Freebase launches today, a new startup that intends to organize the world’s information and make it universally accessible and useful. If that last bit sounds familiar, it’s because it’s actually Google’s mission, but Freebase seems intent on doing it, too.
Like Google Base, Freebase is a massive database. The purpose of the database is to centralize as much data as possible, and allow participants to freely add and access data - developers can extract information from Freebase via a set of APIs and add it to their web applications. It also builds relationships between highly structured pieces of data, something that can’t easily be done with distributed data controlled by different entities. Tim O’Reilly gives a great and in depth overview of the service and why it’s important. The Time’s John Markoff explains it to the masses.
O’Reilly says:
But hopefully, this narrative will give you a sense of what Metaweb [the company that created Freebase] is reaching for: a wikipedia like system for building the semantic web. But unlike the W3C approach to the semantic web, which starts with controlled ontologies, Metaweb adopts a folksonomy approach, in which people can add new categories (much like tags), in a messy sprawl of potentially overlapping assertions.
Now, the really powerful thing about this is that all these categories, these data types and the web of fields that define them, provide new hooks for applications that will be able to extract meaning from the data. That’s what makes Metaweb a kind of semantic web application.
If Metaweb gets this right, this bottom up approach will build new connections between data, new categories and ways of thinking. It will likely be messy and contradictory for a while, but as I told John Markoff for the story on Metaweb that he was preparing for the New York Times tonight, they are building new synapses for the global brain.
Freebase has already sucked in data from Wikipedia and other sources, and individuals can fill in their own data, too. No word yet on how bad data will be purged from the system.
Freebase looks to be what Google Base is not: open and useful. I imagine there will be more than one forehead self-smacked at Google HQ tomorrow, as they think “We could have done this.”
Sounds awesome, but does it actually work? It’s super easy for these tech startups to claim the world - but delivering is another thing altoghter.
Review Mike?
Haven’t seen it, don’t have access. See O’Reilly’s post linked above. He does a walk through with screenshots.
Yes, sorry jumped the gun. The O’Reilly article looks amazing - hope it takes off - although still mildly skeptical as to whether it *really* works.
I can already think of a dozen or so mashups possbile with the api.
So… wikipedia with an API?
wikipidia + lots of other data sources, structured, with an API both for data in as well as data out.
“Danger, Will Robinson.” It looks very cool but gave me a server error when I tried to request an invite. When do we get to see it?
Could they not come up with a better name than freebase? The first thing that comes to mind when I hear that is certainly not a massive online database… I have a feeling that this site is going to get a lot of traffic from people looking to freebase cocaine, crack cocaine, or heroin.
And why are companies content on copying what’s already been done? Being open and useful is not enough of a differentiation to even consider using this site. Hell, Google’s offering isn’t even compelling enough for me to use it either. And why would people use this over the already well-established Wikipedia?
Unless Freebase does it better, makes it easier, and actually innovates somehow… it’ll be in the deadpool before you know it.
The invite request form gives out errors.
The service is interesting because for the first time I see some practical implementation of the so-called ’semantic web’, which makes it easier to understand what it meant by the concept.
Ok…It’s early, I admit it. They are close beta (or alpha)… but… The following message is becoming more annoying every day.
“It appears you’re using an unsupported browser. We recommend one of the following: Firefox 1.5+, Internet Explorer 7.0+”
For a huge database to be filled… they must reach adoption like wikipedia for example, and this is NOT the way to do it.
“Freebase”? Is Richard Pryor spinning in his grave yet?
It is obviously going to take something more that TrustRank – and Backlinks algos to beat Google – this hi tech is one that could begin a new disruption ….. like Link Popularity did in the late 1990s
It could be the dawn of a the semantic web era with new approaches to information architecture and retrieval algos
I think this is an interesting project. I agree that there may be some “d’oh!” moments for folks this morning, as on the surface, the idea of working to relate information in a database to each other instead of just listing them seems like the next logical step in info-taxonomy.
Web 2.0 when explained in its simplest form is a piece of cake but when explained by Tim it gets really complicated. Thankfully, this time around, maybe with limited time, he has done a better job.
Notwithstanding, it would appear Freebase is slightly ahead of its time. I am not sure how any semantic element built into their application will work effectively in a vacuum. From my limited perspective, it would appear that the “backbone” of the Web will have to change substantially to make what they are proposing worthwhile.
It’s a great initiative, but seems Freebase needs to give Tim Berners-Lee and his team more time.
Shoot…. once again, I am a day late and a dollar short.
I came up with this EXACT idea (about a month ago), and went to register the name “Freebase”… and it was taken. I jumped on IFreebase … I think those would be cool T-shirts… “IFreebase, do you?” or.. “iFreebase… it’s addicting”.
Database, plus wiki… it’s a phenominal idea. and I had it too! These guys are geniuses!
Name aside, it’s going to be fun to watch this one.
dangit: you are not the only one who had this idea
Love the headline!
Exacly: Google could have done this.
An open system, and a Creative Commons - Attribution license. Which means I can build a commercial mashup. That buys you a lot of goodwill.
If this is also easy to query, you can bet a lot of people will be uploading their data too, and/or using their URI’s as a namespace.
Sad that it needs a registration to access the data. You can create an account here http://www.freebase.com/signin.html
But why would I need an account to read?
I hope they will support OpenId in near future.
If they are actually any sort of competition, Google will Gooble the up.
-RB
mike….
ummm.. so you think/believe that within google there will be head slapping!!
yeah.. ok. whatever!!!
can this work, sure, possibly. are there extremely huge issues to resolve, yep!
is there really a business model here, perhaps.
but i don’t see google going ‘doh’ on this one!!!
just my $0.02 worth…
peace..
This could be massive. Wikipedia has shown that this kind of mass-edited global resource can work; adding semantic information instantly turns it into a killer app for the Internet. There are doubtless issues to work out, and just as with Wikipedia, not all of the data will be accurate - but what an excellent idea. Good luck to them.
Verbatim dump of my comments to Tim:
Tim,
Please see the Data Web (Web 3.0) project called DBpedia.
This project is about Open Data Access and a Flexible Data Model (RDF). It exposes Wikpedia as a true Database or Web Data Space.
Data needs to be unshackled and this is what the next frontier (Web 3.0) is about; Meshing rather than Mashing which is the only option when you don’t have an Open Data Access principle used in conjunction with a flexible Data Model).
Also look at the Data Web based Start Pages in my blog posts re. Oscar Winners or Major League Baseball. It is all here and happening now
Also look at:
Hello Data Web - Take 3
OpenID & Personal URIs
so the technology is cool. but how will they get people to actually use the thing? the web is littered with cool social products that never reached critical mass. and the more entrants, the harder it is to break through the noise.
man, you people really drink the kool-aid on every new product.
looking forward to being proven wrong. but not expecting it.
I didn’t know John Markoff was in The Time! I
@20 RB - yep. I say google solves almost everything, and google _will_ give a leg and an arm to keep the opinion that way.
Otherwise, they won’t be holding on too long.
I like how their site says, “Information wants to be free” but then they don’t make it readily free in terms of availability. (1) They require a login membership to gain entrance. (2) Their site pops up a warning dialogue that the site isn’t compatible with my Safari browser.
Sounds to me like it’s “information that doesn’t cost money but is hidden behind barriers of entry.” Not very freeing.
dangin, secreeeet — you guys are not the only ones to have had this idea
I must have already written 30 pages of text around this exact idea…identical.
well, clearly it was in the cards…
I mean, you mix wikipedia with structured data, and a query tool, and maybe a ranking tool (like Youtube or digg), and add in a web 2.0 interface, and an api… and every buzz word you can think of… and this is what you have!
But I mean, I even tried to get this domain name!
And yes, I wrote 31 pages about this. I even showed them to my girlfriend, Morgan Fairchild. you can even ask her….
Sounds boomin’… but will it just go “BOOM” like so many other great ideas…?
The interface doesn’t look good and have to register to see what it is. I just don’t feel it is interesting.
ok Dangit - I concede. I wasn’t going for this domain name. You win the hypothetical entitlement award (meaning that you would be entitled to pursue this idea, over me, if they weren’t doing so)
I agree with everyone else on this comment list though, in the seemingly pervasive sentiment that the execution of this objective is by no means a shoe-in. The NYT article (or was it the Oreilly bit) seems to reveal that the Disney Imagineer has failed at at least one other venture…
I got a message saying it doesnt support IE 6. BAHA
To achieve consciousness, Feebase should partner with Hawkins’ Numenta. They have a project called HTM (Hierarchical Temporal Memory) a computer modeled after the human neocortex.
http://www.numenta.com/about-numenta.php check it out.
sounds pretty simplistic to me….semantic web + wikipedia is not groundbreaking. can someone explain to me what was so cool in o’reilly’s walkthrough?
google’s pretty “smart” as it is, and “freebase” will never work if users have to explicitly specify things. google works because the system does this itself.
as of now this is just another run-of-the-mill attempt at semantic web, of which there are currently no useful examples. but I say we get all excited and give it a $500 million valuation anyway!
? I’m rooting for computer consciousness ASAP but I think Blue Brain will get there before Freebase. They won’t kill us as long as we start acting more rationally and don’t try to …. unplug ‘em.
Good title Mike!
Just watch them introduce daily query limits along with a predefined timing gap between queries, and maximum concurrent query limitations, effectively making developers their bitches.
They will then, at the pressure of current or future investors, implement a cliché tiered monetary query chart outlining the cost for XXX queries over the default. Heck, they’ll probably even state that only non-commercial ventures can use the data freely, so they can charge larger companies.
The database has always been a bottleneck for all developers. They try and escape the DB any way they can, by caching or otherwise. By allowing anyone website in the world run queries on your DB an unlimited amount of times, is brainless.
If they can promise me, and other developers contrary to the former, then and only then is it something to get excited about. Settle down people.
Check out http://www.cyc.com to see a group of people who have been working on a pretty similar idea for 20 years and >$100M of funding. I’m not affiliated with Cycorp, but I think it deserves a mention along with this post.
It’s pretty clear that information was going to move in this direction, but I am surprised at the speed at which it is occuring. Google should be slapping their foreheads for a lot of things - one of which is whatever reconfiguration they’re doing with local search results. Check out the forum on webmasterworld. No offence - I would have commented if you had covered it.
I’m curious how much access Freebase is going to give - or if its another “let’s all build it and then I’ll just use it” project of the Tech500. I hope to see some discussion on ethics as we move into the new web.
The problem I have with this is that It seems to to be counter intuitive to a distributed web.
It would make more sense if it was done by a search engine/web crawler that understands microformats and semantic analysis of unstructured data.
More here: http://www.touchstonelive.com/.....ol-of.html
I believe there are some intriguing connections between the Freebase model and Creative Commons licensing - together they are greater than the sum of their parts.
See my post at:
http://community.eqo.com/blog/.....nce_lessig“>http://community.eqo.com/blog/jeff/freebase_the_best_thing_to_happen_to_the_creative_commons_since_lawrence_lessig
- Jeff LaPorte
@Jared - good call, cyc is an impressive project.
@Mike #38 - I agree that the first bottleneck they will come across will be scaling database access to any reasonable set of users.
This Wikipedia entry provides an excellent overview of existing efforts in this space http://en.wikipedia.org/wiki/Folksonomy.
Google may be, in the process of making a MEGA change in the way todays Internet is working….with some of its key aquisitions…
1. Fiber Optic and WiFi pages
http://news.com.com/Google+wan.....37392.html
2. Advanced text search algorithm invented by Ori Allon
http://news.com.com/Google+wan.....37392.html
e.t.c
IF it creates a global network and controls what to put on that internet (controlled environment internet, cinternet or centernet) and combines all of worlds related data and builds in a structured way then no viruses, spyware, OS, nothing…and no competetor and no need of semantic web
just a thought!
Hmmm…”Open” is a great thing, though I’m still not sure about the usefulness of these “base” products. Though, I felt the same way about the original Google CSE projects. Now I think they are great (though still in need of much more refinement).
It will be interesting watching for this one.
Though copying the google concept and moto isn’t really good for karma, nor it will help to get people’s attention since it is something similar - that means only those who are unsatisfied with the current services providers will be interested in trying it out.
How are they going to deal with the obvious trademark violation of the prior work by the same name that is here: http://alumni.cse.ucsc.edu/~jjf/freebase/
Drama 3.0: Hey, if you want to go in on it together, I’m quite sure we could do it better than these guys.
I have the domain name, I have the capital. I need to get the team together. And since they are giving away all their content, we could build on that, and do it better.
What do you say?
Structured content is good…
This is a subject that’s near and dear to my heart and while I applaud what project like Google Base & Freebase are trying to do I’m not sure that this is the best long term approach. As #41 stated, the problem I have with this is that they’re trying to centralize everything into silos of structured content. It seems like this goes against what the web is all about which is de-centralized information.
A better approach would be to crawl this content from websites. And yes looking for Microformats on a page could be one approach but the problem I have with Microformats is that the types and their structure have to be defined by a committee. What I like about Google Base in particular is the democratic approach they take to defining types. Types are defined by the masses and can evolve over time.
Google Base, Freebase, or one of these projects should break their data structures out to an open file format that can be authored by rich client & web apps, uploaded & linked to by web pages, and crawled & indexed by Google, Live, Yahoo, Vista (so I can search for it locally) or any other search engine.
-Steve
It is high time to clean up WEB MESS. I presume your approch is correct. Hope Google and Yahoo will contribute their existing web powers to you, so that ultimate success is possible with your efforts.
I had a similar dream for advertisers “a competitive market place which will result equal opportunities”. I am working towards COPYRIGHTS / ROYALTIES. When I find a correct party to finance, project will go ahead.
Wish you all the very best for your project.
Intriguing technical problems don’t exist in social vacuums. Questions of who controls?, to what end?, at what social cost?, etc. are not techie’s strong points. The simple fact of false information — a big enough problem in “real” life, let alone virtual — and its power is conveniently ignored by the commentators so far. The human propensity to centralize and dominate has been touched upon as well as the undercurent of commercial exploitation of “free” semantic webs. These considerations mute any enthusiasm I might have to a pensive “Hmmm.” I’ll do what I can to keep my own information under my own control for my own ends, thank you.
The concept (I think) implemented on a much smaller scale… WyaBase using Information Magnets… The idea was to use WIKIs and information magnets to collect and tag all kinds of data and let them build their own relationships… then create feeds and applications out of the relationships built…
FYI…Metaweb will have their first public unveiling at the March 21st STIRR event in Palo Alto - http://www.stirr.net/event/12.
I’m using safari which the website claims is unsupported but I can still login. They restricted access to the service to stop the servers becoming overwhelmed (for the alpha). I think total access (for the moment) is limited to 6000.