Dealing with Images

One of the harder parts of a dissertation is getting in your images correctly. My images were typically half a page, which created huge problems in trying to get the text to flow around the images. I finally gave up and just did hard page breaks around my images and longer tables. It was the only way to get it to work to my satisfaction. As an added benefit, it also increases that all-important page count.

Putting it together with a master document

I had hoped to put my dissertation all together as a master document. However, it just didn't work out. The picture were all messed up. I ended up cutting and pasting the chapters together, and then making changes manually. I am not sure there is a way to reliably put together a dissertation using Master documents. While it is theoretically possible, it just doesn't work in the real world. Save your work in chapters, but put it together by cut-and-paste is my suggestion. Perhaps my other suggestion would be to just do your dissertation in LaTex.

On a positive note, I have turned in my dissertation, and did so before the deadline. I have some formatting corrections to do, but they are not too big. Also, I finished my corrections for the third paper. They sent me the author proofs and I sent back a few changes. It should be published soon.

Secret for easy formula entry

I changed the settings on a couple of OpenOffice installations so that control-F brings up the equations editor. You can change it under Tools->Customize->Keyboard. I highly recommend it.

I find it helpful to enter in a number of formulas at once. I can then mix and match them, putting them into the chapters where they will be useful. Entering formulas is something I can do when not mentally sharp, making it useful for mental down-time when I still need to put time in on my diss to meet my daily goals. Then, once you get in the swing of using _, ^, and % followed by a greek letter, it is much faster. It is almost like LaTex.

There are some problems with the Open Office Equation editor. Subscript z's have their central section disappear. I can fix this by using quotes around the z, though that does change the font. I have also used this trick to get the "plusminus" sign to work properly - inserting the plus-minus, then putting quotes around it. In conjunction, it is now easier for me to write out the equation in the regular word processor, highlight it, hit control-f, and it automatically converts it into a formula.

Master Document works! (For the most part)

Today I put everything together in a master document, and added a table of contents. It went fairly well. Here are a few pointers:

1. Make sure you check the option in the table of contents to adjust it manually. It saved a bunch of effort to make it look decent, and I needed a pagebreak to get it on a different page from the title page.
2. Pay attention to "following page" styles. I had to make sure the page style following the title page ("first page" style) was my dissertation page style, and not the "default" style.
3. I have run into a strange problem where I have a number of blank pages before figures. Kind of strange.
4. Making a master document, and then PDF'ing it is really awesome!

For my references I imported my BibTex database from JabRef into Zotero. I then exported it to RTF using the "American Psychologial Association" citation style, which has a nice, full citation format. Then, I developed a "bibliography entry" style, applied it to the entire collection, and it came out pretty well.

Solution to Numbering Problems

I was laying out my chapter and section headings. I thought I had it all down, defining a numbering system named "0 Diss Numbering." The "0" on the front was just so that it would come up first. I then re-defined Headings 1-3 to use this numbering, as well as fonts, small caps, and centering as needed. But, the numbering was not working right. I couldn't figure out how to do it. Finally, failing at my own abilities, I turned to the power of Google, and came up with this posting.

Handling numbering the right way.

Checking over, I needed to go the Tools->Outline Numbering tab, and define each level 1-3 as corresponding to Heading 1-3. It is now working well.

Wild Month

It has been a pretty wild month. I went on vacation with my family, had a paper rejected, interviewed for more jobs, redid the paper for a different journal, and accepted a job that is literally on the other side of the country! It has been a wild ride.

In the midst of all this, I was frustrated with the software setup on my laptop. I decided it was time to upgrade from Ubuntu 7.04 to the next one, 7.10. I thought this would be safe, as the current version of Ubuntu is 8.04, and I didn't want things messing up with a huge upgrade.

I was dead wrong. For years I have never had big problems with Linux. "It just works!" I may struggle with a sound card, have to use a special modem, not be able to do something small, or be addicted to "online update," but I was able to get my work done. Now, I have problems. I can't seem to kill user processes - even as root. That isn't usually a problem, as I hardly ever used kill, but now I have programs that take over and refuse to die. I also am frustrated by Ubuntu's lack of a pre-installed super-user file browser, especially when compared to my previous positive experience with KDE. Having installed Caldera OpenLinux, Corel, Suse, and Mandrake on my computers, Ubuntu has been a step back. I don't think the reality reflects the hype. And I will do clean installs from now on, as they are much, much less work.

I need to get my dissertation finished in some form two weeks before my oral defense, which is likely days before I move my family across the country. It is crunch time. But, I am also really thankful for Craigslist, as it makes it possible to do things much more quickly.

New Week, Fresh Goals

Last week I didn't do much on the diss, but part of the reason was that I was out of town for an interview on Wednesday and Thursday, then on Friday I had a local interview and then toured the place on Saturday. Another big reason was that I read the novel "Dune" that my wife had given me. ;)

While reading Dune may not have been the best use of time, I enjoyed the sense of destiny, purpose, and fate in the book. As I have been pondering my own future, it was both a reminder and a caution.

This week I commit to getting stuff done! Old week is dead, and I need to do the best I can with this one.

Week is winding down

The week is winding down, leaving me little to show for it. I have read through more of the manual today, as well as last night. I hope to finish it today. I did a solid 1x15 on reading, and I hope to link together 3 more during the next hour. I still haven't figured out how to do the bibliography effectively, but I feel much more comfortable with styles now.

Part of what has been taking my time is the job search, as I have had two telephone interviews this week. PhD-level jobs take a long time to find. I do not want to take a post-doc, even though that is the preferred move for an academic career. I have a family to take care of, and I do not want to move them to another post-doc after a year or two, and have a lower salary to show for it. I will likely end up in an industrial job anyway - why take a lower salary for less stability? Unfortunately, it seems like it is really hard to get on at a national lab or work for NASA unless you are working as a post-doc. Still, the first job is the first job -- not your entire career.

I have also used OpenOffice for my resume and CV. In my CV I inserted hyperlinks to my papers, homepage, and email. That way someone looking through it can click on those for further inspection. However, I changed the color back to black, and removed the underlining, so it prints clean. In this way, there is a little surprise for the reviewer. I also have the files converted to PDF. However, I had to play with it a bit so that the PDF conversion did not change the color back to blue. Also, I have not been able to get my headings to come up as bookmarks in the PDF. So, I still have some work to do.

Progress this last week

It has been over a week since I last blogged, and I have very little to show for it. I have been reading the OpenOffice manual, and setting up my installation of OpenOffice to look better. I have also learned a few other tricks.

I have also been preparing my figures. I have a number of figures, including some I didn't use in my papers. Thus, I am using these to increase page count, and hopefully make the writing a little easier. They serve as a de-facto outline.

One of my big decisions is the type of referencing to use -- do I do it by last name or by number? I would like to do it by number, but I would also like to write each chapter in a separate file. I know it is possible to put it together in OpenOffice, but I need to figure out how to do it so it actually works.

However, as I look back, I have really accomplished very little the past few days. I have good reasons for part of it (interviews, job search) but bad reasons for most of it (wasting time on the internet). I am trying to get re-focused today. 15 minutes on, 5 minutes off. Hopefully, I will be able to say at the end of today that I did not waste time online, and I was very productive, even though I have a telephone interview later on today.

Software I am using

Right now I am using OxygenOffice 2.4 to do the writing. It is OpenOffice 2.4 with extras. I just installed it on my laptop (Dell Ubuntu pre-installed) as well on a desktop Windows98 machine at home. It is working well so far, but it is not smoothed and integrated like the native Ubuntu version of OpenOffice. So, I will have to research how to integrate it.

I do most of my diagrams using Inkscape. I have also used Inkscape for diagrams on my last two papers.

My choice for bibliography software is JabRef. It is cross-platform and able to download reference information. It is also able to export in various formats -- including a useful HTML format that allows searching.

I am experimenting with OpenProj for project management. It is not entirely well optimized, as it seems to be for projects involving many people, and not a single-person project like a dissertation. I'm also trying out other open source project management applications, but I am not sure I will use any project applications at all.

Open Office User Guides
Has user guides for all parts of the suite. But, two chapters from the Writer Guide are particularly useful for dissertations, even if they are a little old:

• Ch12 - Creating Tables of Contents, Indexes, and Bibliographies 31-Dec-2006
• Ch13 - Working with Master Documents 31-Dec-2006

Tutorials for Open Office
This is a collection of a wide variety of formulas. Here are some that you may want to check out.

• Construction a Bibliography
• Master Documents
• Introduction To Templates
• Introduction To Indexes
• Create and Modify A Table of Contents
• Different Page Styles In a Document

Should you run into problems, here is the Advanced Writer Forum, in which people talk about such tasks as making books.

Links from Phinished

Probably the best resource out there for writing your dissertation is Phinished. Here are some discussions involving OpenOffice:

Anyone else using OpenOffice for their Dissertation?

Reference Software on Linux - Using OpenOffice Spreadsheet for Bibliography

Writers Tools for OpenOffice

Using Bibliography software with OpenOffice

Welcome!

This is my blog to tell of my trials and travails of using OpenOffice to write my dissertation. I hope you will find it helpful if you are writing a large document with OpenOffice, or if you are writing a dissertation with any other software. Like in the Red-Green show: "We're all in this together. Keep your stick on the ice."