Tiny Tiny RSS and Feed Reading

Posted in hyperweb by Adam

16 March 2013

Discusses the Tiny Tiny RSS (self-hosted) web application as a replacement for Google Reader (which will be end-of-life in July 2013).

Google Reader, a popular web application for organizing and reading RSS feeds (computer-friendly lists of posts from websites), will go away in July 2013. Scrambling for a replacement, many flock to monolithic, hosted substitutes.

I already had a perfect substitute up and running. A few months back I decided to add some feeds that did not fit my usual reading patterns on Google Reader. I looked around for an alternative and decided upon Tiny Tiny RSS (aka tt-rss), a free software implementation that manages and allows easy reading of feeds.

I only added the feeds that didn’t fit my Google Reader usage at the time. I would check tt-rss once a day, and continued checking Google Reader more regularly as always for the bulk of my reading.

Then the Google Reader announcement came, and while I looked at and tried a few of the alternatives, tt-rss fit my use best. Google Reader did not have feed filtering (to my knowledge) where tt-rss does. The ability to “clean up” some of the noise on some feeds means a more pleasant reading experience. For a mobile solution, the web interface has a mobile version (if Google Reader does, I never tried it; I did use their Android application, though), and an Android application can talk to tt-rss as well (requires enabling the API).

Google Reader served many people well for years, but it never went as far as it could have. With people forced out, a lot of projects like tt-rss will hopefully see increased relevance and improvement. But RSS itself never achieved the recognition it deserves.

RSS would serve us best if it were somewhat automatic. Google never took the obvious step of showing you feed items from recently-viewed sites, for example. Auto-management of feeds would have gone a long way toward improving the usage and importance of RSS. Manually adding and removing feeds, not being able to disable them, no filtering, and needing to understand feeds; these things made it a mess.

When you visit a site, they do everything they can to get you to stay. They try to pull you in with other materials, they beg you to subscribe. But as with just about every other industry, they fail to see their role in that process. They should work with browser builders to add next-generation discovery and subscription models.

For its part, tt-rss now has more feeds than Google Reader ever did. I’ve added more feeds knowing I can filter them easily, though I suspect the worst feeds will soon go away. While it isn’t perfect, it’s at least on par with Google Reader. And once I set it up (admittedly easy for someone at my level of knowledge), it’s no challenge to keep using.

The main thing I wish it had (and something I’m continuing to look into for several services I run for myself) is a single-sign-on solution. The last time I checked, my various services had differing support for different authentication technologies. Having just one that worked for them all would be delightful.

Diversity Rules for Social

Posted in hyperweb by Adam

4 February 2012

We got here because social came late. Look at other parts of the Internet. But you have to look carefully, you have to look at social types of activities. Like email. Like instant messaging. Where are we going?

With the Facebook stock selling thing going down, thought I’d talk about why Facebook doesn’t do it.

How we got here

We got here because social came late. Look at other parts of the Internet. But you have to look carefully, you have to look at social types of activities. Like email. Like instant messaging.

Other social services

If email hadn’t existed, it could’ve ended up like Facebook. Instant Messaging did start out that way. And it still is that way in many respects. Used to be that the AOL instant messaging platform dominated. That grew out of the AOL data service of yore.

When the Internet at-large became more important than the data service, people outside of the AOL bubble wanted to talk to AOL users, so AOL expanded to let others use their instant messaging service. Before, though, AOL was a dominant force.

Used to be that lots of television adverts would give you their AOL keyword, just like lots of adverts now tell you to go look them up on Facebook. There was also ICQ, and then MSN came out with an instant messaging deal (which came bundled with Microsoft Windows). And there were some others.

These days, lots of folks use something like Pidgin which lets you throw multiple accounts into one application and use instant messaging on any of them. But the networks are still fragmented. Without one of these polyglot applications, you still have to use separate applications to get on different networks, and with one, you’re still talking to multiple services simultaneously.

Email is the gold standard, though. Everyone can participate simply by knowing the address to send mail to. But that’s a problem too, with SPAM.

To win, you must hold the most users’ attentions

But Facebook came along at a time when Social was still new, MySpace being dominant, and they won that race, were crowned the winner. With a social service, without federation, the winner isn’t the technological superior, but the one that gets the most users. However they do that, and it can include technological superiority, they win.

And as long as you can hold on to those users, you keep winning. It’s a clear enough goal. So you look at why users stay or how to bring them back. And Facebook tries hard.

They try building out their platform, so now you can sign in to various sites and services using your Facebook login, which appeals because then you don’t have to sign up separately. They add games, social games, where you can try to beat your friends and family.

They let their platform leak, so that people say the wrong things to the wrong people at the wrong times. Drama is a great motivator for behavior. It keeps people coming back to the old Globe Theater.

But the main hook is that’s where the people are. Back in old times, people went at night to the place of lights. Usually that was the local tavern or inn. Sometimes it was the church. They tend to follow the voices, especially the familiar voices.

Alternatives require incentives

If you walk the same route to and from work everyday, you’re probably not going to deviate much. If suddenly there’s a detour for construction, it might make you late. Or maybe you like walking past the bakery, being hit with that breeze of warm, moist bread air.

It’s difficult to establish an alternative, unless you first have a draw for some folks. Maybe it’s because they’re college kids who want to do their own college thing. Or it’s based on some existing college tradition, and you’re making it digital.

Maybe they want to go there because they can get laid, or because of who isn’t there.

The question is, what can a diverse social system deliver?

Specialization

One of the big things Facebook can’t do is specialize. If there’s a great feature that will help teachers interact, but won’t help everyone, Facebook won’t add it. It would be dead weight to them.

Moreover, while Facebook can scale a lot, specialized providers can scale better. The cache mechanisms involved with having a niche userbase versus a general userbase means that the IT side of the equation gets easier. Facebook has to plan for all of its users all the time, but if there’s an annual event for a specific group, the specialized provider can ramp up for that and pull back the rest of the year.

School kids have usage drop off during the summer. Facebook probably doesn’t lose a bunch of servers during the summer, but they probably could stand to. A specialized provider or specialized providers could handle that better.

Competition

Even where the features are general purpose, the implementations can have varying success. Monoliths try to inject competition via what’s called Split Testing. It’s just a regular experiment, though. You give one group nothing, the other group something, and see if something does anything.

But a federated system is better at stoking evolution of a service. When one provider finds a new, great feature, others have to match or exceed it. Sometimes the latecomers do it better, or find some more essential reality of that feature. They are fresh eyeballs to the problem, which monoliths like Facebook don’t have.

Diversity is best

There are a lot of reasons why diversity is a great thing. It reduces the potential for provider abuse, including the scenario where you’re excluded from the only game in town. It means that if the social game changes suddenly, you don’t have a mammoth Facebook trying to turn on a dime, flinging millions of pokes overboard, wiping out all the packetfish swimming nearby.

It also means that the social game can change suddenly. With Facebook, there’s little chance they will introduce disruptive changes that threaten their model, where a smaller provider might. Hell, even when Facebook changes something mostly innocuous they already get a lot of backlash from their users.

And if the social web starts to wither, small providers can change to other services a lot more easily than Facebook can change become a different service.

If Facebook is going to thrive for more than a decade, it will need to branch out sooner than later. It will need to prepare for an inevitable change where social becomes federated. If they are really smart, they’ll see it coming and help it happen, so as to maximize customer loyalty.

Their alternative is likely to end up like Yahoo!. Actually, worse than Yahoo!.

Yahoo!’s only real problem has been a chronic inability to reinvent itself. It’s still got a lot of good people doing good things. It’s just not had that true resurgence it’s needed for years now.

Facebook is in a position that Yahoo! was in back some ten or fifteen years ago. They’re doing great, but it’s not clear what their future will look like given the ability for a Google to come along. For that matter, Facebook itself already did to MySpace what will happen to Facebook if they don’t find a better model in the meantime.

The only real question is whether the next champion will be another monolith, or if the distributed social system will finally kick in.

Badges and the Social Fabric

Posted in hyperweb by Adam

23 July 2011

Looking into the social fabric of the internet, at the pitfalls that knowledge badges try to bridge and how to improve the efforts.

Mozilla started a project called Open Badges; they propose to develop of something of a cross between a human-readable Geek Code and traditional Scout badges. They recognize learning on the internet, so that if you put forth the time and effort to learn about a topic, you earn a badge that displays that ability to others.

Screenshot of Google News, Politics Section showing a story about Conan the Barbarian

Google News has initiated its own Google News Badges, where by reading stories about a given topic you can show off your subject prowess through a badge.

Today’s post delves into the social fabric of the internet, and looks at the pitfalls that these badges try to bridge and how to improve the efforts.

Google works hard to make their news more relevant, so please do not take this as a criticism of their efforts. The problem they look to solve holds its ground, and the Google News site still beats the other, non-user-driven news aggregation sites I’ve seen.

Media quality varies wildly, so reading a lot of articles does not necessarily make one informed. Also, for a lot of stories the headline tells the tale, but users receive no credit for a story they understood at a glance. But, possibly worst, taxonomical issues devalue the results.

For the last example, the current Google News results suffice. The second headline under Politics for me is about the movie Conan the Barbarian.

Bad enough in itself, there are two other problems, namely the Murdoch Empire bookending the Conan story. This, despite my best efforts to rid my Google News sections of those sources which I consider too biased to bother with.

Subjectivity abounds in badges for news and similar pursuits, and Google News’ categorization attempts have not been dependable to date. I would not want a badge based on reading those stories, and would not trust someone’s badge based on them either.

This symptom simply represents the larger problem with crafting badges, namely taxonomy.

Someone possessing a given skill, in name versus practice, might meet, exceed, or fall short of expectations. The fact that I read a book or watched a film does not mean I understood it, and the fact that I did not, I could know it implicitly from cultural references (eg, Citizen Kane).

The delicate art of communication riddles us to decipher who knows what in an efficient manner. Our ability to solve large problems depends on such things, and yet we often fail to uncover the knowledge pool.

Studies reveal that groups with more women tend to have higher group intelligence. For example, quoting “Collective intelligence: number of women in group linked to effectiveness in solving difficult problems,” from Science Daily:

When it comes to intelligence, the whole can indeed be greater than the sum of its parts. A new study co-authored by MIT, Carnegie Mellon University, and Union College researchers documents the existence of collective intelligence among groups of people who cooperate well, showing that such intelligence extends beyond the cognitive abilities of the groups’ individual members, and that the tendency to cooperate effectively is linked to the number of women in a group.

While the studies tend to cite sensitivity to the emotions of the group members, it seems plausible that the type of communication, beyond simple sensitivity, holds a key. More social groups construct better social taxonomies (ie, recognition of the role capabilities for the members) and do so more efficiently. A study purely about discovery of the social taxonomy would probably reveal as much.

Badges may improve discovering the group ability. Chiefly, badges should assist in motivating learning and crediting it. But to truly uncover the promise of the internet, both pieces are needed.

One of the ways to improve badges might be to grant special statuses, like teacher, atop the regular badges. Teaching refines existing knowledge, as it challenges you to present information in different ways and to approach the subject differently than as a learner or user.

Most specifically, teaching relies on formalizing the models of a subject, taking them from primordial form to crisp edges and smooth, consistent surfaces.

The other major challenge and improvement involves ascertaining the existing skills. Some websites already work toward that end. For example, the Reddit community, AskScience currently marks members with scientific training so that readers may measure the reliability of their answers on a given topic. If new-web initiatives like badges take hold, those acknowledgments may be transformed into true badges.

The internet’s potential remains untapped, but with all of the experimentation going on, results will come.