I think Jeremy's plans are sound and would be happy to follow his suggestions.
Joshua On Tuesday, September 3, 2013 9:14:21 AM UTC-5, Jeremy Ruston wrote: > > I'm grateful to everybody for taking part in this discussion. It's very > helpful to bring things to the surface, and I'm tremendously appreciative > that there is interest in my activities. > > The original question was about splitting this group, and the rationale > included some comments about the relationship between TWc and TW5 that are > worth responding to in some detail before turning to the original question. > > The first thing to note is that TiddlyWiki classic has been slowly dying > since BT acquired Osmosoft in 2007. Here's the Google Trends graph of > search volumes, usually a pretty good metric of interest: > > http://www.google.co.uk/trends/explore?q=tiddlywiki#q=tiddlywiki&cmpt=q > > That's the time frame that Osmosoft invested significant effort in the > TiddlyWiki universe. It has enjoyed the attention of a devoted but > dwindling core of end users, but it hasn't been enough to stem the decline. > > Against that background of dwindling interest, one might ask why I'm > investing my own time and energy of working on a new version at all. > Fundamentally, the reason is that I think that the success of TW classic > thoroughly validates the idea of a single file wiki but that the execution > was too flawed to sustain the attention it deserves - particularly the > attention of developers. > > I have personal experience of this issue while I was running Osmosoft > inside BT: there was little enthusiasm from many of the developers to work > on the TiddlyWiki core code. Coming from a conventional development > background, TWc is a very strange beast. I wrote most of it while I was > learning JavaScript and HTML, and at a time when TiddlyWiki was blazing a > trail, pushing browsers beyond their commonly understood constraints. There > was no jQuery and so a lot of my effort on TWc was just to get the code > working consistently across a range of browsers. > > At Osmosoft we successfully developed TiddlyWeb, a clean, well engineered, > production quality serverside for TiddlyWiki. But we struggled to get the > TiddlyWiki side doing everything that we wanted. We were working against > assumptions made in the core code, and in the flock of popular plugins. > Over the years I think Osmosoft did a pretty good job on the suite, but > even today there are some troubling limitations - for example, TiddlyWiki > under TiddlySpace can't dynamically reflect the changes made by other users > to the same space. > > Other technical limitations of TiddlyWiki have dogged it from the start: > for example, there is still not a good experience for search engines to > index TiddlyWiki. Not everybody cares about such capabilities, but those > that do have little choice but to abandon TiddlyWiki. > > So, for a while, TiddlyWiki has been an unfriendly environment for > developers, but with lots of limitations that really need developer effort > to overcome. > > Turning now to end users, with open source software a good rule of thumb > is that the more users a piece of software has, the more useful it becomes. > Accordingly, I believe that the best way to assure TiddlyWiki's future is > to try to make it much more popular than before. I bandy around the figure > of making it 100 times as popular. > > I believe that we can make it that popular by making it easy to learn and > easy to use, and ensuring that the underpinnings are flexible enough to > cope with whatever the future throws at us. As I said before, the basic > proposition of TiddlyWiki has always seemed to go over quite well with > prospective users. This time around I think we can do a much better job of > explaining it to new people. > > If we achieve this increase in popularity, then the community of incoming > TW5 users will rapidly outnumber our core community of TWc users. > Therefore, I believe that we should optimise the experience of discovering > TiddlyWiki for new TW5 users, and we need to do that now. > > In my opinion, TiddlyWiki5 and TiddlyWiki Classic are different versions > of exactly the same product, with the same goals and the same basic design. > There are incompatibilities in the way that they are customised and > extended, but the elevator pitch is precisely the same: a JavaScript wiki > that works from a single HTML file. TW5 > > The changes in TW5 are resolutely intended to benefit end users. For > example, the new plugin architecture makes it possible to drag-and-drop > plugins between wikis, not just cut and paste them. And now a plugin can > contain a bundle of related tiddlers (which is handy for developers) while > still being a single unit for end users. > > As I've shown above, underpinning those improvements with a good developer > experience is important for end users because without a stream of > developers interested in working on TiddlyWiki and it's plugins we wouldn't > have the rich world of extensions and adaptations that makes TiddlyWiki so > useful in so many different niches. > > There a few misunderstandings in the thread that suggest that I need to do > a better job of explaining TW5. For instance, the idea that the presence of > the node.js edition of TW5 might somehow compromise/diminish/complicate the > single file edition. The truth is that TiddlyWiki has since 2006/7 had a > complete set of command line tools to enable the developers to work on the > core. They were kind of klunky weird things written in Ruby, with a high > learning curve. All that's happened in TW5 is that that job of building a > single file wiki from it's constituent parts is undertaken by TiddlyWiki > itself. This makes the tiddlyverse easier to use and easier to understand. > > Finally, turning to the original question about splitting the Google > groups, I have a couple of observations: > > * Experience suggests that people find it hard to figure out which Tiddly* > group they should address with a particular post. There are already too > many groups > * The TiddlyWiki group is pretty low traffic; compare it with something > like the node.js google group: > https://groups.google.com/forum/?fromgroups#!aboutgroup/nodejs > * I don't think we can rename the existing group without messing up links > to it > * Based on past behaviour, Google is probably going to kill Google Groups > as soon as they decently can > > So, my response to the original suggestion is: > > 1) We should add a welcome message to the TW groups asking people to flag > their subject lines TWc or TW5 (I've actually already done this) > > 2) We plan a migration away from Google Groups: > a) TiddlyWikiDev could migrate to StackOverflow; several other open source > projects use it in that way > b) The main TiddlyWiki group could migrate to a homegrown > TiddlyWeb+TiddlyWiki5 host > > I haven't had time to address all the points raised in the thread, do > please fire away with any questions or clarifications. > > Best wishes > > Jeremy > > > > > On Tue, Sep 3, 2013 at 7:40 AM, David Johnston > <[email protected]<javascript:> > > wrote: > >> It is amazing how a single knee-jerk response to change can derail >> significant development effort. >> >> Change happens constantly, the early edge of change seems technical and >> often confusing to people because it has not been refined and they are not >> comfortable with it yet. TiddlyWiki 5 is at that stage right now, it feels >> like more effort to use, while TWc feels like less effort. However, with >> refinement and development TW5 will be more capable and easier to use than >> TWc and built upon a much better structure, which will enable it to take >> advantage of modern browsers into the future. Whereas TWc will over time >> become less and less able to perform effectively, it will require more and >> more patching/development effort to keep it limping forward because the >> structure it is based upon will be out of date compared to the browser(s) >> you will view it in. >> >> My advice as a programmer would be to support the development of TW5 as >> much as possible, I feel it is aiming at the correct mix of technologies to >> ensure survival in the future. Jeremy is doing an excellent job from what I >> can see of the code and with support and encouragement will deliver a >> product which will supersede the achievements of its predecessor. >> >> >> -- >> 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] <javascript:>. >> To post to this group, send email to [email protected]<javascript:> >> . >> Visit this group at http://groups.google.com/group/tiddlywiki. >> For more options, visit https://groups.google.com/groups/opt_out. >> > > > > -- > Jeremy Ruston > mailto:[email protected] <javascript:> > -- 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 http://groups.google.com/group/tiddlywiki. For more options, visit https://groups.google.com/groups/opt_out.
