Tags:
create new tag
, view all tags

How Do We Get To Where We Want To Be

Following up on comments in GettingPaidToDevelopTWiki

I make no apologies for the topic name. An old boss of mine once said, when asked what the mission statement of the company was, "We want to get to where we want to be". This is about as useful to a high-tech startup as sliced yoghourt. Bob (for that was his name) was an accountant - basically, he was completely out of his depth with the technology. When faced with questions about adoption, and how the new user feels, and what colour it should be, and why people don't contribute, many of us feel a bit like Bob - we don't know where to start, because we're nerds.

Most, if not all, of the people working on TWiki are highly experienced engineers, used to problem solving. We don't perceive issues that bite new or inexperienced users, and we don't write good documentation because we don't perceive these issues. We are all aware of it, and all try to do something about it, but I - for one - feel like a frog swimming in treacle. All I need is to know is the best way to swim, but all I get is more treacle.

Basically I find it rather frustrating that there is a lot of criticism of the "user experience" with TWiki, but very little in the way of constructive initiatives to improve, or even describe, the situation.

So I am left asking, why?

  • Is it because there is just so much to do, people don't know where to start?
  • Or because they don't feel welcome?
  • Or because they can't be bothered, it's easier to go use another wiki instead?
  • Or maybe because it's easy to chuck stuff over the wall and then complain when it isn't adopted?
  • Or all of the above? Cumulative frustration at a lot of little things?

-- CrawfordCurrie - 10 Apr 2005

I edited in between, because i feel we're falling in a common trap for engineer-types: thinking in solutions. The last part of Crawford's question: '...where we want to be'. And immediately everybody discusses hows and whys.

I've the feeling that discussion keeps popping up regularly. And I have the feeling that it's the where part. Now for my ideas what might be the problem (notice i'm also avoiding the where part :-))

Most of the active contributers are programmers (more or less). I assume they contribute to TWiki for fun. So when someone who cannot contribute as much comes along and complains/explains that the focus should be on making TWiki prettier or more user friendly, i assume (correct me if i'm wrong) that working on TWiki will seem more like work than like fun.

So development directions and feature requests aside: do we really want to develop a succesful wiki engine? Or do we want to develop a wiki engine that helps us in our daily routine?

For the former case, we need to:

  • strictly define the direction (the where) of TWiki
  • give equal importance to all aspects that are needed to arrive at that place (security? speed? user friendlyness? attractiveness? marketing? ...)
  • let non-techies throw issues 'over the wall', and work on these issues in the light of the strategy

For the latter case, we can continue as TWiki is doing right now. And that is not bad at all.

-- JosMaccabiani

Some constructive ideas moved over from GettingPaidToDevelopTWiki. Useless unless someone actually undertakes to do them, of course.....

Make a simple ftp upload, easy configuration for the average person and twiki will go far. Simplicity simplicity simplicity...say it again..simplicity. I would love to trim a version that could be offered as a simple alternative for a basic, pleasant to install and use twiki. -- BruceRProchnau

go for it, Bruce. All it would take (at best) is an equivalent to build-twiki-kernel.pl that builds the simpler package or (at worst) a shell script that trims an existing install and rezips it. The only thing stopping you is you.

... what about some kind of form/survey for the user to fill out that will return a list of suggested plugins that they might consider depending on needs? "Will your TWiki be internet accessible?" if so there is an entire series of plugins they should probably consider that are basically useless if it is to be a LAN site. This is just one example, I'm sure you see my point. Making TWiki more user friendly means helping narrow down the options a bit, making it easier for the user to filter through the noise and get at the bits they want/need -- TravisBarker

great idea Travis - but isn't the fundamental limitation that there is really no ClassificationOfPlugins, so it's hard to know where to look? And really hard to know beforehand if a plugin is any good?

Easy installer

The only thing stopping you is you - I don't agree with this. This is not Bruce's personal problem. I would like to put this (a simpler installer) on the agenda. If I remember well, there was (is?) a unix installer, but no windows installer.

-- ArthurClemens - 10 Apr 2005

Plugins

isn't the fundamental limitation that there is really no ClassificationOfPlugins - That is lacking indeed. But the real problem is that plugins do not give good insight in what they actually do. Because on twiki.org most plugins are disabled, the plugins cannot be seen in action and thus give no insight. Few have screenshots.

This is strange, because one of the strengths of twiki are the plugins. But they are nowhere advocated.

What we can do:

  • Ask plugin authors to write an introduction for dummies, no knowledge required, with the format:
    1. What problem does the plugin solve
    2. Where can it be used
    3. What is special about it
    4. ... please add
  • Use the descriptions to generate lists (manually) o the Plugins homepage
  • Create a Demo installation (demo.twiki.org) with all plugins enabled. Each twiki.org plugin page should have a link to a demo page.
  • Use the demo installation to create screenshots.

Follow up in CreatePluginDemoInstallationOnTWikiDotOrg

-- ArthurClemens - 10 Apr 2005

Sorry, I disagree. It is Bruce's personal problem - he stated that he "would love to trim a version". My point is that nothing is stopping him. Open source software is exactly that - it is open source, and if anyone has a beef with it they can act to change it. If support is no good, then contribute to support and ease the burden on others. If documentation is no good, write better documentation. Otherwise comments like this are just more treacle for those of us who are prepared to act.

On installers; you might as well write off the "existing installer", it is just too long ago to be useful. However, judging from MartinGregory's experiences writing UpgradeTWiki, an installer should actually be quite straightforward to write - all it requires is someone to take the bit between their teeth and finish the job.

-- CrawfordCurrie - 10 Apr 2005

Bruce is not the first to state that TWiki is hard to install. But then again, someone has to do the work.

Bruce, it always helps to break up the problem in manageble parts. Please describe what you mean with "trim" and "easy configuration" in a FeatureRequest topic, and write what steps should be taken, and how you can help.

-- ArthurClemens - 10 Apr 2005

Having a demo area where all plugins are installed would also point out quickly

  • inconsistencies between plugins (where plugins interfere with each other)
  • poor documentation (if users don't understand what the plugin does)
  • when a plugin does not work any more in the current release.

I think this would be a great idea, and really advertise TWiki. Currently very few (if any) truly understand the full power afforded by all plugins working together (and where this model breaks down).

-- ThomasWeigert - 10 Apr 2005

I just read this, but Travis' idea of creating domain related installations (GettingPaidToDevelopTWiki) can be the second step in a demo installation.

Before we create applications, we can first create (overview) domain topics with problem-solution pairs, demos and screenshots.

-- ArthurClemens - 10 Apr 2005

Creating a twiki installer seems like a good example of where it makes sense for GettingPaidToDevelopTWiki. It's not the kind of "sexy" project that would inspire volunteer coders but presents clear benefits to the entire twiki community.

-- LynnwoodBrown - 10 Apr 2005

Hopefully I can feel that I can contribute. Not by "complaining" but constructively. The engineers need the painters to paint the bridge, the painters need the engineers to build it. Complimentary roles work when that realization is there. I guess I felt that while I could do whatever with twiki on my site, that was as far as it could go. I realize that documentaion is difficult, so many variations and differences both in install environments and feature requirements. It becomes a time issue. I have been unsure about even writing anything here in codev, maybe I shouldn't as I'm not one of the developers...but if I can contribute constructively, and even learn by doing so to undertake the part which I can do well, then perhaps its ok.

-- BruceRProchnau - 10 Apr 2005

It certainly is ok. You would be surprised how much impact just one person can have - for example, if it wasn't for AntonAylward testing out the registration and authentication code, and badgering MartinCleaver and myself to support him in doing that, Dakar would be significantly less functional than it is. All he did was to provide feedback and focus, and it worked wonderfully (the treacle got a little bit softer for a while).

-- CrawfordCurrie - 10 Apr 2005

Example of "trim"

http://www.bkdesign.ca/twiki3

A minimal TWiki.Not sure how this relates to the topic, go forwards by going backward? lol, it's hardly going backwards tho, it could become a popular install for small sites.Docs and simple ftp install will follow. I branched this to TestTWikiBasic. This ok?

-- BruceRProchnau - 11 Apr 2005

It's hard to tell how it's been trimmed, apart from the obvious purge of the plugins.

This might well be the first example of a TWikiDistribution.

-- CrawfordCurrie - 11 Apr 2005

In response to Jos' points above, I am certainly more aligned to the former; I want to see TWiki being a successful wiki engine. However, the reason I want it to be successful is that it is then a secure, stable platform for all the other stuff I want to do. Perhaps I am alone in this, which would explain why I am so frustrated. Maybe I should just go and build on MoinMoin instead; at least the people seem to care over there.

-- CrawfordCurrie - 11 Apr 2005

Bruce - I resisted the temptation previously, but I do think it should be said....

Please join the DevelopBranch with us! You have done a pile of work simplifying TWiki in a way that I have not found the time (or even the idea). Commit your changes to Develop (), and they will be apreciated (though there will always be people compaining about decisions, even that is a form of apreciation).

So, please make a "WouldLikeToCheckin" topic, and describe your vision - I'd love to see what becomes of TWiki with your help.

-- SvenDowideit - 11 Apr 2005

Thank you Sven. The comments are appreciated a lot.

I am working out what I'm trying to do specifically. Crawford, the Main web, Sandbox, Default, and a lot of other stuff has been trimmed, and I just started yesterday, building on previous work with twiki and MT. I have a lot to do yet, maybe I listed it here too soon, as its unfinished..,

Lets take MOvable Type as an example. You get a very basic weblog on download. No topics, no extra blogs. You install it and you have a website. No content, no plugins, nothing.

Yet it's been downloaded over a million times. After you install it, you can customize it if you wish. There is probably fifty or more plugins you can add if you want them. Most pick the ones they want, and only the ones they want. They have documentation problems as well, in fact they just hired someone to work on the docs for clarity for new users.

Main point? Most programs come small and very basic, and give users the ability to choose additions to the basic core. What this would be couldn't be called an add on, it's the opposite really...not sure what yet wink

What I am coming out with (?) wouldn't need an installer, not even Movable Type or many others have one. Last night I uploaded this "stripped down" TWiki by ftp to a folder, it was up and running in ten minutes after the normal configuration in TWiki.cfg. I still have an original "out of the box" TWiki on my server. In a year I haven't read all the topics in it, probably never will. I think that the average user is intimidated with three webs and all the content from the word go. Not that this is a complaint, it is an awesome program...but most people want to make their own site from scratch as far as content goes. I know most have the docs on the side to read, not part of the site itself. This would be an alternative to the present premade site perhaps, or a separate thing altogether...I will also say this is not a criticism in any way, perhaps an alternative for the "little guy" - small business/personal small site?

A good example of "trimmed" - TWiki "out of the box" without the pub folder is -5213 kb

The TWiki "trimmed":

http://www.bkdesign.ca/twiki3 - 1531 kb

Not that size matters so much, but this shows a minimal install, perhaps for a specific demand. Now I better shutup as I am using up way too much paper...

-- BruceRProchnau - 11 Apr 2005

This is easier to install than Movable Type.

-- BruceRProchnau - 15 Apr 2005

Bruce, can you please give us a manifest (preferably in a new topic (MinimalistTWiki?)

-- CrawfordCurrie - 15 Apr 2005

Can you define that more pls?

-- BruceRProchnau - 16 Apr 2005

These installs should have the Copycatskin template with a few changes and a new stylesheet, I just deleted the webs and added topics needed to make it work. Some of these are disorganized yet...but its coming along.

Movabletwiki was mentioned on Pronet, dev for Mobable Type, site went from 30 hits/day to 300 hits/day. Good advertising for TWiki.

-- BruceRProchnau - 16 Apr 2005

Along these lines:

  • A basic one web twiki - minimal on everything - simplified, for the non tech iinstaller/user (to be defined).

  • A small business TWiki

  • The present and future Enterprise Edition, as it comes now (the present TWiki) and as it grows.

Then the non tech user won't be complaining, the tech user gets the full TWiki and the non tech gets the simplified personal edition that is user friendly, small and simple. Then as that users skills grow they can addon. I feel this "simple twiki edition" would make twiki VERY popular.

If you want a REAL Wiki - get TWiki.

-- BruceRProchnau - 16 Apr 2005

Edit | Attach | Watch | Print version | History: r24 < r23 < r22 < r21 < r20 | Backlinks | Raw View | Raw edit | More topic actions
Topic revision: r24 - 2005-04-16 - BruceRProchnau
 
  • Learn about TWiki  
  • Download TWiki
This site is powered by the TWiki collaboration platform Powered by Perl Hosted by OICcam.com Ideas, requests, problems regarding TWiki? Send feedback. Ask community in the support forum.
Copyright © 1999-2017 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.