The thing about this is that in a different codebase, even just that search for the trumpet guy could either burn a ton of tokens and come back with nothing or find it but light the ocean on fire in the process. This is because trumpet guy would be named "asset-ds1f23nbiq35421fsda.jpg", and it would be located in an images folder with thousands of other similar assets. So the agent starts by grepping for reasonable names, discovers nothing but finds an image directory, and then searches inside. In the worst case, I guess it starts analyzing each image. Or maybe it checks the article like he mentions, but the codebase uses a component library and the filename thing is still true, so it analyzes every image in the page. This is a bit of hyperbole, of course. Please allow it for the sake of comedic example.
Many of his other prompt examples also illustrate this. The author is massively leveraging the architecture of the organization through familiarity. It could well be the case that even an LLM was used to set that up, but I doubt it happened in one prompt or without critical thought. To be fair, I guess he's sort of trying to point that out, but by saying "Look how lazy I am while it still works" or something like that in order to look cool, I think all he's conveying is "Look how easy this is", when it's potentially not and could increasingly cost him/his company wildly varying amounts of money based on how he goes about it. When tokens were virtually free, no one cared. This is increasingly not true. Many are drawing comparisons to early cloud architecture.
I find this is sort of similar to the problematic idea that in theory anyone can clone your repo and get a local dev environment running in a few seconds. So then you try it with a non-technical staff member and quickly realize how unrealistic that is as you run into abstraction after abstraction that is completely foreign to them.
Why is that problematic in the age of AI? Yes, previously, me being totally comfortable at a terminal where cd and ls are reflexive for me, working with someone non-technical, to get them to open a terminal and move around to type out git clone in order just to get started, was definitely a realization of a lot of things.
That was the past though. I've seen people who were good at computers but didn't know how to code, make some good programs where the AI wrote the code. Yeah, turns out tokens cost money, but they're still pretty cheap.
I really dislike how the dribbble/envato style somehow escaped containment into Claude and now I’m seeing it everywhere mixed with mediocre design layouts. I think to myself that maybe stuff like this will help counteract that but then I see it in use with 12px typography (the kubernetes 3D dashboard thing) and I recoil in horror. Please consider this a PSA: Claude doesn’t understand UX or good design principles out of the box. Please go hug a designer.
UI greebling [0] is something I'm not looking forward to.
People always miss out the fact that older information-dense UIs were... information dense. We get this weird simulacrum of information density where the LLM has slapped on "every pod has a machine", a made-up PDU for pods with a watt rating somehow, and random status labels EVERYWHERE. All it's missing is the green circle "LIVE" label that every claude app seems to have...
Wow, greebling is a good metaphor for it that I never would have thought of.
> "every pod has a machine"
Everything you mentioned is annoying, but I think this one is the worst. I don't know why it loves slapping on meaningless info blurbs like that. I guess it carries over from training on test/demo programs where the functionality is clearly spelled out, but on a production UI it looks so bad.
I had no idea “eyebrows” was a term until I started using Claude to generate designs. That model is eyebrow crazy, every section of the output has them.
Even comprehensive design skills like impeccable miss filtering them out (it only detects “eyebrow chips” and not the prolific barrage of basic eyebrows.)
It will not take superior models or LLM processes to do skeuomorphism again. It was always possible to do skeuomorphic design, it just took hard work and a focussed design effort. Just as it’s now easier than ever to do any sort of UI/UX design now with LLMs, the missing link is the work, the effort. When you want to create a skeuomorphic design, don’t defer the hardest parts to the LLM, you should investigate for yourself the intentionality and thoughtfulness of each element you create. That’s what made the skeuomorphic design of yore so attractive, it was thoughtfully designed, down to the details.
This! That was, IMO, a golden era for UI design. We've lost SOO much to the wild west experimentation brought on by the advent of touch interfaces. A really strong shared design language across apps and vendors is an unbelievably valuable productivity enhancer that lets people apply the intuition they've learned in one app to a UI in a different app that they've never seen before. When everyone tries to make their app look and feel and operate with uniqueness and visual distinction, much is lost.
Consistency in the interface allows users to transfer their knowledge and skills from one application to another. Use the standard elements of the Aqua interface to ensure consistency within your application and to benefit from consistency across applications. Ask yourself the following questions when thinking about consistency in your product:
- Is it consistent with Mac OS X standards? For example, does the application use the reserved and recommended keyboard equivalents (see "Keyboard Shortcuts Quick Reference" (page 309)) for their correct purposes? Is it Aqua-compliant? Does it use the solutions to standard tasks Mac OS X provides? (For more information on these solutions, see "Using Mac OS X Technologies" (page 61).)
- Is it consistent within itself? Does it use consistent terminology for labels and features? Do icons mean the same thing every time they are used? Are concepts presented in similar ways across all modules? Are similar controls and other user interface elements located in similar places in windows and dialogs?
- Is it consistent with earlier versions of the product? Have the terms and meanings remained the same between releases? Are the fundamental concepts essentially unchanged?
- Is it consistent with people’s expectations? Does it meet the needs of the user without extraneous features? Does it conform to the user’s mental model? (For more information on this concept, see "Reflect the User’s Mental Model" (page 40).)
Meeting everyone’s expectations is the most difficult kind of consistency to achieve, especially if your product is likely to be used by an audience with a wide range of expertise. You can address this problem by carefully weighing the consistency issues in the context of your target audience and their needs. See "Know Your Audience " (page 25) for more information on how to define your audience
You were filtering the sidebar for "consistency", that's not a search box.
From 2026 HIG:
> Create a list of common terms, and reference that list to keep your language consistent. Consistent language, along with a voice that reflects your app’s values, helps everything feel more cohesive.
> Build language patterns. Consistency builds familiarity, helping your app feel cohesive, intuitive, and thoughtfully designed. It also makes writing for your app easier, as you can return to these patterns again and again.
Oh you're right, well in that case I'd like to complain about turning the big PDF into a linked web document and not giving it an obvious whole-HIG search box. Worse experience than cmd-F 20 years ago.
What's a good example that comes to mind for you of it done right? I don't do much frontend so I think I just uncritically accept whatever design is put in front of me.
I can’t teach someone design in a day, but I can elaborate on the reference and point out some of the worst habits. A lot of these designs feature ornamentation without purpose. Look critically at the content in a design it produces. Ask yourself if each piece of information is serving a purpose for the end-user. This used to be, at least a little, harder to get wrong when adding noise came at the cost of more effort. The other thing is most often just typography. For a number of reasons, text below 16px should be treated very seriously. Claude throws it around everywhere.
Actually if I must be literal it’s a pretty big part of the culture out there. The playa is rife is with inside jokes that lately revolve around people telling you to have a terrible time, or saying there’s absolutely nothing at some place with a wink. Discovery and resilience is the point.
And sure, there’s a bit of old guard/new guard here, where the new guard is a bit softer about it. Personally, it was never my favorite aspect, but dismissing it says more about you than me.
I didn't say anything about alleged. I said the "apparent facts". I chose the phrase "apparent facts" because prosecutors and police outright lie or, at best, misstate the facts about what happened, such as when ICE agents claimed that the people they shot were driving their cars at them.
Whenever I point out (in agreement with your statement) that the idea that the left and the right are somehow equally bad is a kind of fantasy on HN I usually get downvoted. People seem very committed to "shades of gray", moderate thinking to the point that it's viewed as always correct, which is well, just wrong. Yeah, often things are more complicated than they appear, but I find it almost hilarious how often you can't do that in American politics. And for the record I'm definitely not saying the left is somehow perfect. Obviously the left is a bit of a mess too. It's just that they're not the same kinds of messes across party lines.
> People seem very committed to "shades of gray", moderate thinking to the point that it's viewed as always correct, which is well, just wrong.
> And for the record I'm definitely not saying the left is somehow perfect. Obviously the left is a bit of a mess too. It's just that they're not the same kinds of messes across party lines.
Seems like you've fallen to the same fallacy within the same paragraph, which is quite ironic.
Showing my age, about 35 years ago, someone (wish I remember who, but google probably knows), wrote an article in Analog magazine called "The Dogma of Otherness" that spoke to the tendency of a certain demographic to always try to understand different perspectives, even when those perspectives were objectively hostile.
Some people just can't not think in shades of gray.
the thing is that the "left" in the US (and then on HN) have almost no political representation. The Democrats aren't left, just the leftest available. They aren't better than the Republicans at all.
This is exactly the kind of thing the parent poster was talking about.
The Democrats certainly aren't left, and they're not great, but we live in a world where the Republican party ranges between refusing to condemn and openly supporting white nationalism, gloats about making trans people's lives worse, encourages police to use unnecessary force, etc. even before we get in to the impacts of their economic policies.
Despite being a massive disappointment as far as actually making positive changes, one is still obviously miles better than the other because they're not trying to drive the car the wrong way as fast as possible. Claiming that the two major US parties are no different requires willfully ignoring a lot of bad stuff that one party does for which there is no comparison on the other side. There are absolutely specific issues on which there is often no meaningful difference but as a whole there is a universe of difference.
This is true. Here’s a cautionary tale to that end. My org’s IT recently introduced, without input, an ai filtering tool to automatically filter junk email. It apparently worked too well. In an ironic twist of fate, it filtered a raft of emails (across multiple domains, including tickets and account reps!) to IT from an important vendor about contract renewal into junk that ultimately resulted in leaving the vendor no choice but to suspend service until the bill was paid (they waited until the very large bill was over two months past due to do this, to their credit).
When I learned about this during the post-mortem I was incensed, to say the least. Email mgmt is part of a professional’s job, for better or worse. Accountability for that doesn’t suddenly evaporate because of a scenario like this. I don’t practice zero inbox for fun. I do it for my sanity and effectiveness. Technology can’t magically solve every problem.
> Letting mega-billionaires spend mega-millions to buy the conversation around elections isn't my definition of freedom.
Spell out the details: Exactly who will not be permitted to speak under your definition of "freedom"?
If you say that only people with a billion dollars can be punished for speaking: Do you think the limit won't be lowered over time?
If the issue is that no person should be able to spend more than N million dollars in one "conversation", how do you allocate broadcast licenses when the price in an auction would be more than N million?
America wasn't a democracy before the 2010 Citizens United ruling allowed corporations and wealthy donors to spend unlimited money on elections? That's news to me.
I don't feel more free at all on account of that ruling. Maybe if you said "freedom" one or two more times, that would do the trick.
"The political action committee emerged from the labor movement of 1943. The first PAC was the CIO-PAC, formed in July 1943 under CIO president Philip Murray and headed by Sidney Hillman. It was established after the U.S. Congress prohibited unions from giving direct contributions to political candidates."
The above from Wikipedia.
But I'm wasting my time talking to you, you are just raging and putting words into other people's mouths, because you are too high on your own supply of resentment to want to have a good discussion.
I'll put my opinion clearly: Everybody should have the freedom of association and be able to get together with likeminded to work for any political cause they want to advance, as long as they do it within the limits of just law.
Political donations by large groups of special interests aren't unique to America. Many countries have very strong labour union super-PACs.
Given that I just read this comment after getting off a connecting flight that had me waiting on a fully boarded plane in 100 degree Phoenix weather for half an hour, it’s pretty poignant. I read an article earlier today about how Phoenix may not be inhabitable in twenty years, and I’m here to tell you that it’s already insufferable no matter how well the A/C was or was not working on that plane.
I don’t need a plane that flies faster. I need an airline that puts pure service before profit.
Many of his other prompt examples also illustrate this. The author is massively leveraging the architecture of the organization through familiarity. It could well be the case that even an LLM was used to set that up, but I doubt it happened in one prompt or without critical thought. To be fair, I guess he's sort of trying to point that out, but by saying "Look how lazy I am while it still works" or something like that in order to look cool, I think all he's conveying is "Look how easy this is", when it's potentially not and could increasingly cost him/his company wildly varying amounts of money based on how he goes about it. When tokens were virtually free, no one cared. This is increasingly not true. Many are drawing comparisons to early cloud architecture.
I find this is sort of similar to the problematic idea that in theory anyone can clone your repo and get a local dev environment running in a few seconds. So then you try it with a non-technical staff member and quickly realize how unrealistic that is as you run into abstraction after abstraction that is completely foreign to them.
reply