Looper's Delight Archive Top (Search)
Date Index
Thread Index
Author Index
Looper's Delight Home
Mailing List Info

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Date Index][Thread Index][Author Index]

Re: Repeater and sync from midi in



I've responded to some things down here for discussion purposes, but I
realize that it's all moot anyway because there is no Electrix to appeal to
for documentation changes, etc...

----- Original Message -----
From: "Chris Muir" <cbm@well.com>
To: <Loopers-Delight@loopers-delight.com>
Sent: Saturday, July 19, 2003 2:59 PM
Subject: Re: Repeater and sync from midi in


> At 11:06 AM -0400 7/19/03, Paul Sanders wrote:
> >I don't consider that *right*. The RIGHT way to do it would be to 
>develop
> >such that this like this wouldn't cause obscure problems for people who
> >don't happen to know.
>
> While I agree that things should be as intuitive as they can be for a
given feature set, I don't think that ignorance should be glorified.
>
> In this case the confusion could be avoided by what would be learned in
MIDI 101. MIDI channels are pretty basic stuff.

I suppose that's true, but the notion that, one is going to get *something*
from a device (in this case the midi clock from a drum machine) so, it 
makes
sense they should be on the *same* midi channel isn't unreasonable IMO. The
notion that "I need to get the midi clock from this drum machin, so let me
make sure I set it to a *different* channel" is less than intuitive.

I've seen other posts from people getting bit by this, and others who have
"worked around" it but not really understood exactly what the problem is.
It's a bit obscure an could be documented better. Much better.

I'm just now coming in on all this MIDI stuff though in the past few days
I've learned most of what I've been after. I guess those who have been
dealing with it for some time have more of a "well DUH!" view of these
things because they've know and lived with all these things for some time.
I'm coming in with a fresh perspective. I guess though, it doesn't matter
what my perspective is, it won't change anything.

>
> [re: using MIDI notes for pitch change]
> >but since they chose not to do this they
> >should have plastered an unmissable caveat in the manual about it!
>
> I guess that they expected you to read the section on MIDI from page 33 
>to
38. The answer to your misunderstanding, while not exactly stated as "an
unmissable caveat", can certainly be found in this section.

I just looked at this, and I fail to find the answer to my 
misunderstanding.
I supposed if I were able to read betweent the lines a bit better it would
be more useful.

>
>
> >This is the perspective of a guy who's been a software developer in a
world
> >where things have to work correctly and robustly (high end commercial
Unix
> >systems).
>
> But do you expect people who run your software on these systems to be 
>able
to run it without understanding it, or at least something of the *nix 
world?
No, you probably assume a basic understanding of *nix. You probably assume
the users of your software will read the documentation or at least the man
page, no? And even in your documentation you probably don't have anything 
to
cover the case of "Gee, I did a 'chmod 0000 your_program' and now it 
doesn't
run".
>

I write my code to account for things that I anticipate are likely to
happen.

A decent analogy is if I'm writing a UDP socket program to communicate with
another system via UDP.

UDP, by definition is an unreliable protocol. I'm not going to require the
user to understand UDP, aside from assuring his/her network configuration 
is
appropriately set to communicate on the network it's attached to (this 
would
be the same thing as making sure I have the Repeater and drum machine set 
to
exchange midi clock.

I'm going to write my application such that it handles dropped packets
(which is possible with UDP) and resends them, and this will be transparent
to the user. That way they will never need to verify that their data
transfer had missing packets (or in the case of drum machine sending note
info on the same channel) extra packets that I aren't wanted.

MIDI, at a certain level is like a protocol. It's great that lot's of 
people
know all about it, but there are even more that use it and DON'T know so
much about it.

So, Kim suggesting to me that I should know so much more about the midi 
spec
before I make comments about how things are implemented is lame IMO. In a
smaller way, it's like me telling a customer they need to read the UDP 
spec.

Paul

> Chris
>
> --
>                        | In theory, there is no difference between
>  http://www.xfade.com/ | theory and practice. In practice, there is.
>      cbm@well.com      |               - Jan L.A. van de Snepscheut
>