mca blog

in a recent post, Soumitra Sengupta gives me a nice shout-out over my recent SSDS Examples coding and offers some comments in reply to my recent post regarding Astoria and SSDS. first, thanks to SS for his responses. always nice when i get the opportunity to 'talk' (even only virtually) to some of the folks at MSFT that are making cool things happen.

a key thread in our on-line exchanges has to do w/ getting the Astoria libraries working with SSDS. while SS calms my concerns over any loss of flexibility, i still have a few issues with linking Astoria and SSDS. mainly, that SSDS not become 'a hostage' of the Astoria library.

having earned my stripes using Dynasets (remember?), i've seen my share of DAO,RDO,ADO,ADO.NET, etc. Astoria is just another one in the line. SSDS, however, strikes me as quite different. SSDS is the cloud. SSDS is anytime/anywhere/anyclient. i can use HTTP to do all my SSDS work. the same HTTP that was around when Dynasets were first invented. and i expect to use HTTP to access my SSDS data ten years from now - long after Dynasets and (probably) Astoria libraries are long gone.

SSDS is not a library and i want so very much to make sure that linking SSDS to Astoria, or any other code library, does not change the nature of SSDS. as long as i can always get my data, i'll be pleased. heck, i might even give this new-fangled data-library a try!