For your basic words-and-dates-with-shapes-and-shit posters, the AI ones are annoying, but I’m not gonna die on a hill over it. The hill I _will_ die on is restaurants using AI generated images of food. No. That needs to be illegal. I thought it _was_ illegal.
>That needs to be illegal. I thought it _was_ illegal.
Ummm.... I have news for you, pretty much all ads have used fake food since the beginning of advertisements.
Nobody ever cared that McDonald's ads makes McDonald's food look appealing but all of the sudden when mom & pop places can do the same cheaply and easily there's lashing out saying it should be illegal? Wtf?
Sure, yeah, I understand what foot photography is. McDonalds etc. are at least riding the line instead of ignoring it. Using real food to carefully craft a scene, no matter how "done up" it is, is still representative of what's in the product.
Generative AI art is going to fabricate things. Especially when used by smaller shops who won't fact check it. I go to restaurant, look at picture of food, and expect to receive that. Period.
I’m sure it depends on the type of work, but for mine, Codex is much more helpful. Honestly, it mostly comes down to it being significantly faster, probably because as many have said, it seems tuned to not spit out word vomit, both in its chat interactions, and its code (Claude is obsessed with massive comment blocks that are basically guaranteed to become dead context noise if you ever use it to iterate on code).
I mostly do very obsessive, tightly scoped, carefully thought out small changes on a fairly boring stack, one interaction at a time, verifying functionality and code. I know what I am doing, but I also know what I don’t like doing (the same exact set of things I’ve already done a dozen times in my career)
If every commit has full ai conversation baggage, and you feed that history back to an ai… aren’t you wildly inflating context with stuff that may or may not matter? How does this even work with thousands of “commits?”
There is a universe where prompting with large context will be near instantaneous and pretty much free. In that universe, there would be no reason not to have even more robust and verbose logging like this. For thousands of commits, a chunking mechanism and a running summary of the chunks would roughly get you there.
realized this one about myself earlier this year, it really helped to boil it down to something besides “I am just inherently bad at things.” that attitude worked as a dumb single kid, but it was harming my adult life and relationships.
therapy helped me get there. I have been on bupropion for about a year, and recently started on methylphenidate. it might not be the right one for me, or maybe too small of a dosage. I’m taking it slow and being deliberate with the drugs.
working with a personal trainer to get in shape and lose weight, as well as quitting my fully vested tech job to fuck off and be a cook for a while has also done wonders for me here. it’s cliche, but you really can’t replace fitness with anything else, and that took me 35 years to internalize.
> quitting my fully vested tech job to fuck off and be a cook for a while
I am also fantasizing about this and am only holding off from doing it because of the social stigma (what idiot would quit a well paid full-time job). My biggest issue with the software industry is the feverish shiny-new-thing syndrome that AI is causing (and my current company is all in on this, with "Hyper-Velocity Engineering" panels). Maybe I don't want to move at light speed and would rather stop and smell the roses.
pretty much every day I feel that tickling of “what the hell did I just do?”
but I don’t care. I put 15 years into my tech career. I am good at building software, and I will not let this ridiculous “resume gap” problem stop me from taking a break for my mental health. any tech employer that wouldn’t want me because of it is a place I wouldn’t want to work anyway.
also, to be honest, I’m writing more code now than I ever did in the last year of my tech job… working on a full CMS and custom website combo for my friends bar, such that I can copy that template over for future projects (want to help local businesses escape the bullshit machine). also building a cool web development desktop app. and more! I’m having a great time
> Maybe I don't want to move at light speed and would rather stop and smell the roses.
This really struck a chord with me. I've spent the last 15+ years building up a craftperson's skillset (IMO) akin to a carpenter's or mechanic's. Yet, people still seem surprised when I tell them I'm not willing to run a slop cannon and excrete software which is _good enough_. I actually enjoy the nuts and bolts of writing and debugging software and using AI feels like cheating (if only myself). I'm really not sure where I go from here. I wish I had a work situation like yours to complain about but I know I'd have hit the eject button the minute someone started mandating anything about my workflow, so it's kind of moot.
> Yet, people still seem surprised when I tell them I'm not willing to run a slop cannon and excrete software which is _good enough_.
Cue the usual propaganda: "Oh but that's just real life," "it's the nature of our industry," "I'll agree things aren't perfect, but {nothing must change}," "I understand, but the true problem lies with {mistake we'll gladly keep making}," "it's bad, but at least we're not {killing babies}."
It's become almost automatic to associate pondering with perfectionism, and perfectionism with flaw.
yeah! it's pretty cool when I start doing something that used to be difficult and now it's just... not. weird how long your old form sticks with you in your head like that.
yeah, at a local pizza/taproom place. they share tips with cooks, works out to ~$35/hr. I'm not expecting it to replace a tech salary, but it helps offset the savings burn (which was specifically set aside for this). engaging with real people in real life about real things is also an indreeeedibly nice change of pace (even the stressful/tense moments of shit hitting the fan).
I mean you can burnout on kitchen work as easily as tech work. Maybe easier. I know that first hand. The one nice thing is that when you clock out you can not think about it until the next shift. There is some peace in knowing you'll never get a call at 2am because some bird's nest of technology fell apart and people can't order their custom cat pic coffee mugs.
I have always wanted a web browser with tmux/iterm2 style tabs/panes, as well as the ability to save the window configuration and tab contents to a file that can be checked in to projects and loaded.
Load all of your project's documentation links, local development browser tools (database viewers, etc.) into a set of views that can be source controlled with the project. Don't force people to use their daily driver browsers for this, or hack side-by-side views together with their OS window managers. Zen and friends have split panes, but it's not the robust tab/panel system that I wanted.
There is a simple tab widget system, which so far has:
- Viewport manager: basically 1:1 with what is in Chrome devtools
- Session manager: create and manage browser sessions as a first-class entity, and attach/detach tabs to these sessions. Includes a simple "incognito" toggle as well.
- System light/dark override: stop flipping this on/off for your whole OS to test "system" light/dark mode (tedious)
- Reload trigger: pick a target tab. if that tab reloads, so does this one.
- Log file viewer: if the tab source is a local file, change the tab's view to a structured log file parser with search/filter, play/pause, etc.
- Screenshot/Video capture: not built yet. pretty self explanatory.
Great keyboard controls are a hard requirement for me. It's a little tricky since content in web views can capture this too, so I have a global "nav mode toggle" you enter to move around between panes and the tabs within them. Actively figuring out the correct UX, but I am liking what I have so far.
Toying with the idea of a "tab link" which allows you to store a set of "source" tabs in a view, but create "links" in other views, where the navigation is synced across all instances. Useful if you want to have, say, the Tailwind docs open to a specific page, but have that page shared across different views. For example, if you want to have one view specifically for mobile view work and another for desktop view work, and not have to manually navigate to the same Tailwind docs page in both views.
I'm honestly just using it as I work on another real project, and adding features as I think "hey wouldn't this be nice?" Which is a pretty fun and satisfying process. I don't have it published yet, because I'm not entirely sure if it's worth sharing at the moment, but I feel like I'll discover that along the way here and go from there. Maybe someone here will chime in :P
I just recently decided to replace iterm2 with wezterm when I started moving macbook over to nix. iterm2 is about the only one that didn’t work well for this, since you can’t source control the configuration (import/export doesn’t cut it)
Any of the ones you mentioned would probably work good with nix too. I don’t really care about the config being scriptable at all, it was just the first terminal that easily let me set all of the keyboard shortcuts I wanted, so I stuck with it.
iTerm2 does support source control; I've got my settings in a git repo managed by Chezmoi. In the settings dialog, under "General" -> "Settings", there's an "External settings: Load settings from a custom folder or URL" option.
I've also gone wez from iterm2, and you hit upon why. Don't ever make me click on things. I can't script => modify => export clicking on things. When you make me click on things, you've defined my interface for me, which I did not ask for.
Almost entirely with you on that, actually. But OS and other environment differences frequently demand some sort of tweaking, which I absolutely do not want to do by hand if I've done it before.
It's about time I started using a dotfile manager that I didn't make entirely myself - thanks for the recommendation.
For more and more of the cross-platform headaches, I've actually found myself treating the OS as more of a virtual host, and spending the plurality of my time configuring layers that run in it (modify .zshrc where it can do the work of iterm/wezterm, if it can be done in .emacs then do it there).
I get the feeling that I'm not far off shipping personal nix containers around, but there's still a little too much friction between having containers work and working on the OS itself.
Completely agree, I hate the “hackathon” for so many reasons, guess I’ll vent here too. All of this from the perspective of one frustrated software engineer in web tech.
First of all, if you want innovation, why are you forcing it into a single week? You very likely have smart people with very good ideas, but they’re held back by your number-driven bullshit. These orgs actively kill innovation by reducing talent to quantifiable rows of data.
A product hobbled together from shit prototype code very obviously stands out. It has various pages that don’t quite look/work the same, Cross-functional things that “work everywhere else” don’t in some parts.
It rewards only the people who make good presentations, or pick the “current hype thing” to work on. Occasionally something good that addresses real problems is at least mentioned but the hype thing will always win (if judged by your SLT)
Shame on you if the slop prototype is handed off to some other team than the hackathon presenters. Presenters take all the promotion points, then implementers have to sort out a bunch of bullshit code, very likely being told to just ship the prototype “it works you idiots, I saw it in the demo, just ship it.” Which is so incredibly short sighted.
I think the depressing truth is your executives know it’s all hobbled together bullshit, but that it will sell anyway, so why invest time making it actually good? They all have their golden parachutes, what do they care about the suckers stuck on-call for the house-of-cards they were forced to build, despite possessing the talent to make it stable? All this stupidity happens over and over again, not because it is wise, or even the best way to do this, the truth is just a flaccid “eh, it’ll work though, fuck it, let’s get paid.”
You touched on this but to expand on "numbers driven bullshit" a bit, it seems to me the biggest drag on true innovation is not quantifiability per se but instead how organizations react to e.g. having some quantifiable target. It leaves things like refactoring for maintainability or questioning whether a money-making product could be improved out of reach. I've seen it happen multiple times where these two forces conspire to arrive at the "eh, fuck it" place--like the code is a huge mess and difficult to work on, and the product is "fine" in that it's making revenue although customers constantly complain about it. So instead of building the thing customers actually want in a sustainable way we just... do nothing.
We have to do better than that before congratulating ourselves about all the wonderful "innovation".
Maybe, maybe not, it’s hard to tell from articles like this from OSS projects what is generally going on, especially with corporate work. There is no such rhetoric at $job, but also, the massive AI investment seemingly has yet to shift the needle. If it doesn’t they’ll likely fire a bunch of people again and continue.
Not OP, but this is the part that I take issue with. I want to forget what tools are there and have the LLM figure out on its own which tool to use. Having to remember to add special words to encourage it to use specific tools (required a lot of the time, especially with esoteric tools) is annoying. I’m not saying this renders the whole thing “useless” because it’s good to have some idea of what you’re doing to guide the LLM anyway, but I wish it could do better here.
I've got a project that needs to run a special script and not just "make $target" at the command line in order to build, and with instructions in multiple . MD files, codex w/ gpt-5-high still forgets and runs make blindly which fails and it gets confused annoyingly often.
ooh, it does call make when I ask it to compile, and is able to call a couple other popular tools without having to refer to them by name. if I ask it to resize an image, it'll call imagemagik, or run ffmpeg and I don't need to refer to ffmpeg by name.
so at the end of the day, it seems they are their training data, so better write a popular blog post about your one-off MCP and the tools it exposes, and maybe the next version of the LLM will have your blog post in the training data and will automatically know how to use it without having to be told
I’ll give it a fair go, but how is it not going to have the same problem of _maybe_ using MCP tools? The same problem of trying to add to your prompt “only answer if you are 100% correct”? A skill just sounds like more markdown that is fed into context, but with a cool name that sounds impressive, and some indexing of the defined skills on start (same as MCP tools?)
reply