home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: alt.cyberspace
- Path: sparky!uunet!paladin.american.edu!darwin.sura.net!sgiblab!octela!shaun
- From: shaun@octel.com (Chocolate Flavored Clorox)
- Subject: Re: MEMES (was: RE: Cybrspc Grvty, Def of Cybrspc)
- Message-ID: <1993Jan24.171024.14359@octel.com>
- Organization: Octel Communications Inc., Milpitas Ca.
- References: <HURH-190193203747@131.225.123.13> <1993Jan20.040126.10021@iscsvax.uni.edu> <HURH-200193203017@131.225.123.13>
- Date: Sun, 24 Jan 1993 17:10:24 GMT
- Lines: 43
-
- In article <HURH-200193203017@131.225.123.13> HURH@FNAL.FNAL.GOV (Patrick Hurh) writes:
- >
- >I'm not saying that talking and hypothesizing about the flow of information
- >is impossible (it's necessary for survival) but that a model in
- >mathematical terms (and in any qualitative terms) is impossible.
-
- It is difficult to analyze the flow of information through a network, but it
- is relatively trivial to analyze the flow of data. This does give you a bit
- of leverage; if you know that a node in the network has no users and is an
- archive site specializing in X, you can reasonably subtract the amount of
- input data from the amount of output data, (thereby subtracting pass-through
- data) and be fairly sure that the rest is information concerning X. A sudden
- jump in information concerning X can be correlated manually with other events,
- such as a similar site going down, the release of an album by a popular band
- that is a member of set X, a new release of software that is a member of set
- X, an environmental disaster such as an oil spill, from which data in X is
- useful, etc.
-
- Another example is the set of regularly-posted articles regarding readership
- and traffic of Usenet Newsgroups. That set of articles alone provides a
- pretty good general indication of information flow, at least at the topical
- level. More detailed analysis would be time consuming, but could be done
- via monitoring of email list traffic, say. Much of Usenet is now available
- on CD ROM, so correlating an unexpected rise or fall in traffic dealing
- with a subject could be done simply by popping in the correct CD ROM and
- looking at the articles in the newsgroup in question.
-
- Finally, however, I would suggest that even the analysis of Usenet by the
- above method wouldn't cover the majority of "interesting" information,
- which is probably be sent via email person to person. I know I use email
- much more heavily than news, and deal with far more detailed issues there
- than I do when using Netnews.
-
- (BTW, that was a great article, Pat.)
-
- @Man
- Up The Line CD-ROM Guy
- --
- shaun@octel.com || You can fix that throat with a pair of pliers.
-
- ''Sometimes sysadmins can be cool, sometimes they can be living in a totally
- alternate dimension where the parameters of "right" and "wrong" are mixed
- all up in a high-speed blender.'' -- Mr. K. Cheeseball of Brandon, Manitoba
-