[Grok-dev] a nice research project

Chris McDonough chrism at plope.com
Sun Feb 8 12:20:48 EST 2009

Sebastian Ware wrote:
> 8 feb 2009 kl. 05.13 skrev Chris McDonough:
> Chris! I might have read too much into your post. I kind of jumped the
> second half of the first paragraph and then you lost me on the second
> paragraph. I apologize for spicing my reaction, and I am sure you have a
> great framework, no doubt.

No worries.

> What does worry me however is that one would feel compelled to spend a
> lot of time making a "hello world" app look good in your benchmark when
> that sample code doesn't represent a real world application.

Lots of people have made this point to me over the last day.  Note that the
original blog post made no judgment about how easy it is to write *an
application* using any of the given frameworks.  Instead it calls out how much
work it might be to *extend* or *fix* the *framework itself*.

I would think this concept would be especially attractive to Zope people; we
have a ton of experience in this field.  We spend a *lot* of time maintaining
frameworks.  Do we all really want to continue swapping lots of code into our
heads to fix bugs in the framework or to extend it?

> Maybe one could create a better standardised sample app that is more
> representative of the real world and use as a means to compare different
> metrics that could be applicable to real world use (such as
> performance/scalability and debugging).

Maybe, although, it won't be me, as that wasn't what I was trying to measure.


- C

More information about the Grok-dev mailing list