I think the gist here is interest in 'sugar' to ease Leo's learning curve. To implement this stuff in core is a lot of work, Edward has other concerns. There seems to be a theme of creating more structure for the data which provides configuration, currently it exists as strings in trees which Leo understands, involves users learning to think like Leo. Instead of core Leo learning to translate to natural language, might a tool provide translation layers like: Leo source tree -> leo.cfg and: Leofile.leo -> leobridge -> leo.cfg which would scan and extract mentions of config(xxx), and current values, putting them into well known structure, like json. leo.cfg could provide a source of documentation, and a lingua franca configuration format which, once edited, would be written out to the leoSettings.leo format. It seems doable. If so, it could mitigate the current learning curve without impacting flexibility, and without needing to touch core Leo. Thanks, Kent On Mon, Sep 16, 2013 at 1:12 PM, Edward K. Ream