posted April 3, 2015
I do not feel concerned that Betaworks (which bought Instapaper from Marco Arment) has a database of all the articles that I've saved for later. The vast majority of articles that I've saved in my Instapaper is stuff that I'd be comfortable tweeting out or sharing publicly on Pinboard, so privacy is not an issue. I've mostly been using Wallabag out of curiosity.
Set up was pretty easy. I created a mysql database using the cPanel mysql wizard for my domain, uploaded Wallabag via FTP, configured some settings, and had it running pretty quickly. The app remembers where I left off reading a long article, which is a must. (Unfortunately, the iOS app does not, so I have been using the web app, which is mobile-optimized.)
Wallabag is not supported by third-party apps. I have so many iOS apps that let me save a link to Instapaper (e.g., Nuzzel, Tweetbot, pretty much every Pinboard app). Just having to take the extra step of saving a link to my clipboard and running a script can feel like too much work when there's "Save to Instapaper" in the action menu of an app.
Also, there just aren't that many easy ways to get stuff into Wallabag. I have an IFTTT recipe that saves all the articles from a specific RSS feed to a folder in Instapaper. I can't automate this with Wallabag.
A big problem: Wallabag's rendering of web pages does not reliably show blockquotes, which makes it difficult to know when a quote begins and ends. (I've emailed the developer about this, have not yet received a reply.)
There is an appeal to Wallabag in that all my archived articles are sitting in some database on my server. Sometimes I read an article in Instapaper and want to save it for posterity -- like to Evernote. But knowing that the article is already on my server, I can rest assured that I can access it later even if the host site of the article sunsets.
I went into phpMyAdmin on my server and looked at the Wallabag mysql database. Each article is saved as a BLOB which downloads as an
entries-content.bin file. I can open this up in a text editor, and it shows as regular html within a
div container. Not that I would ever hope to have to go through this effort to find the content of an article that may be taken down from the web one day, but I feel reassured knowing that I can.