ResearchBuzz

I know everybody’s talking about Web 2.0, but that’s not the aspect of searching that’s been grabbing my attention lately. The thing that’s been interesting me is what I’ll call, for lack of a better term, results presentation. Results presentation is what the search engine does with the result list and what it allows you to do with the result list.

A search engine like Google is old school; you get the list and that’s it. On the other hand you have search engines which give you page thumbnails, or open the page itself in a small frame, or even allow you to share the results you find with other people. I find that intriguing as engines spend more and more time with what can be done with results in addition to the algorithms that generate the results. Which brings me to Molu the Search Spider.

Molu, at http://www.themolu.com/ , is a search engine that’s currently in alpha and has some Issues. It’s not a search engine I would use on a daily basis because of the issues. On the other hand, the results presentation was interesting enough that I wanted to talk about it and I’ll probably come back and experiment with it a bit.

But let’s talk about the Issues first. Issue first is that I couldn’t get it to work in Firefox. Works fine in Opera, didn’t test it in IE. Second thing is that I don’t know what’s being indexed. If I had to guess it’d be DMOZ or something like that. Third thing, sometimes the search would be a little wonky.

From the front page you can do a search and get suggestions for your search or spelling suggestions for your search. The spelling suggestion worked well, correcting everything I threw at it. There’s also a search suggestions offering and a recent search tag cloud (which was HUGE and hard to navigate around.)

I did a search for spider. There’s no result count (annoying) but each search result offers several options. You can drop down a page frame of each result (or fold it up again when you’re done with it). Search results sometimes have page previews available. Each result is Diggable, and at the bottom of the search results you have the option to extract all the links on the page. A nice touch.

Sometimes I’ll cover a resource not because of what it is but because of what it’s doing and what it makes me think about. Molu didn’t work in Firefox, didn’t give me a search result count, and I can’t tell what it’s spidering. But the way the results are being handled hints at more possibilities. How else can we share results besides Digging them? Could the extracted links be put into an RSS feed, or could links be extracted from each search result’s pages?

Dear Yahoo, I hate you.

I mean, it’s not like I didn’t have anything to do. I’ve been trying to learn the Second Life scripting language, have a whole mad scientist’s lab of experiments I’m doing on Google, and I STILL haven’t figured out what I’m going to give my husband for Valentine’s Day.

And then YOU come along with your Yahoo Pipes product — or, as I’ve renamed it, Yahoo Nerd-Crack Pipes — and now I despair of getting anything done. That little pfft sound you heard was my free time vanishing like a snowball on Mercury.

Still and all, I might as well write this up for ResearchBuzz. I was going to go to sleep — after all, it’s almost 4am — but as soon as I lay down and closed my eyes I thought of a great pipe idea… drat you all.

Bitterly, Tara

I don’t know how recently Yahoo put together Yahoo Pipes — I first read about it at Jeremy Z’s blog — but it looks like an early beta: there’s some documentation missing and a couple of spelling errors. Still, there’s enough here that I’m intrigued and overwhelmed with ideas.

Yahoo Pipes ( http://pipes.yahoo.com ) allows you to query online resources and “pipe” your results through several filters, or take a set of results from one resource and apply it to a set of results on another resource, or aggregate several results. (You’ll need to have a Yahoo account to use this service.) Yahoo Pipes has an initial set of sources you can use, but to a certain extent you can also use any resource that outputs as an RSS feed — which means, thanks to tools like Feed43, almost anything.

You accomplish this by dragging and dropping a series of modules onto a grid and linking them together. There are also submodules you can create and apply them to your pipes. Some of this is intuitive, but a lot of it is not. (It took me a while to wrap my head around how to feed the results of an RSS feed into a search engine and get an output.) There is a limited amount of documentation available, though tutorials are promised eventually. There’s also a fairly active message board.

Oh heck, this is kind of hard to explain. To see what it looks like, take a look at a couple of pipes I made. The first pipe is a simple eBay search, which you can see at http://pipes.yahoo.com/pipes/KjghjVW32xGFoDPPqu5lkA . The second is a pipe that takes the overall list from Yahoo Buzz and runs each item on the list through eBay. That one you can see at http://pipes.yahoo.com/people/fh2U.zoguGp_C_bLDy8YgTw.faUl . The first pipe, the one for eBay, is used as a submodule for the second pipe, the one for running Yahoo Buzz through eBay. (These pipes did not work properly in Opera, and worked fine in Firefox.)

You can view pipes, make copies of pipes and edit them yourself (this is great for learning how some of the functions work, dissembling other people’s pipes) or run the pipes. Running the pipes means generating the results, the output you can get as an RSS feed or as JSON (JavaScript Object Notation).

I have spent a crazy amount of time with Yahoo Pipes in the last several days. My initial enthusiasm was replaced by frustration, which was replaced by the desire to hit my computer with a hammer, which in turn was replaced by enthusiasm again as I did a lot of trial/error learning. I can imagine — I have imagined — literally dozens of search tools I could create with this. I have an idea for a Second Life gadget I’d create using Yahoo Pipes. (More about Second Life in another post.) There were aspects of the Pipes that I kept looking for (if/then was the biggest one, real simple if/then). I hope this is only an early iteration of what could end up being a very exciting tool. That sucks up all my free time while I sit around and crank out search gadgets.

As I noted before Yahoo Pipes doesn’t have any tutorials up yet, but Mr. Speaker has a lovely tutorial available. Thanks Mr. Speaker!

YAY! Clippy’s Dead!

New database to protect consumers in Indiana. “The database, which can be found on www.IndianaConsumer.com, was established as part of National Consumer Protection Week. It allows users to view a searchable list of companies that have faced legal action for violating the Deceptive Consumer Sales Act.”

New digital museum for Richmond, Vermont: http://mylocalhistory.org/ .

The Top 25 Web 2.0 Search Engines. Wow, some nice stuff in here…

Time to search the Web for summer camps. I went to Target the other day for some fingerless gloves (it’s very cold where I work) and they didn’t have any — that season was over. (It was something like 18 degrees outside while they were telling me this.) Meanwhile, there were aisles and aisles of swimsuits. Rggh.

Kodak and Mozilla team up for photo sharing service.

A little more information on Search Wikia.