Nah, I mostly write short one off scripts that are literally "just need to work". So far AI has completely failed 5 out of 6 times. I think the issue is not that it can't write an effective script but that the scripts I write are to glue things together in my environment and it has no environmental context.

Our environment has *quirks*. My most recent failure was trying to convert an AD group membership on AD joined Alma computers (using SSSD) into .k5login files as a cron task using bash.

AI came up with a perfectly good script using getent except for "reasons" SSSD sometimes doesn't actually fill the getent results for a group membership. So in practice in the environment, depending on the computer you ran it on and the phase of the moon, you got null results for group memberships.

Now, stepping back you could argue we should spend time fixing that bug, but all other IT processes just use LDAP anyway. But there's no way for AI to have known that unless I thought to tell it. However, that's also "knowledge" I didn't think to share, and certainly wouldn't expect to put in a extremely long and detailed prompt a la the "vibe coding". If I'm going to psuedo code it out for an hour or two to make sure I provide a lot of context, I might as well have just written the script by then in many cases.

One obvious improvement is that eventually we'll be able to run a powerful model somewhere with sufficient security and data sovereignty and privacy guarantees that we can "ingest all our written docs" and get the AI at least the context a newly hired person would have. IDK when that political issue will happen though so I can't comment on if that "fully fixes" this or not. Most optimistic estimates I've seen from local stakeholders is 18 months.

Ok, but now I've told it to use LDAP. You'd think this would also be well defined and understood, but then AI proceeded to fuck up string mangling and whatever for another hour till I realized I was debugging awk to a level I'd have done if I just was hacking it together myself anyway and called it as any sort of productivity enhancement for me.

Granted, I was fully trying to "vibe code" this because I don't live in AWK and always have to basically iterate test iterate anyway. After I kinda ranted about this in our group chat, a colleague of mine said: "Why don't you just do the filtering and name conversions in LDAP in your initial query?" To which I said something like "whut?". TBH, I personally minimally use LDAP and it's filter format makes my brain melt. But they gave me a 2 line ldapsearch command that did exactly what was needed, that AI had failed to create and in fact had made pages long loops that I know weren't performant... and also didn't end up working at all.

So, the long and short of it is - if getent / SSSD worked 100% of the time, AI would have 110x my script writing. Because it doesn't for "reasons", and fixing that would -100x us as mostly a tangent, in reality it was a wash at best. I'm still going to try to vibe code using the stuff my colleague gave me and wrap the rest of it just to "keep up with AI" but as is obvious this isn't a priority which is why I was vibing around in the first place.

It doesn't bode well though IMO if toy examples yet tied to a real world complex environment have the same issues. I found the one time AI did work for me it was entirely self contained stand alone windows script that didn't interact outside the OSs it was run on.

I won't say that AI won't fix these issues, but IMO fundamentally the limit there is human / business legal / political and how many people want to or can upload all their docs, existing code etc to an AI company, and then can the AIs handle the context size properly.

I'm pretty sure that's not true given MCP, AI agents, and even just web enabled search. They can check other sources now anyway - but I'm not really sure that would prevent them hallucinating cases without additional training and such, maybe a verifier agent or something. Now of course that it's gotten appealed, they probably did this a couple years ago when AI was far more limited.

https://www.thestudiesshowpod....

Has sources in the list of links on that page. Makes a good case that antibiotic resistance has been overblown over the last 20 years.

I would agree, don't take antibiotics for no reason - but delaying treatment when there's a ticking clock for things like anaplasmosis is probably a worse idea than perhaps taking an unnecessary dose of doxycyclene.

We don't have "really effective medication", we have "somewhat effective" at reducing body weight by ~15% (depending on which medication you use). I hate to break it to you, but many people need to be more like 50% less weight to be considered not obese. These medications are amazing but not cures for obesity - just makes it better.

It's pretty easy to get that to $15-$25 a month by doing pre-paid via a number of options for 5g service. Phones got more expensive because tariffs and inflation, but I know 2 years ago I got a OnePlus for $250 at Best Buy near christmas time - IDK if you'd call it mid range or not, but it does pretty much everything if you're OK with Android.

Conservatives sure - probably many progressives too, though for different reasons. What about MAGA/MAHA? They seem to love all the Trump crowd.

Whereas hardly anyone worries about the problems of multigenerational uses of antibiotics.

I'm not really sure what you mean here - the "problem" of multiple human generations taking antibiotics? I'm pretty sure it's commonly thought that antibiotics have saved probably countless lives and likely improved just about everyone's life since they've been discovered and developed.

And they're needed more now than they used to be with the increase across at least the eastern US of tick borne bacterial diseases like lyme and anaplasmosis.

You all are very lucky - where I live I semi regularly find one pump, completely unlabeled, (and it's a random pump and station, not "I'm too dumb to just avoid the problem pump") that just doesn't work right. It won't take a card, or it won't let me click through all the ever more annoying prompts about various stuff.

This is a non-starter for me. Half the time, I'm not paying for gas, someone else is. (we're on road trips, and they're covering gas, or they borrowed my car and are paying for gas). We should be able to pay cash also if we want to, or use a different card that gives extra points for gas purchases or discounts for gas etc... but for electricity.

All around stupid IMO to tie charging to the car ownership.

How the heck would banks affect the US national debt at all? I'm also relatively certain that when there's a fixed payment of X for Y months, the lender can't just decide, whelp, no, I'm now going to jack the interest rate for...reasons. This whole comment is bizarre leaps of logic.

If you're truly in a position where you can't save even small amounts like that, then your living situation was never sustainable to begin with. Something HAS to give, whether that's more of your free time or where you're living. Example: You're making minimum wage but you feel entitled to live in the SF bay area. That's a self-inflicted financial hardship.

This is more about the bay area liking to have some people around working minimum wage jobs, if everyone doing so moved away... The other issue IMHO is - sure you can go move out to the boonies where it's cheaper to live - but then there's no jobs. Or you now need a car - great, you're not saving money, you've just shifted from rent to car costs. And you're more isolated. This even assumes you can move - there's all sorts of reasons that could be an issue. I always wondered why anyone would be homeless in the northeast (or anywhere with a cold winter), but then I started to think about how realistic it would be with no money to somehow get thousands of miles south... There's clearly barriers here.

Meanwhile, some people think they need to live on campus in an insanely overpriced dorm room, then blame everybody BUT themselves for the fact that they owe $300,000 in student loans.

I don't know if you went to college, or when you may have went, but both colleges I went to required students to live in a dorm for at least the first year, and also required a meal plan. I'm sorry that at 17 I didn't have the wisdom to know I should not go to a "good college" like everyone preached to me for my whole life.

If a person is that stupid, they really had no business going to college to begin with.

Not having the life experience of a 40 year old at 17 is not being stupid. It's far less true for new potential students, but 20-30 years ago there was little to no common cultural idea that going to college would be a bad idea, or that the costs were unreasonable. We were all told it was an investment in our future, and that we'd make back far more money with a degree than we'd spend to get one. Even today, if your parents can brainwash you into a religion, they can brainwash you into college debt - this isn't the students fault IMO.

And little numbers add up to big ones too. Why would you spend $5 on a coffee that costs 50 cents to make on your own?

I'm going to let you in on a little secret here - most coffee shop coffees aren't "pour out of a pot" into a cup coffee - that's diner coffee, and it's not $5 even today, it's more like $2.50 and unlimited refills. Whether you think coffee is that fungible (i.e. just a caffeine delivery system) is up for debate between coffee drinkers, but this is a silly comparison IMO - might as well compare dried black beans to a ribeye - both are protein so why are you wasting money?

I've seen people make "higher end" coffee at home, and there's bunches of youtube channels about it too - yes, the incremental cost may be $1.50 I'd say, but the capital investment tends to be a lot higher in good grinders, proper brewing equipment, filters, mugs, cleaning all that. Plus various milks, whipped cream, time making, cleaning, etc.

Heck, I'd have had more respect for the argument if you treated coffee like cigarettes - you sure don't need either one.

I'm willing to admit I'm probably an edgecase, but I asked AI - Claude 3.7 Sonnet in my case - to do the same sort of thing(translating one scripting language to another) and what it spit out didn't work at all (oh, it didn't error out, it just ended up doing nothing when run). It seems very context dependent.

I also asked it to do something very simple in bash - get the users of a group via LDAP and write a .5klogin file with them. I spent a couple hours trying to massage what it created and gave up. Now, a co-worker did it in 2 lines using LDAP search functions that I wasn't familiar with. It seems to me this is exactly the sort of thing AI ought to be able to find and wow me with the slightly arcane LDAP formats... I also enabled web search so it really should have been able to find all the docs necessary. It just... didn't.

Again, if I have to 100% know enough to do it myself to prompt the AI correctly, or learn a complicated skill to prompt it correctly - I might as well just *do it* myself IMHO.

I'd argue, for me, the problem with AI coding is it's a lot like a slot machine. I have had it work once early on and I was quite pleased, but then each subsequent pull has failed. The issue is when I try and guide it out with suggestions or telling it what and how it failed, I end up spending a lot of time hoping it's going to work - but it mostly doesn't.

Then there may well be things that I would think it could be very good at if I knew how to use it better I guess - I'm just not sure if it's actually possible with the system I have access to. For instance, I have one docx file with a bunch of info in it. I for bureaucratic reasons need to match up the document headings and sub heading numbers and take from a table and freeform prose under it and put it into an Excel template file with pulldowns with the same text options as in the table and paste the freeform text into another field where the row titles match.

But I don't think I can upload a docx and an xlsx and give it a slightly more detailed description and then download a filled out version of the xlsx.

You might think you came out a better man, but I strongly question that as being widely true. PTSD tends to really screw with people as I know from a relative who was in Vietnam.

I see lots of people blaming the user, but if you look at the hype and what so many of the influencers and what the CEOs etc are reported to be saying - he used it exactly right. The way it's promised to be able to be used.

If it's actually just a search tool, then it's not replacing any coders right? You still need people to do all the hard stuff - no one needed AI to make slightly different SO posts to copy paste and massage.

The other issue is - often to use it like you're saying, you need to somehow feed it all the data, which requires you to figure out what data is relevant, figure out how even to feed it all in (context windows are a thing), and potentially do almost everything except typing.

That's a much less interesting sales pitch for people "Look, this thing is a slightly better search that avoids google sludge (somewhat)". Probably not worth a crapton of money for a subscription to a lot of companies then.

Congress has shirked its responsibility to jealously guard its power. That needs to change.

This exactly. I've been saying this since the 90s - Congress needs to stop abdicating its power and actually make laws and balance out the other branches. At least half of our issues are people who as far as I can tell are doing the best they can to make up for Congress doing fuck all, but still having to deal with stuff that wasn't legislated 80 years ago because no one knew about it. The other half are power hungry people. As we've seen very little good comes of this.

It seems obvious to me that we need more than one person to declare a war, or an invasion or whatever. At the very least, it should be someone other than the person wielding the "emergency powers".

I just don't see Congress starting to do anything anytime soon. The incentives just aren't there. I don't know how you get Congress able to do something, because the options seem to be either enable even wilder swings (assuming Democrats ever get back in power and Republicans again after that) through simple majorities OR some fundamental change in people getting along or compromising. The former probably would actually just push us closer to governmental and foreign affairs breakdown, the latter is a Utopian dream.