YouTube would be a bad move for Google.

It is bad for shareholders since it makes Google less competitive on search since it provides a disincentive to make any advances in search technology. It also may create additional legal exposure for Google in terms of click-fraud lawsuits.

It is also bad for advertisers since it dilutes the value of Google’s ad distribution network. Advertisers are the life-blood of Google’s revenue model.

Hurting Search.Search is ultimately the utility that has made Google popular and that has made Google the default start for most people on the internet. Search is still extremely important to the internet and has a ton of value left in it as a function.

There are plenty of competitors ready and waiting to eat Google’s lunch in search.

The more content that is covered by Google’s ad distribution network, the less incentive Google has to fine tune its search technology.This is discussed in previous posts.This is especially the case for the untargeted, un-controlled content of YouTube.

Legal Exposure. The YouTube acquisition also creates another arena for link farms and click fraud. Only, in this case, it would be video click fraud and Google would own the content distribution network.

Google already has a challenge protecting its advertisers from click-fraud and the integrity of its search results from link farms and the such without owning the content or the content distribution network. This has provided a clear line in the sand.

By acquiring YouTube, Google may be opening up a whole new set of boundaries it will have to define in court as to where its liability/responsibilities stop and where it becomes the content producer’s fault.

Bad for Advertisers. It is the kind of untargeted content of YouTube which is a negative to brand advertising. (Unless Google plans on taking a more network Television type approach to having input with the content creators on YouTube -which is highly unlikely and would likely squash YouTube’s popularity). Brand advertisers crave targeted, well defined channels. By aggregating more untargeted content in its network, Google is creating a larger market place for bidding on big umbrella keywords, like deodorant, without adding additional value to the advertisers.

By increasing the demand without providing more value-ad, the short term result maybe higher prices for the keywords (and more revenue for Google). The longer-term results may be the defection of advertisers of all sizes.

Brand advertisers are often cited as a large avenue for future online revenue as big brand companies move a larger portion of their advertising budgets to online advertising.

Microsoft in the late ’90’s? In some ways, the talk of Google and YouTube is reminiscent of Microsoft’s foray into content in the mid to late 1990’s. In hindsight, Microsoft’s acquisitions and investments in those areas now look like a distraction. And while Microsoft was distracted, it’s growth slowed and the door opened to a whole new wave of competitors -including Google.

Does Google’s very economic model cause it to violate its principle to “do no evil?”

In all fairness, the same question could be asked of all automated ad distribution networks, not just Google’s Adwords program.

Looking at how the ad networks work, it seems that the more content there is on a subject (that is, the more popular or broad a keyword is), the more competition there will be for producers of that content to get noticed.

Therefore, producers will bid-up the price of an ad on that keyword. Is this the most efficient way for searchers to find relevant results?

One thing, this potential inefficiency in search relevancy has created an entire industry of search engine optimization specialists. These are people paid to understand the functioning of search algorithms for clients. It has also created an industry of link farms and black-box tactitians who exploit the gaps in the search algorithms.

(Note, this does depend on the assumption that many of the content producers are also vendors or sellers of goods or services within that broad category.)

One would think that the more content there is, the more need there is for the search engine itself to do a better job at sorting the data.

But, if the end result of this inefficiency is more revenue per click for Google, Yahoo, MSN or the owners of other ad distribution networks, do they have any incentive to improve the quality of their search, to do a better job?

Or, are they incentived to hold-back innovation and profit off the status quo?

p.s. For those with a long memory, this reminds of an observation by Jeremy Allaire, founder of ColdFusion, years ago, on why the browser stopped going forward once Internet Explorer became dominant.

Bainbridge OMPG carries on an interesting exploration on the value of automated ad distribution networks to distribute ads across offline media.  That is, does it make sense for advertisers to hire companies like Google, Yahoo or MSN to do their non-online media buying?

(Which brings up another question on how much media convergence is really valuable for brand advertisers -but that’ll have to wait for a different day).

The Bainbridge post talks in terms of relevancy and targeting.

Another angle to bring into the discussion is the supply of online content and the ease with which it can be created. While television channels have proliferated in the last two decades, and video content is ever finding new places to be viewed, it pales in comparison to the amount of content available on the web.

It is really cheap and easy to create a web page (which is not the case with a TV show).And it doesn’t require a license from the FCC to broadcast it.

The overwhelming ease of supply has the potential to destroy the integrity of the search results and therefore erode the value of ad distribution bots. Since the barriers to entry are higher for creating and broadcasting or disseminating?offline content, advertisers and media buyers should continue to get a bigger bang for their buck with offline content. -At least until the search engine networks figure out some way to further fine tune their alogrithms.

Here’s an example of what I’m talking about. Take the keyword “deodorant.” There are nearly 9 million pages Google found that it thinks are relevant to the term deodorant.

There is a neat metric called the KEI (keyword effectiveness index) which compares the supply of webpages which are relevant to a term versus the demand for that term. It is a scale of 0 - 400, with 400 being the most effective keywords. The KEI for deodorant on Google is 0.00. That means that the supply of websites, the supply of content, outstrips demand by a near infinite amount.

An automated ad distribution bot doesn’t know which of those are most revelant to an advertiser. It will broadly distribute those ads. This is fine if the marginal cost of each ad is near zero.

The problem is that with such a large amount of competition, owners of those content sites will likely start to bid up the cost of that keyword to make sure their content or product gets seen. (Which, of course, doesn’t give the ad network much incentive to change since its collecting the money.)

A second problem with the ease of content creation and the volume of content is that it increases the opportunities for click-fraud, link farms, and other practices which further degrade the value of the ad network.

A third problem is that the content doesn’t cost the ad network anything. They are capitalizing on other people’s work and do not have an ownership or any accountability to the content owners. Therefore, they have very little incentive to protect the integrity of a particular ad.

Google, for example, does not have any practice in place to stop a rival advertiser for flat out stealing another advertisers ad.

A television network is dependent on advertisers for revenue. The network and media buyers protect the integrity of the ads and, to some degree, the content of the ad outlet (the content and programming).

In the world of Google and other automated ad distribution networks, it is the content providers and advertisers who are dependent on them. Its not a winning dynamic for a big brand advertiser.

Found an anecdote on Mark Cuban’s blog. Highlights the nature of the information on Wikipedia (and the process on its formation).

Along the same note, here’s a link to a now classic Stephen Colbert report on Wikipedia.

I’ve received a lot of feedback to my position on Wikipedia brought about by a recent Wall Street Journal article. The feedback has been productive. Its helped elucidate some of the key differences between Wikipedia and Britannica.

The foremost of these differences seems to be audience expectations.

There is one world, online, where people want to be part of the process of growing knowledge, of being heard and sharing. Knowledge in a particular area, for them, is something malleable and something they use to identify themselves with a community.

Then there is another world, offline, where people just want the facts, as they are. Doesn’t have to be latest greatest. Doesn’t have to be with a grain of salt. Something straightforward they can learn and then get on with their lives. Knowledge in that particular area, for them, is something they want to find out about, but not something they want to give too much commitment to.

This gets to the core of our definition of knowledge. When someone wants to “look something up” what are they hoping to find? What do that want to do with that information? How do that want to interact with that information?

Wikipedia presents a different way of interacting with knowledge.

Its seems to have unlocked a lot of pent-up demand for people to be heard. But I’m not sure its the most productive way for society as a whole. I’d worry if kids did science reports based on Wikipedia. Or if they protested the whole format of a report and pushed, instead, for an open dialogue with their class on the subject.

There is a place in this world for a fixed grounding -for clear definitions of what we know, delineated from speculation and opinion. Perhaps Wikipedia should have a disclaimer that the contents are a work in progress, take it with a grain of salt. Perhaps it would be more accurately filed in the category of an open, collaborative publishing site and not an online encyclopedia.

The Wall Street Journal has a tense interview between the founder of Wikipedia and the editor the Encyclopedia Britannica.  Raises tons of interesting questions.  The biggest one is whether the Wikipedia approach offers readers a source of knowledge? Or is it, in Stephen Colbert’s term, a source of truthiness? Feels true, we’d like it to be true. But it isn’t.

For those who don’t know how Wikipedia works, it goes like this.  Someone decides that there should be an entry on a particular topic, say project management software.  Or, they decide there should be an entry on a particular piece of software.  The author searches around and sees that other vendors of project management software have entries on their products, so why not this particular software.

They go ahead and write that entry and post it.  Suddenly, this particular software is part of the universal body of knowledge available to everyone.  (And it also benefits from having a great link-source for search engine rankings.)

The post may then be visited by other people interested in the subject or it could be totally ignored.  Or, it could be stumbled upon by someone who decides it doesn’t fit with “Wikipedia standards.” They may decide it is an advertisement and not worthy of being included in the Wikipedia body of knowledge.  This self-appointed editor then goes to work on the article.

Unbeknownst to the author or to the world at large, the self-appointed editor begins an opaque process to remove the article and any mention of it throughout Wikipedia.  Like in the Soviet Union, the article, the software and the company are quietly erased.  They no longer exist.

What remains in Wikipedia are articles which the self-appointed editor crowd decides are relevant. Articles and entities which fit into their world-view or their self-interest. At the end of the day, Wikipedia reflects the interests and views of a large, community clique.

In contrast, Britannica draws on 4,000 scholars - Noble Prize winning, expert in their field type people. Academia, while far from perfect (and in many ways, another, albeit larger clique), is, at least, a recognized tradition of knowledge seeking, debate and fact finding.  The goals of academia are to advance the cause and body of knowledge.

While it may be counter to truthiness, academia is where the human race has drawn its truths.  I’m not talking here about moral truths. I’m talking empirical, mechanical or historical truths.  The same can’t be said of the clique of people who somehow get into the editorial club at Wikipedia.

Further, the editors of Britannica (and other traditional encyclopedias) see themselves as keepers of the academic tradition and disseminators of knowledge.  Unlike Wikipedia, they are not out to “make a point.”

When Diderot and the other encyclopedist set out to organize and publish the entire body of human knowledge, the goal was the educate as many people as possible. To share the fruits of lifetimes of research with the “masses”, to make the repositories of knowledge more accessible.

Wikipedia’s goal seems to be to have as many people as possible decide what the body of human knowledge should be. Its spin and hype a public bulletin board, rather than knowledge.  Like another mass movement of Diderot’s days, the French revolution, this too will be disruptive and dangerous. And perhaps most upsetting, a distraction to education and knowledge.

Search engines may be stifling the success of software companies. How?  By assuming all searchers are the same.

In general, a software company writes for a target audience, its intended customer base. That audience may or may not be people who spend a lot of time on the internet or in the blogosphere. (For brevity, let’s call those who do spend a lot of time on the internet and blogosphere netizens.)

However, it seems that search engine rankings are determined, in large part, by the amount of presence a specific url has on the general internet and blogosphere. The more a particular software is talked about by netizens, on net-based outlets like blogs, websites and forums, the higher the likely ranking of that particular software on a search engine.

Due to their popularity, search engines are often the first place people turn-to when looking for new software. This is particularly true of non-netizens who don’t know the specialized websites or back corners of the internet where they can find insider information on software or technology.

So, the non-netizen goes to the search engine and searches for the type of software its looking for.  The results come back ranked by the amount of buzz by netizens.

The ranking may be a perfect set of choices for one particular user, say a netizen or those who are looking for the same thing netizens value. But the results might not be a good set of choices for the needs of the non-netizen.

The right software might be buried deep on some double digit page of the results.

The non-netizen may never find the right software. The software company misses out on the customer (and therefore the revenue). Both sides lose.

There is a potential paradox here. That those who most need the help of search engines to find the right product, are least likely to find the right fit in the search results.

I’ve found that a lot innovation floating around is cool, but doesn’t necessarily solve any problems or help make people’s worklife easier. This is nothing new. Useless innovation has been around forever. But it seems that the pace has picked up on software being released that doesn’t bring anything new to the table.

Wonder if it has anything to do with more developers spending time on blogs, reading the latest success stories or latest technologies and trying to hop on the wave –without understanding the fundamental problems being solved or the thinking that goes on behind the design of the functionality?

How has software already changed now that so many developers get input directly from the blogosphere? Is it creating software that solves problems better? Or, does it just increase the tempation to create something that rides latest, greatest tech buzz?