So let me get this straight (Pretend I'm 50, a developer since childhood, but I can't CSS to save my life) are there shops where developers, even front end developers, have to talk to designers who are't just sketching an idea for a logo or landing page, but designers who run this Figma thing and maintain the entire products "design" in some "style database"? And the idea is that these designers - who aren't developers - should be able to tweak the look of things without changing code? Or is it usually just the front end devs that run this Figma thing, but they dislike the disconnect between it and their code?

lol yes. At least in agency world, a common approach in the last X years has been that designers create entire pixel-perfect, component-based sources-of-truth in Figma (which evolve! they aren't delivered static and complete) -- these are also what the client sees and approves, or at the very least they see branded deck slides that incorporate the Figma designs. Anyways, front end then re-implements from Figma into CSS, except it's usually best-approximation (not pixel-perfect) partially because, despite Figma allowing you to "copy CSS" for an element, it's unusable, almost inline CSS (and usually not aware of its ascendents and descendents, or any variables you're maintaining in CSS, or any class hierarchies, etc), and partially because the units of measurement aren't always identical on either side. You'll also often have multiple FE devs recreating components independently of each other (as a team effort), which can lead to drift and different implementations, which is fun. Then, depending upon the tech stack, FE might be building these components in something like Storybook [0] as a "front end source of truth", which then are either directly injected into a React or NextJS app or whatever, or sometimes they're partially or fully re-implemented again into BE components in the CMS (ex. Sitefinity). Then people ask which one is the source of truth, but really it's a chain of sources of truth that looks more like the telephone game than a canonical "brand bible". Then throw in any out-of-the-box future client efforts (say, a promotional landing page hosted outside of the main project) and you may have yet another reimplementation of part of the same design, but in a completely different system.

[0] https://storybook.js.org

I've directly experienced this and it is roughly as sane and effective as it sounds.

This is exactly my experience of working in an agency. Made worse by Figma defaulting to 1440px so every design only really works at that width.

We only have the Figma and Storybook layers (product not agency) but these two comments paint an accurate picture of the absurdity. Thank you!

Yah you get this inner platform effect where designers start unwittingly creating their own version of css using Figma and it gets really bespoke really fast.

Why don’t we just teach the designers code / the coders design? This feels very Programmer-Analyst split

Just teach backend engineers to do frontend pixel perfect CSS!

Different fields.

Design management will say that they don’t want to limit the designers’ creativity.

Don’t hire anyone that is a front end designer and doesn’t implement their own CSS. This applied 15 years ago when it was photoshop people faking design skills.

@kevinsync's answer is 100% correct and it's been this way for the last ~~~20? years? at least - only it was "Photoshop files hold the (design) truth" before - now it's figma.

But yes, the "design to code" gap has always been where designers' intentions were butchered and/or where frontend developers would discover/have to deal with designs that didn't take into account that some strings need more space, or what to do when there are more or less elements in a component, how things should scroll in real life, how things should react to a variety of screen sizes, etc.

this short meme video is funny/not funny because it hits too close to home - https://www.youtube.com/shorts/r6JXc4zfWw4 - but yes, "designers don't code and developers don't design", roughly speaking

and then of course you meet some people who do both really well... but they are pretty rare. :-)

It's a real shame that people bought into this false dichotomy, because the base reality is that people who work in web dev that stubbornly pick either code or layout are more of a liability than an asset.

I don't believe that people who can design and code are as rare as folks seem to believe, either. What seems more likely is that there are a LOT of coders who are extremely fluent in CSS but aren't particularly gifted when it comes to making things look good.

It wasn't that long ago that designers understood that they couldn't just hand off a 2D comp of what they want to see. The job isn't done until the output can be integrated into the app. Nobody gets to launch cows over the wall and go for lunch.

It’s just one more example of people realizing that the code is the spec.

Nah yeah Photoshop .PSD's were totally normal for website designs. I got extremely proficient at building functioning websites based on PSD files, going back as far as the days of using nested <table> structures with 1x1 transparent spacer.gif images :) I built hundreds of websites from .PSD files, and Fireworks was pretty much non-existent in my experience.

Those days aren’t over. HN uses nested tables with 1x1 transparent spacer.gif images.

I like that HN’s design is timeless. It’s a shame people usually ignore it, or are critical of it.

The text is barely legible, and you can hardly ever click on the actual target you want (e.g. upvote vs downvote).

It's timeless in that it was never good to begin with, it is just actively malicious now for anyone using it from a touch interface.

I was in web agencies since like 2002-2015, always got PSDs from either clients or internal designers

100%. I was always told to slice the PSD, fireworks never entered the conversation in the agency world I was a part of.

Sure, and also Illustrator sometimes, and Photoshop at other times. Some of the designers I know (very famous for their ui/web work) never touched vector components and just had a ton of layers in Photoshop and air/paintbrushed everything. Hence the meme...

I've even received designs in PowerPoint

Everyone used whatever they were familiar with regardless of the purpose of the application.

I've seen the front end devs get PSD files as recently as 2021.

Many designers stuck with Photoshop sadly. Back when I did agency work it was absolutely normal to get PSDs of mockups.

I think his point was made regardless of his mistake

I wouldn't even classify it as a mistake, just a difference in experience regardless of what Adobe's intentions were.

As someone who has done front-end development for both web and mobile devices for a very long time in the pre-Figma days I was handed a lot more designs that were mocked up in Photoshop than Fireworks.

Yes, Figma is the de facto standard for how designers hand off a UI design to developers in large (and some small) companies.

It's kind of horrible, but I guess it's better than previous alternatives. But, it's not better than a tool that works with code directly and mostly automates away the tedium or translating a visual design into code. I haven't tried Claude Design, but I know I don't find Figma enjoyable (but I'm not much of a designer...I'm more comfortable with code than with pages and pages of options in a GUI).

I'm not sure about "without changing code" but I have definitely seen the believe that Figma represents something authoritative about the product instead of, say, the product being authoritative for itself.

Perhaps because I have a similar bio to yours, I am allergic to this view.

Replace the word figma with the word “specification” and it seems pretty reasonable to me. Same thing. It’s not a belief, it’s a process.

In a bigger company, yes, as it's required to make sure every engineer implements everything in the same way without differences in style over time.

UX designers I encountered have mostly been tasked on ensuring consistency across the various product (A lot of devs are very cavalier about spacing and font sizes). Sometimes they proposed new flows and layouts, especially when the product needs a coat of paint.

So tools like Figma is nice in that regards as it's simpler to iterate on (From simple to hardest: Sketch on whiteboard|paper, Wireframe tools like Balsimiq, Figma|Sketch, css code) because it's pure fiddling with various properties. Figma has direct feedback while the code may require a compilation phase.

That lens (Sigma Cine 18-35/F2) is a big lens, but it looks almost too big there, like it was composited in, or the perspective is somehow strange.

Let's say I have the canonical example of a stack from main via a backend-pr and a frontend-pr. When my stack is done I send it for review to one frontend reviewer and one backend reviewer.

Usually when you develop a "full stack" thing you continuously massage the backend into place while developing frontend stuff. If you have 10 commits for frontend and 10 for backend, they might start with 5 for backend, then 5 commits to each branch to iron out the interface and communication, and finally 5 commits on the frontend. Let's call these commits B1 through B10 and F1 through F10. Initially I have a backend branch based on main wuth commits B1 through B5.

Then I have a frontend branch based on B5 with commits F1 through F5. But now I need to adjust the backend again and I make change B6. Now I need to rebase my frontend branch to sit on B6? And then I make F6 there (And so on)?

And wouldn't this separation normally be obvious e.g. by paths? If I have a regular non-stack PR with 20 commits and 50 changed files, then 25 files will be in /backend and 25 in /frontend.

Sure, the reviewers who only review /frontend/* might now see half the commits being empty of relevant changes. But is that so bad?

> If you have 10 commits for frontend and 10 for backend

In this model, you tend to want to amend, rather than add more commits. And so:

> they might start with 5 for backend, then 5 commits to each branch to iron out the interface and communication,

You don't add more commits here, you modify the commits in your stack instead.

> Now I need to rebase my frontend branch to sit on B6?

Yes, when you change something lower in the stack, the things on top need to be rebased. Because your forge understands that they're stacked, it can do this for you. And if there's conflicts, let you know that you need to resolve them, of course.

But in general, because you are amending the commits in the stack rather than adding to it, you don't need to move anything around.

> And wouldn't this separation normally be obvious e.g. by paths?

In the simplest case, sure. But for more complex work, that might not be the case. Furthermore, you said you have five commits for each; within those sets of five, this separation won't exist.

I also kind of wish Rust had a two-tier system where you could opt into (or out of) the full thing.

The lighter version he describes would make some assumptions that reduce not just the mental overhead but importantly the _syntax_ overhead. You cant have Arc<dyn Animal> if what you describe is an Animal and the readability is infinitely more important than performance. But if you could create a system where in the lighter Swift/C#-like syntax things are automatically cloned/boxed while in the heavier Rust they aren’t, then maybe you could use Rust everywhere without a syntax tax.

What’s the current state of it in terms of stability, performance, features?

When you use LLMs with APIs I at least see the history as a json list of entries, each being tagged as coming from the user, the LLM or being a system prompt.

So presumably (if we assume there isn't a bug where the sources are ignored in the cli app) then the problem is that encoding this state for the LLM isn' reliable. I.e. it get's what is effectively

LLM said: thing A User said: thing B

And it still manages to blur that somehow?

Someone correct me if I'm wrong, but an LLM does not interpret structured content like JSON. Everything is fed into the machine as tokens, even JSON. So your structure that says "human says foo" and "computer says bar" is not deterministically interpreted by the LLM as logical statements but as a sequence of tokens. And when the context contains a LOT of those sequences, especially further "back" in the window then that is where this "confusion" occurs.

I don't think the problem here is about a bug in Claude Code. It's an inherit property of LLMs that context further back in the window has less impact on future tokens.

Like all the other undesirable aspects of LLMs, maybe this gets "fixed" in CC by trying to get the LLM to RAG their own conversation history instead of relying on it recalling who said what from context. But you can never "fix" LLMs being a next token generator... because that is what they are.

I think that’s correct. There seems to be a lot of fundamental limitations that have been “fixed” through a boatload of reinforcement learning.

But that doesn’t make them go away, it just makes them less glaring.

That's exactly my understanding as well. This is, essentially, the LLM hallucinating user messages nested inside its outputs. FWIWI I've seen Gemini do this frequently (especially on long agent loops).

What would be "actual security" in this context?

This isn't about security of the same kind as authentication/encryption etc where security by obscurity is a bad idea. This is an effort where obscurity is almost the only idea there is, and where even a marginal increase in difficulty for tampering/inspecting/exploiting is well worth it.

The one not described as "security through obscurity".

My point is: the "security through obscurity is bad" and "security through obscurity isn't real security" are both incorrect.

They apply to different threats and different contexts. When you have code running in the attackers' system, in normal privilege so they can pick it apart, then obscurity is basically all you have. So the only question to answer is: do you want a quick form of security through obscurity, or do you not? If it delivers tangible benefits that outweigh the costs, then why would you not?

What one is aiming for here is just slowing an annoying down an attacker. Because it's the best you can do.

Somehow your approach was not chosen by Intel ME or AMD PSP, and they remain unbreakable.

That's orthogonal to this. That requires special hardware and using those doesn't really rule this out as an additional measure.

Trusting the messages to contain specific keywords seems optimistic. I don't think I used "emergency" or "hotfix" ever. "Revert" is some times automatically created by some tools (E.g. un-merging a PR).

For the stuff I've worked on, if you want to know about bugfixes and emergency releases, you'd go to Jira where those values are formalized as fields. Someone else in the comments here had a suggestion which just looks for the word "fix" which would definitely capture some bugfix releases, but is more likely to catch fixes that were done during development of a feature.

Yes for a meaningful context you should need both the source repo and the work tracking system. But today most systems have apis (jira, ADO, gh, ...) so this should be fairly doable, especially using a bot like copilot cli. But it's not doable as a little shell script.

We use branch names like `hotfix/some-hotfix-name`, don't think I've seen many commits mention hotfix directly.

That would mean in a way accepting that they are suddenly a service company with the aim to create revenue by selling services to customers for money.

The roadmaps look messy if you look at them as coming from one company. But if you remember that windows and .net (or DevDiv) are more like competing companies then it makes more sense. Then for one side it’s Win32, MFC, WinSDk and for the other it’s Win32 (WinForms), WPF, MAUI

If Microsoft hadn’t been preoccupied with a failed mobile bet then this wouldn’t have happened. It’s a lost decade followed by a (much more successful) cloud pivot. The reason desktop is ignored is because it can be. No one is eating their lunch on desktop.