O__________O

Agree, the karma below -4 counting, but not being shown would be an issue to calculate the average karma per user — but turns out that it appears submissions only add karma to users total karma, so assuming that it is true submission upvote karma is worth 0.5 regardless of context, then to get the average karma per submission simply sum the karma of all submissions, multiple by 0.5, then divide by count of submissions including the flagged and/or killed submissions; to see those, go to your profile, select show dead. To get the average karma per comment, just take the sum of the submission karma above, multiply it by 0.5, then take that amount and divide it by the count of comments by user including the flagged comments.

If I have missed something, let me know.

Possible missing something, but custom GPT + index will not solve the problem of the users knowing what to prompt to use to access the information. Is there something I am missing?

Account is required to play, but getting error when attempting to create an account:

>> “Sorry, something went wrong during registration. Check your input or try again later.”

Given error is not specific enough to figure out what went wrong, I gave up.

Thanks, meaning the public karma of post is only upvotes and downvotes via flags (or any other negative ranking factor) are not applied to this count, right?

My understanding is majority of “Reinforcement Learning from Human Feedback (RLHF)” for OpenAI comes from contractors:

- https://time.com/6247678/openai-chatgpt-kenya-workers/

So the submission score is the sum of the “upvotes minus downvotes” — and not just the sum of upvotes?

Author mixes two topics, that being searching content they control and content they don’t control.

While understand desire to search content created by yourself, in my opinion vast majority of valuable content is created by others, in part because the most valuable content created by yourself is rapidly internalized into your brain.

Meme of Google is getting worse fails to acknowledge that Google is free and makes money from advertisers. More to the point, more valuable the related search is per amount spent on advertising, the more the noise and as result, more resources user will need to spend to enhance the signal. This has nothing to do with Google and everything to do with value of the information itself.

Thanks, I see now and agree; per user, per comment, each user is able to see the the sum of current “upvotes minus downvotes”; to get the average karma per comment for themselves, simply sum the karma of all their prior comments, then divide by the count of prior comments.

Am I missing something?

How even for your own account would a user get their average karma per comment if the true value of karma per submission is hidden? Meaning my assumption is the public HN submission karma is the sum of the upvotes alone, not “upvotes minus downvotes” which I believe is the true karma users receive from a submission.

Aware of minimaxir‘s list — was there anything specific to that thread or his list that relates to topic of formula for calculating average karma per comment?

EDIT: Searched for “karma” on the linked HN thread and minimaxir GitHub page a did not find anything that would answer the question. Am I missing something?

Question is unrelated a user’s total HN karma — it’s about formula to get average karma per comment, per the title.

Waste of time, website returns error when trying to generate image.

In this context, what is “backend” and what does “replying as backend” mean?

Example of prior HN post on topic:

- https://news.ycombinator.com/item?id=33727981

Title was updated, but for context, prior to editing by HN mods, HN’s title for the post read, “Closing your MacBook hardware disconnects microphone, safety/privacy feature” — was confusing edit made by OP; current title now mirrors title on Apple.

My understanding is it is a depth map, which is a 2D layer, not a fully 3D map; think of it as a 3D map from a single point of view rendered in 2D using a predefined gradient encoding.

Understand your point, though doesn’t match my experiences turning a iPhone off fully charged before sleeping (including configuration you mentioned) — and then leaving it off for the night only to find the battery drained significantly in the morning upon turning it on. Beyond that, unlike the OP article that’s subject of this thread, not aware of any Apple support article that clearly states off is off if XYZ is done. Obvious solution would be a switch that physically disconnects the battery, drains capacitors, etc. — though it would likely require redesign of how the system clock pulls power, though might be wrong:

- https://apple.stackexchange.com/questions/87580/how-does-an-...

Since the terms of the deal are not public, anyone have any idea how to estimate value of the deal. For example: how much Init ML raised prior to its acquisition and common comparable multiples; actually monthly active users and comparable price per user; etc.

Anyone have an explanation of how Apple actually decides what merits security and what does not?

(For example, it’s my understanding that turning turn off a iPhone, it’s bluetooth, etc — does not actually completely turn them off. Also appears hardware/OS specs vary from jurisdiction to jurisdiction; for example, it is my understanding China limits a number of iPhone’s hardware/OS specs for domestic iPhones.)

Link to related tweet Emad retweeted:

https://twitter.com/sudu_cb/status/1632674745031483392

Chicken of the woods being Laetiporus, an edible mushrooms:

- https://wikipedia.org/wiki/Laetiporus

Additional edible mushrooms are here; warning: many edibles have non-edible lookalike “twins” - and whole books are written on the topic on identifying mushrooms:

- https://wikipedia.org/wiki/Edible_mushroom

Yes, their names are listed here:

https://arxiv.org/abs/2303.01818

Someone might want to ping the authors to let them know paper is on HN’s home page:

https://arxiv.org/abs/2303.01818

As reminder, from 2011 to 2014, Twitter also 10x their headcount:

https://www.statista.com/statistics/272140/employees-of-twit...

Source?

At the very least per tweet it’s possible to see really view patterns, which to my knowledge what’s possible before.

Let’s not pretend like Twitter has never had downtime, they literally popularized the “Fail Whale” meme way before Elon was around:

- https://www.theatlantic.com/technology/archive/2015/01/the-s...

Also weird given that Twitter is using 3rd-party service which apparently doesn’t even automatically log and post known downtime issues; service being:

https://status.io/

(Assuming this is more of a the site’s not down unless the CTO says it is services.)

Error randomly shows up if you’re not logged in; meaning not logged in I have seen error and error free homepage too.

Worth noting Twitter’s status page claims it is “operational” and no events are listed under history section for past 180-days, which is obviously wrong:

- https://status.twitter.com/

* Not to mention the SSL certificate appears to be expired too.

———-

EDIT-01: Appears there’s a status page just for the API:

- https://api.twitterstat.us/

EDIT-02: Wow, Twitter is using Twitter to post status updates:

- https://twitter.com/TwitterSupport/status/163279294226274713...