But also -- the main thing that might be different is the screenshot algorithm. I'm over on the reMarkable discord; if you want to take up a bit of Rust and give it a go then I'd be happy to (slowly/async) help!
Initially most of the Rust was written by copilot or Sourcegraph's Cody; then I learn more and more rust as I disagree with the code-helper's taste and organization. Though I have a solid foundation in other programming languages which accelerates the process ... it's still a weird way to learn a language that I'm getting used to and kinda like.
That said, I based the memory capture on https://github.com/cloudsftp/reSnap/tree/latest which is a shell script that slurps out of process space device files. If you can find something like that which works on the rPP then I can blindly slap it in there and we can see what happens!
I don't think that is an in-depth on methodology, they seem to be talking about how the WHO does things. And that doesn't seem to translate the graphs.
But regardless, the bigger point is that the default position isn't that Stats Norway data is automatically comparable with everyone else's data. The world is large and complicated; it is quite easy for small details between systems to do surprising things.
I got a great laugh out of that, they've done an impressive job anglicising their website. But it doesn't really change the fundamental point. It doesn't take long to get to "Most of the content here is only available in Norwegian" [0]. And the articles on the Norwegen version of the site seem to be different to the English.
It can take a surprising amount of research sifting through who-knows-what to figure things out. One fun introductory challenge I recommend is figuring out what the components of the inflation index actually are; it usually takes a few rounds of sleuthing unless you have a muscle memory of where the right manual is. It is hard enough in the same language and with a familiar government. It isn't easy to do in a foreign language and unfamiliar government.
If your're most interested in blog posts google translate is great for exotic languages.
But for the data they're all there in English [0].
And if you're after methodology, analysis or understanding medical data, they follow WHO standards and publications are all in English on pubmed.gov [1] for the explicit purpose of international collaboration (which is the norm in medicine and public health for most developed nations).
I applaud the enthusiasm but I'm not that interested in Norway's medical system. I'm making a point about the larger issue of using foreign data. I spend a lot of time arguing with people on the internet for fun and education; and it is extremely common to get a cheerful comment which - after a few hours of investigation - appears to be an incorrect interpretation of data.
It is hard enough to do for systems that are part of the English speaking world or big, easy to track metrics. It is substantially harder to do for fiddly data series from foreign systems where the primary source material is in a different language.
> And if you're after methodology, analysis or understanding medical data, they follow WHO standards and publications are all in English on pubmed.gov
This goes to the main point - if it turns out that they don't follow WHO standards in an area or there is critical data not on pubmed.gov, what is the expected path for finding that out?
Because in English I have a much better chance of being able to figure that out. The countries are familiar and there is a better chance that the criticisms of the major institutions are well known. In a Norwegian context that already rather challenging task is even harder.
EDIT
An example occurs to me a few minutes later; there was an interesting theory that Japan had a lot of old people because there were unusually strong pension & tax incentives to lie about elderly relatives being alive when they were in fact dead.
The Japanese stats office could be following WHO standards and publishing all their information on pubmed.gov and the series would still be incomparable with other countries if there is an unusual incentive for the stats to deceive coming form an unexpected angle.
Keeping on top of that sort of thing in foreign legal systems is simply hard.
For the point of arguing with strangers, yes, I agree that neither PubMed nor any other entities will provide you with what you need. I don't think that it is possible to acquire an understanding of an issue without some domain knowledge, at least on how to get the data.
But to gain a deeper understanding of the flaws of any country's health (or any) system, there is no way around that except by comparing it with data from other countries. And that might be hard, which is why professionals spend a lot of time on it.
I trained as a doctor, then family practice for a few years, but since the middle of my studies, I realised my main passion was the more technical aspects. I did a bachelor's in electronics engineering while working 50% as a doctor the first two years, and with the last year dedicated only to studying. I realised a bachelor's would not be enough to get the engineering jobs I desired, so I went back to medicine and started training as a radiologist. There they agreed to fund me doing PhD research 50% of my time, and I am now doing a PhD using AI diagnosing dementia from MRI scans--basically my dream job, while working 50% as a radiologist which also is fun and intellectually very rewarding.
Wow, that truly sounds like a dream job!
I discovered neural networks in 2017 and managed to write a simple app to scan dermatological images to classify them, amazing technology!
I have worked for one year at a dementia diagnostic centre as a part of my training to become a specialist in family medicine and there is definitely a lot of interesting work to be done. I have also have an interest in radiology, how are you finding the profession so far?
I like it a lot, as I mentioned intellectually rewarding and I feel like most of what I learned at med school is useful. I also quite enjoy the occasional increase in adrenaline when at trauma reception or stroke CT evaluation. Also nice to both be able to work in peace and collaborate with clinicians. And personally I also enjoy keeping up to date on research which is evolving quite rapidly. The negative aspect is mostly what is common to most medical specialities and due to understaffing.
One of the biggest scandals in Norwegian health care at the moment is a botched transition to Epic in one of the biggest university hospitals. Doctor dissatisfaction has gone to the roof at the point where 50% of the doctors are considering quitting.
Maybe they are different versions, or I'm an outlier, but I love epic - I used a relatively new build of the software last year for a new months, just the ability to message other staff in the context of a patient's chart saved huge amounts of time and interruptions. They also had this super low friction way of addending to the chart, so you could, say, review a (paper) ECG and start recording your interpretation straight into the correct chart within a second or two, and be done in around 10 seconds (with dicatation software which is usually from a different company but integrated)
For my part time bachelor's in electronics we made an autonomous drone that could fly a preplanned pattern and recognize people through machine learning. Software: Ardupilot and ROS. Hardware:home made 3d printed drone,PX4, Nvidia Jetson, some Intel camera). Very fun and absolutely something you could tinker with as an hobbyist (it was basically a thesis on a hobbyist project).
We first tried with a raspberry pi based autopilot, it crashed and we rebuilt it again and again until we realized the GPS was broken.
It's just a heads up that it will not be on their timeline to fix. My empathy would be with the developers and maintainers of this amazing project (and all open source maintainers in a similar situation).
I'm in about the same situation as OP. We have a small cluster of Power9 and it's been unmaintained and unused for a while so I will set it up from scratch. Been looking into solutions that would be a good fit, for the moment we are just a few students/postdoc, so manual scheduling is feasible, but eventually we would like to make it available to other students at the institution.
My candidates are also
- slurm + ray/lightning/etc.
- determined.ai (maybe together with slurm)
Some advertise a kubernetes setup with kubeflow but I would imagine that is a bit too complex for a small cluster.
Anyone else with experience in this? Any other suggestions?
To make the environments as reproducible as possible it would be great to also have a setup based on docker containers and maybe nix, but not sure if it is feasible on ppc64. Guix and Spack have also come up in my searches.
