These sub keys (hashes) are generated when a secret access key is requested and propagated to services at that time, right? The way it was explained made it sound like there was some batch job generating keys every day
It was quite helpful. Aversion to risk is aversion to change, not necessarily aversion to growth but it can be, specially when it comes to department and company culture, whether it’s lack of cohesion in team work or growth.
I created a lambda with a function URL. is there a straightforward example for making a call to this URL with fetch, or jquery ajax, etc. ? ( or even axios within react )
tldw: you send your `id` along with fancy `hash(key_for_that_id)`. AWS can then look up their stored copy of `key_for_that_id` and calculate hash(key_for_that_id) on their end. As long as the two hashes match, all is well. [Fancy here means they also take region, service name, date, url into account etc]
Given my political leanings I was happy to see Peltola win in Alaska, but my guess is this is actually an example in which RCV failed to find the Condorcet winner (would have beat either of the other two candidates in a head to head), who was likely Begich. Doesn't mean there was anything about the election that was rigged or unfair, but I don't necessarily think its a great example of RCV best representing voter preference.
Thanks for putting this together, it's cool stuff! One factor that wouldn't be relevant in a better world: it's a lot simpler to talk about the validity of an approval vote. Given that our society has reached the point of people trying to discount first-past-the-post votes based on supposed patterns in when they were counted - it makes me pretty doubtful that we'd handle ranked choice runoffs calmly and logically. That said, people don't talk enough about the benefits of those indirect impacts on coalition-building, so this was neat.
Hi Colm, 0x9 and 0xA are duplicated btw (both Idempotency). Thanks so much for taking the time to do these videos. They are very instructive at many levels.
So, why didn't you just remove region information from the hmac hash to make token work across regions? It must have been very expensive to optimize performance for public/private key encryption considering AWS scale
I had this in my bookmarks when it was on twitch and to my horror it was removed, luckily I found it here so I can finally watch it. Why was it removed from Twitch?
At 53:07 or so. Is there a bug in the code? You aren't guaranteed to fill the list if the data_size > sample_size. I think you need to fill all the samples, then start sampling using random numbers. This manifests when the sample size is only a bit smaller than the overall data size. Sometimes the sample list doesn't fill up but if data is sufficiently large the chance a sample won't be filled trends towards 0.
Yes, I found that too. The fix is easy: until the sample set is full, add items as they come in (as explained in words, but the code does not do that part).
Is there a way to suppress sound of twitch when a user is joining or posting comment ? The sound is very distracting and sometimes we miss important keywords that you are saying because of that sound. Thanks
Loved the content so far. The only issue is that the video quality is 480p so when it gets to the coding section, it's really hard to read the screen, even when you make the font bigger. Other than that, great content. I missed breakfast cause I got hooked on the topic.
