I'm really late to the party, but I wanted to clarify what you experienced with the RSS feed, on why you could get old posts as new. The thing is proxigram grab the information from different sources (providers), not instagram.com directly (at least at the moment of writing this). So, let's say you are subscribing to @instagram feed, and the instance admin has set the maximum number of posts per feed, which is 12, proxigram makes a request to each of the twelve post to get the full information, so you are able to see the full content of the post directly from your reader.

The twelve request are being made at the same time and with each request a random provider is selected to scrape the necessary data, if an error occurs while fetching a post instead of pausing the process of generating the XML file, proxigram just ignore that post an returns what it was able to get (let's say 11 out of 12). Hours later, when you refresh the feed, the post that returned an error before, it didn't this time, but since your reader didn't have the id of that post at first, it treats it as a new post.

I could instead of this return an item saying that there was an error getting the post, and a link to the post. But since errors are pretty common, I think it would harm the experience of using RSS, but I don´t know. If people really want to add this, I could do it. What do you think?

Pixelfed is great! But as much as I like the Fediverse, not everyone is here. :(

But I have a plan, to encourage people to use Pixelfed. Since I have to create error pages for the project, I have one in mind. When an instance gets rate limited/blocked, there will be a notice to use another instance (or create one) and a paragraph (or link) to encourage the use of Pixelfed.

It ain't much, but is honest work (View)

Good news! Someone created a Dockerfile and docker-compose.yml is already merged in the repository.

Me too, just give me the content! I don't want to login. But let's hope it can last long as much as it can.

This is great! You don't know how happy this makes me! For things like this is why I wanted so much to have an API instead of just generating the HTML.

disclaimer I don’t really know what I’m doing, this is more of a yolo I wanna look into Android app development kinda thing.

Don't worry, I don't know what I'm doing either, the thing is learn while doing it. :D

I witnessed the death of two great projects too: Bibliogram and Barinsta, I mentioned before that Bibliogram's death was the reason to create Proxigram. The way Proxigram works is that it scrapes the data off others Instagram viewers online, it has multiple providers and more will be added in the future.

@weirdo_from_space@sh.itjust.works Asked the following:

Comment thread (view)

@weirdo_from_space@sh.itjust.works: So this’ll make use of other Instagram viewers instead of directly going to Instagram itself?

@ThePenguinDev@lemmy.world: Basically, but maybe in the future Instagram support is added, just I didn’t want it to make it dependable on scraping Instagram since they are more likely to block the instance. If Instagram support is added in the future, since a random provider is selected every time, is less likely to get blocked by Instagram.

I just created a wiki page with the available instances, pretty simple at the moment. Thanks to @privacyfrontends@lemm.ee for setting up the first instance!

Thank you so much! I'll be creating a wiki page now.

Edit: https://codeberg.org/ThePenguinDev/Proxigram/wiki/Instances

That would be great! If you do, I could create a list of public instances and add yours. :D

I really hope to see it too! That would improve a lot of people privacy journey.