GazoPa Answers is at http://answers.gazopa.com/. As you might imagine it’s yet another answers site, but it’s based on answering questions about photos/images or which have to do with photos/images. I remember a site that basically helped you identify things by uploading images of them and asking for help, but the name of that site is slipping my mind at the moment. Anyway GazoPa Answers is for asking all kinds of questions about an image, not merely “What is it.”
There is a page for the various categories of questions, though the front page just gives you a bunch of questions and popular tags. Being so new there aren’t many questions, what I saw ran the gamut, including “Who is the best Thundercat?,” “What does this mean on map?” (along with a symbol), and “How can we make this?” along with an image of Saturn Peaches (also known as donut peaches.)
Questions have their own individual pages which show a larger version of the image, any answers (all the questions I looked at had either no answers or just one), and tags. The GazoPa technology was also used to show images similar to the one which was being asked about, which in the case of this map symbol was partially accurate and partially hilarious.
I can easily imagine using something like this, but there’s not much to evaluate at the moment. You can share the questions on Facebook, but in a pretty basic way — when I tried it it doesn’t even include the images. I didn’t see an easy way to share via Twitter. GazoPa images needs more ways to get the questions out there.
It’s kind of funny that I actually found out about Facebook’s new Q&A service via bemused comments on Twitter. Those actually popped up on my radar a bit before Facebook’s actual announcement —
Facebook Questions (in beta) is a new part of Facebook that allows you to ask questions of everyone on Facebook. Yup. Everyone. And if you already have concerns about Facebook’s privacy issues, don’t use this new feature, because any questions you ask will, by default, be public and available to anyone on Facebook. (If you have a question like Should I be worried about this rash?, save it for a status update on your wall.)
The questions feature lives in a menu selection on the left part of your Facebook home page. I decided to test the service by asking a question I really needed answered — the best antivirus/Internet Security for Windows 7, 64-bit. I didn’t find an easy way to specify what category I wanted to use — Facebook seemed to pick the category itself based on keywords in my question.
I asked my question, and then a day or so later tried to go back to the question’s page to get a screenshot for this writeup. No good. I can’t get to the question’s permanent page, and I can’t even get to the answers. (So the questions will be open and public, for a given value of open, public, pageload, and if-I-see-that-spinny-cursor-thing-one-more-time-I’m-gonna-cuss.) Fortunately the answers are also e-mailed to you.
According to my e-mail the question got five answers. The first was from a friend on Facebook, the others were from people I don’t know. There was no snark, just honest, polite opinion. I guess I was expecting something like the anarchy of Yahoo Answers (which does have good content, but also has a lot of answerbombing) and instead got something closer to LinkedIn or Ask Metafilter (my gold standards for question-and-answer communities.)
The problem of course is that Facebook makes it hard to get to the content. I should not have had to dig into my e-mail’s trashcan to get to these great responses. I also can’t browse other questions. I go to the Questions part of my Facebook page. It says there are three questions about Ubuntu, so I try to browse them. Facebook pulls away the football — there are actually no questions about Ubuntu. But there are 34 questions about computers, Facebook notes. So I click on that category. NIX, saith Facebook. There are no questions about computers. But there are 38 questions about sports…
At that point I gave up. From what I could tell from the answers I got, Facebook has a community that’s ready to be helpful in answering questions. Sadly Facebook’s Q&A service has an infrastructure that’s ready to give me an ulcer.
Update: Wednesday — I got Facebook to show me an individual question page! Here’s what it looks like:
(It is my understanding that currently these questions are viewable only with in Facebook, so I had concerns about showing user names and avatars outside the confines of that community. So I blurred full names and avatars. I apologize if it seems excessive, but I figure with an issue of privacy it’s better to do too much to protect it than too little. All these answers were wonderful so if you’re one of the ones who left them leave me a comment and I will give you link love.)
As you can see, Facebook shows the answers and gives you the option to vote them up or down. You can follow a question to see new answers as they’re added, and if a question is offensive you can report it, of course. And if you’re bored with a question you can explore one of the other ones that Facebook helpfully provides you on the answer page.
This is pretty basic, but I like the voting. I just wish I had gotten to explore this page more when I was doing the writeup!
Ask.com formerly Ask Jeeves, formerly all kinds of different things, announced yesterday the new Ask.com, which is currently available in “public beta.” I put that in quotes because the mention of public beta in the announcement is followed by “Now available on an invite-only basis (you can request your invite here), the capability to pose questions to real people is now possible for those complex, subjective and/or time-sensitive queries that, no matter how advanced, computers simply can’t address.” So I guess the redesign is public beta with the ability to play with the actual, y’know, NEW stuff, is available on an invitation-basis.
I must confess that when I first read this announcement my immediate reaction was the overwhelming desire to slag Ask.com. And I do not normally slag. But Ask.com has been goofing around with revamps and changes and overhauls and etc for years, and strangely, it seems like nothing substantive ever happens. And this invite-only foldelrol for the new community question feature. Dear Ask.com — lose the invite-only status as soon as possible, because if you don’t do something quick nobody is going to care. Nobody NEEDS to care anymore about asking questions at Ask. They’re too busy using Quora and Aardvark and Facebook and LinkedIn and even Twitter to ask questions. Open up and quick, or it’s not going to matter at all.
And finish what you start! Over a year ago — a year ago — I wrote about Ask.com’s announcement of its AnswerFarm technology, which was supposed to be a return to its question-answering roots. At that time the new technology had 300 million question and answer pairs. Today Ask.com is touting 500 million pairs for what I assume is the same technology. And then a few months ago, I wrote about Ask.com’s related questions, which weren’t too bad except for a some inappropriate words. It just seems like Ask announces these similar things over and over again, but nothing actually clicks into place, or makes a dent, or changes perspectives, or happens.
Of course despite the snarkiness here I did ask for an invite to the new community question service (I’ve been writing about Ask for over a dozen years, and I’ll probably continue until it either shuts down or vanishes from the search mainstream ala LookSmart.) And because I think that the evolution of searching — or rather, finding — depends on integrating human knowledge and intelligence with vast oceans of data, I hope that Ask’s Q&A service survives to contribute to the social search evolution.
But please, Ask, get on with it. And keep it up.
This is a search engine I haven’t mentioned in a while. Ask announced yesterday Related Questions, where you can ask a question and get suggestions for additional questions which will help your research.
I started with Why is the sky blue? I got an answer at the top of the result page, but the related questions on the right were downright weird. They included “Why Is the World Going to End in 2012″, “Why Do Dogs Eat Grass” and “Do Spiders Have Good Eyesight”. Um, okay. I tried for something more obvious: When was Pearl Harbor? This time I got the answer at the top of the page and related questions that were more focused, like “When Was Pearl Harbor Bombed” and “What Year Was Pearl Harbor Built” There were related searches as well. Clicking on a related question gives you a result and another set of related searches and questions. (Confidential to Ask: You may wish to vet your related questions more; “Why Did the Japs Attack Pearl Harbor” contains an inappropriate pejorative.)
I tried to go a little further afield with the more general question How do you cure a cold? I had mixed results with this one. On one hand I got questions like “How Do You Get Rid of a Cold,” which is great, but then I got “How to Treat a Dog with a Cold” and the completely perplexing “How Do You Season a Cast Iron Skillet”. Um, what?
Related questions is most useful when you use very specific, targeted questions. Otherwise be prepared for a bit of surrealism…
If you have kids or like “Ripley’s Believe it Or Not” You’ll enjoy Life’s Little Mysteries, a Web site that answers all those random questions that you constantly wonder about. It’s available at http://www.lifeslittlemysteries.com/. It’s a terrific reference site, though it does in my opinion suffer from a lack of sourcing.
Life’s Little Mysteries is sort of like a question engine but human-powered, with extensive articles to answer the questions. So the data pool isn’t particularly wide, but it is pretty deep. The front page divides its answered questions into sections, like Body & Mind, Animals, and Just Plain Strange. Ancient mysteries (“Do Fish Sleep?”) exist side-by-side with more contemporary questions (“How Are Oil Spills Cleaned?”)
There is a keyword search available; a search for baseball found six results, including “Why Is Baseball Spring Training in Both Florida and Arizona?” and “Are Left-Handed People Smarter?” (the keyword search is a full-text search, so you will find questions that are far afield.) Questions are answered by articles that range from a couple hundred to several hundred words; terrific detail. I was surprised to see that while many articles had links, some that didn’t didn’t offer much sourcing. For example, “Why Is Baseball Spring Training in Both Florida and Arizona?” offers several quotes and dates, none of them are sourced in the article that I can see.
I bring this up because of the quality of the site wranglers, which you can read about here. Several of the staffers have journalism or science degrees, and that’s one of the reasons I decided to cover this resource. So I know research has been done to create these articles, so why not source it?
Though the site has just launched, it looks like there have been articles added since at least March, so there’s plenty to see here. I just wish there was more bibliography available.
Do you remember when natural language search engines (sometimes called “question engines”) were hot? It was about twelve years ago. Ask Jeeves was all that. Everybody wanted to be able to search in plain English (or Spanish or French or Esperanto or whatever their preference was.) I remember reviewing a search engine called Electric Monk and quite liking it.
But natural language searching went somewhat out of fashion, replaced by the more practical cluster searching, and by Google, which is actually not a bad natural language search engine but which never made a big deal out of it.
Fast forward to 2010, and question engines are hot again, only this time they’re human-powered. I can’t prove this, but I suspect that it’s because people have gotten used to almost real-time interaction in social networks; toss out a question on your Twitter feed and according to the question and your followers, you could get dozens of replies within an hour. Add that to the ongoing presence of deliberately structured sites like Yahoo Answers and the Q&A feature on LinkedIn, and people became blasé about the idea of hundreds (thousands, more) of people hanging around a Web site, ready to answer whatever question you wanted to holler into the void.
In February I reviewed Aardvark (which had been acquired by Google), and now I am hearing about a new Q&A engine called Quora, which is available at http://www.quora.com/. Well, it’s extant at Quora.com, but not yet available; you can leave your e-mail address for an invite.
According to a recent article in TechCrunch, the beta is a hot ticket, so don’t hold your breath. Also according to the article, Quora was founded by ex-Facebook employees and has gotten a recent round of funding. Check out the article for a brief interview with a couple of the founders and one investor.
I think the popularity of Q&A services like this lie at the intersection of real-time searching — how 2001, to post a question on a site and get an answer back in a couple of days! — and researching in a mobile situation, where deep research is generally not necessary but where you might have lots and lots of small questions (“Where can I find parking in Savannah, Georgia?” “Are there any traffic accidents on I-20?” “What happens if I accidentally drive into the ocean?”) It seems to me that successful development will require a good sense of community, the ability to deliver answers a lot of places very quickly, and potentially integration into other social systems where a lot of data is flying around right now — Twitter, maybe, or LinkedIn.
Definitely something to watch.
Oh, Ask, formerly Ask Jeeves, formerly “Of course it makes perfect sense to advertise a search engine by sticking labels on apples” — what a long, strange trip it’s been.
(Actually you can’t blame Ask TOO much for the advertising-by-fruit strategy: it was, after all, 1999.)
Long ago, Ask got its start as a natural language search engine. That meant that you could type in a question and Ask would answer it, if the question was in its database. Then Ask got into being a more standard search engine, and was doing some innovative stuff for a while after its purchase of Teoma in 2001. Now its made a full circle with an announcement last month about its AnswerFarm technology.
AnswerFarm allows Ask to crawl the Web and pull question/answer pairs from all kinds of courses. Now their database has more than 300 million question/answer pairs. I’m not sure if this would be a natural language search engine, but Ask has certainly circled back to being a question-answering engine.
You can ask Ask some questions at http://www.ask.com/?tool=ans&o=10182&l=dir. Ask handled my standard (Why is the sky blue?) and not-so-standard (How many gills are in a pint?) questions with insouciance; however it gave me several different answers to one question (How do I make cucumbers in vinegar?) and confessed complete ignorance to other (How do I get cat hair off my shirt?)
When Ask CAN handle the question, you’ll see them listed at the top of the page, with a page summary that includes the answer and a link to the page itself. If there are more answers you can get a link to an entire Ask.com page of them. In the case of ambiguous questions (“What is 42?”) you can get wide-ranging answers (See http://www.ask.com/q/what-is-42%3F for more details.)
When Ask CAN’T handle the question, you’ll get suggestions for other questions — if Ask can suggest any — and aside from that you’ll get the option to search the Web, images, or news for your question.
It’s been many years since I’ve used it, of course, but this doesn’t feel like Ask’s old natural language technology, in either presentation or execution. The answers being included in a page snippet in the search result makes this a good resource for quick-hit questions that can be answered in a paragraph. And if the site has gone from 100 million question/answer pairs to over 300 million in a few months, I’d really like to see what this resource looks like in a year.