Journal archives for August 2020

August 18, 2020

Member profile - Matt Tank

Writing an Australasian Fishes bio blurb in COVID-19 Australia is an interesting experience. As I write, the country is experiencing the inevitable second wave, during the normal winter flu season. Parts of the country have entered stringent lockdowns, while others have closed borders, sealing in their healthy citizens. Other states are somewhere in between, putting out viral spot fires, in hopes of keeping the pandemic under control, for a second time. Another defining aspect of the country’s response has been the widespread practice of working/studying from home. We now see the most widespread use of this practice in history where never before has such a large percentage of the Australian workforce has been working remotely. In some ways, this is not unexpected, as most futurists have predicted that working/studying/shopping from home will be a wave of the future. That wave has arrived a decade earlier than predicted.
In my past, I was responsible for writing pandemic plans, and have always regarded “working from home” as the weakest link in pandemic response. At the time, it was a frequently cited work-around, however, it was a logistical impossibility. Limited internet bandwidth, application licensing restrictions, a dependence on corporate portals to access corporate applications are among several technical reasons limiting user’s access.
In 2020, I was pleasantly surprised to see that most of these problems were solved before the arrival of COVID-19. Thanks to broadband, we now have the capacity for more users to simultaneously access the internet. Many corporate applications have moved from internal servers, to the cloud, allowing almost universal, unlimited staff access. The reason I bring this up is that our ability to work from home, as a nation, is the result of IT professionals, who skilfully migrated us to these new platforms, allowing numerous government and business operations to continue to function, even though the entire staff are working from their respective homes. Upon reflection, I realise this capability was given to us as a result of IT experts who paved the way for this transition. The subject of this bio blurb is one such IT expert, Matt Tank, an IT consultant by profession, based in the Southern suburbs of Adelaide. Matt explains, “I specialise in Cloud Technologies, mostly this is about adapting existing business systems to be Internet-based for cost savings and flexibility. However, more and more I’m being asked about ways to make these systems better, using technologies that are much easier to implement in the cloud.” I doubt if Matt would have thought his work in implementing this technology would save the bacon of so many companies and perhaps change the fundamental office working relationship for many years to come.
Matt’s introduction to our program, was, of course, the result of his pursuit of alternative, online technologies. He explains that some of the technologies he implements are things we might recognise as iNaturalist users. Examples include Computer Vision and geospatial reporting (those great observation maps we see), are related to the technologies he implements. He notes, “In fact, it was professional curiosity that brought me to iNaturalist. I was fielding questions for clients about Microsoft Custom Vision, and I decided to build a Fish ID solution as a side-project. Image searches for testing images brought me to iNat, and the rest is history. I started posting my own images, and like so many others, was approached by @markmcg to join the project. Unlike most of the regular contributors to the Project. I discovered iNaturalist while working on a Computer Vision proof-of-concept, and realised not only that I could use the site to help me ID the things I didn't recognise, but I could also help others with what I know.”
Matt says that he is neither a scientist nor a photographer, but like most in the project, his love for marine life started early. He recalls, “TV was an inspiration, but not documentaries at first. It was actually the movie Jaws that sparked my interest. I was way too young to be watching it, but also probably too young to be scared by it but became interested in sharks from virtually that day. The documentaries came later, and my interest widened to fish and other marine life. In the 80s and early 90s, most marine documentaries covered the tropics, which shaped a lot of my early knowledge.”
His early passion for marine life has remained as part of his character and he joined Australasian Fishes in February 2018. Matt’s contribution to Australasian Fishes has placed him in 22nd place in observations, with a total of 771. However, many project participants have benefited from his numerous identifications. For our project he has assisted in 6,505 identifications for us, ranking him 7th in that category. Matt has not only assisted us, but has contributed over 2,736 observations to iNaturalist projects, making a remarkable 21,258 identifications for the benefit of citizen science.
This passion for the marine world, however, did not result in a career in science. He says, “Over the years, my future career as a Marine Biologist slipped out of reach (you have to work hard at school, who knew?), and I was only diving sporadically. I wasn’t really interested in joining a dive club, and my friends were starting to move away and get busy (so was I for that matter). A trip to the Cook Islands in 2016 changed all that. I didn’t (SCUBA) dive, but went snorkelling every day, mostly alone, with a GoPro to record what I saw. I realised again how much I enjoyed it, and living 10 mins away from Port Noarlunga Reef, a world-class diving location, I made the decision that I was going to make sure I was in the water as often as possible, even if it was alone with a snorkel (maybe that’s even preferable – not many people want to wait around for 10 mins while you get the perfect shot of a sponge). I usually get into the water about 20-30 times a year now, and mostly in warmer months. With a huge increase in local experience, my interests expanded to include local fish species, and with less fish diversity than in the tropics, I also started to pay more attention to the hugely diverse invertebrate populations of Southern Australia. Ascidians in particular are of interest, because there are so many unknowns, and of course many of them look spectacular. I don’t really have a favourite location for diving in SA, because there is such a great variety of environments, but I always come back to Port Noarlunga, an excellent reef and jetty dive. Myponga Beach is a mix of huge tidal pools and an underwater wall that quickly drops down to about 6m, and the seagrass meadows of Kingston Park are great for weed whiting and leatherjackets.”
Developing his interest in underwater photography was based on a need to feed his thirst for undersea creature identification. Like other leading project participants, he is driven by a desire to know and identify the creatures he sees while underwater. He recalls, “Being the first time I snorkelled with a camera, another thing that Cook Islands trip showed me is how much more you remember individual dives when you can identify what you are looking at, and how much value that knowledge adds to future dives. As I mentioned earlier, I’m not a photographer, so for me a camera is really just a tool for identification. Recently though, I’ve realised that when you’re looking at something really small, a GoPro won’t cut it, so I’ve recently purchased a new underwater camera, an Olympus TG-6 with the flash diffuser attachment. While acknowledging my limited experience with it so far, I would definitely recommend the product if you don’t want to lug around a lot of equipment or are on a smaller budget. Its microscope mode is really handy, for observations like this little guy here:”
Looking back, he feels that using iNaturalist has been very rewarding, in a number of respects. “If you’re willing to put in the effort, it’s a great learning tool. Professionally, I haven’t seen a better resource for building a repository of images about particular species. The AI aspect of it is really coming along too, and I’m looking forward to seeing where this goes. I think in the medium-term, we’ll get to a point where a single model based on all life won’t cut it, and that these systems will start to think in a more hierarchical way like we do. Instead of one model, it will iterate through multiple models; first of all, to decide what type of organism it’s looking at, then a more specialised model based on its decision. It’s going to be a lot more accurate if the system first decides it’s a fish, then used its knowledge of fish to identify the observation, rather than just comparing to the 250K+ list of (current) species. Combine this with technologies that can analyse the description field to work out the submitters intent, and other similar functionality, and we could end up with a system that can do a lot of the heavy-lifting, and leave the experts freer to fine tune IDs rather than constantly fix mistakes. That’s not even talking about the ways that researchers could plug the data into reporting and/or machine learning solutions to make their own findings. This is where a project like Australasian Fishes could really come into its own, and the groundwork starts right here, where we all contribute to building that repository of knowledge.”
In conclusion, Matt wanted to use this bio blurb to express his appreciation to other project members. He says, “This is also an opportunity for me to reflect on the help I’ve received over the years, so in regard to Australasian Fishes specifically, I’d like to thank (in no particular order) - @maractwin, @sascha_schulz, @joe_fish, @davemmdave, @marinejanine, @kendallclements, @rfoster, @clinton and @markmcg… And everyone who contributes the observations that help so much in building my knowledge - and our collective knowledge.”
This journal post was written by Australasian Fishes member Harry Rosenthal.
Posted on August 18, 2020 02:53 by markmcg markmcg | 6 comments | Leave a comment