Nicola Romanò @nicolaromano@qoto.org
- Pronouns
- he/him
Senior lecturer at Edinburgh University and Zhejiang-Edinburgh Joint Institute (ZJE).
Undergraduate Programme Coordinator, Biomedical Informatics at ZJE.
I teach #imageanalysis & #dataanalysis with #RStats & #python.
My research is focused on how #heterogeneous behaviour in #pituitary (and other) cells shapes their function as a population.
I'm also very interested in #reproducibility and #openscience.
Joined Oct 2022
Nicola Romanò
boosted
It's becoming increasingly clear to me that Reflect Orbital's fucking stupid giant mirror satellite, with absolutely NOTHING useful to offer, which will cause countless safety issues, ecological disasters, and destroy the night sky, is going to launch.
A bunch of astronomers and I have sent out a fact sheet about them to a bunch of journalists, but very few are going to write about this. So, let me try posting it all here.
Here's what I know about Reflect Orbital and all the downsides:
re: Long post about data sharing vs not
@jonny @elduvelle I might have been unlucky, but every time I've asked for data to the authors of a paper who said that data was available on request they either ignored me or the data had issues ...
Anyway, two interesting reads about this
Why don't we share data and code? Perceived barriers and benefits to public archiving practices - Gomes et al 2022
https://royalsocietypublishing.org/doi/10.1098/rspb.2022.1113
Why data is never raw - On the seductive myth of information free of human judgment
https://www.thenewatlantis.com/publications/why-data-is-never-raw
Nicola Romanò
boosted
Nicola Romanò
boosted
Oh, no! My R package {safespace} is in a broken state - won't someone (new to PRs) help me fix it???
I'm renewing my offer to guide newbies through the R package building / fixing / reviewing process during Hacktoberfest - see this post
https://jcarroll.com.au/2024/10/01/a-safe-space-for-learning-how-to-make-pull-requests/
Open a pull request on https://github.com/jonocarroll/safespace to get a mentored review of your submitted changes with zero risk of breaking anything valuable if you mess it up completely.
Please boost for visibility!
Nicola Romanò
boosted
Age verification laws haven't been around very long and they're already showing how dangerous they are https://alternativeto.net/news/2025/10/discord-s-recent-support-data-breach-exposes-user-info-id-photos-and-credit-card-details/
Nicola Romanò
boosted
@reinouts @nicolaromano @Kiloku @eb
The explanation for the bafflement is: there is no revenue in LLMs, but there's an outrageous bubble of investment, particularly also by Microsoft. So LLMs get crammed into any product, useful or not, in order to eventually upsell you.
Instead you get an LLM that tries to guess relations in a table - and indeed one can be happy that the answer isn't "Wednesday". But that doesn't make it useful.
@onterof @PalmAndNeedle @Kiloku @eb I suspect some big investors want to see AI in every software they put money in. Doubt the decision came down to a bunch of programmers...
Also, hype is a funny thing. I've seen students develop suboptimal, very convoluted solutions to fairly simple problems so they could put some transformer or whatnot in there. They're baffled when they don't get full marks...
@PalmAndNeedle @Kiloku @eb Because code is a series of words, which is what LLMs generate. They're not designed to do calculations.
Indeed, sometimes when asked to calculate something these systems generate code that when executed gives the (mathematically correct, of course) answer.
The reality is, very often LLMs generate good, working code. The issue is not that. There are big ethical and environmental issues. Also, because the code works often, but not always, you need to double check it at all times, effectively taking longer then writing it by yourself (eg https://metr.org/blog/2025-07-10-early-2025-ai-experienced-os-dev-study/).
@Kiloku @eb If you use a hammer to eat soup you generally don't end up with good results.
If you want to sum numbers use the sum function, not one that predicts the next most probable token. I dislike this type of bashing of LLMs because it's trivial to dismiss (ok they can't do trivial maths, but they can write an entire piece of software for me). There are much more risky outputs that could be used as an example. Funky Excel formulas have always existed...
@typotheque great survey! A couple points.
1. In the initial test in the "tutorial" part, you're asked to change weight up and down until it's normal... I was stuck there for a sec because my normal wasn't what the system expected. Maybe have an "OK, I understand" button?
2. Great to see you thought of data management etc, but you say people can email you if they want to withdraw their data, yet you don't store any identifying data...
Nicola Romanò
boosted
People perceive thickness of letters very individually, and that’s why we are conducting Font Weight Survey to understand the regional preferences of font weight. Could we ask you for 10 min of your time? The findings will be published publicly once we have some results.
http://tptq.com/weight
@dtabb73 Well, the majority of computations during training is run on the GPUs though...
Nicola Romanò
boosted
🌟 Save the Date! BioC2026 is Coming! 🌟
Mark your calendars! 🗓️ The Bioconductor Conference, BioC2026, will be at Fred Hutch Cancer Center, Seattle, WA, from August 10-12, 2026.
Join top experts in bioinformatics, data science, and computational biology to share insights and explore Bioconductor's latest advancements.
Stay tuned for registration, speakers, and program details. Visit: https://bioc2026.bioconductor.org/. 🎉
Nicola Romanò
boosted
"This study implements a multi-year analysis (2019–2023) comparing the performance of Open Access and legacy publishers. Using a set of financial ratios—grouped by profitability, liquidity, efficiency, and solvency, as well as data on firm size (revenues, assets, and employee counts), we assess their financial performance. The results indicate that legacy publishers have enormous scale, stable profitability, and high leverage, but low liquidity and return on equity."
Nicola Romanò
boosted
Software Carpentry Python Workshop – Join Us This November!
We’re excited to invite you to our free workshop, open to all!
📅 : 4-6 Nov
⏰ : 9 am-2 pm Eastern Time (UTC -5)
🛜 : Online
This workshop is designed to be interactive and hands-on, providing basic lab skills for those in research computing roles.
Free for anyone to attend—please feel free to share with your friends, colleagues, or communities who might be interested.
Coming back home to #Scotland the lovely lady at security at #edinburghairport checks our documents and then greets us saying "Welcome back home!" 🏴
Thanks for being such a welcoming place, really this is now #home! 💕
- Pronouns
- he/him
Senior lecturer at Edinburgh University and Zhejiang-Edinburgh Joint Institute (ZJE).
Undergraduate Programme Coordinator, Biomedical Informatics at ZJE.
I teach #imageanalysis & #dataanalysis with #RStats & #python.
My research is focused on how #heterogeneous behaviour in #pituitary (and other) cells shapes their function as a population.
I'm also very interested in #reproducibility and #openscience.
Joined Oct 2022
