The Time of Day

(a clock showing no time)

From my Skype logs [2008-06-19 Dan Brickley: 18:24:23]

So I had a drunken dream about online microcurrencies last night. Also about cats and water-slides but that’s another story. Idea was of a karma donation system based on one-off assignments from person to person of specified chunks of their lifetime; ‘giving the time of day’. you’re allowed to give any time of day taken from those days you’ve been alive so far. they’re not directly redistributable, nor necessarily related to what happened during the specified time. there’s no central banker, beyond the notion of ‘the public record’. The system naturally favours the old/experienced, but if someone gets drunk and gives all their time/karma to a porn site, at least in the morning they’ll have another 24h ‘in the bank’. Or they could retract/deny the gift, although doing so a lot would also be visible in the public record and doing so excessively would make one look a bit sketchy, one’s time gifts seem less valuable etc. Anchoring to a real world ‘good’ (time) is supposed to provide some control against runaway inflation, as is non-redistributability, but also the time thing is nice for visualizations and explanation. I’m not really sure if it makes sense but thought i’d write it down before i forget the idea…

One idea would be for the time gifts to be redeemable, but that i think pushes the metaphor too far into being a real currency for a fictional world where hourly rates are flattened. Some Lets schemes probably work that way I guess…

So I’ve been meaning to write this up, but in the absense of having done so, here’s the idea as it first struck me. I had been thinking a bit about online reputation services, and the kinds of information they might aggregate. Garlik’s QDOS and FOAF experiments being a good example of this kind of evidence aggregation. As OpenID, FOAF, microformats etc. take hold, I really think we’ll see a massive parting of waves, red sea style, with the “public record” on one side, and “private stuff” on the other.

And in the public record, we’ll be attaching information about the things we make and do to well-known identifiers for people (and their semi-detached aliases). Various websites have rating and karma mechanisms, but it is far from clear how they’ll look when shared in the public Web. Nor whether something robust and not-too-gameable will come out of it. There are certainly various modelling idioms (eg. advogato do their internal calculations, and then put everyone in one of several broad-brush groups; here’s my advogato FOAF). See also my previous notes on representing expertise.

Now in some IRC channels, there are bots where you can dish out credit by typing things like:

edd++ # xtechy

…and have a bot add up the credits, as well as the comments. In small IRC communities these aren’t gamed except for fun. So I’ve been thinking: how can these kinds of habits ever work in the wider Web, where people are spread across Web sites (but nevetherless identifiable with OpenID and FOAF). How could it not turn hideous? What limited resource do we each have a supply of? No, not kidneys. And in a hungover stupor I came to think that “the time of day” could be such a resource. It’s really just a metaphor, and I’m not sure at all that the quantifiable nature is a benefit. But I also quite like that we each have a neverending supply of the stuff, and that even a fleeting moment can count.

Update: here’s a post from Simon Lucy which has a very similar direction (it was Simon I was drinking with the night before writing this). Excerpt:

And what do you do with your positive balance? Need you do anything? I imagine those that care will publish their balance or compare it with others in similar way to company cars or hi fi tvs. There will always be envy and jealousy.

But no one can steal your balance, misuse it.

So who wants to host the Ego Bank.

The main difference compared to my suggested scheme, is just the ‘the Web’ and the public record it carries, are the “ego bank”, creating a playground for aggregators of karma, credibility and reputation information. “The time of day” would just be one such category of information…

Foundation Nation: new orgs for Infocards, Symbian

Via the [IP] list, I read that the Information Card Foundation has launched.

Information Cards are the new way to control your personal data and identity on the web.

The Information Card Foundation is a group of thoughtful designers, architects, and companies who want to make the digital world easier for you by building better products that help you get control of your personal information.

From their blog, where Charles Andres offers a historical account of where they fit in:

And by early 2007, four tribes in the newly discovered continent of user-centric identity had united under the banner of OpenID 2.0 and brought the liberating power of user-controlled identifiers to the digital identity pioneers. The OpenID community formed the OpenID Foundation to serve as a trustee for intellectual property and a host for community activity and by early 2008 had attracted Microsoft, Yahoo, Google, VeriSign, and IBM to join as corporate directors.

Inspired by these efforts, the growing Information Card community realized that to bring this metaphor to full fruition required taking the same step—coming together into a common organization that would unify our efforts to create an interoperable identity layer. From one perspective this could be looked at as completing the “third leg of the stool” of what is often called the Venn of Identity (SAML, OpenID, and Information Cards). But from another perspective, you can see it as one of the logical steps needed towards the cooperative convergence among identity systems and protocols that will be necessary to reach a ubiquitous Internet identity layer—the layer that completes the hat trick.

I’m curious to see what comes of this. There’s some big backing, and I’ve heard good things about Infocard from folks in the know. From an SemWebby perspective, this stuff just gives us another way to figure out the provenance of claim graphs representable in RDF, queryable in SPARQL. And presumably some more core schemas to play with…

Meanwhile in the mobile scene, a Symbian Foundation has been unveiled:

Industry leaders to unify the Symbian mobile platform and set it free
Foundation to be established to provide royalty-free open platform and accelerate innovation

The demand for converged mobile devices is accelerating. By 2010 we expect four billion people to have joined the global mobile conversation. For many of these people, their mobile will be their first Internet experience, not just their first camera, music player or phone.

Open software is the basic building block for delivering this future.

With this in mind, industry leaders are coming together to establish Symbian Foundation, to bring to life a shared vision and to create the most proven, open and complete mobile software platform – available for free. To achieve this, the foundation will unify Symbian, S60, UIQ and
MOAP(S) software to create an unparalleled open software platform for converged mobile devices, enabling the whole mobile ecosystem to accelerate innovation.

The foundation is expected to start operating during the first half of 2009. Membership of the foundation will be open to all organizations, for a low annual membership fee of US $1,500.

I’ll save my pennies for an iPhone. Everybody’s open nowadays, I guess that’s good…

BBC joining OpenID Foundation

The BBC have joined the OpenID Foundation. See blog post from Jem Stone for details. He cautions people not to get excited and expect too much too soon. However I can’t help but see this as a very healthy thing when thinking about the medium-term usability issues around OpenID. Talking of which, does anyone have pointers to real-world usability testing of OpenID?

When your OpenID provider goes offline…

 My main OpenID provider is currently LiveJournal, delegated from my own danbri.org domain. I suspect it’s much more likely that danbri.org would go offline or be hacked again (sorry DreamHost) than LJ; but either could happen!

In such circumstances, what should a ‘relying party’ (aka consumer) site do? Apparently myopenid has been down today; these are not theoretical scenarios. And my danbri.org site was hacked last year, due to a DreamHost vulnerability. The bad guys merely added viagra adverts; they could easily have messed with my OpenID delegation URL instead.

I don’t know the OpenID 2.0 spec inside-out (to put it mildly!) but one model that strikes me as plausible: the relying party should hang onto FOAF and XFN ‘rel=me’ data that you’ve somehow confirmed (eg. those from http://danbri.org/foaf.rdf or my LJ FOAF) and simply offer to let you log in with another OpenID known to be associated with you. You might not even know in advance that these other accounts of yours offer OpenID; after all there are new services being rolled out on a regular basis. For a confirmed list of ‘my’ URLs, you can poke around to see which are OpenIDs.

danbri$ curl -s http://danbri.livejournal.com/ | grep openid
<link rel=”openid.server” href=”http://www.livejournal.com/openid/server.bml” />

danbri$ curl -s http://flickr.com/photos/danbri/ | grep openid
<link rel=”openid2.provider” href=”https://open.login.yahooapis.com/openid/op/auth” />

Sites do go down. It would be good to have a slicker user experience when this happens. Given that we have formats – FOAF and XFN at least – that allow a user to be associated with multiple (possibly OpenID-capable) URLs, what would it take to have OpenID login make use of this?

“Stuff I’ve been thinking about” (SocialNetworkPortability WebCamp) – my slides

I’m in Cork, mainly for the excellent Social Network Portability event on Sunday, but am also staying through Blogtalk’08 which has been great. I’ve uploaded my slides from my talk (slideshare in Flash, included inline here, or a pdf). I have some rough speaking notes too,  maybe I’ll get those online. I have no idea how they relate to whatever actually came out of my mouth during the talk :) Apologies to those without PDF or Flash. I haven’t tried Keynote’s HTML output yet.

Basically much of what I was getting at in the talk, and my thoughts are only just congealing on this … is that the idea of a ‘claim’ is a useful bridge between Semantic Web and Social Networking concerns. Also that it helps us understand how technologies fit together. FOAF defines a dictionary of terms for making claims, as does xfn, hCard. RDF/XML, Microformats, RDFa, GRDDL define textual notations for publishing documents that encode claims, and SPARQL gives us a way of asking questions about the claims made in different documents.

A tale of two business models

Glancing back at 1998, thanks to the Wayback Machine.

W3C has Royalty Free licensing requirements and a public Process Document for good reason. I’ve been clicking back through the papertrails around the W3C P3P vs InterMind patent issue as a reminder. Here is the “Appendix C: W3C Licensing Plan Summary” from the old Intermind site:

We expect to license the patents to practice the P3P standard as it evolves over time as follows:

User Agents: For User Agent functionality, all commercial licensees will pay a royalty of 1% of revenues directly associated with the use of P3P or 0.1% of all revenues directly associated with the product employing the User Agent at the licensee’s option. We expect to measure revenues in a confidential and mutually acceptable manner with the licensee.

Service Providers: For Service Provider functionality, all commercial licensees will pay a royalty of 1% of revenues directly associated with the use of P3P. We expect to measure these revenues through the use of Web site logs, which will determine the percentage of P3P-related traffic and apply that percentage to the relevant Web site revenues (i.e., advertising-based revenues or transaction-based revenues). We expect to determine a method for monitoring or auditing such logs in a confidential and mutually acceptable manner with the licensee.

[...]

Intermind Corporation also expects to license the patents for CDF, ICE, and other XML-based agent technology on non-discriminatory terms. Members interested in further information on the relationship of Intermind’s patents to these technologies can contact Drummond Reed at drummond@intermind.com or 206-812-6000.

Nearby in the Web:

Cover Pages on Extensible Name Service (XNS):

Background: “In January 1999, the first of Intermind’s web agent patents began being issued (starting with U.S. patent No. 5,862,325). At the heart of this patent was a new naming and addressing service based on web agent technology. With the emergence of XML as a new global data interchange language — one perfectly suited to the requirements of a global ‘language’ for web agents — Intermind changed its name to OneName Corporation, built a new board and management team, and embarked on the development of this new global naming and addressing service. Because its use of XML as the foundation for all object representation and interchange led to the platform, yet had the same distributed architecture as DNS, it was christened eXtensible Name Service, or XNS. Recognizing the ultimate impact such a system may have on Internet infrastructure, and the crucial role that privacy, security, and trust must play, OneName also made the commitment to building it with open standards, open source software, and an open independent governance organization. Thus was born the XNS Public Trust Organization (XNSORG), the entity charged with setting the technical, operational, and legal standards for XNS.”

Over on XDIORG – Licenses and Agreements:

Summary of the XDI.ORG Intellectual Property Rights Agreement

NOTE: This summary is provided as a convenience to readers and is not intended in any way to modify or substitute for the full text of the agreement.

The purpose of the XDI.ORG IPR Agreement between XDI.ORG and OneName Corporation (dba Cordance) is to facilitate and promote the widespread adoption of XDI infrastructure by transfering the intellectual property rights underlying the XDI technology to a community-governed public trust organization.

The agreement grants XDI.ORG exclusive, worldwide, royalty-free license to a body of patents, trademarks, copyrights, and specifications developed by Cordance on the database linking technology underlying XDI. In turn, it requires that XDI.ORG manage these intellectual property rights in the public interest and make them freely available to the Internet community as royalty-free open standards. (It specifically adopts the definition provided by Bruce Perens which includes the ability for XDI.ORG to protect against “embrace and enhance” strategies.)

There is also a Global Service Provider aspect to this neutral, royalty-free standard. Another excerpt:

Summary of the XDI.ORG Global Service Provider Agreement

NOTE: This summary is provided as a convenience to readers and is not intended in any way to modify or substitute for the full text of the agreement.

Global Services are those XDI services offered by Global Service Providers (GSPs) based on the XRI Global Context Symbols (=, @, +, !) to facilitate interoperability of XDI data interchange among all users/members of the XDI community. XDI.ORG governs the provision of Global Services and has the authority to contract with GSPs to provide them to the XDI community.

For each Global Service, XDI.ORG may contract with a Primary GSP (similar to the operator of a primary nameserver in DNS) and any number of Secondary GSPs (similar to the operator of a secondary DNS nameserver). The Secondary GSPs mirror the Primary for loadbalancing and failover. Together, the Primary and Secondary GSPs operate the infrastructure for each Global Service according to the Global Services Specifications published and maintained by XDI.ORG.

The initial XDI.ORG GSP Agreement is between XDI.ORG and OneName Corporation (dba Cordance). The agreement specifies the rights and obligations of both XDI.ORG and Cordance with regard to developing and operating the first set of Global Services. For each of these services, the overall process is as follows:

  • If Cordance wishes to serve as the Primary GSP for a service, it must develop and contribute an initial Global Service Specification to XDI.ORG.
  • XDI.ORG will then hold a public review of the Global Service Specification and amend it as necessary.
  • Once XDI.ORG approves the Global Service Specification, Cordance must implement it in a commercially reasonable period. If Cordance is not able to implement or operate the service as required by the Global Service Specification, XDI.ORG may contract with another party to be the primary GSP.
  • XDI.ORG may contract with any number of Secondary GSPs.
  • If XDI.ORG desires to commence a new Global Service and Cordance does not elect to develop the Global Service Specification or provide the service, XDI.ORG is free to contract with another party.

The contract has a fifteen year term and covers a specified set of Global Services. Those services are divided into two classes: cost-based and fee-based. Cost-based services will be supplied by Cordance at cost plus 10%. Fee-based services will be supplied by Cordance at annual fees not to exceed maximums specified in the agreement. These fees are the wholesale cost to XDI.ORG; XDI.ORG will then add fees from any other GSPs supplying the service plus its own overhead fee to determine the wholesale price to registrars (registrars then set their own retail prices just as with DNS). Cordance’s wholesale fees are based on a sliding scale by volume and range from U.S. $5.40 down to $3.40 per year for global personal i-names and from U.S. $22.00 down to $13.50 per year for global organizational i-names.

The agreement also ensures all registrants of the original XNS Personal Name Service and Organizational Name Service have the right to convert their original XNS registration into a new XDI.ORG global i-name registration at no charge. This conversion period must last for at least 90 days after the commencement of the new global i-name service.

Over on inames.net, Become an i-broker:

What Is an I-Broker?

From Wikipedia:

“I-brokers are ‘bankers for data’ or ‘ISPs for identity services’–trusted third parties that help people and organizations share private data the same way banks help us exchange funds and ISPs help us exchange email and files.”

I-brokers are the core providers of XRI digital identity infrastructure. They not only provide i-name and i-number registration services, but also they provide i-services: a new layer of digital identity services that help people and business safely interact on the Internet. See the I-Service Directory for the first open-standard i-services that can be offered by any XDI.org-accredited i-broker.

How Does an I-Broker Become XDI.org-Accredited?

Cordance and NeuStar, together with XDI.org, have published a short guide to the process, “Becoming an I-Broker,” which includes all the information necessary to get started. It also includes contact information for the i-broker support teams at both companies.

Download Becoming an I-Broker

In addition the following two zip files contain all the documents needed by an i-broker. The first one contains the i-broker application and agreements, which are appendicies F and I of the XDI.org Global Services Specifications (GSS), located in their complete form at http://gss.xdi.org. The second one contains all the rest of the GSS documents referenced by the application and agreement.

From that downloadable PDF,

What is the application fee?
The standard application fee is USD $2500. However between the GRS opening on June 20th and the
start of Digital ID World 2006 on September 11, 2006, you may apply to Cordance to have the fee offset
by development and marketing commitments. For more information, contact Cordance or NeuStar at the
addresses below.

In the XDI.org GSS wiki,

6.1. Appendix B: Fee Schedules

Taking a look at that, we see:

Global Services Specification V1.0
Appendix B: Fee Schedule
Revised Version: 1 September, 2007

[...]

This and all contributions to XDI.org are open, public, royalty-free specifications licensed
under the XDI.org license available at http://www.xdi.org/docref/legal/xdi-org-license.html.

…where I find I can buy a personal i-name for $5/year, or a business i-name for 29.44, as in individual. Or as “become an i-broker” points out,

For volume pricing, contact Cordance or NeuStar at the addresses below.

Embedding queries in RDF – FOAF Group example

Is this crazy or useful? Am not sure yet.

This example uses FOAF vocabulary for groups and openid. So the basic structure here is that Agents (including persons) can have an :openid and can be a :member of a :Group.

From an openid-augmented WordPress, we get a list of all the openids my blog knows about. From an openid-augmented MediaWiki, we get a list of all the openids that contribute to the FOAF project wiki. I dumped each into a basic RDF file (not currently an automated process). But the point here is to explore enumerated groups using queries.

<rdf:RDF xmlns:rdf=”http://www.w3.org/1999/02/22-rdf-syntax-ns#” xmlns=”http://xmlns.com/foaf/0.1/”>
<Group rdf:about=’#both’>
<!– enumerated membership –>
<member><Agent><openid rdf:resource=’http://danbri.org/’/></Agent></member>
<member><Agent><openid rdf:resource=’http://tommorris.org/’/></Agent></member>
<member><Agent><openid rdf:resource=’http://kidehen.idehen.net/dataspace/person/kidehen’/></Agent></member>
<member><Agent><openid rdf:resource=’http://www.wasab.dk/morten/’/></Agent></member>
<member><Agent><openid rdf:resource=’http://kronkltd.net/’/></Agent></member>
<member><Agent><openid rdf:resource=’http://www.kanzaki.com/’/></Agent></member>

<!– rule-based membership –>

<constructor><![CDATA[
PREFIX : <http://xmlns.com/foaf/0.1/>
CONSTRUCT {
<http://danbri.org/yasns/danbri/both.rdf#thegroup> a :Group; :member [ a :Agent; :openid ?id ]
}
WHERE {
GRAPH <http://wiki.foaf-project.org/_users.rdf> { [ a :Group; :member [ a :Agent; :openid ?id ]. ] }
GRAPH <http://danbri.org/yasns/danbri/_group.rdf> { [ a :Group; :member [ a :Agent; :openid ?id ]. ] }
}
]]></constructor>
</Group>
</rdf:RDF>

This RDF description does it both ways. It enumerates (for simple clients) a list of members of a group whose members are those individuals that are both commentators on my blog, and contributors to the FOAF wiki. At least, to the extent they’re detectable via common use of OpenID URIs. But the RDF group description also embeds a SPARQL query, the kind which generates RDF rather than an SQL-like resultset. The RDF essentially regenerates the enumerated list, assuming the query is run against an RDF dataset with the data graphs appropriately populated.

Now I sorta like this, and I sorta don’t. It may be incredibly powerful, or it may be a bit to clever for its own good.

Certainly there’s scope overlap with the W3C RIF rules work, and with the capabilities of OWL. FOAF has long contained an experimental method for using OWL to do something similar, but it hasn’t found traction. The motivation I have here for trying SPARQL here is that it has built-in machinery for talking about the provenance of data; so I could write a group description this way that says “members are anyone listed as a colleague in http://myworkplace.example.com/stafflist.rdf”. Or I could mix in arbitrary descriptive vocabularies; family tree stuff, XFN, language abilities (speaks-reads-writes) etc.

Where I think this could fall down is in the complexity of the workflow. The queries need executing against some SPARQL installation with a configured dataset, and the query lists URIs of data graphs. But I doubt database admins will want to randomly load any/every RDF file mentioned in these shared queries. Perhaps something like SparqlPress, attached to one’s weblog, and social filters to load only files in queries eg. from friends? Also, authoring these kinds of query isn’t something non-geek users are going to do often, and the sorts of queries that will work will depend of course on the data actually available. Sure I could write a query based on matching the openids of former colleagues, but the group will be empty unless the data listing people as former colleagues is actually out there and in the Web, and written in the terms anticipated by the query.

On the other hand, this mechanism does appeal, and could go way beyond FOAF group definitions. We could see a model where people post data in the Web but also post queries, eg. revisiting the old work Libby and I explored around RSS query. On the other other hand, who wants to make their Web queries public? All that said, the same goes for the data being queried. And since this technique embeds queries inside ordinary RDF data, however we deal with the data visibility issue for RDF/FOAF should also work for the query stuff. Perhaps. Can’t blame me for trying…
I realise this isn’t the clearest of explanations. Let’s try again:

RDF is normally for publishing collections of simple claims about the world. This is an experiment in embedding data-generating-queries amongst these claims, where the query is configured to output more RDF claims (aka statements, triples etc), but only when executed against some appropriate body of RDF data. Since the query is written in SPARQL, it allows the data-generation rules to mention interesting things, such as properties of the source of the data being queried.

This particular experiment is couched in terms of FOAF’s “Group” construct, but the technique is entirely general. The example above defines a group of agents called the “both” group, by saying that an Agent is in that group if it its OpenID URI is listed in each of two RDF documents specified, ie. both a commentator on my blog, and a contributor to the FOAF Wiki. Other examples could be “(fe)male employees” or “family members sharing a blood type” or in fact, any descriptive pattern that can match against the data to hand and be expressed in SPARQL.

Open IDiomatic? Dada engine hosting as OpenID client app

All of us are dumber than some of us.

Various folk are concerned that OpenID has more provider apps than consumer apps, so here is my little website idea for an OpenID-facilitated collaborative thingy. I’ve loved the Dada Engine for years. The Dada Engine is the clever-clogs backend for grammar-driven nonsense generators such as the wonderful Postmodernism Generator.

Since there are many more people capable of writing funny prose to configure this machine, than there are who can be bothered to do the webhosting sysadmin, I reckon it’d be worthwhile to make a general-purpose hosted version whereby users could create new content through a Web interface. And since everyone would forget their passwords, this seems a cute little project to get my hands dirty with OpenID. To date, all I’ve done server-side with OpenID is install patches to MediaWiki and WordPress. That went pretty smoothly. My new hacking expedition, however, hit a snag already: the PHP libraries on my EC2 sandbox server didn’t like authenticating against LiveJournal. I’m new to PHP so when something stops working I panic and whine in IRC instead of getting to the bottom of the problem.

Back to the app idea, the Dada engine basically eats little config files that define sentence structures, and spews nonsense. Here’s a NSFW subgenial rant:

FlipFlip:scripts danbri$ pb < brag.pb
Fuck ‘em if they can’t take a joke! I’m *immune*! *Yip, yip, YEEEEEEE!*
*Backbone Blowout*! I do it for *fun*! YAH-HOOOO! Now give me some more of…

And here’s another:

FlipFlip:scripts danbri$ pb < brag.pb
I’m a fission reactor, I fart plutonium, power plants are fueled by the breath
of my brow; when they plug *me* in, the lights go out in Hell County! Now give
me some more of…

A fragment of the grammar:

“I say, `” slogan “‘. By God, `” slogan “‘, I say! ” |
“I am ” entity “, I am ” entity “! ” |
“I’ll drive a mile so as not to walk a foot; I am ” entity “! ” |
“Yes, I’m ” entity “! ” |
“I drank *” being “* under ” number ” tables, I am too ” adjective ” to die, I’m insured for acts o’ God *and* Satan! ” |
“I was shanghaied by ” entities ” and ” entities ” from ” place “, and got away with their hubcaps! ” |
“I *cannot* be tracked on radar! ” |
“I wear nothing uniform, I wear *no* ” emphatic ” uniform! ” | ….

To be crystal clear, this hosted app is total vapourware currently. And I’m not sure if it would be a huge piece of work to make sure the dada engine didn’t introduce security holes. Certainly the version that uses the C preprocessor scares me, but the “pb” utility shown above doesn’t use that. Opinions on the safety of this welcomed. But never mind the details, smell the vision! It could always use Novewriting (a Python “knock-off of the Dada Engine”) instead.

The simplest version of a hosting service for this kind of user generated meta-content would basically be textarea editing and a per-user collection of files that are flagged public or private. But given the nature of the system, it would be great if the text generator could be run against grammars that multiple people could contribute to. Scope for infinite folly…

OpenID and Wireless sharing

via Makenshi in #openid chat on Freenode IRC:

<Makenshi>: I found a wireless captive portal solution that supports openid.

With the newest release of CoovaAP, some new features in Chilli are demonstrated in combination with RADIUS to allow OpenID based authentication. (coova.org)

I’m happy to see this. It’s very close to some ideas I was discussing with Schuyler Earle and Jo Walsh some years ago around NoCatAuth, FOAF and community wireless. Some semweb stories may yet come to life.

At the moment, the options available for wireless ‘net sharing are typically: let everyone in, have a widely known secret for accessing your network, or let more or less nobody in without individually approving them. Although the likes of Bruce Schneier argue the merits of open wireless, most 802.11 kit now comes out of the box closed by default, and usually stay that way. Having a standards-based and decentralised way of saying “you can use my network, but only if you login with some identifiable public persona first” would be interesting.

OpenID takes away a significant part of the problem space, allowing experimentation with a whole range of socially oriented policies on top. Doubtless there are legal risks, big privacy issues, and lurking security concerns. But there is also potential for humanising interactions that are currently rather anonymous. In the city I live in, Bristol, there’s a community wireless effort, Bristol Wireless, as well as wireless Internet in countless local cafes. Plus commercial hotspots and whatever the city council are up to. Currently these are fragmented, and offer a variety of approaches. Could OpenID offer a common approach for Bristolians to connect? I like the idea that (for those that choose to ‘go public’) OpenIDs could link scattered presence across community sites. Having OpenID-based login used eg. for cafe-based access could be a nice step in that direction. But would people trust their local cafe to know what they’re doing online any more than they trust Google? Should they?