kschaul's comments

On: Feedmaker: URL + CSS selectors = RSS feed

[kschaul]

Glad you’re find the tool interesting! A short blog post behind it: https://kschaul.com/post/2023/04/16/feedmaker-quickly-genera...

And the GitHub url (hopefully easy to host your own instance): https://github.com/kevinschaul/feedmaker

[mustaphah → kschaul]

Looks like you're hosting this on fly.io - PAYG model. You could probably host this for free on Cloudflare Workers; 100k requests/day on the free tier; static content (the homepage) is free & unlimited.

Edit: The catch is the 10ms CPU cap per request - you'd need a super lean implementation. Django's too heavy for that.

[mustaphah → mustaphah]

Well, someone already did with JS: https://github.com/ProfessorManhattan/rss-worker

[0cf8612b2e1e → mustaphah]

Python alone is many milliseconds to start. Unless they give you some allowances for interpreter overhead.

On: I've been advocating for RSS support, and you should too

[kschaul]

I love RSS, but I often want to create a feed from websites that don’t offer RSS support (or the right features). So I built a little web service that, given a URL and a CSS selector, returns an RSS feed of those items. Figured it might be useful for others too. Code -> https://github.com/kevinschaul/feedmaker Hosted version -> https://feedmaker.fly.dev/

On: I ditched the algorithm for RSS

[kschaul]

I love RSS, but I often want to create a feed from websites that don’t offer RSS support (or the right features). So I built a little web service that, given a URL and a CSS selector, returns an RSS feed of those items. Figured it might be useful for others too.

Code -> https://github.com/kevinschaul/feedmaker Hosted version -> https://feedmaker.fly.dev/

On: Your Name in Landsat

[kschaul]

How did they locate these particular images? I would love to read something on that, or similar projects looking across satellite imagery for specific patterns.

[indeed30 → kschaul]

Easiest way is going to be to downsample the images and then apply a pre-trained classifier that can ignore the fact these are sat images. You could probably turn them into 28x28 greyscale and then use a model trained on handwritten characters, like EMNIST.

Whatever approach you take, you'll probably be selecting the final set by hand, so it's just about building the candidate set in an efficient manner. Low absolute accuracy isn't really an issue as long as you end up with a managable set to review.

[AnotherGoodName → indeed30]

There simply doesn’t seem to be that many different images so I can imagine an even easier way than an image classifier.

As you work through your day as a satellite image person (I don’t know the job title, sorry) and see a shape that looks like a letter screenshot and save it. You’ll have an alphabet in a very short amount of time.

Given the repeats of letters I’m seeing I think that’s all they’ve done here.