• 0 Posts
  • 9 Comments
Joined 1 year ago
cake
Cake day: June 29th, 2023

help-circle

  • I’m not sure I understand why the voters decided this was the right way for their revolution, but it just seems like a weird approach to destroying your government. You still have tons of nukes on your soil. You think those can only go off outside your borders? You think you get a share when the launch codes are sold to the highest bidder? Or are you hoping Musk will be in charge of that, because he’s a rocket scientist?

    I hope they get what they want though.







  • I don’t necessarily disagree, but I have spent considerable time on this subject and can see merit in decoupling your own error signaling from the HTTP layer.

    No matter how you design your API, if you’re passing through additional layers, like load balancers and CDNs, you no longer have full control over all responses your clients receive. At this point it may be viable to always signal a successful backend connection with a 200, even if the process resulted in a failure.

    Going further, your API may include partial success scenarios, think batch processing, then the result could be a mix of success and failure that doesn’t translate to HTTP status.

    You could even argue that there is really no reason to couple your API so tightly with a concept of the transport layer it uses.


  • I feel like most people base their decision on license purely on anecdotes of a handful of cases where the outcome was not how they would have wanted it. Yet, most people will never be in that spot, because they don’t have anything that anyone would want to consume.

    If I had produced something of value I want to protect, I wouldn’t make it open in the first place. Every piece of your code will be used to feed LLMs, regardless of your license.

    It is perfectly fine to slap MIT on your JavaScript widget and let some junior in some shop use it to get their project done. Makes people’s life easier, and you don’t want to sue anyone anyway in case of license violations.

    If you’re building a kernel module for a TCP reimplementation which dramatically outperforms the current implementation, yeah, probably a different story