Is there a way to use your Claude Max plan? I checked my token usage (ccusage) if I wasn't on a plan, and last month it would have been over $2000. This constraint prevents me from realistically considering alternatives to Claude Code.
The problem is the training corpus tends towards mediocre code. But with an agentic loop that analyzes the code against those criteria and suggests changes then I think it might be possible. I wouldn't try to get it to generate that right off the bat.
I wonder how many of these rules can be incorporated into a linter or be evaluated by an LLM in an agentic feedback loop. It would be nice to encourage code to be more like this.
I notice you didn't really talk much about types. When I think of proofs in code my mind goes straight to types because they literally are proofs. Richer typed languages move even more in that direction.
One caveat I would add is it is not always desirable to be forced to think through every little scenario and detail. Sometimes it's better to ship faster than write 100% sound code.
And as an industry, as much as I hate it, the preference is for languages and code that is extremely imperative and brittle. Very few companies want to be writing code in Idris or Haskell on a daily basis.
Yes, ADHD definitely seems to correlate with high openness and low conscientiousness. I've also found most self-help and productivity advice to be useless.
I've seen DISC and Kolbe used at companies I have been at. I've learned that I have a low tolerance for structure and work best on large projects with lots of unknowns and autonomy. Explains why I've always preferred startups over larger companies. Seems like few companies can accommodate ADHD people. It's a shame because once I'm focused and interested in a project I do incredible work. But the daily standup, pick something off the queue, weekly sprints don't seem to be the environment I thrive under. Ask me to build a big system or service and I can come back in a month and deliver something incredible.
Also, I really wish social media platforms provided better tools and didn't have policies that penalized you for deleting followers that are bots or junk followers.
As a Las Vegas photographer that works primarily with models, I often have random profiles blasting out my work. These profiles mostly find sexy content and blast it out in hopes of growing their own profiles. This mostly resulted in my followers being 95% men from outside the US. This does absolutely nothing for increasing my engagement with my actual target audience (female models or would be models in the Las Vegas metro area wanting to book photoshoots).
Unfortunately Instagram penalizes you and has actually removed the search functionality from my follower list because I was using it to delete bots and junk followers. They won't say this officially but their support ignores my requests for why this functionality no longer works.
A lot of these social media promotions work by having people with high follower counts blast you out and try to get their followers to follow you.
The problem is that it is not an audience that would normally be interested in or engage in your content naturally. There are often artificial incentives to follow or engage in someone's content. Often there is some kind of prize giveaway from a "celebrity", that you have to follow everyone on a list to qualify. That celebrity then gets paid to blast out the promotion.
Then after the promotion all of a sudden your massive number of new followers aren't engaging with your content anymore. What are the algorithms going to assume now? Naturally that your content is no longer any good.
It's common for influencers to share screenshots of their analytics or publish them on their websites for people looking for influencers. While the numbers might look impressive, unfortunately, due to how the algorithms work -- mainly things like vector embeddings and placing influencers in a some high dimensional space, the algorithms no longer target and recommend your content to an audience that would be interested.
It used to be that brands would look at your follower count and see how many likes / comments you were getting, but even this is faked now. As your engagement (likes / comments as a percentage of your followers) goes down, they are sometimes artificially propped up by purchasing likes and comments. This worsens your engagement and leads to an endless downward cycle.
While someone might survive for a short while as an influencer using these black hat strategies, brands will be unlikely to use you again if they have not seen tangible results.
Also, if you intend to sell a product or have a certain ideal customer avatar you are trying to market to, it makes sense to do as much as you can to get engagement from that (and only that) demographic.
Follower counts might look impressive on the surface but what ultimately matters is whether you see conversions for your business / brand.
I've worked at Silicon Valley startups that were that way. I remember thinking this is really odd how high the concentration was (probably about 70% Indian). Didn't really bother me because it was a great team and honestly one of the best companies I worked at.
It didn't really seem like they discriminated in hiring and I never felt discriminated against at work. It was just that coworkers referred people they already knew.
- The programmer's tool should be a tool for manipulating an annotated AST (not text)
- There should be many different types of UX's for different scenarios, each maps to and from an AST in a UX that is optimal for the developer for that scenario
- We must be conscious of human brain limitations and cognitive psychology and work within those constraints
- "Reading" and "Writing" code should have different UX's because they are radically different use cases
- Use RPN. It models the real world. Humans are designed to manipulate their environment in an incremental manner seeing the result each step of the way. When we have to plan out and write code for an extended period of time, trying to play compiler in our head, we overload our brain unnecessarily and highly likely to make simple mistakes.
- Testing should be a first class citizen in the developer experience and indeed baked into how we develop at a fundamental level that it seems strange that they are even decoupled to begin with.
I think they may stick around because it has the user base and why not, they are already making money so no point giving that up.
But now models will diversify across multiple platforms because the writing is on the wall. They may lose their income at any point so they are going to have backups.
The damage is already done. I was doing a photoshoot at my house with 2 models that have OnlyFans accounts. When the news hit they literally signed up for other platforms within an hour.
The other platforms are capitalizing on the news and finding ways to make migration to their platform as easy as possible.
Used to be the case. They are far behind now. Just read the docs for CCBill. Most of adult is still PHP even for new stuff. And they usually work on the servers directly instead of using version control and staging servers. Only the really bigger adult companies adopt a modicum of modern programming practices.
I highly doubt they are going through CCBill. They are too expensive at volume. Since they are only charging models 20% I suspect they tried to pretend to not be adult, which they probably weren't in the beginning. But not it is almost all adult and CC processors know that and want their higher cut.