Take Google Chrome, for six years the browser of choice for your correspondent. It’s clean, fast and simple, yet increasingly it tries to second guess how I want to browse the web.

Now, fair’s fair, I may not have a typical browsing history: I spend one week a month in the UK offices of PC Pro, and the rest of the month in our French outpost.

Yet, it’s hard to see what business that is of Google. Or why it should decide to interfere and start presenting a host of content and features in French and refuse to switch back to English with the vigour of an agricultural blockade.

After a recent upgrade, Chrome on a main desktop PC in the office switched from driving on the left to driving on the right without a shrug in the way of explanation. It turned the URL search bar from useful feature to irritating deviation.

Chrome on a main desktop PC in the office switched from driving on the left to driving on the right without a shrug in the way of explanation

Instead of using the generic google.com as a default search engine, it switched to Google.fr, meaning all search results were skewed heavily towards sites from France and in French. Fine if you’re looking for information on the Cathar uprising or the perfect tarte tatin recipe, but France hasn’t made the greatest contributions to technology and the web.

After poking around in the browser’s help screen, there are options for forcing the browser bar search to revert to English results by default, but why would Google overide my original settings, when it knows, what with me being signed into Gmail, that by choice I use English?

If that’s as bewildering as a French supermarket’s opening hours, it gets worse. Chrome, apparently with no hint of irony, points out that the page delivering the results is in French and asks if I’d like to translate them into English. Genius.

Google knows I speak English. It’s there in Chrome’s language settings and over the years it’s developed a pretty good understanding of my browsing habits – how else would it know exactly which adverts to show me? — yet it insists on trying to push me into French options based on where I am browsing from. If Google knows where I buy my pants or go on holiday then it ought to know that I habitually surf UK websites.

Still, at least I speak French. Back in the UK for a week’s shift in PC Pro Towers, I performed a search at my parents’ home while signed into Gmail, and shortly after signing out realised that my parents’ edition of Chrome had now switched to French — and mother finds it hard enough to cope with technology without Google sabotaging her computer with foreign languages.

Google is by no means the only offender here, with an increasing number of companies defaulting from .com to the local suffix and language without obvious recourse  – but in a world that’s more mobile and more travelled than ever before, do we really need our technology telling us where we are?

It’s a simple question, do you trust Google? My confusing answer is yes and no. Yes, I trust Google to find more relevant information in less time than other search engines. No, I don’t trust Google to filter out all the cons and scams.

Indeed, the level of trust that I associate with Google search declines dramatically when it comes to those results that appear at the top and side of the page, you know, the ones with the very light text saying ‘Ads’ next to them. I cannot recall ever clicking on a ’sponsored search result’ for a couple of very good reasons:

1. The whole point of using Google is to uncover information that has been deemed relevant courtesy of the hugely complex algorithm at the heart of the search engine’s success, and not which has been dropped onto the page simply because someone paid for it to be there.

2. The bad guys have, for as long as I can remember, been using such sponsored results to lure people to their sites and whatever nefarious activity lies within.

Not that I am suggesting for one second that all such sponsored results lead to malware-ridden, spam-infested, drive-by-downloading and spawn-of-Satan sites. Some are genuinely just trying to buy your attention, because their SEO skills are such that they would just get lost halfway down the organic results list. Others are just covering all marketing bases, such as PC Pro itself which appears at the top of the organic results list when you search for ‘PC Pro’ but also as the solitary sponsored result.

By appearing as an advert on a Google search results page there is something of an implied transference of trust from the Google brand to the advertised resource

Yet, the point remains that many sponsored results are potentially unsafe, and there’s a simple way to test this claim. Search for ‘free downloads’ and if you have a safe browsing tool installed check out how many sponsored results are flagged as having a poor reputation.

I use Web of Trust (WoT) to provide an at-a-glance idea of site reputation as it uses a community-based, crowd-sourced system to determine if a site is trustworthy. A green symbol next to a result indicates no reported problems, amber advises caution while red means go no further.

Taking that ‘free downloads’ search as an example, three of the top ten (or 30%) of the organic results are flagged red with the remainder green, while four out of the seven (or 57%) sponsored results carry a red ratings. It’s 71% if you take the solitary amber warning into account as well. In other words, my trust gets turned on its head with only 30% of the sponsored results being flagged go-ahead-green rather than 30% being flagged stay-away red.

The WoT ratings take factors such as privacy issues, child safety, vendor reliability and trust into account. While not every site flagged red by WoT is going to be run by scammers, there’s good reason why the bad guys like the Google sponsored search results and that’s the big T word.

By appearing as an advert on a Google search results page there is something of an implied transference of trust from the Google brand to the advertised resource, and I’m convinced that’s why people click on them. It is assumed, wrongly in my opinion, that a sponsored result can somehow be trusted more than an organic result.

As if to prove how valuable the trust equation is to the scammers and spammers, take a new pharmacy spam campaign that hijacks the Google brand. MessageLabs Intelligence has been tracking this campaign recently, and notes that it claims to be promoting a “Google-accredited” resource. The truth is that Google doesn’t give approval to any site, let alone an online pharmacy; why would it? The spammers even use the Google logo with the ‘oo’ replaced by a couple of tablets. I love the concept that Google would approve a drugs-related doodle.

Just as I have trained my brain to ignore the adverts that appear alongside many free iPhone apps that I use, so I have trained it to totally filter out the paid-for search results that appear at the top and side of the organic search results. I suggest you start doing the same. Trust me…

To test the app, I performed a series of nine searches using my iPhone’s camera. You can see the photos, which are stashed in my search history, below:

Here’s how the search engine got on with my photos (starting from the bottom right-hand corner, and working upwards from right to left):

1. PC Pro magazine cover

Goggles got off to a rather flaky start, using its OCR software to mistakenly identify the magazine by the headline rather than the PC Pro logo. Hence the search results pointed us towards a Spanish bank. Verdict: miss.

2. PC Pro magazine logo

We decided to give Goggles a second bite of the PC Pro cherry, this time zooming in on the magazine’s logo rather than attempting to scan the cover as a whole. This worked a treat. Not only did Google match the logo to the one on our website, it returned PCPro.co.uk as the top search result. Verdict: hit.

3. Dell XPS M1330 laptop

This is my day-to-day laptop and Goggles did an impressive job of nailing it down. Admittedly, the lead search result was for an XPS M1530, but Goggles was largely relying on the image of the keyboard to identify the laptop, and the M1530 has an identical keyboard to my model. Impressive stuff. Verdict: hit.

4. Nabaztag rabbit

My colleague Nicole Kobie has a defunct Nabaztag rabbit sitting on her desk, for reasons I’m yet to fathom. Nevertheless, it’s a distinctive product and one that Goggles should have no problem identifying. Should. Inexplicably, it returned a search result for Burton snowboards. Verdict: miss.

5 & 6. Apple iMac

Nobody could accuse Apple of not producing distinctive kit. Yet, Goggles had immense trouble identifying the iMac. The first time it returned image results for some Gothic-looking animation, and at the second attempt it identified Apple’s all-in-one as a rally car. Verdict: Double miss.

7. Talk of the Town

Google boasts that Goggles is particularly adept at searching for books, and credit where credit’s due, it did a fine job of identifying Ardal O’Hanlon’s excellent – if now rather aged – 1998 book Talk of the Town. The first search result even sent us to Amazon to buy a copy. Verdict: hit.

8. Twinings Green Tea

A tea round in the PC Pro office is a nightmare: if it’s not Tim’s cinnamon monstrosities wafting across the office, it’s some other herbal nonsense, such as this pack of green tea sitting on the news desk. Goggles failed to identify the exact product, but the OCR picked out the Twinings brand and sent us scurrying to the company’s website, which was close enough. Verdict: hit.

9. Darien Graham-Smith

In Google’s defence, Goggles specifically states that it’s unlikely to recognise animals or people. But we couldn’t resist, so I snapped a photo of PC Pro’s charming technical editor and Goggles returned this:

Verdict: Miss (although a massive hit amongst the PC Pro team).

FINAL SCORE: So, out of a total of nine searches, Goggles produced the correct result in four. Looks like Sergey and Larry might have let this one out of the labs a little prematurely.

]]> http://www.pcpro.co.uk/blogs/2010/10/06/google-goggles-for-iphone-review-first-look/feed/ 5 http://www.pcpro.co.uk/blogs/2010/09/10/google-instant-v-bing/ http://www.pcpro.co.uk/blogs/2010/09/10/google-instant-v-bing/#comments Fri, 10 Sep 2010 16:02:44 +0000 Tom Arah http://www.pcpro.co.uk/blogs/?p=24364 In case you haven’t heard the news, Google has radically changed the entire nature of web searching with its new real time and predictive Google Instant service. You can read about it here or see the introductory video or full launch video.

By far the best option is to see it in action yourself. You can do this simply by visiting the main Google.com home page (on google.co.uk you have to sign in to your Google account to switch Instant on) and then starting to type in the search box (assuming that you’re using one of the currently supported browsers: Chrome 5/6, Firefox 3, Safari 5 for Mac and Internet Explorer 8).

In practice there are plenty of teething problems – as you’ll see from the screenshot, the first prediction if you type “Google Instant” is “Google Instant Not Working” – but if you persevere you should be able to get it up and running.

Once you have seen it in action, I think that most people will agree that Google Instant really is as extraordinary as Google claims.

One of the big questions Google Instant raises is where does this leave the competition? And Bing in particular. I can’t see how Bing can compete on the desktop unless it has the resources and capability to do exactly the same – which, let’s face it, must be a pretty tall order.

I think the problems will turn out to be even more acute in the handheld space (where Google says “we plan to release it soon”). Google Instant’s real-time and predictive approach will come into its own on the smartphone where typing a few letters and tabbing a couple of times will let you intelligently move through Google’s entire index until you find the link you’re after.

If Microsoft had managed to pull off something like Instant itself, it might well have proved the game changer the company has been desperately looking for. As it is, the hardwired Bing button on the new Windows Phone 7 devices may well turn into an embarrassment and a massive tactical mistake.

But at a press conference I attended in London last night, Google fellow Amit Singhal was given a rough ride by a bolshy journalist who demanded to know why the company didn’t intervene more often when search results went awry.

The journalist cited the example of a search for “Martin Luther King”, which apparently returns a “racist” site high in the Google search results – largely because so many people have linked to the site to object to its content, but Google only sees a link as a link, good or bad, and has rewarded the site with perhaps underserved PageRank. He also raised the issue of the infamous chimpanzee “picture” of Michelle Obama, which remains high in the Google Image search rankings today.

Why does Google allow such blatantly offensive results to pollute its search engine, the angry journalist demanded to know? Why is Google so reluctant to intervene?

“We don’t induce subjectivity into search results, because my subjective opinion is my subjective opinion,” replied Singhal. “Yes, the algorithm makes mistakes – I’d be the first to admit that. There will be occasions when the algorithm fails badly, and we look to address that.”

“I’m by no means saying for that query [Martin Luther King or Michelle Obama] we’re helping users. But that failure is no reason to subjectively filter. Where do you stop?” he asked, citing examples such as Scientology sites, which polarise opinion on their offensiveness. A robust and entirely justifiable policy, in my opinion.

However, Singhal did reveal that there are three reasons why Google will manually interfere with its search results. They are:

1. Removing links to sites distributing malware.

2. Removing links to sites that are the subject of legal action, such as DCMA takedown notices issued by copyright holders.

3. When Google’s Safe Search fails and returns pornographic or unsuitable images. Or when the image search is returning instances of child abuse. “There is no way for the algorithms to tell the age of the person depicted,” Singhal claimed.

There could perhaps be a fourth addition to that list: the now defunct Google China search engine, which censored content at the behest of the Chinese Government. But Singhal didn’t dwell on that matter for too long…

On Bing, Microsoft’s very handy and super-relevant search engine, it would seem that Anglian Windows has bought a featured spot that pops up to tell you about the new Government scrappage scheme – not the one that applies to cars, the one that applies to double glazing.

Except that “Windows” has to be one of the most frequently searched terms on the web  -  I put it in almost every search because I’m always looking for Network error messages and their fixes, and if I leave “Windows” out then I get five times as many hits about Linux, which I don’t need to see. I am very unlikely to go from my “Windows” search to Anglian for some new double-glazing, so quite why Anglian’s ad appears when I type terms such as “windows trust failure vmware” into Bing is a mystery.

Thank God they only pay when people click on the ad, or the Government may be bailing out another company.

This adds a few  features across the board, such as a revamp of importing and various interface tweaks, but the clear focus of the new release is on in-depth tagging of images via a new side panel that offers three tabs for applying text-based tags, locational geodata and new face-based tags.

To be honest my heart sank when I heard this – what I’ve always liked about Picasa is that it keeps things simple and doesn’t treat managing your photos as a full-time job. Moreover I’d recently come away less than impressed with Photoshop Elements 8’s new face tagging not so much because the technology doesn’t work (it does though imperfectly), but rather because the gains aren’t worth the effort.

So how does the new Picasa 3.5 shape up?

Brilliantly.

I’m still too lazy to be a huge fan of intensive text-based meta tagging but Picasa clearly needed to improve its feeble floating Keywords panel and the Tags panel certainly does that. Likewise I don’t suppose many users ever bothered to geotag their their images via Google Earth, but now you can simply call up a Google Map in the Places panel and drag-and-drop your images onto it. Excellent.

The real revelation though is face tagging. Select a folder or album and Picasa does a great job of pulling out the faces in your images – or at least all those looking at the camera – and listing them in the People tab with an “add a name” label next to them. I half-heartedly marked up a few photos of family members while idly thinking how nice it would be to have a week off to go through my entire collection of thousands of images like this…

But then I noticed that Picasa 3.5 had added a new People category to the main navigation panel down the left of the screen. I clicked on the “me” tag and discovered that, based on just one image that I had tagged (with comedy glasses), Picasa had found another 17 images that it was confident was me and another 160 or so that it thought was me. Incredibly all of them were, including many in bad hats and blonde wig (don’t ask). Confirm these suggestions and Picasa goes and finds some more, and then more again and all with astonishing accuracy. I generally don’t like having my photo taken (and it’s my camera) so the fact that  Picasa has found over 300 shots of me without a single mistake is remarkable – and more than enough.

It was a similar story with other clearly-defined adult faces, but accuracy with my five-year old twins (non-identical) was unsurprisingly far more hit-and-miss though I think this might well have been because I didn’t bother finding clear images to tag originally (and because I can never get Robbie to look at the camera).

In any case it really didn’t matter because it’s incredibly simple to select the incorrect suggestions and drag them to the right tag. In fact the whole process is actively enjoyable.

That’s something I never thought I’d be able to say about the prospect of tagging thousands of images containing even more faces.

Ultimately human nature means that most people are most interested in pictures of people, and especially pictures of themselves and their friends and family. With another brilliant search technology under its belt – bitmap-based faces – Google makes a previously unthinkable chore into a real joy. Superb.

And did I mention that it’s free?

I mean, obviously it was never going to slay Google on its first day. But after watching Stephen Wolfram’s pre-launch screencast I did believe it was at least going to be a credible alternative information source, offering authoritative and structured answers in a way no traditional search engine could aspire to.

Sadly, now Wolfram Alpha’s here it turns out that it doesn’t bloody know anything.

Depth rather than breadth

All right, let’s be fair: it actually does know quite a lot. If you try out the test queries, and your own variations on those themes, you’ll see that, actually, there is a remarkable amount of information stored up in the back-end.

But as soon as you start to stray outside of those specific areas of knowledge – and start asking it about things you’re actually interested in – the experience goes rapidly downhill.

Here are ten sample queries I’ve tried. To try to be fair, I’ve stuck to the sorts of information Wolfram Alpha is supposed to handle (as described in its FAQ): systematic, public, factual knowledge. And, just for comparison, I’ve made the same queries in Google and clicked on the first result. These were the results:

1. “House prices in Kentish Town”
Wolfram Alpha has never actually heard of Kentish Town. Maybe its existence isn’t public knowledge. Whatever the reason, this makes it pretty damn useless if you’re, say, looking for a new house within easy commuting distance of Dennis Towers.

The first hit on Google was a website detailing the asking and selling prices of houses sold in Kentish Town since 2000, along with trends and reports of how long they remained on the market.

2. “Biggest selling single of the 1990s”
Nineties chart-toppers may not be to everyone’s taste, but the facts and figures are (if you’ll pardon the expression) a matter of record. Sadly, Wolfram Alpha has none of them. It did, however, tell me that Shakira’s surname is Ripolli. Wouldn’t Graham-Smith suit her better?

The first hit on Google was a Wikipedia page telling me that the single in question was “Candle in the Wind 1997”, which reportedly sold a sickening 37 million copies.

3. “Newspaper circulations”
When I entered this as my query, Wolfram Alpha’s first guess was that I might want to know about Sports Illustrated. Second guess: The Idaho Statesman.

The first hit on Google was the Audit Bureau of Circulation. From here I had to engage in a little navigation, but after six mouse clicks I was rewarded with a list of UK national papers and their average net circulation.

4. “Bicycle theft in London”
As a recent victim of this heinous crime, I’ve been wondering how many bikes are stolen every day, where the hotspots are and where, if anywhere, they resurface. Unfortunately, to get any response from Wolfram Alpha at all I had to generalise this question down to just “bicycle”, at which point it told me that the modern bicycle was invented by John Kemp Starley.

The first hit on Google was a blog post advising me that expensive bikes are less likely to be stolen than cheaper ones. Not much consolation there.

5. “Eurovision winners”
After Norway attained the highest Eurovision score ever on Saturday, I started wondering about previous winners. Which countries had done best, and which worst, over the years? Wolfram Alpha guessed that perhaps I wanted to know about either the euro or the AAA Vision aircraft.

Google once again led me to the Wikipedia page on this topic, which offered all the information I could possibly want (and far more beside).

6. “CPU die sizes”
The original Pentium was built with a 0.8µm process. Now we’re down to 45nm chips, with 32nm on the horizon. Can Wolfram Alpha show me a graph of how die sizes have shrunk over time? No. It thinks “CPU” is an airfield in California.

Google’s first hit was an online article defining die size. I didn’t get the graph I wanted, but at least we were in the right industry.

7. “Web browser market share”
Admittedly the data on this one aren’t as solid as some of my other searches, but I was optimistic Wolfram Alpha would know something about web browsers. It doesn’t, but it did invite me to leave my email address so it could alert me when it does learn something on the topic.

For the third time, Google led me to an extensive and informative Wikipedia page, complete with a pie chart, a stacked line graph and numerous tables detailing the specific figures from the various data sources.

8. “UK unemployment 1980-2008”
For once, Wolfram Alpha did understand what I was asking… but alas it was unable to help due to “insufficient data available”. A few follow-up searches revealed that this is because it doesn’t have any data for any of those years. Oddly, though, if you leave off the year it gives you an estimate for 2008 (5.5%, if you’re interested).

The years threw Google off. It took me to an Amazon listing for a live DVD by a band named “The Inmates”, recorded in 1980 and released in 2008. A rare fumble.

9. “World Cup runner-up”
I tried several variations of these search terms, but Wolfram Alpha could tell me nothing more useful than the volume of a standard cup. Apparently, in the US it’s 0.2366 litres – I had to click onward to find the size of an imperial or metric cup. So much for tailoring results to your location.

Yet again, Google went straight to Wikipedia, this time to the FIFA World Cup page. I had to scroll down the page quite a lot to find the information I wanted, but it was there. (It was France.)

10. “Chemical symbol for tungsten”
“Wolfram|Alpha isn’t sure what to do with your input.” Oh, come on – really? Oddly, when I tried knocking off the word “chemical” it did yield the correct result, so I’ll give it half a point for that one.

With Google I didn’t even have to click on a link: the answer was right there above the search results, along with the source.

The verdict

Do you have to ask? Wolfram Alpha is clearly a very clever bit of programming, but if you actually want to learn something it’s next to useless.

All the same, we shouldn’t underestimate its potential. The engine that combines and presents arbitrary information in a comprehensible way is undeniably powerful. If it can only extend its ambit beyond institutional statistics and scientific formulae – if, in other words, it can start answering everyday questions about everyday topics – it could yet leave Google looking laughably primitive.

But before that can happen it needs more data. Lots more data.

However, I did promise that I would tell you more just as soon as I could.

Well, today is the day. Ignore the bitching on ars technica. Instead, go to Wolfram’s blog – Wolfram|Alpha is coming - and then head to www.wolframalpha.com to sign up.

To quote the blog: “Fifty years ago, when computers were young, people assumed that… one would be able to ask a computer any factual question, and have it compute the answer.

“But it didn’t work out that way. Computers have been able to do many remarkable and unexpected things. But not that.”

We’re still two months away from launch, but when that day happens Wolfram promises us his new search engine will have the answer. So sign up and just wait patiently for a little longer.

You’ve probably never given this feature much thought. I hadn’t, until yesterday when I was searching for some obscure technical fact and (perhaps because we’d just been talking about natural language recognition on this week’s podcast) I absent-mindedly started to phrase my search query as a question. I didn’t get far in – just one word in, in fact – before realising this was unnecessary.

But that one word was enough to give Firefox an idea as to what I might be searching for: immediately it brought up the impressively varied suggestions you see above.

I have to admit, none of these was the question I’d had in mind. But, believe it or not, a lot of people must be searching for these things — otherwise they’d be pretty cussed suggestions. And so they provide a fascinating insight into the interests and concerns of the online population.

In this particular case, for example, they reveal that people still have their doubts about President Obama. Questions about the 44th President of the United States knock Santa into a poor fifth place. Indeed, all across the web there’s only one issue more urgent than whether or not Obama is the antichrist.

This set me thinking: what other cultural revelations could this seemingly-innocent box yield? Inspired, I set out to see what other search terms might unearth, and here are my ten favourite discoveries…

1. WHY PEOPLE …  Obviously a lot of people on the internet are confused about human motives. Why do people lie? Why do they commit suicide? Clearly a sizeable contingent of truth-seekers has turned to Google in the hope of understanding why we smoke, drink and engage in other sorts of anti-social behaviour.

2. WHERE DID I PUT …  Some people put a little more faith in Google than they really should. Yes, it can find your motherboard drivers. It can even find your house. But no, dude, it does not know where you left your weed. And it’s not a mind-reader — you could at least tell it the title of your damn book.

3. WHY IS …  This one really brings home the breadth of the internet as an information resource. At once it can deal with the wonders of physics, geology and astronomy… and other, more earthy issues. It also indicates that a decent proportion of internauts have read Alice’s Adventures in Wonderland.

4. IS MY …  This one too bespeaks Google’s authority as a fount of information. People seek its advice on all of life’s important matters, such as finance, technology and affairs of the heart. They also seem to think it can detect ghosts. And some people really ought to pay more attention in court.

5. DO GIRLS …  It’s tough growing up, and trying to fathom the mysteries of the opposite gender. Luckily, today’s adolescents have the world’s largest encyclopaedia at their fingertips. Well, it’s got to be better than asking some of these questions to the girls themselves, anyway.

6. MY SISTER IS …  Of course, some boys are all too well-acquainted with the ways of females, and generally it seems they’re not impressed. Though if my brother kept saying things like this about me, I’d probably run away and turn to drink as well. I like the fact that these aren’t even questions. What are these guys expecting to find?

7. SHE NEEDS …  As boys mature, they generally warm to the idea of girls. But still they’re a troublesome gender. Girls have a tendency to come out with bizarre declarations which the rational male mind cannot comprehend. Maybe Googling the phrase in question will help you understand what she means.

8. WHO STARTED …  This one exposes the more inquisitive, almost academic side of the online population. It’s not all me, me, me: sometimes they want a historical perspective on the events and organisations that have wasted so many lives.

9. CAN A …  Somewhere in Hollywood there’s probably a producer who’d pay good money for the rights to this one. Starring Emma Watson as the middle-class white girl who gets knocked up by a video-game playing homeboy from the wrong side of the tracks. Lil’ Wayne could be the male lead, if he’s still alive.

10. I NEED TO …  A comforting one to end on. No matter how bad things may be, there’s always someone worse off than you, and so clueless they are actually searching the internet for “I need to make money fast”. Perhaps it’s related to number 7. One thing, though — you, sir, do not need to register to vote.