h a l f b a k e r y[marked-for-tagline]
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,
|
|
|
Looking from different timescales at the information that is in the Internet can make different content attention-worthy, because different processes take place at different speeds (see the picture "speed layers").
The search engine should look at the things that have been said about a particular
subject over a specified period of time. For example, the keyword "house" may have continuously (say, 20 years, 50 years, 1000 years, 6000 years) been related with certain descriptions, making the documents with these descriptions relevant.
Having had chosen a large timescale, recent trends would have negligible effects, so that the keyword "Olympics" would turn up the content that was considered important throught long period of time, which certainly has almost no relevance to "Beijing" or "London"; and choosing a smaller timescale would give results more influenced with recent trends.
The default would be the largest possible timescale to make us wiser.
LongNow.org
http://media.longno.../2/speed-layers.png
Speed Layers: Nature < Culture < Governance < Infrastructure < Commerce < Fashion [Inyuki, Jan 23 2010]
Planck Time
http://en.wikipedia.org/wiki/Planck_time
Quite quick .. [8th of 7, Jan 24 2010]
Please log in.
If you're not logged in,
you can see what this page
looks like, but you will
not be able to add anything.
Annotation:
|
| |
problem with that (if I'm reading the post correctly) is that webpages are often given "today's date" every day, no matter how long they've been around. |
|
| |
[FlyingToaster], yes, it is be a problem, but scientific publications are dated, books are dated. Considering that Google scans the web periodically, the newly seen web pages could perhaps be assumed to be created at the date they were seen. |
|
| |
I think Google cycles through the web either every 3 days or 10 days (vague recollection), but I'm constantly getting hits for info that I consider expired because of faulty (in my view) webpage markers, so you need another method of determining timescale. |
|
| |
I mean, if a new document Google saw first appear in the year 2005, it could with very high level of confidence be assumed that this document was created somewhere in the 20th or 21st century. |
|
| |
Of course, for the filter to work well in short timescales, more accurate dating is needed. |
|
| |
// more accurate dating is needed // |
|
| |
[8th of 7], but Google isn't God...
It depends on the needs of a user. |
|
| |
Are you going to tell them ? |
|
| |
// Are you going to tell them ? // |
|
| |
Well, I am not sure. Perhaps, if the need arises. |
|
| |