On Nov 9, 2007, at 11:20 AM, Frank Barknecht wrote:
Hallo, IOhannes m zmoelnig hat gesagt: // IOhannes m zmoelnig wrote:
Luke Iannini (pd) wrote:
So, most of my work should be pretty self-explanatory. One
question: what is the intent of the "Messages" section? Is this "messages understood by the first inlet"? If so, I guess I should move the "Inlet 0" explanations there and put something like "Inlet 0: 1/0
for DSP On/Off + <Messages>" with a link to that section?well, here starts the great definition chaos on what is message
and what is data.for simplicity i guess that you have 2 choices:
- everything that goes to inlet-0 is a 'message'; in this case i
would suggest to remove the inlet-0 section (rather than the messages
section)
- everything that is not one of the predefined types "float",
"symbol", "list" is a 'message'; in this case we would need both a messages-
and an inlet0- section.Hm, but inlets different from inlet0 also may accept different kinds of messages, "pack 0 0 0" vs. "pack s s s" is one example. So I would suggest to remove the "Messages" section and only use Inlet-sections, that explain the various messages and data streams an inlet accepts or expects. But just like Luke I may now understand what Messages was supposed to contain in the first place.
I guess the idea of the "messages" section was to list things like
[reset(, [poll 1(, [set blah( etc. that are messages with custom
selectors. AFAIK, it's very rare for these messages to be sent to
anything but the first inlet. Then the "inlets" section would list
the message types that the inlet accepts: (float, pointer, symbol,
list, anything)
That was the reasoning behind separate sections, I don't have a
strong preference either way.
.hc
The arc of history bends towards justice. - Dr. Martin Luther
King, Jr.