Re: Questions about URLs for Gopher search items
> As said above: All authors are waiting for your input, in form of
> patches, not text written in some e-mail.
> This is Open Source. Sending in a patch to the authors, instead of
> complaining, would have saved much overall liftetime.
You misunderstand me, and the situation. Bombarding the authors of
every Gopher client in existence with patches to "fix" their handling of
a poorly explored part of the protocol would be impudent until
discussion with the community had established some kind of clear stance
on what the "correct" thing to do is. I am not yet complaining about
what these clients do but rather widely surveying the current situation
to see what conventions are in place. Sometimes Gopher behaves, by
convention, contrary to RFCs, e.g. the terminating lone dot is very
rarely seen in the wild nowadays. Maybe we are ignoring RFC 4266 as
> The clean solution is to send what you discovered, as a patch, to all
> clients and where you realized, it did not work.
My statement about clean solutions was with regard to using item type 1
or 7 for URLs with search terms in them. Neither I nor anybody else can
"discover" the answer to this, it's not an empirical question. The RFCs
are silent and most clients opt-out of taking a stance, so this is
basically virgin territory. Asking for opinions, starting dialogue and
trying to build consensus is the obvious first thing to do. If the
stakeholders, i.e. people who run search engines and people who write
clients that deign to tell users where they are, come to an agreement,
then patches to bring everything in line with this, with reference to
the public discussion and agreement as justification, are certainly a
good thing. Sending around patches out of the blue implementing my
personal decisions would not be.
> We have too many people, who only can write e-mails.
I've spent much more time maintaining a popular gopher client and free
gopherspace hosting provider than I have writing emails to this list.