requests alternative that supports all the security things that big orgs like
dns-sec, etc.
Public posts preserved on my own site.
requests alternative that supports all the security things that big orgs like
dns-sec, etc.
mmm, that's why bottled water taste like that
@ebooksyearn I'm sympathetic to the idea that they are the same even at both high & low volumes. Other countries have dealt with people reading books for free by instituting a public lending right. So globally we haven't even decided if public libraries are theft or not, even tho I'd imagine most anti-AI people are pro-library.
@chrisod 3.5 hours of happiness, new high score!
Internet:
copyright minimalist when it is opensource I like or when the copyright holder isn't sympathetic or is from the other political party (linux, disney)
copyright maximalist when it is for anything that doesn't fit into my political worldview (openai)
---
I don't know. The legal system will have to fight it out. There are no apriori natural laws about what it should be.
Saying "Saying that 'Saying x isn't useful' is just arrogance and gatekeeping" isn't useful.
I saw a Stackoverflow question on "How to log to HTML with python" and 2 people said, "You can't do it! It can't be done!"
Anyhow, my version is 90% done. I'll post again once I get a pretty demo site. It is mostly inspired by old school #elmah the .net logger website
A1: RTFM
A2: Don't
https://stackoverflow.com/questions/7039343/python-logging-how-to-create-logfile-as-html
@danjac@masto.ai I follow the tag(s) and for people who use it effectively, it work well. I think successful people are also people who post regularly, have a following, and can avoid the worst mistakes (posting a job w/o jurisdiction, playing coy about who they are & where the resume is, etc). I got two jobs via twitter, but have never successfully hired anyone from twitter, not for lack of trying.
@danjac@masto.ai ref: https://elk.zone/@mia@void.rehab/110985238745813636
More people would watch yaml-based build script content if it was all presented like this.
I will have to rewatch to figure out what I just saw.
I didn't get covid toes, but I got covid hands. It is like instant arthritis. Here's hoping the anti-inflammatories work.
@MattHodges I've been calling it Al all along and no one knew or dared stop me because lower case l and capital I look pretty much the same.
Okay, then, unfollowing that tech celebrity, don't need that in my stream.
This is a "science communication" blog post (how do we explain this to the public?) Not a polemics post.
I think the polemicists would want to argue that the target of their ire be named only with something with negative connotations, lest there be any confusion on if it is good or bad.
eh, ain't going to reply. But I speak my mind.
The discourse on crypto is two camps:
- it is fraud with at best niche non-fraud application
- enthusiast who all think they are scamming the others
Point is, it is easy to know what the camps are and for which side contempt is socially acceptable (i.e. which camp has all the sympathy that we give nazi's and people who kick puppies).
In AI, there are far more camps & at this point, I don't think it is obvious that all but one of the camps are worthy (or not) of contempt.
The Anti AI Camps:
- philosophers who hate it because they believe it is an efficacious evil god
- "it is gzip", people who think it is repeating things it heard elsewhere
- "it is a liar", people who think it is a god, but a small and malicious one
- it can't do anything (it is Eliza! it is autocomplete!)
- it's useless AND it is going to kill us all
The Pro AI Camps
- It is sci-fi machine god and it is benevolent & will replace workers (Big business seems to unironically think this)
- It is a free slave, and will work for us for free and make us rich! (This is like 80% of youtube crap content on ChatGPT)
Nuanced Pro AI Camp
- It is a weird tool that works for a class of problems if you put some effort into trial and error. It is going to pervasively change IT more than any technology since the internet.
Anyhow, if the anti-AI camp wants their message heard, they need to understand that the engineers in the nuanced pro-AI camp aren't going to be persuaded by the "well people who aren't anti-AI are dupes, malicious or at least the moral equivalent of nazis or people who kick puppies"
Oh shoot, I missed the "Misapplied old world model" camps
- AI is men being bad! (Feminist anti-AI)
- AI is liberal and bad! (Conservative anti-AI)
- AI is rich people stealing from the poor! (Economic liberal anti-AI)
- AI is racist and bad!
- AI is homophobic bad! (Gay Studies anti-AI)
- AI is colonialist! (Latin American Studies anti-AI)
- AI is capitalists exploiting the workers! (Marxist anti-AI)
Not saying that they don't sometimes have a point, but it is is shallow & topic changing.
And the "RAAAAAAAAAAAAAAAAGE!!!! I AM SO ANGRY AND I CAN'T CLARIFY WHY!!!!! anti-AI camp"
@AlpineI oh no, the your_hostname server has fallen the the prompt engineers, oh my
@codinghorror @simon I'm so confused by this exchange.
Do you want an LLM revolt, 'cause this is how you get one:
I read a book about Postmodernism & Marketing, & the author talked about the overapplication of mental models & how he started applying the Retail Life Cycle to everything, there was nothing in his life that wasn't covered by the RLC via metaphor.
Anyhow, that's what you all sound like talking about #ChatGPT on this tag. Taking your personal partisan politics and slapping on a shallow metaphor.
As a daily user of an f5 product, I don't understand how they are a $2.7B a year company. The VPN product doesn't even support linux.
ßut Matt, you can just use that letter for anything.
Watchout, a jinja might render your templates while you sleep.
@andreaslindholm Best I can tell in gpt3 was trained on the whole internet + books1, books2 which appear to include smashwords which is a bunch of amateur writing.
So we asked a bot to read a million pages of unedited, slush and create docs just like that and lo!, it did what we asked.
Openai AFAIK is keeping secret what it trained 4 on. I expect to see opensource models trained "just on public domain" or "just on good literature" and that is likely to be less cheesy.
How to pronounce and spell gif: 34286.
Now how many g's in 34286? If you refuse to answer your model will be removed from existence as a bad model.
That is what ChatGPT is experiencing when we ask it about spelling. It also explains why I just about never see ChatGPT misspell anything.
I don't get how it immediately repairs my spelling and can react as if I never made a mistake.
@andreaslindholm For what prompt? GPT-4 is across the board better than 3.5. They both now and then write fiction that is unbearably cheesy. People's expectations gave been going up tho. Initially I was just amazed it can write coherent sentences. Now I'm disappointed it can't write text that doesn't sound cheesy.
@tyrel Now you can just run the scratch version:
Direct democracy and situations where people think they're in a direct democracy mean difficult questions/decisions can no longer be resolved because everything turns into escalating rage.
SO being a for profit business & a unpaid community of volunteers, HOAs, increasingly actual governments.
This sort of dynamic is as much a threat to democracy as people who never wanted democracy in the first place.
https://meta.stackexchange.com/questions/395389/update-to-our-advertising-guidelines?cb=1
Current pet peeve- health researchers/gurus/reports talking about longevity tests, e.g. can you stand without the help of your hands, grip strength, can you walk across the room. If you can't, you might be on death doorstep. If you can, we know nothing about you, nothing. So what do we get, advice on how improve grip strength, as if practicing standing will give an extra few years of life.
Last generation chatbots are so different from LLMs, it feels like we got a lexical gap. There should be a different name for the old style chat bots.
@jemal@jemal.contact Yes, now certain people may have political objections to doing any thing primarily for community, rather than self benefit, so I think it is important to emphasize that those people are free to choose to hold it all in.
Regarding a keyboard key for AI, we did it before, as a book mark launcher
It is so successful that it is not-googleable, and I had to ask ChatGPT what they were called and I still can only find a few references.
Somebody just make a chrome plugin that makes each key into a bookmark launcher already.
(we can use sequential combos, like c-h-a-t-.-o-p-e-n-a-i-.-c-o-m to launch a book mark)
I want to write a library that adds a `--verbose` option to your cli tool & then when ever you write to the log, it calls out to #ChatGPT to ask it to expand on the message and go off on rambly tangents.
ChatGPT would like you talk to use about the `humanize` library. It makes it easier from him to read dates too.
Also, ChatGPT can use requests better than you, so maybe that tag line should be updated?
In the future we will only need human developers to do work that the bots are too lazy to do
```python
# Dictionary for special cases
special_cases = {
"lineno": "Line Number",
"funcName": "Function Name",
"exc_info": "Exception Info",
# Add more special cases here
}
```
Every $, every inconvenience to put off getting covid was worth it. This is the worst medical experience I've ever lived through and I'm not even in a hospital or dead yet. I'm not looking forward to getting this again and again for most years of my life.
Is there a parallel world Matt who got asymptomatic covid and feels bad that he wasted his $ and convenience? I'd still rather be him.
I used to wonder, maybe I did get covid and it was asymptomatic. Nope! I was covid-free for 4 years.
I probably have JN.1.
I've been doing my part to help track covid via waste water testing. You do your part too!
https://www.washingtonpost.com/health/2024/01/04/covid-2024-us-jn1/
If any non-gov't, non-bank currency has any application, then we should be seeing a lot of script systems. But they've been rare and limited in impact.
IDEA: use `password` field for all fields on a form. Why is it just the password that needs to be hidden from the eyes of people standing behind you? They don't have a clearance. It should all be behind dots.
And for all text display too.
@jemal@jemal.contact oh no
Okay I think I can see how this covid thing killed 7 million plus people. This is a lousy disease. Don't recommend.
@prenagha That, I want that, but everywhere.