Well, I kind of agree with Yoann. We just spent some serious time
evaluating whether to use TiddlyWiki as the basis for an open source
application we're building, and backed away for several reasons:
- no well defined community or governance mechanisms
- really sketchy documentation (nothing serious at the architectural level)
- no clear idea of how it follows standards (or doesn't) - which is
really critical as browsers keep evolving
Re. TiddlyWeb - now it's REALLY hard to decipher what's going on under
the hood.
All in all, it's very frustrating. TW is very capable, obviously has a
lot of time and expertise invested in it over several years - but what
could be a really powerful platform for single-page applications has
really become more of a series of hacks than something more carefully
organized.
It's a shame.
Miles Fidelman
--
In theory, there is no difference between theory and practice.
In<fnord> practice, there is. .... Yogi Berra
an email list is not a community and definitely not a governance mechanism
by contrast:
- debian.org
- http://couchdb.apache.org/
- http://drupal.org/community
right now, the most concise collection of info about Tiddly is the
wikipedia page
>
>> - really sketchy documentation (nothing serious at the architectural level)
> One more of Eric's: http://www.tiddlytools.com/insideTW/
>
> TiddlySpace: http://tiddlywikidev.tiddlyspace.com/
that's a start, but it's a far cry from anything that provides an
architectural overview, data architecture, data flow architecture,
control flow architecture -- i.e., how all the pieces fit together
say... like the documentation for jquery
>
>> - no clear idea of how it follows standards (or doesn't) - which is
>> really critical as browsers keep evolving
> http://tiddlywiki.com/#%5B%5BTiddlyWiki%20Browser%20Compatibility%5D%5D
that's an after-the-fact list of problems - not a strategy for staying
ahead of browser compatibility issues
like, say the approach phonegap takes
>> Re. TiddlyWeb - now it's REALLY hard to decipher what's going on under the hood.
> http://tiddlyspace.tiddlyspace.com
> http://docs.tiddlyspace.com
> http://tiddlyweb.peermore.com/wiki/
> etc etc etc....
lots of little pieces - nothing that presents an organized view of how
the pieces fit together
and... lots of tidbits, spread across lots of locations - no indication
of what's current, what's not
>> All in all, it's very frustrating. TW is very capable, obviously has a
>> lot of time and expertise invested in it over several years - but what
>> could be a really powerful platform for single-page applications has
>> really become more of a series of hacks than something more carefully
>> organized.
> TiddlyWiki is hackable - which is why I use it...
> If you're a developer I suggest that you search for plugins and plugin
> documentation on TiddlySpace, TiddlyTools or on GitHub.
> If you are an software entrepeneur PLEASE hire Eric and get the
> fastest most efficient response you can think of..... (And we get a
> happy moderator...)
personally, my suggestion is that Eric, or somebody take a serious look
at developing a business model for TiddlyWiki - it could be a very nice
platform, but not if it's simply a couple of hackers, hacking away when
they can
Jeremy Rustin, or the folks at UnaMesa, or Eric, or somebody needs to
start thinking about what TiddlyWiki wants to be when it grows up - and
how to get there --- they need to start thinking like Ian Murdock, or
Linus, or the folks behind CouchDB....
right now, it's amateur hour
--
As a recipient of your freely given help over the years I sincerely
hope I can help in some way - I am not sure how at the moment though,
apart from cash.
It would be very interesting to see if some TiddlyFans could spin some
kind of business model operation out of TW. I've often thought "I
wish my real world worked as well as TiddlyWorld."
I think I could have sold TW solutions to customers in the past, but
have backed off because I didn't have the experience or expertise to
deliver something professional. I've only every presented TW as
prototypes and they have not been part of convincing enough projects
to generate serious funds.
GET OUR BUSINESS HEADS TOGETHER ?
I wonder if TiddlyFans might want to get their heads together outside
the Google Group? Bauwe and I Skype from time to time, it would good
to talk to the other European TW people, Måns, Tobias, Mario. One of
the things I like about TW culture is its Euro-Centric make up. I may
be quite alone here in Manchester but I really like this aspect of the
"community" - and I do think that is not an inappropriate use of the
term
The point about governance is an interesting one that has come up
before, its an interesting one given the context - of open source
innovation. It would make a great study for a student of business (I
have made some enquiries for myself) - the story should be written up.
But I don't think governance should get in the way of some
entrepreneurial thinking.
Where I am in Manchester UK money is very tight and going to get worse
over the winter as funding is cut from many 3rd Sector and community
projects. It's these organisations which could benefit from
TiddlyThinking, open source and general Wikification but they are the
ones least likely to have decent budgets to spend. It's a price
sensitive end of the market, dependent on volunteers, volunteers would
would prefer "can operate Word, Powerpoint and Excel to proficient
standard" on their CV. TiddlyWiki - or any kind of wiki for that
matter - is quite a difficult sell to organisations without resource
to allocate to trying out new methods of working.
Having said all that, it might help if there was a compelling example,
a case study - an elevator pitch, (perhaps in video form?) - showing
the benefits of using TW and perhaps more importantly, the culture
that's developed around, it to an organisation. (Mån's use in his
school sounds very enlightened. Nordic pedagogy has a great reputation
in the UK, I'd love to see it in action and hear the user stories. If
it can save money and be easy to use then it will catch on)
got to go...
best wishes
Alex
> --
> You received this message because you are subscribed to the Google Groups "TiddlyWiki" group.
> To post to this group, send email to tiddl...@googlegroups.com.
> To unsubscribe from this group, send email to tiddlywiki+...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/tiddlywiki?hl=en.
>
>
What with Jeremy and Erik both looking for some ways to support both
TiddlyWiki and themselves, it seems like this is the time to start
thinking business model and organization very seriously.
And in that regard, two possible models suggest themselves:
1. Zotero (Zotero.org) is a bibliographic tool for managing research
citations. It runs as both a stand-alone browser plug-in, and in
concert with a central server for sharing bibliographic records. It's
as, or more, capable than commercial products and has a significant
following among academic researchers. It strikes me as an effort of
comparable scope to Tiddly.
The project "lives" within the "Roy Rosenzweig Center for Historiy and
New Media" in the Dept. of History and Art History, at George Mason
University - with work funded by several foundation grants. I.e., a few
techies are employed by a grant to develop and maintain the software,
within a university department.
It strikes me that Jeremy and Erik might have some serious discussion
with UnaMesa about how to set up a similar arrangement under UnaMesa's
501(c)3 umbrella, and pursuing foundation and/or corporate sponsors for
the work.
2. Talk to the folks behind CouchDB about how they ended up operating
under the Apache Software Foundation, and how they support themselves
and the continued development of CouchDB.
At some point, Tiddly is going to live or die by whether or not the
folks most committed to it find a long-term model for supporting the
software and themselves.
(Just one man's opinion.)
Miles Fidelman
It's a curious thing that Eric's expression of a need to ensure his
livelihood somehow managed to devolve into a huge pile of complaining,
demanding and introspection on what's wrong with the tiddly* community.
At such times I would think it far better to celebrate and remind
ourselves of the huge contributions that Eric has made over the years to
the community. His work on creating, maintaining and documenting plugins
is second to none and his efforts to ensure that the TiddlyWiki core
keeps its promises have been outstanding, even if the face of sometimes
different priorities from elsewhere.
So, congratulations to Eric on his many years of work. I'm confident
that as a community we will be able to work out ways to ensure that
he can continue to contribute.
I hope that's not at all controversial. The controversial part comes
next, below:
On Thu, 10 Nov 2011, Miles Fidelman wrote:
> At some point, Tiddly is going to live or die by whether or not the folks
> most committed to it find a long-term model for supporting the software and
> themselves.
I completely disagree with this, and I think it is the prevelance of
this attitude in this thread which has contributed to me becoming
angry and finding it hard to find a way into the thread that is
clear and focused.
Tiddly lives and dies by the efforts of its community. Not by
business, business interests, or even leaders.
Tiddly is a thing in its own right with a community of participants
who all have free and easy access to:
* create documentation
* contribute code
* report bugs
* demonstrate cool functionality
In one of his earlier message Miles asks that Tiddly be more like,
e.g., Linux.
Back when I first started using Linux, I guess in about 1992, maybe
93, Linus and a few core devs were responsible for making patches
and distributing tarballs. I installed Linux via SLS, a distribution
packaged by someone who was not Linus nor one of those core devs.
Later as the community grew someone got the bright idea to start the
Linux Document Project (source of Howtos and the like). That wasn't
a core dev either.
A while later I was using Linux to run an ISP. We had features that
we needed in the packages that ran our services. We mades patches to
apache, qpopper, PAM, innd and contributed them back to the
communities that surrounded them. My employer, the ISP, paid me to
make those contributions back to the free (as in beer) software that
made our business possible.
TiddlyWiki is rather unique in its nature as a standalone piece of
software. It is less easy to connect back to a community than say
something like Apache or Linux. This does not, however, obviate the
responsibility the community has to the health of TiddlyWiki itself
and the TiddlyWiki community.
I'm not sure when it happened (because I have not been observing the
community for that long) but at some point the TiddlyWiki community
stopped operating as one. Perhaps it was when Osmosoft was bought by
BT. It sometimes seems like at that point people decided "oh there
is money now, BT will take care of it."
That's never been the case and never should have been the case. BT
bought Osmosoft to understand open source operations yet bizarrely
TiddlyWiki has become less and less operational as an open source
project since the purchase.
BT's engagement with TiddlyWiki ought to be much like the ISP
(above) engaging with various software: It contributes back to the
community those improvements which it finds valuable to itself. For
example BT wanted a certain type of server-side so they paid me to
make one (more on TiddlyWeb below).
The maintenance of the community, though, should have been and needs
to be (for the sake of just distribution of power) done by the
community and the simple truth is this has not happened. There are
presumably a few reasons for this. Some of it is that perceived
"leaders" didn't step up in an effective fashion:
* When Jeremy was sucked up by the BT spaceship his availability
vaporized. Perhaps that will change now with his recent news.
* Martin, who has been the inside Osmosoft lead of TiddlyWiki
development, has not engaged the community with alacrity.
* Eric, though his contributions are extremely valuable, insists on
keeping them in a format that is not accessible to open source
processes such as version control, forking, patching, issues
tracking, etc. Nor has he, despite many invitations, become a
proper contributor in the core code, using git etc.
* Those of us with monetary relationships with Osmosoft (me, Ben,
Jon, Colm, once upon a time FND, but no longer) have
resposibilities which do not prioritize TiddlyWiki but instead
business goals given to them by the people with the money and
their own developing careers.
* The (probably unrealistic) expectation that UnaMesa, in addition
to providing hosting for things like trac.tiddlywiki.org, would
provide human engagement in the community, went unfulfilled.
Some of it is that the combination of those 5 items above created a
space that looked like it was going to be filled at any moment, so
other participants need not apply or try.
Some of it has been a perception that TiddlyWeb and TiddlySpace were
taking some of the energy. I don't think that's true at all. TWeb
and Space don't change TiddlyWiki. And you do not dimish a candle by
lighting another. These projects can coexist but in order for them to
do so the expectation that people working on only one or two of the three
is somehow taking away from the other must go away. For example I
don't work on TiddlyWiki. I just don't. I'm not supposed to and nor
do I want to. That's not wrong. That just is. There are plenty of
other people in the world who _do_ work on TiddlyWiki.
But, I'd argue, in the end it is because the community of users that
surround tiddly* didn't just do it themselves. That's what open source
really means. It doesn't mean getting free stuff. It means signing up
for a club in which in exchange for providing your expertise (whether it
be testing, developing, writing, explaining, promulgating, or simply
being a positive force of feedback and encouragement (btw, an aspect
which people way undervalue and should not)) you get legitimate access
to use some free stuff and the free expertise of the community.
Everything that TiddlyWiki needs to be a fully operational and
working open source community is available for free: source control,
websites, wikis, mailing lists. If things aren't the way you want
them to be, then please, gather some comrades and fix it. Fixing it
can be as much as the whole kit, or it can be as simple as making
sure that the issues you have are more accurately reported in the
issue trackers associated with the projects. It's in _your_ hands.
Which leaves me to talk a bit about TiddlyWeb. It's been suggested
that it is:
* too hard and too inaccessible (by virtue of power, complexity and
language choice)
* that is it not powerful enough (by not engaging in things like
federated wiki)
* that it violates some principle of how tiddlers ought to be
Perhaps, but TiddlyWeb is in fact exactly what it was designed to
be. It is an implementation of an idealized HTTP API for accessing,
storing and composing tiddlers on the web that is designed for
maximum flexiblity and general usefulness for creating
tiddler-based web applications. Note I say "tiddler", not
"tiddlywiki".
By which I mean that it, by design:
* uses a readable programming language
* has a plugin architecture that lends itself to open source
development and visibility
* prioritizes inspection of the code over simple installation (even
though installation is one step on mature systems)
* allows tiddlers to be represented in many ways, not just in
TiddlyWiki
The end result is a toolkit for collaborating via tiddlers but it is one
that considers itself at least as much an education and uplift tool as a
useful piece of software. The original description was that TiddlyWeb
would be a reference implementation of the API, leading to other
implementations, from the community if they were interested, that were
more suited to particular environments, such as installing on PHP
hosting services.
So to join this last section of the message to the first: If
TiddlyWeb is not what you want it to be, you have the power to
influence it, to change it, or even to ignore it. You. The same goes
for TiddlyWiki and TiddlySpace. This is both the blessing and the
curse of _real_ open source projects.
I very much want all the tiddly* projects to be real open source
projects. With active and diverse contributions from the entire
community. It is very clear from the messages in recent days that
there are loads of really good ideas on how things ought to be. As a
community we need to turn these conversations into action.
I hope very much that we hear from Jeremy, Eric and Martin on these
issues very soon. Their involvement is not critical to success but
their expertise will certainly make it easier.
--
Chris Dent http://burningchrome.com/
[...]
When I joined BT I tried to address Eric's needs by persuading BT to
support UnaMesa, which they have generously continued to do after my
departure (albeit the amount has been reduced in recent years in line
with BT's internal cost cutting programme). We failed, though, to
bring any other supporters on board, making it hard to give Eric the
level of income he deserves.
The thing that might not be clear from the outside is that at the time
of the Osmosoft acquisition BT weren't really terribly interested in
TiddlyWiki itself; they wanted to bring me on board to help them
understand open source and to learn how to use it sensibly. TiddlyWiki
was part of my qualifications, but not really of specific interest.
That's changed now; Osmosoft is all about TiddlySpace, TiddlyWeb and
TiddlyWiki. If the projects using them really take hold in BT then
perhaps we'll be able to persuade BT to offer more support to the
community.
Now I've chosen to leave BT because I crave the independence of being
able to work with a much wider range of clients. I don't expect a
great deal of my consultancy work to be directly about TiddlyWiki, but
it will absolutely be *because of* TiddlyWiki. So, one of my first
goals is to give some love and polish to TiddlyWiki, to make it be a
decent calling card. I love coding, and haven't done nearly enough for
a long time.
Now, Eric's work makes TiddlyWiki better and brighter, and so it makes
sense for me to try to support him directly out of my consultancy
income. I'll commit to a regular monthly donation, that will hopefully
rise as I get my teeth into my new working life.
Best wishes
Jeremy
> From a personal perspective, the problem is simple: in order to
> continue providing full-time support to the worldwide TiddlyWiki
> community, I must quickly find a way to replace the loss of revenue
> from UnaMesa, ideally without the distraction of finding and working
> on *non*-TiddlyWiki consulting projects. Of course, TiddlyTools.com
> is an open-source effort, so there are no licensing fees for the
> addons I have published, and my online assistance has also been
> offered free of charge. However, there *is* a "tip jar" form on
> TiddlyTools.com (see http://www.TiddlyTools.com/#Donations) that
> --
> You received this message because you are subscribed to the Google Groups "TiddlyWiki" group.
> To post to this group, send email to tiddl...@googlegroups.com.
> To unsubscribe from this group, send email to tiddlywiki+...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/tiddlywiki?hl=en.
>
>
--
Jeremy Ruston
mailto:jer...@osmosoft.com
http://www.tiddlywiki.com
>> At some point, Tiddly is going to live or die by whether or not the folks
>> most committed to it find a long-term model for supporting the software and
>> themselves.
>
> I completely disagree with this, and I think it is the prevelance of
> this attitude in this thread which has contributed to me becoming
> angry and finding it hard to find a way into the thread that is
> clear and focused.
The core of Mile's observation applies to both of us, though: we're
trying to secure a long term model for supporting ourselves while we
work on the software. I guess your point is that if either of us
failed to establish a viable model, then it shouldn't necessarily
imply the death of the software.
> * When Jeremy was sucked up by the BT spaceship his availability
> vaporized. Perhaps that will change now with his recent news.
Yes, that's what it felt like to me, too. Doing a good job in an
executive role in a big company has very, very little in common with
running an open source project. I'm relishing recovering my
independence.
Many thanks,
Jeremy
> The core of Mile's observation applies to both of us, though: we're
> trying to secure a long term model for supporting ourselves while we
> work on the software. I guess your point is that if either of us
> failed to establish a viable model, then it shouldn't necessarily
> imply the death of the software.
Not only should it not imply the death of the software, I think if it
even worries people about the death of the software, then there is
something very unhealthy happening in the community.
But beyond that I think it is important to keep in mind that though
I'm currently employed as a creator of a tiddly-related code I do not
believe that what I'm paid for is the code itself. The code is free,
it is merely an expression of my expertise. It is the expertise and
associated experience which is being paid for.
When you, Eric, I or anyone else is paid to improve tiddly* it is
because the payer needs it in either a faster or more direct way than
the community can provide OR they are doing what they feel is just in
the face of value they are getting from the community. Organizations
like BT, in general, can use money more easily than they can perform
the committed community participation that individuals provide in the form
of use, bug reporting, documentation, community assistance and plain
ol' writing code.
In the end, whatever the currency, the value obtained is membership and
participation.
Perhaps, more to the point, my observation, across multiple open source
(and other projects), dating pretty far back (ARPANET era) is that
"communities" are very amorphous things that are not capable of very
much without a level of organization. In the early stage, that
organization most often comes from an individual or core group who
is/are most committed to the project (usually the founders), eventually
evolving into an established set of procedures, tools, roles, etc. that
allow the project to move forward without them.
Perhaps the best example of this is the Apache daemon - starting as a
funded R&D project at NCSA (the "NCSA Daemon") with a team of people
behind it, with funding behind them. After a while, two things happened:
- a user community had developed around Apache
- NCSA decided it was no longer "researchy" and decided to kill its
involvement
Those two events led to a lot of turmoil, that, after several
incarnations, led to the Apache Software Foundation as a long-term home,
and the ecosystem, infrastructure, and community that maintains ongoing
support and development. (This is, of course, a simplified version of
the history - a better telling, and one that might be educational in our
context, is at http://httpd.apache.org/ABOUT_APACHE.html)
There are lots more open source projects that have disappeared into
oblivion than have gone on to long-term viability. Survival seems to be
a mix of BOTH doing something useful to a large community, AND a small
leadership group that organizes the effort in a way that puts it on a
long-term, sustainable path. It probably doesn't matter if that group
is doing it for commercial reasons (e.g., building a company around a
core piece of open source code) or other reasons - though generally that
core group needs to find a way to support themselves and their efforts.
Whether it's a supportive employer (perhaps one who uses or otherwise
benefits from the software), or a business organized around the
software, both people and projects have real expenses - and it's a lot
easier to focus if one's "day job" aligns with the project.
Seems to me that, for Tiddly to move down that path, some core group
needs to provide the focus for a year or two - and Jerymy and Erik seem
like the obvious candidates.
Just one man's opinion, of course.
> Seems to me that, for Tiddly to move down that path, some core group needs to
> provide the focus for a year or two - and Jerymy and Erik seem like the
> obvious candidates.
Yes, this I agree with and am fully behind.
I believe that one of the simpler causes is that over the last 4 years
the content of the main tiddlywiki site hasn't been actively updated
and managed. As a result of this neglect (and as we've been discussing
in a separate thread) many of the links at http://tiddlywiki.com/ have
been dying off, and none of the new stuff (like Julio's) has been
incorporated. As a result, to a new visitor, the project feels like at
worst it's dying, and at best that the community resources are
hopelessly hard to find and disorganised.
Part of the reason is that even before working at BT, I was pretty
unreliable at performing updates to the content; the process was
ridiculously tedious and error prone. There were two or three updates
by Osmosoft but they were fairly superficial.
tiddlywiki.com is the de facto focal point for users of TiddlyWiki,
and it should clearly lay out the available resources, and showcase
the activities of the community around the project. It really ought
to, for instance, list the few dozen plugins that are most widely
used; keeping that list up to date and authoritative shouldn't be an
onerous task to someone participating regularly in the discussion
groups. At the moment we all know that the FET plugin is critical to
lots of quite basic TiddlyWiki requirements, but you wouldn't know
that from poking around tiddlywiki.com.
The process to update tiddlywiki.com is much better now; there is a
staging space for the content for the content at
http://tiddlywiki-com.tiddlyspace.com/, and so we've got a lot of
flexibility for multiple authorship, and pulling in content from other
places.
So, now that I've got my independence, as I've mentioned before, one
of the things I want to do is start properly and regularly editing the
content of tiddlywiki.com. Tiddlywiki.com gets quite a lot of traffic,
and doing a better job of showcasing links to community created
resources, and bringing traffic to them may well improve the
incentives to do such things. There's already a separate thread about
the updates, please do shout if you've any suggestions, questions, or
offers of help.
Best wishes
Jeremy
> --
> You received this message because you are subscribed to the Google Groups "TiddlyWiki" group.
> To post to this group, send email to tiddl...@googlegroups.com.
> To unsubscribe from this group, send email to tiddlywiki+...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/tiddlywiki?hl=en.
>
>
--
Well... it's also not all that developer friendly. And THAT's the rub.
We just went through a technology evaluation process for building a
single-page application.
Now I really like Tiddly (except it's basic interface - for what we're
doing outline format, with click to expand/contract sections/subsections
would be a lot more useful) - but I sure as heck can't figure out how
all the pieces fit together to write a different UI on top of the common
internals.
And... pretty much everybody else on my team takes one look at Tiddly
and throws up their hands.
I expect we're going to go off and write our own framework for
single-page applications - because it sure looks a lot easier than
buliding on Tiddly. That's sort of a shame.
I remember reading on this list that Jeremy is a seasoned author:
there was photographic evidence of a book found in a second hand
store, at a seaside town... somewhere like Lynmouth if memory serves.
I can't find it on the internet. It don't know if dreamed this -- or i
saw such a book myself .... Wait!!! its on Amazon -- "PASCAL with Your
Basic Micro" - £12.20 [2]
Alex
[1] http://www.eastgate.com/Tinderbox/TinderboxWay.html
[2] http://www.amazon.co.uk/PASCAL-Basic-Micro-Jeremy-Rushton/dp/0672220369/ref=sr_1_2?s=books&ie=UTF8&qid=1321219955&sr=1-2
> A step back, though: I'd like to point that as tiddlywiki.org has
> migrated (and unfortunately not fully migrated) to [2], one thing got
> worse for sure: when one meets a MediaWiki page, he or she knows at
> once that he *can* contribute. The TiddlySpace version needs big link
> "how to contribute" to make this clear. But anyway, this rapid
> migration haven't solved the main problem: the need of fundamental
> text, not a pile of notes on different common things about TW.
For the sake of history:
The hope, apparently not entirely well-founded, was that starting the
migration would get the ball rolling and the community at large would
do the rest. This didn't happen. A small number of people helped to
migrate additional content beyond what I started. Very small number,
relative to the number of people reading this group.
To me that was evidence that the information was not considered
relevant by the community. That's perhaps harsh, but that's how it
looked and it makes future efforts to improve the situation harder to
get off the ground.
I agree that adding a big "you can edit this!" and "how to contribute"
would be excellent.
>> There's a whole lot of non-geeky people who won't use a tiddlywiki
>> because it doesn't seem user friendly.
>
> Well... it's also not all that developer friendly. And THAT's the rub.
+1
I'm registering this +1 here because I don't want this comment from
Miles to pass by without notice. We all agree that TiddlyWiki is clever,
useful, interesting, thought provoking and revolutionary at its time of
birth. But there's an argument that can be made, evidence being what
Miles has said, that its structure and style resist developers that are
"trained up" on modern techniques and development tools.
This isn't necessarily a bad thing[1] but it is a thing that exists and
is a force that should be recognized as we attempt to evaluate ways
to improve community engagement and support.
I think that TiddlyWiki can't be all things to all people. That's
the job of the tiddler.
[1] There's another argument to be made the modern development
techniques and communities are exclusionary priesthoods and
TiddlyWiki is a power to the people kind of thing, but if that's the
case then the need for a TiddlyWiki bible from which the people
might learn their power becomes a greater imperative than it already
is.
Hi Miles. As Chris states, we've been looking at and building
non-TiddlyWiki based apps on top of TiddlySpace for a while now.
There are some examples around TiddlySpace that you can look at:
* http://apps.tiddlyspace.com - This is a jumping off point for
several apps that Jon Robson (mostly) has written.
* http://tiddlyace.tiddlyspace.com - This app provides a simple code
editor witht the ability to create new HTML/JavaScript/CSS tiddlers,
etc (good for mocking up/developing around new apps). It's based on
the ACE project (https://github.com/ajaxorg/ace)
* http://tiddlybookmarks.tiddlyspace.com - This provides bookmarking
functionality that lets you save websites back to TiddlySpace (a bit
like Delicious). While the interface is currently TiddlyWiki based,
the bookmarklet is not, and is instead built using Twitter Bootstrap
(http://twitter.github.com/bootstrap/)
If you're interested in developing on top of TiddlySpace (but outside
TiddlyWiki, then there are several things that might help you.
Standard frameworks like Backbone.js and Sammy.js should just work
(with some minor modifications to the syncing mechanisms to convert
models to tiddlers and send them to the right URIs, etc).
I've also been working on a new set of libraries, which I'm currently
calling TiddlyLib, that are designed to take advantage of both the
Tiddly way of doing things, and modern web development methods and
best practises. It's currently unfinished, but a few parts that are
involved are now quite stable and usable.
You can find out more information on TiddlyLib in the TiddlyLib space,
at http://tiddlylib.tiddlyspace.com, which I'm hoping to expand with
more info/docs/demos/etc soon.
Hope that's of some use
Ben
> So, this was not about the TiddlySpace technology?
No, in the original thread that discussed the move the goal was to
move to a system that provided two features that tiddlywiki.org on
mediawiki did not have:
* resistance to spam
* using tiddlywiki for sake of tiddlywiki markup
I'd suggest hosting one's own email list - install mailman or sympa on
tiddlywiki.org. There are pretty good archiving options for both of them.
Thanks for the pointers, but they sort of reinforce that it seems less
work to build from scratch.....
Ben Gillies wrote:
> On Mon, Nov 14, 2011 at 4:33 PM, cd...@peermore.com
> <chris...@gmail.com> wrote:
>> On Nov 13, 7:26 pm, Miles Fidelman<mfidel...@meetinghouse.net> wrote:
>>> I expect we're going to go off and write our own framework for
>>> single-page applications - because it sure looks a lot easier than
>>> buliding on Tiddly. That's sort of a shame.
>> If you're into tiddlers, but not TiddlyWiki, you might have a browse
>> around http://tiddlyspace.com/, looking at some of the non-tiddlywiki
>> applications that have been created there. Ben Gillies and Jon Robson
>> ought to be able to point out some good examples.
First off - we're specifically into SPAs that can run locally - that's
the primary attraction of TiddlyWiki. There are LOTS of ways to write
hosted applications.
If we WERE interested in building on tiddlyspace:
> Hi Miles. As Chris states, we've been looking at and building
> non-TiddlyWiki based apps on top of TiddlySpace for a while now.
>
> There are some examples around TiddlySpace that you can look at:
>
> * http://apps.tiddlyspace.com - This is a jumping off point for
> several apps that Jon Robson (mostly) has written.
absolutely no documentation behind any of those
>
> * http://tiddlyace.tiddlyspace.com - This app provides a simple code
> editor witht the ability to create new HTML/JavaScript/CSS tiddlers,
> etc (good for mocking up/developing around new apps). It's based on
> the ACE project (https://github.com/ajaxorg/ace)
Again, no documentation or anything to give a clue what this thing is.
We keep coming back to the same conclusions:
- TiddlyWiki is a great example of what's possible for with SPAs
- it's pretty useful in it's current form (though the changing
browser/landscape seems to lead to compatibility issues)
- it's virtually impossible to understand what's going on under the hood
without completely deconstructing the raw code - hence providing an
incredibly steep learning curve for even making small modifications,
much less using it as a platform
- hence, it's better as an existence proof, and a source of ideas - but
it's easier to redesign from scratch to get to a useful platform
Miles
> Thanks for the pointers, but they sort of reinforce that it seems less work to build from scratch.....
All of those examples are plain HTML and JavaScript so other than the reliance on the TiddlySpace server for persisting data they pretty much _are_ built from scratch.
> First off - we're specifically into SPAs that can run locally - that's the primary attraction of TiddlyWiki. There are LOTS of ways to write hosted applications.
Sure. The point of these apps is that they make use of TiddlySpace, not that they run from a file:/// URI. Some of them do work offline, but that's because of localStorage and cache manifests rather than TiddlyWiki.
>> * http://apps.tiddlyspace.com - This is a jumping off point for
>> several apps that Jon Robson (mostly) has written.
>
>
> absolutely no documentation behind any of those
That's true. I mentioned them more as an example of SPAs that have been written on TiddlySpace without using TiddlyWiki.
>> * http://tiddlyace.tiddlyspace.com - This app provides a simple code
>> editor witht the ability to create new HTML/JavaScript/CSS tiddlers,
>> etc (good for mocking up/developing around new apps). It's based on
>> the ACE project (https://github.com/ajaxorg/ace)
>
>
> Again, no documentation or anything to give a clue what this thing is.
Same with this one.
> - it's virtually impossible to understand what's going on under the hood without completely deconstructing the raw code - hence providing an incredibly steep learning curve for even making small modifications, much less using it as a platform
All of those examples are plain HTML and JavaScript built in much the same way as any other SPA is today. There really is nothing going on under the hood with them (indeed that's sort of the point).
> - hence, it's better as an existence proof, and a source of ideas - but it's easier to redesign from scratch to get to a useful platform
How "from scratch" do you mean here? TiddlySpace provides a backend server that you can write SPAs on in JavaScript "from scratch". As long as you follow the RESTful API, you can use whatever (e.g. Backbone) you want.
On the other hand, if by "from scratch" you mean something like Rails or Django, then yes, depending on what you're building, that may be a better fit.
Ben
> - hence, it's better as an existence proof, and a source of ideas - but it's
> easier to redesign from scratch to get to a useful platform
This is the case with about 95% of the open source out there in the
world, and is not necessarily a bad thing. If everything was already
perfect there would be no evolution and innovation, perhaps?
Which is not to excuse the current state of affairs, nor to explain
it, but rather to just indicate that it is fairly normal and there's
no catastrophe in progress here.
I'm not quite sure why you keep going back to TiddlySpace. What makes
TiddlyWiki so special is the model of a "wiki in a single file" or "wiki
on a stick." Complete with themeing, code updating and a plug-in
architecture. That's the beginning of a serious platform.
If you look at a server-side equivalent - something like WikiMedia, or
Drupal, or WordPress, or perhaps CouchDB, or eXist - the systems have
clear, and clearly documented architectures - it's pretty clear how to
write a new theme, write a new module, access various APIs, write an
application on top of the platform, etc.
That's pretty much missing for TiddlyWiki - and it's a real obstacle to
anyone who'd actually like to use or extend the platform.
to which chris...@gmail.com wrote:
> This is the case with about 95% of the open source out there in the
> world, and is not necessarily a bad thing. If everything was already
> perfect there would be no evolution and innovation, perhaps?
> Which is not to excuse the current state of affairs, nor to explain
> it, but rather to just indicate that it is fairly normal and there's
> no catastrophe in progress here.
Keep in mind that this is a thread on the BUSINESS side of TiddlyWiki -
starting from Eric's statement about finding a way to get paid for some
of his TiddlyWiki work, and then Jeremy's posts about going out on his
own and focusing on "making TiddlyWiki good."
My comments have been offered as feedback in the context of open source
business models - what it takes for a Drupal or CouchDB to survive and
thrive as a platform, complete with a model for supporting a core group
of developers, and a model for attracting and maintaining a larger
development and user community - and what needs to happen for TiddlyWiki
to move into that league.
IMHO, Tiddly has (or at least had) the potential to become a pretty
serious platform, for a wide range of stand-alone, single page
applications. A lot of my comments have been motivated by working on a
project where I would really like to have used TiddlyWiki as a jumping
off point � and just finding too many obstacles in the way - leading to
our conclusion that we're probably better off going our own way (if our
project gets funded for Phase II).
Miles Fidelman
I probably should add: What's really somewhat disappointing is that
Osmosoft, purports to be "a small team of developers at BT
<http://bt.com> who primarily make Web based collaboration tools
<javascript:;>. These tools are used by Osmosoft in products
<javascript:;> developed for BT, and in engagements <javascript:;> with
BT customers and independent organizations."
Small team, in a bigger enterprise, with business needs driving a piece
of software � that's been a pretty good recipe for launching an open
source project with long-term "legs" (e.g., Postfix, Zope, Erlang).
(The other environment that tends to generate open source projects with
long-term viability is academia - e.g., Apache's origins as the NCSA web
daemon, Mozilla, Sendmail.
You wrote:
> I expect we're going to go off and write our own framework for
> single-page applications - because it sure looks a lot easier than
> buliding on Tiddly. That's sort of a shame.
To which cdent mentioned that non-TiddlyWiki stuff is still possible
within (something like) TiddlySpace:
> If you're into tiddlers, but not TiddlyWiki, you might have a browse
> around http://tiddlyspace.com/, looking at some of the non-tiddlywiki
> applications that have been created there. Ben Gillies and Jon Robson
> ought to be able to point out some good examples.
To which I replied with some examples from TiddlySpace (hence why they
all mention TiddlySpace). If the bit you're interested in is
TiddlyWiki as an offline, single page wiki, and not tiddlers in
general, then none of what I've said is likely to have been of much
use to you. That's fine.
> If you look at a server-side equivalent - something like WikiMedia, or
> Drupal, or WordPress, or perhaps CouchDB, or eXist - the systems have clear,
> and clearly documented architectures - it's pretty clear how to write a new
> theme, write a new module, access various APIs, write an application on top
> of the platform, etc.
>
> That's pretty much missing for TiddlyWiki - and it's a real obstacle to
> anyone who'd actually like to use or extend the platform.
Yep. I don't think anyone here disagrees with you on that point.
* TiddlyWiki does indeed lack the level of documentation that one
would expect. Originally, TiddlyWiki itself was pretty small, and for
a long time I carefully maintained the idea of a good read-source
experience: that someone with a little experience of JavaScript could
reasonably expect to read through the source code in a weekend. As
TiddlyWiki has got more complex that read-the-source experience has
got less useful, and the lack of documentation is more painfully
apparent
* TiddlyWiki has suffered from a lack of impetus over the last few
years, in terms of keeping up with the changing browser landscape, or
taking advantage of new things like node.js
* Over the period that I was at Osmosoft we created TiddlyWeb and
TiddlySpace specifically to try to make something that would be useful
in BT, that would take the insights we've learned from TiddlyWiki and
apply them in a way that made sense in the enterprise. Accordingly, as
TiddlyWeb and TiddlySpace have matured, the lion's share of Osmosoft's
attention shifted to them
* My business plan is to try to have a handful of consultancy clients
that pay for half my time, so that I can work on open source in the
other half -- and thus get more interesting consultancy opportunities.
* My open source focus is going to be TiddlyWiki. As I've said, I'm
currently working on two basic things: the documentation and
experience of tiddlywiki.com, and updating the build tools. Then I
intend to pursue the ideas for a next generation TiddlyWiki that I
last explored in TiddlyWiki5.
* I liked Alex's book idea; it is something I've discussed with Eric
over the years, too. At the moment, I'm acutely conscious that I'm
barely keeping up with the writing that I need to do around
tiddlywiki's documentation, and think I might struggle with a big
project like a book. I wonder if it would be possible to fund the
writing of the book through Kickstarter?
* I don't think we need to replace or close the Google Groups. It
takes human effort to transform the transcript of a conversation into
useful, reusable content. One way to improve things might be for the
regular contributors to the group to adopt an etiquette of writing the
answers to questions on tiddlywiki.org directly, and point to those
answers from the Google Groups thread.
* It's for Matt and others at Osmosoft to say, but I suspect that me
picking up the reins with TiddlyWiki will be a welcome opportunity for
them to focus whole-heartedly on TiddlyWeb and TiddlySpace.
Miles, and everyone else, I really appreciate your perspective, and
thank you for your willingness to engage in this dialogue.
Best wishes
Jeremy
> --
> You received this message because you are subscribed to the Google Groups "TiddlyWiki" group.
> To post to this group, send email to tiddl...@googlegroups.com.
> To unsubscribe from this group, send email to tiddlywiki+...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/tiddlywiki?hl=en.
>
>
--
Jeremy Ruston
mailto:jer...@osmosoft.com
http://www.tiddlywiki.com
I think this is a good idea to explore. I wonder if a professional
technical writer might be employed by the funders to edit / draw
together materials. It would be a shame to divert Jerm's and ELS's
attention from TW
In some was the organisational activity of producing a book about
TiddlyWiki may help resolve other organizational issues which when
combined with TW technicalities have stalled in the past.
Alex
Anyway, can someone describe how to contribute in TiddlySpace?
I think this is essential and according to the concept, should be
written to tiddlywiki.org now (the link should be right near the
"welcome" link on the top), despite the fact that it's rather about
TiddlySpace than TiddlyWiki.
Besides, the idea seems to be quite good -- at least it implements
cumulativity right away. Although, the bigger the tiddlywiki.org
becomes, the longer it should take to load it (as I understand); and
for real cumulativity the space should be organized from time to time
(I mean, tiddlers with aspects in vacuum, with only the tag cloud and
search to navigate is quite bad at least for consistent reading, there
should be sort of table of contents).
* I liked Alex's book idea; it is something I've discussed with Eric
over the years, too. At the moment, I'm acutely conscious that I'm
barely keeping up with the writing that I need to do around
tiddlywiki's documentation, and think I might struggle with a big
project like a book. I wonder if it would be possible to fund the
writing of the book through Kickstarter?
This is a very exciting idea. I guess the big question is who is
actually going to be the principal writer or editor of the work. My
focus has got to be ongoing development work, and so it can't be me.
But I'm in a good position to provide help and support to the main
author, and of course can help track down definitive answers about the
code.
Cheers
Jeremy
--
You received this message because you are subscribed to the Google Groups "TiddlyWiki" group.