Measuring when someone doth protest too much.

#alazony

A few minutes ago in Brooksville, Maine.

Boolean searching within a web page.

Does anyone know a Chrome extension or other tool allowing us to run boolean searches within a web page, rather than across web pages? 

With such a tool, I could search for <A or B> and Chrome would highlight all the A's and all the B's on the page. This would be very useful, for example, when I want to search for <"mark twain" or "samuel clemens">, <edges or arcs>, or <ribozyme or "ribonucleic acid enzyme">.

I'd use this every day. Or at least I'd use the disjunction operator every day. I can't think of a use for conjunction within a page. If I search for <A and B> (when A and B are distinct), then I'd get no hits, since no word, no quoted phrase, and no boolean compound, is both A and B. 

[Update: I retract this claim in the comments below. Even the conjunction operator could be useful.]

But because Chrome supports partial matches, I can imagine using negation. For example, a search for <para> will highlight instances of "paradise", "paragraph", and other para* words. Hence, I might want to search for <para not paragraph> in order to highlight just the other para* words on the page.

Then of course I could also search for <(para not paragraph) or eden>. A good boolean search engine would support many levels of nested parentheses. 

Since much of the power would come from code libraries already written, I imagine it would be easier to provide full boolean power than boolean power minus conjunction. Hence we can forget the fact that conjunction would be useless in this domain.

So where's the power? And why hasn't Google -- Google of all companies -- already built this in to Chrome?

#boolean #search #chrome #google

cc +Google Chrome 

The last Word.

I recently wrote a letter of recommendation for a colleague who was applying for a senior position at a major public university in the US. I wrote a strong, thoughtful letter. At the top I included my addressee's full name, title, address, and the date. At the bottom I included my own full name, position, and address. The letter arrived before the deadline.

But it was plain text in the body of an email. Big mistake!

The chair of the search committee soon wrote back to me:

Thank you for your letter regarding X. I apologize for making more work for you, but our process here requires official letters (preferably on letterhead)....We will certainly use your letter at this stage as we conduct our preliminary screenings, but before a candidate may be brought to campus for an on campus interview we require...official letters....Note: this letter can be sent as an attachment to an email.

I didn't understand this, and I still don't. He preferred letterhead but didn't require it. He didn't mention a handwritten signature, and his willingness to accept an email attachment suggested that he didn't require a handwritten signature either. I replied:

I'm sorry to have to ask for clarification. You say that the letter should be "official (preferably on letterhead)." But I'm not sure how it could be more official. It's from an individual, not from an institution. I wrote it and put my name on it. I'm not sure what you are asking, especially if letterhead is unnecessary.

He replied:

I'm sorry not to have been clearer. Your letter is pasted into an email. When I produce a copy of this email, it will not be formatted as a letter. I would want something composed in Word and formatted in the way a letter is formatted.

What's wrong with this picture? 

1. He really did want me to shift from plain text in an email to rich text in a Word doc. (He was silent on the font and margins.)

2. Word was not enough. He also wanted me to format my letter "as a letter" even after shifting it. But I had already done that. As Sam Seaborn said in an episode of West Wing, "I used punctuation and everything." 

3. He really did say that these changes would make the letter "official".

I won't even get into his impression that I wrote the letter in a program like Word and "pasted" it into an email. (I did write in another editor and paste it in. But it was a plain-text editor.) Nor will I get into his request that the letter be "composed" in Word and not merely formatted and delivered in Word. 

I wanted to quarrel or decline, but I was stuck because I didn't want to hurt my colleague's chances. Moreover, I didn't know whether the committee chair was responsible for these rules or whether he was in the sad position of having to enforce foolish rules imposed on him from above. So I replied that I'd send a Word version of my letter shortly, and added only this:

I must say, however, that formatting a letter in Word does not make it any more official.

I've served on more than a dozen search committees, most of the time as chair. I've written hundreds of letters to other search committees, and in the last decade the vast majority of my letters have been plain text in the body of an email. I've run into the need for printouts on letterhead with hand signatures. I accept those requirements, even if they irritate me, because authentication is not arbitrary in the way that Word format is arbitrary. But in this case, the chair's requests had nothing to do with authentication, only with a provincial concept of officialness. 

Is this new? I think I listen to the academic grapevine as much as anyone, but I've never heard of this before. Have you? If it's a trend, is it rising or falling? 

The harm caused by myths about open access.

For a vivid sense of the harm caused by common misunderstandings of OA, read the comments in this survey carried out at the University of Saskatchewan in November 2012 and released this month.
http://ecommons.usask.ca/bitstream/handle/10388/6290/Report%20-%20USask%20OA%20Faculty%20Survey%20Results.pdf?sequence=1

It's depressing how many respondents who like the idea of OA in theory turn away from it in practice because they believe one of three particular falsehoods about it:

1. All OA is gold OA (through journals). 

The truth: Green OA (through repositories) is an alternative to gold OA, and even more plentiful than gold OA. There are several ways to arrange for permission to provide green OA even for work published at the very best peer-reviewed journals.

2. All or most peer-reviewed OA journals charge publication fees. 

The truth: Most (67%) charge no fees at all. In fact, the majority (75%) of non-OA journals charge author-side fees and only a minority of OA journals do so.

3. All or most fees at fee-based OA journals are paid by authors out of pocket.

The truth: Most fees (88%) at fee-based OA journals are paid by the authors' funder or employer. In fact 96% of authors who make their peer-reviewed articles OA pay no fee at all, because they make their work green OA rather than gold, because they publish in a no-fee OA journal, or because their fee at a fee-based journal was paid by their funder or employer.

For details and sources on my corrections to these misunderstandings, see "Open access: six myths to put to rest," The Guardian, October 21, 2013.
http://www.theguardian.com/higher-education-network/blog/2013/oct/21/open-access-myths-peter-suber-harvard

And more here:

* How often do authors at fee-based OA journals pay the fees out of pocket? Feb 1, 2013.
https://plus.google.com/+PeterSuber/posts/KEff1VvD7Ur

* Once more: correcting the canard that OA always or usually costs authors money, Feb 13, 2013. 
https://plus.google.com/+PeterSuber/posts/QqMhLjodN1T

* How to make your own work open access, originally Oct 2012 but frequently updated.
http://bit.ly/how-oa 

#oa #openaccess #apcs #myths #misunderstandings

When did "scientist" become more common than "man of science"?

This Google Ngram has the answer. But guess before you look. 
http://goo.gl/zUT51u

Some other tidbits and quiz questions from the Ngram:

* "Man of letters" became more common than "man of science" about the same time that "scientist" became more common than "man of science". However, soon after that, "man of letters" and "man of science" declined in parallel, while "scientist" grew rapidly in frequency.

* "Scientist" peaked in the 20th century. But which decade? Again, guess before you look. It's been in decline ever since. 

* "Scholar" was more common than "scientist" until just about the same time in the 20th century when "scientist" peaked. After that, both terms declined in parallel.

* Which related term is the the most frequently used today? Hint: It was nearly unheard of until the mid-20th century. 

I create the Ngram to follow up +Melinda Baldwin's fascinating article in The Conversation, "How ‘man of science’ was dumped in favour of ‘scientist’." (Recommended.)
http://theconversation.com/how-man-of-science-was-dumped-in-favour-of-scientist-30132

Milestone for the Open Access Directory. 

"The Open Access Directory (OAD) just sailed past four million views. The OAD is a wiki and depends on the OA community to keep it comprehensive and up to date. Please help us keep the OAD a definitive reference on open access....We welcome your contributions to the lists, ideas for new lists, and comments to help us improve."

#oa #openaccess #oad

I just signed this open letter and hope you will too. Please spread the word.

#copyright #tpp #eff

I'm very proud that this program is part of the Office for Scholarly Communication. Thanks and congratulations to Kyle Courtney.

The open-access policy from the Department of Energy.

The US Department of Energy (DOE) released its open-access policy today, after it was approved by the White House Office for Science and Technology Policy (OSTP) and Office of Management and Budget (OMB).

The DOE policy for peer-reviewed articles:
http://www.osti.gov/PublicAccess/
http://www.energy.gov/downloads/doe-public-access-plan

The DOE policy for data, released last week:
http://science.energy.gov/funding-opportunities/digital-data-management/

Initial coverage from:
* Nature (Richard Van Noorden)
http://blogs.nature.com/news/2014/08/us-department-of-energy-frees-up-access-to-research.html
* Science (Jocelyn Kaiser)
http://news.sciencemag.org/policy/2014/08/u-s-energy-department-make-researchers-papers-free

Initial comment from: 
* CHORUS (on which the DOE plan relies heavily)
http://www.prnewswire.com/news-releases/chorus-looks-forward-to-working-with-the-department-of-energy-to-advance-access-to-research-269845211.html
* SPARC (critical of DOE's reliance on CHORUS)
http://www.sparc.arl.org/news/sparc-responds-department-energys-public-access-plan

My own take: 

We'll soon have more OA than we had before, and that's good. Another major federal research-funding agency now has an OA policy, and that's good. I could go on about why these are good, and in a longer piece I would. But the policy has three significant weaknesses, and it's the first in a series of about two dozen federal-agency OA policies required by the Obama administration. Here I want to focus on the weaknesses in order to do what I can to head off similar weaknesses from other agencies.

1. The policy relies heavily on publisher web sites, when it should have relied on sites independent of publishers. 

The purpose of the White House OA directive was to assure public access to publicly-funded research. Reliance on publisher web sites runs counter to that purpose. Many publishers lobbied aggressively and deceptively against the type of policy that the White House now requires agencies to adopt. The only way to insure the cooperation of publishers is through regulation, and I strongly oppose the regulation of publishers. So do publishers and the White House, of course. (Aside: part of the genius of green OA mandates from funding agencies is that they regulate grantees, not publishers.) However, once we rule out the regulation of publishers, and depend on publishers to provide OA, we should not be surprised if some publishers deliver late, temporary, or selective OA.

The worst part of CHORUS is the recommendation to rely on publisher web sites, the worst aspect of the DOE policy is its concession to that recommendation, and the worst part of today's news is the White House approval of the DOE concession. 

2. DOE seems to think it will avoid the first problem by linking to OA copies of relevant articles in the institutional repositories of DOE-funded authors. But DOE doesn't require DOE-funded authors to deposit in their institutional repositories. Hence, this is smoke and mirrors. It's another reason to think that the DOE is not serious about assured OA, and is content with late, temporary, or selective OA.

Just as the chief fault of the gold-leaning OA policy from the RCUK was fixed by the green OA mandate adopted by HEFCE, DOE could fix the chief problem in its current policy simply by requiring DOE-funded authors to deposit in their institutional repositories. (For those without IRs, DOE could require deposit in some designated universal or residual repository like Zenodo.) This step would provide OA to all DOE-funded research, and provide it from sites independent of publishers.

3. The policy offers no reuse rights beyond fair use. This is a substantive problem because fair use doesn't suffice for the purposes of research. For example, it doesn't suffice for text mining or translation. But the lack of reuse rights is also a problem for important procedural reasons. The DOE policy is supposed to conform to guidelines laid down by the White House in February 2013. Those guidelines called for agency policies to "maximize the potential for...creative reuse." Nobody anywhere believes that fair use maximizes the potential for creative reuse. Fair use doesn't even increase the potential for creative reuse. We already have fair-use rights for all copyrighted works in the US, and to increase the potential for creative reuse we'd have to go further. "Maximize" is a strong word. Creative reuse requires open licenses, and maximizing the potential for creative reuse requires minimally restrictive open licenses, such as CC-BY for texts and CC0 for data. The real problem here is that the White House abandoned its own guidelines. After calling for strong reuse rights, it approved a policy that does nothing to increase reuse rights beyond what we already have. 

For my longer take on the White House OA directive of February 2013, looking at the guidelines for agency policies before any actual agency policies had been drafted or approved, see my article from March 2013.
http://nrs.harvard.edu/urn-3:HUL.InstRepos:10528299

#oa #openaccess #doe #ostp #obama #chorus #sparc #libre   #fair_use