Lev Lafayette's journal

Four days of conference has come to end, starting with the relatively light Carpentry Connect NZ, which was basically the various educators and trainers involved in the software carpentries programme in NZ. About half the day was directly associated with co-ordinating people involved in the various organisations and their programmes, and half was some rather fun and indirectly practical improvisation games - and even with a nod of recognition to roleplaying games. I found myself quickly helping out with the curriculum for Genomics Aotearoa, mainly because I have already written such a course, and offering to help fix the HPC Carpentry, which is in dire need of such assistance.

Following that was the programme proper. I quite liked the opening keynote by Rosie Hicks and in the afternoon of the first day gave a short presentation about the Spartan HPC system in the cloud compute session and its relationship with the NeCTAR research cloud (as well as mentioning the University of Freiburg approach, all part of previous publications). On the following day I mainly got stuck into discussions about the use of Singularity containers in HPC and reminded people of how one can use EasyBuild to make Singularity containers. Blair Bethwaite ran a two-part workshop on the topic which should be converted into a course at UniMelb as well. On that note, today's presentation included an inspiring piece by Jo Lane on how the University of Waikato has introduced a postgraduate Scientific Supercomputing course on the good basis that they were tired of having the subject in the "shadow curriculum".

The conference dinner was at Lanarch Castle which is still charming, despite my multiple visits to this folly. I do take some macabre delight in with the gothic tale of romance and horror. There was a fair bit of Scottish regalia during the evening, including a recitation of Robert Burns' Address to a Haggis. I think it was performed by the same person who did it at the last eResearchNZ dinner in Dunedin, about ten years ago! Also speaking was one Ian Griffin, director of the Otago Museum and spotter of Aurora Australis and has some superb footage of such events (including the SOFIA flight). Anyway, given that his preferred photography spot (a jetty) is no more, I've offered the roof of my Masonic Lodge. The fact that he also has terabytes of unprocessed footage that needs some computational power is also a happy coincidence. I think this is the beginning of a beautiful friendship.

Also during the past few days, I made my way to visit the Otago University Roleplaying Society for my annual visit. We played Shadow Hunters, a rather interesting game of membership to secret factions and secret victory conditions, combining both co-operative and competitive elements. With my plentiful spare time, I have managed to make two submissions on my MSc dissertation, with another pending as the February 29 deadline approaches. Absolutely fascinated by the trends analysis I've conducted on the degree that open-source licensed software has made inroads in even personal devices and not just the server space. Finally, a Facebook group I started is now featured in The Guardian. Yes, it's that little brown rat again. I cannot help but feel a little bit of pride that this is actually happening in a substantive manner.

The past few days have been throwing myself into various activities after getting (mostly) over the cold last week. I have prepared myself for all the bits and pieces required for RuneQuest Glorantha Con Down Under, and I'm feeling quietly confident that it should all fall into place. Tonight we had a committee meeting of the Cooperative and pretty much checked off all the boxes of things that need doing. I've been composing Metaphysical Musings of the Scholar Wyrm which will be one of my three articles for the Convention special issue of RPG Review. In actual play on Thursday, there was another session of Star Wars: Force and Destiny where we're finding ourselves working for the Empire for the greater good (and they want it for the greater evil). Today I ran Eclipse Phase, where the proxies and their sentinels have been assigned a mission into a radioactive zone of sex-crazed mutant exsurgents; should be fun for them.

Last night was nephew Luke's birthday. After dropping off a nice bottle of whiskey for looking after our place during the European holiday, and as per last year. we took him to a local Thai restaurant, Sukho Thai, which does well with food, price, and decor - plus a nice drop of French red. After giving ourselves a hearty meal, we took Luke to see The Gang of Four, which Luke had familiarised himself with from my collection last time he did house-sitting. It was the fortieth year of their debut album, Entertainment!, and it was a pretty damn good gig. I ended up buying Andy Gill's guitar, making it the first electric guitar I've owned, which is pretty funny for someone who has been reviewing music as long as I have. Unsurprisingly, I've already written a review of the night which is available on Rocknerd.

One other pleasant distraction to this was visiting Brendan E., on the weekend who provided us with the viewing pleasure of Jarmusch's The Dead Don't Die. New housemate Hien N. was also present and we all had quite a good time watching this quite tongue-in-cheek movie with some rather good cultural names - Iggy Pop, Tom Waits, Bill Murray, Adam Driver. and a hilarious performance by Tilda Swinton. The somewhat mixed reviews of the film seem to have forgotten that this is essentially a comedy film, a spoof of the zombie genre, and to be honest there were some scenes where he could have turned it up a couple more notches in this regard. But it was jolly good fun and not to be taken at all seriously.

Apart from this, I've been working through the dissertation of my MSc thesis, specifically the draft of the literature review. My supervisor is being a bit ornery about getting all the specifics in and I suspect I'll have to provide a third version of the proposal even though I'm getting close to halfway in writing the draft thesis itself. I've also discovered that he doesn't actually read the entire thing - he made a request for the timetable in the last revision when it was at the end of the document. It really does strike me that sometimes various academic reviewers are not as careful as they should be.

Last night caseopaya and I went to a Viking-themed restaurant, Mjolner, which we decided was a birthday dinner for log_reloaded and Jase. Some of the affectations were nice, such as the drinking horns, and we had some fun deconstructing the artwork which had Norman, Viking, and Celtic elements. The food was OK, but all and sundry was priced in accord to what one expect at such as restaurant (i.e., more than it should be). The pre-dinner drinks at Otter's Promise were probably better. Afterwards was entertained back at their place with a show-and-tell with The Elder Scrolls V: Skyrim. All-in-all, however, a very good night out.

The day was also the completion of the final assignment for my MSc in Information Systems, which has just been submitted, hooray. I also received my grade for the previous assignment which, whilst a pass, was the lowest grade I had received in over thirty years. I asked the class leader what I had done wrong, for self-efficacy purposes, seeming that a similar subject I had done a few years previous averaged around 30% higher. It was like pulling teeth, but I finally dragged enough information out that the tutor was basically painting-by-numbers, judging references by quantity, and preferred recommended sources which were inferior to the ones that I found. In a nutshell, the marker was lazy; I hope I never advocate for people to do independent research and then punish them for actually doing so.

Speaking of such things, I've made a start on an automated summative assessment script for HPC training. It will essentially consist of a shell script and a heredoc with some fifty random questions for a fifty-minute exam. My MHed supervisor is quite keen on the idea, agreeing that its nothing short of absurdity to examine people's HPC skills through any other medium other than on an actual HPC; I am having a real uphill battle convincing people in the Internation HPC Educator's Forum of the virtue of this. Hopefully, I will have it ready for the next courses I'll be running in a week's time, a double-sized class. In addition, because I believe in making things easier for others, have written up some notes of installation, licensing, and testing of the Gurobi optimiser.

As a sign of a combination of good luck, a co-operative effort, a Bohemian lifestyle with professional employment, and maybe a hint of effective adulting, caseopaya and I paid off our mortgages today. If you had told me in my youth that I would be in the situation I am now, I probably would have responded with the old adage of the real path to wealth being "inherit it, marry it, or steal it", which is largely true in nearly all circumstances, Stakhanovite claims notwithstanding. The main thing that this does leave out is luck, which is actually the most significant factor, and is no wonder that on occasion people refer to "good fortune". Anyway, the practical upshot is that banks no longer own half our home which generates a nice sense of independence on one's mental state.

Last journal entry I mentioned that I had a somewhat distant Europe trip planned, centered on Zurich, as a residency requirement for my MSc in Information Systems. Well, since then the college has decided to extend the spots in the October residency, so I'll be going then instead. Hooray to completing this a couple of months earlier, because one really needs a fifth degree for the practical purpose of arguing on the Internet. It's a living example of the public sphere, you know. Be this as it may, I'm currently trying to plan an extensive loop of the following form: Zurich to Venice to Vienna, to Bratislava to Prague, then through Germany (probably Dresden, Leipzig, and through to Bremman), then on to Delft, and then Ghent, Paris, Frankfurt, Stuttgart, Freiburg and finally Zurich again. Anyway, that's the current plan, and there's a bit of time before I climb into the big silver bird.

In the meantime have engaged in the usual activities of work, study, and gaming. For the former, second half of the course, Advanced Linux and Shell Scripting for HPC was taught at the Peter Doherty Institute. I get the impression they were pretty happy with their new found knowledge of RegEx and the power of a useful shell loop when combined with a here-doc. For studies I have an MSc assignment to complete this weekend, primarily comparing different environment scanning approaches according to the proximity of activity. Finally, in the gaming hobby, funontheupfield introduced us to the rules-light Kids on Bikes, based on the pre-mobile 'phone era of which all players were familiar with. The building of the setting and character relationships is a cooperative venture, and we picked up from a previous setting a couple of years back, a not-quite Wonthaggi, for our rather fun game of Cats Against Cthulhu. That was somewhat light-hearted; this one went quite grim real quick. Yet, we've also retained the unknowably sapient cats and the eldritch horrors, which probably means it's going to end up being quite the surreal horror. I'm rather looking forward to it.

A good portion of this week was taken up teaching Introduction to Linux and HPC and Advanced Linux and Shell Scripting for HPC. They were a good class and quite switched on, even for people who were coming in with a relative lack of familiarity. It's a steep learning curve, but on the basis of the questions they asked (I tend to run classes more of an "interactive workshop") they were well on their way. Next week I have a repeat of the classes for immunologists at the Peter Doherty Institute. In my copious spare time at work, I've slipped in another conference presentation abstract and have continued work on my planned course for regular expressions, which I have been somewhat remiss in finalising - plenty of additions on speeding up grep added today.

In other teaching-related activities have expressed my displeasure at the intellectual laziness of the HPC Certification Forum in their continuing suggestions to use multiple-choice questions as summative assessment for the certification. I have argued, with backing in education theory, that they should be using actual practise on a real HPC system as a test of HPC system competency. For what it's worth my MHEd supervisor at Otago University agrees with the approach that I'm suggesting. Meanwhile I am making some progress with the last unit of my MSc at RKC/Salford for the dissertation, however, it seems that they have stuffed up my residency enrolment in Zurich; just as well I hadn't booked the tickets. For people teaching a postgraduate degree in information systems they're not very good at it.

All this aside, did manage to go out during the week, specifically for the Twilight Zone Movie at The Astor (and also an opportunity to visit Duke The Cat). The film was basically in the style of four of the old-style TV episodes, so it wasn't exactly all guns blazing, but it did have some nice plots with a dash of the macabre - and especially so given that people died in its production. Regardless of what is on at the Astor for me it is very much an opportunity to spend some time in front of a classic large single screen cinema and absorb the trappings of an old and slightly frayed deco beauty; after home and work it's probably the third most likely place one is to find me. The following night was science fiction adventures of a different fashion, with a session of Megatraveller, which involved dealing with the treasures of the Sindalian Empire - which turned out to be bacterial and nuclear weapons; whoops. I get the feeling that the ante of this story is arcing up and the poor ol' PCs are going to be on the receiving end of everything going wrong.

In my massive amounts of spare time that I have available, I've started my MSc in Information Systems Management and have completed the first two major online components (a reflective biography based on Drucker's argument of self-management) and a review of technological innovations of the past decade. Plus, of course, there has been a small mountain of reading. I have hoped that the course would be more agreeable thus far, much of the material is heavy on the business and observational side of the equation and less on the empirical. I've dragged as many conversations as possible in the course forums towards the latter, especially comparing predictions to reality. The first marked assignment will be on Sensors: exploring the opportunities and challenges for business. I suspect I'll do something related to the Square Kilometre Array, as that satisfies the criteria of being related to physical sensors, has the challenge of data processing, and I can even argue business opportunities, all of which are criteria for the study. Plus I have a few connections to the project.

The past few days saw the formal preparations for RuneQuest Glorantha Con Down Under fall into place; a total of 53 registered attendees, which is a pretty good number and a surprise last-minute game added to the schedule by a local Pendragon author, Garry Fay (Blood & Lust, The Spectre King), so that means RuneQuest Glorantha, 13th Age Glorantha, Pendragon, and Khan of Khans, all providing a pretty solid gaming stream to go alongside the panels, castle stream, auction, trollball, keynote and freeforms. Of course, I doubt very much that I'll have much opportunity to do anything of the sort myself. Still, I have managed to get a game of Megatraveller and RuneQuest Questworld in this week. The latter was particularly good fun with a new player in the group who named his Dwarf character Mark Almond; the obvious pun has already lead to some character development.

In work-related events I had a little win this week with an DNA application software project, pychopper taking up my suggestion on having numbered releases. There is many, too many, software projects these days which don't bother with old-fashioned things such as numbered releases, which make it a little annoying to track changes in application development. Yes, one can, and sometimes has to, track by commit ID. But who enjoys doing that? Plus, I've been meaning to post about IBM's recent purchase of Red Hat, but needless to scale of the purchase is pretty epic. Finally, and completely off-topic, today I finished the French to German tree on Duolingo. Most probably next course will be German to French.