[ticker-dev] Re: channel subscription rules

[David Arnold]

-->"Ian" == Ian Lister <ticker-lists at lister.dnsalias.net> writes:

  >> in considering the tradeoff between usability and efficient
  >> evaluation, i'm not convinced that making the router do
  >> 
  >> fold-case(Group) == fold-case("MyChannelName")
  >> 
  >> instead of
  >> 
  >> Group == "MyChannelName"
  >> 
  >> is worth the usability outcome.

  Ian> Probably because we attach different relative importance to
  Ian> usability and saving CPU cycles ;-)

I think it's more a case of preferring to distribute the work, rather
than centralise it.

But .. it's not a great deal of work, and it's something that's pretty
simple for the Elvin Router to do, so it's not a major consideration.


  >> i believe the fundamental problem lies in the UI where we require
  >> users to accurately transcribe strings.

  Ian> To pass the issue off to the UI would mean we would have to
  Ian> have made sure that we'd provided everything in the protocol
  Ian> that the UI would need, and that our expectations of the UI are
  Ian> realistic. I don't think that we've done either of these.

  Ian> We have a means whereby group names can be communicated using
  Ian> the presence protocol, assuming two users are already in the
  Ian> same presence group, but this effectively just pushes the
  Ian> problem from finding correct ticker group names to finding
  Ian> correct presence group names.

Right.  This is great for friends, but maybe not so great for friends
of friends, etc.

  Ian> While Sticker may be a model citizen in implementing many of
  Ian> these new protocol features (they seem to have a tendency to go
  Ian> from Sticker to the protocol rather than the other way around)

Which I think is an excellent model, FWIW.

  Ian> most of the other clients are either very slow to implement
  Ian> them or have no plans to implement them. For example, I believe
  Ian> Ted doesn't think XTickertape should implement the presence
  Ian> protocol, or even move away from text file
  Ian> configuration.

I don't think either of these is a problem: if someone wanted to
extend XTickertape to address an audience that wasn't happy to work
that way, I suspect Ted wouldn't mind them taking a copy of the code
(present legalities aside).

Some people are content with having to type things in accurately;
others aren't.  That's ok.

But ... 


So, what's the general opinion?  Is everyone happy to move to using

  decompose-compat(fold-case(Group)) == decompose-compat(fold-case("MyGroup"))

as the general subscription template ?




d