Agree, I personally also think dark mode never goes well with a map-based app...

Especially with this sort of map regarding this specific city. For example, Beverly Hills is a separate municipality contained entirely within the borders of Los Angeles. It appears here like an empty black space no different from the ocean, lakes, mountains, and parks also scattered around the city. The styling of this map makes it impossible to tell if an area is truly undeveloped and if so, why. Throw in the way there appears to be arbitrary selection criteria for some buildings outside city limits that make them appear as lone buildings in undeveloped space, and I think it adds up to potentially give people a wrong picture of the area.

Yes, good point. The LA City data was rich, but the other communities have data in other places. I will see what I can do.

Great, I'm sure getting more data for a complete picture of the area is a huge undertaking, but simply switching to the light map that shows the geography more clearly is a big improvement.

Use Claude Code with Chrome MCP / Claude in Chrome and it'll happily do it for you, 100%

I think it's an efficiency thing. And at scale, efficiency matters. Whether this is the right implementation idk, but when I see Claude Code use Chrome via MCP, it's incredible to see just how inefficient the status quo is.

I guess it's a more complicated thing to implement (!= low hanging fruit), but I'd be more interested in sleep tracking than step tracking.

Sure, but we can't do sleep tracking well. Todays devices mostly guess and map to typical patterns. Might, might not, work reasonably for you. But chances are that if you want to track your sleep because you suspect irregularities then they are about as useful as a potato.

The Apple Watch is quite accurate (according to Apple's own research, at least), and it doesn't guess or map to typical patterns: https://www.apple.com/health/pdf/Estimating_Sleep_Stages_fro...

From other anectdata the apple watch seems to just as bad as everything else.

We just don't seem to have any idea how to track sleep quality even with better devices than just a watch/whatever. Even a binary, is the person I'm attached to sleeping, is hit and miss. And with that in mind you can be pretty sure that the sleep stages etc is even worse.

The problem with sleep tracking watches is that simply wearing them alters your sleep patterns/quality.

YOU ARE CORRECT.

We should start looking at biochemicals soon - melatonin and cortisol release

To sync activity tracking data with your phone

Nice move! I wish it would support even more activity / health features in the future like GPS tracking, sleep tracking, etc. As a full package in the classic design it would make a great alternative to Garmin / Polar / Whoop / Oura etc.

> most of the feed is filled with useless posts

I, unsurprisingly, am making the same observations. Sometimes I do think about shipping a slim Chrome extension (~Ad Blocker) to classify and filter out useless content. Could make it a bit more bearable and useful.

I use LinkOff https://chromewebstore.google.com/detail/linkoff-filter-and-...

It has a 'gentle mode' so you see what gets blocked.

I think it's thanks to it that I don't feel animosity towards LinkedIn anymore.

I just unfollow everyone on LinkedIn

Me too. Linkedin helpfully keeps finding new people for me to follow and automatically follows them.

Hmm that doesn’t happen to me. But it does suggest content from users I don’t follow

Also been observing "productivity loss" by working on multiple Claude Code sessions in parallel (~5-7) and having to context switch between them multiple times during the day. Sometimes I literally have to ask "My brain is fried, can you catch me up? Keep it concise ..." Other times I force myself to not open another session and rather wait for a few minutes for an ongoing session to finish. Otherwise there's the built-in "recap" summary at the end of each chat, but for whatever reason I rarely read them (maybe I should).

Cloud business is making big dollar I suppose

Huh? But only if you give it a tool it can use - which brings you back to Google (or Bing)

Anyone can leverage common crawl and/or run their own crawler, to do so isn’t novel.

https://commoncrawl.org/

Sure, I was just pointing at the fact that the question "And where does ChatGPT get the data for those answers?" has a two-part answer:

1) training data (common crawl as one example web data source), and

2) live data optionally retrieved at runtime.

My comment was about 2), and that part runs via a search engine (Bing?), at least if you look at how ChatGPT does it.

Indeed, but what is the difference between crawl data and model data but decay rate? Models are trained on previous crawl data, but if an LLM provider engages a search engine to get "live data," that data isn't live but previously crawled as well (and perhaps not yet integrated into models as crawl data).

So, why would you use Google as a tool or search target when you can, in some combination, go direct to the website (or whatever the target data endpoint is) yourself as an LLM provider to retrieve the most recent data or rely on your own "hot cache" of that data that was crawled recently but said data is not stale enough warranting a live web crawl to retrieve and present to the user or AI agent? Is this capability to perform retrieval from a data source in real time not similar to an AI agent?

Broadly speaking, I'm just spitballing on the concept of "You must use a search engine for an LLM to return 'live-ish' results" as I think we're directionally headed to where that isn't the case.

> but if an LLM provider engages a search engine to get "live data," that data isn't live but previously crawled as well

I'm not sure frontier labs do it, but imo fetching live data requires fetching both the search engine (1) and the website/page (2). The search engine gives you search results + content snippets (potentially stale), the website/page gives you the actual content fresh from the source.

The tradeoff I see here is between liveness/staleness and cost (hitting an index if of course cheaper and faster than querying live websites again).

Cloudflare: Give any website a WebMCP interface - https://blog.cloudflare.com/webmcp/