The Most Agile of the Dead Tree Outfits Trimming Digital Staff
January 28, 2009
A happy quack to the reader from Australia who alerted me to some News Corp. staff reductions. Greg Sandoval reported here that News Corp. chopped some staff. What caught my attention was that the trimming took place in digital units, including MySpace.com and Photobucket.com. What’s the problem? In my opinion, its a business model issue. Suck a start up into a giant outfit and there’s culture, technology, and financial shock. What’s the dead tree fix? Fire those who work in the digital units. Change the business model or dig into technology? Nope, dead tree outfits know best.
Stephen Arnold, January 28, 2009
More Pay for Traffic Plays
January 27, 2009
Mahalo.com is a social search system. The idea is that humans get involved. Humans, although more expensive than software, can handle certain tasks with greater understanding. Software still makes mistakes. Humans make some doozies as well. In search, I am on the fence about the efficacy of humans versus algorithms. More precisely, I am on the fence about the specific role of humans. In the good old days of commercial databases, humans were the way to go. The costs of people are high. Once software is locked and loaded ongoing operations can be budgeted and, in my experience, certain costs stabilized. I trade off software for humans now, but I don’t agonize over the decision. Costs control is a big deal to me.
Jeff Meiser’s “Mahalo Spices Up Information Sharing with Cash Incentives” here adds a new spin to the human intermediated Mahalo.com search service. Like Microsoft’s pay plans for Live.com search, Mahalo.com is urging searchers to “pay a few bucks” for the answer. The for fee question and answer model is what Find SVP (Paris and New York) was been doing for a long time. I’ve lost track of that company, but I recall that the sales effort to line up “gold accounts” was significant. Find SVP was a good idea in the early 1980s but the cost of sales and the cost of human researchers was brutal.
The whole pay for information sector has been under siege for many years. Dialog made people pay to ask a question and look at abstractions (Type 5 outputs). The problem was that the answer may not have been in the output. There was no easy way to run down some information in 1980 so people paid. As soon as free options became available on the Internet, the bottom fell out of some commercial information services. You can see some of the financial burden if you look at the 10 Ks for traditional information companies in the commercial database business. Certain sectors are flat. Thomson has exited some of its information businesses. A wise move. Reed Elsevier is trying to increase revenues and hold down costs of information operations that involve humans. Tough job. Blue chip consulting firms run pay to ask question services as well. To make this work, the client has to be convinced that the answers come with a pedigree. So humans can improve certain information operations but humans cost a great deal. Add sales and subject matter experts together and you get a Bermuda Triangle of costs.
Mr. Meiser’s article doesn’t dig into the past with pay for answers. That’s fine. He does include information about finding answers on Facebook and Twitter. He provides some color about the smart money that is behind Mahalo.com. For me, the most interesting comment in the article was:
Since no single player has emerged as dominant using the question-and-answer search model in the U.S., Mahalo sees a unique opportunity to become the industry leader.
I am all for optimism. I recall reading some open source information from Googzilla on this function. My hunch is that Mahalo.com may find itself under pressure if the GOOG makes available some of its nifty automated question answering methods. At that point, question and answer services that charge and have humans in the mix will face a long, tough road to success.
Stephen Arnold, January 27, 2009
Video Sites and Search
January 27, 2009
ZDNet posts interesting statistical data, a bit like the old Predicasts’ File 16 data on Dialog minus the type charges. On January 23, 2009, Alex Moskalyuk summarized the comScore video site traffic data. You can view his original post here. I took a look at the data and did some quick fiddling around. I was interested in how much additional traffic one of the video sites listed would have to generate to pull even with YouTube.com in the month reported in the table. Here’s my rework:
| Property | Viewers, 000 | Share | Traffic Needed |
| Total Internet | 146,064 | 100.00% | — |
| Google Sites | 97,928 | 67.045% | — |
| Fox Interactive Media | 58,115 | 39.787% | 39,813 |
| Yahoo! Sites | 39,956 | 27.355% | 57,972 |
| Microsoft Sites | 34,979 | 23.948% | 62,949 |
| Viacom Digital | 27,109 | 18.560% | 70,819 |
| Hulu | 22,456 | 15.374% | 75,472 |
| AOL LLC | 22,442 | 15.364% | 75,486 |
| Turner Network | 20,735 | 14.196% | 77,193 |
| Disney Online | 13,028 | 8.919% | 84,900 |
| Time Warner – Excl. AOL | 12,564 | 8.602% | 85,364 |
| Source: comScore | |||
What’s interesting is that Hulu.com, to pick one video site that gets a lot of hype as a YouTube.com competitor needs to attract 75,486,000 more visitors to match YouTube.com’s traffic. What surprised me what the strong showing of Fox Interactive Media. Fox needs only 39,813,000 more visitors to reach parity with YouTube.com. I don’t watch video but it seems that Google’s 67 percent market share in video may not be as solid as its 70 percent share in Web search. The gap in video is narrower at least for Fox.
Stephen Arnold, January 27, 2009
ChaCha: Social Search Dances for Big Money
January 27, 2009
About 2.5 hours north of the hollow where the Beyond Search goslings write Web log posts is Indianapolis, Indiana. A suburb of Indianapolis is Carmel, Indiana, population 40,000. Carmel has a search engine, and that search engine has “secured $11 million of a $30 million Series C round.” One of the investors is Bezos Expeditions which brushes up against Amazon. Good news for ChaCha.
ChaCha.com is a question-and-answer system. The company reports that it has three million people on record as having tried the system. In terms of traffic, here’s Google Trends for Yahoo.com, Ask.com, and ChaCha.com.
Here’s the interface:
To get the results, the user then creates an account shown below:
The company sends the activation code to a mobile phone. For me, the system reported unknown mobile device.
After working around that problem, the system did not have an answer to my question, “What is a Google Search Appliance?”
I have this company in my monitoring list. Right now, I don’t feel comfortable offering too many “from the hollow” comments. With more money, ChaCha.com should be able to raise its profile and demonstrate its business model to me.
Stephen Arnold, January 27, 2009
Knol: One More Thing
January 26, 2009
The GOOG’s Matt Cutts, writing on his personal Web log, offers up a parental “Four Things You Need to Know about Knol.” Gentle reader, the story is here. I assume the use of “you” by Mr. Cutts meant me. I looked at his points from my goose-like perspective.
I liked the idea that Google doesn’t favor its own products and services. The assertion may be accurate in terms of Knol but if you search for “enterprise search” you get some results that place Google as the seventh hit in the results list this morning. There’s an ad for a Google Webinar. This supports the assertion that Google is not favoring Google services.
Second, he points out that Knol is “doing fine”. This is a bit like Amazon talking about “objects” in its cloud services. The problem is what’s an object and what’s “fine” mean. Knol has about 100,000 articles. I assume that 100,000 is fine. If so, then why is there a Knol for Dummies campaign underway here?
Third, the Knol team is moving. I agree. Subtle changes creep into Knol; for example, the notion of “authoritative” has obviously been tweaked by the Knol team. Mr. Cutts enjoins me to “write a quick article or put some information on the Web.” My question is, “What’s authoritative mean?’
In short, the “four things” are interesting. The one thing that my research Knol is / was supposed to do was provide inputs to the Google knowledge bases. “Some information” does not match up with the disclosures in Google’s public documents–for example, US20070038600–about its knowledge bases.
Knol certainly warrants observation. More on Knol appears in my forthcoming Google study, The Digital Gutenberg.
Stephen Arnold, January 26, 2009
More on Google and Control
January 26, 2009
Search Engine Journal’s “Unable to Change Googlebot Crawl Rate via Google Webmaster Tools” provides some insight into Google’s ability to implement variable controls. The article addresses the situation in which a Webmaster wants Google’s indexing robot to “crawl faster”. For some Webmasters, the Google administrative controls don’t work. Ann Smarty provides some information about this situation. She ignores one reason of which I thought: Google wants to control what happens with regards to indexing a specific site. The GOOG’s ability to implement fine grained controls makes clear what Google’s resources for control of its sprawling infrastructure is revealed in this write up in my opinion.
Stephen Arnold, January 26, 2009
Google and Video Playing Technology
January 26, 2009
I don’t pay much attention to online video. The trophy generation and the short attention span types do. Google tried its hand at its own video player and then shifted to Flash. If you care about video, you may want to check out Google’s invention disclosed in US20090024923. The abstract for this said:
Embedded Video Player Abstract A system, method and various user interfaces provide an embedded web-based video player for navigating video playlists and playing video content. A Web site publisher can create and store a video player with customized parameters (e.g., player type, appearance, advertising options, etc.) and can associate the player with a playlist of selected videos. The stored video player is associated with a player ID in a player database and can be embedded in a Web site using an embed code referencing the player ID. A user interface for the embedded player provides controls for controlling video playback and for controlling the selection of a video from the playlist.
Why is Google noodling a player? Two reasons. Why help out Adobe? Get more control over the experience in the browser or composite application. Google wants control to have some response to this type of situation. The video push is a real deal. Google has a cluster of video inventions, which signals to me an initiative.
Stephen Arnold, January 26, 2009
Information Technology Experts
January 26, 2009
With the proliferation of clueless management, information technology carpet baggers have had a willing if uninformed clientele. I know. You want to tell me that your clients are well informed and understand your expertise. You may even have a core competency. But when you get to be my age, you learn you don’t much at all. What you do know is changing quickly. With each passing day, my knowledge loses currency and I become dumber. Not surprisingly I have, like the turtles that once lived in Beargrass Creek, become cautious, slow moving, and conservative.
What happens when clueless clients encounter carpet baggers? I will get to that in a moment. I want to make sure you know what a carpet bagger is. According to the Merriam Webster online dictionary, a carpet bagger is “a Northerner in the South after the American Civil War usually seeking private gain under the reconstruction governments.” When I use the term, I put a spin on the connotation; namely, “a nonresident or new resident who seeks private gain from an area often by meddling in its business or politics.” I also spell the word with a space between carpet and bagger to remind myself that these were cheap briefcases in the 19th century.
Image source: http://i175.photobucket.com/albums/w127/CAITIpix/carpetbagger.jpg
Consultants who are carpet baggers may not recognize that their firm and its employees are practicing this type of work. Today 20 year olds don’t want to hear from an intellectual turtle in Kentucky that their analysis of a situation might be uninformed. Better to terminate the deal. I fired a client in Manchester, England. My radar lit up, and I jumped off that reed boat.
I read the Bloomberg story here with the chilling headline “Price Waterhouse Auditors Arrested in Satyam Inquiry.” The story does a very good job of explaining how two Pricewaterhouse Coopers’ professionals were arrested for “conspiracy and co- participation”. We have an allegedly crooked information technology consulting firm and allegedly crooked financial consultants. What’s that mean for customers of these two outfits? Here are my thoughts:
- Complex technology requires that an organization’s senior management understand the business problem, the technology that will be used to resolve the problem, and the risks involved. A failure to assume this responsibility makes it easy for trophy-generation consultants and nice guys like Mr. Madoff to practice their allegedly dark arts.
- As complexity of software and systems increases, the likelihood of improper behavior seems to me to go up. Look at the dust ups over content management and enterprise search systems that don’t manage content and can’t locate information. A specific example is the US government’s decision in one matter to terminate two companies’ contracts. I am not sure what subsequent legal hula hooping took place.
- Some vendors’ willingness to allow marketing to be science fiction and technology a work in progress seems to be spreading. Clueless clients or lazy procurement teams just want to be told that a system will do the job. Whether that system can or will do the job is of little concern. Turnover and other uncertainties often mean that there will be no consequences for flops.
I have been around too long to think that any change will take place to inject the clueless managers with knowledge. Hopefully the present financial crisis will allow the those with knowledge and a desire to behave in a responsible manner to make some gains. In the meantime, it’s probably a good idea to recalibrate your use of Satyam’s services and PricewaterhouseCooper’s advice.
Stephen Arnold, January 26, 2009
The Zune Effect
January 26, 2009
Activenetwork has a short news item “Frankly Speaking: For Microsoft, the Pain Is Just Beginning” here. The short item’s headline stated what few want to believe. The snippet included this sentence, “…It’s big because it means Microsoft has begun to hit bottom… Microsoft has never learned the lessons of the original IBM. My thought was that this statement is not 100 percent accurate. Microsoft’s combination of multiple initiatives, technical tangles, and market missteps deserves a name. I suggest “the Zune effect.” Apple trundles happily along. The Zune effect. Google chugs forward. The Zune effect. A working definition, “A mix of bad timing, technology issues, and misreads of the market set the stage for a rough journey.” Thoughts?
Stephen Arnold, January 26, 2009
Word and C#: Now That’s User Friendly
January 25, 2009
This item is not directly about search, but it has to do with content creation. Close enough. The headline that stopped me in my web footed tracks was, “Take the Pain Out of Creating Word Documents by Using C# and XML” here. To be a little fair, Chris Bennett is writing to developers. But the inclusion of a reference to Microsoft Word almost guarantees that non developers will see his write up. The idea is to put a chunk of code on a server that converts whatever is on a Web page to a Word file format. He provides a clear explanation. I particularly liked the detail for the “transformation method” but you may not be as keen on reading scripts as I. You can convert an XML into a Word document using XSLT. My thought was provide the XHTML or XML and a link that spits out a PDF. You will see this approach in action when my new Google patent search system becomes available. One of the new vendors of super fast search technology will provide the stallion for my service. Watch for more details. We’re shooting to release in the first week of February 2009. No C# required either. Free service. More Google open source information. With world finally discovering Ramanathan Guha, I thought it would be useful to provide access to documents for a method that’s approaching the age of four. We don’t want the pundits to rush too quickly toward understanding Google. That effort would take time away from figuring out how to convert a Web page to a Word file.
Stephen Arnold, January 25, 2009
-
- Subscribe to Beyond Search
Feature archive
News archive
-
