Home // Posts tagged "google"

Proper Search – From Google, Even

I’ve blithered before about Google sodding up search, personalising it and “correcting” your input and so forth. I just want to enter search terms, and have a search engine search using those terms. Call me weird, but when I type something into a search box, it’s because I want to search for that word or phrase. I don’t want my search engine making what Google calls:

“…normal improvements such as

  • making automatic spelling corrections
  • personalizing your search by using information such as sites you’ve visited before
  • including synonyms of your search terms (matching “car” when you search [automotive])
  • finding results that match similar terms to those in your query (finding results related to “floral delivery” when you search [flower shops])
  • searching for words with the same stem like “running” when you’ve typed [run]
  • making some of your terms optional, like “circa” in [the scarecrow circa 1963]“

“Improvements”…

Deciding that some of what I type in is “optional”…

Basing the results returned on sites I’ve visited before…

Automatically “correcting” my spelling – from the obscure/technical word I wanted, to some meaningless and out of context foolishness…

“Improvements.” Ok…

20111117-212133.jpg

Anyway, it’s not been just me complaining … well, maybe it has – until recently, when Google did away with the + operator, which previously could be added in front of a word to ensure it wasn’t “corrected”/modified – though that still left all the other “improvements”. Now Google has introduced what it is calling verbatim search, in which none of those “improvements” are made; it simply searches for the terms you type in, and returns the results.

Thank you, Google.

Oh, but … hold on a cotton-pickin’ minute. You can’t set it as default? You have to search, then when the results come up, click on more options or something, and navigate through to “verbatim search”? Sodding pleh.

So. I don’t like that very much. But it turns out I don’t have to. There is a little string, a few tiny characters, that when added to the end of your search string, can make all the difference:

&tbs=li:1

That, my friends, is all you need for searching win.

Update: I’m pretty sure I found that string at Ghacks; I also note that there’s now a userscript called Gooverbatim that adds a “yes, really” button to Google – which obviously performs a verbatim search.

Now, in Chrome, you can set custom search engines. I created one with this string at the end (it also includes another few characters that I added ages ago I think to stop personalised results; I can’t remember … I left them in anyway, though they might be redundant):

http://www.google.com.au/search?q=@search@&pws=0&hl=all&tbs=li:1

Of course you would make it whatever country’s google URL you’re in (or another if you want, I suppose). Set that as the default search engine, and then when you type something into the address/search bar, you’ll get a verbatim search by default.

So. That’s Chrome sorted. What about my other browsers? Atomic on the iPad is easy: custom search engine with that URL: done. Mobile Safari though doesn’t allow custom search engines. Enter Quix.

It can however, be even more betterer. By adding a line to the quix command list (I also deleted a bunch of commands I would have no need for – you can do all that with a simple text editor) you can have it search without having to type ‘g’ before your search terms. The line you need (again, change the country as you wish) is:

* http://www.google.com.au/search?q=@search@&pws=0&hl=all&tbs=li:1 Untriggered search

I would suggest, as well as adding this line, adding &tbs=li:1 to the end of the default search string for google search in the quix command list; that way you still get a verbatim search even if for whatever reason you use the quix command stem (“g”) before your search terms.

You modify the quix command list, upload it (you will of course need somewhere to upload it to), then – in Safari on your computer, not the iPad – go to the Quix extend page, and from there drag the bookmarklet to your browser bar. Synchronise with your iPad (either through iCloud or via iTunes), making sure the info from the computer goes to the iPad (you might need to tick that option in the iTunes “info” pane for your iThing (just realised the iPad isn’t the only iThing with a browser… Come to think of it, it’s not limited to iThings; this is just a JavaScript bookmarklet, so all of my wittering applies to any platform really), and you should have a nice shiny “Quix” bookmarklet in your bookmark bar. Click that, type your search into the box that pops up, and go. Just as quick as the default search box, and proper unmodified search.

I is happy.

Autocorrect: An Informational Plague

I’ve been running afoul of autocorrect, and it really rather bothers me. On the one hand there’s the trouble I have with my iDevices: no matter how many times I cancel their suggestions, they won’t stop trying to “correct” hell to he’ll or its to it’s, or various other mild annoyances. Most irritatingly is their habit of thinking the shortened version of my name is an acronym, and capitalising it all (every single email or message I sign off … It’s got a bit old by now). That last sentence also reminded me of one of the most irritating “corrections”: -ising becomes -izing, unless I watch very closely indeed. I have it set to Australian English – surely the Aussies spell correctly … or has this country gone all US that way as well? I might try changing to UK (he muses thoughtfully).

Anyway, that’s one thing, and in the scheme of things it’s a pretty minor annoyance really. What really grinds my gears ;) is autocorrect by a search engine. I noticed it first with Google, because that’s the one I use, but then found that Bing and Wolfram Alpha do exactly the same thing: if you type in a search term the engine thinks is misspelled or in some other way wrong or unlikely, it “corrects” it for you. It might be helpful if it simply asked “did you mean…” and provided a link to search for the “corrected” terms, but noooooooo: it searches for what it thinks you meant, and if you want to search for what you typed originally, you have to click again.

That might be fine if you’re making some lowest common denominator search (for lady gaga or ke$ha or whatever you crazy kids are into nowadays) and your spelling is poor: then it’s probably helpful. If however you are searching for something less than ordinary, and your spelling is fine, you are penalised, as you realise that the displayed results are not even vaguely related to your intended subject, and have to click the link for your original search.

I wouldn’t even care about the default behaviour if I simply had the option to turn the auto-”correct” off. I don’t. We’re locked-in to this. It’s really quite frustrating. I can’t see google changing it in any way either, as there has already been angry discussion about it online, including on google’s forums, and they seem peculiarly insulated against what their users want – cf their messing about with YouTube.

It doesn’t seem to me an excessive desire: to have my search results be an unbiased reflection of the (indexed) information on the Internet that relates to the search terms I typed in.

Is it?

YouTube Plays Both Sides of Copyright

I’m annoyed. I’m annoyed with YouTube, and I’m annoyed with Warner-Chappell, and I’m annoyed with a music industry made up of frakking corporate troglodytes.

A while back I got a message from YouTube telling me that a couple of my videos had been flagged by their clever-dick system as containing copyright material. On those two videos advertisements appeared. Now while I don’t like having ads on my videos (because I know I hate them when I’m watching someone else’s video with ads), I realise that was fair enough: the copyright holder wants a bit of ad revenue from the few hundred (maybe) people watching me play a song to which they hold the copyright, so they can put ads on the video. Fine. No worries.

Last night I noticed that at some point between then and now, Warner Chappell, the copyright holder for Hotel California, by The Eagles, has muted the audio to my video of me playing the solo from Hotel California.

So I look a right twat in that video. ;)

My first thought was to dispute it. After all, it’s a cover, and one can play covers, right?

… Riiiiiiight??

Turns out the answer to that is no. I’m not the first person to wonder about this, and some have asked the queyion online. Answer is that covers are not “fair use” under copyright law, and therefore remain subject to copyright requirements. What about cover bands??!! They’re playing in a bar or a club which pays licensing fees to ASCAP (I think), the copyright administering jobby, so composers get paid when their songs are played. All good, don’t get me wrong. They should indeed get those royalties.

However, all the aforegoing leads inescapably to one conclusion: Google/YouTube are a bunch of bloody stingey-arse cheapskates who are avoiding paying licensing fees to ASCAP. Consider: if they paid said licensing fees, we’d be able to play covers, and composers/copyright holders would get a share of any resulting revenue. We’re not allowed to play covers, ergo, YouTube do not pay a licensing fee.

Thy know damn well there are a lot of covers on YouTube. I figure they’re hoping that a few take-downs and mutings will keep the copyright holders happy, while they sit quietly back enjoying the fruits of having umpty-million cover videos (not to mention actual songs with slideshows) attracting lots of views.

So that’s why I’m annoyed with Google/YouTube. What about Warner-Chappell? Yes they’re the copyright holder, and yes they were perfectly entitled to put ads on my video. Hey, legally they were entitled to mute it as they have as well. The point is, first they tried the ads. Because I’m small fry with few views, they probably made no money from it. Instead of thinking “oh well that’s completely irrelevant then. We’ll leave the ads there in case, though”, they decided to mute it. Out of what? Spite? What do they gain? In a material sense what do I lose? They would get any ad revenue, not me. It’s just annoying and frustrating; no-one’s any better off.

Bloody troglodytes.

Especially Google/YouTube, who are playing both sides: denying copyright holders their proper revenue by not paying licensing fees (i admit this is an assumption, but it follows logically: if they do pay licensing fees, surely they would not have to remove or mute videos), and they crap on users like me who just want to share and interact with others playing songs we love.

Troglodytes all.

Copyright Clever-dicks

I’m faintly put out. Google have a new clever-dick system on YouTube that analyses your videos and compares them to reference recordings to try to determine whether you’ve uploaded something copyright. Now, all I upload are videos of me playing guitar. Some of those are cover versions, true, but I’m certainly not uploading other people’s music videos or songs or whatever.

Imagine then my annoyance when into my inbox popped a notice that my latest video might be copyrighted material O_o And then another. Further annoyance followed when they stuck popup ads onto those two videos :|

I suppose that google has to pay royalties to the songwriters’ wossname – just like a radio station or a bar or something. And I suppose it’s fair enough they would want then to get a bit of money from videos for which they might have to pay said royalties. But I hate having ads on my videos. :|

So that’s really the point of this little ramble: the ads aren’t my fault, and I do apologise dreadfully for their presence. ;)

The Destruction of Search

Google seem hell-bent on destroying not search (duh) but the validity of your search results. First they came out with personalised search. This ridiculous idea means that the results you get when you search are filtered, based on the sort of sites you usually browse (I would imagine only those sites you browse to from Google search results, but I’m not too sure). Now they appear to be planning another great leap into crap with Google Me, their second attempt at a social network (because Google Buzz was such a spec-tac-ular success).

Why do I think these are bad ideas though? Quite simply, because you no longer get unbiased information when you search the internet with Google – unless you’ve disabled personalised search. (Note: I was unable to do this, as it doesn’t seem to have my web history anywehere. I’m hoping that really means they don’t have it, rather than it’s hidden away and I can’t turn it off.) Anyway, that will bias the results returned by your search, in favour of the kinds of information you have looked at previously. In other words, taking an extreme view, Google will not show you information different from what you already have. That’s a problem.

Soon it seems we might have Google Me, a social layer on top of existing Google services. Apparently. What do Google say about that? Here’s a statement from Google CEO Eric Schmidt that pretty well encapsulates my horror:

With your permission, knowing more about who your friends are, we can provide more tailored recommendations. Search quality can get better.

Not search results. Not information. “Tailored recommendations”. Based not just on your web-browsing history, but those of your friends. Based on your facebook likes and fan pages – and those of your friends. Yuk. Yuk yuk yukkity yuk.

Basically this amounts to filtering and biasing the information we see. As I said in my previous post, in an age defined by information, the nature of the information to which people have access is a crucial issue. I do not think it is Google’s role to filter that information for us. I am concerned that this could adversely affect people’s already crap ability to think critically, as every search will confirm their biases – and every search will further “tailor” – i.e. restrict – the results given next time, and so forth.

I really might have to check out some other search engines – though I fear they won’t actually have the vast reach of Google. It’s a bloody bind: Use a smaller search engine and likely not be able to access all the information out there; use Google, and be denied access to that information which isn’t the same as what you and your friends already think.

To Google has become synonymous with Internet searching, but I think the real meaning is starting to creep; perhaps to Google might come to mean reinforcing one’s biases, cherry-picking evidence, and spreading misinformation.

Statistical data collected by Statpress SEOlution (blogcraft).