TWiki's User Experience (and user interface) - A Vision

The purpose of this topic is to help developing the UI for the next major TWiki release. See ExtendedTWikiRoadMap for more info.

Do we need one?

Actually, yes. We really need one.

Here's why: A vision helps you setting goals, even more, it drives your goals. And finally, goals enable you to do efficient design. If you envision the easiest to use Wiki for collaborative authoring you have different design goals than the ones you would have envisioning the best Wiki for writing a encyclopedia.

The Vision should serve as a kind of tagline for the user interface and lead or guide the development direction.

How do we get one?

1. We might look at other (once) visionary products:
   • Firefox: "Help Mom use the web" 1
   • The original Wiki by Ward Cunningham: " The simplest online database that could possibly work." 2
   • VW Beetle: a car the majority of the population can afford.
   • ...
     
     
2. We think about what we you use TWiki for. More precisely, what goals do we want to achive by using TWiki?
   • Save time (thus money)
   • Let distributed teams work together
   • Share critical information
   • ...
   • please continue
     
     
3. We might start with the most basic interface possible, and then look at user roles, and ask what "extras" they need to get the most out of TWiki. Consider:
   • Young female driver - Citroen 2CV - wheel at each corner
   • Young male driver - Renault Clio Sport - blacked out windows, 140Db sound system, body kit
   • Experienced motorway driver - Opel Vectra - Sat Nav and Aircon, coat hanger
   • Mom driving the kids to school - Ford Galaxy - infinitely reconfigurable interior, wipe-clean seats, sound proofing
4. Use the principle of 'intelligent borrowing' from rather successful competitors (e.g. Confluence)
   • TWiki already has all the necessary features (and more), they are just not visibly enough to the end users.
5. Mix the input from bullet 1 with the goals from bullet 2 and shake it good (while shaking, think about "user interface, twiki, my goals").
   • If you got one, write it down in the next chapter and spread your vision!

I got one! - aka ideas and brainstorming

...

Now that we got one, what do we do with it?

We are not there yet... but as described above the next logical step is setting our design goals.

-- Contributors: TWikiCommunity - 08 Sep 2008

Discussion

By now, this is more a fragment. Meant for ideas and discussions. Please add your 2 cents, share your sources of inspirations, add what you use twiki for. Everyone is encouraged to refactor and not just to comment.

-- CarloSchulz - 08 Sep 2008

I would love to have a fully integrated Web 2.0 UI where you can do inline edit. For instance you get your topic page and when you click the edit button instead of going to the edit URL it just gives you a nice spinner or even a progress bar while fetching the topic source using AJAX before displaying the text editor.

The idea is that your topic page is a powerful Web 2.0 application very much like Google mail from which you can flawlessly edit, attach, move etc...

funny. or sad. I wrote an zero server round trip inline edit proof of concept a few years ago, and we even tried to solicit for funding (the testing and validation was a bit of a bear). Its somewhere in an svn.

• SvenDowideit - 09 Sep 2008

Sven, I was wondering what happend to your WikiwygContrib? Is this the proof of concept you are talking about? When I first tried the demo page I was really excited - such a perfect editing experience. I still believe that implementing such a thing for TWiki would be huge step forward.

• CarloSchulz - 09 Sep 2008

Carlo - no I was meaning the InlineEdit experiment (which I note in svn is quite broken).

Can you please tell me more about what excited you about wikiwyg, so that we can try to address those

• SvenDowideit - 10 Sep 2008

What excited me about wikiwyg (like seen on the demo page):

I read how it works (dbl click on a section), was wondering about the concept for a sec. But when you go and try it, it's really straght forward, no page refresh, no reload (after which the site looks completely different btw). You can also open multiple edit sections at once and there is a fast and easy switch from WYSIWYG to wiki mark-up.

Bottom line: it's fast, the learning curve is close to not exisiting, no page reload/refresh until you hit save, you only edit the part you want to. It is just easy and pleasant to use.

• CarloSchulz - 10 Sep 2008

• StephaneLenclud - 09 Sep 2008

I think that user interface vision is a bit too narrow. We are in search of a user experience vision.

• No objections. - CarloSchulz
• UserExperienceTaskTeam - SvenDowideit

Refering to the Wikiwyg discussion above, we can have something like this for TWiki 'right now'. The above example is only for Mozilla browsers so we have to look for something a bit more universal: jQuery. Basically, using the section extraction code from Michael's EditChapterPlugin, jQuery's JEditable and markItUp editor plugins we can have the same effect only without the WYSIWYG option (there is a demo of TinyMCE working with JEditable but not on IE browsers).

-- DavidPatterson - 11 Sep 2008

David, that sounds great. We should have a closer look at that.

-- CarloSchulz - 11 Sep 2008