Serge Knystautas wrote:
>
> Steve Brewin wrote:
> > Standards are the things which allow complex interactions
> to take place in
> > this big wide world of ours. I'm sure you would be upset
> most times if a
> > standard was not adhered to - think world wide web, human
> rights or your
> > neighbour choosing to drive on the opposite side of the
> road to the rest of
> > you. Standards matter.
>
> But as your first two examples perfectly demonstrate (world
> wide web and
> human rights), you're luck if 80% of the people follow that standard.
>
> For aged standards like SMTP and NNTP, I've found you need to follow
> pervaling practices as much as what's in the RFCs.

80% of people follow human rights standards. I wish!

Anyway, to the point. Agreed, we also have defacto standards where current
evolved behavior is yet to be enshrined as a new 'standard'. This may be
driven by the community or be dictated by a monopoly. We have to handle
this, sometimes by kicking back othertimes by accomodation.

Circumstances dictate the correct approach. I was simply saying that
supporting standards is not 'wrong' as seemed to being suggested in the
original post. Nor is a considered accomodation of defacto standards. By
'considered' I mean keeping pace with prevalent usage while avoiding
implementing work arounds for every buggy implementation of a standard out
there.

-- Steve


---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to