Pinned toot

I am Victor Venema. I was born in , The Netherlands, and also studied physics there. My PhD was about measuring cloud structure. I currently work on the of climate station data in , Germany, and on @GrassrootsReview

I am also one of the admins of . We would like to create a friendly place to talk about science, but feel free to talk about anything. Please notify Frank Sonntag @admin or me early of unfriendly behaviour.

My 2012 paper used simulated data to test homogenization methods.

The article by Gubler et al. (2017) used real Swiss station data homogenized with a lot of metadata to study how well homogenization works if you have much less (meta)data.

It is another great example of a @GrassrootsReview because several important findings are not mentioned in the abstract & may need some expertise to notice in the article. These are highlighted in the Grassroots review.

Show thread

For the above I linked to its @GrassrootsReview This review is still incomplete, but it contains a discussion by the authors (moi) of weaknesses we found after publication.

I had already talked about these weaknesses at many conferences & put the info on my blog, but this review page is where it belongs. Were if in future everyone uses the system everyone will look for more information on articles.

Show thread

I work in homogenization of station data, it is a small field, unfortunately. So maybe this better be

Let me start with my own paper. The reason people know me in climatology & my opportunity to switch from working on clouds to observed climate change. It is a big blind validation study of homogenization methods, including many manual method.

Venema et al., 2012: Benchmarking homogenization algorithms for monthly data.

Here to discover Masotondt and to promote a way of post-publication peer-review: you put your manuscript on an Open Archive (zenodo, OSFpreprints, bioRxiv etc.), you ask for our community to review it. We make a classic but fully open peer review (open review, open answers, open decisions) and if the work is good, we "recommend" it. It is therefore a quality article, citable, and if you need it, you can still publish it in a traditional journal. Everything FREE.

I'll be giving a ~25 minute talk on Tuesday the 28th explaining why patents don't work for complex inventions like software or whatever you're reading this on. There will be bonus slides about how Amazon et al use patents to avoid taxes.

7/28 @11pm UTC.
It's part of the incredible program put together by the Masto instance. Sign up for all their talks here:

Paper DOI: 10.1007/s13132-020-00629-1

I post a (or pop science piece leading to papers), some days. not usually the newest press releases but I really like this idea - spherical harmonics of the entire atmosphere :)

student in one area of , forever trying to understand all of it. spectator.
also posts about , I find,

We now have 55 users. Thanks for everyone helping to promote .

If you look at my first Toots you will find some tips and tricks to survive in this new jungle.

. I studied Physics, got my PhD in Physics, and have coauthored papers dealing with the analysis of climate data, statistical and numerical downscaling and analysis of renewable energy resources. I have coauthored open source code for research, such as pyClimate, a tool for the FAST estimation of multidimensional PDFs based on kernel, extending the 1D "area under the curve" model agreement index, aiRthermo in R or Sailor in R (extension to vector data of the Taylor diagram).

@VictorVenema I'm ready to flood Twitter about #fediScience and #science on the #Fediverse!
Counting the seconds ;)

I wrote an announcement of to explain to people who do not know Mastodon yet why it is good for them and society.

Micro-blogging for scientists without nasties and surveillance variable-variability.blogspot.

The links on Twitter and Reddit are up.

Please help create some buzz on the legacy silos. So that we can move some scientists from to past to the future.

Show thread

Micro-blogging for scientists without nasties & surveillance variable-variability.blogspot.

The announcement of FediScience got 500 boosts.

More important would be to spread the news outside of Mastodon. They need more explanation, so I wrote the above blog post to explain why Mastodon is better for us & for society.

Please spread on social networks this Monday at 3pm (Berlin), 2pm (London), 9am (New York). You will find a post here:

And a Tweet here:

Do you already know the Unpaywall browser extension? On the page of a scientific article, it will indicate whether there is a free version. Very often there is and it will link to it.

Elizabeth Warren wants answers on Facebook’s fact-checking loophole

Facebook pretends opinion articles cannot be fact checked to appease Republicans & their donors.

Warren calls Facebook out. The first example was a fact check I did for Climate Feedback. One of the dumb arguments was that global warming had "paused". In 2019 (!) when warming was exploding! We need better climate stooges.

our #IASGE ( survey closes TOMORROW at midnight, pacific standard time :blobastonished:

we are ~100 responses away from 500, which is the number we really strove for!!

if you have 10minutes and write/reuse code in academia, I would be so grateful if you could take the survey:

and forward it around to all your coworkers/friends/enemies-you-must-work-with we would be EVEN MORE grateful :blobheartcat:

I am a computing scientist at the University of Glasgow in Scotland and my specific interests are compilers for acceleration of scientific code, parallel and heterogeneous programming, GPUs and specifically FPGAs.

I have a PhD in social sciences (information and scientific communication). My research is focused on public and university libraries, bibliometrics, science mapping analysis, digital media, and social networks. My option is always FLOSS and I lead a vegan lifestyle. I have also an account on (, but it'll be a pleasure to read and share also on fediscience!

The “Peer Community in” (PCI) is a non-profit scientific organization that aims to create specific communities of researchers reviewing and recommending, for free, unpublished #preprints in their field (i.e. unpublished articles deposited on open online archives like arXiv and bioRxiv). To a lesser extent, they may also recommend articles already published in journals.

#Academia #Research #PeerReview #OpenScience

Show more

Fediscience is the social network for scientists.