Making Blogs Browsable

Knowledge Management experts will tell you there are three ways of looking for information, when you can’t get it first-hand: (a) searching, (b) browsing, and (c) using alerts & profiles

Searching is looking, just-in-time, for information about a specific subject or event. When you do research, you usually start by searching. When I was growing up, you started with the encyclopedia, and then went to the library subject card catalogue or bibliography. Today, you start with Google, follow the links, and backtrack each time you reach a dead end.

Browsing is reading serendipitously, looking for anything of interest. I can spend a whole day in a bookstore, wandering from section to section and stack to stack, reading bits of several hundred books and coming out with a half-dozen. This is possible, but much harder and less intuitive, with Amazon’s “read inside” capability (which allows you to read online an image of the first few pages of a book, and its table of contents and index). I used to browse the newspaper as well, because I believed its editors tried their best to cover a broad cross-section of subjects and events, and laid them out so I could quickly browse them by starting with the headline, then the first paragraph, and then, if I was interested in more, the details and the “continued on page A15” material. Online, newspapers also provide links to “related stories”, and some (but increasingly few) provide free archives online. Once I discovered that newspapers select the cheapest and most sensational articles for inclusion, not the most interesting or important ones, I stopped browsing newspapers. I still browse some magazines like The New Yorker, Wired, and Fast Company, since they appreciate that some people still want to read about important things.

Alerts and profiles are pre-set, ‘standing order’ requests that notify you, usually daily, of any new information about a specific subject or event. There are four main kinds of alert tools.

• Keyword alerts retrieve articles and new web pages that contain selected keywords or strings.
• Subject alerts retrieve articles about a particular subject (you generally have to pick from a pre-set taxonomy or index of subjects) that have been ‘tagged’ with that subject tag. These vary from very narrow (e.g. New York Times international news headlines) to quite broad (e.g. StumbleUpon has subject categories that include complexity theory, dogs and women’s issues).
• Source alerts retrieve any new articles from a particular source (periodical, website etc.). RSS feeds are a form of Source Alert.
  
• Synopses are distillations of longer works (e.g. Executive Book Summaries) or of all major stories on a particular subject in the last day, week or month (e.g. Innovation Weekly).

Alerts and profiles come in ‘push’ and ‘pull’ varieties. ‘Push’ alerts send something (an e-mail or notification) to your mailbox or desktop. ‘Pull’ alerts store the retrieved articles on a page for you that you must remember to go out and browse (e.g. Bloglines).

None of these three ways of looking for information is ideal in every case, and no information source lends itself perfectly to all three ways of looking. Here’s a current ‘scorecard’ of how the sources stack up today:

Note 1: If you’ve ever set up keyword alerts, you are probably aware that the results you get include a lot of self-aggrandizing corporate press releases (most of them close to spam) and a lot of duplication, as many sources republish the same articles. It’s a frustrating, needle-in-the-haystack exercise.
Note 2: Most alert services do not cover these sources.

The bottom line is that if you’re searching for specific information, you’re going to rely on the online newspapers and online magazines that provide full, free content and archives. Most of these companies are trying to get corporate researchers to pay big money for full content and archives, and it’s possible that thanks to corporate greed, these sources will soon no longer be available unless you have a big expense account. Fortunately the independent online media will continue to provide all of the non-proprietary articles (those from the news services), and bloggers, including an increasing number of freelance journalists who keep ‘mirror’ copies of their articles on their own websites, will keep almost all of this content online, free, indefinitely. It may be a bit harder to find, but it will be out there.

The bottom line for those browsing for information is that libraries (online and the old fashioned kind) are still the only way to go. Blogs could be excellent for browsing, but they’re constrained by the architecture of the current tools.

The bottom line for those who prefer alerts to get their information is to use RSS subscriptions for source alerts, and to be prepared for frustration if they’re using keyword alerts. The world needs much more sophisticated keyword alert tools, which will (a) filter out duplicate items, (b) filter out corporate press releases that are little more than spam disguised as news (“Megacorp announces innovative way to sell nothing for something”), and (c) broaden the reach of alerts to include new content in online libraries, websites and weblogs. None of these improvements will be easy. Many articles go through multiple drafts and rewrites and abridgements for different periodicals, so the ‘duplicate’ articles may be hard to spot. Some corporate press releases actually contain useful information. And there’s an ocean of material on websites and weblogs that information-seekers probably won’t consider useful information. A lot of experiments will be needed to design a keyword alert tool that strikes the right balance.

The task of making weblogs’ architecture more robust should be much easier. Weblog software with more dynamic information architecture would not only make blogs much more valuable to those browsing for information, they would make weblogs much more valuable in corporate environments. The current emphasis on adding ‘tagging’ information is, in my opinion, misguided: That would make their content easier to search, and might solve the information overload problem when they’re embraced by keyword search agents, but it won’t make them easier to browse. Much of the readership of weblogs is serendipitous — people stumble on them (usually through search tools) when they’re looking for interesting reading. Or, they blogroll a weblog because some of its content is of interest to them. What is needed is a way for people to browse through a selected subset of weblog content, all of the articles on a particular topic.

At present, weblogs can only be displayed one way: in reverse date order. After a week or so, the older content ‘disappears’ into the archives. The presumption is that it is no longer worth reading, like an old newspaper. But a weblog is not a newspaper, and many blogs have lots of information that has a long or even unlimited ‘shelf life’. The makers of blog tools have attempted to deal with this by allowing bloggers to set up ‘categories’, with each category separately subscribable using RSS. But each category still displays in reverse date order, and each ‘category’ is essentially a separate blog, with the category articles also disappearing into the archives.

The best analogy to the weblog is the newspaper or magazine columnist. They, too, write regularly on a variety of subjects, and would like to make their content available to people by subject, not just in reverse date order until it ‘disappears’. If they don’t write every day, like Malcolm Gladwell, they can at least fit all their article titles and links on one long page, and it’s awkward but not impossible to find and read everything he’s written on a particular subject. If you’re a weekly writer, like my friend Michel Dumais, finding all the articles on one subject can get ugly. Even a site-specific search bar like mine (which surprisingly few prolific bloggers provide) doesn’t help much if you’re not sure which keywords to use. If you want to read what I’ve written about the Wisdom of Crowds, do you use that phrase, or do you also have to check “collective wisdom” and “group intelligence” etc.?

If you’re working for a company and you want to read everything your company’s technology expert has written on a subject and his corporate weblog only lists the 8,000 articles in his ‘filing cabinet’ only in reverse chronological order, 20-per-page, chances are you’re going to give up before you find what you’re looking for. Yes, you can always search the corporate Intranet for everything containing both the subject in question and the expert’s name, but what if the subject goes by several names? And what if you get too many positives and have to wade through 100 articles one at a time to find the pertinent ones?

I have argued in past that the technology expert’s files, and Malcolm Gladwell’s and Michel Dumais’ columns, and the articles in each person’s blog, are analogous to the content in a personal filing cabinet. There is no ‘universal’ taxonomy that we would all agree upon for indexing this content by topic and sub-topic. But if we had access to the content the way the author would index it in his/her own filing cabinet, even if that isn’t the way we might choose to organize that content, it would certainly be much easier to browse.

In fact, new technology would allow you to index it in more than one way, so that, for example, an article like this, that talks about blogs in business, could ‘show up’ in two different filing cabinet organizing schemes, one by subject (blogs) and one by constituency of application (business). The most important thing requirement for each author to be able to organize content their own way(s), in whatever level of detail makes sense to them.

So how difficult would it be to allow each blog owner to set up their own taxonomy (filing cabinet ‘tabs’ and ‘subtabs’), and each time they write an article to ‘check off’ which ‘tabs’ it belongs under? You would of course need to be able to change and add to this organization, just as you might do if one section of your filing cabinet got too unwieldy.

And then, how difficult would it be to equip the weblog tool with a ‘toggle switch’ to allow readers to view articles by topic and subtopic, instead of just in reverse chronological order? So you would get a ‘table of contents’ on the ‘home page’, which would probably be in ‘outline view’ which you could expand to see the entire detailed taxonomy of the author, and the articles under each sub-topic.

This, I believe, would be all that is needed to make weblogs easy to browse, and hence far more useful as a research tool and as a ‘calling card’ for the author — and more enjoyable for serendipitous reading. And it could be all that is needed to allow weblogs to finally break into the business world and corporate intranets in a big way.