Posts filed under 'Research'

Transcription blues

Two weeks ago I had a window of free time, so I got ambitious and scheduled four interviews. Now I’m deep in the transcription blues. Going to Auckland for three days last week interrupted the flow, but I’m starting to regain some momentum. So far I’ve finished transcribing one interview, and tonight I’ll start on the second. If I’ve interviewed you lately and you haven’t heard back from me, don’t worry, I’m working on it.

Transcribing is a slow process (it takes me around three hours for every hour of interview), but I’m glad I’m doing it myself, rather than involving someone else. Listening to the interviews for a second (and third and sometimes even fourth!) time means that I can reflect on what I’m hearing, and think about what’s going on at another level. I hope that I’ll get the remaining ones transcribed by the end of the week. I have a couple of email interviews to complete, and then I’ll be finished with this stage of my data collection, and will be able to write about my findings.

In the meantime, I’ll carry on musing on whether Ranganathan’s five laws can be adapted for open source software projects. Next posting, the first law, software is for use.

February 19th, 2007

Taking stock

Today I managed to find enough clear time to stop and reflect on where I’ve gotten to with this stage of my research, and what I’ll do next. I’ve scheduled a meeting with my supervisors next Wednesday, and need to write a short progress report before the meeting.

So far I’ve completed 6 face-to-face interviews, involving a total of 8 people (one interview was a group discussion), and 7 email interviews. I’ve transcribed all of the face-to-face interviews and sent the transcripts back for review. One more email interview is partly completed, and two other people have agreed to be interviewed by email, but haven’t yet sent me any responses to the initial questions.

The respondents represent three different open source projects (one large, one medium, and one small), and quite a few different roles (such as core developer, local developer, project manager, project sponsor, and end user). I’ve learned that people see themselves in a wider range of project-related roles than most of the open source research literature suggests. Marketing and promotion is one aspect that many people have mentioned, which suggests to me that a satisfied open source project participant becomes an advocate (or evangelist) for their project.

The approach I’m using for this stage involves purposive interviews, where I select participants from a range of projects and backgrounds, and keep going until I’m not learning anything new. I don’t think I’m quite there yet. Most of the people I’ve been dealing with have been involved with their projects for several years, and I’d like to have a few more relatlve ‘newbies’ involved.

My plan for the next month or so is to carry out more interviews, complete the data analysis, and then start revising the model I developed for my research proposal. Then I’ll be developing a web-based survey to test the model. This is likely to take a month or so to get set up, given what I’ve recently learned about the survey tool I’ll be using. So it looks as if the survey will be ready in May or June.

One of the other things I’ve learned about research is that timing is important, and mine’s not very good. Sending out the first round of invitations at (American) Thanksgiving wasn’t a good idea, and then of course the second lot happened just before Christmas. I’m hoping that things will be a bit smoother now that the holiday season is over, and people’s routines are returning to normal.

January 26th, 2007

Growing acceptance of open source software

Will 2007 be the year open source finally becomes mainstream in the U.S. library community?

Two of the members of the LITA Top Tech Trends panel have included open source software on their lists of 2007 top trends; see the LITA blog entries by Thomas Dowling and Karen G. Schneider on January 13th. I wonder if any of the other panelists will mention it?

In addition, Lorcan Dempsey has blogged about Equinox Systems, a company set up by some of the people behind the Evergreen ILS to provide support and software customisation. Lorcan also highlighted some of Koha’s new features as implemented in the Nelsonville Public Library in another blog posting. LibLime is the company supporting Nelsonville, and I understand that they’re primarily responsible for these features. There are now three U.S.-based companies supporting open source software for the library community: IndexData, LibLime, and now Equinox. IndexData seems to have carved out its own niche, but it appears that both LibLime and Equinox offer support for Evergreen. I see an important benefit of open source software in that it has the potential to free organisations from reliance on a single vendor. Open source software users could even choose to have support from more than one source, for example, choosing a company with expertise in searching to support the discovery module (AKA the OPAC), while they might use their own staff to support circulation. Time will tell to what extent this is a realistic option; it will also of course depend on the types of contracts these companies prefer to use.

I’m not sure if this has any implications for my research project, but I hope that it will increase people’s interest in the results. For people who haven’t started at the beginning of this blog, I’m looking at factors that influence open source software participant satisfaction with the software, and in particular, how the factors vary with people’s involvement with the project.

January 15th, 2007

Back to business (or should that be research?)

It’s hard to believe that it’s already January 8th. The holidays went by very quickly, even though Wellington has had its coldest December since records began.

I didn’t just laze around, though. I did manage to fit in one more interview and finished transcribing two others. I have at least one more interview to do later this week, and there are still a few email interviews in various stages of completion. I am starting to wonder if some of my email messages have fallen into a trench somewhere in the Southern Cross cable (or more likely fallen into a spam trap somewhere), and I’ll be sending out follow up messages later this week.

To pick up where I left off interviewing myself, the next question is about my involvement with open source projects. I’m not directly involved with any library OSS projects, except as an observer/lurker. I’ve written the odd article and given a few presentations featuring case studies of various projects, so the best term to describe me is probably a ’supporter’. I do use some open source software (like this blog) and have set up the LIANZA IT-SIG wiki, which uses the PmWiki engine. I’m currently creating some content for the ‘Five Weeks to a Social Library’ course being organised by a group of North American library bloggers. My role has been to organise an interview between my husband and Patrick Michaud, that will be published as a podcast. They did the interview in late October, and now I need to to a bit of editing before submitting it to the course site. I’m planning on using Audacity (another open source project) for that.

January 8th, 2007

We interrupt this blog …

… for a short holiday. Regular posting will resume in early January.

I’d like to thank everyone who’s participated in my research project so far for making such thoughtful comments and giving me such a lot to consider. I hope you all have an enjoyable holiday season, and a happy and prosperous New Year.

We’ll be enjoying home-made English muffins for breakfast on Christmas morning. I’m including the recipe here if anyone else is interested in trying them. It assumes you have some experience making bread or other yeast-leavened baking. You can leave out the dried fruit if you prefer them plain.

Ingredients

1/2 cup warm water
1/4 cup milk
1/4 teaspoon sugar
1 1/2 teaspoons dry yeast
2–2 1/2 cups flour (the quantity you need will vary depending on the humidity and other conditions)
1/3 cup dried fruit (cranberries for a seasonal flavour, but raisins are also nice)
rice flour or fine cornmeal

Method

Place 1/4 cup warm water in medium size glass mixing bowl, and dissolve the sugar. Sprinkle the yeast on top and leave for 10–15 minutes.

Soften the cranberries by covering them with boiling water in a measuring cup or small bowl.

Heat the milk and remaining water to body temperature, and add to the yeast mixture.

Add 2 cups of flour and knead with your hands until all ingredients are mixed well together. Add more flour if necessary, but keep the dough sticky, not smooth. Drain the cranberries and mix in.

Roll the dough into a ball and place in a bowl covered with a damp, clean tea towel until doubled (1–1 1/2 hours, depending on the temperature). This is a good time to open presents.

When the dough has risen, deflate it by gently pulling it away from the sides of the bowl. Divide into 4 equal-sized pieces, and roll each piece in fine cornmeal or rice flour. Let them rise slightly on a flat tray for 10-15 minutes, while heating a frying pan to a medium-hot temperature.

When the frying pan is heated, cook each side for 8-10 minutes, or until lightly browned. Let rest for a few minutes before serving with butter and jam or honey. Fresh fruit is also nice. Makes 4 muffins. Enough for two hungry people who don’t plan on having much lunch.

Christmas dinner will be goat cheese profiteroles, salad, and methode champenoise, plus fresh berries for dessert. The profiterole recipe may follow later.

December 22nd, 2006

Keeping on keeping on

Even though I said I’d aim for weekly updates, there’s not much to say about my research at the moment.

I’m still scheduling synchronous interviews, and waiting for a few email interviewees to get back to me. I don’t want to say too much about what I’m learning in case I affect the interviews to come. So what I thought I’d do now is interview myself. It seems only fair that I let participants find out more about me, since I’m asking them about their background and skills.

Could you tell me a bit about yourself and your background? For example, what are your educational qualifications? What is your current position, and what types of positions have you held in the past? How would you describe your technical skills? Other skills?

I have a B.Sc. in chemistry and mathematics, and an M.L.S., both from the University of Alberta. I got interested in library work in the mid-1970s when I saw Dialog for the first time, and thought how exciting it was to be able to search multiple years of Chemical Abstracts in a single step. One of my early summer jobs involved going through every volume of Chem Abs looking for everything that had been written about methylmercury. It seemed to take months, and was very slow — so the idea that you could use a terminal connected to a remote computer to speed this up was something I appreciated. I applied to library school and was accepted. I felt very out of place at first, because I was the only one in the class without actual library experience, but I survived.

I’ve already written about my first library job (see: Choosing a research topic for more about it). Learning to be a SPIRES programmer for the library led me to my next job, working as a Programmer/Analyst III in the SPIRES group. I did that for a couple of years, and then moved on to become a project analyst with UTLAS in Toronto. I remember writing a specification for an online catalogue that people thought was very exciting. By today’s standards it was boring and unimaginative.

In 1986 I moved to New Zealand to work as a systems analyst at the National Library, and moved through the ranks to become the Applications Manager. Access to source code was important there, too, because the NZBN bibliographic utility and union catalogue was based on a modified version of the WLN software. In 1990 I started teaching part-time in the New Zealand Library Studies Certificate programme, and I took up my present position of Senior Lecturer at Victoria University of Wellington at the beginning of 1997. In addition to teaching in the MLIS programme, I’m a Ph.D. candidate in the School of Information Management.

The best word to describe my technical skills is ‘rusty’. I tend to be a power user of whatever software I use regularly (although WordPress is currently an exception). I’m still able to troubleshoot and can generally find solutions or workarounds. I know basic HTML and CSS, and can understand simple PHP scripts. I try to keep up with new developments such as Ajax, OpenURL, etc., but am generally more of an observer than a developer.

In many ways I’m either an early adopter, or have been lucky enough to have been exposed to technologies before they are in general use. I used the device-independent TEXTFORM software to produce my M.L.S. thesis in 1978. TEXTFORM used document markup in combination with layouts to format text for different devices. This gave me an early introduction to the power of markup languages. In many ways I still prefer markup over WYSIWYG for word processing.

Coming next: Projects I’m involved with (at least in a peripheral way)

December 18th, 2006

Research design

My research project uses a sequential mixed-methods design. This is research jargon meaning that the project has two stages that collect different types of data: the first involves gathering qualitative data using audio or email interviews, and the second will involve a web-based questionnaire to collect quantitative data that will let me test (and revise) a model of factors that affect people’s satisfaction. This is a standard approach to conducting research, and the only part that’s slightly unusual is the use of email interviews, and even that is becoming more common.

I’m currently in the middle of the first stage. I’m using a purposive sample, which means that I’ve approached a small number of people involved in a range of activities from different projects. The idea is that I’ll get different perspectives and will be able to identify significant similarities and/or differences in the responses, and start thinking about the reasons for these. I’ll then use the results to fine-tune my draft questionnaire.

In the last few weeks, I’ve done a couple of interviews in person, and have sent out a number of invitations to be involved in the email interviews. So far my response rate is slightly over 50%, which is similar to my previous experience of inviting ‘virtual strangers’ to collaborate on a paper about wikis in 2003. Interestingly, it’s almost exactly the same for both types of interview. I’m using my iPod with a microphone to record the in person interviews. The sound quality is clear, and the device itself is relatively unobtrusive. I’m transcribing the audio interviews myself, and am gradually learning a few tricks to make the transcription go more smoothly. I assume this is something I’ll get better at over time; at the moment it’s taking me around 3 times as long to do the transcription as it did to do the interview.

I hope to complete a few more interviews before the Christmas break, and will begin the formal analysis in the New Year.

December 12th, 2006

Choosing a research topic

I’ve been working in the area of libraries and technology for a while—since the late 1970s, in fact . When I was choosing a topic for my PhD, I knew I wanted to do ’something about open source software and libraries’, but turning that vague idea into a researchable topic was a challenge.

I read everything I could find about open source software and projects, and kept an eye on the fledgling research that was being done in the area, but very little of it captured my imagination. I knew that I needed to find a topic that was more than just a description of projects and how they worked. The trick that worked for me was asking myself ‘what is the most compelling aspect of open source software to me?’, while at the same time my husband (who is a contributor to one of the more active wiki projects) said to me ‘come and see what I’ve just done!’. Two things clicked then: people’s abilitity to contribute to projects in whatever way suits them, and the satisfaction they feel when using the software.

My formal research question is ‘what factors influence participant satisfaction with open source application software?’, with two sub-questions:

  • what types of contributions do people make to open source application software projects? and
  • Do the factors that influence satisfaction with an open source application software project differ for different roles? If they do, in what ways?

I’ve also realised that access to source code has been important in several of my previous jobs. In my first professional position after library school, I worked with a mainframe database management system called SPIRES (Stanford Public Information REtrieval System). Bo Parker’s “An Overview of SPIRES and the SPIRES Consortium.” The Public-Access Computer Systems Review 1, No. 3 (1990) has a good overview of SPIRES and its key features.

The library had commissioned a SPIRES-based application to print catalogue cards—even though this now seems an outdated concept, it eliminated a significant amount of time I spent proofreading typed cards. I only needed to proofread the electronic catalogue entries once, rather than checking every tracing on every added entry card. One of my tasks was to liaise with the developer, identifying bugs and enhancements. I don’t remember why he gave me access to the source code, but I can still remember the feeling of empowerment I got the day I phoned him and said “I think you need to change xxx to fix the problem I told you about yesterday”. He said “why don’t you change it yourself?” and I haven’t looked back since. Thanks, Ron.

That’s one of the things that led me to this topic. I’ll say more about some of the others in a future entry.

1 comment December 6th, 2006

Hello world!

I’ve finally managed to install WordPress and start blogging, after talking about it for months. It’s been lots of fun choosing a theme and tweaking it. For more information about what I’ve changed, see the ‘Colophon’ page.

‘Hello world!’ seems like an appropriate title for my first post. I’ve set up this blog for several reasons. First, it will give me a place to document parts of my life, and reflect on what’s going on (hence the blog name). I am also interested in having a place to keep track of key events in my PhD project, and to let participants in the project keep up with my progress.

November 17th, 2006

Next Posts


Calendar

September 2020
M T W T F S S
« Oct    
 123456
78910111213
14151617181920
21222324252627
282930  

Posts by Month

Posts by Category