Research Platform Services Blog
Month
By Paul Mignone
The technical progress in open-source 3D printers (e.g. the RepRaps) in recent years has skyrocketed. Until recently, most of this progress has been limited to 3D printers that produce only polymer or resin based products. The lack of an open-source metal 3D printer means $$$$ for those that want to produce metal components.
Well, not anymore…
A group of researchers at the Michigan Technological University have developed (what appears to be) the first, low-cost, metal 3D printer. The open-access paper is freely available, and provides information on the bill of materials, electrical and mechanical design schematics, as well as basic construction and operating procedures. More information on the printer can be found on it’s appropedia page.
The printer is basically a modified version of the Rostock delta design, using a gas-metal arc welder to deposit the metal. The initial design has already been able to produce functional metal parts, and costs less than $2000 to make. This is AWESOME news for researchers looking to produce higher-strength components for their next experiment.
To put it in perspective, commercial metal printers start at around $500,000!
I know what I’m building next ;-)
Image Source: http://www.appropedia.org/Open-source_metal_3-D_printer
(Every Thursday, 3-4pm, Tsubu Bar)
By Damien Irving.
I was sitting at my desk the other day struggling (as per normal) with some code, when it occurred to me that within a 500m radius of my office, there was probably half-a-dozen (or more) people who could quickly and easily solve my problem. (Conversely, there was also probably half-a-dozen people wrestling with problems that I could solve, but at the time I was too frustrated with my own problems to worry about them!)
This should have been a very reassuring thought, however for all intents and purposes those half-a-dozen people might as well be living on the moon. Like most researchers, I rarely interact (in an academic sense) with people outside of my Department, and if I do I’m certainly not pouring my heart out to them about the programming issues I’m having.
we even got an umbrella at #hackyhour this week! @ITS_Res pic.twitter.com/KZ9AfEo4oL
— Katie Ewing (@katieaewing)
March 13, 2014
In an attempt to address this problem, we started running “Hacky Hour” (puntastic, right?) in early 2014. Every Thursday afternoon from 3-4pm, the Research Bzaar team leaves their offices and converges on Tsubu Bar (just look for the table with all the laptops!). Reseachers who’d like assistance with their coding are welcome to join us and ask questions, or you can also just come down and work on whatever you’re working on, in the company of other researchers who are into coding. We’ll either answer your question on the spot, or link you up with someone in the Research Bazaar community who can. Think of it as a help desk, with the added bonus of food and beer!
Another happy researcher! @action4climate came to #HackyHour today and @DrClimate solved his #netCDF and #Python problems! #ResBaz
— Research Platforms (@ResPlat)
October 30, 2014
Speaking of happy researchers, @Isa_Kiko was stoked to find that @andreabedini could solve her #HTML problems at #HackyHour! #ResBaz
— Research Platforms (@ResPlat)
October 30, 2014
Matchmakers! @andreabedini came to #HackyHour looking for help with @openstreetmap, so we hooked him up with our mapping guru, @stevage1
— Research Platforms (@ResPlat)
October 30, 2014
We’d love to see you there - feel free to join the Facebook event if you need a weekly reminder!
By Damien Irving and Steve Bennett
Data visualisation is crucial to the atmospheric sciences, in which large, complex 3D datasets are researchers’ bread and butter. It’s therefore very fortunate that Martin Jucker - a 3D visualisation guru (and polar vortex expert) from Princeton University - has been visiting with the ARC Centre of Excellence for Climate System Science this month. In between meetings and seminars up and down the east coast, Martin kindly offered to run a half-day workshop on ParaView.
A group of 20 atmospheric scientists from Melbourne and Monash Universities benefited from Martin’s expertise.
ParaView is a powerful scientific visualisation tool that can be used on your desktop, or scaled up to run on an a high performance cluster. If you like the look of the images below, let us know and we can look into offering more courses/support for ParaView. If you came to the bootcamp and would like more information on the routines that Martin presented, check out this paper that he published soon after the workshop.
The first half of the workshop was spent creating this visualisation of the polar vortex - a region of strong winds in the upper atmosphere that was recently linked to some very cold weather in North America.
The final visualisation was of a heatwave that affected south-eastern Australia in March 2013. Shown is the surface pressure (contours) and upper atmosphere winds.
Great post from @peterseibel on code reading groups: http://t.co/9dleujHuwM. Important lessons for our @MelbITSResearch #HackerWin meetings
— Damien Irving (@DrClimate)
January 21, 2014
If you work in a small research team that writes code, this is a fabulous opportunity to get assistance from programming gurus in your field.
If you’re wondering whether you do enough coding to call yourself a “hacker,” this article has your answer…
If you’re a researcher who writes code, this is possibly the most useful paper you’ll ever read. Written by a group of Software Carpentry instructors, it nicely summaries many of the skills that The Hacker Within and ITS Research can help you develop.
By Damien Irving.
At the inaugural #HackerWin meeting to be held in a couple of weeks time, we’re going to discuss what the group should do over the next 12-months and beyond.
To get the conversation started, I wanted to document my thoughts on the topic. First and foremost, I’m suggesting that our mission should definitely be to have all #UniMelb (and eventually all Melbourne) researchers coding like in the movies… ;-)
In order to help researchers achieve movie-like proficiency, my vision for the group centres around regular meetings and three key activities:
Consultations / code review
Informal presentations on packages / tools
Feedback to ITS Research on the needs of the research community
The first of these activities would be achieved by building up a database of willing tutors. For instance, I would be willing to put my hand up to help people who are having trouble with the Shell, Python, or Git. Others might be willing to help people with Matlab, R, and so on. We could then invite graduate students and university staff to request assistance via an online form. They would be linked up with a suitable tutor for a one-on-one consultation, or they could bring their problem to the next fortnightly meeting and have the entire #HackerWin group work on it.
Which of these tools/languages/approaches would you be willing to help other researchers with?
In addition to live problem solving (or ‘hacking’), the regular meetings could also involve informal presentations. For instance, I’ve recently started using a Python style checker called Pylint, so I could give an informal presentation on what I’ve learned about programming with style. This mixture of hacking and informal presentations would provide the ongoing support and development opportunities that a one-off Software Carpentry bootcamp cannot.
Pair programming is a well established practice in the software development industry, and yet researchers rarely look at each other’s code. We need to change that. [Source]
Perhaps most importantly, #HackerWin would also provide feedback to the ITS Research department on the needs of the community. If Greg senses that the graduate students in his Astrophysics and Astronomy group would benefit from a short course on data management using R, then the #HackerWin meetings would be a perfect place to bring this to the attention of ITS Research.
Of course, this is only my personal vision for #HackerWin. I’m hoping others will have many additional and better ideas, so please come along to our next meeting and share your ideas!
By Damien Irving.
While the rest of UniMelb was winding down at the end of last year, 35 PhD students and early career researchers subjected themselves to an intensive two-day (four-afternoon) Software Carpentry bootcamp.
Researchers from UniMelb (plus a few from Monash Uni) working hard at improving their computer programming skills during the recent Software Carpentry bootcamp.
Why do researchers need to learn basic computer programming skills?
In today’s web-enabled world of big data and cloud computing, researchers from all corners of the campus are spending more and more time wrestling with software. Since most researchers are self-taught programmers, they spend hours doing things that should take minutes, reinvent a lot of wheels, and still don’t know if their results are reliable.
Errors in your code can come back to bite… read more about this scientist’s worst nightmare here.
What is Software Carpentry?
Not knowing the basics of programming is an inefficient (and stressful) way to work, so it’s no surprise that Software Carpentry - a volunteer organisation that teaches basic software (i.e. programming) skills to researchers - has exploded in popularity in recent years. They recognise that researchers don’t need to be professional software ‘engineers’ to get their work done, but a solid understanding of basic software ‘carpentry’ skills is now essential for modern-day researchers. Check out their website for more information.
Over the past two years, Software Carpentry bootcamps have been held all around the world (see here for details).
Given the success of our inaugural Software Carpentry bootcamp, the ITS Research Services department is committed to running more in 2014. If you’re a researcher who has to code (via Python, Matlab, Rstat, etc) and would like to learn how to be a better “hacker,” please get in touch.
Beyond bootcamps… The Hacker Within
More bootcamps for researchers is obviously great news, but it’s not enough. The adoption of good programming practices takes many years, so ongoing support through a community of practice is critical to improving the computational competency of researchers. Accordingly, ITS Research Services recently announced a new community called “The Hacker Within,” where researchers can come together to collaborate with one another.
Give a researcher a fishing rod and they can feed themselves, teach a researcher to fish in a community and they can collaborate to invent new ways of catching fish! [Source]
The Software Carpentry bootcamp was the first of many events that The Hacker Within will be hosting over the next 12-months. If you’d like to get involved with this exciting new research community, we’d love to see you at our next meeting.
By Damien Irving.
The problem: how do we help researchers use cutting edge digital tools, so that they can do cutting edge research?
Researchers use more than just the Microsoft Office suite of software to do their world class research. Providing support for all the different tools out there is a significant challenge. [Source: David Flanders]
Traditional answer: The typical IT support model would suggest that the ITS Research department (#ITSResearch) should provide a help desk for researchers, staffed with a bunch of programming gurus. While this would be great, the unfortunate reality is that there isn’t enough money (nor expertise on campus) to support all the discipline specific tools, data and methodologies that researchers use.
Our answer: A better solution is for the graduate students and research staff at #UniMelb (and perhaps even #MonashUni and #RMIT - the more the merrier) to form a community and help each other.
Inter-university rivalries are fun in a sporting context, but not particularly useful in research. We’re hoping our programming community can reach across institutional (and departmental) boundaries.
If Sarah from the #UniMelb Biology Department is having trouble with a particular programming task, chances are somebody else at her university (perhaps from the Earth Sciences or Medicine Department) or at a university nearby (e.g. #MonashUni, #RMIT) has encountered and solved that very problem. In order to help Sarah link up with that somebody else, #ITSResearch wants to assist in establishing a new community known as The Hacker Within (#HackerWin). In other words, a group of researchers who write code, helping other researchers who write code.
The #HackerWin concept was pioneered at the University of Wisconsin a few years ago. This article explains how it started and where the name comes from. [Source]
Who is going to lead the “Hacker Within” community?
That would be me… hi, I’m Damien! As part of my role as a Research Community Coordinator with #ITSResearch, I’ll be leading the #HackerWin community. When I’m not working on my PhD (looking at climate change in Antarctica), I’m a qualified instructor with Software Carpentry and also blog and tweet (@DrClimate) about research best practice in the weather/climate sciences. In other words, I’m pretty passionate about this stuff! Of course, a community is not just one person, so I’d love to catch up over coffee and hear your ideas!
I gave a presentation at PyCon 2013 about Software Carpentry activities in Australia. As I said, I’m pretty passionate about helping researchers improve their programming!
Who has already joined #HackerWin?
Off the back of our first event (a Software Carpentry bootcamp that you can read about here), we already have a core group of people attending our formal monthly meetings:
Artika Nath (Medical Systems Biology)
Bernd Merkel (Radiology)
Errol Lloyd (Neuroscience)
Isabell Kiko (Biomedical Engineering)
Mitchell Black (Earth Sciences)
Katie Ewing (Biomechanics)
Michael Wang (Chemical Engineering)
How can I get involved?
All those interested in being a part of #HackerWin are encouraged to come along to Hacky Hour.
By Katie Ewing
On October 24, as part of the Campus Research and Education Week (CREW) at the Royal Children’s Hospital, ITS-Research hosted a “Speed Networking” session that showcased the different tools and services available to the research community. The format allowed the attendees (including physicians and researchers alike) a chance to briefly “date” each of the tools in five-minute blocks - deciding which resources would be most useful in their practices or research.
Like speed dating, the speed networking session provides an initial encounter, but the real connection occurs after the event. Based on interest in learning more about these tools, we will be providing training workshops on several softwares, including Galaxy, TileMill, and 3D Printing, research data management and video interviewing skills in the near future. Stay tuned for further details via the feeds on our event page: http://its.unimelb.edu.au/research
In 2013 the ITS Research department hosted a number of workshops aimed at training researchers in using TileMill tools, Cloud-based software specialised in polishing and producing maps. The availability of these tools in the Cloud, along with the workshops, provides direct support to academics and researchers looking to give their research a more professional and captivating finish. The image below showcases the final product.
Page 3 of Lauren Gowne’s (The University of Melbourne) ‘Report on the Relationship Between Yolmo and Kagate’, Himalayan Linguistics.
The only questions remain: will the researcher’s paper accumulate more downloads, and will they be able to tell their ‘research story’ better and thus produce more citations?
There’s no treasure map - only time will tell!