TiddlyWeb in layman terms?

[Amzg]

Could someone give a brief description of the planned or merely
discussed features of TiddlyWeb and a word about where things are in
the process and, you know, the general nice-to-know stuff.

For one, what were (are) the unsolved needs that the whole idea sprung
from?

Thank you

/Mat

[chris...@gmail.com]

As the author of TiddlyWeb, I'm not sure I can give you a description
in layman terms. I've been working on it long enough now that I take a
lot of things for granted. If you have specific questions I'm very
happy to answer them as best I can. In the meantime I'll try to give
an overview for which I apologize: it's bound to be inadequate.

= The Idea Whence Sprung

I started the work based on some concepts from Jeremy Ruston, so
perhaps he can join into this thread to provide his own input,
especially on the unsolved needs. I once asked him for an elevator
pitch for TiddlyWeb and he said:

"TiddlyWeb is a practical, readable, reference implementation of a
rigorously designed headless wiki server that may use TiddlyWiki as a
user interface. The combination of TiddlyWeb and TiddlyWiki allows a
complete separation of server and UI concerns."[1]

Which is a mouthful and not particularly down to earth.

The fundamental concept or goal of TiddlyWeb is that it is a system
that (eventually) allows for multiple people and teams to use, edit
and learn from a shared collection of Tiddlers in a security conscious
fashion. This makes it much like other server-sides such as ccTiddly,
TiddlyHome, etc.

Where TiddlyWeb is different tends to veer us off into territory that
is a bit more abstract:

* In TiddlyWeb Tiddlers are first class accessible objects on the
internet that can be manipulated through a clean and simple REST API.
* There is a simple (once the UI is made, we hope) system for
constructing dynamic collections of Tiddlers into TiddlyWikis that
work both on and off the network.
* TiddlyWeb is very extensible and flexible, allowing multiple systems
for storage and access control.

There is currently a demo running at http://peermore.com:8080/ (see,
for example, http://peermore.com:8080/recipes/AutoTiddlyWeb/tiddlers/TiddlyWeb
).

I recently unveiled a fun little hack using TiddlyWeb on the
TiddlyWiki dev list: http://groups.google.com/group/TiddlyWikiDev/browse_thread/thread/5352f22cefb805a1

= Planned and discussed features:

* Server-side generation of TiddlyWikis that can be taken away and
automatically sync back to the server.
* Flexible authentication system (user/password form, OpenID, Google
Auth, others).
* Flexible storage system (files on disk, googledata, MySQL
(planned)).
* Server-side HTML rendering of Tiddly content, for search engines to
see.
* Atom feeds of Tiddler content.
* Easy way to create custom, server-hosted TiddlyWikis by example and
selection.
* Multi-User collaborative wiki system running on top of TiddlyWeb,
using TiddlyWiki interface.
* TiddlyWiki Plugin Library running on top of TiddlyWeb.

From my own personal perspective what TiddlyWeb offers me is a way to
centralize a lot of my thoughts, noodlings, clippings, prevarications
and cogitations in a robust datastore that I can then slice, dice and
manipulate in useful ways using all the wonderful plugins that exist
in the TiddlyWiki universe. It's a library from which I can check out
as much or as little as I like.

= Where Things Are

TiddlyWeb has been in a working alpha form for a few months now.
Getting the basic architecture in place was pretty straightforward,
but ironing out the details has been the slow part. It's always this
way. The code[2] is available in the TiddlyWiki subversion repository
and may be checked out and run. At this stage of the game it requires
some degree of proficiency with the command line, and preferably some
experience installing Python modules. Development is being tracked in
Trac[3] and is ticking along. The major things left to do before we
can call the alpha something more than alpha (maybe a beta, maybe
more?) include:

* Clarifying and simplifying the installation process.
* Developing user interfaces for managing Bags, Recipes and Filters
(these are the tools that describe the dynamic TiddlyWikis).
* Fixing or creating necessary TiddlyWeb related plugins (for example,
login handling).
* Fixing some holes in the security model.
* Providing a more effective administrative interface for the system.
* Others that I can't remember now.

The primary sticking point for the last little while has been
insufficient use by other people. The more people use it, the quicker
it improves and becomes useful. This situation has improved quite a
lot recently but more adventurous souls are always welcome.

Thanks for your question. I'm sorry I was unable to be brief. If you
have more questions, please ask.

[1] http://cdent.tumblr.com/post/42582420/the-web-services-and-applications
[2] http://trac.tiddlywiki.org/browser/Trunk/contributors/ChrisDent/experimental/TiddlyWeb
[3] http://trac.tiddlywiki.org/query?status=new&status=assigned&status=reopened&component=tiddlyweb&order=priority

[Amzg]

Although I didn't understand particularly much I did get some bits
here and there and wish to thank you for your answer cdent.

A specific question: Is the following a description of TiddlyWeb or
something else or is it just my fantasy:

Some system allowing tiddlers to be handled individually and in a
decentralized way, scattered around all over some kind of virtual
tiddler net. And some kind of slimmed down TW only working as an
interface to present the tiddlers that the user creates links to. A
bit like using the IncludePlugin * but with the source tiddlers
actually being separate entities.

If not, is there anyone / any project aimed at this? And, if yes, are
there additional projects aimed at this?

Thanks

/Mat

* http://tiddlywiki.abego-software.de/#IncludePlugin

> for example,http://peermore.com:8080/recipes/AutoTiddlyWeb/tiddlers/TiddlyWeb

> ).
>
> I recently unveiled a fun little hack using TiddlyWeb on the

> TiddlyWiki dev list:http://groups.google.com/group/TiddlyWikiDev/browse_thread/thread/535...

> [2]http://trac.tiddlywiki.org/browser/Trunk/contributors/ChrisDent/exper...
> [3]http://trac.tiddlywiki.org/query?status=new&status=assigned&status=re...

[chris...@gmail.com]

On Oct 2, 12:36 am, Amzg <matiasg...@gmail.com> wrote:
> Although I didn't understand particularly much I did get some bits
> here and there and wish to thank you for your answer cdent.

If you'd like some clarification, feel free to point at the parts that
you are curious about.

> Some system allowing tiddlers to be handled individually and in a
> decentralized way, scattered around all over some kind of virtual
> tiddler net. And some kind of slimmed down TW only working as an
> interface to present the tiddlers that the user creates links to. A
> bit like using the IncludePlugin * but with the source tiddlers
> actually being separate entities.

That could be a description of TiddlyWeb but it is not quite there
yet. The functionality is present in the server design to allow such a
thing, and there has been increased work lately in the concept of
"lazy loading" (including Tiddlers only as they are linked/clicked).
What's not really there yet is the TiddlyWiki side plugin code to
bring it all together in a nice way. Lots of pieces are in place or
being built, but not integrated.

"scattered around all over some kind of virtual tiddler net" is
certain something I'd very much like to see.

You might find this blog posting and the comment that follows it
interesting: http://cdent.tumblr.com/post/46209623/webs-of-tiddlywebs-with-tiddlywebweb

[Morris Gray]

I wonder, rather than try to describe it as a generic system, could we
use a real world application to help come to grips with at least some
of its capabilities.

Let us take a typical manufacturing concern and apply the need to
disseminate various aspects of their Quality Assurance Manual
throughout the organization.

Just about every department has a need to access certain parts of its
specifications and procedures on a daily basis. Starting with the
Purchasing Department to incoming inspection, the manufacturing dept.
in-process inspection, checkout procedures, final inspection, packing
specifications, and dispatch. All of these have their own specific
sections of such a manual.

However in the past this information has been parceled out to the
different stations without regard to the overlapping context that is
sometimes needed. Or it is not parceled out at all and every
department and station has to wade through the whole manuel to get
what they need. In this case copies are invariably made and control
over revisions and changes is lost.

If there was a central repository for the manual and procedures made
up of tiddlers then that would constitute the whole of the procedures
when assembled as a complete manual by a master TiddlyWiki.

Now each department and station could customize their own needs
drawing from the tiddler base those things they specifically need,
even to the point of overlapping other departments and stations which
is a real world need so often neglected.

Revision and change-control integrity is now intact because the master
tiddlers are regulated centrally. Now what used to be uncontrolled
custom folders of photocopies becomes easily and highly customized
TiddlyWikis that are constantly up to date for everyone's specific
need; even for those people on the move with their own broad range of
packaged tiddlers or in remote locations with very specific needs.

This is not to mention how easily annotations can be made as certain
events arise and how easily 'request for changes' can be made to those
with the power to change them by providing them with a fait accompli
so all they have to do is approve it :-)

Would that make a good description of something TiddlyWeb could do?

Morris Gray
http://twhelp.tiddlyspot.com
A TiddlyWiki help file for beginners

[chris...@gmail.com]

On Oct 2, 3:27 pm, Morris Gray <msg...@symbex.net.au> wrote:
> Would that make a good description of something TiddlyWeb could do?

Yes, that's an excellent description of one of the use cases that has
come up a few times. A sort of multi-audience editing and publishing
system where the participants can work on and use some or all of the
content in their own particular customized contexts.

Your example of departmental procedures is very good. I don't
necessarily want to see all the procedures for everyone, but it would
be good to know how to get to them if I do need to see them. And if I
come up with a better process I can make annotations directly, just as
you say.