| August 2002 | ||||||
| Sun | Mon | Tue | Wed | Thu | Fri | Sat |
| 1 | 2 | 3 | ||||
| 4 | 5 | 6 | 7 | 8 | 9 | 10 |
| 11 | 12 | 13 | 14 | 15 | 16 | 17 |
| 18 | 19 | 20 | 21 | 22 | 23 | 24 |
| 25 | 26 | 27 | 28 | 29 | 30 | 31 |
| Jul Sep | ||||||
This site is no longer maintained.
My current weblog.
Why I keep blogging about aggregators and outliners
My Blogging the Enterprise post received a solid rebuttal from A Klog Apart, KM Systems are to Treacle as Weblogs are to Honey:
The best KM is the one people practice. For all I know, klogging may address only ten percent of your KM goals. But try it. It is a critical ten percent. This wedge gets people owning their expertise, sharing it willingly, getting credit, getting feedback, being social about knowledge. How does this compare to any other tools you've ever introduced?
I've been meaning to respond to this for a week now, but I'm fresh out of round tuits. I'm getting there tho, and my explorations into changing aggregation behavior and utilizing outlines are a major part of it.
AKA is preaching a sort of Guerrilla KM, using the path of least resistence. People will blog if you give them the chance. Initially I wanted to write that people will follow KM processes if you ensure that the emphasis is on Knowledge and not on Management processes. I may write about that too, when more round tuits are available.
But the first round tuit I find will be spent writing about how to address the biggest failure of blogging for KM: organizing knowledge. All of the blogging tools that I am familiar with are dismal failures when it comes to organization. The good news is that Radio has unique features can help create order from the chaos, but these features aren't necessarily: fully developed, well organized themselves, or integrated with other features.
There's nothing wrong with blogging for KM, it just needs to be made a little better.
Once I get that out of my system, I'll get back to more geeky Pocket PC and Wireless Data goodness.
Wanting to Outline
Today I was hoping to use the Outliner in Kit to organize a bunch of stuff that I have kicking around in my aggregator. Problems:
- Pastes only plain-text.
- This means that I have to create several nodes in order to outline the text of a news item, any links within the item, and a link to the original source.
- I'm wondering how much effort it would take to write a system tray utility to convert HTML clipboard data to plain-text with the HTML retained.
- Kit loses the links from link nodes when loading an outline. Kit will allow links to be created and saved, but they are destroyed if you load the outline back into Kit. Reported to Kit's author.
- Kit encodes any HTML instead of rendering it. Mildly annoying.
- When saving the outline, Kit encodes some things that Radio's outliner doesn't decode.
- An example is the apostrophe.
- Suspect that this is Userland's problem, not Kit's.
Using Radio's outliner presents it's own set of problems:
- Pastes only plain-text.
- However, it does allow for some WYSIWYG-style editing of an item.
- Creating an inline hyperlink is user-hostile.
- Ctrl-K transforms the whole node into a link (a dangerous function as the outliner will destroy any child nodes if the link node is ever collapsed).
- The UI for inline hyperlinks sucks.
- If you use the HTML | Add link menu, no prompt is presented. Radio will either take the contents of the clipboard or create an empty link.
- The only way to edit an inline link is to turn off the HTML rendering (HTML | Format Text or Ctrl-` toggles this).
- The Outline and HTML menus do not have hotkeys.
- Radio's outliner has other issues and annoyances, these are just specific issues that interfere with my task of outlining news items.
Controlling Aggregation
Turning off the Enable the Aggregator pref doesn't have any discernable side-effects. For on-demand scanning, I've long had the following bookmark:
system.verbs.builtins.xml.aggregator.readAllServices (adrlogcallback:@radio.log.add)
Kit has a feature to run arbitrary UserTalk scripts, so I should be able to build a button for this.
More on the Aggregator
Rogers Cadenhead offers some pointers on aggregation in response to yesterday's wish for finer control of aggregation frequency:
A few things that might help: Kit, a Radio Userland shareware tool by Mark Paschal, can be used to read the news that arrived over a time period you designate (for instance, I start each morning by using Kit to display all of the RSS items that showed up overnight). Also, if you'd like Radio Userland to scan immediately when you load the program, change the Scan on Startup? preference.
I actully turned off the Scan on Startup pref the other day. I want Radio to scan less frequently, in order to achieve that now I have to not leave Radio running. Today I've been wondering if turning off the Enable the Aggregator pref will disable anything other than the hourly scan (update: This seems to be a viable option)
As for Kit, I've been using it for some time for it's other features. Kit's aggregator is able to filter by Keyword, Category, and Timeframe (specific hour, X hours from Y hours ago, or between specific times). A deal-stopper is that a Timeframe must be specified. Also, I don't like that it uses Radio's categories -- I prefer the arbitrary groupings of myRadio.
I've organized my thoughts on this, see Better News Aggregation.