tharkun__

I don't understand how "wall of text" is related to git large file support. The wall of text is a problem for me, the human. Sure, there are ways, like "be brief", caveman etc. In a large repo with lots of different people over time, I can't see how it won't just be wall of text again. It's just too much. TL;DR. And coz DR, the LLM will have buried bullshit in that text, which future session might read and "believe".

As for "no dude", no that can't be put down into rules. Not all of it anyway. We have stuff encoded in the repo wide md file, I have my personal one etc. and the various agents still don't do what we tell them to in all cases or a new model comes out and it no longer works. For example, for finding the root cause of a bug, it's very important to have actual proof and references. It's getting there w/ my instructions in the .md but it doesn't always work and I do have to "dude" it from time to time.

Is that back and forth valuable to have in files that are going to be part of the repo? I very highly doubt it. Having new rules that came out of the back and forth in a checked in AGENTS.md, sure, that is valuable. Or nowadays in individual "skills", like a "root cause finder" skill that can have very specific instructions about being thorough in proving its "found the smoking gun" BS ;)

I've seen enough PR descriptions created by the agent. Fluffy wall of text that looks good but is factually wrong. Seen it way too many times. Too many people just look at whether it looks good and then pass it off as truth. I'm tired of it and making that into "nice HTML" doesn't make it better. It just makes it look even nicer but not more true.

Re: reproducibility. My parent poster (and I guess you as well) wanted to have the prompt/conversation as "documentation". I don't see why that would be helpful. The only reason I could see would be for "reproducibility", which you won't get with an LLM. I don't see why else, but do tell me.

What I can agree could be valuable are the "why"s. I.e. the stuff that already should have been part of the ticket/requirements document. If you want to store that inside the repo as text files, instead of the original tickets or documents, that's fine of course. But I don't see how a "recording of how the code came to be" is valuable. It's like having a recording of all my IDE keystrokes and intermediate code state in pre-LLM days. Not valuable. What's valuable are the requirements and the outcome (i.e. code). Not "the thing in between".

Now don't get me wrong. Recordings of how people code/use their IDE can be a valuable teaching tool. Both as good and bad examples. And the same can be true for an agent coding session.

FWIW, BDD and frameworks like Cucumber don't work at all in my experience. The people that'd need to fill these out don't do it properly (they can't) and then we, devs, are stuck with brittle and un-debuggable stuff that's worse than if we just used regular code to encode what we understood from them.

It's the same reason (most) PMs armed with an LLM still won't get anything usable done. They can't do it properly. They still need devs. But the gaps are shrinking. Some few PMs can get stuff done w/ both Cucumber, could wireframe UX with previous tools and can now do so much easier and better with an LLM.

    It would be nice if people also committed their initial prompt and chat session with the LLM into their codebase

I doubt you'd want this. It's a chat session for a reason. It's gonna be huge wall of text, especially if you meant to actually include all the internal prompting the LLM did while it was working. You'd also have all my "no dude, stop bullshitting me! I told to ignore X and use Y and to always double check Z and provide proof".

It would only "work" if every single piece of feature you wrote was 100% written by the LLM from a single, largish and well defined prompt, the LLM works for a few hours and out comes the feature. And even then you have no reproducability (even if you turned around and gave it to the exact same model, no retraining, newer model, system prompt etc.).

I know you're trying to be funny but ... technically it's 100% clear: You should talk HTTP, because that's the URL scheme here. The port makes no difference. You just happened to use a port by name. For all we know I run my HTTP server on some NFS related port so all the script kiddies try all the wrong exploits on it or something ;)

    “people like me are still needed” is just a desperate form of self-persuasion.

No, no it's not. I've seen what "PM armed with an LLM" will do. Trust me, if you're a decent enough Full Stack software engineer that can take an idea and run with it to implement it, you'll have a leg up over the PM with the idea that has no idea how to "do computers".

Most of what these PMs can produce nowadays turns boardroom heads, sure. But it's just that: visuals and just enough prototype functionality that it fools the people you're demoing to. Seen enough of these in the recent past.

Will there be some PMs that can become "software developers" while armed with an LLM? Sure!

But that's not the majority. On the other hand, yes there are going to be "software developers" that will be out of a job because of LLMs, because the devs that were FS and could take an idea from 0-1 with very little overhead even in the past can now do so much faster and further without handing off to the intermediates and juniors. They mentor their LLM intern rather than their intermediates and juniors. The perpetual intermediate devs with 20 years of experience are the ones that are gonna have a larger and larger problem I'd say.

The Staff engineer that was able to run circles around others all along? They'll teach their LLM intern into an intermediate rather than having to "10 times" a bunch of perpetual intermediates with 20 years of experience.

If it's anything like the animated Babylon 5, no thanks :(

If you love animated stuff I guess it might be fine.

Damn, now I want to binge-re-watch Firefly.

You'd be (half-jokingly) amazed at how many people are entirely incapable of understanding and debugging an existing code base.

Like, literally: "Error message has string 'abdc-1234-something-whatever'". They can barely figure out to maybe search the code base for that error message. Unfortunately they can't find the full string. Now they're stuck and can't think of anything else to try.

So, effing, amazing. How do these people ever get through (coding) life? Ever heard of substring search coz error messages frequently have parts that are concatenated/variables inserted? Search for parts of it until you find something. It's not that hard dude, yes the 1234 probably is some dynamic id, so search for just something-whatever and you'll instantly find the relevant code and you can debug further.

But no, this "Senior" can't think of anything when not finding the full string anywhere in the codebase and would rather throw up their hands and let others figure it out.

Either a really dumb "Senior" that somehow got through so far at previous companies or they're silent quitting during probation period already.

If this continues it's not gonna be silent.

    For every class with a required language, it was not taught during class. 

This.

We did have the oblig "learn language X" course. Which used to be C and had just switched to Java at our uni. But it was just to learn a language.

The Operating System class used Minix in a VM to teach concepts. You were expected to just use a Minix vi (that was fun!) or somehow figure out yourself how to best use something outside while still having a fast enough feedback loop. Nobody taught us C (since we were already a Java language year. You were just expected to figure it out. Quite a bunch of people couldn't and literally quit CS entirely over that course. I loved it!). We had to add a memory compactification algorithm to Minix to pass the course. That was a fun learning experience! And made me appreciate modern (well, I guess this was like 25 years ago now lol!) operating systems.

Scripting languages course: Pick one and do a project and turn it in. That was it. (I turned in something I had written in Perl as a side-job for hire and got an A without doing any course work at all (except for the two initial attendance checked classes)

How many military bases does the US have in Switzerland again?

Oh, the number is zero?

Germany? Well guess what, the US has a very prominent airbase and listening station in Ramstein and a bunch of other military installations there. Also: History.

Withdrawing 5k troops that are probably just rotational troops that ... go home without replenishment is different from actually shutting down bases.

Unless Germany denies the US Ramstein airbase and spying operations, you can safely bet they'll stay. Even if they withdraw everything that isn't required to just keep operating the airbase and listening posts, they'll at least keep those around for as long as they are able to actually use them.

What's every day?

Exactly, lots of different things.

When I alt-tab I care about latency.

When I ssh I care about latency.

When I download a 25GB game I care about throughput for the download to a certain extent that is probably mainly ISP bound rather than local system bound. I don't care if the download takes 10 or 11 minutes as long as I can still use my system with zero delays meanwhile. And whether it takes 11 minutes of 3 hours depends on my ISP mostly. But being responsive to me while it downloads is local latency bound.

The Youtube example you have makes sense, sure.

Maybe I'll have to try something new at some point. Fair. It's been a while.

I just googled this and what I found was this for example:

    The Sony WH-1000XM3 is much better at canceling noise above 100Hz than the Bose is. However, because the Bose QC35 II can block out more sub-100Hz noise, it does a better job at killing unwanted car engines and low rumbles.

So sounds like it's just gonna be a different kind of noise that will still come through. So instead of still hearing voices, but much clearer I might hear more of the AC humm. Sounds like a wash unfortunately. And one the company won't pay for ;)

One thing that immediately turned me off when finding the Sonys on Amazon: It says "Alexa". Sorry, immediate and 150% no thank you, see you, bye.

Which is now out the window.

And yes I definitely remember Colbert quite some time ago quipping about exactly that (paraphrased from memory): US no longer reliable NATO partner and nuclear deterrent. So Europe needs to step up. Let's have Germany have nukes. What could possibly go wrong!

The obviously funny thing being, that the US has, for a long time and Trump doubled down, asked Europe including Germany to spend more on military. And the "moderate forces" in Germany are not an issue in that regard. Those are the ones not wanting Trump to use Ramstein airbase in a war he started.

But would you want the AfD to come to power and wield those ramped up, potentially now nuclear, forces? The party that was ruled as "definitely extremist right wing aka neo nazi" in some federal states by Germany's own "Federal Office for the Protection of the Constitution". Oh and also was that not the party a certain Elon Musk and Trump were trying to prop up? Which is doubly funny because of the AfD's alleged ties to Putin (sometimes more than alleged).

This. And on top of that, headphones at office suck, at least for me.

They don't drown out enough even with large, well insulated cups. So you add noise cancelling. Which drowns out more but not everything. In fact it keeps some very annoying stuff around that is suddenly actually audible VS being drowned out without the headphones. And having noise cancelling on for 8 hours straight for days in a row actually creates some significant pain in my ears. The next idea is music to drown out what's left but that just distracts me too.

Remote is the only good way.

In fact, being remote means I have "social interaction budget" for the family again VS it all having been used up during work hours (being an introvert)

Meh. YMMV as always.

We tried that on the team when Covid hit and we all went remote. Lasted like a week and we were sick of it. Never reintroduced.

The US did not keep bases in all of West Germany though.

There were different sectors. The US had essentially the South. There were also the British sector and French. The Soviets were the fourth sector but we all know how that one was quite different from the other three.

While the French and British have mostly left, the US stayed. Though to be fair even the British still do have some bases it seems as NATO troups. But no more large garrison in many larger cities.

The US on the other hand is still there with much larger force. Like think back to "Air Force One" (the movie with Harrison Ford) which used Ramstein Airbase in the movie (though they didn't actually film there) and that airbase has come up in the Iran conflict as a conflict of its own. Meaning Germany didn't want the US to use it as a hub for US operations (supply logistics) for the Iran war.

See, you are probably right about that new hire. But the way you say things here, you do come across (at least to me) as "that guy". You know? "That guy" that says "we've always done it this way and that is why it's good and why we still do it that way".

What happened to "never change a running system" is that if the system is barely running at all, you better do change that system.

If I'm the new guy and you tell me how to do things and those things seem bad and there's no explanation for why they need to be that way, I'll also ignore you, coz I know how I want to do things and how things can be done better. Don't tell me to do things X way. Many roads lead to Rome and some are better than others. See my other reply. At my very first job I was also told how things work and how to do things. But things sucked and so I made them better anyhow.

Now, in the other part of the thread you also did mention how they just sent you error logs without reading and thinking about them themselves and such. That definitely is a red flag and the kind of thing that will make me fail someone's probation period. Definitely. But just because someone doesn't think that "the way we've always done things" is a good reason to keep doing something bonkers is where I'm no longer with you. And again, probably just your wording/what you disclose in various parts of these threads but it explains why "we concentrate on certain things only" ;)

Hehehe, definitely been there and dealt with that guy that I definitely didn't let pass his probation period. I see we do understand each other.

And I might not silent quit right away and try to actually improve things and see how it goes/how you "let me".

This does actually remind me of my very first job actually. It's been many, many, many years now. I found a stinking pile of disparate shell and perl scripts that made up the "backend" of the application I was hired into working on. Grown histerically for almost 15 years prior to me being hired fresh out of university. I started extracting common library code out of every single one of these every time that I was tasked with adjusting one of those scripts. I introduced a proper deployment from source control to production the second time I "broke" something, because someone had previously fixed a bug directly in production and forgot to check the fix into source control (and no, I didn't believe that it was not the guy that was working on the project w/ me for one second after seeing how they coded and how they defended everything that was bad about that pile of poo and how "it's too complicated to do X".

Well guess what, this new grad did all of that anyway. Without AI and without IDE refactoring support (I mentioned Perl and shell scripts, did I?) and without a single QA person in sight. And yes, every single one of the readers here very probably has bought a product that was "touched" by that software, without knowing, since it was an in-house administration tool.

See, I would probably have been the guy that ignores the dest part. On purpose. Just to see whether this pile of poo shits itself and how much.

I would also recognize what happened when I see the error messages though and then silent quit until I've been with you guys for long enough my resume doesn't take a hit just because your interview process duped me into starting at your place.

Yes this sounds harsh. I know. Nothing against you personally. But I've been at too many such places. One can do things better.

They do which is fine.

Where's my Matrox?

Matrox Mystique was a combination 2D/3D consumer card, which at the time, was still something that mattered. Sure a Voodoo addon card mattered more, very soon, but then quickly things shifted back to combination 2D/3D card with Nvidia!

Also, how is a "first $2000 consumer card" something that "matters"? That's precisely the kind of thing that doesn't matter. My entire laptop cost less than that and I play games with it. What matters much more is that I can play quite a bunch of games that are even pretty recent with a laptop that cost less than that, all with an integrated graphics chip from a company that is precisely known for having abysmal 3D performance: Intel (I have an Iris Xe)

Haven't done it in a while, but I've done some tasks with both Codex and Claude to compare. In all cases I asked both to put their analysis and plans for implementation into a .md file. Then I asked the other agent to analyze said file for comparison.

In general, Claude was impressed by what Codex produced and noted the parts where it (i.e. Claude) had missed something vs. Codex "thinking of it".

From a "daily driver" perspective I still use Claude all the time as it has plan mode, which means I can guarantee that it won't break out and just do stuff without me wanting it to. With Codex I have to always specify "Don't implement/change, just tell me" and even then it sometimes "breaks out" and just does stuff. Not usually when I start out and just ask it to plan. But after we've started implementation and I review, a simple question of "Why did you do X?" will turn into a huge refactoring instead of just answering my question.

To be fair, that's what most devs do too (at least at first), when you ask them "Why did you do X" questions. They just assume that you are trying to formulate a "Do Y instead of X" as a question, when really you just don't understand their reasoning but there really might be a good reason for doing X. But I guess LLMs aren't sure of themselves, so any questioning of their reasoning obliterates their ego and just turns them into submissive code monkeys (or rather: exposes them as such) vs. being software engineers that do things for actual reasons (whether you agree with them or not).

Hah! Funny how everyone here seems to be thinking the same: They said something about "finally 4k moon rocket images" and the stuff we got in the news was like blurry 800x600 type with lots of JPEG artifacts and such.

Even the smallest resolution images I see in the link that the parent edited into their comment have better quality than what news outlets posted.

I want TIFFs that takes ages to download and I need to scroll around in/zoom out on!

"sammeln" can have multiple translations. "Collect" would be more like "einsammeln". In the context of "Pilze sammeln", you'd use "forage". You forage for food.

Not quite the scenario from my parent. They said "high cost weapons taking out air defenses". Whatever the US equivalent of an Iskander would be (I used a Tomahawk as an example), the S-400 (i.e. Patriot "equivalent") would be used to defend against it at first/in his scenario.

If you want to turn it around, sure. Let's see how you'd want to take out a Patriot: high cost weapons, like an Iskander might try it? Costs about as much as a Tomahawk? Would need multiple ones, because the Patriot would defend itself against even multiple ones? But the Patriots cost as much and you want multiple interceptors for each Iskander sent its way?

What if I could send, for less money/resources, a drone swarm that also takes out the Patriot or at least expends more money/resources in interceptors shot from it, than I had to spend on the drone swarm?

I totally agree, it's "just a race". If I build an offensive drone swarm for $x, which is less than your high cost interceptors, you better build an "anti drone whatever thingie" (which might be anti-drone drone swarms) that's even cheaper.

But, thanks, essentially you're agreeing with me: Don't use your high cost stuff to take out SAMs and then use cheap drones. Instead, use cheaper stuff to swarm it out of existence. Just gotta be faster at being cheaper. Doesn't matter if you're the attacker or defender.

Zerg vs. Protoss.

    namely your high cost weapons take out air defense capability, so you can stop using them and use cheaper more numerous systems to hit the now undefended targets.

That makes no sense to me. Why would I spend millions times, dunno how many do you need for a guaranteed kill for an S-400, if you could spend hundreds of thousands on cheaper ways to kill the same S-400, while the S-400 still defends itself with millions worth of its own missiles?

That's precisely what Ukraine was/is doing and has developed. The West provided lots of military support, including the US of course, but way not enough as we can see now play out in even the US itself vs. Iran. They developed cheap drones that can shoot down cheaper Shaheds. Shaheds that are way too cheap to use regular interceptors for. But even cheaper drones tip the scales back.

Why would I want to waste Tomahawks 1:1 vs. S-400 interceptors, if I can kill it with a much cheaper drone swarm?

Not saying those precise conditions/weapons exist today. I have no idea. But if they did, why would I still waste my high cost weapons.

This so much. As a user, especially a private user, I want my apps I can install and run locally, no internet connection, nobody forces updates on me for an app that does exactly what I need and I'm used to it.

As a developer, SaaS all the way. I really really love not having to deal with versions, release branches galore, hotfixes on different releases and all that jazz. I'm so glad I could leave that behind and we have a single Cloud "version" i.e. whatever the latest commit on the main branch is. Sure we might be a few commits behind head in what's actually currently deployed to all the production envs but that's so much more manageable than thousands upon thousands of customers on different versions and with direct control over your database. We also have a non-SaaS version we still support and I'm so glad I don't have to deal with it any longer and someone else does. Very bad memories of customers telling you they didn't do something and when you get the logs/database excerpt (finally, after spending way too much times debugging and talking to them already) you can clearly see that they did fudge with the database ...

The big "test" there for Steam will be when "Gabe goes away". It's gonna happen sooner or later.

That was literally the first thing I thought reading from OP comment down to your parent.

Then I thought: Sure but management made the devs promise these things. We don't do it of our own volition (exceptions prove the rule - some people are conditioned to do it of course).

    I would argue that we are seeing a new emerging group of coders come into the realm of programming and we are judging them at their worst and comparing them to our best

Nyes. I think what we're doing is that these new guys are coming in and using AI and trying to tell us how super awesome and powerful they are because of AI and that nothing could ever go wrong.

    It is quite insane to me to expect someone who just started to fully build google.com and all of it's infra,security,etc.

But it's not us expecting them to do it. It's them telling us they can do it coz they have AI.

Look, I've been using Claude and Codex agents for about 6 months now, full time for coding (when I code) essentially (coz I can't ask my people to use a tool I have no experience with myself, so I purposefully forced myself to use the agent and only the agents as much as I could bear, only resolving to manual changes in very very few instances. And there have been many many frustrations, believe me).

The amount of times that even Seniors have just verbatim pasted Claude analyses as truth to me, when it was apparent after the first read through of the output, that it wasn't true is amazing. How we expect juniors that have way less developed "spidey senses" to successfully navigate that is beyond me. Most people are trusting by default. They shouldn't be, but it's human nature for most of us. For some it isn't, like myself. I'm already the dude that asks too many questions of humans when they're not clear on what they assumed vs. have verified.

Like, example, I showed an analysis, full page in a slack thread recently to one of my Seniors (made by some other Senior) and tell me where they think it shows that it's BS and not true. He couldn't do it. He tried over and over and he was unable to. I read it and the second paragraph out of lots of them was BS and just not true. Easy to verify. Claude didn't have access to the actual information (because of various circumstances) but just made something up. Said the relevant code was deployed, thus XYZ was true. Listed lots of extra analysis after that, which sounded reasonable and probably was, if the premise was correct. Just it wasn't. The code had never been released at that point.

I've been doing the same kind of "spidey senses are tingling" comments and questions back to people for lots and lots of years. And others are usually not good with that sort of thing (exceptions prove the rule). Coz people do the same kind of "BS-ing" that Claude et. al. do. Claude is generally "better" about questioning his/her (yes, it works both ways) judgement actually than people, which in many cases have feelings attached to their investigations (even if they very blatantly didn't check something and just assumed it - pre AI - all by themselves).

Oh so much this, in a sense.

Look, as a software dev myself, I really like that my company lets us use our computers the way we see fit. Pre- or post-AI with no restrictive lockdown. Been there, hated that.

But I totally get the freaking out over "normal devs". The amount of stuff most people think is reasonable, AI or not, is mind boggling. For myself of course I like to just be able to be responsible myself. But as a security team I'd also be freaking out.

Like, the amount of people that find our super boring, totally corporate "security training videos", helpful and insightful and "oh dang I'd never have thought of that!" is mind boggling all by itself. Never mind any actual security training that'd be useful to someone with half a brain. You can literally just click through the 8+ hours of stuff you're supposed to watch / answer / do in 30 minutes.

I use Claude. I use Codex. I've never heard of or used Meta AI. Nor do I have a Facebook account. Never have, never will.

I am also a software developer. So while the numbers of "people" that use one AI or another may be higher than either of these, it's not a useful metric for myself.