Pd-vanilla added pow~, abs~, exp~, and log~. Pd-extended had those
from cyclone by default since 0.39.3. IIRC, pow~ is the only one that
has issues since it it incompatible with the cyclone one. So I am
wondering how best to handle this in the next Pd-extended release.
I have started making a 'vanilla' libdir that is all of the pd-
internals from 0.42.5. This gives us the option to keep the cyclone
pow~ as the default and then access the new pow~ via vanilla/pow~ and
the standard declare/import styles. So all you who use pow~ with Pd-
extended, what do you think the best plan is?
The other related issue is perhaps its time to remove the included
libs from loading by default... Then people who want complete vanilla
compatibility would just load 'vanilla' and nothing else by default.
That would also provide a mechanism for backwards compatibility using
libs like vanilla-0.42.5, vanilla-0.41.4, etc.
.hc
"[T]he greatest purveyor of violence in the world today [is] my own
government." - Martin Luther King, Jr.
i think you have to look at which method would benefit the greatest amount of people. so, i think you'll find that far more people will benefit from proper vanilla compatibility straight out of the box, than will benefit from the cyclone library.
If that is true, there could be a lot of patches out there that use
[pow~] as cyclone's pow~ and would need to be changed. Therefore we'd
need to make some kind of way to alert people which one they are using
and the differences between them. The SVN is pretty clean, there is
only one patch in there that uses [pow~]: (nusmuk/distortion.pd)
The vanilla pow~ does make more sense, IMHO. I think the biggest
concern is minimizing breakage and helping people avoid annoying bugs.
.hc
On Nov 11, 2009, at 8:40 PM, hard off wrote:
i think you have to look at which method would benefit the greatest
amount of people. so, i think you'll find that far more people will
benefit from proper vanilla compatibility straight out of the box,
than will benefit from the cyclone library.
If nature has made any one thing less susceptible than all others of
exclusive property, it is the action of the thinking power called an
idea, which an individual may exclusively possess as long as he keeps
it to himself; but the moment it is divulged, it forces itself into
the possession of everyone, and the receiver cannot dispossess himself
of it. - Thomas Jefferson
While I like the idea of backward compatibility, I would prefer more cross compatibility. If Pd-extended continues to default to cyclone's [pow~], there will still be issues with compatibility between different versions of Pd in the future and the [pow~] problem will persist. While old patches may experience problems, future patches won't. Better to take care of it now than later.
I don't know how annoying it would be, as I don't use it that often anyway (I usually stick something like that in [expr~]) but what if a little warning message showed up in the Pd window when a patch loads with [pow~], at least in the next release or two? I think that would just help people catch it quicker.
.mmb
Hans-Christoph Steiner wrote:
If that is true, there could be a lot of patches out there that use [pow~] as cyclone's pow~ and would need to be changed. Therefore we'd need to make some kind of way to alert people which one they are using and the differences between them. The SVN is pretty clean, there is only one patch in there that uses [pow~]: (nusmuk/distortion.pd)
The vanilla pow~ does make more sense, IMHO. I think the biggest concern is minimizing breakage and helping people avoid annoying bugs.
.hc
On Nov 11, 2009, at 8:40 PM, hard off wrote:
i think you have to look at which method would benefit the greatest amount of people. so, i think you'll find that far more people will benefit from proper vanilla compatibility straight out of the box, than will benefit from the cyclone library.
If nature has made any one thing less susceptible than all others of exclusive property, it is the action of the thinking power called an idea, which an individual may exclusively possess as long as he keeps it to himself; but the moment it is divulged, it forces itself into the possession of everyone, and the receiver cannot dispossess himself of it. - Thomas Jefferson
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
I don't know how annoying it would be, as I don't use it that often
anyway (I usually stick something like that in [expr~]) but what if a
little warning message showed up in the Pd window when a patch loads
with [pow~], at least in the next release or two? I think that would
just help people catch it quicker.
although this might imply cooperation from the developers (I guess), I
find it a good solution.
about the library loading thing: is it possible to put vanilla loading
before all other libraries as default? that should ensure
vanilla-compatibility before anything else.
another developement could involve a more user-friendly startup window:
instead of the current one, maybe a dynamic list which scans the folders
of extra and adds checkboxes to them? the checked fields will be loaded by
next time. also with "select all" and "unselect all" options. but maybe
this involves too much effort?
João
.mmb
Hans-Christoph Steiner wrote:
If that is true, there could be a lot of patches out there that use
[pow~] as cyclone's pow~ and would need to be changed. Therefore we'd
need to make some kind of way to alert people which one they are using
and the differences between them. The SVN is pretty clean, there is
only one patch in there that uses [pow~]: (nusmuk/distortion.pd)The vanilla pow~ does make more sense, IMHO. I think the biggest
concern is minimizing breakage and helping people avoid annoying bugs..hc
On Nov 11, 2009, at 8:40 PM, hard off wrote:
i think you have to look at which method would benefit the greatest
amount of people. so, i think you'll find that far more people will
benefit from proper vanilla compatibility straight out of the box,
than will benefit from the cyclone library.
If nature has made any one thing less susceptible than all others of
exclusive property, it is the action of the thinking power called an
idea, which an individual may exclusively possess as long as he keeps
it to himself; but the moment it is divulged, it forces itself into the
possession of everyone, and the receiver cannot dispossess himself of
it. - Thomas Jefferson
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management ->
http://lists.puredata.info/listinfo/pd-list
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management ->
http://lists.puredata.info/listinfo/pd-list
about the library loading thing: is it possible to put vanilla loading before all other libraries as default? that should ensure vanilla-compatibility before anything else.
IMHO there ideally shouldn't be overlap between vanilla and other libraries, at least in names, so that if someone comes out with some different really funky [dac~] they should call it [funkydac~] and not [dac~] but I may be wrong there.
This makes me also think: for compatibility it might be interesting to think of some kind of 'dependency checking' aid for when one wants to distribute stuff, helping you check what additional stuff is needed rather than vanilla, especially for slightly bigger projects.
Kind regards, Lorenzo
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
Lorenzo wrote:
about the library loading thing: is it possible to put vanilla loading before all other libraries as default? that should ensure vanilla-compatibility before anything else.
IMHO there ideally shouldn't be overlap between vanilla and other libraries, at least in names, so that if someone comes out with some different really funky [dac~] they should call it [funkydac~] and not [dac~] but I may be wrong there.
that's debatable, but i think it's not the point.
e.g. Pd sorely misses [pow~]. so somebody who needs it, writes an external [pow~]. at some point in time, an upstream author decides that it really would be a good idea of have [pow~] in Pd an includes it into vanilla.
so what is the library author supposed to do? unwind the time?
fgmasdr IOhannes
vanilla pd should always have priority.
That doesn't take you far when your patches break because of changes.
We can make a way where we aren't crippled by name conflicts. There
are many examples to follow: Lua, Python, Tcl, etc. etc. etc. I tried
to cover all the issues that I could find in my PdCon paper:
http://at.or.at/hans/Let%27s_Make_Libraries_-_PdCon3.pdf
.hc
On Nov 19, 2009, at 12:08 PM, hard off wrote:
vanilla pd should always have priority.
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
Programs should be written for people to read, and only incidentally
for machines to execute.
- from Structure and Interpretation of Computer Programs
Hallo, Hans-Christoph Steiner hat gesagt: // Hans-Christoph Steiner wrote:
Pd-vanilla added pow~, abs~, exp~, and log~. Pd-extended had those from cyclone by default since 0.39.3. IIRC, pow~ is the only one that has issues since it it incompatible with the cyclone one. So I am wondering how best to handle this in the next Pd-extended release.
I have started making a 'vanilla' libdir that is all of the pd-internals from 0.42.5. This gives us the option to keep the cyclone pow~ as the default and then access the new pow~ via vanilla/pow~ and the standard declare/import styles. So all you who use pow~ with Pd-extended, what do you think the best plan is?
I didn't use pow~ with Pd-extended, but I've used Cyclone before (though not its pow~, I prefered the expr~ pow then.)
Anyway there was a big discussion about which pow~ is the right pow~ in regard to inlet order on pd-dev. Consensus was that the right pow~ should have its inlets as it is in vanilla now. Andy Farnell's book on sound design uses this version of pow~ all over the place. As it is an internal in vanilla now, I would suggest to make this definitive version the default in Pd extended as well and not deliberatly create an incompatibility between the same versions of Pd extended and Pd vanilla.
Ciao
Frank
-
Frank Barknecht -
Hans-Christoph Steiner -
hard off -
IOhannes m zmölnig -
João Pais -
Lorenzo -
Mike Moser-Booth