h a l f b a k e r yBaker Street Irregulars
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
Vernon-like theory post warning for those who'd rather not bother.
There's rather been a dichotomy of opinions out there as to whether we as a global society are getting generally getting *better* or *worse,* in matters of a per capita average of high culture, intelligence, mass education, creativity
potential, ethics, manners, civility, whatever. (Personally I think it depends on the topic). However, there is an observational relative reference frame problem that I see cropping up; if you're lumbering along at the same speed as everything else, you really don't know which direction you're headed, or at what speed you're getting there, of if you're moving at all. Your view is polluted with relativity issues, limited life-time issues, and hampered by the fact that measurement of any system response requires a faster "sensor" than the system's capable response.
But if there were a standout, a radical member that was getting more intelligent or less so much more quickly than the general population as to be observable, and yet was moving in the same general direction, it would be noticed.
If a fast enough "sensor" is put into place, then some differential response can be effected. Consider, for example, the use of birds as indicators of radiation. Because they are more sensitive to radiation exposure, they serve as an effective early-warning system that there's a radiation leakage issue somewhere, and the crew can then shut the system down. There was a fast sensor comparison difference, (bird was living, now dead), that caused a system response, (engineering fixes radiation leak), i.e. a feedback loop consisting of several semi-independent systems.
But where do we look to find a good early-warning indicator that the world is getting dumber or smarter? The measurement system would need to have an observable comparison of states of where we were versus where we are. It would also need to have a monstrously vast scope to capture as much data as possible. And it would need a signal response much faster than the general population's rate of change.
The net will arguably fulfill all of those requirements once a true representative population cross-section is hooked up; I argue that it serves as a perfect early indicator of global trends. It is likely to continue to evolve quicker than the general population culture; it is consistent with the direction we're heading, and it's searchable, storable, and transferrable.
So, in order to get a more concrete fix on global trends, one simply compares the activites of the net with the same activites a few years ago, and "subtracts." While it still doesn't tell us if there happens to be a cliff at the end of the road, it makes for a steady course.
Now, I'll acknowledge that there are many inherent assumptions that are totally unwarranted in this; like the assumption that there's a simple positive multiplier correlation between the activities of the net and the rest of the world, for instance. But, could be useful.
Google Zeitgeist
http://www.google.c...ress/zeitgeist.html
[krelnik, Oct 04 2004, last modified Oct 06 2004]
[link]
|
| |
Kind of like the Google Zeitgeist, but focused on more important stuff? |
|
| |
Don't worry, I was confusing myself when I wrote it. Another one of my hard-to-explain theories. Maybe I should chew on them for a week or so before posting. |
|
| |
Thinking about this a little more, there's really not that much to it besides using the net as a leading indicator of cultural shift forwards or backwards. |
|
| |
Yeah, something like the Google Zeitgeist would be a good place to start, with a team of analysts rating high-hitting concepts on some scale. Add in a spider that analyzes 'blogs and sites for estimated reading level and caches the results every 5 years or so. |
|
| |
Rods, I've thought about that, and I'm not certain it matters, as the difference could be assumed to be a near constant for the before and after states. If we're all connected in some way, then whatever state the education level of the net is in now, then a decrease would indicate a general overall population decrease as well. Ie... if the smartest of us are perceivably dumber than a comparitive cross section a few decades ago in measurement x, would that indicate that the rest of the world is dumbing down with them, or that things are leveling out between classes? |
|
| |
'Course the theory assumes a near steady-state in the way the net is used and its benefits. A radical technological departure that affects communication in some radical way could throw the whole thing off. |
|
| |
So, uh, what do you actually look for? Are there keywords that are "smart" or "dumb"? Could you really say anything useful? I completely cannot fathom how you could hope to mine a "cultural progress meter" out of petabytes of porn and All Your Base and a/s/l and Nigerian spam and whatever else fills up the net. |
|
| |
Lessee, what would be a good set of predictors:
Estimated reading level.
Size of vocabulary used versus previous takings.
'chunkiness' or repetition of topics, ie the tendency for the masses to follow popular trends in mindless swarms versus defining their own input.
The rate of turnover in new topics and significant creations.
There's probably hoards of other indicators.
One interesting test would be to adapt blaise's heuristic patent search techniques into an experimental generator of net topics, events, and issues, and then run a comparison of whether the machine or the man would win out. If the test group beats the machine, or vice-versa, then by how much? Run the test once or twice every decade. Note that it would be important to NOT upgrade the machine, as it serves as the control benchmark. |
|
| |
I think as the net ages, there would be a great many things to learn about cultural shift in general in tracking its development over a long period in time. It will simply be interesting to have a few decades of net history in the wayback machine. |
|
| |
Okay. We have to be careful to weight appropriately, though. Because certain text appears in great volume doesn't mean it's representative. |
|
| |
For example, the introduction of spam probably worsens many of your metrics, since there's a lot of e-mail going around that's repititive and badly written and full of lame commercial content -- but that's all the actions of relatively few people and not representative of society as a whole. |
|
| |
You could try to filter spam out, but in general as the "shape" of producer/consumer relationships and ratios and protocols change it will be hard to track just what people are actually spending their time reading and writing (as opposed to what is being transmitted). |
|
| |
None of this is insuperable and I agree that historical data will be interesting to mine in all sorts of ways that weren't practical in the dead tree era. |
|
| |
It may be that the net mirrors the culture to som degree, and the term "net" helps connote the fact that cultural change is not unidirectional. I envision culture as like a slime mold, that sends out pseudopodia in all directions: productive directions result in a shift of bulk and resources down that pseudopod - unproductive pseudopodia are pruned. I think that leading indicators are unlikely to be found on the net, just as the hippest clothing is not advertised on prime time. Some other index would need to be found. |
|
| |