Recently in IT Category

There is a very good (well, I like it) summary of how to use Orgmode as a day planner at Lost in Technopolis.

I've been a big fan and user of Remember the milk. I found it a great way to manage my task, and so on, in a GTD like manner. As a bit of an aside, the GTD methodology really works for me ... and every time I find work getting a bit out of control its because I've not been ruthlessly applying the principles of GTD.

Anyway, I've switched to using org-mode (aka orgmode). I don't recall how I stumbled across Org-Mode: Your Life in Plain Text. No before I go any further, let me say that I don't think orgmode it for everyone. Its an extension to the emacs text editor, and in this WYSIWYG world a texted based programme may be 'a bridge to far'. But, as an old hand a unix (albeit a a vi/vim user) the transition to emacs hasn't been too hard.

But why change? Is this just part of the ongoing search for the perfect tool [Andrew might comment here]? Is it just displacement behaviour (avoiding doing PhD work)?

As I said at the beginning of this post, when I find things spinning a little out of control (or what I've come to regard as control) it's because my GTD system isn't out of date/wack. So, moving to a new central store (orgmode) is a mechanism for me to get things back in order (and on track). So, today I've finished the migration from Remember the Milk to orgmode. No doubt I'll have some ongoing tweaking of orgmode ... but I have deleted my Remember the Milk account.

There are lots of useful resources on using orgmode in a GTD way. For example, this one, this one or this one. Most of the articles assume you have emacs up an runnign already (which isn't hard), and that you know how to configure emacs (which is harder). Fortunately, emacs now ships with orgmode included, so there isn't a lot of additional work that needs to be done. you can just jump right on in.

Orgmode understands lists and dates ... and a whole lot more; but lists and dates are enough to get started with.

First of all, one needs to turn on orgmode. There are a few ways to do this. I do it thought my .emacs file (which, is normally in your home directory; if it isn't there you can create it). In .emacs I have the following text.

;;
;; Peter's setup for orgmode in .emacs
;;
(add-to-list 'auto-mode-alist '("\\.\\(org\\|org_archive\\|txt\\)$" . org-mode))
(global-set-key "\C-cl" 'org-store-link)
(global-set-key "\C-ca" 'org-agenda)
(global-set-key "\C-cb" 'org-iswitchb)
(add-hook 'org-mode-hook 'turn-on-font-lock)  ; Org buffers only
(setq org-log-done t)
;;
;; Some more settings
;;
(setq org-agenda-files (file-expand-wildcards "~/org/*.org"))
;;
;;
;;quickly open my gtd file - "M-x gtd"
;;
(defun gtd ()
	(interactive)
	(find-file "~/org/gtd.org")
	)
;;

The main parts of this code are (1) the first line turns on orgmode for any file that is ends with .org .txt or .org_archive. (2) I keep all my orgmode files in a directory off my home directory, i.e., ~/org. any files ending in .org in that directory are scanned when orgmode puts together my agenda. (3) The last little bit of code opens my main 'to do' file called gtd.org when I execute M-x gtd (for non-emacs users that means pressing ALT-x then typing gtd and pressing return). This provides me with a quick way to get to my my to do list. Oh, the "org-log-done" line results in my to do items getting a date attached to them when I mark them as being done.

Now for some fun stuff. First I create my gtd.org file (using c-x c-w org/gtd.org ... that means Control-c Control-x and then typing the path/file name). I can then start making my to do list, vis:

* Work
** Review APR
** Revise article on governance
** Order textbook for INTBUS 202
* Home
** Pick up milk on way home
** Buy a copy "The Wire" (Series 5).

So far so good. I have to have my APR review done by Monday, so let's put a deadline beside it. Positioning my cursor somewhere on the relevant line I type c-c c-d mon (and press enter)

The line changes to be:

** Review APR
   DEADLINE: <2010-01-18 Mon>

And I what to schedule doing the article revision on next Monday. So, on the appropriate line, I type c-c c-d +2mon (and enter). The line changes to be:

** Revise article on governance
   DEADLINE: <2010-01-25 Mon>

I also want to schedule buying "The Wire" today. On the correct line I type c-c s and press enter. The line changes to be:

** Buy a copy "The Wire" (Series 5).
   SCHEDULED: <2010-01-16 Sat>

Now I have dozens of tasks to do. And looking at them all at once doesn't always add clarity. So let's look at my agenda. Pressing c-c a a will bring it up (no need to press enter). My agenda looks like:

Week-agenda (W02):
Monday     11 January 2010 W02
Tuesday    12 January 2010
Wednesday  13 January 2010
Thursday   14 January 2010
Friday     15 January 2010
  gtd:        Scheduled:  Buy a copy "The Wire" (Series 5).
Saturday   16 January 2010
Sunday: 17 January 2010
  gtd:        In   1 d.:  Review APR
  gtd:        In   8 d.:  Revise article on governance

And that's probably enough examples for now. So to sum up, c-c c-d marks an item with a deadline, c-c c-s marks an item with a scheduled date. Oh, and to save your file (the currently open buffer) use c-x c-s

I really like Zotero. The more I use it, the more useful and powerful I find it. Also, the technical support for the product (from the community of users and especially from Dan Stillman is outstanding.

Dan pointed out a really useful feature to me today. When adding items to Zotero, sometimes the title of the item is published ALL IN UPPER CASE or sometimes In A Mixture Of Upper And Lower Case (i.e., title case). It is really tedious to fix by hand; but there is a better way. I viewing mode (thats in the right-hand-side panel), right-click on the title and choose "Transform text->lower case". It can then be quickly tweaked to be sentence case.

Since moving to Zotero, I've found my 'work flow' to be much smoother. These few tips have made things even easier. I particularly find it useful when I've found a link to the full-text of an article, just to drag the link to the item in Zotero (and have Zotero auto-magically download into my library).

In writing my piece on referencing, I just Googled 'apa referencing' and here are the first page of search results.

I was surprised by the results. First, I used google.co.nz instead of google.com (I normally use the later), and secondly the University of Auckland's referencing resource Referen©ite didn't feature at all on the first page.

I see:

• www.waikato.ac.nz
• www.aut.ac.nz
• www.fmhs.auckland.ac.nz
• library.ucol.ac.nz
• library.curtin.edu.au
• library.canterbury.ac.nz
• library.manukau.ac.nz
• www.southernoceansoftware.com
• owl.english.purdue.edu
• aut.ac.nz.libguides.com (again)

But there is no sign of Referen©cite. I wonder google doesn't refer to Referen©ite on the first page (that's a rhetorical question; I know how search engines rank sites).

Here we tend to use nVivo for qualitative research/analysis. It's pretty robust, but the handling of PDF files leaves a lot to be desired. What nVivo seems to do is to convert the PDF into some sort of RTF-like format. Unfortunately the fidelity of the converted file isn't that good. In fact, it can be darn difficult to work with most converted files if they are anything more than plain-old-text.

The problem is so bad, and I am so keen to analyse/work with PDF files I'm thinking of changing to Atlas/ti. That programme handles PDF files in their native format ... there is no loss of fidelity when working with them.

What I can't decide is whether the learning curve/downtime is worth the change.

Sometimes, it is really necessary to display a complex formula; such as this:

I use MathTran to achieve this, rather than doing a complex server-side install of Latex. Then I can use a URL like the one below to dynamically display the equation above:

That reminds me, there is a Google API for doing charts on the fly. The simple example they give is:

Although much more complex graphs¹ are possible:

¹ Thanks to http://www.felipebarone.com/plot-function-google.php [now defunct] for this example.

Jo Badge talks a bit about her personal learning environment, and so I thought I'd draft out mine. I'm really web-centric and spend most of time 'in' Firefox (I sometimes think about switching to Chrome, but I'm hooked on Zotero).

I found it interesting to consider what is missing here ... there is no Powerpoint. I so rarely use 'slides'; often all I have is a single slide so students coming into class know they are in the right place. Increasingly, when I need slides I'm trying/learning to do them using Beamer. Having said that, for the odd complex set of ideas I do use MindManager in presentation mode to show how I see the connections between things.View image

Since writing this, I've come across Hull, Pettifer, and Kell (2008). They present an interesting figure, which seems to show how much I can actually do in Zotero (which is most of it).

Figure 4. A typical workflow for using a digital library representing a subset of the literature. Tasks represented by white nodes are normally performed exclusively by humans, while tasks shown in blue nodes can be performed wholly or partly by machines of some kind. The main problematic tasks that make digital libraries difficult to use for both machines and humans are ''GET'' (publication) and ''GET METADATA''. These are shown in bold and discussed further in the Identity Crisis section of this paper. doi:10.1371/journal.pcbi.1000204.g004

References

I've just moved to beta version 1.5 of Zotero. It was a trial.

1. My database wouldn't convert, so I had to upload it to their db repair tool.
2. Upgrading the Word add-in irrevocably changes your documents so you can't go back to 1.0. That isn't a problem in its self, but it looked as if the upgrade was hanging--it took 20 minutes to upgrade a 144 page, 220 reference document.
3. The sync option often gives errors ... but it always manages to sync in the end.

Now, all I have to do is figure out how to use webdav with Amazon's S3 to back up all my PDF files (Yes, I know I could use Jungle Disk, but I like my solutions more open source than that (generally)).

Anyway, it is working fine now and its nice to be able to share my bibliography.

Like many other people, I'm concerned about the recent changes in Facebook's terms of service. Especially, when I read how their terms of service compare

This is doubly true now that I've read the New Yorker article on the subject.

So, being unable to delete my Facebook account, I've:

• turned off as many "applications" as possible
• limited my visibility to others as much as possible
• deactivated my account

Facebook is dead to me, and hopefully, I'm dead to Facebook.