Do you still use directories to find information? I do, actually, occasionally resort to a directory tool but mostly for finding brick-and-mortar businesses. In my current personal search habits, the combination of Web directory and mapping tools available online makes my life easier.
It’s no longer enough that Web directories list Web sites and briefly describe them. Even the fact of Web site categorization has become less relevant to my needs because directory search tools make it easier to find listings. I rarely drill down to a DMOZ listing through categories, for example.
The combination of directory plus crawled Web listings is no longer useful because anyone can spam a directory (even Yahoo! and DMOZ, both of which include many spam sites) and because anyone can spam a search engine. But when you expand the scope of the directory or crawled search’s data beyond simple Web site listings you introduce a new factor that is less spammable (but don’t be fooled — any database can be polluted with low quality data).
Today we have directory tools that show us maps and contact information, user reviews, and proprietary algorithmic evaluations. A “proprietary algorithmic evaluation” is something like a Quantcast listing or a McAfee Site Advisor. We also have annotative directories like Wikipedia and Answers (which is a Wikiclone but tends to have better data because the Wikicommunity cannot edit out useful self-promotional content).
Annotative directories would actually be more useful than Wikipedia in several ways, particularly if they allowed people to take ownership over their own name spaces. So Britney Spears could go into a properly designed annotative directory and tell people about herself without having to worry about someone using the page to bludgeon her reputation. Wikipedians take no responsibility for the emotional harm and distress they cause people by improperly enforcing their standards. A community where each listing can only be modified by the listing owner can compensate for self-promotional hype through user feedback (like star ratings, pre-set comments that express approval or disapproval without being spiteful and immature, etc.).
Although the Web presently lacks truly useful annotative directories their day is coming because Local Search has proven that the concept can work. You can take control over your Local Search listings without depriving the search services of the ability to leverage social value. Social valuations may not always be accurate or reliable (any hostile entity can use sock puppets to hurt your credibility) but you at least get to provide the bare facts about your business in Local Search.
People Search tools sort of try to do this too but they have so far taken the wrong approach. We need a People Search Web Directory that looks and feels like a Web directory, that works like a Web directory, that allows people to take charge of their name space, and which allows third parties to annotate the listings without being unreasonably hostile or manipulative. The Wikiconcept has failed miserably in this respect but it might not take much tweaking to fix the Wikiflaws.
The fact that annotative and listing-owned directories exist does provide us with some useful opportunities for search optimization. That is, old school Directory Optimization has pretty much died out (although we have successfully transferred some of those directory optimization skills to on-page optimization). But Local Search Optimization and People Search Optimization have begun to develop as skill areas in SEO. Annotative Search Optimization would be a better name for the general group of optimization skills we need to develop.
One of the problems with People Search today is that it is still driven by Web content. The search tools crawl the Web to find references to people and try to identify who is whom. I recently looked at my name space on one People Search service and found several listings that referred to me. One listing was in Seattle, one listing was in Houston (although another Houston listing applied to a different Michael Martinez), and there were a couple of geographically neutral listings.
Web content does not tend to consolidate itself, and therefore consolidating Web content doesn’t really work well. If we could incorporate digital fingerprints into our Web profiles we would be able to help People Search tools identify people more precisely. That is, if I could include a digital fingerprint in every blog comment, forum post, social media tag, user review, etc. where I contribute content then People Search tools (and general search tools) would have an easier time of finding my content.
Today’s Web content software doesn’t provide any digital fingerprinting capabilities so we have to wait for a new generation of software products to help us consilidate our content. Worse, there are no universally accepted digital fingerprinting technologies. In the 1990s news group and email discussions were peppered with digital signatures but that technology was never fully adopted.
If the digital signature concept can be revised so that all Web site registrations allow the inclusion of such a signature we can actually gain better control over some of the abuses that people tend to engage in. That is, if a reliable digital authentication process is used people would be restricted to using relatively few digital signatures. A trustworthy digital signature would have to be tied to some very private information for verification (very much like your bank account information or passport information is).
Of course, people would immediately try to breach such a system but Trusted Digital Signatures would work because — in order to include a digital signature in a registration process — you would be required to prove that the digital signature actually belongs to you. I think a safe way to do this would be to require the registration process to work like this:
- You complete a registration form for some Web service, including your digital signature
- The Web service pings the digital signature authority with the request to include your digital signature
- You have to log in to the digital signature authority service to confirm that you initiated the request
- The digital authority sends an authorization ping to the Web service
- The Web service acknowledges the authorization and inserts your digital signature into your profile and all your contributions
It would probably be reasonable to require periodic reconfirmation of the digital signature just in case user accounts are breached. A breach could be detected if the digital signature inexplicably vanishes from a profile or contributions. Whether the signatures should be archived or managed dynamically might be a question for some people but I think the third-party verification process makes it safe to leave digital signatures around the Web. Simply requiring email verification is insufficient because email addresses can be rerouted, changed, etc.
Once you build the infrastructure for a trusted digital signature you can begin organizing data about people on the Web — but the people would have some influence over what data (about them) is organized. You can choose to add your digital signature to content you want people to find and associate with you. You can do this for your directory listings as well as for your community memberships. People Search thus becomes more meaningful and relevant.
And if we can sustain a verifiable but secure digital signature for a person then we can do the same thing for an entity like a business or organization. Hence, businesses and organizations can take more precise control over their public statements by digitally signing statements (a modified process would have to be used to ensure that only authorized representatives are speaking on behalf of the organization). In this way, Web content would become more trustworthy and people would be more accountable for what they do and say.
Which brings us back to Web directories. Web directories are considered to be “trusted sources”. In fact, search engines tend to crawl the Web starting with major Web directories because they feel that human review somehow improves the value of the content in the directories. Unfortunately, today’s major Web directories link out to many dead, altered, and spammy sites. They are not personally useful to me any more and we need a new generation of Web directories.
A good Web directory should provide unique, useful, relevant content but in this respect I use “relevance” to refer to the user’s needs, not what the user is searching for. A relevant Web directory complements my search process by refining the data it shows me. Annotative search is much more beneficial than algorithmic search. Yahoo! only provides algorithmic search (the descriptions are not truly annotative). DMOZ only provides algorithmic search. Yahoo! Local Search, AskCity, Google Maps, and Live Search Maps offer truly annotative search (although it’s still very crude).
Although privacy advocates may be concerned about technologies that incorporate digital signatures into their user interactions, I think that if a reliable authentication process is developed we can reasonably build a level of trust into Web content that doesn’t yet exist. Instead of favoring Web sites that are link popular search engines could favor digitally signed Web sites and focus more on relevance and trust, less on algorithmic voodoo and nonsense.
Web spammers would have to digitally “out” themselves in order to compete if search engines found enough authenticated digital signatures to reliably organize data on such concepts. But how would you digitally sign a whole Web site? The authentication process would have to work a little differently. For one thing, I think the digital signature would have to become a meta tag and used in conjunction with a universal verification system. That is, instead of using Google Webmaster Tools and Yahoo! Site Explorer to authenticate or verify your site with those services you could digitally sign your site and they would be able to ping the authentication service.
Once you confirm the authentication request, the search engines could activate the full Webmaster tool features thry provide as well as give your site a little more trust. After all, if you digitally sign a Web site then the moment you start spamming an index all of your sites can be held accountable. Now, what’s to prevent people from creating multiple digital signatures? Again, we’d need an authentication system that requires enough private information that you’re not likely to be able to spoof it — certainly not over and over again.
Maybe you could create 2 or 3 digital signatures (which is not desirable) but the authentication system should prevent you from creating 50, 100, 1000 digital signatures. Digital signatures have to be tied to you the same way Local Search listings are tied to a specific busines. Digital signatures have to be reliable enough that people want to use them to declare what they are willing to be accountable for.
If we adopt digital signatures across the Web many of the online abuses that have plagued people (and search) would gradually diminish. But more importantly, we would have the option of influencing which information is associated with us, how that information is organized, and how that information is made available to others. Search tools would be vastly improved because they would have the means to algorithmically determine who is responsible for content. Site operators can be distinguished from users, so a spammy post in a forum doesn’t have to lead to a penalty for the entire forum.
The search environment would ultimately determine whether digital signatures become so relevant that we must compel people to use them, and that idea is truly scary. The day the search engines favor digitally signed content over unsigned content will bring cheers and jeers. People will be able to look at more relevant search results because the link-based analysis that has failed so miserably will go away; but people will also have to ask if there are any boundaries that the digital signatures should not cross.
The moral and ethical power that digital signatures might confer upon search engines needs to be engineered and carefully monitored well in advance so that we don’t find ourselves ruled by the machines we created to help us. It is not enough that we are aware of the risks entailed in adopting such a technology. We need to consciously plan and sculpt the technology so that our freedom of choice is respected while reinforcing the integrity of our search visibility.
It’s time to start planning a new generation of information technologies. In fact, plans are being drawn up already. The search optimization and Webmastering communites now have a real incentive as well as an opportunity to influence the next generation of information technologies, as well as the generation after that. Now is the time to start creating annotative directories that don’t inflict biased content on people the way Wikipedia does. Now is the time to start creating cooperative technologies and tools that provide people with the means to influence or control where and how their online visibility is shaped.
Think about it.
{ 0 comments… add one now }
You must log in to post a comment.