Aug 18
Blinkx or bollokx?
This summer, companies are taking an interest in search, again. Two years ago, no one was seriously talking about competing against Google.
However, we all knew Google’s dominance would eventually be challenged. Yahoo! bought Inktomi (search) and Overture (distributed keyword relevance text advertising), to compete directly in the two areas that have made Google a success. Microsoft has also thrown its weight behind getting search, although many will never trust them to offer non-biased results. Even Amazon has had a go, although its skinned version of Google - A9 - is hardly innovative.
There are new names, though, hoping to find the relatively quick, word-of-mouth driven success that Google has had. IceRocket.com is one of them. It looks like Google and, some of its results are generated by Google. IceRocket brings together a few nice ideas - thumbnails of each search result, a quick view of the site in an iframe, a meta-search of FaceParty style sites - but there are problems that need to be sorted, if it’s to have any real success. The quick view, for example, takes just as long as opening the link in a new tab but gives you only a tiny window through which to see the content. Thumbnails are available for only around 20% of the results returned. And without their own deep crawl of the web, no number of bells and whistles will help them to make a major impact on search users.
If IceRocket is a little “me too”, then Blinkx’s approach, at least, appears to be original. I say “appears” because I’m sure similar, admittedly less ambitious, widgets have been around before.
Blinkx appends a small blue bar to the title bar of certain applications - MS Office, IE, Mozilla/Firefox, some email apps. It then scans the content of the current document and offers what it promises are relevant local files, news stories, web links, blog entries and products.
Here’s the problem: very little of what it has offered me has been relevant. Looking at the press for Blinkx, someone in their PR team is fond of the phrase “Google killer”. That puts a mighty weight of expectation on the service. I’ve been generous to Blinkx: I’ve given it a whole morning. People who don’t write blogs or work for computer magazines, will give it five minutes; that’s if they can be persuaded to install it in the first place.
“Google killer” means that Blinkx will give me better information than Google, in a more effective form. Blinkx sits on top of my window and changes colour very slightly when its found something that might be relevant. Click one of its tiny icons and, depending on which one you’ve stumbled on, it gives you either:
- news
- general web results
- blog entries
- products
- video and audio
- local files.
Products is where Blinkx hopes to make its money and you have to elect to click on it - “please advertise at me”. Admittedly, what its thrown up has been fairly relevant.
News has only been relevant when I’ve been looking on news sites. It’s let down by the first link almost always being to the actual story I’m reading.
Video and audio, apparently one of Blinkx main USPs, has given me nothing. (As an interesting aside, one of Blinkx’s co-founders claims this will mainly throw up BBC content, as “the BBC posts its digital TV free on the internet”, which is news to me.)
Local files has given me nothing. I gather from the Blinkx website that this is because it takes around a day to index your local files. Call me impatient, but nothing else the tool did made me want to keep it on my system that long.
Blogs: my favourite one. It seems that there are seven blogs, which it cycles and none of which were relevant to anything I was looking at.
Web search, the all important one, has not given me a single result.
So, as soon as I’ve published this post, I’ll be uninstalling Blinkx. Google is a success because of its simplicity: you go to Google.co.uk, type in your search, it gives you relevant results. There’s nothing to install, nothing to learn. Blinkx is only for Windows (Mac coming soon), providing they’re using a compatible browser and are happy with very poor results (in my experience). Oh, and it takes hours to become fully usable.
Perhaps Blinkx has a future, if some UI problems are tweaked and they don’t pitch it as a Google killer (they really don’t want to kill Google, when they rely it on its search results). If it was sold as a local search system, that happened to include web searching, users may be prepared to wait a day or two for it to get up and running and to learn how to use it. Competitors to Google should be easier to use, not harder.
Maybe I’m just not using it properly but that’s because of the ways it’s been sold to me. Bad marketing also effectively killed Tivo, in the UK. The advertising spoke of pausing live TV: that’s a cool feature but probably the least important to most people. The Sky + marketing - a rebranded Tivo with a digital satellite decoder built-in - has focused on watching what you want, when you want. It’s been a relative success. So Blinkx, forget killing Google: improve your UI and focus on local search.
I welcome competition to Google: we need it. I just don’t think we’ll find it in a product that has none of the advantages of Google but, instead, has gathered a little too much publicity, through an echo around various blogs, many of whom seem not to have used the thing.
August 20th, 2004 at 12:17 pm
Hear you loud n clear but what you gotta remember is even Google started somwhere…some guys basement in fact! In all fairness, blinkx have stated they arn’t competing wiht Google and it says on their website they don’t aim to replace them. Still, i’m more optimistic myself. I think they have a really unique product that, with some refinement, could really revolutionize the way we work. I’m surprised to hear you’re not impressed with their local search. I’ve actually found it very relevant and am finding myself becoming increasingly reliant on it. All local search utilities take a few hours to index…and for what it has helped me achieve this was certainly worthwhile. I don’t necessarily see blinkx replacing Google, blinkx’s linking idea is different, why not use both…?
August 20th, 2004 at 12:27 pm
I’m happy to admit that I may just not understand what Blinkx is all about. In fact, I pretty much say that in my blog post; the problem is that the *feeling* I get from what’s written about Blinkx is that it’s meant to be a Google killer, whether the Blinkx team expressly make state that aim.
I’ve given Blinkx another go, every now and then, and still haven’t got anything useful from it. I understand it’s early days, though. There’s also the question of whether my expectations match what it’s designed to deliver: a marketing problem. Hopefully, they’ll sort the UI problems, the relevancy of suggested links and then they’ll have a good product.
Several bloggers and even BBC News have written about Blinkx in the context of “the next big thing”. I think Blinkx, if it’s most obvious problems are quickly fixed, has a future as a tool that some people find useful. I’d suggest that, unless I’m really missing something, it ain’t the next big thing.
August 20th, 2004 at 3:53 pm
Fair enough. But what you’ve gotta realize is that you can never control the press. There are plenty of journalists out there who can’t wait to get a sensational story in their paper and the best way to do that is to compare Blinkx to Google, a search engine everyone knows. Every journalist knows how to hook readers with headlines such as ‘Is Blinkx going to kill Google’.
I hope they sort out the UI problems too, and make the links even more relevant, but then, actually, a lot of the time when I’m looking for something on Google it returns lots of irrelevant stuff too. Think in the long run technologies like Blinkx will prevail as they have more potential to be accurate than other search engines (since they are based on understanding concepts and not just keywords). People are getting lazy and the focus is shifting from going out to search for things to having them delivered to you, take internet food shopping for example…brings the supermarket to your home! Not long before people will move away entirely from searching to personalized desktop applications that deliver the information to you.
Blinkx is moving away from the old model of typing in a query and getting returns to linking together data and delivering relevant content to your desktop while you work. It’s a different idea to Google altogether.
August 23rd, 2004 at 2:27 pm
I understand what you say about not being able to control the press.
I’m giving Blinkx more time and it has - finally - started giving me fairly relevant results. I’m still not finding it particularly useful, though. I shall write a blog post with more detail, another time.
I truly hope Blinkx finds its niche. I’m sure there’ll be a hundred and one other services like this, in a year’s time. With any luck, some of them will be good enough that they can make using our computers and the internet much easier. The main thing I hope for, though, is that there are enough of them in competition, so that we don’t have a monopoly provider.