oleg_log
1.72K subscribers
1.88K photos
131 videos
9 files
2.82K links
Shelter for antisocial programmers "Oleg"

halp: @olegkovalov
web: https://olegk.dev
fov: @oleg_fov
chat: @oleg_log_blabla
podcast: @generictalks
Download Telegram
Нашел в закладках, источника не будет, искать лень.

What fruit flies could teach scientists about brain imaging. As programmers we are used to working with dumb stuff. Dumb data is stored and manipulated by smart stuff residing elsewhere. That’s not how life does it. Cells are the building blocks of biology and cells are smart. This experiment directly measures how individual neurons use energy inside intact brains using new sensors that measure ATP levels. When you increase the activity level of neurons wouldn’t you expect ATP levels to go down? Work requires energy. That’s not what happens. The cell is smarter than that. The cell increases energy production. It’s guessing how much energy it will need in the future, so it increases its energy production to meet expected future energy needs. Brains are not computers, they are implemented by cells. We’ve taken dumb stuff programming as far as it will go. We need to program with smarter stuff.
Forwarded from moki.codes
Two Hidden Instructions Discovered in Intel CPUs Enable Microcode Modification

lamoa

https://www.infoq.com/news/2021/04/intel-hidden-instructions/

demo: https://twitter.com/h0t_max/status/1375871849079406597
oleg_log
топ вин https://vc.ru/services/256051-internet-sloman-sboy-infrastruktury-fastly-vyzval-pereboi-u-twitter-twitch-reddit-amazon-paypal-i-drugih-servisov
Порофлите еще, Terraform Cloud тоже кашляет из-за CDN https://status.hashicorp.com/

Угадайте у кого пайплайны ушли на перекур? (не у меня)
Неужели свершилось, неужели теперь не будет ломаться совместимость между версиями https://www.hashicorp.com/blog/announcing-hashicorp-terraform-1-0-general-availability
Wu Dao 2.0 is multimodal. It can learn from text and images and tackle tasks that include both types of data (something GPT-3 can’t do).

Wu Dao 2.0 was trained on 4.9TB of high-quality text and image data, which makes GPT-3's training dataset (570GB) pale in comparison. Yet, it’s worth noting OpenAI researchers curated 45TB of data to extract clean those 570GB.

1.2TB Chinese text data in Wu Dao Corpora.
2.5TB Chinese graphic data.
1.2TB English text data in the Pile dataset.

https://towardsdatascience.com/gpt-3-scared-you-meet-wu-dao-2-0-a-monster-of-1-75-trillion-parameters-832cd83db484

Hua Zhibing, Wu Dao 2.0’s child, is the first Chinese virtual student. She can learn continuously, compose poetry, draw pictures, and will learn to code in the future.
хех
oleg_log
Порофлите еще, Terraform Cloud тоже кашляет из-за CDN https://status.hashicorp.com/ Угадайте у кого пайплайны ушли на перекур? (не у меня)
https://www.fastly.com/blog/summary-of-june-8-outage

On May 12, we began a software deployment that introduced a bug that could be triggered by a specific customer configuration under specific circumstances.

Early June 8, a customer pushed a valid configuration change that included the specific circumstances that triggered the bug, which caused 85% of our network to return errors.
Кто из вас ведет блоги? я не про телеграм каналы, именно полноценные блоги (ахахха 😭 ), Medium, LJ, Blogspot, WP, а может и на своем домене.

Давайте, хвастайтесь в коментах, почитаем и вас. Даже если не айти.
Чуть подсмотрел в чате по Fuchsia (https://t.me/fuchsia_dev)

Твиттер одного из разрабов, я так понял, что он начал вялую и неофициальную серию твитов, так что же там в Фуксии?

https://twitter.com/adambarth
Твои оправдания?

I currently have 10 fully remote engineering jobs. The bar is so low, oversight is non-existent, and everyone is so forgiving for under performance I can coast about 4-8 weeks before a given job fires me. Currently on a $1.5M run-rate for comp this year. And the interviewing process is so much faster today, companies are desperate, it takes me 2-3hrs of total effort to land a new job with thousands to chose from.

Yes writing code, I provide enough actual progress that people must assume I'm either extremely methodical or woefully incompetent. I put in about an hour per job per day. I'm logged-in on slack/teams/etc and disable idle indicators. I use a virtual KVM to switch between laptops as needed. The biggest problem I run into is video meetings occurring at the same time, those I usually just disable video and blame it on internet problems.
Many of these positions are mid-to-senior level in "must have" positions like Data Science or Data Engineering where measuring progress can be difficult for even full-time contributors. I target overly funded growth-mode companies where they're focused on adding unnecessary headcount to work on poorly defined projects.

https://news.ycombinator.com/item?id=27454589
Что-то я проиграл, offensive programming так сказать https://twitter.com/jimmyzelinskie/status/1403079612331089929
Люблю такие багульки.

Github на странице репо показывает в топе контрибьюторов 2 аккаунта ботов (1 обычный акк и 1 прилож), но вот на странице контрибьюторов их уже нет)))

Так и вижу, как разные сервисы используют разное апи для одного и того же действия.

Напомнило эту штуку, как по 1му кадру можно было угнать приватный видос с Youtube, гениально и просто https://bugs.xdavidhu.me/google/2021/01/11/stealing-your-private-videos-one-frame-at-a-time/
Ого, оказывается сам Python можно спонсорить. Ну..оно очевидно, некоммерческий проект или как там оно, но почему-то со страницы Github меньше всего ожидал.

Вы кого-то спонсорите? Кроме вебкам стримеров.

https://github.com/sponsors/python
Вспоминаем Спольски и про переписывание софта) хотя конечно интересно
Forwarded from Sysadmin Tools 🇺🇦
Vagrant 3.0 находится в alpha стадии, и он переписан на go

#vagrant #hashicorp