Hi Matabele,
 

> Could Tobias please document the process of setting up a plugin library 
> like his?
>

In the context of this thread, I think it would do a lot of good for all of 
us if we started to give the */dev* wiki about the same attention (if not a 
little more, for the time being) as the core documentation at 
*tiddlywiki.com*. Yes, I can write a handful of instructions, but possibly 
they all should eventually be on *tiddlywiki.com/dev* and not where ever I 
would quickly publish things.

On this score, in order to use Tobias's library, I must first install a 
> plugin. Therefore, why not include plugin library plugins into the tw5 
> plugin library as a 'hook' into 3rd party plugin libraries?
>

That is not exactly true. You simply need to import a tiddler that is NOT a 
plugin. All it does is to define a path to the library and some field 
describing that plugin resource. There can sure be a list of trusted plugin 
sources that actually come shipped with the core... where user's can feel 
confident that they're not going to install entirely experimental features, 
but possibly / hopefully the more stable ones.

Also, the entire process of handling the plugin library needs a bit more 
*finishing* when it comes to:

   - catering for different plugin libraries
   - updating installed plugins
   - opening a plugin library listing ...as well as closing it

Perhaps it would be more convenient for some plugin authors to submit their 
> plugins to a 3rd party plugin library than to the core?
>

All that really boils down to what works best. In act, you are totally 
right that there could be a "public" plugin library where all authors may 
want to push their plugins to be listed... one that is not necessarily part 
of the tiddlywiki.com repo. For the moment, I would think the approach of 
having:

   1. either the core plugin library as a source for plugins
   2. a given author's plugin library as a source for that

...is good enough as it somewhat makes the task of curating all of that one 
that is well defined in terms of responsibilities: For 1. it would be 
Jeremy to take the lead ad for 2. a given author. So, yes, there could be:

3. a "community plugin library" where someone would take it upon themselves 
to:

   - review any contributions, possibly also in terms of code
   - actively ask authors for contributions to that listing
   - manage the whole thing as a repository, possibly published on GitHub

All that is not "far out there" but I imagine it takes a considerable 
amount of work to actually do... and some(one) willing to pull through the 
whole process and keep it up... because nothing is possibly as useless as a 
"community resource" that eventually dries out for there not really being a 
community to actually maintain it.

In summary: the core plugin library should contain mostly plugin library 
> plugins as hooks to 3rd party plugin libraries rather than the plugins 
> themselves -- then, rather than submitting my contribution to the core 
> plugin library, an author can submit to a 3rd party plugin library (or 
> maintain their own 3rd party plugin library.)
>

I'd very much advocate this model. So, the task would be to develop a given 
level of confidence / trust into which of these resources are officially 
listed / shipped with the core (in terms of "listings"). Quite possibly, 
the actual plugin libraries themselves could be (selectively) fetched via 
some form of TWederation:

   - "Please give me a list of all plugin libraries, their descriptions and 
   their plugin lists / descriptions ...so I can choose which of those I'd 
   want to include in my wiki to later on pull plugins, even content from."

This could also help with the TW federation infrastructure: a 3rd party 
> plugin library could be used as a intermediary for communication. In this 
> respect, it might be advantageous to produce an optimised, uglified and 
> stripped down version of the core specifically for the purpose. As it's 
> currently necessary to download the entire wiki to extract a single post or 
> comment -- the version opened in the iframe needs to be as small as 
> possible (all unnecessary functionality removed.)
>

There is definitely room for making servers cough less when shipping 
TiddlyWikis... I imagine it's a matter of working out the details to that 
workflow and actually doing it... knowing that this eventually will not be 
a good environment to debug issues in. ;-)

Best wishes,

Tobias.

-- 
You received this message because you are subscribed to the Google Groups 
"TiddlyWiki" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].
Visit this group at https://groups.google.com/group/tiddlywiki.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/tiddlywiki/89382493-8db2-4aef-9628-2d0f4dec1653%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to