an open source model code: osd 2002

[Rod Dixon]

Thanks David. Your points are helpful. We will add a definitions section
and include a definition for free software. With regard to the model code
for Article 2, we would like to amplify the meaning of source code in a
practical sense rather than a conceptual sense. In doing so, it might be
helpful to state that the source code should be in a form that others can
actually use. Hence the requirement that the code be doucmented or well
documented. As you state, there is a good deal of code that is not only
poorly documented, but extremely cryptic. If we don't set some minimal
standard here, then the "open" source code requirement is form over
substance. What good is the source code if programmers can hardly read it?
In some respects, the argument that "source code is speech" depends on the
assumption that source code not only expresses ideas, but is INTENDED to
do so. As we think about enhancing the OSD, whether the source code
provided by the licensor is provided in a form that demonstrates the
source code was written not to just run machines, but to express the ideas
it contains so others might modify the code for innovative purposes should
be an important factor for OSD compliance. I agree that providing well
documented code is one way to achieve that goal and, perhaps, it is an
onerous objective. what I am unsure of is whether we should be satisfied
with the status quo or whether we should amplify Article 2 with something
more than just saying the source code should not be deliberately
obfuscated. Poorly expressed source code need not be deliberately
obfuscated to end run the objective of what it means to provide "open"
access to the source code. Agree?

Rod




On Wed, 22 May 2002, Charlie Root wrote:

> Rod Dixon wrote:
> >
> > Please take a moment or two to download a draft of the framework for our
> > work on the OSD. We have only posted Article 1. We would like to hear your
> > thoughts on the framework. It is our view that a model code is the most
> > helpful framework for augmenting the Open Source Definition (OSD), but we
> > would like to hear what the folks on license-discuss have to say.
>
>
> 1-2.1: The terms "free software" and "open source" are intermixed. There
> is nothing inherently wrong with this, but the term "free software" has
> not been previously defined.
>
> 2-2: "The code should be well documented." I hesitate at this one.
> Although I strongly favor well commented source code, there are
> innumerable examples of poorly documented source code in the Open Source
> community. It would be very hard to distinguish code that is poorly
> documented because the writer is ignorant, lazy, or arrogant, from code
> poorly documented as an excercise in obfuscation.
>
> David
> --
> license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
>

--
license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3