With the close of the year approaching, this issue marks the 14th week of DLTJ Thursday Threads. This issue has a publisher’s view of Amazon’s strong-arm tactics in book pricing, research into the possibility that academic authors could game Google Scholar with spam, demonstrations of how Amazon’s Mechanical Turk drives down the cost of enlisting humans to overwhelm anti-spam systems, and a story of multispectral imaging adding information in the process of digital preservation.
As the new year approaches, I wish you the best professionally and personally.
DLTJ is in a bit of flux now. After updating some underlying packages on my 9-year-old Gentoo-based personal server, I’m finding that I can’t start the web server process without the 1-minute load average climbing to roughly 60 in the span of about 5 minutes. (Translation: the machine is working very hard but getting nowhere fast.) Increasingly, the server has also been hard to update — lots of strange errors, etc. — so after 9 years, it is clearly time to rebuild it. In the interim, I’m in the process of moving the blog over to an Amazon EC2 cloud computing instance. If you see this post, you are reading it on that virtual server. The DNS entries should catch up with the migration in a couple of hours.
This is a preview of DLTJ In a State of Flux. Read the full post (165 words, 40 seconds estimated reading time)
When I say “<blank> is a question answering system. A question can be posed in natural language and … <blank> can come up with a very precise answer to that question” — what comes to mind to fill in the <blank>? If you guessed a system developed by IBM to appear alongside human contestants on Jeopardy, you’d be right. That quote comes from video posted by IBM earlier this year that is the topic of the first DLTJ Thursday Threads entry. This weeks other entries look at possible erosions of copyright first sale doctrine, the state of open access publishing, and a proposition for new definitions to terms of art in data modeling.
If you find these threads interesting and useful, you might want to add the Thursday Threads RSS Feed to your feed reader or subscribe to e-mail delivery using the form to the right. If you would like a more raw and immediate version of these types of stories, watch my FriendFeed stream (or subscribe to its feed in your feed reader). Comments and tips, as always, are welcome.
This is a preview of Thursday Threads: Digital Reference Librarians, First Sale Danger, Open Access, Data Modeling. Read the full post (1279 words, 5:07 minutes estimated reading time)
I’m trying an experiment over the next couple days/weeks. I’m turning off the reCAPTCHA requirement for blog commenters (the figure-out-these-words-and-type-them-in anti-spam scheme I turned on three and a half years ago). The only automated scheme in place now is Akismet. This change was made Friday night, and over the weekend a few spam comments got through to “approved” status while 550 were in the “spam” queue. With reCAPTCHA in place, I would typically only get 10 or so comments that would make it through reCAPTCHA only to get caught by Akismet (and none through to approved comments). I could easily go through 10 or so comments a day looking for ones that would accidentally get trapped (maybe one a month), but I’m not going through 200 or more a day. So, if you comment on DLTJ and don’t see it immediately posted, please do let me know and I’ll fetch it out of the spam queue.
I was reading a story last week about the Linux Foundation‘s third annual report [PDF] of the Linux kernel, and in it was a section that talked about the affiliation of the programmers that contributed to the development of the kernel. This got me thinking about the affiliation of programmers in the library open source community. More on that after a brief detour to explain what the “kernel” is.
This is a preview of When Closed Source Companies Contribute to Open Source Communities. Read the full post (908 words, 3:38 minutes estimated reading time)
This week’s Thursday Threads highlights includes two legal cases that bear watching. The first is the case of SkyRiver/Innovative Interfaces versus OCLC (covered on DLTJ previously); now that the case has been moved to OCLC’s home court (the federal district court located in Columbus, OH), it is asking for the case to be dismissed. The second legal cases is the UCLA streaming media case, with issues ranging from fair use to licensing terms to DMCA violations; if this one goes to trial we might get some new case law surrounding the intersection of copyright and libraries. The remaining two pieces are a look at how publishers (and librarians) should avoid paving cow-paths and the origins of the hash symbol.
This is a preview of Thursday Threads: OCLC Moves to Dismiss SkyOCLC, UCLA Sued For Streaming, Paving Cow Paths, Origins of #. Read the full post (1385 words, 5:32 minutes estimated reading time)
I’m pleased to announce that the Fall 2010 issue of NISO‘s International Standards Quarterly (ISQ) is done and available online to NISO members and ISQ subscribers. Print copies are scheduled to be mailed on December 28th. The individual issue is available for purchase (see the form link to on the issue homepage), and some of the articles are freely available on the NISO website. The theme for the issue is resource sharing, and I was privileged to be the guest editor for the issue. Included below is my introduction letter to whet your appetite for the full issue.
This is a preview of “Do More … With Someone Else” — Guest Editor Introduction to NISO ISQ Fall Issue. Read the full post (700 words, 1 image, 2:48 minutes estimated reading time)
It has been another busy week, and unfortunately Thursday has slipped into Friday. There have been a few updates to earlier Thursday Threads items, so I’m turning this into “Friday Followups” instead. We’ll attempt to get back new items next Thursday, but in the meantime take a look at these updates.
The highlights of the past week are around publishing — first with a model proposed by Eric Hellman in which consumers can pool enough money to pay publishers to “set a book free” under a Creative Commons license, then with an announcement by the University of Pittsburgh offering free hosting of open access e-journals. Since we have to be able to describe and find this content, their bibliographic descriptions are important; John Wilkin proposes a model for open access to elements of bibliographic descriptions. Rounding out this week’s topics are a report of a master’s degree program in business using Facebook, and tips for planning an unconference meeting.
This is a preview of Thursday Threads: Open Publishing Alternatives, Open Bibliographic Data, Earn an MBA in Facebook, Unconference Planning. Read the full post (1350 words, 1 image, 5:24 minutes estimated reading time)