Hacker News Reader

Comment by badlibrarian | Hacker News Reader

Or maybe you'll have better luck in a week when somebody familiar with RAW and a 20 year grudge against Adobe chooses to spend a day of tokens on the problem. Maybe me.

Rome wasn't built in a day, nor Adobe destroyed in a single git push. But they both shall fall.

View on HN

People are underestimating heavily what has happened here and to our industry in general.

Adobe employee here. We have access to LLMs and tokens as well.

And 500 levels of org tree, and grumpy customers that freak out if you change anything about their workflows, and decades of legacy code across dozens of micro-services, and backwards compatibility requirements with 20 year old formats...

You have:

1. LLMs

2. an executive team that hates your customers. this problem with company culture goes back at least 13 years: https://fstoppers.com/video/adobes-ceo-deftly-dodges-questio...

Unfortunately any benefit from #1 is neutralized by #2

How are LLMs and tokens going to help you retain customers who you've spent the past decade pissing off?

Those customers are already pissed off, yet record sales every quarter. This will be great to offer this diminished experience to them until they start using AI to do all of their editing.

You also have managers actively hostile towards your customers.

That is true, but does Adobe have another Photoshop to build? And in how many hours will that be copied as well?

That's exactly what I was thinking. It's only a matter of time. Surely the models, when well directed, can get most or all of the way there in short order.

Surely the models, when well directed, can get most or all of the way there in short order.

"Hey ChatGPT, what would the sufficiently smart compiler argument look like in the LLM era?"

I did this a few months ago and it's not simple: different sensors have different pixel arrangements, files can have different data structures, and the default tone mapping is provided by the app (so opening the same raw file in darktable vs preview will have different results). There are different algorithms for how you assemble the raw data as well, and the issue is that you can get something _passable_ that isn't _optimal_, and it's really hard for the LLM to recognize that something isn't right. Is it the pixel processing, or is it the tone mapping, or is that just how the image looked?

And on top of that, you need a bunch of different files from different sensors to test with. Canon, sony, apple, fuji, nikon, olympus, etc. -- if you're building a truly general application, that's a concern.

Or if you're like me and you shoot fuji, you just build a fuji raw parser and call it good. The "TAM of 1" world of software development...

Reply on news.ycombinator.com