Search Engines: Is Google Building on Shaky Foundations?
I'm sure that there must be some sort of wise old saying along the lines of 'the loudest noise starts with the quietest whisper'. This is something that I'm experiencing a lot at the moment - as I spend time on the Internet, or teaching, or even just talking to people I'm beginning to hear a little whisper along the lines of 'What's up with Google these days?' This is not a line from the experts - it's just ordinary people who are beginning to wonder if Google is... well - running out of steam. To be fair, and to abuse my own metaphor, it's not entirely a whisper, and in fact more and more people are raising the issue with every passing day. This really isn't anything new, and in fact I wrote about it in Ariadne back in April 2004 , but what I think is slightly different is that people are less surprised these days when Google does something that doesn't work; there is almost an expectation that when something new is released it's going to be less than perfect, or that there are other, better resources available.
In this article I'm going to look at a few of the things that Google has released over the course of this year and see how they have been received and what, if any, other resources there are that either do a better job than Google, or from which Google should be learning.
The Google Approach to Product Development
There is no doubt that Google has been quite prolific in 2006, releasing different utilities on a very regular basis. We've had the news archive search function, the image labeller game, the ability to download the full text of out-of-copyright books, webmaster central, music trends, the addition of malware warnings - and that is only for August and September!
Clearly Google is doing exactly what it promised; in July Marissa Mayer said 'We believe that we should be launching more products than what will ultimately become phenomenally popular. The way you find really successful new innovation is to release five things and hope that one or two of them really take off.'  This tends to ring alarm bells for me, and I may be guilty of misreading her comments, she seems to be saying that Google is basically going to throw a lot of stuff onto the wall and see what sticks. In fact, later in the same piece she says"We anticipate that we're going to throw out a lot of products. People won't be able to remember them all, but they will remember the ones that really matter and the ones that have a lot of user potential." This is all well and good, but again, I find it somewhat worrying. It seems to imply that Google doesn't really know their market - they're not sure what will work and what won't, so they're just trying everything. It's an interesting marketing strategy - most companies do extensive market research before releasing something, but I suppose if a company has as much money as Google does they can afford to take a different approach. My concern at this point is rather less with Google, and rather more with the end user. People use products for a variety of different reasons, but generally in both the hope and expectation that the product will continue to be improved and supported. Of course, if something is in a beta release it's necessary to accept that things might not work as well as they should, or that it may break now and then, and that is perfectly reasonable. However, given that Google keeps products in beta mode for a very long time they are clearly demonstrating that they have a different view and understanding of exactly what 'beta' means. It is an unfortunate fact of using Google-based products that users will not always get a good service; possibly because their development teams are expending their energies running around producing new products rather than improving and developing existing ones. It was only in August of this year that Google paradoxically moved Blogger (the weblog utility that they purchased) into beta mode as they added new features such as categories. The worrying point here of course is that they are already way behind the times; the ability to categorise weblog posts has been around for a long time, and on a personal note it's the reason why I finally moved my weblog from Blogger to Typepad, over a year ago.
To take one other example at this point, if you run a search for the phrase 'Google is broken' you'll probably get something in the region of 10,500 hits. If you then limit your results to the last 3 months there are 10,100 hits, for the last 6 months 10,200 hits and for the last year there are 9,980 hits. I'm not entirely sure how you can have more results in the last 3 months than you can for the entire year, but clearly this functionality is unreliable, at best. The ability for search syntax to work within Google also depends on the user taking into upper and lower case - a search for Link:www.ariadne.ac.uk provides a very different figure to the search link:www.ariadne.ac.uk and these are absolute nuts and bolts basic errors.
The concern for the user is therefore the lack of certainty that a product will actually develop over time, and one of the concerns for Google should be that because of this uncertainty there will be less take-up of their products than they might hope for. Although the search giant is still by far and away the most popular search engine on the Internet - statistics continually show that it commands the majority of searches in the United States  and the UK  for example - studies are also illustrating that for the majority of their other products they are in fact running quite far behind their major competitors.
Google Search Offerings and Other Options
There is no doubt that in the area of search Google is still number 1 by a country mile - that's not in dispute at all, as the previously mentioned figures illustrate. However, just because something is popular doesn't mean that it's always the best in its class, and Google could actually learn rather a lot from its competitors. When I run a search I may simply need to obtain the information as a one-off, but quite commonly I want to run that search on a regular basis to see if any new Web sites are emerging in that area. MSN Search  has a very nice little function hidden away at the bottom of the results page - RSS. By adding that RSS feed to my news aggregator Bloglines  I get an update every hour of new Web pages that the engine has found for me based on the search strategy that I've used. This allows me to follow changes to the top results on an almost instant basis. This is a powerful and effective current awareness tool and one that I use a great deal - I have a lot of searches saved and they are almost constantly running in the background for me.
Now admittedly I could get something similar using Google News , with their news alert function, so I can keep up to date via email or RSS, but Google News is something of a blunt instrument at the best of times. I wrote about Accoona in my last Ariadne column  and when it comes to targetting the news that I'm really interested in, Accoona provides me with far superior methods of doing this than I get with Google.
The Yahoo! Mindset  is a research project that Yahoo! has had ticking away in the background for over a year now. Essentially searchers can run a search and then by using a slider bar rearrange the results to emphasise either the research or shopping angles of a subject. Consequently a search for 'mobile phones' allows searchers to focus more on sites that are offering to sell them mobile phones, or sites that are looking at say, the dangers of using them. MSN Search, with the Search builder application, allows users to re-rank search results based on freshness and popularity for example. Exalead  allows us to re-rank by date. Accoona  also lets users emphasise particular aspects of a search to highlight a particular search term. With Google we have 'Google sinkers' where by users can repeat a search term several times to change the emphasis of a term and to affect the order of results. Try a search for 'football' and then try a search for 'football football' and see what I mean. I'd like to think this is a well thought-out strategy from Google, but I suspect it's nothing more than a happy accident.
There are times when we all have to search for information on a subject that we don't know terribly well, and as such we're fairly limited in the terms that we can use. I wanted to find out more about the Zoroastrian religion recently and a search for the term on Google gave me a list of pages that I could look at, but wasn't overly helpful. A search for 'define:zoroastrian' was rather more helpful, since at that point Google did suggest some other terms that were related. However, since there were only 3 offerings, I was still fairly much in the dark. However, moving across to Ask  and running the same search I was presented with a listing of 14 different searches that I could run to narrow down and focus the search. Admittedly in this instance I didn't get any suggestions for running a broader search, but that's quite unusual.
If I want a quick view of a page then I'm going to turn to Exalead for their thumbnail approach, and that's also the search engine that I'll use if my spelling is really bad, since their option for phonetic spelling works extremely well.
Google Utilities and Other Options
Google does of course provide users with many other resources, not just basic search. However, in these areas it is also finding it quite hard to make any sort of impression. The ability to post and to search for video footage is swiftly becoming a very popular tool; we're now very long gone from the days when we could really only consider searching for text or images. Google does have a video search function , and is promoting it very heavily - the function is now linked from the home page, relegating Froogle down and into the 'more' menu popup. However, when looking at the usage of the service in comparison to the competition as highlighted by Hitwise , Google video is lagging behind all the other major providers. Youtube  is probably the most popular utility at the moment, but speaking personally I have always found that Yahoo! Video Search  has always given me good solid results that are absolutely on topic. The same can also be said of image searching in fact: although Google has a larger image database  than Yahoo! , the search results always seem to be just that slightly bit more accurate, although this could always simply be my own perspective. Since this article was first written Google has purchased YouTube.
After an initial popular start with Gmail when people were apparently offering to swop a car for an invite to use the system (this may or may not be true, but wouldn't surprise me), Hotmail  and Yahoo! mail  are still proving more popular with 52% and 23% respectively in comparison to Gmail's 2.2%  Many people were concerned that their mail was being read, in order for Google to serve appropriate advertising to them and while it's true that users do get adverts based on the content of emails, the company is not employing thousands of people to sit and read users correspondence. The incorporation of Google Talk  into the system also failed to arouse a great deal of interest, only claiming a user base of 3.4 million in comparison with 204 million using MSN Messenger .
Over the summer Google introduced the Shakespeare collection of works  as part of their Book Search service. The theory was that people could download copies of all of Shakepeare's plays, but unfortunately this proved to be incorrect due to perceived copyright restrictions. Ironically the UK was one of the countries that was blocked, and in most instances it was not possible to download copies of all the plays, just a small selection of them. In comparison, the Internet Shakespeare Editions site  among many others provided a far superior resource and makes the Google offering look amateurish and poorly thought out. Moreover, Clusty has produced Shakespeare Searched  which has been very well received by most commentators with its clear interface, flexibility and search functionality - all things lacking in the Google version. More recently the Google News Archive Search  has proved popular with users, though it has been pointed out that since many of the resources are only available under commercial arrangements it is not as useful as it might otherwise have been. As Gary Price was quick to point out , libraries have the same type of information and have been offering it free for a very long time.
The list goes on and on, very depressingly for Google. Their blog search option  has never really put their competitors such as Technorati  under any pressure. The Google notebook, Google calendar and Google spreadsheet options were all launched with great fanfare, but after the initial burst of interest and curiosity they have failed to make much headway at all.
Industry pundits and observers always look at new Google releases or updates with keen interest; it would be stupid to do anything else. However, when it comes down to practical options and choices the number of times that I, as a trainer, recommend Google is becoming more and more infrequent. Google doesn't appear to be overly concerned with this at the moment - Marissa Meyer is prepared for a failure rate of products running at somewhere between 60%-80% . As long as Google can sustain that level of failure they can continue to do pretty much whatever they please. However, I think that the danger is that as more people are continually disappointed by Google's performance, the more they will look at the company overall, and this will include a re-evaluation of their Internet search flagship. A good example of exactly this can be found in Rafe Needleman's article 'It's time to reconsider Google' with his quote "Google is good, but it's not worth unyielding devotion.".
The fact that 'limit by date' still hasn't been fixed becomes less of a whispered 'Google is broken' and a rather louder shout of 'Let's go somewhere else'. There are plenty of companies, organisations and utilities already filling the gaps. A clear example of this is the weblog entry from Sarah Houghton-Jan entitled 'Ten reasons librarians should use Ask.com instead of Google.' . Reasons include less advertising, better functionality and most importantly - a better search experience. If this doesn't concern Google then nothing will. Of course, none of this means that Google is going to disappear over night - or at all, since so many people like the simple-to-use interface and approach; but, just maybe, it's illustrating the fact that it's not the invincible monolith some people think that it is.
- Phil Bradley. Search Engines: the Google Backlash, April 2004, Ariadne Issue 39
- Inside Google's New-Product Process, 30 June 2006, BusinessWeek online, Technology
- Google continues to be the biggest search engine in the US, 27 May 2006, Pandia Search Engine News
- Heather Hopkins. MSN and Yahoo! Search Tied for Second in UK, 1 June 2006, Hitwise UK
- MSN Search http://search.msn.com/
- Bloglines http://www.bloglines.com/
- Google News http://news.google.com/
- Phil Bradley Accoona: Super-charged Super Target Searching, April 2006, Ariadne Issue 4
- Yahoo! Mindset http://mindset.research.yahoo.com/
- Exalead http://www.exalead.com/
- Accoona http://www.accoona.com/
- Ask http://www.ask.com/
- Google video search http://video.google.com/
- Bill Tancer Google Video, YouTube and MySpace Vids, 16 August 2006, Hitwise US
- Youtube http://www.youtube.com/
- Yahoo Video Search http://video.search.yahoo.com/
- Google Image Search http://images.google.com/
- Yahoo! Image Search http://search.yahoo.com/images?&ei=UTF-8&p=
- Hotmail http://www.hotmail.com
- Yahoo! Mail http://mail.yahoo.com
- Google Talk http://www.google.com/talk/
- Google Talk fails to find audience. http://www.siliconvalleysleuth.com/2006/07/google_talk_fai.html
- Shakespeare Collection http://books.google.com/googlebooks/shakespeare/
- Internet Shakespeare Editions http://ise.uvic.ca/index.html
- Shakespeare Searched http://shakespeare.clusty.com/
- Google News Archive Search http://news.google.com/archivesearch/
- Gary Price Google launches news archive search: commentary, 6 September 2006, resourceshelf
- Google Blog Search http://blogsearch.google.com/
- Technorati http://www.technorati.com/
- So much fanfare, so few hits, 10 July 2006, Business Week Online
- Rafe Needleman, It's time to reconsider Google, 12 September 2006, CNET News.com http://news.cnet.co.uk/software/0,39029694,49283482,00.htm
- Sarah Houghton-Jan, Ten reasons librarians should use Ask.com instead of Google. 22 September 2006