PhD status

Mount Everest as seen from the Tibetan plateau
Mount Everest by Joe Hastings

A quick note that appearances do suggest that I am in the final weeks of my PhD, with submission in late May. I am reluctant to say this because I’ve been wrong before, but this time my supervisor agrees. So.

I probably will be pretty absent for several weeks. And if I am not, I may be very tired.

Also an explanation of how this works in Australia, because it’s quite different to North America. Mostly writing this so that people don’t start addressing me as ‘Doctor’ in June.

Short version: this work is me preparing my thesis for initial examination, and this is hopefully the hardest bit. But I won’t graduate for at least six months.

First, I finalise my thesis document (we don’t call it a dissertation). I submit this to the university where it is examined by three external examiners: ideally at least one from an Australian university. At this point my work on it is in deep freeze.

Unlike in North America in general, these examiners are anonymous to me (chosen by my supervisor) and were not involved in my PhD studies prior to this point. This means in theory that they might not like it: in practice I am told that around 99% of students who submit at all eventually graduate.

Examination in theory takes six weeks, it could take as long as six months (since appointing a whole new examiner might be slower than waiting for a late one). They submit reports which my supervisor reads and makes recommendations on (most commonly I agree, Mary should indeed fix all these things, followed by I almost entirely agree, Mary should indeed fix all but a few of these things). This is fed into the higher degree research committee (who usually agree with the supervisor, but they might come up with a different answer if the examiners’ recommendations varied a lot) and then there’s a huge range of possible decisions that come out of the HDR committee:

  1. pass as is
  2. make minor amendations to the Library copy and pass
  3. minor revisions to be checked by supervisor (for which I’d be allocated a month) and then pass
  4. major revisions to be checked by supervisor (for which I’d be allocated two months) and then pass
  5. revise and resubmit to examiners a second time
  6. only award a Masters degree (possibly in combination with revisions): recall that in the Australian system I don’t already have a Masters degree
  7. fail

The most likely decision by far in my research group is minor or major revisions: I’ve never heard of anyone avoiding them. Some people in fact prefer major just because you get a bit more time to revise. (In other faculties, it isn’t unheard of to pass without revision.) No one wants to be re-examined: this usually means re-enrolling and re-doing experimental work and similar.

Unless re-examination is needed, after any required revisions it is pure administrivia: the HDR committee must pass it, and then the university Senate. I submit a bound copy of my thesis to the university library and (far more importantly now) put it on my website and submit to the university’s digital collection.

I think at that point I am finally a graduand and can use the title ‘Dr’ in academia and so on. Actual graduation would take place in either September/October or April/May, so in the pathological case it could be a while between finalising the thesis and actually graduating.

I do not do an oral defence (a three hour or so session where my examiners ask me questions in person).

linux.conf.au 2011 call for papers

We’re seeking a wide range of papers across the whole spectrum, encompassing programming and software to desktop and userland, education, community and law…

Some typical topics (but not limited to these) include:

  • Aspects of kernel development, including recent data structures and algorithm developments
  • Database and File system developments
  • Desktop topics, covering aspects of the user experience
  • Networking topics, from device drivers to servers
  • Novice user’s introduction to exploring FOSS
  • Professional development, including Software Engineering & System Administration techniques
  • Scalability, both embedded and enterprise
  • Development topics, including concurrency and toolchain advancements
  • Open Source Software usage, including business, education & research
  • Graphics & sound advancements, from low level drivers to end-user applications
  • Open Source culture, including open content creation

More information is here, and submissions close August 7.

Unlike for the previous three conferences (Melbourne 08, Hobart 09, Wellington 10), I’m not heavily involved in selections: I will be reviewing abstracts but not (co-)chairing the process. I probably won’t even be at the infamous day-long meeting to finalise selections, a long awaited visit from Andrew’s sister will clash.

I’m hoping to attend the conference — it’s fairly safe to say by now that unlike this year, it will not clash with giving birth — but I’m not sure yet. It probably will clash with producing a PhD thesis, and ACL (computational linguistics’ main conference) usually has a deadline that nicely clashes with LCA too.

The end of 2010

Although there’s always a possibility, it seems that regular childcare will not be available for my son Vincent until early 2011 (yes, I’ve had his name down for ages, although not since birth, since I was pretty sick afterwards).

What this means: my supervisor and I have agreed that I can’t resume my PhD studies if I am doing fulltime baby care. My experience so far meshes with this: I can poke at my code and read the literature, I can’t really pound on it in forty-five minute chunks a few times a day, and it would be a waste of both of our time if I was to be formally re-enrolled now.

So it looks like I have six months or so to kill, and I’m thinking about what to do with that time. I may or may not have some paid work lined up for some hours of the week. I’m hoping not to spend the rest aimlessly noodling around on the ‘net, well, not all of the rest. Options include more paid work (if you for some reason have paid work that can be done by a — skilled! — programmer/writer/researcher in forty-five minute chunks, get in touch), hacking, writing. Looking forward to coming up with some projects!

Plans for 2010

I don’t intend to write much about my just announced pregnancy over on the thoughts/geeky side of my weblog, but there are a couple of geek implications, assuming all goes well:

  • it’s unlikely in the extreme that I will make it to OSDC 2009 in Brisbane in November (it’s very close to the deadline when airlines will stop letting me fly, in addition to discomfort and so on); and
  • it’s completely impossible that either Andrew or I will make it to linux.conf.au 2010 in Wellington in January.

I’ll not be especially available for additional volunteer tasks in 2009, since I will be trying to finish my PhD work with a small human trapped in my abdomen.

For Andrew’s other 2009 and 2010 availability check with him, I can’t see that I’ll be volunteering to travel before at least April 2010 and that might depend on someone donating a nanny to accompany me.

I am cut about lca2010, especially considering the effort I’m sinking into it. I really hoped that wouldn’t happen, but it’s turned out to have the worst possible timing. If a generous donor offers to fly, say, 20 of the speakers to Sydney afterwards and re-stage the conference for the sole benefit of me, I will not say no.