Category Archives: Tech

Tech

EFF, AT&T, NSA, and Voting

I realized a little while ago that any money I dumped toward Washington would have a far greater effect than my one-in-several-million vote(s). To that end, I started financially supporting the EFF, and so far they seem to be putting my money to good use. Here’s the latest:

San Francisco – A federal judge has refused to dismiss the Electronic Frontier Foundation’s (EFF’s) case against AT&T for collaborating with the NSA in illegal spying on millions of ordinary Americans, setting the stage for a congressional showdown over proposed dramatic changes in federal surveillance law.

EFF filed the class-action suit against AT&T in January, alleging that the telecommunications company has given the National Security Agency (NSA) secret, direct access to the phone calls and emails going over its network and has been handing over communications logs detailing the activities of millions of ordinary Americans. The government intervened in the case and asked that it be dismissed because the suit could expose “state secrets.” But Thursday, U.S. District Judge Vaughn Walker refused: “The compromise between liberty and security remains a difficult one. But dismissing this case at the outset would sacrifice liberty for no apparent enhancement of security.”

“We are gratified that Judge Walker rejected the government’s overbroad claims of secrecy, and that our case on behalf of AT&T customers can go forward,” said EFF Staff Attorney Kevin Bankston. “Judge Walker correctly found that the government, after having already admitted to and extensively commented on the NSA’s spying program, cannot now claim that it is a secret and sweep AT&T’s role under the rug.”

The whole post on the EFF blog is here.

Daily Work Log 2006-07-21

I arrived at work this morning to find the AC off. There was a scheduled shutdown last night, which I knew about, but I neglected to apply that knowledge and stay at home this morning.

I’ve spent a little time this morning trying to install Ubuntu Linux on an external drive attached to my PowerBook 17″ (G4). I had some difficulty getting the advanced install CD to do what I want, so now I’m downloading the LiveCD to try that.

…

The LiveCD wouldn’t install a bootloader either. I don’t have time to mess with it today.

We spent some time meeting the the IB/VRE people, and they’re off to lunch with Rob and Umar for the moment. Therefore, I’m going to do my weekly review, because it’s that day.

…

Finished the weekly review. I’m really feeling tired today, and there’s a lot of the day left. I’m taking care of my end-of-week activities, and the next one is to windex my desk (it’s glass). Therefore, I think that’s enough for the day.

Today has not been very productive. This week has not been very productive. I’ve done a lot of things, most of which really needed to be done, but my projects have not moved forward satisfactorily. Blah.

Research and Reference Management: Part The Second

Yesterday I posted a flow chart and a description of the first half of the process that it diagrams. Today, I’ll explain the more elaborate second half. First, for convenience, here’s the flowchart again:

The first part of the process gives me a prioritized list of articles to read. It helps, when I have time to read articles, if I can simply go to the list, get a few PDFs, print them, and read them. Therefore I typically go down the list and acquire PDFs for tens of articles in one go. That’s covered in the diagram. For each item on Cite-U-Like, there’s a link to PubMed. The important thing about that is that on PubMed, (while I’m on the University’s network) there are big icon-button things indicating whether I have full-text access to the article, and where. If I have full-text access, I follow the link and download the PDF. I then upload it to Cite-U-Like, which does three things for me.

  1. It keeps the file online for universal access (for me only)
  2. It renames the file to something sensible for me
  3. It puts the file in a nicely organized system with the rest of my article information

If, however, the article is not available to me in full-text (some institutions have more extensive access than others… *cough*), then I have to follow the “no” branch of the flow chart. First, I check the library’s online catalog to make sure that the library does not have the issue in question. They might have it. Ironically, they seem to have extremely obscure journals, but none of the really relevant ones. At least in my field. They also might have the article in J-STOR or Ovid or something like that. If so, then it’s time to descend into confusing multiple-electronic-database hell. I left that off the flowchart.

Anyway, so, the more likely case is that they don’t have it at all. That’s actually a relatively painless scenario, and it’s getting better. Most universities, including mine, have a nice online interface to request photocopies of articles via Inter-Library Loan (ILL). You request your article, and in a day to several days it arrives. Historically, the article would arrive in the form of a mailed or faxed photocopy. This was sometimes inconvenient. Why?

  1. You have to go pick up the article. If you’re doing this several times a week it can get pretty time consuming and annoying.
  2. The article is not digital. If you want to take the copy with you somewhere, you need to lug the dead trees. It is possible to scan the article using an auto-document feeder on a nice digital copy machine. This is yet another step, though.
  3. Sometimes they do a really bad job of copying
  4. Color figures usually become something between mostly useless and entirely useless after being xeroxed, especially if it’s not done with a careful eye for darkness settings.

The new thing is that most articles are delivered electronically. You get an email saying, “your article is in!”, you go to the ILL site and download however many arrived-articles have piled up, and you’re good to go. Of course, the issues of bad copying still apply, since nobody seems to use color scanning.

As you can see from the flowchart, no matter how the articles arrive, you can make them into PDFs somehow, and get them back into the Cite-U-Like library. From there, you can download the PDFs in order on the prioritized To Read list, and churn through them.

How to properly read and make use of the knowledge you gain from reading is worthy of several other blog posts. Questions?

Research and Reference Management: Part The First

To help me keep up with daily blogging, I’m going to try keeping the post window open all day so that I can add stuff easily.

Also, how do you like the new favicon? I thought it would be appropriate.

Today I’m spending a lot of time pulling references from various papers, adding them to my Cite-U-Like collection, trying to get the articles in PDF form if possible. Then, if Tulane doesn’t have the article I want via online access, I request it via inter-library loan.

I feel like I should have a flowchart for this. On my next break, it shall happen… (time passes)… Here’s the flowchart, and though it’s missing a few details, it’s mostly complete.

Here’s a little explanation. When I read an article, I circle the numbers of references that are attached to concepts I’m interested in. So, if the paper says, “widgets were frobbed using Magical Frabulation [12]” and I want to know more about Magical Frabulation, I circle [12] in red pen and continue reading. When I’m done reading the article, I go back to the article, find each circled number, and then circle the number on the actual reference in the reference section at the end. Then I use my hubmed search engine in Firefox to look up each reference. Usually I search using the last name of the first author and some pertinent words from the article’s title. If the first author has a last name like “Smith” or “Li” (do you know how many people named Li are on PubMed?) then I use a different author, if available.

I’m using the TabMix Plus extension, so every time I enter a Hubmed search a new tab automatically pops up. If the search worked, I carry on. If not, there’s a handy little search box at the top of the Hubmed results page that I use to try a different set of search terms. If it’s a medical article, I’ll eventually find it. If not, I have to look elsewhere and then my procedure has to be modified a lot. I left that out of the flowchart because it’s very case-specific.

Assuming it’s on Hubmed, I use my Cite-U-Like bookmarklet to post the item to my library. That’s the end of the first part of the process. I do that first part for each reference I want from the article. Depending on how I ranked those items in terms of how badly I want to read them, they will then show up in the proper order on my Cite-U-Like To Read list.

Daily Work Log 2006-07-12

In an effort to increase my blogging, I’m going to start making daily logs before I leave work.

I spent most of today reading journal articles to prep for writing the discussion and introduction of my paper. The methods and results have been coalescing for a while, and I think are nearing completion. I received some helpful comments from my coauthor via trans-Atlantic fax this morning, and addressing those comments is next on the agenda for that project.

In other news, I’ve begun work on my prospectus. It needs to be defended before I leave for Baltimore, and that doesn’t give me very long. Luckily a lot of the groundwork has already been done for other reasons, so it should be mostly synthesis. The department does require that I write it up in NIH format in addition to the standard format required by the Dean’s office, so I get to join the wonderful world of people who’ve filled out all of that stuff. I hope it’s simpler than my AHA fellowship application was.

Tomorrow I have to take the day off to go to Mandeville, so I’ll follow my first day of daily posts with a day of no posting. My apologies!

One last note before I go home. Last night I found a great meditation timer for Palm OS, PocketDoan. It’s good for much, much more than meditation, though. People have siezed on it for use in work “dashes” particularly the (10+2)*5 method discussed at 43 Folders. I used it for that today and it was fantastic! I only got 3 dashes in today, and I only managed to finish one of the three without an interruption (though pausing is easy). I’m going to keep trying this and try to squeeze more 1-hour periods in per day. My ideal goal would be 7-8 per day.