- Your latest published article has a graph in it. If the eResearch police asked you to reproduce the plot exactly using the original data you’d:
- Check out the code archived with the article, and re-run the make-file, which would not only re-generate the plot using Knitr, but the whole article, which would also be made available as an interactive website using Shiny with an option to re-run the models on data which is crowd-transcribed from the logs of 17th century slave ships.
- Redo the diagram in Excel, using the clearly set out method and supplemental material from the article.
- Find the data (by borrowing back last year’s laptop from a postgrad), then fiddle around with what you think is the right spreadsheet make something that looks pretty much like the one in the paper.
- Plot? What plot? And what was all that babble in option A?
- Turns out that some of the photos and recordings you made when documenting a research site contain images and sounds of a Yeti. If you can provide complete records of where and when you collected this data, you can collect a $1,000,000 prize from a cable TV station. Your next step is to:
- Provide the DOI to the dataset which you have archived in your institution’s data repository. The repository record with the data attached provides all the information required to support your claim.
- Scan the relevant pages from your field notebook and annotate these with supporting information specific to the Yeti sighting.
- Rummage around the office: you last saw that scrap of paper you scribbled on during the fieldwork with the pile on top of your filing cabinet.
- Quickly throw together some handwritten notes and scorch them with a candle so they look old. No actually, you couldn’t be bothered. Also you don’t believe in Yetis or Santa.
- You have so much data to analyse and your models are getting so complicated that your laptop is getting hot, so you:
- Use Docker to create a 128 Node compute cluster in the NeCTAR cloud, get some results, archive all the code, data and outputs with DOIs and go home early.
- Enrol in Intersect’s High Perfomance Computing (HPC) courses and learn how to run your job on shared infrastructure.
- Give it to one of the PhD students to sort out.
- We have you mixed up with someone else – your iPad never gets hot unless you watch too much YouTube in the sun.
- When archiving data you always:
- Take care to use standard file-formats that are easily machine readable, and make sure all code and as much provenance information as possible, are also archived.
- Fill in the metadata fields on the institutional data catalogue application as carefully as you can.
- Try to change the worksheet names on your Excel files from Sheet 1 to something more meaningful, if you get time.
- Use the shredder in the research office. It’s more fun than the old technique of scrunching up the envelope on which the data were written and trying to get it in the bin for a three-pointer.
- The best place to store research data during your project is:
- On a secure, backed-up, cloud storage server (with data held in an appropriate jurisdiction) which you can access from anywhere with an internet connection, and share with designated collaborators.
- On a secure, backed-up drive accessible only from your office.
- On a Dr Who USB stick.
- You delete your raw data after you’ve analysed it. Although, actually, sometimes raw data doesn’t agree with you; so you cook some up to better fit your conclusions.
- A data management plan is:
- An important tool which facilitates planning for the creation, storage, access and preservation of research data. Creating this at the start of a research project and referring to it as a living document informs the research workflow and specifies how data will be managed.
- Something to think about once you’ve collected some data.
- More paperwork to bog down the research process, like Ethics. Oh for the good old days when we used to be able to electrocute the students without filling out so many forms.
- Data management plan? I’m not even in management so don’t interupt me, I’m enjoying my holidays
- Collaborative research is:
- Enabled by eResearch technologies and supported by Open Access to published research data.
- Maximising the funding universities receive by sharing resources and equipment for a research project.
- Popping next door to ask a colleague a question.
- Not something you’re interested in. Your data will die with you.
- If you wanted to share your completed research dataset with others you would:
- Contact the Library or eResearch and discuss publishing the data and related methodology to the institutional data catalogue, which can then also be included in the Research Data Australia discovery portal. The data would be described using appropriate metadata, and linked to related collections, fields of research, people and facilities.
- Publish the data on your personal website and ask people to contact you via a hotmail address for more information.
- Email the file to colleagues you think would be interested
- You told us before – your data will die with you.
Mostly As – We’d love to talk to you about becoming an eResearch champion. You have embraced the benefits of eResearch technology and methodology and have put comprehensive plans in place for the use and re-use of your valuable data.
Mostly Bs – You understand that technology is a useful tool but you’re hesitant to rely on it for your research. Try putting aside your trust issues and play around with one new tool or habit this week – it might spark an idea or save you valuable time. There are lots of opportunities to attend training or do a self-paced online course to increase your comfort level.
Mostly Cs – It might be time to chat to the eResearch team about joining the 21st century. Although your existing research process may be valid, eResearch boosts the research process through opportunities to add computing power, streamline workflows, and collaborate with like-minded researchers from around the globe.
Mostly Ds – Bah, humbug.
How eResearch-y are you? An extremely serious quiz comprised of eight multiple-choice questions. by Peter Sefton & Katrina Trewin is licensed under a Creative Commons Attribution 4.0 International License.
Thanks Kim Heckenberg for your input and sorry Alf, we didn’t put in anything about multi-screen immersive visualization.