ATTENTION: You are viewing a page formatted for mobile devices; to view the full web page, click HERE.

Main Area and Open Discussion > Living Room

A Very Simple Ethical Principle for Search: Google Fails Miserably

<< < (5/9) > >>

dk70:
Settle for enjoying Google with CustomizeGoogle (including links to other search engines) and/or some Greasemonkey scripts like http://userscripts.org/scripts/show/8477 and http://userscripts.org/scripts/show/6136 Not to forget Bumblesearch http://www.bumblesearch.com/bsearch/home/ check the shots. Dont be too idealistic on Googles motives or power of top 10 hitlists, dump default search page and make advanced one default, increase hits to 20 or higher. Lots of ways to improve searching http://www.googleguide.com/

Seems to be there are more important issues with Google position than dusty claims of evil cookies/ID tracking. Think even CustomizeGoogle author jokes a bit about the "paranoid" user. Ive yet to see some proof of unauthorized or not the same as everywhere else tracking. Even tiny requirements will bring some people in trouble as was seen when they launched Google Notes with "public" as default setting. Stories that snowball and NOW Google is insecure blah blah... Everyone is free to change to alternatives, like the safer Yahoo, Microsoft!

Google search is like a so so program which becomes way better once tweaked. Considering size of money machine they are very friendly, still! Or they are very good at appearing friendly  8)

Dont worry too much about "risky" ads. Google kick their own behind if not fighting scams/spam is top priority. Their algorithms of ranking will probably forever be internal secrets. So some SEO tricks are obvious but when to be sure? Will always be good question. Should be possible to Google without having full faith in everything they deliver - the opposite may be more scary!

I thought the conflict from users point of view was that if noting was "done" to Google we eventually will end up having to like Google cause it is the only one. MS give up, Yahoo dies out - plans of new comers to attack Google not likely to succeed. Advertising/revenue logic vs. organizing and guiding world information is just how internet works.

broken85:
Very well said superboyac!  :Thmbsup:

I also tend to use google now-a-days less for "searching" and more for "getting there" because there isn't much that I'm clueless on how to find anymore. Sometimes even if I know a company's website I'll search for the page I want in google so I don't have to look for the particular link in some other navigation scheme.


Slightly off topic of advertising, but about link relevancy in google...
I don't think any search engine currently developed can actually only return useful information... because people try very hard to get their content as far up the search results as possible. With Google this is much more complex, but the fact is, many relevant sites could care less about search engine placement, and many irrelevant sites will use every trick in the ever-expanding book to increase their search ranking for the very keywords you're looking for; and who's to say which will have the higher PageRank or be listed first? Something that simply parses code and displays results based on the very algorithms thousands of sites try to "persuade". And that's before you even factor in paid advertising which may or may not, now or in the future, be involved in the behind-the-scenes algorithms.

But it's the human aspect in us that lets us easily filter through even Google's results quickly and click on the links we *know* are at least somewhat relevant. I think it is (or should be) the ever-expanding goal of any all-encompassing search algorithm to do more and more of this work for the user so that we can start to see one page of relevant results instead of having to weed through them ourselves. But how does a search algorithm get around the limits of its static nature and those trying to trick it?

I suppose just by becoming more and more complex, and appearing less and less static. But I guess the question then is where do you draw the line? What I consider junk or advertising some users might consider the most relevant results. Perhaps since Google is already doing personalized searches and keeping account profiles, they could implement an adaptive set of algorithms that remember patterns (or the lack of certain patterns) in the results you actually click on and start ranking those types of results higher for you.

Or we could do away with automatic page ranking altogether and each end user could visit and rate every page in google's index for themselves. It might take several years (lifetimes?) but hey, we'd filter out the junk alright!  :D

Or, less sarchastically, it could be a collective project... have an open search engine which parses google's results but allows page rankings to be submitted by users (through a firefox extension maybe) which would average out, or be weighed somehow, and which would override google's rank for that particular site and place it accordingly in the results. Granted it would take some time before the database was large enough to make a collective difference, and would then be subject to its own need for filtering out junk that is submitted, but it would be one way of adding the 'human' aspect into google's search results without losing google's great algorithms and without relying on the page's webmaster or their code to define its relevance.

But I don't know how easy or fast it would be to filter through large numbers of google results in order to re-organize them on the fly. Google is quite fast though!

*EDIT*
More on point with mouser's topic, the above mentioned collective project could also allow junk sites, advertising sites, etc. to be re-ranked, or filtered out entirely based on user submittals... and maybe even its own algorithms and/or filters based on common junk/spam techniques or domains/URLs

Cpilot:
Maybe you could set up something like Digg, where users submit urls and everyone votes for em based on how commercial or not they are.
If their voted out then they get deleted and only the ones approved by the majority would be left. Then you could use a site search feature that would return only user approved links.

BTW mouser, I want to accumulate more than just enough to get by.
The prospect of eating dog food in my old age is very unappealing.  :mrgreen:

app103:
how about something like copernic. it returns a set of results by combining the results from several search engines.
-nudone (May 04, 2007, 04:44 PM)
--- End quote ---

I had Copernic a long time ago...and I loved it. I stopped using it about 5 years ago and I don't quite remember why. I think it was just easier to use Google.

I didn't even know they still had a web search product. I thought they switched over to desktop search and gave up trying to compete with Google.

Maybe it's time I give it another try.

And guess what? I get to upgrade to the most recent pro version for free! How cool is that?  :-*

Of course I'll still be waiting for StumbleUpon to become a full fledged search engine and not just a social bookmarking site.  :)

nudone:
i have to say i don't really have any complaints about google either. when looking through the results it returns, i will often view them way past page 10 - past page 50 if need be. like superboyac said, you can pretty much spot what looks like a relevent link so it doesn't take that long to go through a load results. and, of course, usually you'll find a site that will provide further links that are even more relevent to your original search. i think it's called web surfing.

google still seems to be the best out of the bunch. it will be interesting to hear how app get along with copernic.

Navigation

[0] Message Index

[#] Next page

[*] Previous page

Go to full version