I had that in mind when I set up newsbeuter to sync my NewsBlur feeds. Yeah, I am on a commandline kick, and it is rocking.
So I looked at my 129 feeds and nearly 7,000 saved items and thought to myself, “maiki, what are you doing?”
I needed to get this under control, because as awesome as NewsBlur is, the saved items bucket is not maiki-friendly. It just gives me a place to hide articles and never consume them.
I don’t want to be a hoarder.
Hence, a great purge!
Now I am at 75 feeds, and they are currently all caught up.
How did I decide what stayed and went? I came up with some simple rules, but first a few exceptions:
- If it is a site I host, I subscribe. This has been useful, as I can see if a client is writing for the web, or if there is an error with the feed.
- If it belongs to a friend. My friends get to be stupid with their feeds, I love them. Also, I will probably fix their feeds anyhow.
- They are job feeds. These aren’t great, but are kinda worth the extra hassle if they deliver. Think subscribing to a Craigslist search.
Despite those exceptions, most of those feeds worked out. Turns out I host a substantial amount of my friends blogs, so they were always legit.
Okay, now for the criteria!
- Full article only. I don’t have time to click through and potentially deal with a crappy, illegible type system that I have to increase or render differently. Also, most of my saved articles are summaries that I think might be cool, but they just stack up.
- Served over HTTPS. If anyone wants to read what I am reading, they are going to have to grab the public feeds just like NewsBlur!
- No feed proxies. I am reading your ideas, not opting into your analytics engine.
- Has to have been updated in the last year. This just saves resources, as RSS is a polling tech, and there isn’t a reason to keep checking the feed for updates when none are coming.
A better bucket
Now that the feeds are in order, I need a better process for emptying my feeds bucket. Well, turns out there is a method in newsbeuter that works for me: saving the article.
As I am reading my feeds I will want to comment on one, or reference it later. In newsbeuter the default binding is
s, and it will save the article as a text file; I configured my
~/docs/news/, so it even gets synced to Nextcloud, in case I need that reference on the go.
Aside from reference, I am trying to publish more regularly, and am playing with weekly list o’ links. That
news directory comes in handy on Saturday when I go to compile it. I’ve processed a few of these already, but here is a current
tree of the directory:
0 directories, 20 files
The cool part of this is, once I’ve done my link and commentary, I just delete the article from my filesystem.
This new bucket is better for my sanity, and has the added benefit of being offline friendly; I carry my wifi spot wherever I go, but sometimes I just want to sit under a tree and write, ya know?
My hopes for the new system are less distraction and more writing. My goals are to encourage the best practice of sharing information.