2017-10-22 19:52 GMT-02:00 Miller Puckette <msp@ucsd.edu>:

I've been thinking about some sort of such extention of loadbang - also
to allow you to specify the message, as in "loadbang 5" - I think that would

get a lot of use.

It would certainly do! There is already an external for this though, it is [cyclone/loadmess], I know I use it all the time. It's one of the new ones we included in cyclone 0.3

But I don't have a solid sense of what the design should
be (and am always afraid of implementing something before I understand all
the issues; if I do something stupid then its there forever :)

The thing about connecting [loadbang] to a message is that you can at least click the message on the patch while you're working on it to test things. That's also the case with a plain loadbang, but you can just also connect a bang GUI. The thing is that it'd be more annoying to deal with that if you had a message. The way [loadmess] works is that you can just click on it and it sends the message as well. I like that idea and it's easy to implement it in Pd, and it could be useful for a plain [loadbang] as well.

As long as loading messages go, what I don't like about [cyclone/loadmess] is that you can't give it a comma and send more than one message at once. That would be cool and I was even considering creating another external so it could do that.

2017-10-22 16:44 GMT-02:00 Dan Wilcox <danomatika@gmail.com>:

Or [loadbang] could have a second outlet for init bangs...

that actually sounds better and cleaner, and totally compatible with Miller's idea of allowing it to send messages as well. I don't see how that would raise any issue.

cheers