• 0 Posts
  • 39 Comments
Joined 6 months ago
cake
Cake day: January 14th, 2024

help-circle
rss







  • i once had a customer who wanted their fish “blackened”. Everyone else on staff had tried at some point and the customer always complained it was undercooked. So me being the new gal said id do it. I threw it in a pan and left it there for god knows how long. Then flipped it, left it there longer, and finally threw it into the back of our oven for another 15 minutes. It was literally charcoal when i was done. I was sure they would complain that it was overcooked but those fuckers loved it. They specifically requested me to cook their meal whenever they came in after that. I was just fucking with them the first time 🤷‍♀️






  • Id set up a static website with Hugo. You can preview and build locally. Or put it on your home network and vpn in if you need remote access to make an entry.

    In your content folder you could do content/[year]/[month]/[day]/index.md, and have a _index.md in the year and in month folders so there would be pages with automatic collection of articles under that year/ month. You could also subdivide the content folder into health/ general/ shower thoughts and other “types” of journals

    They have support for tags, categories, and custom taxonomies. So if you wanted to have “people” category you could, and then a “thing” category or any other sort of way to tag the content.

    https://gohugo.io/



  • Its not any different than how it already was. Initially the GenAI models were all being trained on masses of unlicensed data including data from reddit. The problem is some companies like New York Times are suing for training an LLM off of their data. So in response companies like OpenAI are now trying to reach partnerships that basically license the use of the data (that they already had). This also means that they will be able to continue to have future access to that data as long as the partnership is in place. Whereas some companies without a partnership could start to ban scraping activity or update their terms to forbid training AI off of their data.

    Overall these partnerships are a good thing. Licensed training data is good. But from a privacy standpoint, the AI models were already trained on reddit data. This is just formalizing the relationship





  • Yeah your experience matches mine. Id really encourage people to not blanket trash on all mods. It is a lot of work that goes into moderating communities and it is either done by people who love the community, or by someone who loves that power dynamic. I’m not saying all mods are perfect, but give a chance for individual mods to prove themselves. It’s generally a thankless job, especially lately. By trashing on all mods we’re just going to scare away the good intentioned people and all that is left are the power hungry ones.