Came to ask this. I suppose if the edge of the sun glows in a different color than the rest, it would tint the edge of the shadow too? So maybe appropriate for sunsets, where the sky near the sun is red but the sun itself still glows bright white. Honestly just guessing.
Kudos for engaging! It’s been a while since I used Photoshop on a daily basis, but my impression was always that the UI felt a bit stuck in time. Like no one had thought about how to make little things in the UI better in a way that improves daily work. You’d see new ideas crammed in on top, but very little refinement of what was already there. (Is the ”Wind” filter still only possible to apply left or right, not up or down?)
I think a nice outcome of this would be if Adobe recognized how much these things matter to power users, and that it’s possible to improve existing workflows without disrupting them, and without just adding something new that sits awkwardly side by side with the existing features. Maybe rather than fixing the issues that were introduced, you could aim for something that is thoroughly better, as you need to work through everything anyway.
Thank you for the constructive feedback. I agree power users are a "keystone species" and improving their workflows will benefit everyone.
Improving existing workflows without disrupting them is extremely hard to do, and often "improvement" is in the eye of the beholder. To be clear, I am not excusing issues within the application that we must fix. The team is working hard across multiple departments to gain consensus on how best to move Photoshop forward, including gathering feedback from users.
Clearly broken and unfinished modals such as those in the blog post don’t require much more than a couple of devs to fix, and yet this behaviour is still present in the latest version shipped to customers.
I find it hard to believe that the team is “working hard” to gain consensus on how best to move forward when such simple things make it to production.
Does anyone at Adobe ACTUALLY use Photoshop? Didn’t anyone stop for a moment to think that shipping in such a taste was a terrible idea?
Yeah, I understand how difficult working at a company like Adobe can be. But it's still hard for me to sympathize when these dialogs that don't contain much more than a bunch of text fields and buttons are just halfway done and then shipped.
It's not like focusing on the first field when a dialog opens requires months of work. I'm genuinely confused by how this stuff happens; I feel like a regression like this should have been caught in the first PR review and fixed.
Are there OKRs for converting as many dialogs as possible to the new UI library? Or how does that happen?
Yes, it is clear - for every compute unit you have that's more efficient and effective I have 10,000,000,000 compute units. I can use the same models that you do. How would you win?
This says something about why LLMs simultaneously feel like incredible productivity enhancers and at the same time don’t have as clear of an impact on productivity in the bigger picture. It probably does increase productivity, but doesn’t widen the bottlenecks or reduce the friction of tech debt. Instead it pushes more work through the existing bottlenecks, clogging them up even more, and makes code even harder to understand as no person ever necessarily had to understand it in the first place.
”Thinky puzzle games” is a specific subgenre and community, revolving mostly around variations of sokoban, but really has an appetite for any game that deeply explores how a few mechanics can be combined and lead to interesting consequences.
I tried uploading our design system. Claude Design’s environment was so limited it had to reimplement it from scratch in HTML, JS and CSS. Doing that burned through more than half the token limit. Along the way it completely changed it and made up things that don’t fit in at all, neither visually or as code. The output of making a mockup is one huge HTML file with minified CSS that just can’t be used meaningfully for anything.
I guess I had expected something like Claude Code with visual tools added on top, but that’s not what this is.
I imagine lots of games do already, but offline. Most games don’t have gameplay that requires real-time terrain generation. The idea of generating terrain procedurally is not new, but this technique that gets a great-looking erosion effect in real time is.
> But for I, like most millennial Americans
Someone who turns 45 in 2026 is a millennial though.
reply