Welcome to WebmasterWorld Guest from 54.198.205.153

Forum Moderators: phranque

Message Too Old, No Replies

Identifying web communities

     
10:43 pm on Mar 27, 2002 (gmt 0)

Administrator from US 

WebmasterWorld Administrator brett_tabke is a WebmasterWorld Top Contributor of All Time 10+ Year Member Top Contributors Of The Month

joined:Sept 21, 1999
posts:38182
votes: 77


We generally think of Hubs and Authorities research when "community" identification comes to mind. That isn't the entire picture. Check out some of the papers linked by the author citations: [webselforganization.com...]

* (link found on Jills blog [cmc.uib.no])

11:18 pm on Mar 27, 2002 (gmt 0)

Senior Member

WebmasterWorld Senior Member 10+ Year Member

joined:Nov 29, 2000
posts:1425
votes: 0


Great link Brett!

Was wondering what Steve Lawrence has been up to lately.

12:08 am on Mar 28, 2002 (gmt 0)

Senior Member

WebmasterWorld Senior Member digitalghost is a WebmasterWorld Top Contributor of All Time 10+ Year Member

joined:Feb 23, 2002
posts:3687
votes: 0


Was that a government funded study? How much "study" did it take to realize that birds of a feather...

Arpanet proved this more than a decade ago. Hobby groups, study groups, organizations with like interests all tend to reference, quote and study each other. In terms of the web, they link to each other. Does anyone find it suprising that link structure would point to groups of sites with similar interests?

The study shows that a remarkable degree of order emerges from the independent linking actions of individuals.

Any time spent at a college campus would provide enough information to support the fact that people of diverse cultures, backgrounds and goals gather together to form groups where similar interests are shared.

Sites that link to Rivest deal with encryption, now there's a stunner. Looks like a lot of math to determine what a spider that analyzes only anchors could do.

The definition of a community, (a collection of sites with more links within the community, in either direction, than outside) seems elementary. The interesting part that I see is an automation of the process to determine the community. The benefits of which I suppose webmasters could determine for themselves. As far as the actual point of the paper, sites that link together typically form communities, there isn't anything new there.

DG

5:22 am on Mar 28, 2002 (gmt 0)

Full Member

10+ Year Member

joined:Aug 26, 2001
posts:277
votes: 0


>> How much "study" did it take to realize that birds of a feather ...

I have noted in my life that many people do not believe it is so untill the right person sayes it is so. Or, to borrow from another saying "untill the fat lady sings".

So, i am left wondering if it is now so, that communities link togeather or are we waiting for another source to say what we know to be true?

5:53 am on Mar 28, 2002 (gmt 0)

Senior Member

WebmasterWorld Senior Member marcia is a WebmasterWorld Top Contributor of All Time 10+ Year Member

joined:Sept 29, 2000
posts:12095
votes: 0


DG quoted:

The study shows that a remarkable degree of order emerges from the independent linking actions of individuals.

Which implies that groupings of microcosms are defining the macrososm.

>automation of the process to determine the community

That's the challenge, since what sometimes is normally independent action is not that at all in certain cases, where it's programmed to appear so automatically. The whole concept can be and is being used for less than noble purposes.

>Does anyone find it suprising that link structure would point to groups of sites with similar interests?

When groups of sites are linked in logical communities and the individual microcosms of like sites collectively add up to over 9K instances of generated third level domain spam with links unrelated to those individual communities strategically interspersed, an automated process can prevent them from inhabiting the SERPs without human intervention, or at very least automatically
trigger an alert.

It's not unlike how explosives can be used for destructive purposes or in positive ways, in urban renewal projects to bring down decayed structures to make room for new. Yin and yang, positive and negative.

6:47 am on Mar 28, 2002 (gmt 0)

Senior Member

WebmasterWorld Senior Member digitalghost is a WebmasterWorld Top Contributor of All Time 10+ Year Member

joined:Feb 23, 2002
posts:3687
votes: 0


The whole concept can be and is being used for less than noble purposes

True, and I lay that fault at the door of search engines that place undue emphasis on links. The environment created by the Google algorithm was quickly exploited requiring Google to adjust the algo to weed out the inflated link popularity counts manufactured by link farms.

The current environment in which webmasters are afraid to link to other sites for fear of losing position isn't healthy for the web or for the business sites that were unduly penalized by the PR0 bug

A brand new site with great content should not be penalized by a search engine because they don't have the requisite inbound links to satisfy an arbitrary algo. Inbound links have absolutely NO correlation with content relevance or in the initial stages, with the importance of the site. An important idea is important regardless of whether it is new or old. I had hoped the days in which we ostracized Copernicus were past. :) Google wouldn't have ranked Copernicus well, papers and books that referenced him were burned.

In regard to the topic at hand, community is the sine qua non of the web. I can't believe a study was required to validate it and then announced it as if it were something new. Struck me as odd as being suprised that gear heads show up at classic car shows.

DG

7:18 am on Mar 28, 2002 (gmt 0)

Senior Member

WebmasterWorld Senior Member 10+ Year Member

joined:Dec 11, 2001
posts:2725
votes: 0


Digital host - I agree and disagree.

I would argue that if 20% of your outbound links are reciprocal (but to many different quality sites/pages, that themselves also adher to this policy) Google will not penalise your site nor the others with a PR0 penalty and if so I would yet have to see an example of such a penalty. You can still create a good community in that way.

I totally agree with your Copernicus analogy. The biggest flaw of Google is the need of inbound links to be indexed by Google (at least that is what they claim in their interviews). I know of competitors of mine who have decent regular businesses and nice websites who are not findable in Google simply because they do not understand this need of links - it helps me, but not the searcher.

More importantly Google should allow (near) immediate crawling and indexing if the the site or page you submit to Google contains highly topical actual information. This should not be difficult to automate for Google, they could relate the theme or topic of the specific page with the content of their news page for example.
Ok these new submitted pages will not rank high for regular querries because of the lack of links, but searchers could still be helped if they search with very specific multi-word querries for which pagerank and linking matters less.

(i guess this message is a bit to much Google related for this thread).

9:55 pm on Mar 28, 2002 (gmt 0)

Senior Member

WebmasterWorld Senior Member digitalghost is a WebmasterWorld Top Contributor of All Time 10+ Year Member

joined:Feb 23, 2002
posts:3687
votes: 0


Hello Vitaplease,

>Too much Google related for this thread

Well, maybe not. A number of my friends have sites related to higher maths. It comes very close to being a closed circle of links and quie a few of them were hit with the PR0 bug. That didn't matter much to them as they weren't worried about their rank in Google, only accessibilty. The community is valid though, provides a service, has fresh content and also meets all the requirements of community outlined in the paper. Google chose to dismiss that.

The requirements of community outlined in the NEC paper come suprisingly close to what many webmasters consider dangerous linking practices now.

The fact that webmasters are using any criteria other than relevance and quality to determine which sites to link to indicates that Google has too much influence and that there is a flaw in their algorithm. A valid community could be nixed for having "incestuous" links.

The paper still reminds me of many government studies, hopefully, the publication will be the end of it. :)

DG

 

Join The Conversation

Moderators and Top Contributors

Hot Threads This Week

Featured Threads

Free SEO Tools

Hire Expert Members