You're not going to believe this, but a new search engine has just appeared and, well, it may be better than Google.
Obviously, that sounds slightly ridiculous but after having spent a day devising weird and wonderful searches and comparisons, not only has it stood up to the test but it's so good that you realise how much of an effect Google has had on your thinking when it comes to searching the Net.
You can go try it now - it's at Turbo10.com - but for God's sake, before you start emailing and ranting and raving, read the rest of this story as it will probably cover what you're going to say.
First of all, let's get it out the way and slag Turbo10 off. One, it's called Turbo10 - brings to mind something slightly 80s with silver streaks. It's called Turbo because, spokeswoman Megan Hamilton explains "we wanted a name that connoted speed. Also the word 'turbo' has the same meaning across a number of major European languages".
The "10" is vital to the search engine. Megan says: "The 10 is used because we show 10 results per page, we connect up to 10 target engines, and we show the top 10 topic clusters for every search." It's still called Turbo10 though. However, if you remember correctly, everyone thought "Google" was a bit of a silly name when it first appeared in Beta.
Turbo10 is cluttered. It's cluttered and the colours (dark blue and purple) will not be to many people's tastes. A far cry from the liberating white space in Google. Is this a return to the bad old days of over-complex search engines? No, because when AltaVista was king, all those extra bits were useless add-ons and got in the way. Every single bit of Turbo10's page has a very real and very useful function.
The results are tightly packed in and the description of where your search request is in the document is not that great. It sometimes contains a meaningless jumble of words and it seems a little buggy in that it sticks the wrong thing in sometimes. We imagine this will be ironed out as it comes out of Beta.
Turbo10 does not let you use punctuation marks or logical expressions (well, it does, but they have no effect). Now, your immediate reaction is "that's rubbish - how the hell will I get at what I want?" But after playing with the site a bit, you realise that this reaction is due to the Google (and many previous engines) mindset you have for searching the Web.
If you assume for a minute that you can get exactly what you want without having to use the various refining techniques that we have all learnt, then this non-use of logical expressions is a positive thing. Basically, Turbo10 has a different philosophy of Net searching and after a while you start to get it.
Let's give the Turbo10 PR information here. "Turbo10 has launched the first fully automated system that creates and maintains connections to online databases en masse. 'Connecting to 1000 engines is just the tip of the iceberg. By creating an automated system we can connect to thousands more,' Nigel Hamilton, Turbo10's CEO said." That's the start of the company's press release.
The more interesting bit is further down: "Turbo10 searches the Deep Net - a vast array of specialist databases that range from business associations, universities, libraries, and government departments. These specialist search engines are inaccessible to traditional crawler-based engines such as Altavista.com and google.com who can only index static pages. Turbo10 is the first commercial metasearch engine to connect to hundreds of these specialised engines en masse, broadening the depth and range of search results for the online searcher."
That's it in a nutshell. Okay, so now you're thinking "this is just as academic, specialist search engine and it can't even be compared to Google". In fact, it can be compared directly to not only Google but also specialist search engines - plus the ones in between like Teoma.
Turbo10 let's you select up to 10 search engines to run a search through. But unlike AskJeeves where all the results are irritatingly put in different boxes, Turbo10's genius to combine them all in one weighted listing - it's the search engine of search engines. So if you wanted, you could select google.com, google.co.uk and news.google.com to search in.
If you only select these three, Turbo10 will run a search and also choose another seven search engines it sees as best-fits for your search. Then, the results will be displayed according to either speed (Turbo10 doesn't want to be accused of being slow) or relevance, irrespective of search engine, but with the search engine that the result came from shown. Although we note that suspiciously few results come from Google itself.
The reduction of search results from this initial list is then done by the use of clusters in a box on the left-hand side. So, for example, a search on "the register" - two very common words - gives Register.com as the first choice. TheRegister.co.uk only makes it to number eight. However, among the clusters on the left are "news" - which links mostly to news stories and is less help - but also "theregister content" - all of which links to stories on this site. Once you use the clusters a few times, you end up getting to what you're looking for faster than with advanced searches or logical expressions.
So what? Google still gives you everything you need and Turbo10 is too much trouble. Well, quite possibly true - but then Turbo10 has a lot of advantages.
New search mentality
Rather than list results in a long line, causing you to have to scroll, Turbo10 has ten fitted on the page and you use arrow buttons to move onto different pages. You may like this, you may not, it doesn't matter, it's there. But you will notice that there are only spaces for 10 of these pages - so, a total of 100 results for any particular search.
In most cases however, there are only three of four pages - so just 30 or 40 results returned. Here again is where there is a different mindset. While Google tends to go "oooooooooo" right across the page, Turbo10 sticks to its guns and throws out very few.
This is the crux: if you want something obscure in Google and most other engines, you keep changing the search criteria until you get it or you finally find it on page 36. With Turbo10, you simply choose different search engines.
When we first started trying out the engine yesterday, it had 1102 search engines available. At the moment, it links to 1108. But the time you've reached the end of this article, it will be probably be more.
The core of Turbo10 is that you select your own engines. This is all done simply online. You select the ten you want, give the "collection" a name, and stick on your email address (it promises to never sell it) so you can be recognised and then this is added to a list of "collections" that you can select using a drop-down menu on the main search area.
So, basically, you could use the Turbo10 default setting (about, altavista, bbc, dmoz, encyclopaedia, goggle, msn, yahoo) for general Web searches but if you want a news story, you set up a collection of 10 news search engines and call it "news". You then select this one is what you want is news stories. Then, if you want say something technical, you set up another collection with 10 technical search engines.
It's a different way to find what you want, and if you think about it, far more logical than learning different search tricks and techniques. Plus - and this is the real killer - we reckon that aside from all the "how it works" and "what it means" stuff, Turbo10 produces BETTER results for a given search. In fact, after using it for a while, you tend to notice that Google slightly over-rates geeky and sub-culture comments in place of solid, useful links and facts.
Turbo10 has a vision for how search engines will work in the future. Spokeswoman Megan Hamilton confirms that Turbo10 is going for the broad market and so competing with Google. "We want to become the Hotmail of search - personalised to each searcher," she says. We're not sure that is the best comparison to make, but the idea is to use the latest Internet technology to make search engines less impersonal.
Plenty of new features are in the pipeline. "We are going to use Amazon-style recommendation algorithms to further personalise searching. Searchers will be able to view their own search profiles which will also act as bookmarks. Most search engines have no idea who their users are - we want to change this completely - to user-focused search. Each user's search profile will interact with others ... we want to create a blog-o-sphere of search users and their profiles. More browsing options and result annotation will also be included." (Hopefully with this personalised approach, you will be able to change the colours.)
Of course, whether this plan is a truly wonderful and helpful idea or another irritating way for a company to learn too much about you depends entirely on how you view the organisation running it. Turbo10 is brand new and so has yet to build up trust with its users, so we'll hold off until time gives a better perspective.
However, from what we've seen, this looks like a really good and useful search engine. Better than that, it actually provides a different philosophy of running Internet searches. With Google having come to dominate the market (to the extent that people seem to forget that it is a company and not some kind of United Nations for the Internet), the competition and new blood can only be a good thing.
I'm sure I've forgotten lots of bits and bobs but this review is too long already, so get out there and see what you think. Oh, Turbo10 doesn't have the Google cache feature complete with highlighted words, which seems to be becoming more important every day, which is a shame.
Also, please don't bother emailing saying that there's no way this will put an end to Google: of course it won't, Google is Google and is pretty much here to stay. This is an alternative.
Apart from that, our only fear is that it will be knocked offline by the demand that will inevitably follow. ®