December 2008 Archives

unraveling ravelry's social software success


This is the first holiday break I can remember that doesn't seem to be rushing past me before it's even begun. I've been slowing down a lot, and indulging myself with binges in everything from baking (double chocolate walnut biscotti, my annual gift to our department office staff) to reading (I tore through Malcolm Gladwell's Outliers in a day, loving every second of it, and am also really enjoying Neal Stephenson's Anathem), and crocheting.

That last one is what leads me back to social software musings.

During my annual day-after-Christmas bargain hunting with Alex, I came across some bargain-priced sueded yarn, so I picked up a bag full of it. After I got it home, I had a hard time deciding what exactly to make with it. I tried a few patterns from books and magazines, but none seemed to work well with that particular yarn. Then I remembered Ravelry, a social site for knitters and crocheters that I'd joined this past year.

What's interesting about Ravelry is that it isn't just glorified forums--there are plenty of very active forums for crafters out there, but that's not what I was looking for. Ravelry is far more like Flickr or LibraryThing than it is like Crochetville or Craftster. That's because Ravelry is based around objects--yarn, patterns, projects, people--rather than conversations.

Like Flickr, LibraryThing, delicious, and other successful social software tools, Ravelry entices you to enter data and metadata because it's useful to you. It's helpful to enter your yarn stash into their database because once you've done it, you have quick and easy access to a list of all your available yarns--without having to dig through boxes and bags. Most serious needlecrafters have enough yarn to make that well worthwhile. And once you've done that, it's easy to add new projects that you're starting with your yarn.

But the real power comes from the aggregation of that information. After I'd entered my new yarn into my online stash, I could immediately see that there were over a thousand projects listed on Ravelry that used the same type of yarn. And following that link allowed me to further sort the projects by craft type (crochet only), by yarn color, by type of project, and more. That's what makes Ravelry so very useful to me (and the thousands of other needlecrafters using it)...the ability to pivot on different aspects of the data--the yarn, the pattern, the designer, the crafter, and more.

Ravelry pulls together a number of the factors that make social applications work:

  1. It meets the specific needs of a specific community. Facebook did this for college students, Flickr for photographers, LinkedIn for people looking for jobs or workers. If you build it, they won't come...unless you're helping them solve a specific problem.

  2. The initial hook is not the value that you're providing the community--it's the value that you're providing to yourself. Storing and re-finding your own bookmarks is the hook in delicious; being able to store, organize, and share your photos pulled people into Flickr.

  3. The value to the user grows as the community expands. This isn't true for most forum based sites, which simply don't scale well. But site that are focused on objects rather than conversations provide increasing value as they scale to more users. The successful sites scale at a reasonable pace, as well--including Ravelry, which continues to add users from its waiting list at a rate that doesn't overload their infrastructure, thus avoiding the "fail whale" problem that can lead users to decide a site isn't reliable or useful enough to invest time and energy into.

  4. The sites are well designed to pivot on critical data points that are relevant to their users. That's most likely to happen if the people building the site come from their target community of users and therefore understand the nature of the information needs. Mark Zuckerman was a college student. Stewart Butterfield and Caterina Fake were both photo enthusiasts. Joshua Schachter wanted a way to store his own bookmarks efficiently. And Ravelry was started by an avid knitter (Jess) and her coder boyfriend (Casey).

There are plenty of communities of interest out there that can benefit from online sites that help them connect in useful ways. The success of those sites, however, will depend on how well they get the above things right in their implentations.

big news for the lab for social computing


My Lab for Social Computing has just been given the official green light for something we're really excited about--we're going to become part of the RIT Libraries!

Rather than being a somewhat orphaned group with no formal home, we're going to become a full-fledged separate organizational unit of the library system, which will give us access to their extraordinary team of administrative staff, a wonderful office location in the library itself, and a college/department-neutral space that doesn't leave any of the many faculty working with us feeling like they're second-class citizens.

Anyone who knows me knows that libraries have always been a big part of my professional life, and this move feels like it's perfect for both the lab and the library.

We'll be having a "grand re-opening" on Friday, February 13th, and we've managed to convince the amazing David Weinberger, philosophy PhD, Berkman fellow, marketing guru, author of Everything is Miscellaneous, and all-around wonderful guy, to be our featured speaker.

So...make sure to block out some time on that Friday the 13th to hear David talk, and to check out our new digs!

(Why no link to the Lab website? Well, it's under grand re-construction itself! We should have a new site (Drupal-based, yay!) up at the beginning of January, and all the information about our grand re-opening will show up there. I promise I will blog and twitter and email that information around as soon as it's live!)

is twalala my solution for twitter filtering?

| No Comments

I love Twitter, but because my primary use of it is to keep up on the day-to-day lives of my friends, I don't like reading "live tweets" of events. They tend to create a flood-like effect in my stream of updates that drown out the personal things I care about.

I can always unfollow someone who does that a lot, but often they're people that I really like and care about and want to follow. So I've been looking for something that would allow me to hide only their tweets on specific topics (typically those that contain an event-specific hash tag, like #etech).

It looks as though Twalala may do what I want. Here's what the filter options look like:


The only downside is that I usually check Twitter on my computer, and Twalala is obviously designed for use on an iPhone screen. I'm wondering if there's a way to add a user-defined stylesheet so that I can get the Twalala filtering goodness in a full-size-browser-friendly format...

one of your classmates has been killed by a werewolf...

| 1 Comment

This quarter I'm teaching a graduate seminar on online identity, community, and group behavior. I taught it last year for the first time, but now that it's not all new prep I'm able to have a bit more fun with it.

This week we're just starting with identity concepts, and I was trying to think of an engaging way to get them doing something related to identity issues. When I woke up this morning, I had the idea of having them play Werewolf, which is at its core all about knowing/deducing identity based on contextual clues.

I hand-wrote roles onto pieces of postcard stock, and handed them out to the sixteen students after a very brief overview of the rules. (One or two of them had played before, but most were new to the concept.) We played a very simple version of the game--for those of you familiar with werewolf, we had no healer and those "killed" did reveal their identity upon death.

It was spectacularly successful. Most of the students were game design & development students, and listening to them process all the game theory issues out loud was fascinating and immensely entertaining. Best of all, it forced even the quietest and most introverted students to engage in discussion (and misdirection), required everyone to learn each others' names, and created a sense of engagement and fun that's so often lacking in grad theory classes.

In other words, today was full of win. :)

Some useful links for people wanting to try werewolf themselves:

prop 8: the musical

| No Comments


best techno-thriller ever

| No Comments

Back in February, Joi Ito posted about a book entitled Daemon that had been recommended to him by Stewart Brand. It sounded interesting, and I made a mental note of it. Then in May I ordered it for my summer reading pile.

Once it arrived, Gerald started reading it, and later Lane started reading it, and it migrated out of my sight for quite a while. When I was packing for the trip to Alabama last week, I grabbed it and stuck it in my bag--and Saturday I finally started it. Then on Sunday I stayed up until 2am to finish it!

It was awesome. Seldom do contemporary writers combine engaging narrative and plot with genuinely knowledgeable representations of technology, but this one really does. It's a must-read for anyone who's interested in gaming, security, AI, distributed systems, or just good fiction.

Unfortunately, you can't buy it right now. The book was apparently self-published to begin with, under the pseudonym Leinad Zeraus (which is the real name of the author, Daniel Suarez, spelled backwards). After it became popular (and with Stewart Brand and Joi Ito recommending it, that was inevitable), Penguin Books' imprint Dutton purchased the rights, and they're re-releasing it in hardcover in January. Go pre-order it now! Really! It is a wonderful book. (I'd tell you to get the older edition from the library, but according to WorldCat only fifteen libraries in the entire US have it in their collections. Sad.)

In the interim, our pseudonymous paper back version seems to have shot up in value--used copies are going for $80 and up! I'm not selling, though. We'll keep this one in the permanent lending library.

About this Archive

This page is an archive of entries from December 2008 listed from newest to oldest.

November 2008 is the previous archive.

January 2009 is the next archive.

Find recent content on the main index or look in the archives to find all content.


Category Archives