On Wednesday, January 16, 2013 4:51:20 PM UTC-5, Jeremy Ruston wrote:
>
>
> I'd see both as experiments into the same idea of harnessing the power of 
> a command line interface, and updating it to the GUI and/or web eras. 
> Definitely of great interest to me.
>
> For lots of reasons, most of the discussion around UI design is rather 
> oriented towards users who are assumed to be too busy to learn new things, 
> unwilling to invest anything more than cursory attention in the interface 
> that we craft. In ecommerce we visualise prospective buyers falling away at 
> the first sign of anything that they don't understand. So, we design 
> checkout flows to be familiar above all else, to reduce the cognitive load 
> on our users. That's all fair enough, but I'm fascinated by user interfaces 
> that trade a steeper learning curve in return for ultimately being more 
> useful. I believe we can craft user experiences that combine a visceral 
> initial appeal with the progressive disclosure of a small number of 
> concepts that combine in powerful ways. Such interfaces aren't appropriate 
> in all circumstances, but they're much more thrilling to work on.
>
 
I also think that UI's can be unduly limited (e.g. Netflix) to suit the 
masses and so significantly lower the experience for those who want to 
achieve more than the basics. For me, the ideal knowledge management tool, 
has to permit effortless capture and curation. For the capture part, I 
could envision some kind of HUD where you'd work as usual with the tools 
you normally use, but at the touch of a key or as a result of a gesture, 
you could bring focus to the knowledge capture mode of the knowledge 
management system and indicate by way of selection, lasso, etc. what part 
of the screen/document you wanted to capture/link to. The capture tool 
could introspect the underlying app(s) and, through OCR, inspection of its 
temp/working files, automation API, etc., grab what needs to be grabbed 
with a minimum of hassle for the user. Some of this is what I had in mind 
when I asked you about node-webkit, earlier.

With current tools such as Evernote or OneNote you can capture with 
relative ease if the source is a web browser, but for many other sources 
the user has to go through a few too many steps to export data into the 
knowledge system. The ideal, again IMO, is a way of capturing that allows 
you to immediately "position" the new data in the context of existing stuff 
in order to minimize the curation needed later and to maximize returns when 
searching/visualizing right from the start. A sort of an intelligent 
"auto-completion" for insertion would be great, where you could be 
presented with different ways of integrating new data into the old.

The other place where existing tools suffer is in the ease with which you 
can curate a growing knowledge DB. When I think of how I'd ideally like to 
deal with information overload, I can imagine ways to easily create and 
re-arrange links between info atoms. For some things a spatial (2D/3D 
depending on the data in question) layout is what's best with links being 
visible, color coded, labeled, etc. Things that are not so important could 
be dimmed out or sent to the periphery. VUE (http://vue.tufts.edu), allows 
for some of the kind of layouts I'm talking about, but, IMO, something like 
that could be just a particular view onto your data and not the only way to 
manage it. As an aside, the lack of spatial "stability" of tiddlers is one 
of the things that, depending on what kind of visualization I'm after, kind 
of bothers me about the original TiddlyWiki. I'm kind of OCD about some 
things and I want them to stay where I put them! :-) Of course I also 
appreciate locality of reference at other times. 

Anyhow, I'm blabbering on here and I'm sure you've got your own vision for 
this stuff, but I'm hoping what you'll end up with will be flexible enough 
for others to build some non-wiki-like visualization and management of 
linked data. I'm also glad you mentioned interoperability with SFW as I'd 
hope that if the ideal PIM can't be had with any one tool, maybe it could 
be possible with the right amount of deeper integration among disparate 
tools.

Regards,
Adrian


-- 
You received this message because you are subscribed to the Google Groups 
"TiddlyWikiDev" group.
To view this discussion on the web visit 
https://groups.google.com/d/msg/tiddlywikidev/-/w6m45y9CA2IJ.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/tiddlywikidev?hl=en.

Reply via email to