[gopher] gopher++ (gopher1) protocol

Cameron Kaiser spectre at floodgap.com
Mon Jan 11 17:35:01 UTC 2010


In fairness to Kim, we did talk a little more favourably about having the
server handle transcoding a few months ago.

I don't mind server transcoding, btw, but my problems with Kim's ideas
are:

- Performance issues for the server figuring out if it has a gopher0 or
  gopher1 response. This is the big kicker for me, though I appreciate the
  attempt at backwards compatibility.

- The server needs to be able to tell the client it *refused* to transcode,
  or couldn't.

  Conversely,
- The server needs to be able to tell the client there is nothing to
  transcode (or simply a file not found), even if it is willing to.

I support putting much of the work on the server, because I think Gopher's
biggest benefit is a VIC-20 could be made to be a gopher client. Indeed,
implementing gopher clients is now very popular in retrocomputing circles
because it has little or no barrier to entry. I think this is a key way to
play to Gopher's strengths.

That said, the three issues I have above would need to be resolved to my
satisfaction before I'd support such a scheme in, say, Overbite or GopherVR.

-- 
------------------------------------ personal: http://www.cameronkaiser.com/ --
  Cameron Kaiser * Floodgap Systems * www.floodgap.com * ckaiser at floodgap.com
-- Seen on hand dryer: "Push button for a message from your congressman." -----



More information about the Gopher-Project mailing list