From aja at si.on.ca Fri Nov 9 14:23:47 2007 From: aja at si.on.ca (Anton J Aylward) Date: Fri, 09 Nov 2007 14:23:47 -0500 Subject: [Informl-talk] Introduction Message-ID: <4734B3C3.8050709@si.on.ca> Let me take this opportunity to introduce myself. I'm Anton Aylward. I've been using Wikis since 2000, installed my first in 2001 and became involved in the TWiki project in 2002 and have installed a number of TWiki Wikis since then. TWiki is probably the most "5th generation" wiki out there, a complete set of tools and application builder rather than just a "white board". It has a number of ideas "done properly" that have crept half heartedly into other CMSs. However it has a number of very serious shortcomings that are best described as "quick and dirty hacks to get a 'proof of concept' system up and running" that have become key and immovable parts of its architecture. That and an obsession with backward compatibility. Its current revision is battling with matters that are no-brainers or built-ins for Ruby or RoR. As a result, I'm unhappy with it and looking for a new wiki. Informl is still in its infancy but has great potential. I'm going to post some follow-ups here with ideas that I'd like to discuss. I don't want to hijack the agenda and I'm not that experienced with Ruby, and I'm sure that while I can say "look see, that's a cool idea", other minds than mine can say "well I can do it better/smaller/faster". -- The emphasis should be on "why" we do a job - W. Edwards Deming From aja at si.on.ca Sat Nov 24 11:51:39 2007 From: aja at si.on.ca (Anton J Aylward) Date: Sat, 24 Nov 2007 11:51:39 -0500 Subject: [Informl-talk] What we can learn from other Wikis - 0 Message-ID: <4748569B.4080300@si.on.ca> I see Wikis in terms of 'generations'. How familiarity has allowed people to become more proficient at integrating web technology into making them into applications rather that just whiteboards. The First Generation Wiki was Cunningham's original written in perl. It was a plain whiteboard with n access control or filtering and no styling or decoration. The clones that followed went in a number of directions, but the plain whiteboard stayed the key of a wiki. Very few developers saw this as a platform for developing applications, and the wiki ethos of an open platform with no access control, that everyone should be polite nd relying on backups and history has predominated even in spite of various abuses. So I define the second 'generation' of wiki-dom to be concerned with styling and decoration. Wikipedia is an example of this; the use of style sheets and 'framing' and navigation tools. Note: I'm not concerning myself with the underlying technology, language, storage or whatever. A third 'generation' brings in tools for management, better searching, for example of the body rather than the title, renaming & redirection, roll-back, better editor interfaces, attachments and of course access control. Access control is controversial in early wiki-dom. A Wiki is based on open free-form editing as a collaboration tool. Early exponents believed that peer influence and editing would be enough. The assumed a benign community. The Internet showed that wasn't the case. Other claimed that corporate intrAnet wikis were different, but even there access control was needed for different reasons. Most Wikis stop there. Sadly. WikiMedia may be immensely popular,but its still essentially _just_ a whiteboard. There are just a few wikis that allow for applications to be built. Most 'tools' end up going the other way, they are like Documentum or Moodle - http://moodle.org/ - where the 'wiki/whiteboard' is a plug-in that is a scratch-pad or a substitute for a BBS. So the Fourth Generation of Wiki-dom is application builders. This means database, form for inputs and queries. The only 'mature' Wiki with all that is Twiki. The only other potential candidate is Informl. That is why I think we can learn from Twiki. We can also learn what NOT to do as well. In "The Mythical Man Month", Fred Brooks suggested "build one to throw way". Twiki was a great 'proof of concept' implementation, light-weight using the UNIX file system, RCS, grep an other tools. But somewhere along the way that should have been thrown away. Despite the evidence of any other systems, he originator wants to hold on to the flat file format with embedded metadata. I have long criticised this. Some developers have pushed for implementing the 'rcs' and 'grep' in perl and showed it to be faster. But a good DB schema would simplify many execution paths. In subsequent mailings I'm going to discuss abstract ideas and use Twiki as an example of how these are useful. -- The surest way to corrupt a youth is to instruct him to hold in higher esteem those who think alike than those who think differently. -- Friedrich Nietzsche From tech at onghu.com Tue Nov 27 22:17:57 2007 From: tech at onghu.com (Mohit Sindhwani) Date: Wed, 28 Nov 2007 11:17:57 +0800 Subject: [Informl-talk] What we can learn from other Wikis - 0 In-Reply-To: <4748569B.4080300@si.on.ca> References: <4748569B.4080300@si.on.ca> Message-ID: <474CDDE5.6090200@onghu.com> Anton J Aylward wrote: > I see Wikis in terms of 'generations'. How familiarity has allowed people > to become more proficient at integrating web technology into making them > into applications rather that just whiteboards. > > <..snip..> > > In subsequent mailings I'm going to discuss abstract ideas and use Twiki as > an example of how these are useful. > > Hi Anton Thanks for the very informative and interesting post. I look forward to the subsequent messages. Cheers, Mohit. 11/28/2007 | 11:17 AM.