h a l f b a k e r yPlease listen carefully, as our opinions have changed.
add, search, annotate, link, view, overview, recent, by name, best, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
Please log in.
Before you can vote, you need to register.
Please log in or create an account.
|
Google, same as it is now, but with a
little
thing on each idea that denotes a
negative aspect. The
idea is that: You have a Google listing of
results, and you're currently pecking
through the list, only to find a result that
is
nothing like the desired objective.
Clicking on this 'minus'
button will not
only remove this entry from your results
list, but cleverly remove all others that
might similarly have been included for
similar reasons. It's just an additional
way
to sift and sort, but this time: within
results; and without using words.
Waxy.org links
http://waxy.org/links/
Look for May 16, 2005 [waugsqueke, May 20 2005]
Search Engine Watch
http://blog.searche.../blog/050923-132042
It doesn't work exactly as Ian describes, but certainly a step on the way there. [zen_tom, Sep 28 2005]
Matt Cutts' Blog
http://www.mattcutt...blog/remove-result/
Interesting blog by Matt Cutts, Google SEO [zen_tom, Sep 28 2005]
AllTheWeb
http://www.allthewe...g=crv&_sb_lang=pref
might do what you're saying. I haven't played with it for a while. Try searching for Bakery... It'll suggest other options which suit/don't suit [Dub, Oct 01 2005]
Report "spam" pages to google.
http://www.google.c...act/spamreport.html
[jutta, May 10 2007]
Clusty, the clustering search engine.
http://clusty.com/
Sorts results by tags, by source and by domain type. I tend to use this instead of google nowadays. [DrBob, Nov 27 2007]
[link]
|
| |
This would be similar to the Amazon music recommendations system (which used to be the FireFox music recommendations system). If it recommends something which you don't like, saying you don't like it will rebalance all the other recommendations. Applying this to search results would be nice though. |
|
| |
It might be useful when used in a manner opposite to the stated goal. In the creative thinking/problem solving process you might filter out the "good stuff" and look for inspiration in the remaining orthogonally related references. |
|
| |
Perhaps if the "votes" (filter set) were applied only to your search result set. Guard that cookie! |
|
| |
//Clicking on this 'minus' button will not only remove this entry from your results list, but cleverly remove all others that might similarly have been included for similar reasons//
[Briz], does this scentence not protect the cookie? |
|
| |
Google already does something with the opposite information, i.e. clicking a link you did like. Look at a Google results page: on the <a href=> link that takes you to your result, there is also an "onmousedown" event defined that calls a small function to log which ones you clicked on. |
|
| |
Maybe I'm missing something, but I read this as only applying to your search in progress, like the [similar pages] option after each result, but in reverse: [remove ones similar to this]. I like it, if that's what it is, but already it's pretty easy to remove a lot of wrong results by adding -(keyword). |
|
| |
I do what [oxen] does (-keyword), and I suspect that would be far more precise than a button. |
|
| |
True, but that only works if you know
what you're doing. I'm becoming aware
that although google is about as simple
an interface as it can possibly be, it's
still beyond a lot of ordinary people to
create a useful search string beyond
directly typing in the words in one block
as they think about it. There's rarely any
modification or optimisation to turn it
into an effective search string. |
|
| |
Yes, to clarify, this only applies to (and
within) your own results. |
|
| |
Waxy picked up on this link (see link about link) and pointed out that it's "badly needed, and would also provide aggregate stats to Google for removing spam." |
|
| |
(mis-annos duly yanked. Apols to
anyone who said something that I
shouldn't have deleted - by all means
say it again). |
|
| |
I've just noticed this feature pop up in my last few google results - so this idea would seem to be in the oven. |
|
| |
Having nosed around, it would seem to be in testing and is only available to people using the Personalised Search feature. |
|
| |
The level of data that Google must be starting to amass, on each and every one of us, is disquieting. Pray it's never nationalised by the US Gov't. |
|
| |
Bun! Google does sometimes have an option "More like this" (or something)... this'd be great. Sort of genetic search engine... Take a look at the link, too. |
|
| |
Aboot the Amazon/Fire Fox search engine thing, in order for Google to do such a thing, they would have to place all of the sites in to categories, then further break them down in to relations... |
|
| |
Too much work, since they have a LOT of info... |
|
| |
They actually already do this (to some degree). You make a search, and it notes all the sites you looked at and immediately left and those that you passed over. Upon making the same searh another similar search(even days later), these results are shuffled to the very bottom (usually a number in the 100K's or so depending on the speciality) |
|
| |
They use some great algorithms for this. Somewhat baked? |
|
| |
I was thinking that a scheme like this, along with registration of users to avoid unjustified attacks, would enable Google to eliminate the bogus content sites from its engines. |
|
| |
this is a nifty idea: but Google's search success lies in having a simple UI: improvements to it should ideally be invisible to the user. Perhaps you could modify the algorithm to factor in information about which links actually get clicked... that would be transparent, and might improve search results. |
|
| |
Good idea. Searching for a topic in different ways often keeps getting the same annoying irrelevant pages. Manually removing them and their associated pages would be very handy.
I do wonder, though, doesn't Google push pages with lots of Google-ads first nowadays? Even if they're annoyingly irrelevant? |
|
| |
That would surprise me very much. Not ever selling placement in the real search results (as opposed to the "sponsored links") is a pretty important core principle of theirs, and people's trust in, as ask.com's ads call it, the "algorithm" is arguably their greatest asset - so if they broke with that, and you could prove your claim, that would be a pretty big deal. (And would piss a lot of people off.) |
|
| |
I've been tinkering with such an idea, by adaptively applying results through Bayesian Classifiers. |
|
| |
A similar feature I would like would be an "other-than" notation, which would ask for instances of a word which occurred somewhere other than in a particular phrase, but which would not exclude all pages that happened to include that phrase. |
|
| |
For example: "(otherthan-Joe's)-Crab-Shack" would return pages containing "Fred's Crab Shack", even if they also contained references to Joe's, but would not return pages which didn't any reference any "Crab Shack" other than Joe's. |
|
| |
[supercat] just put a minus in front of the term you don't want. -joes +"crab shack" is what you want. Works fine. |
|
| |
That does the job in practice, but it's not quite the same. Consider
A "Joe visited Fred's Crab Shack"
B "Joe visited Joe's Crab Shack"
C "Fred visited Joe's crab shack"
D "Fred visited Fred's Crab Shack"
E "Bob visited Joe's Crab Shack and Fred's Crab Shack" |
|
| |
Supercat's expression matches A, D, and E because they have a Crab Shack other than Joe's.
Your query matches only D; A, B, C, and E all contain the word "joe", and -joe excludes them.
My own knee-jerk solution of -"Joe's Crab Shack" +"Crab Shack" matches A and D, but not E, because it happens to mention both Joe's and Fred's. |
|
| |
Well what you really need is a search engine that accepts regular expressions. //(^joe\'s) crab shack// |
|
| |
Also, there should be an explanation of why it doesn't fit as well. However, every single search result is like this when you get to about the 3rd or 4th page. |
|
| |