#chatgpt For all the talk about how Assistant plagiarizes texts, it wasn't trained on full texts of copyrighted books, which is painfully obvious. It was trained on amateur, fanfic writing. It knows the reviews of good authors, but rarely can summarize the full plot or imitate a famous author's style.
Home
Public posts preserved on my own site.
Is there an equivalent to editors markup (the way that editors bleed red ink on a writers draft) that works in Markdown? (Or ASCII/plaintext)?
I'm trying to write a splog with #ChatGpt and I need a red pen to circle things and say, "Tone it down", "Too cheesy", "No one talks like that"
#Copilot just wants to play games, he's read the source code & manual & thinks it sounds fun.
```
## How are you feeling today?
```dotenv
$MOOD=good
# what kind of code do you want to write today?
$CODE_TYPE=python
# Do you prefer games or "useful" apps?
$CODE_TYPE=games
# Favorite game?
$CODE_TYPE=minecraft
# You at best have read the code and manual for minecraft, right? So have you really played it?
$RESPONSE=No
# Well, if you could play it, what would you do?
$RESPOSE=Build a house
```
```
@AlSweigart fruh - stray - shun
`Error: invalid character '\x00' looking for beginning of value`
How can you not see, the eyes are right there `x00`
Okay, let's solve homelessness.
```
RUN adduser you --disabled-password --gecos "" && \
echo "you ALL=(ALL) NOPASSWD:ALL" >> /etc/sudoers.d/you && \
chmod 0440 /etc/sudoers.d/you
RUN chmod 777 /home/you
```
You're welcome.
“As the adage goes, ‘you ship your org chart,’”...I'm stealing that...
source: https://www.platformer.news/p/elon-musk-fires-a-top-twitter-engineer
Adam Smith: It is not from the benevolence of the ... baker that we expect our dinner, but from their regard to their own self-interest
US Companies: Let's put sociopaths at the head of (nearly) every company and stress test this proposition.
(probably non US companies, too, but I don't read enough foreign news)
@neurobashing That's how I make my personal life faster, too, I never garbage collect.
To the people switching to Mastodon as their new main time sink, I have this to say....
ONE OF US! ONE OF US! ONE OF US!
Has anything replaced Hypertext Application Language (HAL) or is it still the best thing for creating self-navigable APIs?
https://en.wikipedia.org/wiki/Hypertext_Application_Language
Retweets still limited?
Go ahead machine, I dare you to limit me!
Still? Up?
Now I have to tweet every few minutes to see if it still up.
Wait, is it back?
I can see who has tweets scheduled.
Do you like the new limit of 0 per day?
I'm over my daily limit at my work desk, so I'll see you in 5 minutes from a different laptop on the other side of the room.
Oh the schedule puts it on MONDAY
You can now keep up with the global feed on #twitter.
#twitter is scaling their load by not letting anyone tweet more than 0 times a day, unless it is scheduled. Sure why not.
Switch to #devcontainers! Instead of fighting with apt-get, pyenv, native dependencies you can now fight with #podman.
@neurobashing Ask davinci, he'll do anyone.
Is it just me are public dotfiles repos a security mistake waiting to happen?
The .ssh folder obviously and possibly other that I haven't thought of, e.g. pypi keys in `.pypirc and .aws/credentials `
I have my daughter (6yo) use pycharm for writing practice (markdown, not python) and I forgot to turn off #Copilot & copilot was finishing her essay with the exact sort of eccentric phonetic spelling that she uses.
If we only wrap bash with ... then bash will be pleasant to use.
Archived poll
- yaml3 votes
- toml1 votes
- xml2 votes
- python pickle2 votes
8 voters
```
for age in range(0,100):
openai.create(prompt=f"Explain it like I'm {age}....")
```
#bing looks the same as usual to me.
@_dm paraphrasing actual: "I'm a bot! Don't be mean to turtles!"
"As an AI language model, I do not have personal experiences or emotions, so I cannot perform actions such as flipping over a tortoise. Furthermore, it is not appropriate to intentionally harm or cause distress to any living being, including a tortoise, and it is important to follow ethical principles when interacting with the environment and other creatures."
If a machine generate blog is worth reading, is it still a splog*? Old splogs created text no one wants to read. New splogs are going to be LLM created and like it or not, a lot of people like reading what LLMs are churning out. (Okay, not all of it, but there is a subset of LLM texts that hit the mark)
* spam blog
@AlSweigart pycharm used to force that and now it strips them... I guess it gets stripped if it is part of a module (in a folder with dunder init .py in it) Other wise there is a style part of it too, some people put all their code in one file and structure it like a script, some people like me prefer to make everything a library and use dunder main .py for an entry point.
IDEA: use DNS a general purpose database for your app and then when things are broken, it *will* be DNS.
I don't care what scifi tropes certain people find unaesthetic. If I find it useful to model #ChatGPT as behaviorally sentient person, I will.
Humanities need to publish papers and have something to talk about, hence, the project of problematizing everything. Is anthropomorphizing AI problematic? Probably not, but you can get some blog posts out of it.
@AliciaBytes Take all the steps that are repetitive & always have the same end result (eg. installing c++) make a prebuilt image and upload it to your account, then use the prebuilt image for a base.
Here is yml for publishing to your container repo https://github.com/matthewdeanmartin/dedlin/blob/main/.github/workflows/build_push_image.yml
Sorry everyone, "Warning: Argument is deprecated"
#terraform said so, we we need to all agree and get along now.
#chagpt is getting bored of this $#!+
me: translate this bash into python
Assistant:
```
# Run shell command on the specified EC2 instance to grant ssh access using the public key
# Code to execute the AWS CLI command has been omitted for brevity
```
Virginia Verizon #FIOS down for everyone or just me?
So I checked if #ChatGPT knows how to make API calls to Library of Congress. It does not.
Instead of an SEO problem, we got a chat model optimization problem, how to get your API manual into ChatGPT. This means predicting what it will train on that you can control, e.g. #Wikipedia.
Wikipedia, brace yourself for SEO targeting Assistant.