John

I think we're now debating- nothing to gain from that imho.

On a technical level:

>>The issue is also about manageability. Imagine a remote view that is composed of multiple joins. You now need to update multiple tables. This is a bear - if not downright impossible. <<

What do you mean, "need"? I know of no requirement to update multiple tables just because they are in a joined view. You can update just one (or none) if you so choose.

Equally, if you "choose" to update multiple tables in a remote view... I'm not sure why issuing Tableupdate() is so "impossible". Please explain.

>>The issue is also about security. You may be forbidden to render client-side SQL. i.e., you may be forced to deal exclusively with SP's.<<

I do have clients who demand SP- mainly against Oracle, though. But in response to your point: one might equally say, "the client may demand that you use Cache". Your answer to that was "I don't know about Cache and have never heard of it". Perhaps I should reply "I don't know about SP and have never heard of it" as if that supports my position. < bg >

With web or http apps, such "security" issues are far less important- the client can't access the data directly anyway. Not that I would criticise others who have not discovered the advantages of such methods.

So: it seems that apart from declarations of "better performance" from use of proprietary methods, you have little else for a competent developer to assess. No "real life" performance stats, nothing.

It also seems that having decided to interface with backends in a certain way because you perceive a performance advantage, and having labeled VFP "lousy" for not doing same, you are disinterested in even looking at alternatives that may offer even more advantage in your terms. I'm afraid that casts doubt on your claimed pragmatic reasoning.

What you are really saying is that you have chosen to use a certain mechanism to interface with certain backends. That's fine, but if you couple that with absolute criticism of those who use a different method while ignoring options that may provide even more advantage in your chosen area... well, don't be surprised if not everybody is impressed with your reasoning.

Regards

JR