hi roy, > but how i see the current design of (micro)sling - there is not any > intention of 'storing' actions in the uri. of course a user can still > code some 'actions' in his script - but this would not be the fault of > sling. the 'selectors' are primarily used to resolve to a different > 'view' of the content, or to pass additional render information, e.g. > the dimensions of a resized image.
i completely agree with toby. i see selectors exclusively as a means to display the same logical content in different ways for example: as html, as pdf, as a small navigation-images, as an html "teaser" in a list, etc... generally, i think that selectors are generally only relevant for GETs anyway. from what i can see there is absolutely no intention to put http methods into a selector. i think we never did that, and i am 100% with you that this would be bad design. is there anything in particular that makes you think that there is a danger that this would happen in (micro)sling? maybe we were not able to articulate the intention of the script mapping when it comes to methods and selectors properly... regards, david
