News from sleeper_must_awaken

  1. To give some context, on Azure for an E32 spot node we were at some point paying 0.20$ per hour to azure for the VM and 1.2$ per hour to Databricks in DBUs. So basically 600% increase to the price of the VM to run it on databricks.

  2. I have done an extensive cost analysis of Databricks on AWS. The calculations I did showed that DBU cost is more or less equal to the price of an on-demand VM.


  4. Most studies I've seen say omicron is less likely to cause long covid than delta at least.

  5. Something like snowflake is super hands off and just uses the XS warehouse with PowerBI pulling once per day.

  6. For that data scale, just use postgresql or similar. Don’t worry, refactor later.

  7. Any virus. Post-viral diseases are more common than people believe and there's nowhere near enough research about them. Long COVID is just the flavor of the month and getting the attention it is because of the sheer volume of people who have had it.

  8. If that would be true, we’d have 10% of the population already suffering from post-viral disease before 2020. That seems unrealistic.

  9. Before COVID, millions of people were already suffering from post viral diseases.

  10. Well, we see people getting less sick after vaccination. This is why we keep doing it.

  11. As a veteran Scala developer, I would argue against investing heavily in Scala. The ecosystem is fragmented, many repo's are abandoned and the JVM is surpassed by alternatives, such as Rust and Go.

  12. Most of the hard problems lend themselves to being solved by very expressive languages in small teams. With which tends to come the money. The people who use Scala/Rust/Haskell/C++ et al aren't (normally) doing it for fun it's because the alternative is throwing huge numbers of developers/expertise at the problem to basically mould the language to your requirements (e.g Instagram's frankenPython) which isn't viable without FAANG money.

  13. Yes, you make a very good point here. We aren't doing it for the joy of FP (sometimes, I admit), but mostly to fix developer inefficiencies.

  14. Besides the study using patients who were infected with acute COVID-19 illnesses, there is not scientific test to diagnose “long covid” right?

  15. There is no official test. However, there are some tests that show microclots. Also, active SARS-CoV2 has been found in patients with symptoms.

  16. Is that the same mechanism that caused the rare blood clotting issue with some vaccines, or is it just a coincidence?

  17. There is no-one officially linking the two together. However, we do know that in rare cases the vaccin can end up in the blood stream (instead of in the muscle tissue). When that happens, it may cause clotting due to the spike protein according to some experts.

  18. For the same reason developers wrote JSON configuration tools 5 years ago and XML configuration tools 10 years ago. Although these tools are born in best intentions, they often have a deleterious effect. Decision making around abstraction is difficult and often the results are leaky and difficult to understand.

  19. Ok, apart from technical reasons, it is important to realize the ecosystem of support and development each of these environments provide. Both system provide distributed, scalable, batched data processing functionality as its prime selling point.

  20. Covid doomers and anti-vaxxers are becoming one and the same...."Just wait 6 months".

  21. The difference being that the doomers have been right. We see consistent excess mortality in the last 6 months. We see an unprecedented level of sick leave (at least in the Netherlands). Both can be explained by a deregulated immune system.

  22. So I mean, just how much more 'transmissible' can this thing get? Is there a not a ceiling of sorts? My immediate family caught it over the last week +. Have no idea if it is/was XBB, but it sucked regardless.

  23. There is a ceiling as there is a limit to human connectivity. You don’t meet (on average) more than, say, 500 people while being infectious. That would cap the R to 500.

  24. It's inversely correlated with mortality in practice.

  25. What makes you believe this? AFAIK, there is no reason for more transmissible viruses to be less lethal.

  26. Not exactly underrated per se, but not very well known - Eva Cassidy. My all-time, all-genre favorite vocalist, if I had to choose just one.

  27. Slide 1: don't mix green and red if you want to stay friends with the 2% colourblind people.

  28. It can be true. He might have been considered the greatest baseball player in New York by himself. Something I find very believable.

  29. Russia is not a democratic system in which there is no plural and free media and no real opposition, so no one can detect lies and deception..

  30. White does not only reflect heat from the sun, it also reflects your body heat back into your body.

  31. If you want to take a deep dive, this book covers everything around nutrition in endurance sports:

  32. I'm one of the children that had to grow up under the effects of the toeslagen affaire for 15 years. I think I have reasons enough.....

  33. And again everyone is indifferent. On to the next scandal, while the previous ones are forgotten.

  34. They are still solving this one, but it's all very hush hush 🤫.

  35. That's indeed very sad. Moreover because the Efteling makes most of its income from upselling during your visit at the park (coffee, toys and other kinds of food).

  36. For running I use goodr’s which is excellent. Perhaps it works well with cycling too, but haven’t tried.


Leave a Reply

Your email address will not be published. Required fields are marked *

You may have missed