I think it succumbed to Lost syndrome. Don't get me wrong. I enjoyed the novel despite the issues. But it was definitely created in an episodic SCP way which means rising mystery and fewer and fewer answers. I think that, like Lost, the author worked themselves into a corner they didn't fully understand the way out of. The building mystery works because everyone wants to know what the fuck is going on. But the delayed payoff fails more and more often until the project collapses. There Is No Antimemetics Division isn't unique in this. It describes most Blake Crouch and Peter Clines novels. Lots of interesting mystery buildup with completely unsatisfactory conclusions to them.
> It describes most Blake Crouch and Peter Clines novels.
I vaguely recall liking Dark Matter. I've still got Recursion on a shelf around here somewhere waiting to be read. Still haven't read any Peter Clines. Now I'm not sure I want to. :-)
Absolutely. There are many axis with which to judge books. Some have terrible characters and a mediocre plot, but amazing world building (I'm looking at you Brandon Sanderson). Some have amazing characters and decent plot in a forgettable world. It's rare you see authors good at all the various things they could be good at. I read enough to view books like a lot of folks view TV. Some of it is just filler. It's okay to forget about it. It was just there to distract you from one point in time to another and there's nothing deeper than that.
Both would make a great foundation for this sort of app. OFF is crowdsourced and does include barcode information. I have no idea how robust the dataset is for your geography though. If I were to build something like this for personal use, I'd be looking at a PWA that can leverage the camera for barcode scanning. I'd work with the existing crowd sourced database as well as provide a mechanism for "manual" entry which should just be scanning a barcode and taking a picture of the nutrition information. I've personally built systems like this before and all of these things are well within the capability of most SOTA LLM to build out.
I am not going to even try until it becomes a literal weekend project with zero maintenance and perfect reliability. Maintenance is crucial - I would not want to become the maintainer of two dozen apps that are genuinely useful to me. At 10€ a month the economics are not there. Even researching if the codebar database has perfect coverage in Spain like MFP sounds like more time than I would like to invest. I have a thousand other projects I’d like to build.
I've seen a lot of the issues mentioned in the issue. The attempts to end the session early are particularly annoying. We spend a while iterating on a plan and after every phase of implementation I get some variation of "That's a lot of work for today, should we wrap up?" like it's actively trying to drive sessions to a close. I wouldn't say it's useless for these tasks. But it's requiring more effort and guidance than it used to. It's also more likely to jump right into changes from a question I ask rather than addressing the question which is very annoying.
My temperature tolerance has varied quite drastically over my lifetime. Born in the south, raised mostly in Iowa as one of those kids who was wearing shorts as soon as the temp was above 20 degrees. To a pathetic SoCal resident who reaches for a jacket or hoodie when it hits 65 degrees. You absolutely do adapt and rather quickly.
I should probably drop to that plan. I'm averaging around $800/mo in token usage based on ccusage, but I never hit plan limits and am told to wait. I've used it quite extensively this week with a lot of changes to local infrastructure, but still showing 0% utilization across current and weekly sessions according to /usage.
They do make a difference. Statements like that put Claude into different "effort levels". I believe the latest UI update shows the thinking level it's operating at. At one point telling CC to "think really hard" had the same effect as using the #ultrathink prompt. I mostly use the clearly defined prompt instead of more vague statements, but still sometimes tell it to "research deeply" and it does make a difference in how many external sources it checks. The official way to do this to use the /effort command but that changes the session effort level, not for a specific request the way the other prompts do.
I tend to have multiple sessions running when doing stuff with CC. But rarely in the same space / domain. I'll have one working on some programming task, another making changes to my NAS and another doing research for some other topic (most recently networking gear upgrade paths).
reply