KISS

[Daniel Jalkut]

And, ergh. I guess I am sort of defying Brent's request to "brainstorm first, decide 1.0 later" … but I do mean this in the spirit of brainstorming. I.e. let's keep in mind and start to think about what kinds of features WOULD need to be baked in to the product itself, vs.  being able to be added as plugins of some kind.

Daniel

On Jan 16, 2010, at 6:07pm, Daniel Jalkut wrote:

> 
> Yes, smooches to all of you.
> 
> Keep it simple, stupid. I think that should be the mantra for this project, if it's going to succeed. I like the brainstorming that is going on now, because it's probably getting people excited about the prospects of such a project. But each and every feature that gets "accepted" into a plan for 1.0 will be a nail in the coffin of that ever being completed.
> 
> To this end, I definitely agree that extraneous features such as ToDo management, calendars, etc., etc. should be kept out of the project. But furthermore, even some relatively common expectations should also be left out: no filters, no signatures, etc.
> 
> The suggestion to support plugins is huge, and important to making this KISS plan realistic.  The "shared database" idea is also conducive.
> 
> 1.0 doesn't have to be actually useful as a replacement mail client in order for it to be a success. The most important thing 1.0 can achieve is to prove the product's own viability. I would suggest that in order to do that it should:
> 
> 1. Possess a robust, though not necessarily bulletproof IMAP implementation, capable of writing/reading/deleting from a significant percentage of popular IMAP servers.
> 
> 2. Establish a common database format for local caching of mail content, and implement a standard library for interacting with that.
> 
> 3. Establish a variety of plugin points for further extension of the feature set.
> 
> The idea would be that 1.0 aims to implement nothing that can be reasonably implemented by way of a plugin. In this way, the core of the app can be made available as soon as possible, and the features most desired by other developers will be filled in at a prioritized pace, in the form of plugins.
> 
> An extremely limited focus for the 1.0 gives it the best chance of success, and provides contributors with a really easy mantra to repeat as they barrel ahead towards that goal.
> 
> Daniel
> 
> _______________________________________________
> email-init at lists.ranchero.com mailing list
> List help: http://lists.ranchero.com/listinfo.cgi/email-init-ranchero.com