[ticker-dev] Re: channel subscription rules

Ian Lister ticker-lists at lister.dnsalias.net
Thu Apr 22 22:32:50 CDT 2004


On Tue, 20 Apr 2004, David Arnold wrote:
>people are bad at copying things exactly -- regardless of whether it
>is getting the case of letters right, or just getting all the right
>characters in the right order.

Yep, so making it easier for them is important. Getting letters in the
right order is a lot more basic than getting their case right. As long as
you're dealing with ordinary single words (even if they're joined together
DNS-style) then people are usually pretty good at spelling, but case is
still pretty much an arbitrary choice (and some people will even still put
things in all upper case, just because they don't realise that's the done
thing).

>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.

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

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

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

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

While Sticker may be a model citizen in implementing many of these new
protocol features (they seem to have a tendency to go from Sticker to the
protocol rather than the other way around) most of the other clients are
either very slow to implement them or have no plans to implement them. For
example, I believe Ted doesn't think XTickertape should implement the
presence protocol, or even move away from text file configuration. Unless
these UI changes actually have some chance of happening, it's only fair to
make the protocol itself friendlier as well. (And Matthew, who's off in
Sticker-land, still sees the need for case-insensitive group names).

Also, while we can add mechanisms (such as integrated presence/ticker)
that allow people to bypass manual transciption in many circumstances,
there are many other circumstances where group names are best carried by
some non-digital (or otherwise non-automatable) means.

>  Ian> <...> and by the time people need to specify directory paths in
>  Ian> URLs they're more often cutting and pasting or just clicking
>  Ian> than having to type from scratch.
>
>exactly.

... _because_ by that point so they're so long and complicated people
don't try to remember them. Ticker group names aren't so long and
complicated that they preclude the possibility of humans remembering them
and/or transmitting them verbally, and I don't think they should (or will)
be.

>so you'd recommend
>
>  decompose-compat(Group) == decompose-compat("MyChannelName")
>
>as the basic pattern for subscriptions?  (potentially with fold-case
>wrapped around or inside that)

Yep.

Ian




More information about the ticker-dev mailing list