> > If you go through the links I have been pin-pointing here, you would find > many concrete examples of data-models and how they are used in SQLA's ORM.
I have gone through the links (not the full presentations, though), and I don't think I've seen anything that looked obviously easier than the DAL equivalent (though, maybe we could add an abstraction to simplify many-to-many relationships). > You are basically asking me to get much more proficient in the DAL, as > well as in SQLA, and to design the solution myself and present it. > Not really. It's a bit odd, though, that you would claim with such confidence that web2py needs an SA-like ORM without fully understanding the DAL's capabilities or any concrete notion about how such a hypothetical ORM would improve the development experience. A better starting point might be to say, "Here's something I can do quite easily with SA -- how can I achieve something similar with the DAL"? Then maybe we can discover specific areas where some additional abstraction on top of the DAL might be beneficial. > This is a valid expectation in a developers-group, not a users-group. The > whole point of sharing ideas and expertise, is that someone can suggest > a conceptual/abstract direction, and this entices someone else who is much > more proficient, to come-up with a concrete solution. > Yes, but even in the users group, it helps to have concrete domain examples and preferably even code examples. It sounds like you're saying, "Hey, I suspect that an SA-like ORM could make some types of data modeling a lot easier, so you experts please go out and investigate whether that is really the case and in what situations it is the case, and then build a solution that covers those cases." I'm not sure this is a sufficiently well-defined problem to entice anyone to come up with a concrete solution. > I am inclined to hoped I would not be needing to become an expert myself, > in order to propose an idea for improvement. > It doesn't hurt to propose ideas, but frankly, most new features are developed by someone who needs the feature, develops and refines it, and then offers it to the community (at which point it may then undergo further revision/refinement). It's not likely that a volunteer developer will be motivated to spend tens or hundreds of hours of his time because one person on the users list says he needs an ORM. If you want something to happen along these lines, it's likely you'll have to get at least somewhat involved beyond just making some very general suggestions. Anthony -- --- You received this message because you are subscribed to the Google Groups "web2py-users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/groups/opt_out.
