Monday, 22nd November 2010
Some Very Strong Words About Google Print Initiative From a UK Columnist and Other Google Bits
Let's begin with news of a new limited trial from Google and then to comments about Google book digitization.
Google has started a limited test of a plug-in that allows those who use Google Documents to work with Microsoft Office Products and vice versa. The technology comes from DocuVerse, a company that Google acquired in March 2010. The new product is named Google Cloud Connect for Microsoft Office and the test announced today appears to be for Google enterprise customers.
Btw, Microsoft offers a free "cloud" product named Microsoft Office Web Apps. It provides free web-based access to limited but still useful versions of Microsoft Word, PowerPoint, Excel, and OneNote. Documents are stored in the cloud and each account gets 25GB of free space via MS SkyDrive. You do not have to have a license for MS Office to use this service.
Now, comments about Google and book scanning.
Today, The Guardian* published some extremely negative comments about the Google Print Initiative (GPI) that we've read in a long time. These comments were written by Robert McCrum, the "On Books" columnist for the paper/web site.
The column's is headlined, "French publishing giants cave in to Google's great copyright heist."
"Will no one stand up to Google? French publishers used to be in the vanguard of opposition to the internet giant's mass digitisation programme, the so-called Google Print Initiative (GPI). Not any more. Last week, Hachette, France's biggest publisher, made an agreement with Google to scan thousands of out-of-print French titles for Google's online library."
"On the face of it, Google has pulled off yet another copyright heist under the skull and crossbones of "free content", putting yet more pressure on all the remaining hold-outs in the library world."
"Google (motto: "Do No Evil") insist that they are working for the good of the reader, liberating otherwise moribund texts from the darkness and isolation of library shelves. That's been its consistent position, but I just don't buy it, long-term.
"That day has not yet come, but as the ebook revolution gathers momentum in the USA and worldwide, I predict that Google will find a way of "revisiting" the noble, altruistic stance of the GPI."
The column also includes a mention of a recent article an other comments from Robert Darnton about a national digital library for the U.S.
Btw, Roy Tennant analyzes the Darnton plan along with another national digital library plan by novelist and TeleRead founder, David Rothman. Rothman's planwas published by The Atlantic earlier this month.
"Dueling National Digital Library Visions" by Roy Tennant, Library Journal.
Rothman also comments on Tennant's post.
* The first column on this page includes a bit of biographical info about Mr. McCrum.
That's not all on the Google front.
In another Library Journal blog post (online today), Roy Tennant writes some thoughts about how Google doesn't do well in responding to questions or comments from users. His post is titled, "Google to Users: We Donít Want to Hear From You."
Roy's comments are 100% on the money. Earlier this month it took three days for Google to respond to a ResourceShelf post about some inaccurate polling place info in their election info database. It used to take a day at the most to get a response. The post that documents our experience is here with a follow-up here.
For a story we will post after the holiday we've been monitoring the web search help forum for the past few months and have seen very few responses to questions and complaints that individual users. This is also the case even if the same question is being asked by a large number of users over the period of a few days or longer. The only time we've seen Google say something in a forum is when they have something to announce.
Of course, Google can't answer every question or reply to every complaint but if the same question is being asked over and over again maybe it's time to share some info?
Monitoring the web search forums for the past few months have been extremely educational and something that you might want to do. Why? First, they provide real world examples of problems users have (the users who can find where to post an issue or concern) with searching and also documents what many of us know, some users need some very basic search training, a 30-40 minute introduction would do wonders. Second, the spyware/malware problem is larger than we could have ever imagined. Many questions asked in the forum lead us to believe that spyware/malware is likely the problem or part of the problem. An example, would be Google and other web sites not to load, take the user to the wrong page after they click the search button, the Google homepage and/or pages format improperly on the screen or when printing, are just a few examples of many.
Kudos Roy on the blog post and we will have more to say on this topic after the holiday.
Finally, a story first reported last week continues to get some attention.
The reports deal with Google possibly giving better placement on a results page to services that come from Google. Stock quotes and health info are two examples.
Both Technology Review an the IDG News Service have articles about the research and analysis by Harvard Professor, Benjamin Edelman.
You can read Professor Edelman's report titled, "Hard-Coding Bias in Google "Algorithmic" Search Results," here.
Here's the Abstract:
I present categories of searches for which available evidence indicates Google has "hard-coded" its own links to appear at the top of algorithmic search results, and I offer a methodology for detecting certain kinds of tampering by comparing Google results for similar searches. I compare Google's hard-coded results with Google's public statements and promises, including a dozen denials but at least one admission. I tabulate affected search terms and examine other mechanisms also granting favored placement to Google's ancillary services. I conclude by analyzing the impact of Google's tampering on users and competition, and by proposing principles to block Google's bias.
Some web search and IR experts might say that every search engine algorithm has to have some form of bias towards what is or is not relevant for the users of the system. It's a very good argument.
However, the primary issue (in our opinion) that Edelman points out is that Google--over the course of many years--has used the word "unbiased" to describe their search results and have also said that there is no manual manipulation of the results while at the same time offering more and more results in what Google calls a "OneBox". However, Google says one thing in discussing their web search product but is actually doing something else.
If Google would have not used words like unbiased over and over again to market their product it would likely be less of an issue. Also, some disclosure and explanation about what a OneBox means in terms of unbiased results being placed at the top of results pages mean for algorithmically generated results would have also helped. Said a different way, every result can't be at the top of a results page. Again, it becomes less of a problem if you disclose and explain before or as changes take place.
If nothing else, Edelman's in-depth research and report should make for some lively discussions.