[Schevo-devel] Re: Benefits of schevo vs. traditional RDBMS

[Patrick K. O'Brien]

Patrick K. O'Brien wrote:
> Tom Locke wrote:
> 
>>>The benefits haven't been documented, so I'll just rattle off a few:
>>>  
>>
>>ok - sounds good. And how about benefits vs. ODBMSs?
> 
> 
> Let's start by limiting ourselves to the ODBMS products available for
> Python: ZODB and Durus.  Both are fine products.  But they are pretty
> low-level in that they provide persistence for just about any kind of
> python object, or collection of python objects.  Exactly how those
> objects are structured is left up to each and every developer.
> 
> Schevo goes beyond those capabilities by providing structures that give
> you relational-like integrity, automatic indexing, a nice API for
> accessing database objects, transactions-as-objects (as opposed to
> transactions as milestones within procedural code), import/export
> capabilities, a mostly-declarative schema syntax, etc.

Forgot to mention a few things.  Schema evolution is another major
feature than is in the works.  Again, we'll make as much of that
declarative as possible, and provide nice APIs for the rest.  We also
know that an application requires a good interface.  So we have lots of
UI hints and/or API stuff specifically to ease the burden on UI widget
developers.  We also intend for Schevo to supply custom widgets for
popular toolkits such as Qt, Nufox, TurboGears, and anything else than
makes sense.

And, last but not least, Schevo plays nicely with PyCrust.  In fact,
Matt pulled a rabbit out of his hat when he came up with a way to launch
PyCrust (which uses wxPython) alongside the PyQt Navigator such that you
can interact with the database in PyCrust and all your changes are
immediately visible in the PyQt Navigator.  Now that's one cool trick if
I do say so myself!  :-)

-- 
Patrick K. O'Brien
Orbtech       http://www.orbtech.com
Schevo        http://www.schevo.org