Multivrss 0.2 up close

Translated from the original Italian with the help of AI. If a sentence sounds a bit off, that's probably why.

In my original plan, MultivRSS was supposed to be simple and minimal, but how can a Product Manager resist the temptation to add a ton of useless features? After piling on enough of them, I had to build a page just to keep track of the changes, and try to impose some discipline on how I publish updates.

Changelog and version tracking

Here you'll find the changelog page, with a concise description of every change made in each version.

Readability for in-app reading

The feature I'm proudest of is the integration with Mozilla's Readability library, which lets MultivRSS users read articles directly inside the dashboard.

And I'm even prouder of the fact that, as of today, you can:

• copy articles as plain text (and maybe paste them into an LLM prompt)
• download articles as markdown (and maybe drop them into your favorite second brain. And if you don't have a second brain, I'd suggest starting here)

A few other things

• The app's internal search engine originally ran on Meilisearch: blazing fast, extremely capable, and also the textbook definition of overkill. I realized Postgres's full-text search was more than enough to search across feeds and saved articles, so I backtracked and removed Meilisearch entirely. If this project had any users, they probably wouldn't have noticed a thing
• Account deletion: if this project had any users, they would now be able to delete their own account
• Progressive Web App: from Chrome you can install the site as an app and use the PWA on both desktop and mobile. I personally use it on both, with great results
• The ability to tag saved articles
• You can paste a YouTube channel link and follow it as a source
• Noticeable improvements to the suggested feeds page, hopefully one I'll keep updating regularly
• A FAQ page
• Integration with Sentry for error reporting

There's definitely more, but this is already long enough.

Until next time.