Wednesday, February 21, 2007

Next Search Engine...a Google Killer? -- Part 2

One internet trend that I failed to include in my last blog was Wiki, what is Web 2.0 without Wiki right? The first aspect to help make it easier to find really relevant sites is to create independant "Metadata," that is data that describes other data. This metadata would be in a wiki format so that anyone can add or correct information about the website. So what kinds of data could be included in the metadata?

  • Short abstract about the site - if relevant, have site broken down into different categories those categories could be listed.
  • What kind of site is it? (i.e. corporate, informational, news, social content)
  • How time sensitive is the data? (Temporal Info)
    • When was the original data published?
    • How recent is the last changes made?
  • How location specific is the data? (Spatial Info)
    • Is it specific to a small geographical area or region?
    • What is the geographical area or region?
    • How relevant is it to the topic outside of that region?
  • Keywords, tags, categories, labels, etc?
    • What category does it fit into? (i.e. Computing > Smart Phone > iPhone)
    • What are keywords or tags that describe the site and its content?
  • What is the intended audience of the site?
    • Interest (or Occupation) specific (i.e. business finance, nuclear science)?
    • Age relevant?

The above are just some suggested metadata that could make it much easier know about a website and if it is relevant to you. Of course the above will not help with the initial sorting through the sites, only when you have narrowed the search down. In the next blog entry I will look at how the above could be used to help with the initial matching.

-Karl

Friday, February 16, 2007

Next Search Engine...a Google Killer? -- Part 1

Google is great, but not always. If you don't know exactly what you are looking for (i.e. the exact phrases) you are screwed; you have to go through page after page of useless links and waste a lot of time. How do you fix that? It's going to be a ton of trial and error, but I think combining aspects of:
  • Digg.com - digging or burying sites/(mostly news articles now) that are good, relevant, interesting or sites that need to be kicked to the curb. Adding comments can be very helpful (of course the comments themselves should be rated so that the best show up first).
  • Del.icio.us - tagging the contents of each site or page is a very cool concept. The personal tagging is cool, but when it is done collectively and shared it is ground-breaking. Admittedly they are doing a lot more with this lately, but it can still be hard to find the relevant sites to what you want. What is up with squishing words together to tag them (i.e. BusinessStartup) other sites like Google sites allow spaces between words.
  • Yahoo.com (old) - directory type portal with categories. People like things pigeon holed into categories, but they also have drawbacks like confusion over which category contains the type of site I want. The new Yahoo! is way to commercialized, it is like a TV channel or a walled garden - whoever has the money gets featured.
  • Technorati.com - probably the closest thing to what I see as a complete part of the killer search engine, but it is mainly for blogs. The "Where's the Fire" is very cool, and you can add "blurbs" about topics.
  • Google.com - to me it seems that they are sitting on a ton of techniques and functionality that is spread throughout the Google world (I am a big fan of Gmail, Google Docs, Blog Reader, and many more). I think that they may be getting ready to play the card to end the race for the next search engine, as a lot of the puzzle pieces are there but they have not been put together in an effective manner yet.
  • Amazon.com - the Recommendations, however annoying they may seem, are very cool. A form of pseudo-AI using personality profiles to find relevant sites or news for people could be a key component.
  • Academic Search engines - the articles have keywords and abstracts to help you narrow down which articles are worth reading. This type of information about websites would be a huge boost to finding the right site.

NOTE: To clarify, what I am proposing is not a way to do a better word search like Google, but to change how people can find information that they need.

Next, I will look at the actual components to combine into the Next Search interface.

-Karl

First Try....Is Blogging For Me?

I have sat on the sidelines during this whole blogging explosion. I thought that they were cool but not for me. I did not know what to talk about, but now I think I do. Two things that I have always been facinated with are Innovation and Technology. I want to make things happen, but it seems that you need a PHD for any big company to even talk to you.

I am always thinking, trying to come up with the next big thing. It's fun, but also very frustrating as I usually do not have the skills or tools to develop the things that I come up with. I will be honest it is usually not a totally new concept, just a spin or miss-mash of things I see or read about. So, what I hope to do with this blog is to post ideas that I would love to see in the future. Feel free to leave comments or feedback on them. Some things maybe only seem cool to me, but I hope that others will strike a cord with you.

- Karl