dewey

I don't think GitHub stars are what makes GitHub sticky. Having a single UI / account and being able to jump between work projects, private projects and random open source projects where I want to open a quick issue or submit a PR is much more sticky.

There's basically no moat for OpenRouter, you can very easily switch them out to any other of these AI gateway companies.

I guess it's pretty useful to see the payment volumes of all these startups and then "inhouse" the fees as extra margin, especially if it's a relatively thin wrapper like OpenRouter.

This is the way! It's great to just point frontend developers to the Buf documentation for new features and they easily have all the types and generated clients available in their projects too.

Sure, but having a generic LSP that works with all editors vs. an editor specific integration is still a very different thing.

It's a bit odd for the parent commenter to say "Weird post" and implying that having an editor specific feature would somehow make having an LSP that works for everyone be any less interesting.

Miniflux also does that, you just have to enable it by feed or by default.

> Fetches the original article and extracts only the relevant content using a local Readability parser.

How? Seems like just a fork without PG won’t won’t fix buggy feeds. I already have Postgres running on my server so the DB isn’t a selling point for me.

I’m also a happy user of Miniflux and that specific feature. I still sometimes want to click out to the article easily to double check if it feels like something didn’t parse correctly. Rare but happens.

The advantage is that I don’t have to use Telegram and Pushover can just be triggered by curl in any scripts I use.

In my regular RSS reader I often have the problem that some feeds are not full feeds, or some images are missing (In case they have some gallery) and I have to click out to the page, even just to verify I got the full text. That always kinda stops me from going the eink route where I don’t have a proper browser.

Depends on which feeds you subscribe to I guess, probably works really well for long form, reliable feeds.

Not Open Source but I’ve been using https://pushover.net/ for this use case since forever. The website still looks the same as on the day I signed up for it probably more than 10 years ago. Never had an issue and using that in all my scripts or cron jobs.

For me it’s about building a routine. I go to a coffee shop every morning with my kindle and phone. Phone in my pocket on do not disturb and just read for an hour.

It’s not that hard to disconnect if you really want to.

Ah you are right, it's mentioned on the career page. I thought he might be an advisor there.

There's many competitors these days, one that Shkreli always seems to use in his streams is https://godelterminal.com (Not open source though).

What’s sensitive about displaying public information in pretty graphs?

I think Kagi Small Web filter would give you very similar results.

Agreed! It's one of these apps (Similar to Flightradar) that just feel like a great use of technology and add a little bit of joy.

Just because they started out as a code editor doesn't mean it has to stay that way. I was looking forward to a more modern Sublime Text in Zed but I also know that nobody is going to make money off a simple code editor.

> This thing is now just a white-label version of Visual Code.

That's a wild hot take, and pretty ignorant to the people working on that for years.

For Google it's pretty straight forward to throw away fake crawlers by just only allowing their published list of crawler IPs so you don't accidentally allow someone from a random GCP IP to crawl you if unwanted (https://developers.google.com/crawling/docs/crawlers-fetcher...).

My problem is usually that I bookmark in the app, but then forget about the bookmarks as they are in the walled garden. I've built something for that a few years ago (https://getbirdfeeder.com) that just sends me my Bluesky (And X, Mastodon etc.) bookmarks straight to my email inbox once a month.

That's not true. If your ad placements don't perform, people are not going to book them. If ads are loaded or not is also often verified by independent third party trackers that you have to embed on your page next to the ad. There's whole industries built on this stuff and it's not as simple as "we sell the slot and don't care if people see it or click on it".

It's still good etiquette to disclose affiliation in online discussions.

You forgot the "Yes, that's my company" part in your reply (https://ghking.co)

That's why engineers are usually bad marketers (And I'm including myself there) and their technically sophisticated side project goes nowhere because they don't know where their potential clients hang out or they think doing marketing is "cringe".

If you are the best consultant but you don't tell anyone about it, you'll not have many clients. In this case it's not about what you think is the best and most pure platform, but where your potential customers are hanging out.

More, as you have the advertising identifiers from Apple / Google on mobile and you are also not able to block tracking scripts / tracking ads like on mobile web (At least not without more effort or installing additional software or a vpn).

It's never about the technology or if it's theoretically feasible but that tracking, push notifications and ad blocking are much better handled in mobile apps for most platforms.

For a social media site on mobile, yes.

Not having the app installed but being a monthly active user, not just someone who's driving by because they clicked on a link on mobile seems like an edge case and I'm pretty sure it won't make a difference to the numbers.

I guess you didn't really read what I wrote, buddy.

You are basically just describing targeted advertising and it's already working like that, just not by email and probably more accurately than people putting themselves into buckets.