On Thu, 16 Dec 2010, Chris McCormick wrote:
On Wed, Dec 15, 2010 at 09:57:08PM -0800, Jonathan Wilkes wrote:
In many cases it is replaced by the effort required to make a hack to replace the functionality of the missing external.
Yep. In my experience, the cost-benefit balance usually falls on the side of restricting myself to not using many externals,
What's the cost of getting used to restricting yourself in such a manner ? The effect on your problem-solving habits with pd... you'd have reasons of using externals, such as concision, completeness, correctness, expressing yourself at an appropriate level of understanding, but instead, you'd rather learn kludgy workarounds by heart until you don't have to think about them anymore ?... (though you can compensate for some of it using abstractions, but how much are you compensating ?)
Why don't the handheld-pd-without-libdl crowd pick a set of most useful externals and compile them as part of vanilla, statically ? I mean only things that would be portable anyway : there are lots of useful things you can do as externals for pd, that don't require any libraries whatsoever (apart from a minimal libc).
What do you do to get [fiddle~] loaded, on a system on which you don't have the dynamic loading support running ?
I guess I view it in a different way. Pd-msp is a constrained software environment. I choose to match my patching style to those constraints so that I don't have to do more annoying and time-consuming work.
If you wanted to avoid annoying and time-consuming work, you'd use externals.
It's like writing a haiku.
Haikus don't get any work done.
(And I'm not even convinced that they _say_ anything either !)
And if you cared about getting patches to remain as small as they can be, you'd care a lot more about externals than you do.
| Mathieu Bouchard ---- tél: +1.514.383.3801 ---- Villeray, Montréal, QC