LuisMondragon

There is something worse than appeal to authority: appeal to non-authority.

Activity from Amazon's real-time digital advertising auction may be responsible for much of the traffic overwhelming the firewall, according to internal Slack messages seen by Insider.

Even inside AWS, however, information on the outage remains sketchy. One AWS employee speculated that the outage was caused by an "orchestrated DNS attack," while another employee downplayed those concerns, saying it was more of an "internal thing" related to networking and firewall saturation.

"It's the fog of war," said an AWS manager.

In the early days, the distinct feature of the web was bidirectionality. Unlike TV or radio, on the web the communication went both ways. The user could finally give their opinion and comment. We have gone full circle, with YouTube hiding dislikes, and even with some channels disabling comments, we have unidirectionality again.

I use the empty dark new page extension for chrome/brave. The empty new tab was too bright for my dark environment.

To attenuate my hand pain issues, I am using a mix of a tenkeyless 60% keyboard, AutoHotkey, espanso and dictation. Also, instead of focusing on using key combinations and typing fast, I am trying to practice 'typing mindfulness'; that is, really to be present and type consciously, accurately and slowly, instead of fast and erratically.

Imagine if this were the case in other field. A "no words" movement for people that want to publish a book but don't want to bother with the writing part.

I translate games and use Android emulators on Windows to test games. I think I could use the Android subsystem, I would give it a try.

AFAIK the chat backup is optional. I've never stored a backup.

I agree. It's a typical error of judgement from people that know things superficially. "All electronic music sounds the same", "all FIFA games are the same", "all soccer matches are the same". We simplify things that we know little about or things where we have poor proficiency. But experts can see the slightest differences and nuances.

My issue with this kind of reasoning is the comparison and reference to the human brain. The potential and reach of AI transcends the brain. We never had to master the "mystery" of how birds fly to invent aviation. It was never necessary to compare the number of turbine revolutions of early airplanes to the number of an eagle's feathers. Maybe birds were an inspiration or a metaphor, but thankfully aviation has not been limited to the means of propulsion of the beautiful yet humble pigeon. The potential of aviation has taken us into space exploration and massive international travel. I don't know where AI will take us, but I don't think it will be constrained by this temporary organ called 'human brain'.

New AI fallacy: appeal to size

Thank you! This is a very comprehensive tutorial, I'm sending it to my Kindle to study the rest of the day.

I only wished the headline was a little bit less generic. There are many tutorials about clustering with scikit-learn, but this one is specifically dealing with historical and religious data, which is interesting and unique.

Also, I really like the concept of "programmer historian". I can understand this perspective, since I'm defining myself as a "programmer translator/linguist" and I know that these kind of methods can be extremely useful for humanities professionals.

I'm in South America and see the same message

Is there an objective benchmark to measure the performance of voice assistants? Usually, NLP tasks have ways to measure their performance.

Facebook employees unable to enter buildings this morning to begin to evaluate extent of outage because their badges weren’t working to access doors.

https://twitter.com/sheeraf/status/1445099150316503057

Some Oculus Quest owners can't use their device https://www.reddit.com/r/OculusQuest/comments/q18xwy/faceboo...

> Facebook has realized that if they change the algorithm to be safer, people will spend less time on the site, they'll click on less ads, they'll make less money.

> But Facebook’s research tells a clear story, and it’s not a happy one. Its younger users are flocking to Snapchat and TikTok, and its older users are posting anti-vaccine memes and arguing about politics.

I'm an editor and proofread texts daily. (In Spanish). The spell check process in Word was always straightforward, without significant delays between each check.

But guess what. Now the humble spell checker has evolved and now it's AI and cloud based. Nothing wrong with those technologies if properly implemented, but now each word that gets flagged makes a web request to try and find a suggestion, most of which are useless.

It's so frustrating. My work was becoming noticeable slower. Since I get paid per word, I need to work fast, so now before spell checking I have to disconnect from the internet. And I paid for Office! I paid for a software that helps me work slower and earn less.

I watched Joe Rogan's interview with defector Yeonmi Park and, if I understood correctly, she said that this is the case in North Korea only, because the government banned the word "I"!

You can't say "I like this", only "we like this".

In South Korea people do use the first person singular. I hope a Korean person could correct me if I'm wrong.

I remember a discussion on Twitter about abolishing possesive pronouns, because they allegedly express a relation of power and, well, possession. Like, you shouldn't say "my wife" because she doesn't belong to you.

The thing is this: for a lot of things we have experts, and we listen to them. We know we should listen to medical doctors instead of improvising uneducated opinions.

In the case of language, there are also experts: linguists. They can educate us about the various uses of possesive pronouns, which express more than just a naive idea of possession.

But somehow in this kind of heated discussions nobody consults linguists for guidance.

They turned something beautiful and poetic into a Black Mirror prompt.

Not the same. CLIP is trained with pairs of images and texts, whereas VideoCLIP uses pairs of videos and texts.

> There needs to be a new fallacy or something: "Appeal to AI."

I applaud and welcome this idea

This tutorial was excellent. I got everything working in a couple of hours. On CPU alone the similarity search is mindbogglingly fast. Thank you!

I run a script to autotag my Diigo bookmarks. (Using their API).

Wow, there was a thread today about moderating user generated content, a HN comment told how users try to get away using symbols from other languages that look like English to post offensive content. I was wondering if one could make an ML system that takes into account the appearance of words. And now I find your post, which is doing something like that, albeit with regexes. Interesting!

> Found out, they have learned the trick and now using symbols from different languages to make it look like English.

I wonder if you can train a ML model using text as images. For example, taken as strings, "porn" and "p0rn" are not very similar, but visually they are.

How do you handle typos and spelling mistakes? I wrongly wrote "steve irwin cocrodile" and got zero results. Then I corrected to "steve irwin crocodile" and got many results.

Interesting NLP tech:

> Amazon claims it’s the first company to offer this kind of privacy-first technology on smart speakers. Last year, Amazon unveiled the AZ1 Neural Edge processor, which powers current Amazon Echo devices, but its new Echo Show 15 will be powered by the AZ2 processor. Amazon said that this processor can do 22 times more TOPS (trillions of operations per second) than the previous generation.

> The company built this module alongside MediaTek, and it will allow for on-device neural speech recognition for new products.

Aaand, as I expected, it is for English only:

> Amazon says these latency benefits will be available first for English in the US, but it plans to expand to more languages “in time.”