Here is the link for the ad age search fact pack 2008. It covers SEO / SEM and is packed full of pie charts, stats and short stories on Google, Microsoft, Yahoo. Took me about 20 mins to go through it – nice overview of what’s happening in search. Covers industries like travel, politics, entertainment, what are the top search terms in these industries. Also lists the top 25 search firms.
Oh and yes we can! Obama becomes president – congrats.
Matt Ho.
Folksonomy is a term I’ve come across recently in a couple of articles and on my company’s blog. To begin with, taxonomy is the science of classification – that’s my understanding. The way in which content creators/librarians/administrators, etc… create categories to classify information.
Folksonomy is play on the word taxonomy. Its how “folks”, normal people like you and me and the general web community can add our own labels known as “tags” to information. A good example is Youtube, Del.ico.us and Flickr and blogs. When you upload information or media, each of these allows the user to apply our own “tags” to create our own categories to search. For example when I finish writing this post on this blog, I have the option of creatiing new tags which allow others and myself to find it when the search function is used. It also will put it in a category allowing others to see it. So I’ll add tags like “taxonomy”, “folksonomy”, “metadata”, etc…
A classical example of a taxonomy is dewey decimal system used for books in libraries. It is a system to classify the hundreds of millions of books, so that users can find books. Folksonomies differ because the user can add their own labels, effectively creating our own kind of dewey decimal system.
The strength of folksnomies is that it allows more users to create labels (tags or if your really nerdy – “metadata”) and is more useful in a system where there are even more information to categorise. In the past, taxonomies were restricted to administrators like librarians or authors (content creators). But now, with the explosion of information, videos, pictures, websites being uploaded and shared on the internet – folksnomies has had its greatest use. It allows scalability – it can grow with the more users and more information that is shared.
I’ve actually being using tags for a while – for my own blogs. It is an easy to keep track of everything and to search for posts. Because I know in my mind what kind of tags I would use.
The second strength of tags is that users are better equipped to label their own information because they are more engaged with the content and have a deeper interaction with it. For example, if I read an article that I like and add it to the social bookmarking site “Del.ico.us” (now www.delicous.com), I will add in labels about what I feel the article is about. I would have read the article, thought about it, etc.. and put related tags opposed to a search engine that uses web crawlers that would scan for the keywords in the article.
However, this is also a weakness of folksonomy because the user is biased as to what that article is about – different users can potentially put vastly different labels. Further, users can use the same labels for different things. As this paper mentions, the tag ANT can refer to “Actor Network Theory” but also “Apache ANT” a programming language. So folksnomies is not without its problems.
One of the interesting things I found with Delicous is that when you are about to save a link and add in your tags, it will actually recommend tags that other people have put on there. I look at them and think “ok, that sounds alright” and you can click on those links – that is one way to help users add labels and create consistency across tagging. I can also add in my own tags as well and ignore the recommended ones.
Folksonomies allow users to create new terms and to start creating paths for finding things. It also can introduce you to new information or media, because you can click on a category or tag and discover what other people have posted up under the same tag.
And to clarify, metadata is the search terms or tags you put into a page, allowing web crawlers to find your information. It is important when putting in the terms for metadata to think of all the various terms a user might type in to find something – for example if you have a company webpage about soccer, you also want to use the term “football” which is what the rest of the world calls it. So the page wont get rejected if people dont type in soccer.
I’m out like folksonomies,
Matt.
Originally written on 9 May 2008.
————————————————————————————————-
In a world of discovery, we are all searching for something
We look to find ourselves, to find other people like us, and to find the answer to life’s greatest mysteries. We believe the answer to most things lies somewhere……. out there. In our search for answers, often the first point of call is not religion, the trusty encyclopedia, the dictionary or the wise old sage.
Instead we turn to mankind’s answer to all things: the search engine.
What is the capital of Kazakhstan? Obama or Hillary? What should I study? The Celtics or the Lakers? How do I fly there? Why are people protesting about Tibet?
Millions of these kind of queries are fed into search engines each day. The search engines never sleep. It never fails. It always computes.
More than a decade ago, when we were still using Altavista and living in a pre-IPOD world, it was thought that it was all about internet portals. The idea was that you had to keep the user there as long as possible on the search engine page. More content they screamed! They did not want you to leave to find your search results. Refining and gathering the most relevant answers to our queries was not as important.
Internet companies had also not figured out how to make money from search. A voice in the distance cried out “Sell the keywords!”. Someone had come up with the idea that advertisers could be charged for buying the keywords and having their ads displayed with relevant results. That voice was Bill Gross of Idea Labs, a serial entrepreneur. He decided that advertisers could bid for the keywords in an auction style system, paying differing amounts according to the clicks they generated. Thus this was a way to monetise search engines.
When he told the world (i.e. Silicon Valley), they laughed at him and declared him crazy. Clearly, a man before his time. His company, Overture would later be sold to Yahoo to run the advertising on their search engines for billions (Interestingly enough, Yahoo has been the subject of a failed takeover bid by Microsoft in an move to compete with Google)
At the same time, two Stanford PHD students had begun working on the worlds most audacious search project – Google . Two other Stanford PHD students had co-founded Yahoo. These guys were developing something special, but they just could not figure out how to make money from the internet and looked around for a solution. They saw the Overture model, liked it and took it. When they told the world in 2002 about this idea to auction off the keywords, they were hailed as kings. Later Google, would settle out of court with Overture but it was too late – Google was already too dominant in the world of Search.
Google’s secret sauce lies in its PageRank methodology. The heart of it lies in something known as backlinks. The two academics, Larry Page and Sergey Brin looked at the concept of academic research – how a thesis is considered more authoritative the more times it is citied. They applied this concept to the world of search and analysed the links linking back to a website. It is the equivalent of hitting the back button on the internet browser and figuring out how many times it is linked. Using this method, they could measure how relevant a website is. A crude but effective measure. The PageRank formula also looks at a million other variables, but that is the core of it.
In a world of search, ideally you want your website to appear in the first three results on the first page of the search results. There is no benefit being on page 10. You want to be considered relevant. There are ways to optimize your website by putting more of a particular keyword in the headings, in bold, using it more times on a page, etc…This is a practice known as search engine optimization.
Another practice is known as search engine marketing (aka paid search). These are the results that turn up on the right hand side of Google’s search results as if someone had magically put them there. But no, someone is going to pay money for that. The amount an advertiser pays is worked out in one of two methods. You can pay only if someone clicks on the website, and nothing if they do not click – a practice known as cost per click (CPC’s). The other method is on a views basis i.e. the number of times your page has been viewed. This is known as cost per impression (CPM) and is based on 1,000 views.
Most advertisers would prefer the CPC basis because they only pay is someone clicks. However, there are times where advertisers would prefer a CPM basis as it is more predictable and the budget can be forecasted better.
Different advertisers will pay varying amounts depending on click through rates (CTR’s) and who they are. A search engine ideally wants the most relevant ad to be placed next to relevant results for a particular keyword. They would prefer a good advertiser and can ban someone from advertising if it may cause more harm, even though they are prepared to pay more for that keyword.
As an advertiser it is important to understand what kind of words a user will be entering into a search engine to find a particular product / service. For example, a business selling mobile phone will want to buy the following keywords: phone, mobile, 3G, handset, nokia, network, Telstra, call, SMS, etc…
Ah, the world of search, where you can discover anything and pay to be discovered.
———————————————————————————–
If you want to read more, I can recommend “The Search” by John Batelle and the Googlepedia (a brief encyclopedia) about Google.
Or maybe you could just search for it?
