Erics Tech Blog » search

Sortfix Or A Next Generation Search Tool

eric — Tue, 16 Feb 2010 14:45:25 +0000

Although SortFix has been around for a while, I hadn’t heard of it until recently. I also didn’t really have a reason to give it a try because the vast majority of my Google searches were providing me with the results I was looking for.

The idea here is that you can do a normal search for something and then you will be presented with a screen that will allow you to narrow down your search using a graphical user interface (GUI). This will produce a search similar to something that a power searcher would do in Google. After doing this a few times, it becomes fairly evident how its done and one could start “power searching” on their own.

Now what’s exciting to me here is the potential. SortFix currently only uses Google as a backend search engine. There is nothing wrong with that considering that most of us only stick to one search engine anyway. However it would be great if SortFix could act as a meta search engine (a search engine that searches multiple other search engines) to do power searches and search the major players using your search terms, pair down the dupes and give you the results. Technically speaking, I am aware that this is no trivial task. But a unified search engine like that would certainly help quite a bit with finding what you’re looking for in a difficult search.

If you haven’t seen SortFix, then I’d head on over there. It’s a concept with some great potential and a cool user friendly searching style.

Social Search’s Effect on SEO

eric — Fri, 12 Feb 2010 16:36:50 +0000

There always seems to be something affecting your SEO rankings. Most recently, people have had issues with personalized search. Another kink in the chain may be social search.

The first thing is that Social Search can’t replace Hyptertext search (such as Google/Yahoo!/Bing/etc). Social search has 3 inherent drawbacks, time to receive information, credibility of the source, and subject/objectivity. Technically subjectivity is an issue with hypertext searches as well, but since the information is considered more permanent, people tend to be a little less opinionated and a little more objective (if the information is factual).

According to a recent NY Times article, the majority of social searches are answered within 10 minutes. In our culture, 10 minutes is a long time to wait. We are all about instant gratification. With regard to credibility, most of the time social searches are requests for opinions, so credibility isn’t an issue. What one thinks of a restaurant or some other establishment is generally an opinion based question.

Ultimately I don’t believe there will be a direct impact on SEO. Social search just aim to fill a different niche. People have a different goal in mind when performing a social search than when they are performing a non-social (hypertext) search.

The one thing that this article in the times neglected to mention is that a lot of the hypertext search results are becoming social. With the hypertext search engines indexing sites like Yelp and TripAdvisor for example, they are already allow social search results in a hypertext search engine. Therefore social search engines will need a wholly different approach to searching to be effective.

Model Specific Formatted Search Results Using Thinking Sphinx

eric — Mon, 08 Feb 2010 14:15:23 +0000

Having recently implemented Thinking Sphinx on one of my web sites, I thought it would be cool to be able to search every indexed model. With Thinking Sphinx, it’s easy to have a bunch of different classes returned in the results. The tougher part is displaying them in a way that is organized (although admittedly not very DRY).

Let’s start out with the controller method for search results (app/controllers/search_controller.rb):

def result
if !params[:model].blank?
@model = params[:model]
else
@model = "ThinkingSphinx"
end
@query = params[:search][:query]

@results = Search.model_search(@model, @query)
end

Also, let’s make sure we define the actual search model. I define it in a separate search library (lib/search.rb). This is the relevant snippet of code:

def self.model_search(model, keywords, var = {})
@search_options = { :page => var[:page] || 1,
:per_page => 15 }

@search_options.merge!( :order => "@relevance DESC",
:sort_mode => :expr,
:sort_by => "@weight * @weight")

model.constantize.search(keywords, @search_options)
end

Assuming that we are working with the all models search, where the above model is ThinkingSphinx, let’s iterate over the search results with this code in the view (app/views/search/results.html.erb):

< % if @results.total_entries > 0 %>
<div id="localLocationList">
<ul class="list">
< % @results.each do |result| %>
<li>
< %= display_search_result(result) %>
li>
< % end %>
ul>
<div class="clear">div>
div>
< % end %>

And the view helper (app/helpers/search_helper.rb):

def display_search_result(result)
eval "render :partial => '#{result.class.to_s.downcase.pluralize}/result',
:locals => { :result => result }"
end

The interesting thing about this bit of code is the eval. The eval on each iteration decides which search result partial to display based on the class and then passes the result to the partial for display. So if the result has a class of Business, the partial app/views/businesses/_result.html.erb will be rendered. This is a quick example of a search result partial:

This is useful because all models don’t have the same characteristics. By creating a search result partial for each model type, this can be reused for consistent looking search results around your webapp. If the search is model specific, the same result partial will be used in every iteration over the results. If the search is model agnostic, then you can display your search results in a consistent manner.

Being Smart is all about Being Resourceful

eric — Mon, 25 Jan 2010 12:45:48 +0000

The internet us the ability to not while keeping up the appearance that we do. Now that’s not to say that you should be a know it all, but you should definitely know how and where to get information if you need it. If you use a specific open source technology at work, then you need to know how to support (because odds are, it was written by a few interested people and doesn’t have a company behind it). So you should know where the forums are, where the documentation is, where the mailing lists and the mailing list archives are, etc. Do they have an IRC channel where you can talk to live users who might be able to help on a more immediate basis? Maybe there was an even a book written that you can get your hands on, a PDF, or even a screencast. If you lucky, you might write a Tweet about your frustration and one of the products creators will answer (which happened to me recently).

The best part about all this is the internet is the enabler. It’s searchable, indexed and provides you many methods of finding the answer to your (or other peoples) questions. Why am I telling you all this obvious information? Because its why I write my blog. People who need help need to be able to search for it. Most of the things I write about here are specific problems with answers that I figured out or aggregated from reading various things around the web. So be resourceful, use what’s out there and then document it. You may save yourself (and maybe me) a lot of time.

Bing! Hunch! Decision Engine!

eric — Wed, 03 Jun 2009 14:48:55 +0000

I know, the title is an awful play on Batman from the 60s, but I thought it was funny, so tough. Anyway, Bing as most of you know is Microsoft‘s attempt to fix search (if you think search is actually broken, but that’s a whole other post).

Bing (which for those of you who don’t know is: Bing Is Not Google) is touting itself as a decision engine. If I understand what a decision engine does correctly, it helps you take a bunch of variables related to the outcome and depending on your feelings about those variables, helps you to get to the end state (a decision).

(overly simplified) example: Should I live in New York City?
Variables: Noise, Transportation, Money
Q 1: Do you mind a lot of noise at night?
A 1: Yes, New York City is fine.
Q 2: Do you like driving everywhere?
A 2: Yes I like driving. New York City is better for people who like mass transportation. Parking and timeliness of movement can be a problem.
Q 3: Do you have the money to live in New York City?
A 3: New York City is one of the most expensive cities in the world to live in. No, I don’t make enough to live in New York City.
Outcome: 2 of 3 answers are contrary to living in New York City. Therefore you should probably not live in New York City.

I don’t see how Bing does this for you.

Enter Hunch. Hunch is an actual decision engine (or a much closer version than Bing). Just to give you an idea of how Hunch works, I decided to ask it whether I should get a netbook or a laptop (even though I know full I need a laptop and I love my MacBook Pro). This is what Hunch did for me.

Using multiple choice questions for everything, Hunch asked me about my usage plans. I told Hunch that I need it for photos, videos, music, etc. Then Hunch asked me how much I would be willing to pay. I need power and I know that comes at a cost, so I told Hunch greater than $1200. Hunch asked me about my travel habits and I said I travel a lot, but I still need power. It asked me about an OS (which I of course said Mac). It asked me about my keyboard size preference, which I prefer a larger keyboard. Finally it asked me whether this would be my primary computer. I said yes. It came up with the suggestion that there is an 85% chance that I should get a MacBook Pro based on my needs. Sounds good to me . Hunch will even tell you why it came to that conclusion (based on your answers).

As an aside (since I am an email administrator by day), I found it interesting that Hunch, when sending their welcome email, sends a vcard to ensure that their email address is properly added to your contacts. And it is located just a few blocks from my office. Small world.

I am really excited for this product to go fully live. I think it is an absolutely outstanding engine and once live will be a great asset to the web (no I am not being paid to say that). I just think that its about time stuff like this happened. Now if they can just get the experts involved for people who want more advanced information…

More Wolframalpha Easter Eggs

eric — Tue, 19 May 2009 02:30:20 +0000

While finally getting to play around with the new Wolframalpha search (computational knowledge) engine. I found a few more Easter Eggs in addition to the ones posted on Mashable here and here. It appears that the creators of WolframAlpha believe that proverbs (more commonly referred to as aphorisms) are a handy trait for a computational knowledge engine to have. Regardless, they are fun to play with.

I should probably listen to this one myself. Starting out with this cause it is just amusing.

All Work and No Play Makes Jack a Dull Boy
Another instance of WolframAlpha being health conscious.

An apple a day Keeps the doctor away
It wants to be smarter so it tries to learn from its mistakes (not really, but we are being proverbial here anyway).

Fool Me Once, Shame On You
In the spirit of the Back To The Future (and other cult film references):

1.21 Gigawatts
Another great proverb reference ignorance.

Out Of Sight, Out Of Mind
This isn’t really an Easter Egg. I was just expecting a response more like, “There is a season” or something a little more 60s ish. So I am including this one purely for amusement’s sake.

Turn Turn Turn