InformationOverload

InformationOverload is the problem of dealing with the sheer amount of information available. It is the common ground between sorting through the news, sorting through email spam, and the problem of how to build a search engine. It is the problem of reading articles on the web, and getting the feeling that each one was important and valuable, and then afterwards deciding that you spent too much time reading articles on the web. The first problem is that there is too much junk to sort through; the deeper problem is that there is too much good stuff to sort through.

As the amount of information increases, here are some bad aspects of InformationOverload:

• A decrease in personal time as people have to spend more and more time keeping up (reading the news, learning new technology, responding to email, surfing the web).

• Potential limit on the rate of technological growth. Technological development is perhaps a function of the ability of researchers to find techniques and findings relevant to their goals. As knowledge increases, a model of hierarchical specialization will become insufficient, as most knowledge sought by any given specialist will turn out to be analogous to an already-solved problem in a different area. I don't believe this has happened yet, but it is beginning. Example: i get the feeling that a lot of the advancement of the field of machine learning has been essentially a "refactoring" of the field of statistical inference.

• Potential limit on the rate of social change. Speed of social and political development is perhaps a function of the speed of effective propagation of the "conclusions" of debates. For instance, it seems to me that it is accepted that spending money on education is (via reducing poverty) a more efficient way to reduce crime than to spend money on enforcement/imprisonment. Yet this "conclusion" has not been effectively propagated; our political systems look as if they assume that spending money fighting crime directly is the most efficient way to reduce it. Similarly, it seems to me that US scientists reached a broad consensus that global warming is occurring a while ago, but US politicians have not yet received that information.

For me, this is an information overload problem. The opinions of ordinary people in a republic (to a large extent) drive these sort of decisions. Yet few people have time to read lots of political commentary. That is enough of a problem in itself, but a further problem is that it is hard for people with a very limited amount of time to choose to read only good "summaries".

The problem is that, if something is 80% agreed on but it is a "hot issue", to anyone who just hears about the issue casually through the media, it will sound as if it is not settled at all and both sides are "equal". This is a consequence of the necessity of not shutting out the 20% side (after all, they may still turn out to be right).

I feel wikis are a powerful weapon in the battle against InformationOverload, and I feel they could become even more so in the future. See WikisVsInformationOverload.

-- BayleShanks

How does one OrderChaos?

My diary [entry] for New York City was partially about InformationOverload. (More like advertisement overload.) I don't believe that it's a crisis because the world is already adapting to it. People are becoming more "media savvy", or at least media ignorant, by protecting their psyches from a bombardment of messages. Ignorance is not a big deal, because we can always learn what we need to know "just in time." (Compare with NoLogo) -- SunirShah

A very vaguely related concept is InformationLoss (this is the best place I could find to link from).

InformationOverload happens when people over-react to IntelligenceFailure.

Ironically, there's another page on this subject at CommunityWiki:InformationOverload.

I spend more time these days glossing, skimming, and occasionally reading a broad and depth amount of material, most of which is interesting and worth relating to Meatball. I keep a lot of ideas in my head or on my memo pad (which is more available than the Internet), but I need a way of organizing it. Research blogs are popular amongst Ph.D.s who have to organize their reading over a long period of time, but they are not AppropriateMedia for the task of finding relationships between concepts, and their chronological ordering has nothing to do with the task of note-taking. I think wikis are a better model.

I also think that with all these researchers in the same field as I am suffering the same problem there is a serious need to combine resources so we don't all drown. Understandable summarizes, relationships between concepts, SocialNetwork?s, and lists of AcademicCitations constitute the majority of an academic's work, and this level of work can be shared. I have strong hopes for something along the lines of the GuildModel--or more simply stated, a wiki--to solve this problem. -- SunirShah, trying desperately to just-in-time learn at the Wizards of OS Conference

I feel the same way -- BayleShanks

If there is a profitable demand, there is a business idea for it. There could be training companies that train workers and professors to take their materials and really make it easy to process versus some obtuse writing style. Companies who will take your work and then process the information to be easier to process for others.

One thing I find about wikis is that they are too easy to write in and become overloaded. It allows people to be mentally lazy when they write and not make it well processed writing. Lazy writing and the infinite loop nature of wikis can lead to people eating more than they can chew wandering the links to other pages. --MahyarMcDonald?

There is also a certain threshold stopping WikiGnomes? from rewriting less ordered, "tacked on" (like this) or opinion-like comments in order to separate the informational wheat from the chaff so to say. Likewise to improve the language of text, especially if it has a signature on it. It seems that the article vs discussion model of e.g. mediawiki/wikipedia might alleviate this. Easy writing going to the discussion side and processed text going to the article.

Funny thing is, to throttle InformationOverload, people would actually need more information. (Instructions, methods, programs, algorithms, contacts...) If you just step outside the flow and try to pick at the edges, you aren't likely to hit many good bits. More likely you'll just get spam.