আমরা বাংলায় আলাপ আলোচনা চালানোর জন্য এই গ্রুপটিকে ব্যবহার করতে পারি। এই একাউন্টটিকে ফলো করুন, এবং বাংলায় লিখলে এই একাউন্টটিকে @ মেনশন করলেই হবে। যাারা এই একাউন্ট `@bengali_convo@lemmy.ml` কে ফলো করবেন, তারা সবাই এই একাউন্ট থেকে পাঠানো পোস্ট পড়তে পারবেন।
প্রথম পাঠ | বাংলায় বিশেষ গ্রুপ নেই | একটা গাপপি গ্রুপ আছে বটে, তবে এই ধরণের গ্রুপ যেখানে খোলা মনে যা কিছু লেখা যায়, শেয়ার করা যায়, এই ধরণের একটা গ্রুপের প্রয়োজন রয়েছে।
www.medrxiv.org
In total 120 studies in 130 publications were included. Length of follow-up varied from 12 weeks to over 12 months. Few studies had low risk of bias. All complete and subgroup analyses except one had I2 ≥ 90%, with prevalence of persistent symptoms ranging between 0% and 93%. Studies using routine healthcare records tended to report lower prevalence of persistent symptoms/pathology than self-report. However, studies systematically investigating pathology in all participants at follow up tended to report the highest estimates of all three. Studies of hospitalised cases had generally higher estimates than community-based studies. Interpretation: The way in which Long Covid is defined and measured affects prevalence estimation. Given the widespread nature of SARSCoV2 infection globally, the burden of chronic illness is likely to be substantial even using the most conservative estimates.
github.com
Comparative genomics toolkit
www.medrxiv.org
Covid-19 has caused more than 1 million deaths in the US, including at least 1,204 deaths among children and young people (CYP) aged 0-19 years. Deaths among US CYP are rare in general, and so we argue here that the mortality burden of Covid-19 in CYP is best understood in the context of all other causes of CYP death. Using publicly available data from CDC WONDER on NCHS’s 113 Selected Causes of Death, and comparing to mortality in 2019, the immediate pre-pandemic period. Our findings underscore the public health relevance of Covid-19 to CYP. In the likely future context of sustained SARS-CoV-2 circulation, pharmaceutical and non-pharmaceutical interventions will continue to play an important role in limiting transmission of the virus in CYP and mitigating severe disease. #Academic #discussion
An article on reading scholarly articles.
arinbasu
2 years ago
•
100%
Or ... the art of creative data analysis
cross-posted from: https://nrsk.no/post/112817 > I made a "Share to Lemmy" bookmarklet to easily share content to Lemmy, maybe you want one too? If you don't know what it is, it's a bookmark that runs a small piece of code - In this case, it shares the page you're currently on to Lemmy. If your browser have a bookmark bar, you're always just a click away from a pre-filled share screen on your home instance. > > It should work in all desktop browsers. > > - Copy and tweak the code, replace `nrsk.no` with your home instance > > - Create a new bookmark, the tweaked code goes in the URL field > > - Click the bookmark while on a page you want to share > > `javascript:void%20function(){var%20t=document.URL,o=document.title;window.location.href=%22https://nrsk.no/create_post%3Furl=%22+t+%22%26title=%22+o}();`
github.com
This github repo contains lists of servers and users.
An interesting paper where the researchers conducted a cross sectional survey of people aged 18 years and above in Mexico. They divided people into two groups, one group who had toxoplasma infection in the past, and the other who did not, and for each group they measured their "attractiveness" (using other people) and also some parameters of "health" and "well being" and having good looks (such as "width of the face divided by height of the face", so the lower the value, the more attractive they are and so on), and for women measured their BMI. They found that those who had toxoplasma infection had "better looks" and scored high on "attractiveness" measures. If you read the paper, (free to read the full text), can you identify how many errors there are? What are they? Common sense, you do not need to be expert on anything on beauty or biology but if you are, even better. :-)
arinbasu
2 years ago
•
100%
Yes and the state they are writing about is the study of one state within the union of India. While it is not easy to provide entirely publicly funded healthcare, it is possible to strengthen primary care provision. This is a good warning article in some sense.
healtheconomicsreview.biomedcentral.com
An interesting secondary analysis of different data sets. The paper compared how much households and government spends on healthcare in private and publicly funded health services in a state in India. The findings suggest that on an average, the spendings on private healthcare is three times that of public funded healthcare. The paper concluded that, "Using public resources for purchasing inpatient care services from private providers may not be a suitable strategy for such contexts."
Researchers studies 130, 000 records of individuals who were hospitalised or who died in the US. Then they studied the association between omicron strain and the deaths/hospitalisations after adjusting for the effects of vaccination but not treatment. The evidence is intuitive: omicron is as deadly as previous waves. What seems to happen is vaccines blunt the impact. This is a preprint
cross-posted from: https://lemmy.ml/post/63108 > We are all in love with decentralized social topology, aren't we. But to make society reasonably decentralized, we need to remodel more than one level of it. I would like to bounce around some thoughts that may help establish a multilayer model of decentralized society. > > ::: Longwinded > 1. Assumptions. > > 1.1. I use communications as an example of social activity that is a key to all other processes. > > 1.2. I assume that a decentralized network of heterogeneous communities is a good model for human society that we need now, as the all-crisis unfolds and neither democracy nor (even less) capitalism can offer any constructive approach. > > 1.3. I assume that the **minimal** provisions for an individual must include the right to participate in more than one community at the same time, the right to opt-out peacefully at any moment and the right to form a community (and participate in the network) on an equal basis. > > 2. Layers > > 2.1. "Fediverse". > What we now see as fediverse is an implementation of communications pattern, where instances of various services can be associated with specific communities and their local users considered community members. Federation protocol provides a routine way to regulate interactions with other communities. > > 2.2. "Community Intranet". > To control their collective memory, their policies/rituals and their boundaries, communities need to have control over the physical infrastructure of their "village intranet". It applies mostly to "natural" (local) communities, while "virtual" ones may need a trusted and neutral virtual hosting environment. The control should not, however, influence individual participation in remote communities. > > 2.3. "NetCommons". > To keep the information flowing, society needs a non-owned, collectively managed transmission backbone. We can draw analogies with watershed management that is a known example of advantages and shortages of the commons approach. > > 2.4. "Platform Cooperatives". > Economic (and, effectively, political) control of the means of production is a key to stability of the ecosystem. Thus, decentralized ecosystem of user cooperatives provides cohesion to the whole multilayered model. Every user becomes a member of the co-op(s) operating their community(ies) infrastructure. Community co-ops then form the "NetCoooperative", managing and maintaining the backbone systems and providing support, R&D and exception handling to communities in need. > > 3. Essential question is, whether such a model is comprehensive and complete enough to provide scaffolding for an attempt to implement proof-of-concept project. > > Questions and comments welcome. > > ::: > This is a repost.
plos.org
A tutorial on writing a peer review
www.overshootday.org
A marker of sustainability.
A great summary on decentralised social networks, including scuttlebutt protocol and Aether. Worth a read.
A good critique of Twitter and future direction of social networking apps.
arinbasu
2 years ago
•
100%
The only mastodon instance that I can federate with so far is qoto.org ; all other instances, if I search for, brings up my username i.e., arinbasu@lemmy.ml and allows me to follow. But then, they state, "request sent". In lemmy itself, I do not see anywhere I can approve such requests. I think friendica works, but have not had enough experience, and it is not a mastodon instance. I also tried to use pleroma, but pleroma too has the same issue, i.e., request sent and that's it.
stackoverflow.blog
An essay on why you should read computer science essays to understand the history of concepts, and figure out how to solve a tricky problem from the first principles. This essay also highlights the [paperswelove](https://paperswelove.org/) project.
Getting my head around
Interesting article
arinbasu
2 years ago
•
100%
Yes, I agree with you here, "I’m not sure how it can work with one big community for all fields"; I am not sure either. Let's say the poster who posts the link will take the trouble to make it easy for other people to understand the complexities in the paper and talk about it. Perhaps these best serve for individual fields.
www.curvenote.com
I find curvenote as a great tool for research writing. It combines Jupyter notebook with a markdown rich text editor based content authoring. Bibtex (reference management), figures, tables everything flows easily.
www.qeios.com
I created this community with a view to discuss academic and non-academic research papers. If you are a student, academic, researcher, we can bring in our papers and we can build a community around it. I have added the link of a website where we can add reviews of the papers we may want to review and also link the papers here. What do you think?
Mastodon