Please Police Your Google Search Parameters December 10, 2011 8:39 AM Subscribe
I'd like to respectfully ask everyone to please clean up their Google search URLs.
Having to sift through
This is especially true if the reader has no idea what to look for amongst all that gobbledygook. Even if they do know what to look for, nobody needs to know that a poster's host language is English ('hl=en'), that their browser is 1205 pixels wide ('biw=1205'), or that they search with SafeSearch off ('safe=off'). Indeed, perhaps posters don't want people to know these things.
Note that there are two 'q' parameters in that search. The first one has a misspelling and the poster clicked on the suggestion, which is why 'spell=1' appears in the URL. A poster knows they clicked on a "Did you mean…". A reader can only deduce this from the other parameters in the URL.
Unless you're demonstrating the use of one of the other parameters to a Google search or using an advanced search feature, the only thing you need is the 'q' parameter. For the above that would be
I note for the record that while this post is not specifically about this morning's post about Google now graphing equations, I admit that some of the examples in that thread did lead me to kvetch about it here.
Having to sift through
<URL: https://encrypted.google.com/search?q=kvetchin&ie=utf-8&oe=utf-8&aq=t&rls=org.mozilla:en-US:official&client=firefox-a#hl=en&safe=off&client=firefox-a&hs=kdQ&pwst=1&rls=org.mozilla:en-US:official&sa=X&ei=qYXjToeYLoTMtgfU8qDZBA&ved=0CBgQvwUoAQ&q=kvetching&spell=1&bav=on.2,or.r_gc.r_pw.r_cp.,cf.osb&fp=ee869b7db81e9425&biw=1205&bih=1020> to find the relevant 'q' parameter can be maddening. This is especially true if the reader has no idea what to look for amongst all that gobbledygook. Even if they do know what to look for, nobody needs to know that a poster's host language is English ('hl=en'), that their browser is 1205 pixels wide ('biw=1205'), or that they search with SafeSearch off ('safe=off'). Indeed, perhaps posters don't want people to know these things.
Note that there are two 'q' parameters in that search. The first one has a misspelling and the poster clicked on the suggestion, which is why 'spell=1' appears in the URL. A poster knows they clicked on a "Did you mean…". A reader can only deduce this from the other parameters in the URL.
Unless you're demonstrating the use of one of the other parameters to a Google search or using an advanced search feature, the only thing you need is the 'q' parameter. For the above that would be
<URL: https://encrypted.google.com/search?q=kvetching>. This is much clearer for both novice and expert alike.I note for the record that while this post is not specifically about this morning's post about Google now graphing equations, I admit that some of the examples in that thread did lead me to kvetch about it here.
Unless google offers that as an option somewhere on the results page that I have never noticed, I will never, ever bother to do that. Why would a reader want to look for anything amidst the goggledygook? Isn't the point for them to click the link and view the search results?
posted by jacquilynne at 8:47 AM on December 10, 2011 [17 favorites]
posted by jacquilynne at 8:47 AM on December 10, 2011 [17 favorites]
I think you need to explain why, exactly, you think this is a problem before you ask everyone to change their behavior.
posted by MadamM at 8:59 AM on December 10, 2011 [9 favorites]
posted by MadamM at 8:59 AM on December 10, 2011 [9 favorites]
You need to suggest an easy way to do this.
posted by Kwine at 9:10 AM on December 10, 2011 [1 favorite]
posted by Kwine at 9:10 AM on December 10, 2011 [1 favorite]
I think you'd have more luck getting Google to change their behavior than the Mefi user base.
posted by Roger Dodger at 9:10 AM on December 10, 2011 [7 favorites]
posted by Roger Dodger at 9:10 AM on December 10, 2011 [7 favorites]
I try to clean my stuff up when posting youtube videos and such, but I get annoyed when the link doesn't work after I cut out when looks like junk to me. Then I have to go back and try again and that's even more annoying.
posted by cjorgensen at 9:24 AM on December 10, 2011 [1 favorite]
posted by cjorgensen at 9:24 AM on December 10, 2011 [1 favorite]
I guess I understand what you're saying, but I have no idea how to clean up a google search url.
So, how exactly do you clean it up?
posted by insectosaurus at 9:25 AM on December 10, 2011 [5 favorites]
So, how exactly do you clean it up?
posted by insectosaurus at 9:25 AM on December 10, 2011 [5 favorites]
What are you talking about?
posted by dfriedman at 9:29 AM on December 10, 2011 [11 favorites]
posted by dfriedman at 9:29 AM on December 10, 2011 [11 favorites]
URLs in google search results actually route you through Google's servers for metrics and other secret voodoo Google do, not directly to the site. So if you grab a url from a Google search and not from the actual page, it'll be loaded with all the Google server metric cruft.
posted by crunchland at 9:33 AM on December 10, 2011 [1 favorite]
posted by crunchland at 9:33 AM on December 10, 2011 [1 favorite]
The way you clean it up is to actually grab the url from the page Google sends you to, not from a Google search result page.
Is there some sort of a Greasemonkey script I could use that would just take all the other cruft off of them? --- While we're on this subject, does anyone know of a greasemonkey script that automatically chops off Amazon referrer codes?
posted by crunchland at 9:35 AM on December 10, 2011
Is there some sort of a Greasemonkey script I could use that would just take all the other cruft off of them? --- While we're on this subject, does anyone know of a greasemonkey script that automatically chops off Amazon referrer codes?
posted by crunchland at 9:35 AM on December 10, 2011
Does anyone know of a greasemonkey script that renders splenetic techno-speak into language discernible by humans?
posted by dfriedman at 9:39 AM on December 10, 2011 [8 favorites]
posted by dfriedman at 9:39 AM on December 10, 2011 [8 favorites]
crunchland, I think the OP is talking about URLs for search results (like this: Google search for Metafilter) not the links google uses to send you to the pages on its search results (like this: link to Metafilter through Google). I think you're talking about the second.
posted by MadamM at 9:44 AM on December 10, 2011
posted by MadamM at 9:44 AM on December 10, 2011
Did you mean tilling at windmills?
posted by Rock Steady at 9:45 AM on December 10, 2011 [4 favorites]
posted by Rock Steady at 9:45 AM on December 10, 2011 [4 favorites]
If you do a google search for "duck soup", the first result you get from google is IMDB.
The url for that first result is : http://www.imdb.com/title/tt0023969/
But if you right click and copy the link address from the google search page, this is what you get: http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&sqi=2&ved=0CDoQFjAA&url=http%3A%2F%2Fwww.imdb.com%2Ftitle%2Ftt0023969%2F&ei=uJrjTuj6BKn50gGNh6zjBQ&usg=AFQjCNEnkZIHNq3nWKC6K3JnGXSNRfvYdA
posted by crunchland at 9:48 AM on December 10, 2011
The url for that first result is : http://www.imdb.com/title/tt0023969/
But if you right click and copy the link address from the google search page, this is what you get: http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&sqi=2&ved=0CDoQFjAA&url=http%3A%2F%2Fwww.imdb.com%2Ftitle%2Ftt0023969%2F&ei=uJrjTuj6BKn50gGNh6zjBQ&usg=AFQjCNEnkZIHNq3nWKC6K3JnGXSNRfvYdA
posted by crunchland at 9:48 AM on December 10, 2011
crunchland, I think MadamM is right, you and the OP are talking about different things. Look at the cited example post - it's about full of links that search google, not links that are bad copy/pastes of google search results.
posted by donnagirl at 9:52 AM on December 10, 2011
posted by donnagirl at 9:52 AM on December 10, 2011
While we're on this subject, does anyone know of a greasemonkey script that automatically chops off Amazon referrer codes?
Why would you do want to increase Amazon's margins? Are you a shareholder?
posted by michaelh at 9:53 AM on December 10, 2011
Why would you do want to increase Amazon's margins? Are you a shareholder?
posted by michaelh at 9:53 AM on December 10, 2011
crunchland, yeah, that's exactly what my second example is, if you look at the URL (and I actually think people should avoid that, because it hides where the link is actually going). That isn't what the OP is talking about.
posted by MadamM at 9:54 AM on December 10, 2011
posted by MadamM at 9:54 AM on December 10, 2011
Ah. Ok. Carry on, then. Hope I didn't add to the confusion.
Why would you do want to increase Amazon's margins? Are you a shareholder? --- No. I'd just rather not increase the margins of some of the sites I visit... (And no, I don't mean Metafilter.)
posted by crunchland at 9:55 AM on December 10, 2011
Why would you do want to increase Amazon's margins? Are you a shareholder? --- No. I'd just rather not increase the margins of some of the sites I visit... (And no, I don't mean Metafilter.)
posted by crunchland at 9:55 AM on December 10, 2011
Is this something that could be done on Metafilter's side when comments are posted?
posted by egg drop at 9:57 AM on December 10, 2011 [1 favorite]
posted by egg drop at 9:57 AM on December 10, 2011 [1 favorite]
I completely agree with this suggestion. Posting a google link with all this crap in it is the geek equivalent of forgetting to put your pants on when leaving the house: not a big deal, but it makes you look like an idiot and we probably don't want to see it anyway.
posted by Dr Dracator at 10:03 AM on December 10, 2011 [2 favorites]
posted by Dr Dracator at 10:03 AM on December 10, 2011 [2 favorites]
For what it's worth, I hate hate hate that thing you're talking about, crunchland, and specifically have gotten more consistent about clicking through and grabbing the URL to avoid accidentally sticking crazy cruft into pastes like that.
If there's some similarly slam-dunk-easy method for getting a minimal, cruft free link for the use case ob1quixote is talking about, that would be totally great, but I have to agree with the sentiment that de-crufting these links by hand is an unrealistic expectation. I feel pretty strongly about my anti-cruft stance but even I wouldn't bother to do it on more than exceptional occasions.
posted by cortex (staff) at 10:06 AM on December 10, 2011
If there's some similarly slam-dunk-easy method for getting a minimal, cruft free link for the use case ob1quixote is talking about, that would be totally great, but I have to agree with the sentiment that de-crufting these links by hand is an unrealistic expectation. I feel pretty strongly about my anti-cruft stance but even I wouldn't bother to do it on more than exceptional occasions.
posted by cortex (staff) at 10:06 AM on December 10, 2011
I searched a bit -
There's a bookmarklet that works in Firefox, but not Chrome.
There's a Greasemonkey extension, which doesn't work for me in either browser.
posted by Pronoiac at 10:07 AM on December 10, 2011 [1 favorite]
posted by Pronoiac at 10:07 AM on December 10, 2011 [1 favorite]
I have a very faint grasp of the concept here, honestly, but would running it through a tinyurl type service do the trick? For me - if I bothered - that would be a lot faster and easier than trying to decipher the code.
posted by Occula at 10:27 AM on December 10, 2011
posted by Occula at 10:27 AM on December 10, 2011
Oh, there's an updated bookmarklet that works okay in Chrome & Firefox.
The comments point to Matt Cutts originating the idea, who likes the cleaner links but has some caveats. Those caveats might be outdated; the pages look the same to me with or without the cleaner links.
posted by Pronoiac at 10:31 AM on December 10, 2011
The comments point to Matt Cutts originating the idea, who likes the cleaner links but has some caveats. Those caveats might be outdated; the pages look the same to me with or without the cleaner links.
posted by Pronoiac at 10:31 AM on December 10, 2011
would running it through a tinyurl type service do the trick?
We really try to keep short URLs off the front of the page [unless they're site specific like flic.kr or nytim.es] because they're even more obfuscatory than the long Google URLs.
posted by jessamyn (staff) at 10:32 AM on December 10, 2011 [2 favorites]
We really try to keep short URLs off the front of the page [unless they're site specific like flic.kr or nytim.es] because they're even more obfuscatory than the long Google URLs.
posted by jessamyn (staff) at 10:32 AM on December 10, 2011 [2 favorites]
What does "clicking through" mean in this context? Here's a search I did for kittens:
The first search result is kittenwar.com - you're saying I should just link to that?
posted by desjardins at 10:32 AM on December 10, 2011
https://www.google.com/#sclient=psy-ab&hl=en&site=&source=hp&q=kittens&pbx=1&oq=kittens&aq=f&aqi=g4&aql=1&gs_sm=e&gs_upl=570l1485l0l1735l7l6l0l0l0l0l212l914l1.4.1l6l0&bav=on.2,or.r_gc.r_pw.r_cp.,cf.osb&fp=3672847e84815f7e&biw=1280&bih=909The first search result is kittenwar.com - you're saying I should just link to that?
posted by desjardins at 10:32 AM on December 10, 2011
Running it through a url shortener wouldn't solve anything, it'd just add another layer of obfuscation: you'd have an information-free shorturl leading blindly to an information-poor crufty url.
Long urls aren't a problem on metafilter in the sense that they are long—we expect people to wrap links up in <a href> tags as a rule, so the length of the url is immaterial. The reason crufty urls are frustrating is because it can mean that what would otherwise be an additional source of information on what's behind a link, namely the url's well-structured domain-and-path text, is less useful because it's hard to pick out the informative bits of the url's path and parameters from all the junk no one cares about.
posted by cortex (staff) at 10:35 AM on December 10, 2011 [1 favorite]
Long urls aren't a problem on metafilter in the sense that they are long—we expect people to wrap links up in <a href> tags as a rule, so the length of the url is immaterial. The reason crufty urls are frustrating is because it can mean that what would otherwise be an additional source of information on what's behind a link, namely the url's well-structured domain-and-path text, is less useful because it's hard to pick out the informative bits of the url's path and parameters from all the junk no one cares about.
posted by cortex (staff) at 10:35 AM on December 10, 2011 [1 favorite]
desjardins, here's a cleaner url for that search:
https://www.google.com/#q=kittens&fp=1
It's easier to inspect, which I think is the reasoning behind the FPP's request.
(That link's what I got automatically from the bookmarklet, so, uh, dunno about the fp= bit.)
posted by Pronoiac at 10:42 AM on December 10, 2011
https://www.google.com/#q=kittens&fp=1
It's easier to inspect, which I think is the reasoning behind the FPP's request.
(That link's what I got automatically from the bookmarklet, so, uh, dunno about the fp= bit.)
posted by Pronoiac at 10:42 AM on December 10, 2011
What does "clicking through" mean in this context?
Clicking through means that if what you're trying to link to is a site that you found via search, you should click through to the site instead of right-click-copying the url from the google search results page because google will often interject a bunch of cruft.
If your goal is to link a page of search results, copy the url in your browser after you do the search and there you go. The url may be sort of crufty, but there's not often an obvious and simple approach to making it less ugly.
However, if your goal is to link to Kitten War but you don't remember the URL by hand, and you go to Google and type in "kitten war", you can do one of three things to create your link from there:
1. click on the Kitten War result and get to the kittenwar.com url that way and copy that url from the address bar to paste into your link.
2. right-click on the kittenwar.com url listing on the google results page and copy that to paste into your link.
3. say "oh, http://kittenwar.com, okay" and then type that manually into your link.
The problem is that (1) is more steps than (2), but (2) may silently interject a different url into your clipboard when you copy—a big long google intermediary url full of analytics stuff that is useful to Google but not to anyone else.
And because google does this a bit on the sly by presenting the apparent link text of the search result as the raw url rather than their mungy bullshit, a lot of people don't realize that (2) is a potential mess and so they do that instead of (1).
(3) Works great if you like typing and the url is memorably short, but it's obviously impractical for a lot of applications.
Hence, this weird clusterfucky counter-intuitive situation that's only tangential to the thing that ob1quixote is talking about.
posted by cortex (staff) at 10:43 AM on December 10, 2011 [1 favorite]
Clicking through means that if what you're trying to link to is a site that you found via search, you should click through to the site instead of right-click-copying the url from the google search results page because google will often interject a bunch of cruft.
If your goal is to link a page of search results, copy the url in your browser after you do the search and there you go. The url may be sort of crufty, but there's not often an obvious and simple approach to making it less ugly.
However, if your goal is to link to Kitten War but you don't remember the URL by hand, and you go to Google and type in "kitten war", you can do one of three things to create your link from there:
1. click on the Kitten War result and get to the kittenwar.com url that way and copy that url from the address bar to paste into your link.
2. right-click on the kittenwar.com url listing on the google results page and copy that to paste into your link.
3. say "oh, http://kittenwar.com, okay" and then type that manually into your link.
The problem is that (1) is more steps than (2), but (2) may silently interject a different url into your clipboard when you copy—a big long google intermediary url full of analytics stuff that is useful to Google but not to anyone else.
And because google does this a bit on the sly by presenting the apparent link text of the search result as the raw url rather than their mungy bullshit, a lot of people don't realize that (2) is a potential mess and so they do that instead of (1).
(3) Works great if you like typing and the url is memorably short, but it's obviously impractical for a lot of applications.
Hence, this weird clusterfucky counter-intuitive situation that's only tangential to the thing that ob1quixote is talking about.
posted by cortex (staff) at 10:43 AM on December 10, 2011 [1 favorite]
Okay, I see I was thinking that we were talking about linking to a page of search results. I'm fine with un-crufting a URL found via search; that's not any more difficult than copying from the address bar.
posted by desjardins at 11:05 AM on December 10, 2011
posted by desjardins at 11:05 AM on December 10, 2011
dfriedman: What are you talking about?
I don't mean to speak for the OP, but let me see if I can explain the issue as I understand it.
Suppose you are writing a comment on Metafilter and wish to refer to a page of Google search results, using a link. So, you go to Google's main page, you type "kvetching" into the search box, you hit enter, your browser displays the search results page. You copy the URL from the address bar of your browser and stick it into
The thing is, the URL that the Google search form produces (which your browser then uses to request the search page) is full of information which Google wants but which doesn't have anything to do with the query you actually made (and which you wanted to refer people to). In the post, ob1quixote gives several examples of such information.
Why is this a problem? Here are three answers:
1. It isn't a problem. The link works. Stop your complaining.
2. It's a privacy problem. The URL includes information about your computer or browing history that you might not want us to have.
3. It's a communication problem. People who try to read the URL to understand what they're being referred to (instead of just clicking on the link) will have trouble because of all the irrelevant information.
What can you do if you agree that this is a problem and want to produce "clean" Google search URLs? Well, you have to learn a bit about URL syntax. You can mostly get by just following the template
dfriedman: language discernible by humans
I suppose you were frustrated when you wrote this, and I understand how that feels, but please don't do this. Technical knowledge and language are human works, devised by humans to serve human purposes. We who use them are as human as you, no matter how alien you find our interests.
posted by stebulus at 11:43 AM on December 10, 2011 [20 favorites]
I don't mean to speak for the OP, but let me see if I can explain the issue as I understand it.
Suppose you are writing a comment on Metafilter and wish to refer to a page of Google search results, using a link. So, you go to Google's main page, you type "kvetching" into the search box, you hit enter, your browser displays the search results page. You copy the URL from the address bar of your browser and stick it into
<a href="URL">text</a>. Right?The thing is, the URL that the Google search form produces (which your browser then uses to request the search page) is full of information which Google wants but which doesn't have anything to do with the query you actually made (and which you wanted to refer people to). In the post, ob1quixote gives several examples of such information.
Why is this a problem? Here are three answers:
1. It isn't a problem. The link works. Stop your complaining.
2. It's a privacy problem. The URL includes information about your computer or browing history that you might not want us to have.
3. It's a communication problem. People who try to read the URL to understand what they're being referred to (instead of just clicking on the link) will have trouble because of all the irrelevant information.
What can you do if you agree that this is a problem and want to produce "clean" Google search URLs? Well, you have to learn a bit about URL syntax. You can mostly get by just following the template
http://www.google.com/?q=kvetching, though you'll run into issues if your search terms contain spaces, or characters special in URL syntax.dfriedman: language discernible by humans
I suppose you were frustrated when you wrote this, and I understand how that feels, but please don't do this. Technical knowledge and language are human works, devised by humans to serve human purposes. We who use them are as human as you, no matter how alien you find our interests.
posted by stebulus at 11:43 AM on December 10, 2011 [20 favorites]
stebulus, thank you for that explanation. I wasn't even aware that I was sending data about my browser and computer when I linked to things.
posted by vincele at 11:56 AM on December 10, 2011
posted by vincele at 11:56 AM on December 10, 2011
Well, often you're not. Lots of URLs contain only the information needed to identify to the page being linked to. Google search results pages are an exception, though a frequently encountered one. (Amazon links are often full of cruft too.)
posted by stebulus at 12:11 PM on December 10, 2011
posted by stebulus at 12:11 PM on December 10, 2011
Good post. This is a problem, for the same reason we don't allow posts with mystery meat, link-shortener-shortened URLs. The address should be a clear indication of the destination and shouldn't include unnecessary Google marketing cruft.
posted by Blazecock Pileon at 12:22 PM on December 10, 2011 [1 favorite]
posted by Blazecock Pileon at 12:22 PM on December 10, 2011 [1 favorite]
The problem is that (1) is more steps than (2), but (2) may silently interject a different url into your clipboard when you copy—a big long google intermediary url full of analytics stuff that is useful to Google but not to anyone else.
Yeah, this is frustrating, and something google only started doing not that long ago, I think. Why the hell do they even do that? They know your browser fingerprint already, so why replicate that information when some third party clicks on a link?
Anyway, I appreciate what the OP says because I always look at the browser footer to see what I am clicking on, but I don't think directly editing the URL is going to be the way forward for most Meef.
posted by Rumple at 1:01 PM on December 10, 2011
Yeah, this is frustrating, and something google only started doing not that long ago, I think. Why the hell do they even do that? They know your browser fingerprint already, so why replicate that information when some third party clicks on a link?
Anyway, I appreciate what the OP says because I always look at the browser footer to see what I am clicking on, but I don't think directly editing the URL is going to be the way forward for most Meef.
posted by Rumple at 1:01 PM on December 10, 2011
Ah, I guess I helped with the original request, but not the related problem. So I searched. Google Tracking-B-Gone is a Greasemonkey script that stops Google click tracking, so you can copy useful links from the search page again.
posted by Pronoiac at 1:02 PM on December 10, 2011 [1 favorite]
posted by Pronoiac at 1:02 PM on December 10, 2011 [1 favorite]
Couldn't MetaFilter just do the cleanup automagically like it does to Amazon links? (Pretty please?)
This is just one of many things that pisses me off about Google. The other big thing is that lately I'm constantly having to clarify, 'No, goddamn it, I'm not trying to search for everything, just thing.'
posted by Sys Rq at 1:07 PM on December 10, 2011 [3 favorites]
This is just one of many things that pisses me off about Google. The other big thing is that lately I'm constantly having to clarify, 'No, goddamn it, I'm not trying to search for everything, just thing.'
posted by Sys Rq at 1:07 PM on December 10, 2011 [3 favorites]
Here's a fairly quick way to get a decrufted URL for search results:
(1) Run the search.
(2) Click "Images" on the left (or "News" or "Shopping" or "Videos", if it's something you'd rather not see pictures of).
(3) Copy the part of the url up until (but not including) the first &, which should come right after the q=. For example:
https://www.google.com/search?q=foo&hl=en&safe=off&prmd=imvnsr&source=lnms&tbm=isch&ei=ysjjTv7MJOHv0gGY3ZT2BQ
&sa=X&oi=mode_link&ct=mode&cd=2&ohjesusgod&ved=0CBYQ_AUoAQ&whatisallthisstuff&biw=1050&bih=627
The part in bold is a link to a web (not image) search for "foo".
posted by jhc at 1:09 PM on December 10, 2011 [1 favorite]
(1) Run the search.
(2) Click "Images" on the left (or "News" or "Shopping" or "Videos", if it's something you'd rather not see pictures of).
(3) Copy the part of the url up until (but not including) the first &, which should come right after the q=. For example:
https://www.google.com/search?q=foo&hl=en&safe=off&prmd=imvnsr&source=lnms&tbm=isch&ei=ysjjTv7MJOHv0gGY3ZT2BQ
&sa=X&oi=mode_link&ct=mode&cd=2&ohjesusgod&ved=0CBYQ_AUoAQ&whatisallthisstuff&biw=1050&bih=627
The part in bold is a link to a web (not image) search for "foo".
posted by jhc at 1:09 PM on December 10, 2011 [1 favorite]
I do this when I post a link to Google books, as it removes any keyword highlighting I might have used locating the quote I wanted which will likely be distracting to a different reader. Otherwise can't recall ever having recourse to post a Google search here; do agree about how irritating the obscured interstitial link is when going from a set of search results - it'll often trigger automatic Internet censorship here (because that seems partly sponsored by Baidu to screw up Google) where the plain link won't.
posted by Abiezer at 1:10 PM on December 10, 2011
posted by Abiezer at 1:10 PM on December 10, 2011
Is there some sort of a Greasemonkey script I could use that would just take all the other cruft off of them?
This script has been updated more recently than most others and seems to work well enough.
posted by Lorin at 2:00 PM on December 10, 2011
This script has been updated more recently than most others and seems to work well enough.
posted by Lorin at 2:00 PM on December 10, 2011
Out of curiosity, how hard would it be to automatically strip all that junk from Google links (the way that Amazon links all get the Metafilter referral code added?) At least, strip "&hl" and everything after it?
posted by SMPA at 2:19 PM on December 10, 2011
posted by SMPA at 2:19 PM on December 10, 2011
Question - if I use the crufty link in a comment and 500 people click on it, are those 500 clicks attributed to my browsing history? In other words, if I link to a Google search for kittens, am I all of a sudden going to be inundated with cat food ads?
posted by desjardins at 2:21 PM on December 10, 2011
posted by desjardins at 2:21 PM on December 10, 2011
It should be noted that even for a simple, "clean" Google search link, such as: https://www.google.com/search?q=matt+haughey, your results are probably different from my results which are probably different from Matt Haughey's results, because Google is using information like search history and location to tailor results for you. (Some info here.) To test this, try the above search with your browser in incognito mode, and compare that with the normal result.
Once upon a time we all saw the same results in the same sequence on any particular search term. So readers clicking through on a search link all saw the same Google results page. But now, including a search results link in a post may not show people what you want it to show, at all.
posted by beagle at 2:27 PM on December 10, 2011 [4 favorites]
Once upon a time we all saw the same results in the same sequence on any particular search term. So readers clicking through on a search link all saw the same Google results page. But now, including a search results link in a post may not show people what you want it to show, at all.
posted by beagle at 2:27 PM on December 10, 2011 [4 favorites]
Couldn't MetaFilter just do the cleanup automagically like it does to Amazon links? (Pretty please?)
Out of curiosity, how hard would it be to automatically strip all that junk from Google links
Here's a way to automagically clean up the links after they're posted to Metafilter. For example, you could use it on this page to clean up the four example Google searches upthread. This works even if people don't change their posting behavior.
I don't use Greasemonkey, but I cobbled together a bookmarklet that scans all the links on any given web page, finds links that look like Google URLs, scans the query string for salient bits and then rewrites everything in place without all the googlecruft. No reason it couldn't be incorporated into the site or done server side, but this way allows it to work without any intervention from the powers-that-be.
Just install it as a bookmarklet (or, if you're industrious, convert it into a Greasemonkey script) and you're ready to go.
A few caveats. It's designed for Google web searches; I haven't tested it on links to Google images, maps, books, etc... so no guarantees. I just tested the bookmarklet on this thread but it might not work in your browser if you're not using one that I use. Take it as a starting point for experimentation.
posted by Jeff Howard at 2:32 PM on December 10, 2011 [1 favorite]
Out of curiosity, how hard would it be to automatically strip all that junk from Google links
Here's a way to automagically clean up the links after they're posted to Metafilter. For example, you could use it on this page to clean up the four example Google searches upthread. This works even if people don't change their posting behavior.
I don't use Greasemonkey, but I cobbled together a bookmarklet that scans all the links on any given web page, finds links that look like Google URLs, scans the query string for salient bits and then rewrites everything in place without all the googlecruft. No reason it couldn't be incorporated into the site or done server side, but this way allows it to work without any intervention from the powers-that-be.
Just install it as a bookmarklet (or, if you're industrious, convert it into a Greasemonkey script) and you're ready to go.
A few caveats. It's designed for Google web searches; I haven't tested it on links to Google images, maps, books, etc... so no guarantees. I just tested the bookmarklet on this thread but it might not work in your browser if you're not using one that I use. Take it as a starting point for experimentation.
posted by Jeff Howard at 2:32 PM on December 10, 2011 [1 favorite]
For those who missed it, Pronoiac's link to a bookmarklet takes care of this nicely.
I just now dragged the bookmarklet to the place just to the right where my link to Google is on my Chrome bookmark bar. (I prefer searching directly from Google.)
posted by Ivan Fyodorovich at 2:46 PM on December 10, 2011
I just now dragged the bookmarklet to the place just to the right where my link to Google is on my Chrome bookmark bar. (I prefer searching directly from Google.)
posted by Ivan Fyodorovich at 2:46 PM on December 10, 2011
Can I get a Greasemonkey script that eliminates all mentions of Greasemonkey Scripts ?
posted by sgt.serenity at 3:50 PM on December 10, 2011 [1 favorite]
posted by sgt.serenity at 3:50 PM on December 10, 2011 [1 favorite]
I asked a question about this recently, but the answers haven't worked particularly well for me. Maybe now that it is in MeTa something will surface.
posted by fake at 3:53 PM on December 10, 2011
posted by fake at 3:53 PM on December 10, 2011
Damn, desjardins beat me to it. LMGTFY is the clear answer or question that you're looking for.
posted by arcticseal at 4:39 PM on December 10, 2011
posted by arcticseal at 4:39 PM on December 10, 2011
Having to sift through BLARG to find the relevant 'q' parameter can be maddening.
Oh God. I'm going to start a charity for you at once. And, very possibly, a subreddit. Please write a distressed letter to The Times, also. This vileness is not to be borne.
posted by Decani at 4:39 PM on December 10, 2011 [2 favorites]
Oh God. I'm going to start a charity for you at once. And, very possibly, a subreddit. Please write a distressed letter to The Times, also. This vileness is not to be borne.
posted by Decani at 4:39 PM on December 10, 2011 [2 favorites]
stebulus: "I don't mean to speak for the OP, but let me see if I can explain the issue as I understand it."
So I posted this and then wound up away from the computer for the rest of the day. Stebulus has it more-or-less correct. I might add that this is only really a problem when the whole point of the link is the search term, as in this morning's graph plot examples.
I used a contrived example because I was trying to not to call out twoleftfeet or anyone else. I could have been a little clearer, I guess, and for that I apologize.
As far as automatically cleaning up the links, I think programmatically trying to find the 'q' parameter will run into problems when the poster has clicked on a suggestion as in the example. You wind up with two 'q' parameters and I'm not sure there's any way to tell the difference between them easily. That's why I thought that posters might be in a better position to clean out the URL, as they know why they were searching and why they wanted to link it on MetaFilter.
posted by ob1quixote at 6:09 PM on December 10, 2011
So I posted this and then wound up away from the computer for the rest of the day. Stebulus has it more-or-less correct. I might add that this is only really a problem when the whole point of the link is the search term, as in this morning's graph plot examples.
I used a contrived example because I was trying to not to call out twoleftfeet or anyone else. I could have been a little clearer, I guess, and for that I apologize.
As far as automatically cleaning up the links, I think programmatically trying to find the 'q' parameter will run into problems when the poster has clicked on a suggestion as in the example. You wind up with two 'q' parameters and I'm not sure there's any way to tell the difference between them easily. That's why I thought that posters might be in a better position to clean out the URL, as they know why they were searching and why they wanted to link it on MetaFilter.
posted by ob1quixote at 6:09 PM on December 10, 2011
stebulus: " What can you do if you agree that this is a problem and want to produce "clean" Google search URLs? Well, you have to learn a bit about URL syntax. You can mostly get by just following the template http://www.google.com/?q=kvetching, though you'll run into issues if your search terms contain spaces, or characters special in URL syntax."
Thank you for clarifying this issue in easier-to-understand terms for me. Until this moment I had never thought about it.
posted by zarq at 6:20 PM on December 10, 2011
Thank you for clarifying this issue in easier-to-understand terms for me. Until this moment I had never thought about it.
posted by zarq at 6:20 PM on December 10, 2011
"I asked a question about this recently, but the answers haven't worked particularly well for me. Maybe now that it is in MeTa something will surface."
Via Pronoiac's link to Matt Cutts's site, here's his explanation of what to do to get this functionality but only with your browser's built-in search text box. This will alter/create the/a Google search parameters for your browser ("create" in the case of Chrome because it won't let you change the search settings for its Google search entry) so that it always returns a search URL that is the bare-bones version you want.
I use Chrome, so I had to add a new search entry in "Options/Basics/Search/Manage Search Engines". I called it "Google Clean". You can use any keyword you want—I used "google-clean". And the query URL should be:
Then you'll want to set it as the default search engine when you search via the Chrome omnibox. (With Firefox or IE, it would be your browser's web search box.) Once you've done that, it should appear at the top, as your default search engine.
Here I'm about to do a test search from the ombibox; and here are the results. You can see the bare search URL in the address bar.
This only solves the problem when you search via your browser's built-in web search functionality. If you browse to Google Search directly, you'll still end up with the crufty, long URL.
To clean up one of those after you've searched via Google directly, then you'll want to use something like the bookmarklet that Pronoiac linked to above.
posted by Ivan Fyodorovich at 7:26 PM on December 10, 2011 [3 favorites]
Via Pronoiac's link to Matt Cutts's site, here's his explanation of what to do to get this functionality but only with your browser's built-in search text box. This will alter/create the/a Google search parameters for your browser ("create" in the case of Chrome because it won't let you change the search settings for its Google search entry) so that it always returns a search URL that is the bare-bones version you want.
I use Chrome, so I had to add a new search entry in "Options/Basics/Search/Manage Search Engines". I called it "Google Clean". You can use any keyword you want—I used "google-clean". And the query URL should be:
{google:baseURL}search?q=%sThen you'll want to set it as the default search engine when you search via the Chrome omnibox. (With Firefox or IE, it would be your browser's web search box.) Once you've done that, it should appear at the top, as your default search engine.
Here I'm about to do a test search from the ombibox; and here are the results. You can see the bare search URL in the address bar.
This only solves the problem when you search via your browser's built-in web search functionality. If you browse to Google Search directly, you'll still end up with the crufty, long URL.
To clean up one of those after you've searched via Google directly, then you'll want to use something like the bookmarklet that Pronoiac linked to above.
posted by Ivan Fyodorovich at 7:26 PM on December 10, 2011 [3 favorites]
though you'll run into issues if your search terms contain spaces, or characters special in URL syntax
Common symbols may be rendered in URLs as follows (i.e., as %##), starting with the space:
posted by Sys Rq at 7:41 PM on December 10, 2011 [4 favorites]
Common symbols may be rendered in URLs as follows (i.e., as %##), starting with the space:
%20 = %30 = 0 %40 = @ %50 = P %60 = ` %70 = p
%21 = ! %31 = 1 %41 = A %51 = Q %61 = a %71 = q
%22 = " %32 = 2 %42 = B %52 = R %62 = b %72 = r
%23 = # %33 = 3 %43 = C %53 = S %63 = c %73 = s
%24 = $ %34 = 4 %44 = D %54 = T %64 = d %74 = t
%25 = % %35 = 5 %45 = E %55 = U %65 = e %75 = u
%26 = & %36 = 6 %46 = F %56 = V %66 = f %76 = v
%27 = ' %37 = 7 %47 = G %57 = W %67 = g %77 = w
%28 = ( %38 = 8 %48 = H %58 = X %68 = h %78 = x
%29 = ) %39 = 9 %49 = I %59 = Y %69 = i %79 = y
%2A = * %3A = : %50 = J %5A = Z %6A = j %7A = z
%2B = + %3B = ; %51 = K %5B = [ %6B = k %7B = {
%2C = , %3C = > %52 = L %5C = \ %6C = l %7C = |
%2D = - %3D = = %53 = M %5D = ] %6D = m %7D = }
%2E = . %3E = < %54 = N %5E = ^ %6E = n %7E = ~
%2F = / %3F = ? %55 = O %5F = _ %6F = o
Occasionally you'll see %25##, the result of a URL being processed twice, with the original % getting converted into %25.posted by Sys Rq at 7:41 PM on December 10, 2011 [4 favorites]
(And, yes, there is a difference between this and this.)
posted by Sys Rq at 7:54 PM on December 10, 2011
posted by Sys Rq at 7:54 PM on December 10, 2011
stebulus: You can mostly get by just following the template http://www.google.com/?q=kvetching
Fact-checking fail. The (or a) right template is
Fact-checking fail. The (or a) right template is
http://www.google.com/search?q=kvetchingThis is BREAKING MY MIND.
posted by two lights above the sea at 9:25 PM on December 10, 2011
posted by two lights above the sea at 9:25 PM on December 10, 2011
My pet peeve is the complete inability to pick up clean links for PDFs from a Google search - I'm having to cut and paste into notepad and then physically clean them up before pasting grar grar grar... I have a vague memory of back in the day when the URL would show up in the browser bar even if the PDF was open in Adobe but perhaps I'm wrong.
Would appreciate any pointers for this specific problem, thanks
posted by infini at 9:50 PM on December 10, 2011
Would appreciate any pointers for this specific problem, thanks
posted by infini at 9:50 PM on December 10, 2011
Jeff Howard would your bookmarklet help me with the PDF url problem ?
posted by infini at 9:52 PM on December 10, 2011
posted by infini at 9:52 PM on December 10, 2011
Also, one could use duckduckgo search results. Here's the link for search "Matt Haughey" on DDG:
Sweet.
posted by taz (staff) at 11:05 PM on December 10, 2011 [5 favorites]
http://duckduckgo.com/?q=Matt+Haughey.Sweet.
posted by taz (staff) at 11:05 PM on December 10, 2011 [5 favorites]
DDG is pretty nice actually. I've been using it exclusively for the past few weeks - I get the feeling it's not as comprehensive or as fast as google, but it's got a great signal to noise ratio.
posted by Dr Dracator at 5:08 AM on December 11, 2011
posted by Dr Dracator at 5:08 AM on December 11, 2011
Google Tracking-B-Gone is a Greasemonkey script that stops Google click tracking, so you can copy useful links from the search page again.
This is what I use, and I'm a big fan.
And while we're complaining about stupid link cruft, I don't like the youtu.be shortener either. (And I don't do inline videos here, but my understanding is that it breaks them.)
posted by inigo2 at 8:46 AM on December 11, 2011
This is what I use, and I'm a big fan.
And while we're complaining about stupid link cruft, I don't like the youtu.be shortener either. (And I don't do inline videos here, but my understanding is that it breaks them.)
posted by inigo2 at 8:46 AM on December 11, 2011
infini: My pet peeve is the complete inability to pick up clean links for PDFs from a Google search...
Try these Greasemonkey extensions:
* For Firefox, Google Link Cleanup.
* For Chrome, Google Tracking-B-Gone.
posted by Pronoiac at 2:21 PM on December 11, 2011 [2 favorites]
Try these Greasemonkey extensions:
* For Firefox, Google Link Cleanup.
* For Chrome, Google Tracking-B-Gone.
posted by Pronoiac at 2:21 PM on December 11, 2011 [2 favorites]
I don't like the youtu.be shortener either. (And I don't do inline videos here, but my understanding is that it breaks them.)
I believe it did break things for a bit when it was first introduced, but pb has since updated things to work with shortened youtu.be urls. Those ones don't bother me too much, as they have all the same info in them asa regular Youtube URL. Remembering that http://www.youtube.com/watch?v=unreadable-hash is equivalent to http://youtu.be/unreadable-hash is one more thing to remember (or look for in your code), but it's still clear that the link is going to a youtube video, and it's no more unclear which video you're going to.
posted by JiBB at 5:39 PM on December 11, 2011
I believe it did break things for a bit when it was first introduced, but pb has since updated things to work with shortened youtu.be urls. Those ones don't bother me too much, as they have all the same info in them asa regular Youtube URL. Remembering that http://www.youtube.com/watch?v=unreadable-hash is equivalent to http://youtu.be/unreadable-hash is one more thing to remember (or look for in your code), but it's still clear that the link is going to a youtube video, and it's no more unclear which video you're going to.
posted by JiBB at 5:39 PM on December 11, 2011
would your bookmarklet help me with the PDF url problem ?
No, the bookmarklet just cleans up links to Google searches that people have posted elsewhere (like in this thread). It doesn't change Google's behavior onsite for PDFs or for regular search results. Pronoiac's links look like a good bet.
posted by Jeff Howard at 10:08 PM on December 11, 2011
No, the bookmarklet just cleans up links to Google searches that people have posted elsewhere (like in this thread). It doesn't change Google's behavior onsite for PDFs or for regular search results. Pronoiac's links look like a good bet.
posted by Jeff Howard at 10:08 PM on December 11, 2011
The suggested alternatives aren't great: lmgtfy is as abrasive as ftfy, and the equation graphing that sorta spawned this thread simply doesn't work in duckduckgo.
posted by Pronoiac at 10:50 PM on December 11, 2011
posted by Pronoiac at 10:50 PM on December 11, 2011
I am a n00b when it comes to Greasemonkey - I clicked on Install (i do use Firefox and even disabled adblocker on that page) and I just got a page with the actual script progam - nothing else happened. What do I do next?
posted by infini at 11:58 PM on December 11, 2011
posted by infini at 11:58 PM on December 11, 2011
Oh God. I'm going to start a charity for you at once. And, very possibly, a subreddit. Please write a distressed letter to The Times, also. This vileness is not to be borne.
posted by Decani at 12:39 AM on December 11 [+] [!]
Dude, it's not a huge problem, which is why he brought it up on a no impact messageboard, not at the UN. It is allowed to talk about small things that bug us here.
posted by Catfry at 3:19 AM on December 12, 2011
posted by Decani at 12:39 AM on December 11 [+] [!]
Dude, it's not a huge problem, which is why he brought it up on a no impact messageboard, not at the UN. It is allowed to talk about small things that bug us here.
posted by Catfry at 3:19 AM on December 12, 2011
Now that you've suggested this, please do not get annoyed when people ignore it.
posted by smackfu at 6:24 AM on December 12, 2011
posted by smackfu at 6:24 AM on December 12, 2011
If this isn't a First World problem, I don't know what is. Click it and move on.
posted by MexicanYenta at 6:28 AM on December 12, 2011
posted by MexicanYenta at 6:28 AM on December 12, 2011
Those of you who think this is a silly thing to care about: Again, not speaking for the OP, but I get a bit annoyed about things like this because it feels like I'm being stymied in applying my reading skills. It's like if you've learned how to skim text by reading, say, the first and last sentences of each paragraph, and then you encounter text with no paragraph breaks at all, or with paragraph breaks in weird places — the writer's ignorance of the conventions of paragraph structure has made your skill useless, slowing you down, preventing you from realizing the rightful benefits of learning the skill.
Doesn't that seem frustrating to you? Have you nothing like that in your life?
MexicanYenta: First World problem
How about this? I'll agree to feel ashamed for caring about the technical nuances of communication in online media if you'll agree to feel ashamed when your ignorance and/or incompetence inconveniences other people. Deal?
posted by stebulus at 8:58 AM on December 12, 2011 [5 favorites]
Doesn't that seem frustrating to you? Have you nothing like that in your life?
MexicanYenta: First World problem
How about this? I'll agree to feel ashamed for caring about the technical nuances of communication in online media if you'll agree to feel ashamed when your ignorance and/or incompetence inconveniences other people. Deal?
posted by stebulus at 8:58 AM on December 12, 2011 [5 favorites]
Or you could, you know, click it and see what it is.
posted by MexicanYenta at 10:07 AM on December 12, 2011
posted by MexicanYenta at 10:07 AM on December 12, 2011
infini: I clicked on Install ... and I just got a page with the actual script progam - nothing else happened. What do I do next?
Huh, that's happened to me too, I think before I reinstalled Greasemonkey. I think what helped was:
* making sure Greasemonkey was enabled.
* reloading the displayed script.
* clicking back, then clicking "install" again.
posted by Pronoiac at 10:21 AM on December 12, 2011
Huh, that's happened to me too, I think before I reinstalled Greasemonkey. I think what helped was:
* making sure Greasemonkey was enabled.
* reloading the displayed script.
* clicking back, then clicking "install" again.
posted by Pronoiac at 10:21 AM on December 12, 2011
* making sure Greasemonkey was enabled.
clarify this a little more please
posted by infini at 10:30 AM on December 12, 2011
clarify this a little more please
posted by infini at 10:30 AM on December 12, 2011
infini, if you've installed Greasemonkey properly, there will be a happy little monkey head in the browser. In Firefox 8, it sits to the right of the location bar. You can click it to disable Greasemonkey, or set options.
posted by Pronoiac at 10:37 AM on December 12, 2011
posted by Pronoiac at 10:37 AM on December 12, 2011
Ok, so what I am understanding is that I have to install greasemonkey itself first, and then install an app within it? I did not know this
posted by infini at 11:05 AM on December 12, 2011
posted by infini at 11:05 AM on December 12, 2011
Greasemonkey is an add on. It's like a little script-running engine that is attached to Firefox. So once Greasemonkey is installed, you can run little scripts in it which can be used to affect the way web pages are displayed. These scripts are found in places like userscripts.org [here are scripts for metafilter] and are installed by clicking on a link which will prompt you to click a button to install the script. The scripts are text files and you can inspect them by going to the "manage user scripts" option within Greasemonkey. A lot of scripts are just functional on one site. I have one, for example that changes the Post/Preview buttons into a unicorn an a narwhal respectively. It's sort of great. These scripts work with Firefox, as far as I know. I think there are ways to run them within Chrome, but I know less about that.
posted by jessamyn (staff) at 11:11 AM on December 12, 2011 [1 favorite]
posted by jessamyn (staff) at 11:11 AM on December 12, 2011 [1 favorite]
Or you could, you know, click it and see what it is.
There are a few conflicting schools of thought in MeFi.
posted by zamboni at 11:38 AM on December 12, 2011
There are a few conflicting schools of thought in MeFi.
- If you don't know what something is, why don't you click on it?
- If you don't like something, why did you click on it?
posted by zamboni at 11:38 AM on December 12, 2011
Huh. I usually scrape out the extra parameters when I linked to google, mainly to remove any personal information that might be there. I don't really see why I should care if anyone else does, it whatever is in the URL doesn't really affect me.
posted by delmoi at 8:03 PM on December 12, 2011
posted by delmoi at 8:03 PM on December 12, 2011
empath: "Why not use lmgtfy links?"
Because I used it once on a girl who was a problem user on my forum (always asking "what is that?" 5 posts in a row when she could have easily Googled it) and she got all offended and threw a fit about how rude I was. Bound to happen again. - more people all offended. (And yes, I am speaking of LMGTFY and not JFGI.)
In addition to Google Tracking-B-Gone, the script UTM Stripper will help clean up links from other sites. I think Google Reader uses UTM so it would help with links copied from there, at least.
posted by IndigoRain at 12:27 AM on December 13, 2011
Because I used it once on a girl who was a problem user on my forum (always asking "what is that?" 5 posts in a row when she could have easily Googled it) and she got all offended and threw a fit about how rude I was. Bound to happen again. - more people all offended. (And yes, I am speaking of LMGTFY and not JFGI.)
In addition to Google Tracking-B-Gone, the script UTM Stripper will help clean up links from other sites. I think Google Reader uses UTM so it would help with links copied from there, at least.
posted by IndigoRain at 12:27 AM on December 13, 2011
I hate UTM with a vengeance and endeavour to strip it from everything that I can, within reasonable oversight.
Thank you for the very helpful overview, Jessamyn, on greasemonkey and its attendant utilities.
posted by infini at 12:57 AM on December 13, 2011
Thank you for the very helpful overview, Jessamyn, on greasemonkey and its attendant utilities.
posted by infini at 12:57 AM on December 13, 2011
LMGTFY is meant to be patronizing, right? I can't imagine anyone using it to be helpful.
posted by smackfu at 5:16 AM on December 13, 2011
posted by smackfu at 5:16 AM on December 13, 2011
You are not logged in, either login or create an account to post comments
posted by jessamyn (staff) at 8:44 AM on December 10, 2011 [3 favorites]