the text for the module. The proposal module could then look for a parameter along the lines of "Amend: True" to find out that this is an amendment to a previous proposal, for example, and the dice roller module can use these parameters to find out who to send the resulting message to. 5. Generality The key advantage of this is that it can be provide a common authentication framework for many different systems, so that we could add future modules to it easily and they'd known immediately how to determine who sent a message and how to parse basic parameters from the top of the message. Am I crazy, or would this work? -- Wonko _______________________________________________ spoon-discuss mailing list spoon-discuss@xxxxxxxxx http://lists.ellipsis.cx/mailman/listinfo/spoon-discuss