The Evolution of Information Grazing

by Joshua Porter | February 15th, 2006 | shortlink: http://bokardo.com/p/357

Update: James Corbett disagrees completely.
Update #2: Well, not completely.

One lens through which to look at the recent innovation in the memetracker space is frustration. If you look at where the frustration is in how we track memes (ideas), you can get a decent picture of where the innovation is going. If you want to predict the future, find the frustration!

Like an antelope eating grass on the Kalahari, grazing is eating small quantities of food at frequent but irregular intervals (Apple Dashboard dictionary). Recently, the term grazing has been adopted to describe our efforts at finding information on the Web. The following is a very general picture of the four types of “grazing” we’ve gone through, or are going through now. Each level had it’s own share of frustration, which led (or is leading) directly to the next level.

Site Grazing

Site grazing is typing in URLs to see if there is any new content. The URLs often come from a list of bookmarks or memory. This difficulty was noticed early on in the blogging world by Rebecca Blood, who says that even as early as 1999 it was too difficult to read all blogs:

“At this point, the bandwagon jumping began. More and more people began publishing their own weblogs. I began mine in April of 1999. Suddenly it became difficult to read every weblog every day, or even to keep track of all the new ones that were appearing.”

The frustration with site grazing is what led to the popularity of RSS, as it immediately relieved the need to visit sites over and over without knowing if something was new.

Feed Grazing

Feed grazing is what most of us do in our feed readers: checking a bunch of feeds to see if there is anything new. Usually the reader notifies us of new content, by either starring the content or sorting it based on newness, or both.

In my feed reader, Shrook, the feeds that have been updated more recently are presented first. The feeds that haven’t been updated are pushed to the bottom. Whereas site grazing is a bunch of HTML urls, feed grazing is a bunch of RSS urls. So feed grazing works at the feed level. A feed list is often called a Reading List, and is captured in an OPML file. Here is my reading list.

The frustration with feed grazing is that we soon have too many feeds, and many of the feeds overlap content. Ironically, however, we still want to add more feeds if they are relevant to us, and so we prune our feed list over time.

Grazing Lists: These are a variation of feed grazing, and where the term “grazing” came into play via James Corbett. (obviously, I think that grazing is a great term to describe all of these levels) Anyway, a grazing list is a dynamically updated feed list like the one that Adam Green has created out of the tech.memeorandum site. In this paradigm, you subscribe to a Reading List that dynamically changes over time, so that you never know what feeds will be on it.

The frustration with grazing lists is that it is unclear as to why feeds are in the list. Is it because the feeds themselves are about the hot topics of the day, or because they happened to have one good post in them? Adam, on the other hand, likes to use them to discover interesting feeds.

Post Grazing

I would say post grazing is getting nearer our end goal. It is grazing for the latest, most interesting posts, regardless of what feed they come from. Right now, there are a few services out there putting out feeds with which we can post-graze, such as tech.memeorandum, Tailrank, and Findory. As I was telling Kevin Burton of Tailrank the other day, I’m completely in awe of folks who create these services: Gabe, Kevin, and Greg, in this case.

Each of these services offers different levels of relevance, and clustering. As far as I can tell, the service that provides the best solution along these lines will certainly gain a strong following, at least from the tech bloggers I read.

Relevance Relevance is the make or break of these services. Many of them use a link-counting method to find the most talked-about content, often combining that with some measure of authority to present the most relevant information.

Clustering Clustering is extremely important because otherwise the content becomes an echo chamber. It is very useful to have related posts clustered so that we don’t see the same news over and over. Just this morning I read that Google bought Measuremap in about 85 different places. Enough!

To use an analogy from offline, post grazing is similar to reading a big-time paper, which filters news from various news organization (Reuters, AP, etc) in addition to their own content. They are getting those stories that are of the most interest and providing a one-stop place to read them.

When you move from one paper to another, however, you realize that there are different filters, and that each has it’s own strengths and weaknesses. For example, Fox News is a completely different filter than is NPR. So there is a level of editorial as well. So then another frustration sets in, you don’t know what editorial is being applied to the filter you’re looking at. Instead, we want news that is filtered by ourselves, or based on our needs.

Personalized Post Grazing

I think this is the golden path. This is when we get post-level news recommended to us based on our personal preferences and reading behavior. The three services I mentioned above are moving in this direction, but as this comment thread on Scoble’s site shows, they are not quite there yet. (this is a *great* thread, if you consider how it might change the way that we read news). (hat tip – Father Richard)

So, in addition to working at the post-level, this works at the personal level. Combine what the group has found to be valuable to what the person has found valuable in the past, and you get a great sense of what might be valuable to the person in the future.

The Evolution of Grazing

However, there may still be frustrations at the personal post level as well. One is our changing tastes. What if we track tech news for years and then find ourselves burnt out, and yearning for a different kind of news? What if we don’t feel that we’re discovering enough diversity in our personal recommendations…what if we feel like we’re missing out? With every level that we reach, we’re happy for only a short period of time. We will continue to want increased efficiency, and increased denseness of information.

Make them Care! - Struggling to communicate the value of your product or service? I'm writing a new book that shows you how to make people care about your product or service by clearly communicating the most important bits. For designers and marketers creating product web sites. Find out more.

On Changing the World

On Noise

Links to this Post

Comments

1. James Corbett 4:51am, Fri 17th, 2006

Josh, I’m just after posting to my blog about your term ‘Post Grazing’. Sometimes when I’ve been thinking about something too long I get stubbornly set in my own views which why is its been terrific to widen this discussion on Grazing.

It just struck me last night that your term ‘Post Grazing’ is actually much more apt than my term ‘Feed Grazing’ within the context of a fully dynamic hierarchy, the likes of which we’ll soon see many more of. The one Adam is working on for tech.memeorandum is the first I’ve seen. And certainly the kind of trees I envision del.icio.us generating will be restless organic entities.

All along I’ve been thinking about Feed Grazing more in terms of the transition from where we are now with (rather) static OPML hierarchies and Reading Lists. But as we move towards entirely dynamic trees the idea of Feeds will fail to make any real sense and the latest Posts will be all that count. In that context ‘Post Grazing’ is the term that will apply.

Thanks for clarifying my thinking on this.

2. chillu 5:23am, Fri 17th, 2006

Interesting post! I’m just starting my bachelor-thesis on this topic, essentially doing concept and development of a “social rss-reader”. though it will still use “feed grazing” in your terms, planned features are personal recommendations by other users, personalized relevancy-filtering etc.
greets from germany

3. Hooman Radfar 11:54am, Mon 20th, 2006

Joshua, great post. I think that folks in the blogosphere are much more cognizant of the concept of “information overload” that is swiftly becoming an important problem in an increasingly connected world. From an end user perspective, I think that it has become clear that we need more innovative visualization and filtering mechanisms. One such mechanism you had mentioned is clustering. The problem of information overload, however, not only requires revamping the end user experience, but also working on our fundamental information representation schemes. Unfortunately, information still exists in an unstructured format and makes it difficult for designers/developers to create more useful visualization paradigms. That being said, I think that end user frustration will continue to drive content providers to provide data in structured formats and – eventually – the web truly will transition from a publication mechanism for unstructured documents, into a platform for structured online services that can be more effectively leveraged by end users in tandem to get things done.

4. Easton Ellsworth 12:39pm, Mon 20th, 2006

Very interesting post. You have some great ideas here!

You called this “a very general picture of the four types of ‘grazing’ weâ€™ve gone through, or are going through now.” Is there really just one “we” – and if not, who are the different “we’s” and how important is each group in influencing the evolution of information consumption?

In other words, besides grazing gazelles, you’ll find a host of other animals with distinct eating habits – from the occasionally binging alligator to the constantly foraging … well, some little animal that’s constantly foraging. And the amount of food available in the area to each type of animal changes over time in response not only to what the antelopes are consuming, but also to what the other animals are consuming.

What do you think – over the next several years, what group or set of Web users will most significantly influence the way people consume information? Who’s the most important animal to be watching on the grassland?

Thanks for the thoughtful post. Hopefully this will get a lot of people thinking about their own information consumption habits and how they are (or could be) influencing the habits of others!

5. Dennis D. McDonald 6:14pm, Mon 20th, 2006

The concept of “information overload” is an ancient one that bloggers did not invent. It goes back to the 1940′s explosion in R&D that drove investment by the US government in technology-aided clearinghouses, indexing services, and current awareness services that, at some level, relied on some form of expert review and analysis to ultimately decide what should be published on a list that would make a reader aware of “something new I need to know about.” We’re going though a similar cycle now. This is why services like Technorati have added “authority” sliders that do some filtering based on some automatically generated measure related to linking. Eventually we get back to measures of influence to help us decide what to read — such as an “A-lister” publishing his/her OPML file. Right now, for example, the Web 2.0/blogging world seems to be divided into two camps – those people who attended that party last week, and those people who didn’t. I didn’t attend the party but, by God, I know two people who did! Therefore, you should read my blog and subscribe to one of my 5 RSS feeds:
http://www.ddmcd.com/rss-feeds/

6. Josh 7:50am, Tue 21st, 2006

Easton, I think you’re entirely right. I think grazing fits a specific type of information gathering, that of news gathering. I should have made that more clear.

For gorging, we still go to books. For snacks, we still go to Amazon reviews. etc…

7. dan 12:26pm, Fri 3rd, 2006

I understand what yer saying here but part of me really wants to write a blog entry entitled “The Evolution of ReWording…” – what is it about the webosphere lately, there’s been a rash of posts lately to restate the consistant reality of what people do and then give a cool name to describe it by… in six weeks grazing will be out and sniffing will be in. I don’t graze for new posts in bloglines, I clearly sniff them out…

Before it was surfing, now its grazing. Martha, you don’t surf.. you graze… ahhhhh i see.

Before I had a website, no.. now thats a blog. Oh wait and that radio show you had, thats sooo not the right word, its podcast, and its totally new and revolutionary!! (sure sure rss, got it, thats technobabble keycard access in the end people want the end product.. and that hasnt changed)

I dunno gang. Ok sure its grazing, but thats a fancy word for something i still consider surfing, that i know will in fact change into sniffing, or peeking or whatnot. I like the thinking here, just dislike the trend analysis aspect of it, nothing really new. Pointing out the cues of the system and what people do ok, sure i get that, but naming it.. oh man.

Its dancing with jargon, cool, neat.. evolutionary? Well for now maybe.. but when sniffing hits.. ya’ll be old school!!

8. Lukas 6:12pm, Wed 3rd, 2007

Interesting post thank you

9. Sergey 2:46pm, Wed 24th, 2007

The problem of information overload, however, not only requires revamping the end user experience, but also working on our fundamental information representation schemes. Unfortunately, information still exists in an unstructured format and makes it difficult for designers/developers to create more useful visualization paradigms.

10. Markus 10:37pm, Mon 19th, 2007