RSS ('really simple syndication') is a simple old-school way of keeping track of feeds on the web. RSS is the tech that allows podcasters to say 'wherever you get your podcasts' - there's no vendor lock-in like there is on e.g. youtube. If you've ever come across a blog, publication or generally anything cool on the web and thought 'I should come back here': RSS is the easy way to get updates.¹

RSS is useful for academia for a range of reasons, particularly for getting journals alerts in a searchable format. This means you don't need to open up those Nature TOC emails every week just to check if it has something relevant to you.

RSS Readers

First you're going to need an RSS reader, this is just some app that tracks the RSS feeds and let's you manage them or read their contents. I personally like newsboat, I'll get to why in a moment but just note it's command-line only. If you're just starting off, pick anything you find with a quick web search that's free to start. You used to be able to use MS Outlook, you can definitely use Thunderbird, if you don't want another app/web service. If you like the RSS life, you might want to try out FreshRSS as I recommend below.

Finding RSS Feeds

So you've got some sort of reader situation, you need some feeds. What you're looking for is a link that you copy into your reader. Often it'll be with a the RSS logo on sites, in particular blogs. You can also install extensions to your browser that auto-find feed links for the site you're on. On academic journals you often need to hunt a bit harder, but at least in physics they all will have one.

I recommend you go with what you're interested in and the feeds will follow.

API Feeds

Some sites such as the arXiv and Internet Archive Scholar, have direct RSS output from their API. Basically this means you can have an RSS 'alert' for your chosen keyword/search terms. Here's some examples I use:

https://export.arxiv.org/api/query?search_query=all:%22nitrogen%20vacancy%22+OR+all:%22quantum%20diamond%20microscope%22$id_list=&start=0&max_results=20&sortBy=submittedDate&sortOrder=descending
https://scholar.archive.org/feed/rss?q=nitrogen+vacancy+centre+widefield+quantum+diamond+microscope&filter_type=everything&filter_availabilty=everything

Check out their respective docs for the details (on IAS you can just search on their site and it'll let you copy the rss feed for that search). These tools are very powerful, and almost mean you don't need to sub to the journals' feeds manually!

FreshRSS

FreshRSS is a 'self-hosted' RSS reader (browser/web only) and server. This means you have to setup a server somewhere to run it - but don't get scared just yet! There are 2-click solutions to do this for you, for free (with some light restrictions), or for $2/month, which is very reasonable IMO. Before you do any work, check out the demo to see if you like it (note: there's a nicer orange theme available once you have your own instance called 'Ansum'.

See the install instructions, or just do what I did and set it up on pikapods in a minute. I think cloudron is free, but the first paid tier is more than pikapods and I was happy with the $2/m. Note these services also have a bunch of other self-hosting apps you can run for cheap, I'm keen to try some of them out myself.

You can actually use FreshRSS as a server/aggregator, and then send the results to a different reader, like newsboat. Pretty cool! Looks like I've only just scratched the surface of FreshRSS too.

Scraping websites/custom RSS feeds

If you do come across a website without an RSS feed, but clearly something that should be, FreshRSS has functionality to let you make own! I don't want to give a full tutorial here, so I'll just recommend you use the 'XPath' option and use F12 to open up your browser web tools to see the html tags for the elements that make up a feed 'item'.

Query Feeds

Query feeds ('user queries' on FreshRSS) are a powerful (IMO requisite) feature of some readers that allow you to save a search through your RSS feeds for keywords etc. I have seven query feeds set up to organise my reader, and I mostly don't go through the feeds independently.

In newsboat you set these up like so:

"query:|> OptSpinD:title =~ \"diamond|nitrogen vacancy|nitrogen-vacancy|hbn|hexagonal boron nitride|spin defect|nv center|nv centre|ODMR|colour centre|color center\" " 

where I've given the feed a name (|> OptSpinD) and match all items with a title that contains the list of terms between pipes (|).

In FreshRSS make a search (there's some filter syntax to learn), then you can save it as a user query. It's not quite as neat as newsboat (nothing is 😉), but combined with filters on incoming items (e.g. to auto set them as read if they DON'T match what you want), you can replicate the newsboat behaviour.

Make sure you have some sort of search system set up, it really is core to the experience!!

Sharing/backing up your feeds

Libraries can be shared between readers or computers with Outline Processor Markup Language (OPML) files. I occasionally export an OPML file so I have a backup of my (~300) feeds.

Outstanding Questions

• Newsboat can be a bit slow to update all of the feeds, and query feeds in particular cannot search article contents (not just title) for keywords above snail pace. Can off loading some of this work to a server (FreshRSS) improve the experience?
• How simply can I self-host 'manually', both on a cloud provider and my own hardware. Setting up on a raspberry pi or old laptop would be pretty satisfying.
• Are there any other pikapod apps I want to try/will be at all useful to me?

References