AUC Conference 2005 – So it began

The conference proper kicked off with the welcoming reception at 5:30 or so on the Sunday night. Everyone gathered in the foyer area of the conference section, and Stephen Young said a few welcoming words. There was a whole host of assorted drinks going around, along with (from memory) some nibbles. Ashley, Brett & I gravitated together as familiar faces to one-another, and maintained an orange juice black-out in our immediate vicinity – since none of us were drinking, and on each tray of half a dozen drinks there was only one or two orange juice’s, one had to be quick. :)

There was also food going around – chicken on skewers, rolled up meaty things, all the usual party foods. All very good, of course. The waiters and waitresses quickly learnt not to bring the fresh food trays by us first, as we consistently laid waste to them in no time flat. :)

But beyond the food – did I mention the food? ;) – the first night was nicely casual. Everyone chatted, got to know each other a bit, all the usual introductory socialising. From memory this was when I was introduced to Daniel Woo, who as it turns out works with Ashley (at uni). There was James Lucas, as well, whom I spent quite a lot of time with over the course of the conference. And I also met many other people, who will have the awkward pleasure of re-introducing themselves to me next time I see them, seeing as I’ve largely forgotten all their names. :)

Once that wrapped up, which was fairly early, I retired to my room, probably wrote some emails, did a little coding, before hitting the sack. Knowing I’d be looking at a 6am waking the next day, I was careful to be asleep by a mere 9:30 or so – which felt like it was barely even midday to my twisted student body clock. Oh well. :)

The next morning I fought with the crazy shower for a long while – it was one of those “water saving” ones which never give you enough flow to actually wash yourself properly, and had the crazy bath-shower hybrid system I grew oh-so fond of in San Francisco… shudder. Anyway, I was up and down to breakfast in good time, amongst the first to arrive. I found it a pleasant, albeit unnecessary touch that Natalie Evans – one of the handlers from ID Events & Meetings, with whom the AUC entrusted the conference’s operations – was at the door in these wee hours to greet people (and, let’s be honest, check they had their conference badges, but hey, let’s be nice :) ). Breakfast was a little skimpy for my tastes – the kind of “continental” breakfast you get in U.S. hotels which mostly involves toast and muffins. Well, muffins are all good and well, but I’m a cereal man myself. There was some cereal there, but with tiny little bowls, with which I was too embarrassed to go for seconds and thirds lest I scare away other attendees. ;)

I’ll digress blatantly at this point, to point out – yes, ironically – that I keep digressing onto the food. It’s a genetic problem inherited from my grandfather, I’m afraid. Although I’ve been lucky so far in seemingly not inheriting his desire to visit every bathroom in Australia. Anyway, onwards…

Breakfast was a good chance to socialise again, and meet a few new faces. After that, it was up to the main hall for the opening speeches. Stephen Young kicked off again with the usual housekeeping stuff, before handing off to Tony King, the managing director of Apple Computer Australia. He talked about where Apple was, where it was going, and all that sort of thing. He was a good speaker, and kept interest up, although for myself – someone who reads all the Apple news sites religiously – there wasn’t really anything too new. Nonetheless, I good “state of the union” style opening for the conference.

After that, Stuart Lynn – past President of ICANN among many other impressive accolades – spoke on the topic “Build it and they will come…”. His main point, I think, was impressing the importance of natural evolution over intelligent design, if I can needlessly incite furious debate with that pointed analogy. :) He was excited about where the future was taking us – particularly relating to networking technologies – and left a good impression. I regret I cannot now recall more details of his talk – the conference proceedings contain only the papers presented by other speakers, not an overview of all the sessions, so I’ve got nothing to jog my memory with… I’m sure snippets will come back to me as I write on. But one key thing I do remember him emphasising was the importance of providing the infrastructure without immediate justification – i.e. just building it, and seeing who comes. It’s an important message – let’s face it, 90% of the stuff we propose today is unnecessary for our immediate uses, but it’s because of the recklessness and the over-provision of facilities that we get bold new revolutions. For example, sure, the average person doesn’t need a 20 megabit internet connection for their email and web browsing, but it brings with it all new possibilities – VoIP, IPTV, advanced peer to peer technology like BitTorrent, etc. But I digress to my own opinions… Stuart’s talk was kind of a “feel good” thing about all this, and how in the end things tend to work out pretty cool for everyone.

After Stuart there was a break for morning tea – won’t talk about it, promise :) – and then we resumed sessions. I think – although I’m hazy – that it was Joe Cox who spoke for the Apple Plenary session, on MacOS X technologies. It was an interesting talk, but much like the “Apple Directions” talk by Tony King earlier, really just a rehash for someone like myself, already “in the know” on all things Apple. Still, it was a good intro for those new to the platform, or without a technical background on it.

At noon we kicked off the streams. For the first session, while I did want to see Sandy Shuck’s talk on “Teachers as Producers, Students as Directors”, I gravitated in the end towards James Steele’s talk on “Streaming TV via IP”. This was a fascinating talk, where he discussed the streaming video system at Australian National University (henceforth ANU) from it’s prehistoric days as a satellite TV relay via analogue coax, up to it’s current incarnation as a true video over IP solution. It was interesting from all points of view – seeing how they originally started out with their coaxial system, and hearing of the problems and costs associated with installing such niche infrastructure, in sharp contrast to their modern system, which utilised the existing networking infrastructure.

It was also amusing to see how they had tackled many of the interfacing issues with their various TV sources. They have numerous satellite dishes amongst their sources, as well as AM/FM radio and possibly even terrestrial broadcast TV as well. For the AM/FM radio they actually had, in their server rack, half a dozen car stereos, each tuned into a single station and outputting via [presumably] a standard 3.5mm stereo connection. I think they also had some VCR’s and other crazy equipment linked in somehow.

They used numerous PowerMacs for the real-time encoding, and ultimately piped their channels out using m3u playlists which could be viewed in VLC (VideoLanClient). I don’t think they were at the point of streaming directly to Quicktime or similar, which is a shame, but I seem to remember him mentioning that it was certainly something they hoped to do in future.

It was also funny how the project came in to real, financial being – originally the AUC provided funding for a single Xserve so that they could prototype a digital system… but then somehow they landed over a million dollars in funding from some other project that had been cancelled or run underbudget, or somesuch… so they – like all good kids in a candy store – went all out and built up the whole system in quite a short time. It’s always funny, things like that – of course it never rains, it pours. :)

After that session was lunch – which was good of course – followed by two more stream sessions. I should note that some time during morning tea I started to get quite nervous about myself having to talk later that afternoon, so I began ferreting through my Documents folder, rounding up anything that might be demoable at short notice. Similarly, I whipped up some horrible but succinct notes on what I wanted to say. Lunch was spent similarly.

The next session I went to was by Paule Bourke, from Swinburne University, on the topic of “Using Mac OS X to drive immersive displays for science visualisation and education”. This was a really cool session from a “whizbang” point of view, as he showed off a lot of their technology and research into immersive displays – that is, 3D displays and very wide-angle displays (e.g. planetarium-style).

It was impressive to see how they were dealing with the problems within stereoscopic displays and immersive displays. The immersive displays, for example, that they used utilised a quarter-sphere mirror to reflect the image onto a large portion of the room (or dome, or whatever was available). They could then use precisely calibrated texture maps in OpenGL to warp their outgoing images so that they would appear perfect on the final surface. Very interested stuff – we’ve all probably seen or heard of people doing simple fish-eye distortions for dome projections and such, but these guys were working on arbitrary mesh distortions, so they could use any shape of surface they like. Very cool.

And the stereoscopic stuff was good, although they were a bit down on the fact that MacOS X doesn’t really make life easy in that area… none of the graphics cards currently available support gen-locking, which is where two graphics cards are synchronised to output frames at exactly the same times, necessary with stereoscopic and multi-panel displays to prevent tearing. There were also some issues, I believe, with actual stereoscopic support… OpenGL has supported this pretty much forever, but they were having some issues making good use of it…

Anyway, overall it was pretty cool. I love seeing scientific visualisations in these ways – it really is so vital to facilitating understanding, and can really increase interest in the sciences.

The next session I went to was titled “Wicked problems and shared meanings: Evaluating design competence”. The speakers were Grant Baxter and Nick Laird, from the university Otago in New Zealand (although we didn’t hold that against them ;) ). The name of the session doesn’t mean a whole lot, really, I know. But what it was about was an online community for design students, which encompassed online coursework as well as submission and review. It really stood out to me in two ways:

a) It didn’t suck horribly. In fact, it was pretty fantastic.
b) They really “got it”, with regards to what students and staff really want out of such a system.

The first point is pretty easy to define – if you’ve ever used WebCT, you’ll know about the horrible sucking. Their system was the opposite – it was smooth, elegant, aesthetically appealing and simple. Students could obtain their assignments online, then submit them, and lecturers could keep track of everything with great ease. The marking system was great – marking was broken down into weighted categories, which could be assigned marks using a slider. In addition, each category had a whole lot of checkboxes which could be filled out to provide more detailed feedback, e.g. “Understood the application of soft lighting in this scene” or somesuch. And there was of course a provision for free-form textual feedback as well.

And there was a cool feature were you could see the distribution of overall marks in various histograms, and could adjust the current student’s mark to see how that changed the distribution. This is essential for lecturers to ensure their markings are consistent – on the first run through they mark each submission in an absolute sense, and then they go through and adjust the marks slightly to ensure the better students actually get better marks than the others. All too often I’ve seen marking go haywire for this reason, because it’s difficult with traditional submission schemes (i.e. paper) to do all this adjustment.

There were also other really nice touches, such as personalised home pages for each student which listed relevant information, news, etc. There was also a real-time slide show feature of student’s works, updating as they were handed in. In this way students could see what everyone else was submitting, and get a real sense of engagement. The speakers noted that they’d ended up setting up dedicated computers in public places to showcases these slideshows, which gathered large audiences around submission time. Very cool.

It really emphasised, in my mind, how crap La Trobe’s systems are. We have a generic, boring home page for each department – all completely different styles and arrangements just to make things difficult – and online submission consists of a jumped-up cp clone (on the Unix machines), or a web form. No marks or feedback or usually given online, or if they are it’s just a crappy Excel document.

In fact, feedback on formal submissions is generally non-exist at La Trobe. Very counter-productive.

I really hope Ric (from La Trobe ITS) saw this session – someone really needs to light a fire under the administrata at La Trobe, and get the whole system out of the 19th century.

After that session was afternoon tea, followed by the Student presentations. Yikes! I was getting pretty nervous at this point, and really running out of time to get my presentation in order. But, more on that in the next instalment. :P

Leave a Comment