Dissertation writing

I'm working on writing up my dissertation, at the end of my PhD now. I've made good progress. I have another chapter or so to go before I should be able to submit. My supervision team has been reviewing and editing individual chapters as I finish them. We are already working on a paper, and the pre-print for this had gone through multiple rounds of review. So the chapters that came from the paper didn't need too much work. Another chapter is based on the reports I wrote for my assessments (back in 2015!), but having it all written down also made it easier to edit and add to the dissertation. I am extremely glad that my supervision team got me to maintain a daily lab journal. It makes writing the dissertation so much easier.

Research fellow at the Silver Lab at University College London

I was fortunate enough to secure a research fellow job at the Silver Lab while working on my dissertation. Generally, we PhD candidates submit our dissertations and then spend a few months hunting for positions. From all accounts, it's not meant to be an easy transition to a post-doc position. The number of PhDs entering the job market easily outnumbers the number of available research positions. So, I was also preparing for a few months of unemployment---saving money primarily---while I hunted for positions after I had submitted. In my case as an international migrant/expat/job stealer, my employers would also have to sponsor my visa, and not all employers do that. If I hadn't managed to find a position before my current student visa expired, I'd have to return to India and continue applying from there. That makes it even harder, and all the time one is not working on research, one is already falling behind. So, transitioning on to a position while still writing up was a very very lucky break for me.

The research group does a lot of experimental work, but it is also where the Open Source Brain project is based. Given my computing background, and experience with FOSS in Fedora, a large component of my role is to work on the development of the Open Source Brain platform, and liaise with MetaCell who do most of the core development. Along with that, I get to work on modelling and other research projects. I was looking to work in a group that included experimentalists. I think that it is important for me to develop as an independent researcher in neuroscience.

It's early days yet. I am only two weeks in, so I am still settling down. Since I am on a Tier 4 student visa at the moment, I can only work twenty hours a week. That's about two days a week, and is really not enough to get a lot done, especially given that it is meant to include the various meetings that I am to attend. HR are working on getting me my Certificate of Sponsorship (CoS) so that I can apply to transition on to a Tier 2 work visa. It is similar to the Certificate of Acceptance of Studies (CAS) that international students must get from their universities before they can apply for their Tier 4 student visa. Sponsoring organisations can sponsor a limited number of internationals each year, so we're waiting for the new cycle to start this month in April when UCL will be able to sponsor more of us.

In the two weeks that I've been in, I've been learning the development process that Open Source Brain follows: attending sprint meetings, and the sort. I'll write a post dedicated to this in a near future. It is on my to-do list. This week, we diagnosed and fixed an issue with the current deployment. Sendmail was blocking the server because the Docker container hadn't a FQDN as its hostname. The things we learn. If you do find any issues with the Open Source Brain platform, please file an issue (or e-mail me if you prefer).

There's more work to be done: the deployment is being moved off AWS on to the Google Cloud Platform. It's simple enough, but of course, the deployment needs to be tested and validated before it can be declared live and the AWS instance torn down.

I've also been learning how the research group works: getting to know the people, what their interests are; attending the group discussions and journal clubs; keeping an eye on various journals to share new science that may be interesting to us. I've already learned quite a bit from the discussion. Given the COVID situation, we're all working from home, so everything is happening over Slack and video calls. I had managed to go to UCL for my induction in my first week. I was the only one there for that particular session. Again, I was lucky, since I expect that was the last induction session before UCL decided to limit access.

Fedora

We're nearing the Fedora 32 release, so I worked on the bits remaining for the new CompNeuroFedora lab image. Based on the discussion at the NeuroFedora meeting, I passed all the information needed to set up a page for the lab to the Websites team.

The general package updates continue. I just updated Brian2 to the new version this morning and pushed an update with a test case. The test case takes one through the tutorial, so if one is looking to learn how to use Brian2, this is a good way of doing it while contributing to NeuroFedora. Another few bugs were fixed and updates pushed too. I've got to work on packaging a few new tools that are on the list.

On the Fedora-Join front, we've had a few more folks join the community to help out. It was lovely chatting with new folks and discussing where and how they'd like to work with the community. Needless to say, lots of cookie giving has occurred in the IRC channel.

I've also been thinking about the lack of a process for Community Changes in Fedora. Why isn't there something similar to the Change process that we use for dev changes? I finally filed a ticket with the Council. It's being discussed on the council-discuss mailing list. I've also asked Mindshare and CommOps to weigh in this morning. Please feel free to jump in and discuss how we should go about this. A change process that focusses on community is important, in my book.

The Git forge discussion continues on the -devel mailing list, so I've been keeping up with that. I would prefer Pagure myself, and I do understand the CPE team's view even if I don't necessarily agree with it.

Organisation for Computational Neuroscience: OCNS

Things are quite quiet in OCNS. The Board has been discussing how best to handle the CNS*2020 conference. An announcement will be made once a decision has been reached.

Review comments

This turned out a lot longer than I'd expected. As I settle down to a weekly post, it should get shorter. However, I do see that I've got lots going on, and perhaps I do need to be more disciplined when accepting/volunteering for tasks, and prioritising them once I've taken them up. I'm not cookie-licking at the moment, so that's quite good. Still, lots to do.