I think I’ve finally found a good replacement for FeedDemon (Windows) and Reeder (iPad), the two commercial clients I had been using to access the Google Reader backend. Time was running short, with Reader shutting down in July and those two clients not supporting any other synchronization scheme. The most popular choice, Feedly, does not appeal to me. Its interface is too simplistic, and I’m deeply suspicious of another free service without a clear business model. Instead, I settled on NewsBlur. (Update: No longer, please see below.)
Samuel Clay’s paid service has been around since 2010. Its rather modest 1,500 daily users exploded to 50,000 when Google announced the shutdown of Reader. By now the flood has receded a bit, and NewsBlur has rapidly upscaled to accommodate a larger and growing customer base. Speed and availability are currently fine. According to NewsBlur statistics, 5,581 “premium” (paying) and 6,436 “standard” (free trial) users were online yesterday. At $24 per user & year, that looks comfortingly sustainable. Check out Adrianne Jeffries’s article for more information on NewsBlur’s history and its recent redesign.
NewsBlur Features
Aside from the service’s solid three-year history, successful expansion after the Google Reader influx, and sustainable business model, these are my reasons for picking NewsBlur.
-
Fantastic interfaces. The web app for desktop browsers is easily on a par with FeedDemon, a native Windows application. You get the same configurable three-pane layout with a large number of hotkeys, and even nearly the same speed.
-
The free iOS app lacks some functionality and polish by comparison, but is still perfectly usable. Remarkably, both web and iOS app can show articles in either Lucida or Georgia, and at a variety of font sizes to boot! (There’s also an Android app I didn’t try.)
-
Articles can show corrections within the text, using strike-through and colors, whenever NewsBlur had fetched an earlier version of a story and then later a changed version. I don’t think I’ve ever seen that feature in another news reader.
-
NewsBlur is fully open source, including the apps, and offers a documented API. This is not relevant to keeping the servers running, but it’s a nice gesture nonetheless, and possibly enables third-party clients, as well as enhancements and bug fixes.
NewsBlur also offers features I wouldn’t count as advantages, namely its own private social mini-network in the form of “shared stories” with comments and followers. Fortunately you can ignore this nonsense if you don’t want it.
Update: NewsBlur Trouble
Immediately after posting this recommendation I discovered one important drawback that you should be aware of. Just like Google Reader, NewsBlur updates smaller feeds very slowly if they don’t send push notifications. I haven’t seen the days of delay threatened in the NewsBlur FAQ, but my own feed got classified with a 9.5–12 hour update cycle. This delay is visible in each feed’s statistics panel (web client only). Push notifications would force an update but WordPress doesn’t send those, and I’m no longer going through FeedBurner which did.
Fortunately, the feed options (on both web and iOS) also include “Insta-Fetch Stories” which appears to instantly refresh the feed in the shared database. So I’ll have to do that manually whenever I make a new post. Interestingly, neither the NewsBlur FAQ nor NewsBlur support who told me about feed statistics mentioned insta-fetch. I’m guessing they don’t want people to overuse it, but for now this leaves a bad taste in my mouth. I would have expected a paid service to check each feed at least once per hour. I still maintain my recommendation in view of NewsBlur’s excellent clients and a lack of good alternatives, though.
2013-06-01: After digging deeper, I decided to cancel my account as NewsBlur appears to have pervasive problems with feed updates that they cannot fix and refuse to acknowledge.
Reader Emigration Trouble
One problem did occur during the transition to NewsBlur, but that turned out to be Google’s fault. Specifically, a few of my ca. 150 subscriptions had not been imported. As it turned out, that was because they had not been in my current FeedDemon export file either! Apparently Google Reader had stopped accepting new subscriptions a while ago, and FeedDemon had silently omitted those rejected subscriptions from its OPML export file, even though they appeared correctly in FeedDemon’s own feed list. Something to watch out for if you’re making a similar transition from a third-party client coupled to Google Reader.
I had to dismiss another promising service, Feed Wrangler, because its OPML import simply aborted on multiple attempts, after about a third of my subscriptions. To their credit, they refunded my subscriptions within a day. I’ve heard of a similar problem with The Old Reader but haven’t tried that service myself. Wherever you go, you’ll want to double-check that all your subscriptions have actually arrived at the new service.
Rolling Your Own
I briefly considered writing my own multi-device feed reader without server-side code, just using JavaScript and “dumb storage” (FTP or WebDAV). Open-source readers that require a full server already exist, e.g. Tiny Tiny RSS, but I wanted to see if I could avoid that.
Parsing OPML and fetching feeds using XMLHttpRequest should be straightforward. Synchronization is a bit problematic since JavaScript cannot write to FTP servers. XMLHttpRequest is supposed to do that, but browsers don’t implement any protocol other than HTTP(S). That leaves WebDAV which is accessed through HTTP. So the reader would have to request WebDAV credentials, presumably store them in a local cookie, and then write state data to the WebDAV server. So far, so good.
But what state data should be written? Storing the read state of every single article, let alone the full articles themselves, would quickly grow to a volume that needs a proper database. So realistically, I’d only store the last read date per feed and declare all newer articles unread. However, that means that all feeds and articles need to be fetched from the original site in every session. This makes looking through a feed’s history slow and unreliable, or outright impossible when an old article has dropped out of the current XML feed file.
That’s not really better than just using a single-device reader with a proper local archive, never mind an excellent web service such as NewsBlur. For now I’ve shelved the idea of writing my own minimal reader.
RSS Praise & Nostalgia
The outrage over Google’s shutdown of Reader has died down, but March and April saw a couple of noteworthy articles that I wouldn’t want to leave unquoted. So here’s another batch of hopefully premature epitaphs on the red-headed stepchild of social media.
MG Siegler, What If The Google Reader Readers Just Don’t Come Back? RSS is not just a major traffic source in its own right but also seeds social media linking.
Reader’s users, while […] relatively small in number, are hugely influential in the spread of news around the web. In a sense, Reader is the flower that allows the news bees to pollinate the social web. You know all those links you click on and re-share on Twitter and Facebook? They have to first be found somewhere, by someone. And I’d guess a lot of that discovery happens by news junkies using Reader.
Marco Arment, The power of the RSS reader. If you don’t see the point you’re doing it wrong.
The true power of the RSS inbox is keeping you informed of new posts that you probably won’t see linked elsewhere, or that you really don’t want to miss if you scroll past a few hours of your Twitter timeline.
Annalee Newitz, Magazines have finally killed blogs – but in a way you never expected. Reader-controlled syndication is an aberration. The historical norm are publisher-controlled bundles and walled gardens.
That’s why RSS readers were so remarkable — they let you take information from everywhere and organize it however you like. Your Wired stories were filed in the same place as your Entertainment Weekly stories. Everything was mixed together in an information jumble. […] Information in the world of RSS is not organized into silos that resemble magazines or social networks.
Ryan Holiday, Our Regressive Web. RSS gives readers power over what and how they consume, and that’s exactly why publishers and advertisers hate the format.
When a site stopped delivering a quality product, I had the satisfying ability to withdraw my subscription. Apparently that power was threatening. […] RSS is impervious to blogging’s worst, but most profitable, traits: [banner ads, slide shows, click baits, pagination, etc.] No wonder nobody ever pushed for widespread adoption of RSS.
Meanwhile at Google
While shutting down Reader and retiring its RSS browser extension, Google introduced Google+ Notifications, another Chrome extension that tracks website updates. Of course, as Panayotis Vryonis complains, those websites must be Google+ pages, you must be logged into Google+ yourself, and Google will analyze the sites you visit to check for Google+ links.
That’s a terrible replacement for a proper RSS reader, but it fits Ben Thompson’s interpretation of Google+ as a global identity and tracking service. Contrary to what pundits usually focus on, it’s quite irrelevant how many people chat on G+ compared to Facebook or Twitter. Google cares more that people are constantly logged into G+ while browsing other websites than whether they are active on G+ itself.
Every feature of Google+ – or of YouTube, or Maps, or GMail, or any other service – is a flytrap meant to ensure you are logged in and being logged by Google at all times. […] Make no mistake, Google+ has been a massive success. Credit to Google for their willingness to be misunderstood and portrayed as a loser even as they mine information Mark Zuckerberg can’t even dream of.
