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 |
My Topics:
k-log (66)
radio (56)
blogging (50)
RSS (46)
politics (36)
knowledge-management (34)
business (32)
topics (30)
tools (25)
software (25)
trackback (20)
google (17)
community (17)
shrub (15)
java (15)
humour (15)
metadata (14)
culture (14)
XML (13)
corruption (13)
XFML (12)
microsoft (12)
Gulf War II (12)
collaboration (12)
American culture (12)
XTM (11)
the middle east (11)
paolo (11)
information (11)
licensing (10)
learning (10)
publishing (9)
knowledge (9)
intranets (9)
blogplex (9)
outlining (8)
networking (8)
life (8)
Gurteen (8)
email (8)
wiki (7)
trust (7)
rant (7)
pax Americana (7)
palladium (7)
organisations (7)
open-source (7)
big media (7)
terrorism (6)
privacy (6)
PKP (6)
patents (6)
marketing (6)
law (6)
JIRA (6)
copyright (6)
broadband (6)
activeRenderer (6)
Wi-Fi (5)
tv (5)
the state (5)
spam (5)
sharing (5)
semantic-web (5)
security (5)
project management (5)
Lisp (5)
leaky pipes (5)
hope (5)
content-management (5)
consultancy (5)
CMS (5)
Business Journalling (5)
unemployment (4)
surveillance (4)
start-up (4)
programming languages (4)
pigopoly (4)
pagerank (4)
P2P (4)
leadership (4)
identity (4)
ideas (4)
groove (4)
Frontier (4)
connections (4)
career (4)
aggregators (4)
website (3)
warblogging (3)
visualization (3)
the economy (3)
test (3)
telecomms (3)
teaching (3)
social-networking (3)
selling (3)
RSI (3)
RIPA (3)
research (3)
referrers (3)
Novissio (3)
multimedia conversations (3)
memory (3)
media (3)
london (3)
investment (3)
innovation (3)
IM (3)
history (3)
e-government (3)
drm (3)
daypop (3)
communication (3)
Amazon (3)
XSLT (2)
xml-rpc (2)
XKM (2)
workflow (2)
words of wisdom (2)
webservices (2)
visibility (2)
UNL (2)
test topic (2)
tacit knowledge (2)
strategy (2)
storytelling (2)
spamblocking (2)
search tools (2)
Ryze (2)
RDF (2)
productivity (2)
PingBack (2)
organisational-development (2)
opml (2)
MovableType (2)
metalogue (2)
listening (2)
knowledge metrics (2)
information-overload (2)
InfoPath (2)
IE (2)
health (2)
hardware (2)
gpl (2)
faceted classification (2)
explicit knowledge (2)
European Union (2)
environment (2)
enron (2)
effectiveness (2)
edublogging (2)
Creative Commons (2)
CoP (2)
conferences (2)
bots (2)
big oil (2)
wizards (1)
Web Services Architecture (1)
UK culture (1)
transclusion (1)
TKP (1)
the-game (1)
text-analysis (1)
symantec (1)
structure (1)
stress (1)
State of fear (1)
stability (1)
socialtext (1)
sfa (1)
sensuality (1)
search-engines (1)
search heuristics (1)
s-l-a-m (1)
ROI (1)
respect (1)
quotations (1)
Process logging (1)
presentations (1)
PIM (1)
patterns (1)
ontology (1)
obituaries (1)
neighbourhood (1)
multi word topics (1)
morals (1)
manifestos (1)
M$ (1)
liberty (1)
kcafe (1)
jobs (1)
Italy (1)
issue tracking (1)
hypertext (1)
game-theory (1)
gadgets (1)
future-publishing (1)
FOAF (1)
films (1)
fibre (1)
failing fast (1)
faceted browsing (1)
enterprise streaming (1)
e-learning (1)
Dynamic DNS (1)
Dublin Core (1)
dns (1)
dieting (1)
dhtml (1)
deep-linking (1)
CyberWar (1)
CRM (1)
creativity (1)
conversation (1)
conflict (1)
complexity (1)
competition (1)
Colonising Space (1)
brands (1)
boycott (1)
bookmarklet (1)
backlinking (1)
annoyances (1)
algorithms (1)
agents (1)
adverts (1)
accessability (1)
academia (1)
Blogroll:
[Macro error: Poorly formed XML text, we were expecting . (At character #172.)]Recent Items:
3/27/03 |
|
3/27/03 |
|
3/27/03 |
|
3/26/03 |
|
3/26/03 |
|
3/26/03 |
|
3/26/03 |
|
3/25/03 |
|
3/24/03 |
|
3/24/03 |
|
3/23/03 |
|
3/23/03 |
|
3/19/03 |
|
3/19/03 |
|
3/18/03 |
|
3/18/03 |
|
3/18/03 |
|
3/18/03 |
|
3/18/03 |
|
3/18/03 |
|
3/18/03 |
|
3/17/03 |
|
3/17/03 |
|
3/17/03 |
|
3/17/03 |
|
3/13/03 |
This work is licensed under a
Creative Commons License.
Staying on theme...
I see MT's TrackBack has the ability to generate some RSS output, but it doesn't actually look like RSS (as in, I don't think you could just subscribe your aggregator to it).
Here's what I have in mind. I want to know whenever someone pings any of my pages. I register with the TrackBack server a pattern like http://radio.weblogs.com/0107808/*
Whenever someone pings a page under that URL it gets added to a dynamic RSS feed with a URL generated for me. I can subscribe to that feed in Radio and see all the trackback pings appear as news items.
If I'm interested in who is writing about things Dave is saying I might add a pattern http://www.scripting.com/* and get his TrackBack results as well.
Does MT's TrackBack do this already? I couldn't really tell what it's RSS was doing...
Rahul has the solution. Trackback and comments suffer from the same problem: they aren't seemlessly integrated into the act of posting to a weblog. Rahul's points out that a combination of RSS UIDs and blogrolls would seemlessly pull this off. It would automatically thread the conversation back from your post. This is a solution that doesn't detract from the weblogging experience. All you need to do is post, and the rest is taken care of. The requirements to participate would be that you need to have a weblog and be on someones blogroll (in someones neighborhood). To screen out inconsiderate particpants, just put their blogs on your block list. It gets really simple if this is an orgnizational setting: a list of all the people that run weblogs could provide a definitive blogroll, a one-stop-shop for all participants. Nice. [John Robb's Radio Weblog]
» Rahul's idea is interesting, but it isn't TrackBack. It still assumes a near-line level of connectivity between two blogs. Maybe that's okay in some situations but I don't think it should be a requirement.
As to seamless integration. The Radio plugin I have devised provides that. Any url you referencein a post gets pinged when that item is published. No need to do a thing.
Of course some people will want more control. That's okay too...
I've had a couple of comments to the effect of "why implement TrackBack in XML-RPC when its perfectly fine in HTTP/XML?"
I guess the answer is that Frontier & Radio make it so easy to implement XML-RPC clients & servers that it was the easiest way to glue together a TrackBack solution that worked the way I had in mind. I guess there is no no reason why it couldn't also support the same method's as MT as well. I have no technopolitcal axe to grind here.
Something that really bothers me about TrackBack in MT though is the idea of enabling. Have I got it right? You have to enable a post for TrackBack before it can receive pings? And the trackback url isn't the same as the link to the posting? This seems clumsy to me. Can someone explain what's going on with that? Am I confused?
While I was in the hospital in June, the Movable Type folks implemented a feature called TrackBack.
I'm not exactly sure all that it can do, but here's at least part of the story. (I'm posting this so I can get corrected if I don't understand the feature. It occurs to me that this post could use the feature, heh.)
Anyone, anywhere can send a message to any Movable Type server to associate a URL with a weblog post. That URL will be shown in the list of TrackBack links for the post.
Further, based on an email from Matt Mower, for some reason that I don't understand, this can only work with Movable Type servers. I doubt this, because from all outward appearances it is using HTTP, which could be emulated by any program capable of doing HTTP. Matt thinks this feature should be implemented with XML-RPC. I'm not sure it'll take off no matter what it's implemented in.
Here's the problem. By design it seems to assume that everyone plays fair. But eventually we all attract a relatively small number of people who would mark up every post with trash talk, if given the chance to. It's a predictable process. That's why I don't have a discussion group here (I used to), or a comments feature. It's why MSNBC is moving to weblogs over discussion software. It's basically why weblogs have a future for thoughtful discourse where mail-list-like collaboration tools are dead-ends. When I think about evolving weblogs, I try to avoid features that turn them into discussion groups.
» I think there has been a misunderstanding between Dave and I. Maybe I misphrased something or it was misinterpreted. Either way:
I'm not suggesting that TrackBack can only be implemented in MT. Just that, as it is implemented in MT it can only be served by MT and is most useful to MT users. It doesn't suit me very much. I also don't like the way you have to TrackBack enable things, use special URL's, have bookmarklets etc.. all that gets in the way to me.
I envisage an open XML-RPC based system. The TrackBack data should be available to & from any system and can track arbitrary URLs (no requirement to TrackBack enable anything).
Also with the prototype Radio client all the work of pinging is done for you automatically. As part of the publishing action Radio will figure out all the posts being referenced and ping them automatically. That's how I want it to work, you might want it different which is why I say it's a prototype.
As to the inherent design problem in trackback, well, I agree with the comments made. From a certain viewpoint.
However I see TrackBack not so much as a weblogging tool but as a k-logging tool. It gives you the ability to know what someone else is contributing to projects you are working on and that could be vital. As are discussion forums and all the other collaborative tools that help people do useful work.
Will TrackBack be absused? Sure. But so can any technology. If the abuse becomes a problem we can evolve strategies for addressing it. For me this is a time for experimentation, it's too early to abandon a potentially useful idea like TrackBack because it has a potential for abuse.
Example (and shooting from the hip) : Problem: "nusiance pings appearing on my TrackBack report." This seems a lot like the problem of spam email to me. Collaborative spam filtering looks set to deliver good results here, maybe it could do the same for TrackBack?
[Disclaimer: TrackBack - I am a believer!]
In order for klogging to be successfully I think it is going to have to come to an understanding with Big-KM.
Example: BigCo has invested half a million dollars in a big knowledge management system for their world-wide operations. This kind of investment can become a lode-stone around any other systems neck. For klogging to thrive here it is going to have to integrate.
Here's one idea I have for how this could work.
- Extend Big-KM System-X so that it can aggregate RSS feeds like Radio, MT and others do now.
- Extend your klogging software to allow per-post meta data. (liveTopics does this for Radio)
- For each project in System-X define a set of topics that will act as trigger phrases for that project
- Get the kloggers to use those topics when they want to involve a post in a particular project
- Now subscribe System-X to every klog in the organization and watch as it indexes and archives all that information. Each project grabbing only those postings that are appropriate (by use of the trigger phrases)
- This means that the klogs add value to the big-KM system. Suddenly it doesn't just have the dry dusty project documention, but all the live vibrant stuff that people are really doing!
- Now extend System-X to generate a per-project RSS feed.
- If I am on the project I can subscribe to this feed. Now instead of receiving email from System-X or having to go to an arbitrary web page, I get all the "official" project stuff (new documents, forms etc...) delivered in my RSS stream.
Closing the loop between the big-KM and the klog so that they both add value to each other.
Just an idea....
Envisioning projects a la klog?
How can weblogs contribute to project visualization?
Annotation.
- Associate each project post with one or more tasks, issues, milestones, and deliverables on a given project.
- Enable a few extra attributes for a post: Red/Yellow/Green Priority (U.S. cultural bias).
- Create a view into a team's weblog posts organized by the work breakdown structure, another by priority
PM is about the conversation more than formal modeling. It is how we come to appreciate project dreams and know project reality. We discover our colleagues' capabilities and limits. We negotiate commitments. We make the thousand mid-course corrections to the project plan. My project communication templates help you script some of those conversations.
But conversation is narrative and auditory. How do we get the best characteristics of project conversation into visual media? Into electronic visual media?
Thanks to experience designer Diego Lafuento for the Tufte pointer.
[aka design]
[a klog apart] [Ron Lusk: Ron's K-Logs]
» This is an interesting idea which I need to think about some more.
The idea of klogs integrating with other systems already has me interested and this adds a new dimension.
This inspires another post.