ResearchBuzz

What’s the deal with the applications named with an r at the end instead of er? The latest one I’ve found is called Moviemappr, and allows you to find movie film locations on a Google Map. Nice idea, though some locations provide more information than others.

Moviemappr is available at http://moviemappr.com/ . When you get there you can either browse the map or search for a location in the US. The first one I searched for was Wilmington, NC — lots of movies and TV shows get filmed there. I got a Google map with locations picked out, and on the right a list of locations and movie/TV show titles. The locations ranged from very specific (Laney High School) to not specific at all (”Wilmington”. Some cities I looked at had no specific locations at all, just the city name). Clicking on the location takes you to it on the map (or to the center of town if the location is just a city name) while clicking on the show/movie name gives you a popup with artwork from Amazon (how about a link to the IMDB or Rotten Tomatoes?)

If you don’t want to search by location there’s an advanced search option that allows you to search by movie name, and that’s how I discovered that Redneck Zombies was filmed in Delmar, Maryland. This search is amazingly open ended — it includes full words and parts of words. Do a search for zombie and look at how many of the results cluster in DC, Michigan, and California.

I’ve mentioned ZapTXT in my RSS talks as a way to get RSS feed updates by e-mail if you just can’t stand the thought of using an RSS feed reader. Recently I heard from one of the folks from ZapTXT, pointing out that their monitoring service has expanded far beyond e-mail updates.

ZapTXT is available at Http://www.zaptxt.com. You will have to create an account to use ZapTXT, but that’s free and doesn’t ask for much besides the information required to contact you (e-mail, IM, and mobile #. Obviously if you don’t have all those you don’t have to fill them in.)

ZapTXT in its simplest form, and the way I’ve been recommending it, allows you to monitor a single feed for keywords you specify, and get the results by mail. However, ZapTXT also allows you to get results by instant messenger or mobile device. Furthermore, ZapTXT allows you to monitor entire groups of RSS feeds, instead of one at a time.

Once you’re logged in, ZapTXT gives you the option of creating your own “ZapTask” or using the “Quick ZapTasks” panel. I decided I wanted to monitor gadget blogs for upcoming news on the Nintendo Wii. I clicked “Gadgets” entered the keyword “Wii”, and chose from a checklist of gadget blogs to monitor, including Lifehacker, Gizmodo, and Engadget. Choose where you want your updates sent (e-mail, IM, or mobile, or any mix of the three). Clicking next prompts you to name and tag your ZapTask, as well as specify how often you want to receive alerts (as frequently as they appear, once an hour, or once a day.) Later you can go back and edit the ZapTask to add some feeds you want to monitor.

(If you just want to get all the contents of a feed or feeds, leave the keyword field blank. However bear in mind that this can lead to a lot of mail/IMs/mobile messages.)

There is apparently no limit to the number of feeds you can monitor with a ZapTask — woo hoo! (And even better, if you manually set up a ZapTask and use the advanced button, you can import an OPML file. And if you export a ZapTask, you can export the URLs into an OPML file. >swoon<) With that in mind, I would ask this: “how specific should I get with my query, considering that some RSS feeds provide only summaries and some RSS feeds provide full text?” Words like Wii, being unique, are probably okay, but if I wanted to monitor a topic, I’d probably do some experimenting with my searches. (Once you’ve set up a ZapTask, they’re all available in a control panel, and easy to edit/turn off/delete, etc.)

If you’re looking for a simple way to get RSS feeds by e-mail, ZapTXT is usable. But where it really shines is when you want to monitor controlled lists of URLs for specific keywords and get your results in a variety of formats. And that OPML support is a really nice bonus.

I sat down with Gary Price for a little while during Web Search University and got a tour of his new cell phone (Gary is the king of cutting-edge cell phones) as well as a demonstration of some of Ask’s “Smart Answers” features, brought to you via RSS.

If you’ve used the search engine at Ask.com at all, you’ve probably noticed that some of your searches bring results at the top of the page, above the advertisements and the regular Web search results. Do a search for George Washington, for example, and you’ll get a picture, a brief description, and a pointer to more information.

There are also results based on RSS feeds. If you do a search for “search engine watch”, you’ll get as the first results RSS feed results from the last three Search Engine Watch stories. What I didn’t realize is that Ask.com is using the RSS feeds to expand the kind of information you can get. For example, if you do a search for Hurricane Warnings, you’ll get both general information and pointers to hurricane information, but also a feed from the national hurricane center. Naming a specific hurricane, however (”Hurricane Helene”) brings you slightly different results; pointers to information about that hurricane as well as the latest information on Helene. (And defunct hurricanes, like Fran, merely point you to an encyclopedia article.)

The interesting thing about this setup is how customizable it was. Gary noted that they could customize the Smart Answers based on current events, evolving names and keywords, and so on. That seems to have a limited effect, however; searching for spinach recall didn’t bring me information from either current news results or CPSC about the spinach recall. However, searching for cpsc recalls did give me the CPSC’s current feed.

(By the way, it’s AMAZING how much word order effects the results in Ask. Do a search for spinach recall and recall spinach. Good grief.)

This is very cool, but I’d like to challenge Ask to go a little further and start using keyword-based feeds in addition to the static feeds it seems to be exclusively using. Take for example the keyword autism. If you run a search for that you’ll get a Wikipedia result as well as some way to narrow and expand that search. Based on the Expand Your Search options, Ask seems pretty aware that autism is a medically-related query.

That being the case, why not set up something that automatically scans HubMed or PubMed and gets the results as an RSS feed that’s put at the top of the search results? So if I ran an Ask.com query for autism I would get the Wikipedia pointer, a list of the latest HubMed articles containing the word autism, and then Web results. Or maybe I get HubMed information and the latest news results. It’d be adding a little bit of metasearch to the Ask.com engine, except the external results you’d be getting would be much fresher, and if you stuck to certain kinds of sources (news search, PubMed, etc.) much more authoritative.

This posting originated at ResearchBuzz.