Do you trust your friends with your URLs?

"Facebook's data feed a data leak?" over at Lawgarithms:

Please correct me if I’m wrong about this; I want to be wrong about this. Or I want to learn that Facebook has already considered and dealt with the issue and it’s just not readily apparent to me. But I’m thinking that Facebook’s feeds for Status Updates, Notes, and Posted Items must in many instances be at odds with privacy settings that attempt to limit users’ Facebook activities to “friends only” (or are even more restrictive).

Denise is both right and wrong.  The basic issue is that once you give out a feed URL (which is not guessable) to a friend, they can then give it out to their friends and their friends... ad infinitum.  These people can then get your ongoing updates, without you explicitly adding them.

Of course, this requires your friends to breach the trust you placed in them to guard your bits.  Notice that even without feeds, your friends can easily copy and paste your bits and send them on manually.  It's a simple matter to automate this if a friend really wants to broadcast your private data to whoever they want.  So as soon as you open up your data, you are vulnerable to this.  To prevent it you'd need working DRM; not a good path to go down.

It would be possible to access control the feeds; there's even a nascent standard (OAuth) for doing this in a secure and standards compliant way.  But even this doesn't prevent your friends from copying your bits.

A much simpler approach is to hand out a different URL for each friend.  They're still obfuscated of course.  You can then block a friend (and anyone they've shared the URL with) from seeing future updates at any time.  This is about the best that can be done.  Update:  This is apparently exactly what Facebook has done.  Denise is still concerned that friends could accidentally or purposefully re-share the data, since the feed format makes it easy to do so.

Facebook's messaging could definitely be improved.  Suggestions?

Tags: facebook, feeds, security, friends, oauth

RESTful partial updates: PATCH+Ranges

Over the past couple of months, there's been a lot of discussion about the problem of partial updates in REST-over-HTTP [1][2][3][4][5].  The problem is harder than it appears at first glance.  The canonical scenario is that you've just retrieved a complicated resource, like an address book entry, and you decide you want to update just one small part, like a phone number.  The canonical way to do this is to update your representation of the resource and then PUT the whole thing back, including all of the parts you didn't change.  If you want to avoid the lost update problem, you send back the ETag you got from the GET with your PUT inside an If-Match: header, so that you know that you're not overwriting somebody else's change.

This works, but it doesn't scale well to large resources or high update rates, where "large" and "high" are relative to your budget for bandwidth and tolerance for latency.  It also means that you can't simply and safely say "change field X, overwriting whatever is there, but leave everything else as-is".

I've seen the same thought process recapitulated a few times now on how to solve this problem in a RESTful way.  The first thing that springs to mind is to ask if PUT can be used to send just the part you want to change.  This can be made to work but has some major problems that make it a poor general choice. 

• A PUT to a resource generally means "replace", not "update", so it's semantically surprising.
• In theory it could break write-through caches.  (This is probably equivalent to endangering unicorns.)
  
• It doesn't work for deleting optional fields or updating flexible lists such as Atom categories.

The next idea is generally to simply use POST to update the resource.  This does work in many cases, but conflicts with the use of POST to add a resource to a collection.  That is, if you POST to a collection, are you trying to add an element to the collection, or perform some other update to the collection's metadata?  It's possible disambiguate using MIME types but it feels fragile.  It also doesn't capture the fact that the operation is retryable; POST in general is not retryable.

A good solution to the partial update problem would be efficient, address the canonical scenario above, be applicable to a wide range of cases, not conflict with HTTP, extend basic HTTP as little as possible, deal with optimistic concurrency control, and deal with the lost update problem.  The method should be discoverable (clients should be able to tell if a server supports the method before trying it). It would also be nice if the solution would let us treat data symmetrically, both getting and putting sub-parts of resources as needed and using the same syntax.

There are three contenders for a general solution pattern:

Expose Parts as Resources.  PUT to a sub-resource represents a resources' sub-elements with their own URIs.   This is in spirit what Web3S does.  However, it pushes the complexity elsewhere:  Into discovering the URIs of sub-elements, and into how ETags work across two resources that are internally related.  Web3S appears to handle only hierarchical sub-resources, not slicing or arbitrary selections.

Accept Ranges on PUTs.  Ranged PUT leverages and extends the existing HTTP Content-Range: header to allow a client to specify a sub-part of a resource, not necessarily just byte ranges but even things like XPath expressions.  Ranges are well understood in the case of GET but were rejected as problematic for PUT a while back by the HTTP working group.  The biggest concern was that it adds a problematic must-understand requirement.  If a server or intermediary accepts a PUT but doesn't understand that it's just for a sub-range of the target resource, it could destroy data.   But, this does allow for symmetry in reading and writing.  As an aside, the HTTP spec appears to contradict itself about whether range headers are extensible or are restricted to just byte ranges.  This method works fine with ETags; additional methods for discovery need to be specified but could be done easily.

Use PATCH.  PATCH is a method that's been talked about for a while but is the subject of some controversy.  James Snell has revived Lisa Dusseault's draft PATCH RFC[6] and updated it, and he's looking for comments on the new version.  I think this is a pretty good approach with a few caveats.  The PATCH method may not be supported by intermediaries, but if it fails it does fail safely.  It requires a new verb, which is slightly painful.  It allows for variety of patching methods via MIME types.  It's unfortunately asymmetric in that it does not address the retrieval of sub-resources.  It works fine with ETags.  It's discoverable via HTTP headers (OPTIONS and Allow: PATCH).

The biggest issue with PATCH is the new verb.  It's possible that intermediaries may fail to support it, or actively block it.  This is not too bad, since PATCH is just an optimization -- if you can't use it, you can fall back to PUT.  Or use https, which effectively tunnels through most intermediaries.

On balance, I like PATCH.  The controversy over the alternatives seem to justify the new verb.  It solves the problem and I'd be happy with it.  I would like there to be a couple of default delta formats defined with the RFC. 

The only thing missing is symmetrical retrieval/update.  But, there's an interesting coda:  PATCH is defined so that Content-Range is must-understand on PATCH[6]:

The server MUST NOT ignore any Content-* (e.g.  Content-Range) 
headers that it does not understand or implement and MUST return 
a 501 (Not Implemented) response in such cases.

So let's say a server wanted to be symmetric; it could advertise support for XPath-based ranges on both GET and PATCH. A client would use PATCH with a range to send back exactly the same data structure it retrieved earlier with GET.  An example:

GET /abook.xml

Range: xpath=/contacts/contact[name="Joe"]/work_phone

which retrieves the XML:

<contacts><contact><work_phone>650-555-1212</work_phone>
</contact></contacts>

Updating the phone number is very symmetrical with PATCH+Ranges:

PATCH /abook.xml

Content-Range: xpath=/contacts/contact[name="Joe"]/work_phone

<contacts><contact><work_phone>408-555-1591</work_phone>
</contact></contacts>
  

The nice thing about this is that no new MIME types need to be invented; the Content-Range header alerts the server that the stuff you're sending is just a fragment; intermediaries will either understand this or fail cleanly; and the retrievals and updates are symmetrical. 

[1] http://www.snellspace.com/wp/?p=683
[2] http://www.25hoursaday.com/weblog/2007/06/09/WhyGDataAPPFailsAsAGeneralPurposeEditingProtocolForTheWeb.aspx
[3] http://www.dehora.net/journal/2007/06/app_on_the_web_has_failed_miserably_utterly_and_completely.html
[4] http://tech.groups.yahoo.com/group/rest-discuss/message/8412
[5] http://tech.groups.yahoo.com/group/rest-discuss/message/9118
[6] http://www.ietf.org/internet-drafts/draft-dusseault-http-patch-08.txt

Some thoughts on "Some Thoughts on Open Social Networks"

Dare Obasanjo:

"Content Hosted on the Site Not Viewable By the General Public and not Indexed by Search Engines:  As a user of Facebook, I consider this a feature not a bug."

Dare goes on to make some great points about situations where he's needed to put some access controls in place for some content.  I could equally make some points about situations where exposing certain content as globally as possible has opened up new opportunities and been a very positive thing for me.  After which, I think we'd both agree that it's important to be able to put users in control.

Dare: "Inability to Export My Content from the Social Network: This is something that geeks complain about ... danah boyd has pointed out in her research that many young users of social networking sites consider their profiles to be ephemeral ... For working professionals, things are a little different since they may have created content that has value outside the service (e.g. work-related blog postings related to their field of endeavor) so allowing data export in that context actually does serve a legitimate user need."

It isn't just a data export problem, it's a reputation preservation problem too.  Basically, as soon as you want to keep your reputation (identity), you want to be able to keep your history.  It's not a problem for most younger users since they're experimenting with identities anyway.  Funny thing, though:  Younger users tend to get older.  At some point in the not so distant future that legitimate user need is going to be a majority user need.

Dare: "It is clear that a well-thought out API strategy that drives people to your site while not restricting your users combined with a great user experience on your website is a winning combination. Unfortunately, it's easier said than done."

+1.  Total agreement.

Dare: "Being able to Interact with People from Different Social Networks from Your Preferred Social Network: I'm on Facebook and my fiancée is on MySpace. Wouldn't it be great if we could friend each other and send private messages without both being on the same service?  It is likely that there is a lot of unvoiced demand for this functionality but it likely won't happen anytime soon for business reasons..."

Will there be a viable business model in meeting the demand that Dare identifies, one which is strong enough to disrupt business models dependent on a walled garden?  IM is certainly a cautionary tale, but there are some key differences between IM silos and social networking sites.  One is that social networking sites are of the Web in a way that IM is not -- specifically they thrive in a cross-dependent ecosystem of widgets, apps, snippets, feeds, and links.  It's possible that "cooptition" will be more prevalent than pure competition.  And it's quite possible for a social network to do violently antisocial things and drive people away as Friendster did, or simply have a hot competitor steal people away as Facebook is doing.  Facebook's very success argues against the idea that there will be a stable detente among competing social network systems.

Tags: socialnetworking, facebook, open api, social software, open social networks, cooptition

Relationship requires identity

Nishant Kaushik:

Let's face it, relationship silos are really just extensions of identity silos.  The problem of having to create and re-create my relationships as I go from site to site mirrors my problem of having to create and re-create my identity as I go from site to site. The Facebook Platform might have one of the better Identity Provider APIs , but all the applications built on it still have to stay within Facebook itself.

Yup.  Which is the primary reason that I've been interested in identity -- it's a fundamental building block for social interactions of all kinds.  And think of what could happen if you could use the Internet as your social network as easily as you can use Facebook today.  As Scott Gilbertson at Wired discovered, it's not hard to replicate most of the functionality; it's the people who are "on" Facebook which makes it compelling.

Tags: identity, silos, openid, facebook, social software, social networks

Google += Joe Gregorio!

+1!

cat Google Spreadsheets | Venus > my.feed

Sam Ruby (prompted by Alf Eaton) combines Google Spreadsheets and Venus to let people manage Venus subscription lists (or whatever) using Spreadsheets.  The lingua franca is of course CSV-over-HTTP.  Like Unix pipes running over Internet, um, pipes.

Note that this requires the data to be publicly readable on the Spreadsheets side, which is fine for this use.  A lot more uses would be enabled with a lingua franca for deputizing services to talk securely to each other.

Tags: cool, pipes, mashups, web as platform