The Facebook Developers Garage went well thanks to the planning of E27. We had interesting speeches from Battle Stations, a Singaporean MMORPG application that took off, Duration Inc. on how they’re using the .NET framework to build applications, and Mindwasabi about the non-rational, beneficial effects of people working together. I’ve published the unabridged version of my talk, “Getting love from the Facebook Platform,” on Slideshare: ∞

Four new as-of-yet undocumented methods appeared in the Facebook API Tool to developers today, indicating that applications will be able to read listings from the Facebook Marketplace. No word yet on further details, such as whether or not an extended permission will be added to enable write access, allowing applications to post listings*.

• marketplace.getCategories - returns an array of all available categories (currently: forsale, housing, jobs, other, free, forsale_wanted, housing_wanted, jobs_wanted, other_wanted)
• marketplace.getSubCategories(category) - returns an array of subcategories, (calling this with “FORSALE” returns: books, furniture, tickets, electronics, auto, general)
• marketplace.getListings(listing_ids, uids) - returns one listing, as described in the search method
• marketplace.search(category, subcategory, query) - returns an array of results. results are an array with the keys listing_id, url, description, price, poster, update_time, category, subcategory, image_urls (an array of urls)

This functionality will be helpful to applications that involve goods, but unfortunately for the developer, no money can be made from these free postings. If I have a book reviews application, for example, now I can share with my users marketplace listings where they can buy the book, but as a developer, I won’t make any money from doing so. This once again echoes the worries of developers (1,392 words) that monetizing on Facebook might prove to be too challenging. However, at least while developers scramble to figure out how to monetize their applications, the users will benefit from easier access to another useful, free feature of Facebook.

*Update (10/13): The API Tool now shows two more methods which use extended permissions, marketplace.createListing and marketplace.removeListing. All of the new methods have been documented, as now linked in this post. ∞

One month ago, we saw Facebook slip momentarily and expose a few unreleased methods to developers. The new methods, when combined, told us that there will be lists of users, and these have something to do with friends. At the time, I tried some input (did I have any “friends lists?”), but couldn’t get an interesting response back from the Facebook servers.

In a move pretty unusual for a competitive tech company, Facebook has confirmed our speculation on their What’s New page. “Sort out your friends. We’ll let you organize that long list of friends into groups so you can decide more specifically who sees what.” You could argue that they’ve revealed it to us because, unlike features worth hiding, this isn’t an innovative move. Instead, it’s a necessary one, and one for which they probably receive a lot of requests.

Since the story broke, we’ve learned a little more information. Here’s a recap of what we now know.

• What we know:
  • Friends Lists are like buddy lists, in that they are ways to group your contacts.
  • Friends Lists will be located at www.facebook.com/friends/. Before the publicity from the original story, a few users who guessed this URL got to see this feature in the works. The URL now forwards to your normal list of friends.
  • Friends Lists can be used for composing messages. The URL, when working, suggested this was the main functionality.

  • Friends Lists can be used as for privacy control. This wasn’t certain in the beginning, even though it’s a great, obvious feature. There was some fear that the lists might only be used for composing messages. The phrase “so you can decide more specifically who sees what” on the What’s New page verifies our hopes.

    Privacy has always been a founding block of Facebook. Their closed-network structure essentially made the site, and the News Feed backlash emphasized the sensitive role of privacy on the site. Friends Lists will be a perfect way to block your parents from seeing those wild photos of that trip you took on their charge card without telling them.

• What we don’t know:
  • Will we compose these lists from scratch, or will they be generated automatically from the current social graph features? When you add a friend right now, the functionality already exists to ask how you know that new friend. You can say you’ve gone to school with them, on a trip with them, hooked up with them, etc. I think that they’ll introduce something new, though, since the older feature isn’t widely used. If we’re lucky, though, maybe you’ll be able to import this information.

Since Facebook has already taken a month on this without releasing it, it’s safe to say this is a big feature. By naming it an upcoming feature, they probably don’t have plans to release it in the next few days, but my completely ungrounded guess would put the release somewhere between two and six weeks. ∞

People seem to be talking about the small, routine case Facebook just won for the domain face-book.com. Domain squatting is an extremely popular activity, despite the fact that it’s illegal. When a company goes large, squatters assume the company will pay them for their slightly different domain name. Squatters fail to realize that a simple complaint to the UN’s WIPO can invalidate their ownership. Squatting’s annoying, but it’s not going to change any time soon.

This reminds me of some of the other domains that Facebook owns, that as of yet haven’t been discussed. Face-book will surely forward to the Facebook homepage, but these other domains are more ambiguous. Here are some of the interesting ones. You can find all of these by searching for domains sharing namesevers with facebook.com.

• zuckerberg.net - The big, mysterious domain. Does it belong to Mark? Will it one day be his personal blog?
• somethingtoputhere.com - This one is likely for side projects, and maybe some private testing outside of the production environment. The name you might recognize from one of the original curious quips at the bottom of a Facebook page (“One day I’ll find something to put here”). The domain is already used to host the Which FB Employee Are You? application.
• facebooklet.net - At first, it sounds like “bookmarklet,” but the bookmarklet Facebook gives to its users uses the usual facebook.com domain. The kind of fun name sounds like a fan or employee’s nickname. Maybe it too could be a weblog one day? Why is it connected with the official facebook servers, though?
• facebookmail.com, which handles “over 50M email messages a day” according the probably outdated Facebook jobs posting for an email admin
• wirehog.net - I’m not sure if anybody remembers Wirehog, but Facebook endorsed it as a way for users to share files. These days, the project is dead… for now.
• Facebook also has lots of subdomains. Most of them are probably just what they sound like, but a few raise some eyebrows.
  • mirror.facebook.com - It’s not commonly known that facebook is a mirror for several software organizations. With their other contributions to the community, such as memcached, phpsh, Thrift, and PHPEmbed, this reflects great on the company
  • code.facebook.com
  • presence.facebook.com
  • extftp.facebook.com - obvious, but interesting to know
  • moat.facebook.com
  • themob.facebook.com - hopefully they won’t come to get me for publishing this

If you know more about the domain repertoire, feel free to contact me. ∞

Typeahead, Quick Friends, Friend Jumper — who knows what to call it, but Facebook has launched one of those very small, but extremely useful features.

This new friend finder integrates beautifully into the existing search bar. I can’t imagine what percentage of my searches were for friends, so this is certainly a welcomed addition. One of the great things about this new feature is that you can see just what you need to. Before with search, if I searched for “Annie,” my first result was always “Anne,” then “Annie.” With this, I can see exactly what my results are, and which one I’ll be going to if I hit enter. I can also see primary networks if I’m feeling particularly braindead. As you see in the picture, another great part of this is that it matches any part of the name. Whatever you call it, this is another great, recent addition to the site. ∞