This release adds a bunch of features that should hopefully help.  Even for the case where there are multiple authors editing the same app, there’s now at least a warning at the bottom of the file list alerting you to that fact:

However, we’ve also gotten the feedback that people prefer working within their own clone of an app anyway (rather than co-editing the same app)… but that they then need the ability to merge changes between apps.  Taking some baby steps in that direction, we’ve upgraded the Changes tab in the App menu to list both parent (cloned from) and children (cloned to) apps:

When you click the View Changes button next to an app you will see a new dialog that lists all of the affected files.  This includes new files, missing files, changes in content and even changes in metadata (e.g., different mime-type).   Clicking on a filename shows a diff of the changes between the current file and what it will look like after applying the changes.  Any merge conflicts are inserted at the top of the file.

You can then choose the files to which you’d like to apply the changes to your local copy (in-browser).  If you choose “Apply & Save,” then the changes are also immediately saved back to the graph.  In this release, there is no two-way merge capability… it’s only pulling changes in one direction.  However, if you’re an admin of both apps, you can still click the ID of the parent or child app in the Changes tab to switch to the other one and pull changes in that direction.

Finally, given all this new app-wide diff functionality, we also threw in a couple of long-requested features for diff’ing different versions of the same app.  There’s both a button in the Versions tab for viewing all the changes from the previous version before creating a new version… as well as a button in the Changes tab for viewing all app history. Similar to how applying changes works, you can then revert to the previous version if you so choose:

Nearly all of these new features were built using a new (for us, at least) diff/merge/patch library. It’s been a nice library to work with so far, however it’s really geared more for free text than code, so there may be some gotchas.  We’ve already been tweaking it some to address these, but please let us know about any specific problems you run into at http://bugs.freebase.com.

There has been considerable interest and discussion within the Freebase community about connecting Freebase to Wordnet. Wordnet contains a rich set of linguistic relationships between concepts but is lacking in instances. A concept in WordNet is represented by synset, which is a set of synonymous words. For example, connecting the synset corresponding to “democracy” to “/en/democracy” allows you to find out in Freebase the organizations, books, quotations, nobel prizes won related to democracy; while WordNet tells you that “democracy” is a “political orientation”, and therefore has sibling concepts of “monarchism”, “collectivism”, “communism”, among many others, a piece of knowledge not currently modeled in Freebase.

Reconciling synsets to topics is challenging, and we are setting exacting standards for linking them. We need your help in reconciling WordNet synsets with Freebase Topics. The WordHunger application allows you to pick the best matching Freebase topic for a WordNet synset out of candidates generated by our algorithm. Your judgments are going to be used to train and tune the Wordnet-Freebase reconciliation algorithms. WordHunger is a “matchmaker” application, which means it allows you to pick the best matching candidate, it is available here: http://wordhunger.freebaseapps.com/

We have loaded the entire WordNet graph into Freebase, which is browseable using: http://wordnet.freebaseapps.com/

The schema for a Wordnet synset is available here: http://www.freebase.com/type/schema/base/wordnet/synset

Note the /base/wordnet/synset/equivalent_topic property. The following query shows you the links that we have reified at any point:

 [{
   "type" : "/base/wordnet/synset" ,
   "/base/wordnet/synset/gloss" : null ,
   "equivalent_topic" : {
       "optional" : "required" ,
       "id" : null ,
       "name" : null
   }
 }] 

WordNet is about 117,000 synsets, and as of now this query will show you about 4,000 links for the synsets that we have identified to be equivalent to a Freebase topic. With your help, we expect to dramatically increase this. Please go to http://wordhunger.freebaseapps.com/ and match some synsets to topics and tell us what you think!

Thanks,

- Praveen

One of our talks was by Barak Michener, who presented JGD, a Python implementation of a simple graph database and MQL interface. JGD lets him work with Freebase data offline, mash it up with other data, and all sorts of other clever things.

Here are his slides:

The source code — just 700 lines of Python — is available on github.

This is an awesome hack. Thanks to Barak for presenting it, and thanks also to our other speakers, Kirsten Jones and Will Moffat. Unfortunately their presentations were in the form of live demos, so we don’t have slides for them.

Next month’s meetup will be April 14th. As usual, we’re on the lookout for speakers, so if you’re in San Francisco Bay Area and would like to show us how you’re using Freebase, drop me a line.

Last year a twitter conversation among three data geeks accidentally turned into a 30+ person data meetup — with no planning, nowhere near enough chairs, and yet a total success. Clearly, SxSW needs a gathering of data geeks.

So here it is, bigger, better, and with free food. We’re still keeping it simple, though: get a whole bunch of really smart data geeks together, set up group discussions and a round of lightning talks, but leave the majority of time for people who admire each other’s work to meet and exchange ideas.

It’s at Opal Divine’s on Sunday March 14th, 6-9pm. RSVP here. I know there’ll be a couple of Freebase people in attendance as well as many from other open data projects.

Meanwhile, at home in San Francisco, we’ll be having our regular monthly SF Freebase meetup on Wednesday March 10th. The meetup starts at 6:30 with pizza and beer, then from 7pm we’ve got some great talks:

• Will Moffatt will be talking about MQL extensions
• Kirsten Jones will be demoing a Netflix/Freebase mashup
• Barak Michener will be showing off a Python implementation of a mini triple store and MQL which you can use to query Freebase data offline, or to mashup your own private data against Freebase.

And as usual we’ll wind down with some free time for Q&A, hacking, or whatever. RSVP at meetup.com to let us know you’re coming!

This Wednesday, Feb 10th, is the second Wednesday of the month which means it’s time for the monthly SF Freebase meetup. This month our talks include:

• A new Freebase plugin for Wordpress
• Using machine learning techniques to type topics in Freebase
• Reconciling Wordnet data with topics in Freebase
• Connecting open government data (especially Recovery Act data) with Freebase

As usual there’ll be pizza, beer, wifi, and time for general Q&A/demos/hacking.

RSVP on the meetup page.

First up, R. Morgan announced announced a beta release of phpFreebase library.

I just wanted to inform the group that a beta release ver.: 0.6.0 of phpFreebase has been released and the project is now hosted on SourceForge.net at: http://phpfreebase.sourceforge.net

The library will currently allow querying of the Freebase system. I will soon be adding write operations and cleaning up the code. This is the first release of the php class library code so many changes may be made between this and the first release version 1.0.0.

Shawn Simister also posted his Java API and power tools:

The Java API currently supports reads and writes through the Freebase API. MQL queries can be loaded from files and parameterized for both read and writes. Cursors are also supported but error handling still needs some work.

The Power Tools project contains a collection of cross-platform, command-line tools that I’ve built to help me work with large data sets. Most notably, it includes my spreadsheet loader which allows you to map spreadsheet columns to MQL parameters and upload complex data sets.

More information about libraries and tools for various languages can be found on the Freebase documentation wiki.

Here’s the slide deck from Alec’s talk:

You can also check out the Objective C Freebase library at Google Code. The iPhone app itself is, I believe, forthcoming.

Also, a reminder that meetups are held on the second Wednesday of each month at Freebase HQ in San Francisco. Join our meetup group.

A reminder that we have a meetup happening in San Francisco next Wednesday, December 9th, at Freebase HQ. We have three talks lined up:

1. Alec Flett will talk about an iPhone app using Freebase’s geodata to encourage photo contributions
2. Arthur van Hoff will demo the Ellerdale Project, a new website mashing up Freebase with realtime data streams
3. I’ll talk about guessing people’s gender based on Freebase’s data about given names

Details at meetup.com — please RSVP there!

Community meeting next Monday

Next Monday is the first one of the month so our regular internal Monday community meeting will be open to all via Skype.

When: 2:30pm PST (GMT-8), Monday December 7th

Where: Metaweb, 631 Howard St, 4th floor, San Francisco

RSVP: Please email kirrily@metaweb.com before 2pm Monday if you plan to attend corporeally or digitally (and provide your Skype username in the latter case).

More details at Community meeting via Skype on our wiki, or see the agenda.

NY workshop is full

Our full-day workshop in New York has filled up quickly, and we have no more open slots for general signup. However, if we get any dropouts we’ll be sure to let you know! If you’d like to be put on a waiting list, please email Robert.

Freebase has passed a notable milestone.  On Sunday, at about 11:00am PST, we zoomed by our 10 millionth topic — and by the time you read this post, we should surpass the 11 million topic mark.  A year ago, Freebase stood at just over 4 million topics.  That’s an annual growth rate of over 100%.

A great deal went into achieving this milestone — contributions from prolific community members like tfmorris, pak21 and sprocketonline (see our contribution leaderboard for more); new Data Team tools like the recon service, RABJ, and the spreadsheet loader; and continued growth in traditional data sources like Wikipedia.  But the largest segment of growth came from our continuing efforts to build a comprehensive repository of high-quality information about media in all its forms — especially music, movies, TV and books.

In October, we rounded out our TV domain by synchronizing with the excellent user-curated TV fan site TVRage.com.  Combined with earlier data loads from thetvdb.com, we now have comprehensive coverage of nearly every TV show and episode created in the United States.  It includes cast and credits, as well as links to key TV websites like tvguide.com and Hulu — nearly a million topics in all!

But the load that took us over the 10 million mark was the final load of editions from Open Library.  Compromising 650,000 authors, almost 2 million books and 2.1 million book editions,   this load pushed new boundaries in our data acquisition, curation, reconciliation and QA processes.

In the months ahead, we’ll be continuing to both curate and extend our media data loads with more high-quality data sets.  We plan on continuing to reconcile authors and books already in Freebase, as well as loading more books from curated bibliographic catalogs.  We’ll also be fleshing out our data about movies with data from Netflix, as well as restarting our regular synchronizations with MusicBrainz and their Next Generation Schema.

Congratulations to everyone who helped get us to this point.  It’s been an exciting year — with more great data to come!