Last week I took a look at searchme, a very slick-looking visual search engine. Since then, I’ve used it a handful of times and found it to be fairly useful. Though the results are not as good as google, searchme definitely solves a common problem I run into when searching.
No That’s Not It
Let’s pretend for a second. Let’s play make-believe.
<aside>
I think it’s always good to play make-believe at least once a day. As kids, we spend a lot of time pretending, but as adults, we almost never do. And honestly, hasn’t every great invention started with "What if?" or "Imagine if…"
</aside>
So, let’s pretend we are working on a web site. And for the life of us, we can’t remember how to, using php, automagically set the site root in a url. We don’t want to use any absolute links, and we don’t want to set a base href, but we want to make sure our urls get all the way back to the root. Come on. You know what I’m talking about. You’ve been there.
So, we decide to go searching. We fire up trusty ol’ google, and try our best to properly describe what’s going on. We get a page full of search results:
That first one sure sounds sexy, doesn’t it? So you click and………
WTF?
Without reading this page, I immediately know my answer will not be found here. The decision to leave the page and head back to the search results happens in less than a second. Faster than a blink. Quicker than something universally recognized as speedy.
Which is why something like searchme can be very useful. Let’s do the same search on searchme:
See the slider there? I can just slide on over to see screenshots of the results pages.
It removes a step. Nice.
Questions
So, it’s fairly easy to see that a nice, slick, visual search engine has a use. But there are some remaining questions that need to be answered before I start using searchme seriously:
1. Relevance- The first hurdle is relevance, and searchme admits that they’re working on it. They’ve decided to test out the UI semi-privately before perfecting their relevance engine. Search is a tough market to be in. The bar is very, very high. If you can offer a great, easy to use interface ontop of a great engine, you’ve got something. If you just put lipstick on a pig, you do not. Will searchme create a great search engine that happens to be beautiful? Or will they create a beautiful thing that happens to be a poor search engine?
2. Frequency of updates- When I look at blogs in the searchme interface, I see old screencaps. The visual results are old. Unless searchme is able to ramp up their update frequency, the value of seeing the page drops to nothing.
3. Anything new- Finally, searchme has to offer something new, and they’re trying. They bring up category based filters to help give further context to the search query. For instance, searching for blogstring will bring up blogs, marketing and advertising. Clicking on blogs will filter the results that are closer to the "blog" category than the results in the "marketing" category.
If searchme isn’t able to add something new (other than pretty search), they could basically be replaced by a plugin.
The verdict: I like what they’ve done so far. The interface is all cover-flowy, and shiny and black. I like that. And they let me filter out the results I know I’m not going to like. All good stuff, but it all comes down to the quality of their results. We shall see, my friends. We shall see.