I hope their search quality improves. I understand it's an arms race with spammers, and they do give a lot of value for free minus ads. I guess I'm so attuned to identifying bad results by now that I haven't noticed much variability in results quality. But echo-chamber curation doesn't feel right, and the feel is important.
I miss literal search, and then they removed +keyword - but fortunately, quotes do it now. For example, rip and "rip" give slightly different results, with different emphasis within them: https://www.google.com/search?q=rip https://www.google.com/search?q=%22rip%22
Google's predictive search while you type is amazing, especially because it is so fast - you can select the search you want. It's tab-completion for the web. And it can be informative in itself, if you aren't sure of the right terms. (i.e. the deep search problem of not knowing what you are looking for).
I think they are doing a great job of searching for what you want - which is their goal. Technically precise searches aren't their main goal (though I wish they retained it somewhere).
However... artificially ranking their own links, ranking that isn't objective, is deadly to the utility of their search, and to their brand. In the past, google products have seemed to rank highly, but google claimed this was their natural rank. What is the evidence against them now? EDIT ok, the evidence is explicit and damning: http://www.focusontheuser.org/examples.php EDIT2 ok, I'm not getting the results they claim, at all. I'm using "google.com" but from Australia. Maybe that's why.
For example, I'm sure Google "knows" that I am a monolingual English speaker, but I can read enough German, French, and Spanish to struggle through a blog post telling me how to diagnose or work around a problem with my computer when that's my only choice. (Especially now that Chrome will run the whole thing through Google Translate for me.) I set an option in Google to show me pages in other languages, but these days I don't see non-English results anymore unless there are only a few pages of search results. It seems like non-English pages are penalized heavily, regardless of how high they rank as search results. (Or did everyone who used to post about technical topics in German switch to English over the last five years?)
Even if that's not true, the fact that I even think about it is a bother. I wish Google would be more simple-minded so I could have more confidence in how I understand the results.
Try setting German at "For Google text" (first on language preference page). I did, and got several german hits. Whereas, merely adding "german" for the search results (second on language preferences page) didn't add any. It's a hassle, but a solution is to change your "For Google text" setting. for each language you want.
It seems to me that the vastly greater number of english results would simply out-weigh the german resuls. The german results will still be there, but ranked much lower.
What's curious to me is that you say Google didn't formerly do this. Maybe they changed the weighting of the various languages from being equal to being weighted according to the results.
Probably searching with german terms will get german results, but again a hassle because you have to do it for each language.
EDIT following guelo's comment on "verbatim" under "All Results -> More Search Tools" on the left hand side, there's also a "Translated foreign pages" option. This gave me several results in other languages. Sounds like what you want.
BONUS: I noticed a "Search English and German pages" just under "Search the web" on the left hand side (I added German as a search language above). I think this is precisely what you wanted.
Google is impressing me.
Really? All that to get what I used to get for free? Not to mention that German is just one language, and "translated foreign pages" sounds like it might search the translated text, which is nice, unless it doesn't deliver results in untranslated pages, which leaves me wondering whether Google translates all pages or just some, and whether it's sufficiently integrated with Google Translate to filter out words that can't be translated, which are often exactly the words I want. That puts me right back where I started, sitting in front of Google wondering how the hell it works and what it's filtering out of the internet as a result.
Google aims at working without thought. But it can only do so for one target audience, and unfortunately for technically minded folk, that target has to be mainstream users. Fortunately, technical folk are usually curious to explore and customize technology.
EDIT just confirmed that it does use added search languages, as expected. e.g. if I add german and french, it provides: "Search English and French and German pages" Note: I'm not doing this for you; but because I'm curious.
Honestly, you shouldn't be so sensitive that you perceive criticism of a web app as criticism of your advice, or assume that I didn't check up on it. I just wasn't happy about what I found. Per my first post, all I want is to be able to use Google and understand roughly what it's doing, and it failed very badly in this case, since the one thing I was counting on it to do -- include foreign language pages, because I asked it to -- was actually the thing it wasn't doing.
You can add more than one language. If you'd checked up on it, you would have seen that. I perceive this as complaining about advice without following it, and I find it annoying.
If you're still angry (and it sounds legitimate, especially if there was a subtle hint of a problem, but not enough to register fully), that explains where you're coming from, which puzzled me.
Aside: Unfortunately with webpages/webapps, they can and do get updated continuously, without explicit version changes or any notification at all. Ideally, all changes would be back-compatible, but then you end up with a mess (both in user interface and implementation). Continuous notifications amounts to spam, and is usually ignored (I do). Change is a problem. I'm not sure there's a solution.
I think back-compatibility is a deep, abiding, practical problem, in general.
EDIT "if the choices I made have been superseded and contradicted": just realized this is more than merely "backcompatibility of UI". In effect, they have destroyed your data. Although that sounds overdramatic, since the data itself is very minor (perhaps one bit, a binary checkbox), it's a clear trigger for notification. There might be some grey area for when your configuration is still there, but has a different meaning - but I think you're saying that your config is no longer there, and/or no longer has an effect.
I used to use Yahoo finance, and they had a nice display of 9 user-configured stocks at the top. In their latest redesign, this disappeared, and I couldn't find an option to bring it back. This was enough to make me spurn their product, and switch to google finance. It makes me think they have no idea what they are doing. I guess in this case, I had notification (as I could see it was gone). But if they had an explicit reference to it going away, even if it said "sorry, this is going away", I would feel better about it. And I have a sneaking suspicion that maybe it is still available somewhere, and I just couldn't find it.
So, changing the meaning of a user-specified configuration should probably trigger an explicit notification (and force the implementer to consider it); and deleting user-specified configuration should never happen, even if it has no effect. I think it should be treated as sacrosanct as other data stored (e.g. gmail emails, dropbox files etc). If changes render it ineffective, it should still be there, but with a notice: "this no longer works, use this instead". Again, this forces the implementer to thin about what they are doing.
Re the "just one language" thing, I can imagine I'm not making sense, because I'm not really communicating the context of how irrationally enraged I am over this. The truth is that I only enabled a couple of languages because I was just too angry and frustrated to enable the rest. I know it doesn't make any rational sense, but my reptile brain is telling me that goddammit, Google is not going to make me click on fifty little boxes just to fix them erasing my preferences. Not today. Not this user. There is a part of the brain that is trained to recognize when someone is jerking you around or abusing you, and when it gets triggered inappropriately, it is impossible to turn off. (I'm sure anyone who has worked in customer service will recognize what I'm talking about. Right now I am the bug-eyed paranoid who is convinced they are intentionally screwing with me for their own amusement.) So long story short, yes, it seems immensely important to me that I have to check fifty boxes instead of one, and I still don't have my unfiltered search results, not until I calm down enough to go click on all those boxes.
As for the problem of change, I can sympathize with the problem they face. I'm sure that globally minimizing user confusion and frustration is one of their goals. As for me personally, they'll have to count me in the "used the product incorrectly for years without realizing it and is now very angry about it" category.
UPDATE: EVEN ANGRIER:
So I swallowed my pride and went to work clicking.
"The page at https://www.google.com says:
The number of language restrictions is limited to 8, so only 8 of the languages selected will be saved."
That doesn't bode well. But then: "The page at https://www.google.com says:
Your preferences have been saved"
Ah, but guess which time it was lying :-/
So unless there's another way to turn off language filtering, Google will filter out every language besides your primary language and seven others. (Yeah, your interface language counts as one of the eight.)
That's terrible. Even without Google Translate, I swear I got useful information from a page in Polish once. Plus there are pictures, video, miscategorized pages, mixed-language pages (what happens with those?) and the simple coolness of finding out that someone in Vietnam is using an open-source library you committed a patch to. And of course, Google freakin' Translate! Unless Google is translating every single page it crawls. Which it might be -- that's something to think about. But it does very poorly with some pairs of languages, and if it filters out every term it can't translate, then it will tend to filter out commands and symbols. And here I am speculating about how it works again, which goes back to my original point.
With my track record today, I'm sure I'll find out I'm misinterpreting this somehow, but I'm starting to feel a little bit justified in my anger. Ah, I can't lie. I feel much less angry now that I feel like less of a fool :-)
I'm with you on the inappropriately triggered jerking-me-round neuron. I am definitely there from time to time :-). You put it well. When in that state, I don't really believe that someone in the real world is messing with me - but inside my world, I believe someone is. I think that I can't quite see the real world in that state (or don't want to). Funny thing is, if I find out someone is messing with me, I also feel less angry...
Oh man, that's awful. And that "limited to 8" dialog pops up for each checkbox. Seems a really badly designed interface in general. I wonder why the limit? Maybe they keep languages separate, because effective pagerank criteria differ across languages/cultures; and they're limiting internal bandwidth needed to combine the results from distant databases.
I was going to say you could just include the languages you can read (and where you often found useful info before); but I appreciate your point about seeing the whole web (I guess, the www), across languages.
I've had a play with google translate, and I noted that untranslated words seem to pass through, unaltered (using the text-based one, eg: http://translate.google.com/?hl=en&tab=wT#auto|en|Je%20n... ) You could probably see by trying example pages til you find a word it doesn't know. It seems much more useful to include an untranslated word than to omit it, so hopefully they do it that way.
Given the 8 limit, this seems the best option for seeing the whole web - and if the translation is any good at all, it might be better. I might try this (though I recall reading something that I didn't notice was auto-translated, and after reading for a while, suddenly becoming aware that I had no idea what it meant...)
I wonder if just plain more information about a situation reduces anger, because anger relies on one's internal view of the world? The more acquainted we get with the real world, the harder it is to sustain anger. (Aside: The Buddhists say all negative emotion is based on ignorance or misconception - of seeing something as 100% absolutely negative - and so negative emotion can be reduced by deeper understanding.)
I think the notion of objectivity in search is a fallacy, and I would hesitate to characterize the addition of social search as artificial inflation, at least not any more artificial than any other part of the algorithm. The fact is: Google has more information about how to rank G+ profiles than any other profiles. That extra information adds google juice. Yes, other social networking profiles are indexed and ranked, but Google doesn't have the appropriate extra information on them that it does wrt G+.
This doesn't seem particularly hard to understand, and I'm not sure how the focusontheuser.org bookmarklet illuminates the issue. I agree that the G+ profiles are probably given too much weight, and that's probably for two reasons: 1. It's brand new and requires more data for calibration; 2. It props up G+ as a product.
From Google's perspective, even if search quality is negatively impacted in the short term while they calibrate and funnel users into G+, as long as in the long term search quality is bolstered by the additional subsequent data, it's a win.
Artificially inflating that, for the purpose of pushing their product, is the problem (that's how I'd define "artificial ranking" - ranking on a basis other than returning a page of search results for the user). This avoids the problematic term "objective".
I don't buy the "means to an end" argument. If google really is doing that, it means worse search results, loss of user confidence, and antitrust litigation (abuse of market power).
BTW: the first search example "people and places" seems to require a google+ account (as it wouldn't give me results unless I signed up). I'm happy for a google+ search to yield google+ results.
It's like saying Facebook is also "abusing its market power" because it only shows you contacts on Facebook and not on Google+. Sorry, but this cannot work this way, when you are on one network, you agree with the principles of that network, or you use another. There's no lack of options out there.
Your point about alternative search engines, and Windows being hard to avoid, is relevant for another kind of antitrust.
Aside: though, I think google is still far and away the best, in terms of response time (especially for predicting search queries as you type - that requires huge local data centres all over the world). e.g. Duck duck go has great features, but it's just too slow (at least, over here in Australia). Who can afford data-centres everywhere? (maybe Amazon...) Google also has one of the strongest brands in the world. So, the choice is kind of figurative. It's an earned monopoly, not earned through abuse, not bad or wrong; but a monopoly nonetheless.
I think the term "abuse of market power" has very thin legal ground. Anything that becomes popular tend to appear like it "abuses" its power. That's usually what competitors say when they simply give up competiting: they use lawyers and politicians to weaken the leader.
What matters in the end is what value the service provides, and if there is choice available for users to turn to other alternatives. If Google does not provide value anymore (irrelevant results) it will consumers/users in the end, and that will spark demand for better alternatives. The market will adjust. I do not really see an issue with the current situation.
Regarding search engines, Bing and Yahoo may not be as relevant but they also do the job most of the time. There will be new competitors coming, web contents are changing, the way people use the web are changing, all this environment creates opportunities to challenge search on the way. Looking through pages of results of links is hardly something that can be optimal over the long term - we will need better tools, I have no doubt about that.
Yes, change may bring new competitors that could upset google - disruption seems the destiny of most tech companies, it's just a matter of time. Interestingly, what we're seeing right now is a valiant effort on Google's part to shift from pagerank to "social-rank", in response to facebook, a new competitor. Going by history (as in "The Innovator's Dilemma", google as the incumbent will succeed if the technological innovation serves their business model - which it does.
If you're not seeing the Google+ results on some pages, it's just because Google is still in the process of rolling out the feature.
There is the 'Verbatim' search option hidden on the SERP page on the left hand side under All Results -> More Search Tools -> Verbatim
The one thing that annoys me most these days, however, is that it also searches for "bike" when I type "motorbike"; I get endless irrelevant results about cycling in many longer queries. I recall other search engines in the past implementing cluster-based analysis to disambiguate homographs (with user selection on the SERP), but Google still isn't clever enough to fully figure out how "bike" means something different on motorbike-related pages than on cycling-related pages.
You could use the form at - https://www.google.com/quality_form?q=google+search+quality&...
I doubt Amit Singhal would object to receiving a specific report about bad search results.
See, for example: http://news.ycombinator.com/item?id=3277867
I don't work on search quality, but I can pretty much guarantee that if you send specific examples to the search quality team, either through the feedback form, or some other means - heck send it to Matt Cutts - it will be looked at and used to improve search quality.
The more specific you can be, the better.
"Searches for bikes suck for me" isn't very useful :-)
"When I enter the query 'blah blah blah', the following urls (url1, url2, url3, ...) are completely irrelevant to me." Even better, say why you think the urls are irrelevant.
If you submit reports like this, I can promise you that the search quality team will take them seriously and will praise your name (if possibly only internally :-) )
https://www.google.com/search?q=low+tire+pressure+motorbike
Lots of stuff in there about mountain bikes, Sheldon Brown, etc.
I'm a bit confused by the recent outcry about relevance. Do people disagree that social signals can add to relevance? If not, then how does Google get these social signals, if they're locked out of FB and twitter?
It seems like a pretty clear strategy to me of simultaneously trying to grab as much social signal as possible by creating a meaningfully competitive social product, while also demonstrating to its competitors the value of allowing Google access to social signals.
I find it strange that everyone seems to want Google to just keep doing what it's been doing, to, in a word, stagnate. I, for one, am happy to see Google continue to do big things, undertake big initiatives, respond to a changing world.
In essence, if they can't get those social signals via fb or twitter, they simply shouldn't use them. Trying to build your own social network that is getting questionable traction (at least among the majority of the people I know), then using that network to judge search results, means that your search results suffer. Your social network has to be the one people are using before the social signals are actually useful and can be said to truly help with relevance.
That is the fundamental problem here, and unless the promotion of Google+ content results in better (not higher, because they've got plenty of accounts, but better) traction for that social network, then they're shooting result quality in the foot, if not directly in the head.
Bootstrapping the social product by funneling your actually successful product's users into it seems like a fairly obvious and reasonable approach to me. (I'd rather not get into an antitrust argument here, since that's been done to death, and I find it pretty uninteresting to begin with.)
You can wave your hands around in pretty patterns all you want – the fact remains I'm a user who used to enjoy the product and now is frustrated by it. I'm not alone in this. Have they annoyed enough users for it to be a big problem? We'll see. But big or not – it's a problem. Adoption curves are led by early adopters. Do they really want to send the sort of person who was once a Google advocate into the arms of a competing service?
> I find it strange that everyone seems to want Google to just keep doing what it's been doing, to, in a word, stagnate. I, for one, am happy to see Google continue to do big things, undertake big initiatives, respond to a changing world.
I don't think we actually disagree here. If Google does nothing, it's dead.
I'm just not sure that the something they've chosen to do is going to be a good idea. If they lose the plot on search, there's not much to fall back on. But, sometimes you do have to bet the farm.
I'm not sure what you mean by "wave your hands around in pretty patterns." I'm not denying that these things annoy some or even many users. I'm saying Google's pretty obviously in the best position to be able to tell, and famously and religiously acts from that position to a fault.
We seem to agree that Google can't do nothing. I guess my perspective on it is that Google's willingness to change even its core source of revenue continues to draw admiration from me.
The quality of their product degraded and they just kinda... did nothing about it for awhile. It got so bad it had to turn into a press thing before they did something about it. The press is damn lazy, so for my money, that means something.
As to the rest – I share you admiration for that kind of gumption. Money can so easily blind and poison progress. So the strategy to continue evolving is sound! The tactics, though, are dubious.
I'd say that any changes to search need to be made with relevance, convenience and user satisfaction as inviolable factors. Anything else is playing with fire in a library, if you're Google.
Or maybe not. We'll see.
I just really want my 2007-era Google search back.
I do agree that the tactics are often dubious. I certainly don't always agree with them. But I can simultaneously understand the motivations behind them. I would have loved to see more grand, open initiatives like Wave, even as it failed spectacularly, but I can understand the trepidation.
You said it. I'm reminded of a line from Stephenson's Snow Crash:
Hiro runs down the center of the Towne Hall... He turns off all the techno-shit in his goggles. All it does is confuse him; he stands there reading statistics about his own death even as it's happening to him. Very post-modern.
Google's big mistakes will be the best documented mistakes in human history, especially in retrospect when historians have sufficient time to sort the data from the chaff. But that doesn't mean the mistakes won't happen.
The issue, for me, continues to be the quality of the results. I still have a lot of issues with duplicate content polluting the results (particularly on programming related queries). I still have to dive a few pages deep way too often to find what I really want.
Although I've yet to find a competitor who consistently does better (or even as well).
http://googlesystem.blogspot.com/2011/11/google-verbatim.htm...
But, is there really such a market? I remember reading a blog post from a FB employee (don't have the link at hand) saying that Google over-dramatized FB's willingness to go into search, be it "social search" or whatever. What brings people to FB, and more to the point, what makes them spending hours (literally) each day on that site is not their search feature (which I'd say is not that good) but the fact that everyone and everything has a profile or a page or something (yes, maybe even the stray-dog from around my building's corner that a lot of my neighbors feed regularly - try creating a page for him on Google+).
Let's say fb/twitter release a public api to access the social signals. So things like number of likes on a particular news page. In return, is google willing to release the data on number of clicks on a particular search result for all queries? Seems a bit hyprocritical no ?
And Google's proposition to all web sites everywhere is that in exchange for indexing their content, they provide exposure through the best search engine and the main interface to the web for most people. This is also what FB and Twitter are complaining about: that they aren't receiving proper exposure.
It seems like a fair deal to me, and for Twitter and FB to be complaining about lack of exposure when they're trying to charge for the data that would appropriately affect their ranking is what seems absurdly hypocritical to me.
I don't understand how so many people are missing the point demonstrated by http://www.focusontheuser.org/. It uses information available from Google search to inject twitter, facebook, etc results into the "your world (as long as it's google+)" module. The ranking it uses is Google's ranking of the relevant results. No special API access. No back-end deals. No money. It's stuff that's available to Google and that Google already has and has already ranked.
If they choose to dilute the worth of their search engine by injecting less-than-relevant (according to their own relevance algorithm) results in order to attempt to grow Plus (presumably hoping that'll ultimately increase the revenue to offset the losses on the search engine side of the business), they are perfectly within their rights to do so. Just don't buy their "oh we'd be happy to show more stuff but boo-hoo we don't have the data" line -- they totally have the data, and it doesn't require sweetheart deals. The only thing they can't do is grab your personal social graph and re-rank according to it. So they use something they know is sub-par (Plus's currently very partial graph) and don't bother dialing in relevance signals against their own best-in-the-world search engine.
Not related to you, but I did some experimentation on my own and published it at: halfaclick.blogspot.com/2012/01/focus-on-thy-users.html. It is not that FB (contributors from where helped develop the FOTU) shows any more focus than Google.
Also the way Focus On The User portrayed the entire thing was not that they are giving a tool that makes Google better but that Google was not using their information. And why just Facebook, Twitter and MySpace? Why not HN, Quora, SO and others?
I honestly don't really see the downside to them opening their data to Google, though. They were in a better position before Google started G+. Google didn't want to make a social network. They just wanted the data. They had the open social initiative and were trying to make social networks interoperable, but FB and Twitter resisted. They sort of forced Google into the position of competing directly with them. Now they're complaining about it.
What really confuses me is what was the downside to them opening up to Google? (Esp. Twitter, who actually had a deal where they were getting paid for access to the firehose.) Google wasn't competing with them directly. In fact, Google was in a position to direct a lot of traffic to them.
The only thing I can imagine is that both FB and Twitter thought they were going to be able to obviate search and tried aggressively for that. I don't know why they thought Google would sit still for that, though.
It is obviously damned if it just allows Facebook to cruise to victory. That was the consensus about what was going to happen before the launch of Google+ last year.
But it is also damned if it fights back with all the resources at its disposal. Either the goliath is being mocked for losing to David, or it is being attacked for actually giving David quite a beating.
Moreover, in this particular case, Google doesn't have full indexing access for Facebook and Twitter (Twitter revoked access to the firehose sometime ago). So if they really want to play rope-a-dope, Google could use this to get FB and Twitter to give up indexing rights. Yet without Plus and SPYW, Google wouldn't have had any negotiating leverage with FB/Twitter to get those indexing rights.
Then they had to start pretending that things had more upvotes than they really did, which made it worse.
The place ghetto-ised in a scarily quick fashion.
I'm not sure though how much the earlier 'Digg-revolts' over senselessly mucking around with the UI contributed. Had they already destroyed or eroded a lot of the goodwill of their users?
Personally I believe the problem started back when they decided to boot Apple stories off the front page, and twiddled their algorithm to achieve it. Yes, Windows fanboys hated seeing those stories popping up all the time, and would go in and troll the Apple fans something chronic (and then complain that it was justified because the Apple users were doing the same thing - even though they weren't)... but doctoring the results like that just meant that what was (at that time) basically one of the few IT companies doing new and interesting things, instead of just being a "me too" copycat.
The HBS case study on Google won't even mention "don't be evil" except maybe as an example of how amateurs shouldn't be permitted to do corporate communications.
There are plenty of companies that we love (hello, Apple) that don't pretend to have lofty corporate values. Yet we keep buying their products and they keep making assloads of money. That's the way it is.
Speak for yourself... there are plenty of us who cheerily invite Apple to go straight to hell. You can certainly count me among their number, for one.
I have no use whatsoever for their anti-hacker ethos, their locked-down, proprietary, "no user repairable parts inside" attitude, and the arrogance with which they presume to know more about what the customer needs, than the customer does.
I'm sure this is going to be unpopular, but sorry, when it comes to corporate success, values don't mean shit.
That's hardly a settled point. There's evidence on both sides of the argument, to this day.
Honestly, from my point of view, just myself counts as "plenty" in this context. But I'm pretty sure the number is actually higher than that. Just count up all the F/OSS ideologues who refuse to use closed-source / proprietary software as a matter of principle, for starters. And I'm guessing there are a handful of folks who don't count themselves as total GNU hippies, but still prefer to shy away from companies that sell products based around such a locked-down, proprietary, closed-off ethos.
Apple was crowned as the biggest PC maker in the world this week.
I never said they weren't successful... I'm just saying that not everybody buys into their bullshit or likes their shitty proprietary products, that's all.
...So you've chosen to swallow someone else's bullshit line then instead? That's way better.
We may not all value the same things, that doesn't mean we need to be deliberately ignorant about the appeal others see in a product.
...Huh?
You appear to be replying to me, but I'm puzzled by what you've typed. Who mentioned anything about values? This is about ruining a core product's user experience and the repercussions of any strategy that allows that.
Definition for spyw: Web definitions: (SPYWS) Stroke Prevention in Young Women Study. www.health.am/acronyms/s7/
you're welcome! (edit: to be fair though, the first link is a good one and explains the whole situation) (edit 2: oh wait! That's because of my personal search results. This is too funny)
This is my perception of how Google evolved:
In the early days, it used to be like running grep on the whole internet but getting your results sorted by a very relevant algorithm (page rank). Then over the years they added a lot of features that fell into three categories:
* operators to run more powerful queries
* adding more complexity to the sorting of the results (localization, penalizing websites trying to game the algorithm, ...)
* adding widgets given the query (ups tracking, weather, calculator, ...)
These were all good, and I loved the fact that you could run some very clever and complex queries to get exactly what you wanted (and the googledorks were born...).
But recent improvements to Google have all been about "guessing what the user wants" (the best illustration of that is the search as you type) and while it's hard to argue about the validity of this approach my problem with it is that it won't give me any fallback options when it guesses wrong.
To be honest, Google does a very good job at guessing what I want. But the frustration caused by the few times it doesn't get it just makes me miss the "old google". And with every new recent improvement I keep being reminded of the MentalPlex [1]
I use search at home, when going out, at home, everywhere. And in every different context I have very different expectations. Precisely sharp results for anything in my domain of expertise, blurry and "good enough" results for generic searches, multilingual results for broadly relevant searches, very local results for everyday related searches. And these expectations might change depending on any random factor.
I don't think google can weasel out what I want in what context. And I don't feel like adjusting dozens of options everytime I do a different kind of search.
Before google started making search smarter, the burden was on my shoulder to do the work to get relevant results for my queries. Now that the algorithms do more and more assumptions, I get to fight with it more and more, and before the blame lies on my search skills, it goes to google for making it hard.
My solution was to use separate engines (including multiple instance of the same ones, like ncr, US only, jp only etc), but a fast a dumb engine is really the way to go.
I hope that duckduckgo will get that it doesn't need to compete with Google at the "guessing what the user wants" and keep providing clear and concise ways to make very specific queries.
I'm pretty sure that most people like this. I hated this feature at first; I would search for "obscure error message" and Google would say 'Did you mean "non-obscure error message"?'. It's annoying because that's not what I meant, but I realized that there were simply no pages on the Internet that answered my question about "obscure error message". So Google assumed I made a tiny spelling mistake, and tried to give me relevant results.
For everyone who's not searching for programming error messages, I think this works really well. I've watched my family and friends use Google and they can't spell or even come up with good search terms to save their life. Despite that, Google always gives them relevant results, because 99.99% of the time, it knows what they want.
Remember, Google doesn't change their product to make it worse. Every move is designed to improve relevancy (or revenue, perhaps), and features that don't work get reverted. (Remember Code Search or Google Video? They didn't work out, and they're gone now.)
Social is the 2012 version of "let's try to make search better". We'll have to wait a while to see if it does. Going with our gut will tell us to refuse change and keep everything the same forever. But data will tell us what users really think, and Google+ hasn't been around long enough to collect the data.
Meanwhile, SERP quality began to deteriorate as well. We suffered for something like 18 months under the regime of those Stack Overflow scrapers and their ilk... and now there's the comically titled SPYW.
Don't you think having your friends affect search results is a good solution to this problem? Google couldn't figure out which site with SO content was legitimate. Now it sees all your friends linking to StackOverflow.com, and it knows that it should probably trust that content more.
I think geeks dislike social networking (I do), and so they assume Google+ is going to fail. But I think that having my friends carve out a little part of the Internet that's most relevant to me is going to be helpful in the long run. More helpful than "Your Friend has given you 86 potato seeds" that you get from social networking on Facebook, anyway.
Matt Cutts wrote a good article about this, BTW: http://www.mattcutts.com/blog/search-plus-your-world/
EDIT: Never mind. Didn't know you are a Googler. of course you agree with Matt Cutts. :)
I have no idea whether they're right.
http://insidesearch.blogspot.com/2011/11/search-using-your-t...
Or turning off Search Plus Your World?
http://3.bp.blogspot.com/-hf5K14O6Fwc/Twufv9LK1PI/AAAAAAAAAL...
No, seriously... Knowing that these two things exist, are you still unable to search what you want to?
For example, if I could I'd set synonym replacement on but turn off assuming I misspelled what I typed. Where's the option for that?
Where, you typed a word, that looks convincingly like another word had been misspelled, but in fact is correctly spelled - but has synonyms that you want to look for?
...but, sometimes you want to type something that looks like it should be autocorrected, but you want it to not autocorrect it...
...and it's not enough that you can do these things - you want them to happen automatically?
I'm sorry, but that is not "a search tool that does what you ask for," that's MAGIC that you want.
If you think a search engine exists that does this better, please tell us all about it, because I think it would be an example of an intelligent machine, in the Turing Test sense.
My suspicion is that Google is going to treat us to a very painful object lesson: what happens to your business when you compromise the values of a beloved core product.
Here's the thing: having the most complete, most accurate, most relevant search results means never having to say you're sorry. You can add ads, you can do additional products, you can add a kind of clunky single-signon, whatever. Because at the end of the day, the user needs the thing you have that no one else does.
Now, Google had this. But one day, maybe around 2009, something happened. Where once I was delighted with Google's search, it started getting annoying. Things like automatically showing you a SERP for a different spelling of your query, because Google thought you were looking for that. Then they started matching to synonyms (tear and rip, say). And so this tool that used to do exactly what it was asked became too clever by half.
Meanwhile, SERP quality began to deteriorate as well. We suffered for something like 18 months under the regime of those Stack Overflow scrapers and their ilk, with plenty of equivalent nonsense in other verticals (hello, ebooks!).
So, already, Google took its eye off the ball for the one thing that previously had been inviolable. And now there's the comically titled SPYW.
Google's not stupid: they get that the rise of mobile and specialty apps/services that go with it are going to whittle away at the searches they'll be asked to do.
They've bet the company on Android and Google+ giving them an out. Android was a good call, as it puts them in the driver's seat for a lot of this mobile action. But they need the web, too. Can they get away with compromising their search with this nouveau portal strategy?
My hunch is that they're going to pay dearly in the process of finding out the answer is no. They're too big and too smart not to make the transition to whatever the next big thing is after search.
But boy – screwing with the golden goose that earns both reputation and cash?
Well, give 'em this: they're willing to take risks.