mediajunk

This year, search became the internet industry's red hot sector. Many companies set their sights on emulating Google, whose impressive performance has led to a renewal of business interest in the internet.

But does Google need to worry about increased competition? According to figures published by OneStat.com (via Search Engine Lowdown) Google now accounts for over half of all web searches.

Here's the breakdown:

1. Google 56.1%
2. Yahoo 21.5%
3. MSN Search 9.4%
4. AOL Search 3.7%
5. Terra Lycos 2.3
6. Altavista 1.9%
7. Askjeeves 1.6%

Consider that Yahoo's search is powered by Google (type the same search into both engines and you'll see very similar results), and you could argue that Google's share is actually 77.6% -- over three quarters of all the web's searches.

Google's dominance of the search market is growing, not shrinking. Potential competitors are in for a tough ride.

Then again, Google didn't exist six years ago...

Google has just undergone a major update of its search algorithm, nicknamed “Update Florida” by the ever-growing community of search engine watchers.

Google trawls the web and modifies its results pages on an almost daily basis now. But updates to its algorithm happen less frequently and cause massive shakeups in the results that appear in its SERPs (Search Engine Results Pages).

Earlier this year, Google underwent two major updates, nicknamed “Dominic” and “Esmerelda”. The Florida update adds the F to this series and is a much bigger update than its predecessors, with greater consequences for site listings.

A thread devoted to discussing the Florida update, which began last Friday, is already the longest, fastest-growing thread I have ever seen on any topic on any webmaster bulletin board.

The reason the Florida update is causing ructions among webmasters and SEOs (Search Engine Optimisers) is because the company has launched a major offensive against what it calls “spammers”.

In the search engine community, spam does not refer to dodgy email marketing, but to any unfair or underhand tactics used by SEOs to gain high positions in SERPs for the websites they represent.

Not everyone in the webmaster community agrees on what does or doesn't constitute spam. While most agree that hidden links (making the colour of links the same as the page background) is cheating, for example, less blatant tactics such as adding URLs to blogs or guestbooks are approved by some, but frowned on by others.

For Google the situation is clear: any webmaster, marketer or SEO who sticks by the guidelines outlined on its website is a “whitehat” (a good guy, basically), while anyone who breaks these rules is a “blackhat”.

The Florida update has seen a major crackdown on blackhats. In doing so, Google has once again improved the quality of its results, polished its integrity-rich brand, further upped the ante on those who try to cheat its system – and increased its power as the sheriff of the information superhighway

There has been much speculation lately about Microsoft's plans to steal the top spot in search technology by integrating search with all applications, not merely the browser.

Today Google responded to this threat with a pre-emptive strike: the company released a beta version of its Google Deskbar application, which integrates web search functionality with the desktop, without launching a browser. Search results are previewed in a small inset window that closes automatically.

The Deskbar application has been greeted with much hurrah by Google enthusiasts, who see it as a brazen challenge to Microsoft.

In truth, however, the Deskbar application is not a significant development. For me it has a gimmicky feel – much like recent additions to the Google toolbar, such as automatic form filler, which were similarly heralded as "technological breakthroughs".

Just as Google cannot claim to have exclusively developed the features in its toolbar, neither can it claim to have invented deskbar searching. There are already a number of deskbar search tools available. Most notable among these is Dave Bau’s Search Deskbar – which is not limited to results from Google.

While Google is clearly maintaining the quality and integrity of its search results, it has also started pandering to the market, using tactic, strategy and gimmick to keep its brand – and its popularity – to the fore.

Just days after Amazon’s launch of its Search Inside the Book feature (see my October 26 entry on this), news of Google’s book search plans are beginning to emerge.

“For the last few months, Google has been courting publishers, hoping to convince them to turn over book content that could be used in Google's database, say people close to the discussions,” says Publisher’s Weekly.

“According to a report from one publisher, Google has said it has reached agreements that allow it to enter as many as 60,000 titles in its database and also presented extensive mock-ups to publishers of how book-relevant searches will look.”

Google’s move reinforces the notion that the web will soon be a true virtual library, with book texts that are currently only available in print becoming fully searchable and downloadable.

Google is also working on a pilot project with the Online Computer Library Center (OCLC) aimed at making the vast catalogues of books, periodicals and other printed materials held in US libraries searchable by the general public.

The project involves titles and other descriptive information, not the contents of these documents. Nevertheless, it would be a massive coup for Google, since the pilot scheme contains two million documents – and a further 51 million could be made available to Google by the ODCP in the future.

“After users locate an OCLC record citation from a Google search,” explains Info Today, “clicking on the citation will link to an interface that requests a ZIP or postal code, state, or province. This in turn will locate the nearest libraries holding the item.”

It will be interesting to see whether Google adds a “book search” tab or whether it will mix printed book results with regular results. In any case, the move is a sign that competition in the search sector is now red hot, and that the internet is still rapidly evolving.

I admit that I am a frequent practitioner of “vanity Googling” – the act, I have just learned from a Febuary 2003 article in the Boston Globe, of looking yourself up in Google.

The article by Neil Swidey explores the voyeuristic tendencies that Google's effectiveness is cultivating among a growing population of users. Swidey makes many thought-provoking points, backed with good examples of how the search engine that sorts out the wheat from the chaff has simultaneously created a raft of new privacy issues.

“While most of your embarrassing baggage was already available to the public,” he writes, “it was effectively off-limits to everyone but the professionally intrepid or supremely nosy. Now, in states where court records have gone online, and thanks to the one-click ease of Google, you can read all the sordid details of your neighbor's divorce with no more effort than it takes to check your e-mail.

‘It's the collapse of inconvenience,’ says Siva Vaidhyanathan, assistant professor of culture and communication at New York University. ‘It turns out inconvenience was a really important part of our lives, and we didn't realize it.’”

Swidey’s piece is a good primer for anyone who wants to know why managing your identity on the internet has become all-important.

The article also made me aware of the fascinating – and tragic – story of 20-year old Amy Boyer, who was being stalked by a former student of her high school. 21-year old Liam Youens, kept a diary-like site that detailed various acts of stalking he had carried out on Boyer since 8th grade, and his intentions to kill her.

In late 1999, he did exactly that, shooting Amy Boyer and then himself.

Amy’s father is now a campaigner for privacy rights (since Youens used a "detective" company to find out where his daughter worked) – though he feels that he may have been able to prevent his daughter's death if he, or Amy, had only come across Youens’s site. In the interests of promoting awareness, he has reproduced the original site.

It would be hard to imagine the horror Amy Boyer would have felt had she come across her stalker’s site. But the simple act of “vanity Googling” could have helped her avoid her terrible fate.

There's a great interview in the Online Journalism Review with Krishna Bharat, chief boffin behind the Google News portal.

The article tackles the debate about whether Google News -- and the internet in general -- is a threat to journalism, or whether it is an expansion of that profession.

There are discussions about the limits of automation and the potential problems that arise when there is no human editor on hand to evaluate articles. Still, the piece makes you realise just how easy Bharat and his colleagues make it seem -- the truth is that collating news stories from around the world and serving them up them in a "democratic" manner to a global audience is no mean feat.

In many cases, the automated system has an advantage over people in performing this task. "We get 100,000 articles a day," Bharat explains. "A human editor couldn't read that many."

One of the most curious things about the article, for me, was that Bharat still thinks that personalisation is the holy grail of internet news.

I've never been convinced of this, and it hasn't really worked in the past when providers such as Yahoo! have tried it. Part of the fun of reading a newspaper is that an article might catch your eye in a section that you wouldn't normally read. This would be far less likely to happen with personalistion, where you would only ever be served up those items that fit into certain categories...

So, personally, I'm against personalisation!

The latest cause of excitement among Google-watchers (or should that be Google-worshippers?) is a search by location feature, currently in beta.

Personally, I'm quite happy with the location-based results I get by typing the name of the region I'm searching directly after my keyphrase, e.g. Internet Usability Consultants Ireland ;)

Google determines the location of a web page (which is never obvious -- all my websites are hosted in the US, for example, even though I'm in Ireland) through what it calls "signals" about "the geographic nature of a page".

I imagine these signals include IP addresses (often misleading -- as per my own example), geo.location meta tag information (rarely used) and the text on the site (again, a regular Google search will find this text anyway).

For now, the Google location search only handles US addresses, but G-Ws have already speculated that the UK, France and Germany will be next -- simply because these are the only other countries served by Google's partner, MapQuest.

Interesting how ideas come about. In January, Biz Stone -- blogger, self-proclaimed genius and author of Blogging: Genius Strategies for Instant Web Content -- came up with a humorous logo (above) for a dating service by the web's top search engine, which he called Google Personals.

Stone didn't elaborate, except to say that the idea "came to him in a dream". The proposal was all just a bit of fun until a site called LoveCompass decided to examine it more closely.

LoveCompass makes some interesting arguments, mainly that one of Google's main rivals, Yahoo!, offers personals. Why shouldn't Google, already diversifying into contextual advertising, offer dating, now one of the web's most popular services?

GoogleGuy (the company's spokesperson who posts regularly on the WebmasterWorld bulletin boards) recently confirmed to me that the company was "happy to take suggestions from anywhere" so, who knows, perhaps they'll consider this one.

Of course, Google would have to do things differently. Perhaps it could offer a
dating service specifically aimed at geeks, based on compatible genome sequences?

Hmmm, I feel (another) science fiction novel brewing...

As is often the case with Google, the first hint of a major new development was a quiet announcement on its site -- in this case, the jobs section. The ad for Director of Product Management, Googlette, leads immediately to the question -- what the hell is a Googlette?

According to the ad, a Googlette is "a start-up within a start-up" and that there will be "a wide array of them". The Director of Product Management will "define Google's innovation engine and grow the leaders of our next generation of businesses."

In an absence of any press release or comment (as yet) from Google, here's what I surmise:

• This venture – rather, these ventures – have grown directly out of "Google labs", which has spawned many new, exciting ideas. Google wants to manage some of these innovations as separate projects, recognising they may require as much nurturing as Google itself did when starting out four or five years ago.

• This is Google sticking to New Economy-style thinking, which so many companies rushed to emulate in the late 1990s, then quickly abandoned after the dot-com meltdown. Not Google, which understands that its toughest competition in the future may come from within its own offices.

  Thus, instead of trying to quash or control emerging ideas and/or personalities, Google encourages them. It offers to create a support network for potential leaders, in exchange for loyalty and an ongoing network relationship. Thus, Google's vision's of the future is one where it is the "queen bee", with many satellite companies hovering around it, at once feeding it and gaining support from it.

• Google is reacting to external competition, which is fast toughening. Witness the recent frenzy of mergers and acquisitions in internet search and advertising. Even Bill Gates is developing a search engine designed to rival Google's. Realising that the best form of defence is attack, Google's strategy is to start working on the next generation of technologies to change the web landscape.
  

I was intrigued by the evolution of the web “ecosystem” during the 90s; now we’re seeing the emergence of dominant web “organisms”, such as Google. I know, I know: I need to get out more.

By the way, did you know that Google recently bought Iceland? Yes, the entire country! ;)

Really, I've never seen so much fuss over a squiggle.

Admittedly, it *is* a nice feature. Wonder if I'll ever remember to use it?

When I first heard that Google was releasing its APIs -- the code that lets you access the behind-the-scenes areas of the leading search engine -- I thought we would see some great Google spinoff sites within a year or so.

Instead, the results have mostly been gimmicky. Like BananaSlug, which seeds your search query with a random word, so that you get a result from way down the listings.

Yeah, it's fun. For a couple of minutes. But where are the genuinely useful sites built on Google's APIs?

To display the usefulness and range of its spell checker, Google has listed every single misspelling of the name "Britney Spears" that its search engine detects over a three-month period.

I did a quick count, and realised Google had detected 592 different misspellings of "Britney Spears". I wouldn't have guessed there could be so many wrong spellings of a name that has a total of three syllables. Some are typos, but many seem to be genuine spelling faults.

For example, there were 40134 searches for "brittany" spears; 36315 for "brittney"; 24342 for "britany"; 7331 for "britny"; 6633 for "briteny"; 2696 for "britteny"; and 1635 for "brittny".

Allowances must be made, I suppose, for the fact that a lot of people searching for Britney Spears don't actually speak English.

Nevertheless, I wonder whether the Google spell checker is improving people's spelling -- or further deteriorating it. As with word processors, users may grow to rely on spell checkers doing the "thinking" for them.

But maybe it's not wrong to depend on software to correct our spellings -- rather than, y'know, actually having to use our brains (sheesh, what a drag!).

I'm spell-checking this post, after all. And I confess, I didn't really count the Britney misspellings -- I cut-and-pasted them into MS Excel, and let the spreadsheet count them for me!

Hmmm... maybe I could devise an algorithm to maintain this blog...

Google has launched a beta version of its upgraded toolbar, version 2.0. I didn't bother downloading as the new features are essentially gimmicky -- like ad blocker and form auto-fill -- and do not enhance web searching.

In keeping with my previous post about Google's (poor) sense of aesthetic design, check out the centred red text on this page. Eeewgh!

We all admire Google's commitment to usability. But usability and sexy design should not be mutually exclusive.

Google started out well: its homepage is usable but aesthetically elegant. As more pages have been added to the site, however, Google has failed to maintain the balance between pragmatism & style.

The latest addition to Google’s site -- the adsense page -- is the least stylish yet. The page balance is askew; the layout is asymmetrical. Left-aligned elements compete for visual attention with centered elements and boxed elements. The result: the page looks too busy, too clumsy.

This inconsistency is found too in the haphazard use of borders -- some are hairline, others are thick – which, along with the general lack of “breathing space” between elements, adds to the overall sense of untidiness.

Worst of all is the extremely thick border around the submit buttons. Talk about overkill. Relax, Google. We know the buttons are there. We see them. Sheesh. (What next: < blink> tags?)

Frankly, https://www.google.com/adsense is ugly. Perhaps Google has too many computer science PhDs and not enough -- if any -- aesthetically-sensitive employees?

Google has agreed a deal with Lycos to deliver contexualized ads on the portal's member pages.

Lycos offers free web space to members, in exchange for the right to put annoying ads on the resulting sites (take a look at this random-chosen and typically ad-blighted Lycos member's page). Google will presumably use its infamous algorithm to glean the meaning of the page (though such rudimentary meaning-gleaning is a far cry from genuine AI), and thereby select an appropriate category of advertisements.

The move confirms widely held suspicions that Google is moving aggressively into hard-nosed business mode (at the risk, perhaps, of its heretofore pristine brand). But it also tells us something about the nature of advertising on the web.

The banner ad never really took off on the web. Advertisers learned that, when given a choice, users prefer to ignore ads, and certainly do not care to interact with them. Clickthrough rates for banner ads are typically less than 0.01%. The advantage that advertisers had in other media -- namely, the passivity of the audience -- has been lost on the web.

To counteract, then, advertisers have had to get smarter, and respond to the wishes of consumers. But is contexualization really enough to make web advertising take off?

I doubt it. Modern consumers -- especially young, savvy web users -- are techno-literate enough to understand how targeted advertising works and not to be unduly allured by it. Unfortunately, advertisers will therefore probably continue to adopt more underhand techniques, such as pop-ups; pop-unders; uncloseable windows; and other spammy strategies that have already been tried and tested by the most shameless of them all -- the porn industry.

Since Google first made its APIs availabe for any and all developers to use, many quirky new utilities -- based in one way or another on Google -- have surfaced on the web.

One that I've found interesting is Poodle, a tool that spiders any given site and points out "warning" areas that, if improved, could increase the site's ranking with Google. It takes a few minutes to figure out what's going on when Poodle gives results, but eventually you get a feel for it's colour-coded "advice".

Another example of a use of Google technology within a different application is Googlematic, which lets you get Google search results through Instant Messanger programmes, such as AIM.

Some developers have used the Google APIs to create fun sites, such as Googlism. Type your own name in there (or just your surname, if you don't see results) and see what you learn about yourself!

Other spin-off sites/tools are listed on Googuide.

Danny Sullivan, on his excellent site Search Engine Watch, provides an analysis of the co-adaptive relationship between blogs and search engines.

But his essay misses out on a key insight, one which was teased out by Andrew Orlowski last week in The Register, concerning the power of a small number of A-list bloggers to influence search results for particular phrases. Orlowski has coined this ability to hijack the search results of certain phrases "googlewashing".

In a somewhat muddled article, Orlowski nevertheless chooses a good example -- "second superpower". The first thirty or so Google results for this phrase lead directly or indirectly to a newly created blog by James Moore. How did Moore manage to generate such a high Google ranking with a new blog? He didn't; the A-list bloggers (such as his colleague, Lawrence Lessig) did, because they all linked to him.

I do not suggest (though Orlowski seems to) that A-list bloggers conspire to "googlewash" certain phrases, but I think the phenomenon of search engine results being influenced by a handful of "elite" bloggers is of interest to those (like me) who (sadly) devote time to imagining how the web will evolve.