2017-07-30 15:56 GMT-03:00 Miller Puckette msp@ucsd.edu:
I think the way forward is to somehow get the "path" mechanism to do what people are asking for.
Fair enough, and I see that would mean getting the mechanism to: *A)* Integrate with Pd's Help Browser; B*)* Integrate somehow to [declare] and *C)* Allow 3 levels of locality (1- Application-specific, 2- User-specific and 3- Global).
Out of these, I find "C)" very complicated to deal with. And it also opens to the question: how to deal with the current "standard path" structure? Removing or replacing the current structure seems like a major overhaul.
So here's my analysis/revision.
I always meant "standard path" to refer to "stuff distributed with Pd"
That seems to be what the "extra" folder is, referred here as the "application-specific standard path". Let's keep it as it is then, we can just refer to it as the "standard path", and detach the other two scopes (namely *user* and *global* ones) from this concept.
It seems the "standard path" concept grew to encompass these two other scopes (*user* and *global*), which would actually correspond to your concept of what the "path" mechanism should be (a place to point to libraries). Let's then incorporate them into this revision of the path mechanism.
By the way, it's the common practice to have both of these scopes (user/global) clearly defined and hardcoded, to point libraries not part of the software distribution. Let's stop calling them "standard paths", and treat them as "*default paths*" for both user and global in the "path mechanism".
So, besides newly incorporated "defult paths", what did the "path mechanism" already have? A freely defined path for the *user* only. therefore, bringing them all together gets us with a default *user* and *global* path, that Pd sees out of the box, plus the ability to define other folders for the *user* scope.
So, what kind of overhaul am I proposing? Basically none, just a new way to look and reinterpret what we have.
And I think it'd be reasonably easy to expand the functionality of user defined paths to do what we are asking for, meaning: *A)* Integrate with Pd's Help Browser and *B)* Integrate somehow to [declare].
Who's following me?
cheers
A) Integrate with Pd's Help Browser and B) Integrate somehow to [declare]. Who's following me?
I've just written basically the same :-)
Gesendet: Sonntag, 30. Juli 2017 um 22:00 Uhr Von: "Alexandre Torres Porres" porres@gmail.com An: "Miller Puckette" msp@ucsd.edu Cc: pd-list pd-list@iem.at Betreff: [PD] Revision of Standard Paths (was Re: New users and external path struggles)
2017-07-30 15:56 GMT-03:00 Miller Puckette <msp@ucsd.edu[mailto:msp@ucsd.edu]>:I think the way forward is to somehow get the "path" mechanism to do what people are asking for. Fair enough, and I see that would mean getting the mechanism to: A) Integrate with Pd's Help Browser; B) Integrate somehow to [declare] and C) Allow 3 levels of locality (1- Application-specific, 2- User-specific and 3- Global). Out of these, I find "C)" very complicated to deal with. And it also opens to the question: how to deal with the current "standard path" structure? Removing or replacing the current structure seems like a major overhaul. So here's my analysis/revision.
I always meant "standard path" to refer to "stuff distributed with Pd" That seems to be what the "extra" folder is, referred here as the "application-specific standard path". Let's keep it as it is then, we can just refer to it as the "standard path", and detach the other two scopes (namely user and global ones) from this concept. It seems the "standard path" concept grew to encompass these two other scopes (user and global), which would actually correspond to your concept of what the "path" mechanism should be (a place to point to libraries). Let's then incorporate them into this revision of the path mechanism. By the way, it's the common practice to have both of these scopes (user/global) clearly defined and hardcoded, to point libraries not part of the software distribution. Let's stop calling them "standard paths", and treat them as "default paths" for both user and global in the "path mechanism". So, besides newly incorporated "defult paths", what did the "path mechanism" already have? A freely defined path for the user only. therefore, bringing them all together gets us with a default user and global path, that Pd sees out of the box, plus the ability to define other folders for the user scope. So, what kind of overhaul am I proposing? Basically none, just a new way to look and reinterpret what we have. And I think it'd be reasonably easy to expand the functionality of user defined paths to do what we are asking for, meaning: A) Integrate with Pd's Help Browser and B) Integrate somehow to [declare]. Who's following me? cheers_______________________________________________ Pd-list@lists.iem.at mailing list UNSUBSCRIBE and account-management -> https://lists.puredata.info/listinfo/pd-list%5Bhttps://lists.puredata.info/l...]
2017-07-30 17:04 GMT-03:00 Christof Ressi christof.ressi@gmx.at:
I've just written basically the same :-)
I saw it :) well, it seemed better to have a fresh reboot on the discussion now that the issue is clearer. Do you agree with the rest of this?
2017-07-30 17:20 GMT-03:00 Christof Ressi christof.ressi@gmx.at:
with user paths you automatically have global/user space simply because some folders are global (C:/Users/All Users/Pd/libs) and some are for users (C:/Users/Christof/Pd/libs) :-).
yeah, you can add globals too, that is if all users to the same, but you're right, nice catch.
I think we can leave the standard path mechanism as it is and just add
features to the user path, as Miller suggested.
Yeah, and all that changes is an interpretation of what we got, and the fact we'd call it something else on the documentation, but no changes under the hood. Instead of "Standard Paths" for user and global scopes, we just have a "default" one. Which would be, by the way, just a hardcoded suggestion, allowing the user to decide on something else if they want - seems like this sort of freedom and flexibility is highly appreciated in the community :)
-
Alexandre Torres Porres -
Christof Ressi