Friday 18 November 2011

November writing challenge update

Back at the end of October I put my hand up to participate in a science writing challenge for the month of November by finishing up two papers that really should have been written long since. On 1 November I started my new job, and didn’t write at all that first week as I focused on the 1001 tasks necessary to starting a new job, including moving into my office, rearranging the office furniture to suit my working needs, meeting my new colleagues, obtaining employee cards, library cards, etc.

Early in week two my office computer arrived, and another two potential writing days were lost installing programs and generally making the computer ready to use. However, on Thursday and Friday of that week I actually sat down and finished up the current draft of the paper based on my PhD research. The previous version I had done (while back in Australia in July) was good in terms of presenting the basic facts of what was done and what the results were, but I had been kind of sketchy in the discussion and conclusions section. (I think that is a common failing on my part that I should work on—I am totally comfortable reporting facts, and I am totally comfortable with editing a previously existing discussion/conclusion section, but actually confessing what *I* think is relevant or important about those facts? That is getting kind of personal.)

Be that as it may, since it was necessary to finish the paper I somehow found the necessary motivation/inspiration to just write it—I went back to my PhD thesis, looked at the points covered in that section there, and chose which ones to address here. Not only did I write it all down, in many cases the version in this paper is much clearer and more eloquent than what I had typed when finishing up the degree. Spending a couple of years thinking about other areas of geology actually helped give me some new insights on that project.

In this, my third week on my job I have continued to split my time between needful tasks for this project (doing background meeting, arranging a trip to the mining company with whom I will be working, obtaining a card to let me use the uni gym (free to employees during business hours), etc.) and finishing up previous papers. Since I had sent a copy of the PhD paper to my erstwhile advisor in Tasmania on Friday that meant I could focus on working on the paper from the experimental post doc position I finished last December.

This week I have managed to do some editing of the text and make major progress on a set of figures that should have been done long since: BSE images of every experiment, annotated to label the mineral phases present.

Why hadn’t I created such images previously? Because I documented each microprobe session in CorelDraw. My standard operating procedure was to look at the sample, determine a region to work on, take a picture, transfer it to my personal computer, open it in CorelDraw, and create a new layer for the day’s session in which I would make colour-coded circles superimposed over the picture at the locations for each analysis point. I would give each circle a name that matched the name recorded in the microprobe (such as RC1-NMg1 for the first garnet analysis on sample #RC1-NM). Repeat for each analysis, taking additional photos as needed.

This works very well for recording things, and one can easily go back and compare the results with the appearance of the phase analysed. However, the layers can get confusing for those samples with multiple microprobe sessions (due to the difficulties in getting good results for some phases).

The new, improved, pdf images I am creating for each sample make things much easier—they are all labelled with the pressure and temperature of the experiment, which bulk composition was used, and which phases are present. The phases are colour coded to indicate the quality of the data—if they are in bold print I had five or more good analyses of that phase for that sample which agree. If they are in normal print there were three to five good analyses, if they are in red there were two (or fewer) good analyses, and if the red text has a question mark next to it I know that the phase is present, but it was too small to get a clean analysis so I do not actually know the composition of the phase. Having this information right there with the photo of the sample is very helpful. It is also pretty easy to see why the red text is in red in most cases—those phases really are smaller or too amorphous to get good readings—one can see that in the photo, too.

There is still a week and a half left in November—I think I may be able to finish compiling these images today, so there is a chance that I will be able to complete all of the other tasks necessary for this project before the month ends. However, even if I do not complete this goal, I still thank Anne for having inspired me to set it—I am certain that I would have found plenty of tasks to keep me busy instead of writing these older papers if I had not stated publically that I would do it.

No comments: