Recently another round of discussion has started on the web about how RSS  is riding to the sunset. I think there is some irony that most of us were alerted to these posts either from our feed reader or other aggregation site like Techmeme.

Your newspaper doesn't show unread count, so why does your feed reader?

This time the debate originate from a blog post at ZDNet. And I think that as long as the title of the post was that RSS readers are becoming meaningless, the post makes some sense. And it’s true, there’s not much innovation in RSS readers these days and some of the design mistakes were listed here. The idea that a user imports a RSS document and reads just it, that’s dead. We’re still far from what’s possible when computers work on feeds.

Another thing this means is that as feeds become more and more part of the web’s infrastructure (see for example Google’s GData), it’s not really interesting for end-users. This in turn means that there just isn’t any money in it. For certain websites, this of course equals to that tech being dead.

One of the blunders in feeds was the dichotomy between RSS and Atom standards. While the former is used today as an umbrella term for feeds in general, it’s really, really inferior to the latter. The problem with Atom is that it came late to the game and while it can be as simple as RSS, but it can also be used for many other things than just blog posts and most RSS readers couldn’t be bothered. This is why the RSS format is dead in the water. The Atom format is much more flexible and is used in many other ways than just one-way polling (see above-mentioned GData for example).

Feeds are here to stay, they are not Web 1.0 stuff, but an integral part of Web 3.0. They just can be so much more than “seeing what’s new”. A site like Techmeme could not exist without feeds. It’s just that we haven’t unlocked the potential. It’s not sexy and it might negatively affect web ad revenues. This is why I think Techmeme shines, just like Friendfeed; they follow the “River of news” approach to new items that was proposed early on. Other readers, like most desktop apps and Google Reader, put new items into an inbox, pretending that each new item has an equal value to us.

Feeds are really immature technology, we’re still unsure about formats and how to consume feeds. And, on top of it all, how could we use this technology the improve the experience of having a discussion on the web. I propose we take a look to ancient computer history.

Before the Internet, on the dial-up BBS services it was a common due to the call costs to download all the new discussions on that box’s forums to your “offline reader” and disconnect. One could then peacefully go through and answer to any threads that were interesting and upload these back to the BBS. But it wasn’t limited to just one board, an offline reader was one inbox for all your discussion on all your BBS boxes. The Usenet newsgroups could be “consumed” using a similar logic. But, today, as Diaz says, our “sources of for reading material are scattered across the Web” and this approach doesn’t work right now. But it could in the future.

I’m not sure that we can stop and concentrate on discussion anymore, because Facebook and Twitter have made “discussions” move so quickly that concentrating on just one is impossible. But if we could go back to those more peaceful times, I’d like to have these “offline readers” back. Of course, they wouldn’t need to be offline today, but real-time.

Discussion on the web is not in good health. It’s scattered and disjointed. I’m not calling for a centralized solution, I’m looking for a standardized solution – something that’s already possible with Atom. We subscribe to blog posts, but we don’t subscribe to the comments. It’s a hassle even if the blog you read happens to use WordPress’ e-mail subscriptions or Disqus, Intense Debate or some other solution.

There are some major obstacles, one of them being that the income of sites are tied to ad impressions. The other huge problem is that we need to lay down the infrastructure first. Pretty much all sites support the one-way RSS today, but only a handful support Atom Publishing Protocol (which is a different thing from the simple feed itself). Also, none of the forum software, as far as I know, support anything like this. Instead of using the web interface, it would be possible to access the discussions using another, more suitable interface. Most of blogging tools are APP aware, though.

We don’t listen to music by going to individual bands’ websites, we have collected our music to a single source (be it iTunes, Spotify, Winamp or something else). I don’t know about Google Reader’s long term roadmap, but it wouldn’t surprise me if something like participating to comments is there. Yes, you can “like”, “share” and “comment” the posts there, just like in Friendfeed et al., but you can’t participate to the discussion on the original site.

We can rebuild discussions on the web. We have the technology.

Image by FastIcon.com

Like

Unlike

I noticed that I’d migrated unconsciousnessly into using Google Reader as my main RSS Reader. Previously I’d used NetNewsWire Lite like many others (even occasionally using it with NewsGator Online).

There were many underlining reasons for this. First of all, NetNewsWire Lite doesn’t archive posts, so I found out that if didn’t read posts from some strangely (mis)configured feeds, they would just vanish from the app as they were removed from the RSS feed document. I don’t consider myself an internet addict, who needs to read all the feeds the instant they update, so this was a major problem as I was losing half of Freakonomic’s posts. This isn’t wasn’t a feature I was willing to pay for, as Google Reader does this even better by archiving each and every feeds each and every post. The other reason was that I’d sometimes like to access my feeds from different locations. NetNewsWire (and FeedDemon) can be used with NewsGator Online, but its interface sucks and the syncing between the website and app didn’t always go flawlessly.

The advent of Google Gears will probably make other desktop apps also irrelevant. It’s strange how even though for a long time there have been things like Java and XUL and now there’s AIR and Silverlight, they just never caught on and instead people kept hacking on with JavaScript to make a dynamic web. The downside to web apps is that the web browser UI is not good in usability sense and you’re missing out on many of the underlying OS’s frameworks – in this case it doesn’t really matter if you use any of the above mentioned platforms as they don’t unlock your OS’s special advantages but naturally try to maintain a stable cross-platform experience.

I was sure that I’m not alone with my migration pattern and tried to do some research. Unfortunately WordPress doesn’t let us see where our RSS traffic originates from in client sense, but Netvibes (which has been mentioned here before) and Google Reader get some hits as referrers every day. So, as for desktop RSS readers, we don’t have that data. I found a poll from Read/WriteWeb, which seems to support my argument. I was surprised how many people used browsers own RSS capabilities, though. Is there are better source for information on web-based rss reader adoption? (Yes, I know)

There are similar stories on the net though. Here’s how Ades went to Google Reader, and here’s how Alex did it. I agree with Ades that it’s a major fricition-remover that so many people already use Google’s services everyday that they don’t need to login to Reader. Alex, on the other hand, sees this as a privacy issue. K-inthehouse argues that there are benefits for bloggers to use dekstop clients instead. In my opinion points 6 (individual feed refresh) and 9 (archiving) don’t matter at all on web-based readers and points 1 (labels) and 4 (grouping) are, in my opinion, pretty well achievable on web-based readers also. What I do agree is point 3 (search) and I too find it ironic that it’s not implemented in Google’s Reader. Of course, I rarely blog about other blogs (the “echo-chamber” effect) – there are other ways to put other posts on the radar, for better or worse – so maybe some people like desktop clients for that.

It looks like Brent Simmons did sell NetNewsWire at exactly the right time. It was a great achievement to develop a Mac-only RSS reader, which had the largest market share (or, more to the point, user base. so, user share?) across all platforms. I doubt this is anymore possible now even on any platform as Netvibes, Bloglines and Google Reader take on more and more people, platform-independently.

Kari has relatively few feeds he follows on a weekly-basis, so things are probably different for you. Sound off your opinion in the comments!

Like

Unlike