New Search Engines in 2006
It’s very easy simply to concentrate on the ‘Big Four’ search engines - Ask, Google, Live and Yahoo, while missing out on what is happening elsewhere. I know that I’m as guilty of that as anyone else and so for this column I thought I would look back over 2006 and see which search engines have come to my attention, what I think of them, and see how well they have actually fared. This is of course by no means a comprehensive list, and I will inevitably have missed out some but I hope I will have caught the main contenders. Finally it’s worth emphasising that my comments and opinions are just that - mine. You may well feel completely different and may well love one that I hate or vice versa, so I would encourage you to visit and try out as many of these search engines as you have time for, and make up your own mind as to their value.
Collarity
Collarity [1] is a social search engine that combines a variety of different types of functionality to produce results in a new and interesting way. The basic concept is very straightforward, and is something that we’ve seen before - searchers begin to type in their search and as they do a little box pops up suggesting appropriate terms for the search - a little like Google Suggest [2] but rather more sophisticated. I found the approach that Collarity is taking to be very intuitive, and, based on a very small sample of searches, extremely accurate.
The bottom line: it’s worth looking at, most especially if you take the few seconds to register and start to personalise your searches.
Healia
Healia [3], described as a personalized consumer health search engine, has a very clear interface (think Google and you’re there). When you’ve run a search you’re rewarded with a dazzling array of information. Firstly you are offered similar searches, more general and more specific, with various suggestions. Worth noting that this is at the top of the results, where you might expect to see the sponsored links (they’re over at the side). We then get straight into the results, with title, key word in context summary, more from this site, and a link to the cached version. On some results we’re also offered an ‘Attributes’ function, that details things such as ‘advanced reading’, ‘privacy policy’ or ‘fast loading’.
Down the left hand side there are a number of filters, allowing you to filter the content, and a few of these I’ve never seen before. There are filters for/about Professionals, Male, Female, Kids, Teens, Seniors, African/Asian/Hispanic/Native heritage, basic/advanced reading, easy to scan, fast loading, for text browsers and interactive tools. There is also a search history for the last 10 searches that have been run.
But wait, there’s more. Running a search for a specific drug pops up a little information box (think quick answers and you’ve got the concept) with more data. I tried ‘aspirin’ and was given information on what it might also be known as, data from MedlinePlus, an image link from Drug Digest, and a quick overview. Within the results I was given tabbed options for ‘All’, ‘Dosage’, ‘Uses’, and ‘Side effects’. There’s also a useful ‘help’ option as well - something far too many search engines are ignoring these days - plus the option of choosing from 3 different font sizes.
Healia was created over a 4-year period under an award from the National Cancer Institute, and while this is obviously a good thing it does allow me to mention my only real criticism of the engine which is the strong US bias to it (as you might already have noticed with one or two of the filters) and this is a real shame. I’d like to see another set of filters for Geographic content for example, but perhaps I’m just being really greedy at this point.
The bottom line: excellent resource and one that I thoroughly recommend.
Huckabuck
Huckabuck [4] is a metasearch interface that takes the words you type into its search box, and queries Google, Yahoo!, and MSN simultaneously to deliver results that are more comprehensive and more relevant than results from a single engine. You can fine-tune it to give more weighting to one search engine, but if you’re going to do that, why not just use that search engine in the first place? Bottom line: seems to be a perfectly fine multi-search engine, but with nothing particularly exciting or innovative to recommend it.
Hot Daddy
Hot Daddy [5] is another Google ‘all-in-one’ search interface. This one is slightly different because it also allows you to search Wikipedia, Digg, Creative Commons, IMDb, AltaVista, Wisenut, MSN, Teoma(?) or Yahoo! I suppose that can be helpful, but you do need to know what all the little icons mean in order to use it quickly (though a mouseover tells you). What is odd though is that once you’ve done a search and inspected the results - you can’t use your back button to return to the interface. Bottom line: useful in that it has a lot of options available on the page in an easy-to- use format, but the inability to go back to the page quickly renders it close to valueless.
iBoogie
iBoogie [6] isn’t bad at all, and it reminds me very much of Clusty, the clustering search engine. Type in your search as normal and get a bunch of results. On the left- hand side you can see a really good collection of categories to narrow down your search to something more useful. There is also the option of adding in your own tabs to include searches on a couple of dozen or so other resources.
Bottom line: worth trying out.
IncyWincy
IncyWincy [7] has spidered and indexed 150,000,000 pages so far. It’s a reasonable search engine, and had found a few things that I’d not come across before, but it doesn’t cluster results, so a test search on my own site turned up a lot of results one after another, which wasn’t helpful.
Bottom line: if you’re looking to search the hidden/invisible/deep Web, this is worth trialling, but be prepared for some very unusual results.
Jatalla
Jatalla [8] is a neat idea - users rank results. You get results and vote (sorry, you lexivote) for a result that works for you, and this pushes it up the rankings. Only… it’s very hard to get anything to work with. My usual egotistical search for “Phil Bradley” produced no results. The solution to this problem is to do another search that does pull up results on which you can then vote, making it more likely that your preferred site does come up next time around. What other searches bring up my name? “Internet Consultant” usually works well - 0 results. “Internet librarian” - 0 results. “CILIP” (UK version of ALA) - 0 results. “Search Engine Watch” - 0 results.
Clearly this approach isn’t working, so let’s just work more broadly - BBC gives me 3 results; not exactly a stunning result. I really dislike being negative, but surely a search engine needs to start somewhere - and to be blunt, what’s in it for the users? I can already get good results, or I can personalise my results, or I can use a search builder to really focus what I’m doing, so where is the value for me in this system?
The bottom line: I think I’ll pass on it and use an engine that gives me some data.
Kahzam
Would you like to go back to the early years of the decade and take a peek at what it was like? Visit Kahzam [9] and do just that - it has a very strong ‘early 2000s’ feel to it - specifically it reminds me of Yahoo’s home page though I’m not sure why. Anyway, Kahzam has little to recommend it to users, I’m afraid. The home page is full of junk, and is very long, with big adverts and horrible little module headline boxes. To be fair, it does offer options for Web, Images, Audio, Reference, News, Directory and Shopping, but it has precious little else going for it.
What made me laugh was that when I tried to do a search for say, “country search engines” it just stripped the double quotation marks out when it loaded the results page, and a voice in my head kept saying ‘Nope, we don’t like double quotes - you’re not having them!’ No advanced search, no help screens - it really doesn’t have much to offer.
The bottom line: in the ‘hit/miss/maybe’ stakes, this is a definite miss, I’m sorry to say.
Kosmix
Kosmix [10] comes in several flavours of vertical search - health, video games, finance, travel, US politics and automobiles. I checked out the health version, and results are available in two formats - ‘classic search’ and ‘smart search’. The latter appeared more useful (funny that), and displayed data in a more helpful style, with links into news items, basic information, support groups, natural remedies, professional content and specific patients. There was a strong US bias, with the option of finding a local hospital via zip code, and links to top US hospitals.
The results were clear, easy to read on the screen, with no distracting advertising, and fairly intuitive. Each vertical engine, while working in the same way, was obviously tailored for the specific subject area, and so the video games engine offered results in categories such as news, hints, reviews, media and what other users were saying. I did however notice advertising in the games and automobiles engines.
The bottom line: worth taking a peek if you have an interest in one of its areas of subject coverage, although I would be surprised if it were to replace your current preferred option.
Mojeek
Mojeek [11] comes in two flavours, a .com and a .co.uk versions. Results for the British version appear in a rather unusual manner - rather than the usual listing down the page, each hit is displayed in its own little box on the screen in a two-column grid, with alternative grey/white backgrounds. I have to say that I found it very confusing, and it was really difficult to make any kind of sense of what I was seeing on the screen. At the top of the screen were two further boxes, one giving information on the search result (scope, approximate matches and so on) and related news stories, which on the searches that I ran appeared wholly unconnected with the subject of the search.
The global version looks much more like we normally expect with search page results. Interestingly there were options to change the ranking method, although they were labelled as ‘one’ and ‘two’ - the ranking did change, but unfortunately it wasn’t obvious (at least to me) as to what algorithm had been used to effect the change, and no indication on the site of how I could find out either. Other interesting things about the search engine were no adverts of any sort and the opportunity to create your own personalised search engine. Unfortunately though this was somewhat spoiled by the fact that you had to contact them directly if you wanted to set this up, and if the designers are hoping to make a success of this search engine that is something that they need to fix really quickly, since there are plenty of other search builders out there now.
The bottom line: not particularly interesting and of very limited value or appeal.
Opsdo
Opsdo [12] is a multi-search engine providing access to over 70 different search engines. Results are displayed on a search engine by search engine basis in a series of small windows, allowing users to move easily from one set of results to another. The search engine makes use of lots of icons and is quite confusing; I found it really hard to navigate my way around it. However, it’s worth using if you want to try some different search engines.
The bottom line: Give Opsdo a try if you have some time available to experiment with it, but be prepared to persevere.
Quintura
Quintura [13] is the latest in the line of visual search engines such as Kartoo [14, Mooter [15] and WebBrain [16]. Quintura basically takes your search term, runs a search and then translates the results into a tag cloud effect on the screen. Users can then simply look at the results (powered by Yahoo) listed under the tag cloud and click on the link as per normal, or they can explore words displayed in the semantic map to focus the query more closely. After the search runs the search terms appear on the screen and are surrounded with other hopefully appropriate terms. My search on ‘search engine watch’ for example returned keyword suggestions such as ‘blog’, ‘forum’, ‘search engines’ and so on. The closer to the search terms, the larger the keyword suggestions (both in terms of font size and bold), the more relevant they are deemed. Holding the mouse over a term - note that you don’t need to click - will display a new set of results in the bottom window and will also show another keyword cloud overlaying the original, which does get a little confusing at times, and it’s quite hard to work out exactly what you’re searching on.
The bottom line: it’s an interesting approach to search, and users who enjoy different approaches to the display of search results will enjoy using it.
Scandoo
Scandoo [17] is an interesting search engine. It’s a bolt on interface to Google, MSN, Yahoo and Ask. Type your search as normal, and you then get taken to the results page from that engine. However, what Scandoo then does is to flag each result with a ggreen tick, red cross or orange question mark to indicate the trustworthiness of the sites in the results. I did a couple of quick searches - it clearly indicated that my site was ok for my ‘phil bradley’ search, but gave a red cross against the porn sites for my namesake. I also tried a search for ‘gay’ and it again correctly identified good informative sites as well as the porn stuff. However, other searches did manage to fool it rather more easily, so it’s not a replacement for common sense.
The bottom line: it’s a good idea, but still keep your wits about you.
SearchMedica
SearchMedica [18] is a search engine that has been specifically designed for GPs (General Practitioners or Doctors) to use. The search interface screen is clean and clear, giving searchers the opportunity to search for their subject limited to medical sites as chosen by doctors, NHS (National Health Service) sites, or the entire web. Once a search has been run the searcher has the option of running it a second time, with the option of further limiting to various sub categories such as evidence, patient information, guidelines or reference. There is also an option to limit results to the UK only. There is a very neat and tidy option to narrow or broaden a search - clicking on the link slides out various suggested approaches. The main results are presented in the usual fashion; numbered, with the title, brief summary with keyword in context and link to the site/page.
The bottom line: quite simply it’s excellent and will become a firm favourite, and not only with medical professionals.
Search Stomper!
SearchStomper [19] styles itself ‘The lean mean searchin’ machine’. Straightforward - choice of websites, images, videos, news, products and local links. Results give some images (regardless of asking for them) first, then the ads, then the data, with thumbnails in most cases. I didn’t find the results particularly impressive, and it missed out on a lot of sites that I’m expecting to see with my test searches.
The bottom line: less of a stomp, more of a tiptoe.
WebFetch
WebFetch [20] is another multi search engine, searching Google, Yahoo!, MSN, Ask, Kelkoo ‘and more’. It’s reasonably good, though I prefer eZ2find [21] myself. To be fair, it has further options for searching images, audio, video, news, business and people. There’s also a ‘Are you looking for’ sidebar, plus a recent searches box, a maps option, weather, horoscope and latest news links.
The bottom line: multi-search engines are ten a penny and this one really doesn’t have anything outstanding to offer.
WebWobot
I’m mentioning WebWobot [22] for the sake of completeness really, and because it was brought to my attention. It is a very basic engine using Google ads for revenue. It can do a global search as well as just within the UK, but there’s nothing yet to particularly recommend it, I’m afraid. Perhaps it’ll improve with more functionality in a few months. However, since it’s already had 4 months and I can’t see anything having changed, I’m not too optimistic.
The bottom line: I really can’t see any value to it at all.
Whonu
This is an interesting search engine - Whonu (which I guess should be read as ‘Who knew?’) [23] It’s basically a meta-/multi-search engine of sorts. Input the search terms that you want, and it’ll produce a bunch of links for you to click on and these will then take you to the appropriate search engine and run the search. Engines included are MSN, Ask, Clusty, Google, Yahoo (and Google/Yahoo images and video).
You can focus your search by choosing one of the icons under the search box to limit your search to people, place, thing, health, word, history, question and so on. While it doesn’t collate the answers for you, the way that most engines do, it does provide some clever short cuts for you to use. The interface is unfortunately very complicated though, and I think it’s going to discourage the vast majority of users right from the very beginning.
Bottom line: worth fighting with the interface for a while because it can bring back useful data that you’d find difficult to obtain elsewhere.
WordIQ
WordIQ [24] is an interesting search engine. It’s really good for words - you can search for definitions, e-books, reference, articles, the Web and dreams (i.e. as in their interpretation). WordIQ worked reasonably well, although some of the links to the actual articles it found just took me back to the home page, which was a bit of a shame. Not bad if you’re looking for words and want a bunch of resources all together in one place.
The bottom line: a very nice niche search engine and worth adding to your bookmarks for just that moment when you need to know all about a particular word.
Yurnet
Yurnet [25] is a nice simple-to-use search engine. Clean and crisp front page, with two boxes - one for the search term and one above it for the type of search you prefer. It has various options, such as Google versus Yahoo! and Google versus MSN (with results from both engines appearing as frames on the same screen). It also provides other options, pulled from various search engines, such as blog search, images, person, school/college, library, Web site and so on. It’s a nice fast way to search.
Bottom line: try it out - you may very well like it.
Zepti
Zepti - Search the Web [26] or not, as the case may be. I’m only mentioning this one because I try to mention all the search engines that I discover. Zepti appears to be an index/directory type of affair, but the options available are a bit strange - ‘bodyart’ gets the same prominence as ‘medicine’, for example. The search option seems only to search their (small) collection of resources, and has next to no value I’m afraid.
Bottom line: Don’t even waste your time going there.
Zoo
Zoo [27] is a search engine designed for children. Actually, it’s not particularly designed for children at all, other than having the funky backdrop to the page, and the use of a simple font. What they really mean is that they’ve provided access to a subset of data from Google, Yahoo and Wikipedia, with news from ABC, Fox and Yahoo.
I can’t say it really does much to attract me. All Zoo has done is impose a blanket block on certain search terms in certain situations, as my own test searches demonstrated. For example, the term ‘sex’ is totally blocked. As is ‘tits’. So much for youngsters who want to do research into biology and bird-watching. I accept that was a cheap shot at the search engine, but just because a searcher happens to be a child it does not mean, per se, that he or she is not going to have serious queries. Completely blocking the use of certain terms is a cheap and unjustifiable measure. All that will happen is that children will shrug their shoulders and move to a search engine that will provide the information that they want. Blocking children’s legitimate searches in this manner just isn’t going to work. Moreover, the engine is not even consistent in this apparent policy: the search functionality is only limited to banning sexual terms, so I could still do searches for ethnically derogative terms for example, without any difficulty.
A further and major source of irritation is that some of the results are ‘sponsored’. I have no problem whatsoever with search engines trying to make money - far from it. However, this really is not the way to do it, because you receive the title of the page, a bit of description and then ‘Sponsored by: <name of sponsor>‘ instead of a URL. As a result it’s not always possible to see exactly where I’m going when I click on the link.
The bottom line: I really cannot see the point of this search engine at all; I certainly wouldn’t recommend it in a school situation!
Conclusions
As you might expect from a sample of search engines based entirely on chronology, it’s rather a mixed bag. Everyone will have their own particular favourites and ones that make them shudder, and they’ll probably be different to mine. Of all of them however, the ones that stand out for me are Healia, because it’s so flexible and has a interesting variety of options that I haven’t seen elsewhere; and Collarity, which, though flawed in places, is very clever.
References
- Collarity http://www.collarity.com/
- Google Suggest http://www.google.com/webhp?complete=1&hl=en
- Healia http://www.healia.com/healia/en/index.jsp?
- Huckabuck http://huckabuck.com/
- Daddy http://www.hotdaddy.com/
- iBoogie http://www.iboogie.com/
- IncyWincy http://www.incywincy.com/
- https://jatalla.com/beta/
- Kahzam http://www.kahzam.com/
- Kosmix http://www.kosmix.com/
- Mojeek http://www.mojeek.com | http://www.mojeek.co.uk
- Opsdo http://www.opsdo.com/
- Quintura http://www.quintura.com/
- Kartoo http://www.kartoo.com/
- Mooter http://www.mooter.com/
- WebBrain http://www.webbrain.com/
- Scandoo http://www.scandoo.com/
- SearchMedica http://www.searchmedica.co.uk/
- SearchStomper http://www.searchstomper.com/
- WebFetch http://www.webfetch.com/
- eZ2find http://www.ez2find.net/
- WebWobot http://www.webwobot.com/
- Whonu http://www.whonu.com/
- WordIQ http://www.wordiq.com/
- Yurnet http://ww2.yurnet.com:8080/
- Zepti http://www.zepti.org/
- Zoo http://www.zoo.com/