in my experience
LazyWeb Theory.
January 9, 2003 8:57 AM
I had my doubts about the signal to noise ratio that LazyWeb might produce, and Clay Shirky had a few opinions he wanted to share with me (seemingly in advance of this article at O'Reilly). I want to share all of that with you (thanks to Clay for permission to reprint) in its original form...
January 9, 2003 8:57 AM
I had my doubts about the signal to noise ratio that LazyWeb might produce, and Clay Shirky had a few opinions he wanted to share with me (seemingly in advance of this article at O'Reilly). I want to share all of that with you (thanks to Clay for permission to reprint) in its original form...
From: Clay Shirky Date: Fri Jan 3, 2003 9:43:43 AM US/Eastern To: Daniel Kapusta Subject: LazyWeb It's unlikely that the LazyWeb.org pings will be numerous and random, because it raises a threshold of actually using the Trackback URL, and even tiny thresholds have a huge filtering effect. And LazyWeb is only random if you think of it as a site -- its real usefulness is as an RSS feed, so that third-parties can, say, subscribe only to those posts that say "Chat" or "Wiki" in the title, thus pulling something useful out of the general feed. -clay ------------------------------------------------ From: Daniel Kapusta Date: Fri Jan 3, 2003 10:06:12 AM US/Eastern To: Clay Shirky Subject: Re: LazyWeb On Friday, January 3, 2003, at 09:43 AM, Clay Shirky wrote: > It's unlikely that the LazyWeb.org pings will be numerous and random, > because it raises a threshold of actually using the Trackback URL, and > even > tiny thresholds have a huge filtering effect. I suppose you are right about the threshold part, but a general resource will be more general. The only threshold that exists is 'those who can send trackback pings and are motivated to do that.' That threshold does little in terms off content likeness since bloggers are in general motivated, and there are many MT based blogs. Within the scope of the interests of the trackback pinging public, randomness indeed exists. But, if I were successful at getting 10 or 20 other like minded bloggers to ping my site on a regular basi, would that threshold be any more valuable? I think so. But maybe not, and there's only one way to figure that out. > And LazyWeb is only random if you think of it as a site -- its real > usefulness is as an RSS feed, so that third-parties can, say, > subscribe only > to those posts that say "Chat" or "Wiki" in the title, thus pulling > something useful out of the general feed. I don't think that is necessarily true. I could post a note on my site saying that I had a chat with my mom about her visit to my home this weekend and ping lazyweb. That's useless to the blogosphere, but I wasn't prevented/filtered from shoving my minutia in front of the blogeratti who are interested in 'chat.' Don't get me wrong though, I like the lazyweb idea enough to have come up with it on my own. It's already better than the weblogs.com recently updated list (which really is useless imho). Dan ps, do mind if I post this as a comment to my post? It's useful discussion. ------------------------------------------------ From: Clay Shirky Date: Mon Jan 6, 2003 9:10:18 PM US/Eastern To: Daniel Kapusta Subject: Re: LazyWeb > I suppose you are right about the threshold part, but a general > resource will be more general. The only threshold that exists is 'those > who can send trackback pings and are motivated to do that.' That > threshold does little in terms off content likeness since bloggers are > in general motivated, and there are many MT based blogs. Within the > scope of the interests of the trackback pinging public, randomness > indeed exists. Sure, but the randomness is constrained by payback. The karmic value of pinging something is low, unless there is some response. Of course, if LW takes off, then there will be spam problems, where free riders identify a high-quality message stream and attempt to steal its reputation, but that’s a problem of success, not failure. > But, if I were successful at getting 10 or 20 other like minded > bloggers to ping my site on a regular basi, would that threshold be any > more valuable? I think so. But maybe not, and there's only one way to > figure that out. IN many cases, it will be more valuable, because in general you will characterize problems of interest to your readers. But you don't have to give that value up to use the LW as well. The two questions are "Is the additional value of pinging the LazyWeb non-zero?" and "is the additional value high enough to be worth the time it takes you to add the trackback URL?" My guess at the answers right now are "Probably, and It depends." > ps, do mind if I post this as a comment to my post? It's useful > discussion. Not at all, post away. -clay ------------------------------------------------ From: Daniel Kapusta Date: Tue Jan 7, 2003 9:51:59 AM US/Eastern To: Clay Shirky Subject: Re: LazyWeb On Monday, January 6, 2003, at 09:10 PM, Clay Shirky wrote: >> But, if I were successful at getting 10 or 20 other like minded >> bloggers to ping my site on a regular basi, would that threshold be >> any >> more valuable? I think so. But maybe not, and there's only one way to >> figure that out. > > IN many cases, it will be more valuable, because in general you will > characterize problems of interest to your readers. > > But you don't have to give that value up to use the LW as well. The two > questions are "Is the additional value of pinging the LazyWeb > non-zero?" and > "is the additional value high enough to be worth the time it takes you > to > add the trackback URL?" My guess at the answers right now are > "Probably, and > It depends." I would guess 'yes and yes' due to the ultra low cost of a cut-and-paste. That leads me to believe many people will ping LW (if they are aware of it, which is another filter of sorts) leading to a higher noise ratio. Part of the basis for the idea is that LW doesn't have a well defined semantic basis (imho). It's new. It's open to anyone for any reason, at any time. I suppose I'd like to see Dublin Core metadata sent as part of the TB ping, and then be able to scrape the LW rdf feed for things I am interested in, but I read the site every day now anyway. Cheers Dan