Google, on Monday August 22, released a beta of a new desktop toolbar, known as Desktop Search 2. Google claims it to be an upgrade to its previous Desktop Search tool, and this new search application goes far beyond typical search.
Keeping in mind that I do like Google, it seems many things that the company has done in the past year are encroaching boundaries of privacy issues and intellectual property rights. Eventually, this most likely will have their legal counsel spending months defending lawsuits.
Sometimes I wonder why other search engines feel a need to blindly follow Google's maneuverings and adopt some form of competitive technology without looking at the intelligence of the programs. Most of the top search engines are now offering some type of desktop search tool, and as far as I’m aware Google’s is the most popular of these tools.
To see what the newest Google toy now does with my data, I downloaded and installed the Desktop Search 2 Beta Toolbar. First, Google demanded that it shut down basically every open program: Windows Messenger, Instant Messenger, FireFox, Word, and Outlook. Perhaps this is so that it can integrate itself with them. Afterwards, you can configure they tool for personal settings using the screen below.
Of the three options of how to display the tool, the sidebar has the most functionality. And although it offered to index my Gmail, I declined; search engines can give me information overload as it is, and throwing my 100,000s of e-mails into a search interface just doesn't make me comfortable.
As the installation completed and the indexing of my hard drive started, I looked to my sidebar and found some nice user enhancements they added. It has a nice collection of information, clicking on items makes a popup slide out to the left to display more information. This even resembles the idea and features of the sidebar that Microsoft has been advocating for their upcoming Windows Vista. Unless Microsoft has much broader design plans, their new sidebar could be far less notable. This also goes head-to-head with Yahoo!’s newly purchased Konfabulator widgets, providing the same kinds of services (albeit less customizable).
Some of those services include weather for your local city and a stock indicator by default set to the Dow Jones Index (DJI). There is a “What's Hot” tab, that shows you what news and searches are hot or trendy online. It is automatically generated, using multiple resources, to show what people are talking about online.
Because of privacy issues mentioned later in this article, Google could be using information gathered from Google Desktop users to provide this “What’s Hot” information. Data collected from user searches could both define what’s how and also show how well developed organic result pages are.
Some other useful features offered are what appears to be an RSS feed of Google News and Web Clips, which displays RSS updates of web pages I visit. For webmasters, this is a clear reason why you need to be able to export your site content via and RSS feed. RSS has many applications, and reaching Google Desktop is just the newest one. This is where true marketing can occur, allowing websites to convert sales using desktop information feeds; it has much lower costs and higher views than any other ad production.
There also is a Scratch Pad where you can type notes and they will be saved automatically for you. There is also a tab for your photos, which cycles through all photos on your PC. So, if you collect pornographic photos and share a home PC with your family, you may want to rethink any decision to download this nifty little Google application (unless you feel confident in beating the Google indexer to the block files). I think Google could do better with the photo feature by making each user on a computer could have the ability to customize it for themselves. An administrator feature for parents would go a long way in gainging the acceptance of the end user Google needs.
Finally, there is the Quick View. For me, this tab pushed me to uninstall the Google Desktop Search Tool 2. The very first file that Google indexed for me reads "secure.overture.com" which is where I log in to manage my clients’ Yahoo Overture Pay Per Click campaigns. I am not about to let Google see this, nor can I run the risk of this information being exposed online in any way. Another issue is I also manage Google Adwords campaigns for clients. I run more Adwords campaigns than Overture, and log into the My Client Center more often than Yahoo Overture. The trouble is that Google did not feel that this should be in my top three Quick Views. That has me wondering why not, and why Overture should be?
Even downloading the Desktop Search tool is a monumental step for me. Having been online now for twenty years, I have developed many sound thoughts about protecting myself online after having trusted some things a bit too much. When speaking to a potential client from Australia the other day, I had told him I felt we all had to suffer one online disaster as a rite of passage. I do not want this disaster to sound unavoidable, but sometimes we can invite problems in. I have had many “small disasters”—a few spyware programs, trojans galore, adware left & right, backdoor robots—but each of these were overcome. My biggest disaster was losing two years of client reports when my web host’s servers were confiscated by Federal Authorities from the reseller hosting company I was using. One bad apple destroyed 135 other innocent websites. So, please excuse me if I sound a little overly cautious about inviting in a web technology.
While installing the Google Desktop search tool I bumped into a warning from Google. This is not the usual modus operandi, so I read it very carefully.
Hmmm. I need to read that privacy policy. In it, Google fortunately covers the most important subject in my eyes and that is what is Google able to see and not see, and what do I give them in return for them allowing me to download this supposedly cool search tool:
What does Google Desktop do with the information on my computer?
So that you can easily search your computer, the Google Desktop application indexes and stores versions of your files and other computer activity, such as email, chats, and web history. These versions may also be mixed with your Web search results to produce results pages for you that integrate relevant content from your computer and information from the Web.
Your computer's content is not made accessible through Google Desktop to Google without your explicit permission.
For more information about what Google Desktop can do, please see the Product FAQ.
So basically, if you use the Advanced Features, there is the possibility Google can use your computer’s Google Desktop index and activity reports. I didn’t see a clear way to tell Google it can or can’t access my computer, so how do I know it isn’t?
One major issue I have here is that Google indexes every file on your computer before you can control which ones it is using. Then if you do not want them to access certain files, you need to go back and takes steps to remove the content. This isn’t necessarily easy for everyone to do.
You can delete any item from Google Desktop's index. Once you've removed an item from the index, it won't show up in your Desktop search results ever again. The one exception is if you re-read an email in Outlook or Thunderbird, in which case it will get reindexed and added back to your search results.
But remember, you're only deleting Google Desktop's version(s) of the item. The current version is still in your computer, mailbox, or on its website. If you want to get rid of the original, live version of the item, you have to delete it from where it lives, the same as any file, email, or web page that Google Desktop doesn't know about. If you delete the original, live version of an email, file, or web page, and there are copies of it in your Google Desktop cache, those copies will not be automatically deleted. To get rid of the cached copies as well, you'll have to explicitly delete them from Google Desktop.
Google should rethink how the toolbar installs as I should be able to control the indexing before it happens and not have to go through extraneous steps afterward. You can also tell Google to exclude indexing certain directories, but this doesn’t help with Outlook email unless you want to exclude it all. And again, those directories will be indexed shortly after installation, and you need to change things afterwards.
This is something else that has me wondering how much is too much:
Your copy of Google Desktop includes a unique application number. When you install Google Desktop, this number and a message indicating whether the installation succeeded is sent back to Google so that we can make the software work better. Additionally, when Google Desktop automatically checks to see if a new version is available, the current version number and the unique application number are sent to Google. If you enable Advanced Features, this unique application number is included in the information sent to Google. The unique application number is required for Google Desktop to work and cannot be disabled.
Now, is this a good thing? I am not sure, they have me now like a tagged deer, don't you think? We can all have little ear tags applied and allow Google to see our every move online. I wonder how happy they would be if I did all of my internet searches through Yahoo.
Okay so perhaps I am a bit paranoid, but I have always been an early adopter of many new technologies. I hounded my ISP when I knew they started offering DSL broadband (I am on them now for FIOS Service), and take pride in being first in for such toys. However, I think the Google Desktop is a sign of Google wanting to be everything to everyone. If you step back and think for just a moment about where Google has ventured and is headed, you can almost envision a world of Goog... a place where you buy a home from Google real estate, Google controls all the utilities to the home as well as cable, phone service, broadband, and your entertainment choices are all in your Desktop Toolbar. Wait till Google adds control for your washer and dryer into the sidebar.
Semantic Mapping, Because a Keyword Isn’t Good Enough
( Page 1 of 4 )
So you’ve built a webpage and pulled it to the top of Google. You may not want to hear it, but that might not be enough. You can sit in the top tier or results for your keyword and lose all your clicks the guy below you. Optimizing for search engines means far more than simply getting a high ranking; it means getting people from search results to your site.
Being in the front should only be viewed as an opportunity to get a visitor. Not an end goal. It may get a few “dumb” click-throughs just by being there, but rank is not the sole factor in determining how much good exposure you will be getting. There are other aspects of SEO in the works here too.
There are a couple things that can keep a searcher scanning right past your result. First is that the person searching often has more on mind than two keywords and page rank. There is a whole world of reasons that person is searching, which you need to appeal to. At this point, search engines aren’t powerful enough to really understand what people want to find, so you will have to help your searchers find you.
The second thing that keeps people scanning past entries and moving onto ones with lower rank is that they do not read every word of description. They may scan titles. If a title seems promising, they scan for a phrase or word that pops out of the description. After being burned too many times by high-ranking link farms and irrelevant sites, web searchers with more than a day of experience are going to be moving through results similarly.
This is where many blackhat SEOs fail. This is where not optimizing your page means failure. Not only does your page have to return a legible listing in the SERPs, but you have to be fairly certain it will really spark interest in somebody who is searching for you. The idea has been loosely termed “semantic mapping.”
The idea of “search semantics” or “semantic mapping” is all based on the idea that a person browsing the web is not a dumb clicker. They have entered your keyword but may not be interested in your page. There is an overlying concept behind the words the person chose which is more than those words can express.
Semantics, the study of meanings and connections between words, can prove to be a friend to a good SEO. After all, search engines are textual programs. Their ability to judge context and meaning in the items they index is severely limited. While Google Labs is debuting a My Search History feature (to track browsing patterns and supply customized results), it is still a far cry from a truly intelligent search. This is where your job as an SEO comes in.
To make up for the shortcomings in search engines, a responsible SEO has to understand how sites will appear to human searchers. To clarify, I’ll recall a recent report from Enquiro, an eye tracking study. It raised questions about how people view search results. You may have heard about Google’s “golden triangle,” which is derived from the report. Basically, it shows that people view the full width of search results at the top of the page, then they scan down the left side of the page.
The scanning follows an F pattern, moving down the left of the page and shooting over to the right if something catches their eye. As they get lower down the page, it takes a bit more help to get their eyes over to the right of the page. This is how the lower arms of the F are formed.
The search engine users in the study moved their eyes quickly between results and used their peripheral vision to notice appealing words and phrases. They did not click top results simply because they were at the top. Likewise, simply because sponsored links were there, surfers did not necessarily even see them.
So optimizing for search engines is great, but useless unless you consider how the page will look to a human being. A search engine is not going to browse your page and buy 500 widgets a week. People who are attracted to your result will, though.
Naturally, you cannot always predict precisely what Google or other engines will pull as the description of your site, but you do control it. Look at your page to find what text search engines see. Look at what would be unappealing to a person if the search engine randomly cut a few words for the search results.
Also, read through the important parts of your webpage where your keywords appear. Make sure the keywords are in good context and that the search engine can pull information from before or after them. Do not put unrelated ideas near each other without transition. You know search engines can’t tell one paragraph from the next, and neither will people checking search results. We’ll start looking at more detail of this later.
The problem is that the initial idea of a search is often bigger in scale or less defined than what can fit into a search box.
Let’s take a look from the perspective of a searcher. Say I want to make a nice dinner and want to find a new recipe. I loosely know what I feel like and what’s in my fridge to work with, but I just don’t care at this point to try narrowing down the search. So, in the search box, I type “dinner recipe.” When going to search for something, people often start with broad terms.
The idea is larger than what I have typed. Even though I am searching for a dinner recipe, it doesn’t mean I want to see anything that is returned on it. The keywords I chose are simply my best attempt to express what I want to find.
A few of the high ranking results have “turkey” and “meatloaf” in the titles. As I scan down the page, I know I don’t have any turkey in the fridge, and meatloaf makes me cringe. I see a result with pasta in the title, which sounds good. But the description has nothing that catches my eye. Upon further inspection, I see that is because it is declaring the date and name of pasta cooking award. I move to the next page. I almost click on a link to a Mexican layered dinner, but the description looks like an email header.
“Dinner recipe” is a rather broad term to optimize for. Once I have decided that pasta sounded good, I may refine my search above: “pasta dinner recipe.” Now we have gotten more specific, as searches often do. Though this search is more specific, it faces the same issues as the last. For instance, there will be plenty of results that I don’t really want and plenty that don’t attract a click. One site was definitely optimized (arguably over optimized) and returned this description:
“Easy Gourmet Recipes Entertaining Dessert Recipes Seafood Recipes holiday recipes pasta recipes dinner recipes party holiday party party recipe party…”
That’s a keyword dump, if I ever saw one. Keyword dumps and other obvious over optimization often make web surfers ignore SERPs. This site is pushed to the top, and is seen as less relevant as something that reached the first page legitimately.
Even if the page has the perfect dinner recipe, I would never find it if the search results looked like garbage. The result listing has to trigger a match with the semantics of my search. If the title doesn’t appeal quickly, it’s passed. If the description is poor, it’s ignored.
To attract people, you want to optimize your site for ideas, not keywords for web spiders. Of course, your ideas will bring about keywords. Search these keywords and make a list of what good, eye-catching phrases and terms stick out among your competition. Keep these in mind when organizing your site. When placing your keywords in context, you’ll know how to distinguish yourself yet provide significant phrases for the SERPs to pick up.
When you have a draft of your site finished, read it as if you were a search engine. Are there any sentences around your keywords that would not attract a click for any reason? Keep an eye on unrelated ideas being near each other in the text even if an image or horizontal rule separates them. Watching keywords within content can be difficult, but it can pay off.
Of course you’re aware that you need to be thinking of what a person typing your keyword is thinking. Sure, somebody typing “dinner recipe” is probably hungry, but what situation are they in? Obviously, they don’t have a specific dinner in mind. Providing a general description is okay. But when I typed “pasta dinner recipe,” but of the follow appeared:
“Thousands of recipes submitted by home cooks. Searchable database, and menu ideas. Most recipes are reviewed and rated by users.”
“Pronto Pasta Dinner ... Sodium: 550 mg Carbohydrates: 35 g Dietary Fiber: <1 g Protein: 13 g. Recipe and photograph provided courtesy of Land O Lakes”
Which one is more click-worthy is obvious. Would you rather start hunting for pasta all over after clicking on a general description, or just right to what looks like the desired content?
I don’t mean to ignore titles while I talk about descriptions. They are also important. In the general search, “turkey” and “meatloaf” in a title will draw some attention from a few people. But because it was so specific it easily clashed with what I wanted. If your keywords happen to be “turkey dinner recipe,” that is a pretty pathetic cry for attention. It achieves no distinction from other turkey dinners.
You want to watch your title on these kinds of pages. Although the title is great for getting high on search results, it can cause somebody to quickly skip reading your results. This is a balance you have to play with.
There may well be more possibilities for semantic search to cover, but this is a starting point. Suffice it to say that optimizing for search engines means more than looking good to a machine. It means getting clicks from humans.
Eventually, we may even see more intelligent searches that rely on highly complex Latent Semantic Indexing (LSI), and keeping your content and keywords relevant and integrated like this will help you rank well in them. It is those who try to trick search engines and those who don’t optimize properly (or at all) that look very bad to people staring at SERPs and LSI searches.