klysm

19 ms·

1. 181.▲
   by klysm 1y ago

   PWM is not to be confused with Switch Mode Power Supplies in the general case
2. 182.▲
   by klysm 1y ago

   Please don't give anybody any ideas!
3. 183.▲
   by klysm 1y ago

   WP high vs low might depend on which chip is used.
4. 184.▲
   by klysm 1y ago

   To elaborate for the uninitiated, that means they are storing it in plaintext somewhere.
5. 185.▲
   by klysm 1y ago

   Soon this will be contaminated as well unfortunately
6. 186.▲
   by klysm 1y ago

   The important thing about vendor lock in is allowing another company to have pricing leverage over you.
7. 187.▲
   by klysm 1y ago

   Not once have I ever seen this happen
8. 188.▲
   by klysm 1y ago

   In the same boat, also with ADHD.
9. 189.▲
   by klysm 1y ago

   Convert to money as late as possible
10. 190.▲
    by klysm 1y ago

    It’s very much hidden, which I guess is a success of the design. You need not concern yourself with the web of ASNs when using IP.
11. 191.▲
    by klysm 1y ago

    I disagree with this approach. Being very very specific in what you accept and very very specific in what you emit seems better to me.
12. 192.▲
    by klysm 1y ago

    Reward hacking is literally just overfitting with a different name no?
13. 193.▲
    by klysm 1y ago

    Have the compute modules had similar problems?
14. 194.▲
    by klysm 1y ago

    I see MiB in usage a lot and I always make the distinction myself
15. 195.▲
    by klysm 1y ago

    Posturing and FOMO + shoveling money to friends
16. 196.▲
    by klysm 1y ago

    I agree with the sentiment, but formal methods are tricky
17. 197.▲
    by klysm 1y ago

    But it’s interrupted by an ad break
18. 198.▲
    by klysm 1y ago

    I think about branches a lot too when optimizing
19. 199.▲
    by klysm 1y ago

    I like react because it gives you escape hatches like this. You can take a range of approaches, starting with simple and declarative, only moving to gross and performant if required.
20. 200.▲
    by klysm 1y ago

    As the sibling comment points out, react does not bring in dependencies like that. It is not that heavy but people associate it with heavy things
21. 201.▲
    by klysm 1y ago

    I think anybody using this term has a shallow understanding of network security and just bundles it all mentally into a “thing” that stops all the bad stuff from happening.
22. 202.▲
    by klysm 1y ago

    Milligram accuracy isn’t a really a direct property of the load cell. A lot of it comes down to creep and hysteresis behavior.
23. 203.▲
    by klysm 1y ago

    I think they have totally different goals typically. For example, let’s say we are doing a sampling procedure. How do you estimate the sampling error? I’m not aware of a machine learning technique that will help, but you can use Bayesian an
24. 204.▲
    by klysm 1y ago

    You haven't pointed out anything specific to FR4, which is what this would be replacing. This is merely a ploy at getting funding, and I'm very skeptical about it because I've seen 2 or 3 companies do the exact same pitch and
25. 205.▲
    by klysm 1y ago

    Greenwashing - this kind of idea has been floating around for years and I don’t think it’s really that big of a problem
26. 206.▲
    by klysm 1y ago

    Good to understand, but most likely better to stick with react for production stuff anyway
27. 207.▲
    by klysm 1y ago

    I’ve had this in my head as well “constraints yield art”. But it’s also required to engineering
28. 208.▲
    by klysm 1y ago

    I’m highly skeptical that this has been established with any real merit
29. 209.▲
    by klysm 1y ago

    Cool idea. It’s like pumped water storage but backwards
30. 210.▲
    by klysm 1y ago

    MCC also generalizes to multi-class well. I wish it had a better name though. It seems like F1 score has better marketing