Mark Ramm wrote: > Just my 2 cents, but this approach seems a bit heavy-handed to me. > > Do you really think that *none* of the other Python ORM's provide any > of the features listed here? If so, I think you may be missing one or > two projects. ;)
Of course other Python ORMs provide some of the features listed here. I don't think any of the others meet *all* the points listed here though, and if they do the list should be extended as the goal is to point out how come Storm is different and has a reason for existing. And if such a list cannot be made, it really shouldn't exist as it would just be a reinvented wheel. The front page should give developers enough information to decide if Storm might be right for their project or not. It is also good to list things that Storm doesn't do, so perhaps * Does not generate your database schema. Which is either a positive or negative depending on your POV, so we cannot categorize such a list into positive and negative features. Indeed - Storm's ability to work easily with multiple datastores could be seen as a negative, as it generates a little complexity and abstraction that is not needed for the vast bulk of projects out there. Also 'not dictating your data model' - some people love object databases because they don't have to do all that 'icky database design stuff, and an ODB or an ORM that generates a schema can be great for small projects that have no need to invest time is designing their storage layout. > I'm hoping that we can avoid some my "ORM is better than yours > flamewars" (perhaps initiated by people out side of the Storm > community) and actually have a discussion of relative technical > merits. And the "nobody else has ever done this" tone of this > message seems a bit incendiary, so I'm hoping it gets a bit of editing > before it becomes the "official" marketing message. Sure. I tried to be non-incendiary, but posted it to the list so it can go through the 'political filter' rather than editing the front page directly so wording changes and suggestions are fine by me :) It would be extremely bad form to belittle other projects for Storm's benefit, particularly since it is really standing on the shoulders of earlier projects (even when these projects showed what *not* to do in some cases - if they hadn't made those design choices Storm devs might have and shot themselves in the foot). However, Storm does need to point out what it can do or it will just be ignored, with devs not even bothering to download it when evaluating tools for a project. At the moment, it just tells people it is 'an object-relational mapper for Python' which is as nearly content free as possible :-/ Wearing my Python developers had, if I was evaluating ORMs and wasn't using external recommendations, I would not get beyond the front page of Storm - it would fall of my radar after one click. Which is all extremely non-incendiary and non-confrontational, but not really fair to Storm which happens to rock. IIRC Storm was developed and adopted because one of the Canonical projects needed a feature that just did not exist in anything else *right now*, and it was determined that it was quicker to implement the basic ORM with that feature that integrate it into any of the other projects that where looked at. And given how quickly the basics where implemented and integrated into that project, I think that the call was right. Should something like this (or the true account IIRC incorrectly) be there too? From other messages in this thread, we should also list: * Pure Python, running everywhere Python does (Linux, Windows, Mac OSX, and many, many more). (I'm pretty certain all the existing ORMs meet that, and they should all mention it because Python rocks too). -- Stuart Bishop <[EMAIL PROTECTED]> http://www.canonical.com/ Canonical Ltd. http://www.ubuntu.com/
signature.asc
Description: OpenPGP digital signature
-- storm mailing list [email protected] Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/storm
