• Home
• About
• Writers
• Landscape
• Wizards Index
• Fraud
• Wanna Be Like Larry?

LinkedIn and Faceted Search

I know the social search revolution has taken place. The parade has left me unimpressed. I am content in the goose pond. Every few days I swat a pesky comment about my lack of excitement about social search, social networking, and social communications. Folks who are “social” seem to be triggered by my goosely musings. But let’s focus on reality. LinkedIn has become a giant job hunt and self promotion service in my opinion. That’s fine. I learned yesterday from a person talking about JobAngels.org that unemployment may be higher than the official government figures. Almost any service that promises an opportunity, no matter how slight, to land a gig that pays will attract attention. That may say more about the nature of social life in 21st century America than online in my opinion.

Source: http://www.emergingspirit.ca/files/images/Unemployment-LR2.preview.jpg

I grant that there are some interesting chunks of content on services such as LinkedIn. One of my colleagues is active in the social search discussion group. In my opinion, high quality content is the exception, not the rule. As the economy worsens, those without a job are looking for online services to be more than search and retrieval. Online is one more communications tool. It is, therefore, not surprising to me that the “traditional methods” of generating income that engaged my father are anachronistic or dropping like flies when the country sprays 2-4-D over this goose’s mine run off pond.

LinkedIn has been tough to use since I “joined” the service years ago. I have one of the goslings manage the requests to join my network. The few times I have offered a substantive comment, I received email about my inability to provide what the “members” of my “network” wanted. Tough. If folks don’t like my opini0ns, there is an easy answer. Click away.

I heard at a conference on December 14, 2009, that Endeca is providing the faceted search system for LinkedIn. Let’s assume this is true. Is search what LinkedIn needs? A related question:

“Will the Endeca search system and “true guided navigation” improve LinkedIn?” My answer, “Some.”

I can hear LinkedIn saying, “Better search means a better search experience.” Okay, okay. But what’s the point of putting a new coat of paint on a house without getting a home inspector to look for those glitches that make the difference between a good deal and a money pit. LinkedIn is one example of a user-fueled information system. It is easy to allow these services to operate without any significant editorial oversight. But is that what I want?

No. I want professional services to have more than eBay style controls and the “wisdom of crowds.” I expect professional services to add knowledge value. Frankly I don’t think search does that. An externality like search software cannot remediate certain structural and economic factors that make an information what  it is. I anticipate that lots of search vendors will disagree. That’s okay, just check out the editorial policy of this Web log before you try to reform me with your opinions.

The new interface. Search cannot solve content quality problems. Marketers may make this assertion. The reality is that finding baloney does little to help me locate Kobe beef. Fancy search is not able to address the substantive content issues. More sophisticated content processing is required. Where’s a reputation score? What about lineage?

First, in the new faceted LinkedIn interface, the inclusion of hot links that allow a person to locate related information is useful. However, when the people using the service are on the hunt for jobs, I see no substantive change in the trajectory of LinkedIn knowledge value. Software is external. By itself, code does not impart knowledge value to information that may have deeper issues. What about  a user like me? I am not looking for jobs. I am not looking for classmates. I am not looking to find an expert. In short, the information payload of LinkedIn for the type of work I do is unchanged with a better search system. The content is the issue, not the access. Search in many organizations is a bandage that is too small to cover the deeper problems in the service.

Second, guided navigation is now a commodity. One can implement the function with open source tools or just get a low cost solution from Microsoft, one of the leaders in driving down the perception that facets are better, faster, and cheaper. With a content base like LinkedIn’s, adding facets is not rocket science, and I wonder why the firm has not moved more quickly to implement a more modern search system. The slow shift to a more robust search system is encouraging. I hope that the firm dives into the deeper issues of its social service. LinkedIn is not bad, but I think it could be significantly better.

Third, search will not address the interface issues, the begging-for-dollars screens, the confusing set up of the site itself, and the deteriorating quality of the content on the site. Obviously when users can build a profile and post information, the “quality” is in the hands of the users. In a professional service (which I assume LinkedIn wants to be), the proliferation of job ads, specious assertions about expertise, and the recycling of content by those who want to inform group members creates noise. LinkedIn has an editorial obligation to ensure that the system has content that merits inclusion. In my opinion, a search system that finds information goes half way. Smart search needs to calculate quality, filter baloney, and assign “reputation” scores. The “owner” of the Web site must experiment constantly to find ways to minimize noise as perceived by a paying customer from the useful content objects in the system.

A quick example is that several people post links to the content of this Web log in the search group. I don’t have a problem with this. My concern is that I write articles that are often humorous. I suppose I could put a label on the write ups that are intentionally Swiftian such as my statement that the USPTO’s search service is great. It is not. It is awful and it should have been fixed years ago. Posting one of my Swiftian stories without context creates the impression that a particular article is “real”, not  a jibe. Posting multiple stories from my Web log is laziness, not substantive information. I would prefer that a person cite my writing and then add new information to the topic. I see similar laziness when a person asks to be on my “network”. I pay someone to be “me” in the social space. When I get a name and zero context, I have told my online counterpart to ignore the request. Perhaps LinkedIn could provide a sample of what a “add me to your network” email looks like. Seems simple enough to me.

I am delighted that Endeca or whatever search vendor sold this job made a sale. I want to be clear about my opinion. LinkedIn needs more aggressive, direct action in these areas before a search system will deliver a payoff to me; for example:

1. Filtering of job ads. I find pleas for people to help a recruiter earn some bucks or Euros by placing an Autonomy SME annoying
2. Enforcing some editorial controls on the content and putting in place guidelines for cross posting. The goal should be to add value, not point to articles out of context
3. Providing a service that can be navigated without recourse to a search box. I have to tell you that when the person I pay to be “me” on the social networks shows me what LinkedIn looks like, I am baffled. I can’t tell what’s what. When I walk through the response to a request to join my network, I need more than “met you at a conference”. Not a chance that I will remember this person. I instruct my agent to ignore such requests. Detail needed!
4. The fee approach is okay. Just make it clear what is provided. As far as I can tell, paying gets more of what’s free. I will pay for substantive services. Why not make it clear what one gets for how much. Maybe I don’t use the system enough to see this basic information? Perhaps LinkedIn may want to look at their core presentation of information with the trifocals of a 65 year old addled goose?

I suppose LinkedIn will point out that I am indeed little more than an addled goose. That’s par for the LinkedIn course. High value content is needed and lots of it. Then search is useful.

I would have used a different approach by putting the money into improving content and the core navigation logic. Searching flawed content does not constitute a net gain for me. You may find the LinkedIn guided navigation just what you need. For me, more substantive work is required. Just my opinion.

Stephen E. Arnold, December 16, 2009

Oyez, oyez, I want to report to the American Battle Monuments Commission that I was not paid to write about LinkedIn, its content, its navigation, and its business model. I anticipate a dust up in the comments section of this Web log. That’s the price of not looking for work, not being a social goose, and not rolling over when search is positioned to solve problems that information retrieval cannot address no matter how much dough is wrapped about the spiced apples.

Comments

Copyright © 2008 Beyond Search · ArnoldIT.com · Disclaimer · Revolution News theme by Brian Gardner · Log in