Ross Mayfield's Weblog

Vacation Email Page

Luis Suarez is on week six of giving up on work email.  Not entirely, but constraining its use to the proper while moving other communications to social software.  This week he found the reliability of those other tools made people default back to email.  While he doesn't mention what services went down on him, its an obvious distinction between consumer and enterprise.  Jon Mell builds upon his experiment to explore ROI and I can add that its common for our customers to reduce email volume by 30%, eliminating occupational spam.

Ironically enough, I find myself using Twitter, Facebook, and wikis of course, to help close a couple of deals this quarter.  The modality choices perhaps say more about the customers, but its a way to cut past the noise of the inbox and stay within peripheral vision.

Late next week I'm heading to LA for spring break with the kids.  I'm going to set up an vacation autoresponder with this message:

I'm on vacation March 26th-31st and will scan email when I return.  If you need immediate Socialtext assistance, contact info@socialtext.com or 1-877-GET-WIKI.  If it really is an emergency and you have my mobile number, use it.

But if you really want to help me with email overload, edit my vacation page if you are a Socialtext employee or edit my public Vacation Message Page.  You may find its okay to communicate in public and the communication can be group and summarized with others.  Thanks in advance!

I have a feeling this is either a really great or really bad idea.

I am not People Ready

Status Contests and Attention Aggregators

I find myself updating my status, or answering the question "what are you doing?" across Twitter, Jaiku, Plazes and Facebook.  This is made easier through clients like Twitterific, Juhu, Plazer and some Facebook hacks that are less attractive.  I'm using these clients for more than updating status easily, however.  They are a new kind of attention aggregator -- bringing the status or lifestreams of my social networks to me in real time.

They are pretty cool tools, if you haven't tried them.  For Twitter, having a client fits my laptop-oriented daily use, so I can mostly turn the mobile client off (text "off" to 40404, and "on" when you are roaming around) and subscribe to a larger social network.  For Jaiku, having a client brings continuous partial presence to my laptop, that is far richer because people are lifestreaming (adding feeds from other tools like blogs to enrich their presence).  For Plazes, the Plazer has always let me share my location, but now it is richer with status sharing and a reverse-chronological view of your network.  Facebook will surely have a better client soon as part of their quest to be the social operating system, and hopefully incorporate your Mini-feed.

But all this client development seems like one-offs, status service providers have stayed out of development perhaps wisely, and there is ample room for innovation.  As is usually the case, we've been here before -- a lack of standardization and cooperation yields less and leaves room for a large service provider to monopolize.

I don't think we want to wait for Google, Yahoo or Microsoft to provide a status and lifestream integrated user experience that flows through clients, let alone browsers or operating systems.  I do think this will happen as status services fill a valuable niche in our demand for social interaction, let alone being a new command line for web services.

Compare where we are to how RSS and Atom provided common standards for developers to innovate.  There was a time where almost every graduate CS student would write a news aggregator for fun and new startups proliferated.  Adriaan Tijsseling leveraged Atom from the earliest versions to create client blog editor, Ecto, and went beyond the Flickr Uploader to also offer aggregation with 1001.  Some, like Newsgator, plunged deep into clients across operating systems, email clients and mobile clients to let you work across them.  But the result isn't the case where one vendor, could own publishing, syndication and reading including clients.  We have a healthy marketecture that continues to innovate.

There is a new kind of aggregator, for more real time attention, that needs to be build to work across status services.  I'm not sure if it will be built into existing news aggregators, if existing status clients will evolve into them, or it will be something new.  I just know it is coming.  It will leverage status service providers and Lifestreaming you find in services like Dandelife and Jaiku.  You will be able to edit your status and perhaps more, like location to Plazes or a blog entry.  Maybe it will be built on Apollo or Google Gears, maybe a Firefox extension or a mobile version on WidSets.  But it won't happen too soon.

The problem is, while the REST APIs are easy to work with, they aren't standardizing.  Maybe they will converge on using the Atom Publishing Protocol.  Maybe they can work out a way to let you write your status once, publish everywhere, and remove dupes when aggregating.

Jaiku Tips the Tuna?

Did Jaiku tip the tuna yesterday?  Leo Laporte  jumped ship from Twitter to Jaiku, his 4,000 followers followed.  The Twitter herd debated platforms, has herds do when chosing to migrate.  Suddenly the story was Twitter vs. Jaiku and Jaiku team dealt with digesting a big chocolate Easter Bunny.

Let me provide some context first.  I was exposed to Jaiku at Aula in Helsinki last June.  From my notes:

Jyri Engestr?m and Mika Raento on social peripheral vision.  Phones are designed with the assumption that you know who you want to call before you do.  You need to process social signals before using the device.  Jaiku, their startup is looking to augment basic functions of a phone by pasting onto it what is happening on the internet.  If you can't find anyone in your contact book, you can search a directory made of everyone's contacts. Calendars let you share future events to let you plan together.  The demo shows very rich profiles based on phone usage (automatic data) and more social signals (more manual) -- which provides a different form of Presence.  In usage, people still call regardless of presence, but when someone doesn't answer, you leverage the presence to understand why. Integrated IM is more convenient than SMS, and includes group messaging.

Since then, Twittr came on the scene and Jakiu's web interface got a major upgrade.  It's important to understand the significant differences between the two services, their design thinking and strengths.  Joi Ito:

Looks like a bunch of people are trying out Jaiku after "tasting" co-presence with Twitter. To me, Jaiku, which existed before Twitter, is a bunch of Helsinki mobile jocks getting into the Web 2.0 of it all whereas Twitter is the Web 2.0 crowd "getting" co-presence...

Jaiku comes from a "presence" background allowing bluetooth proximity, phone idle time, ringer mode and other things to trigger state changes - the messaging came later. Twitter, on the other hand, is primarily messaging, which as we all know, is just a flexible and manual vector for presence information.

To understand where Jaiku is coming from, I encourage you to read this interview with co-founder Jyri Engestr?m and his post on social peripheral vision (the ability to have your finger on the pulse of your friends, family, and colleagues).

  Twitter on paper 
  Originally uploaded by jack dorsey.

In digging around for some of the thinking behind Twitter, I found Jack Dorsey's napkin design for Twitter:

from a note circa Jan 2006.

casual awareness.
"what are you up to?"

multiple entry point to set status
- web
- email
- phone
- sms
- im

multiple ways to "subscribe" to status
- web
- email
- phone
- sms
- im

3 aspects
- set status
- timeline (collaborative)
- configuration

The interesting thing is that I found it on Jack's Jaiku page where he had included his Flickr stream as part of his presence.  For a long time I've wanted the Xfire for social software, and today Jaiku provides this kind of persistent presence.

Jaiku lets you incorporate feeds from your blog, bookmarks, photos, location -- and Twitter if that is where you prefer to post status.  Every post of any kind becomes an object for conversation, through comments.  This works easily in the web UI, but it also works in the Nokia mobile client because presence isn't overwhelming. Presence is something you can glance at, not an SMS interruption. 

Unfortunately today this requires a Nokia phone, but they are working on a Java version that also specifically supports commenting (kind of like Radar.net, more on that later).  People coming from Twitter won't expect the ability to add their attention breadcrumbs to their attention stream (developers will) and will probably expect something they can adopt on their mobile easily.  In the US, this is a significant barrier (Sidenote: fuck you Cingular.  Making me change calling plans to switch SIM cards from my Blackberry and claiming the handset wont work because you don't sell it even though it runs the same software is an easy way to lose me as a customer, as if I had alternatives.).  Jaiku isn't ready to Tip the Tuna until their next mobile client comes out. 

But until then I'd expect a lot of people to use the web version as an attention pool.  Posting to Jaiku via Twitter is a no-brainer and I'd hope you can do the opposite without loops and dupes soon.  Rafe asked the right question:           Is it possible Twitter and Jaiku will end up sharing users, instead of hoarding them like the IM services did early on? I responded: systems of record are being replaced by systems of discovery. 

In other words, in the first web I would worry about which service I would commit my social network, presence and persistence to.  But services are increasingly making data discoverable and discovering data from other services.  We used to worry about transporting our FOAF relationships, but then I think we realized that each tool is different and being able compose a different social network was a virtue (not just because of faceted identity, but that different tools need different filters and the social network is the filter).

UPDATE: This post was written in haste before going out for Easter.  Jaiku released their API and developer site.  I forgot to highlight Marko Ahtisaari's why I use Jaiku:

1. Silent sociality - checking up on what my friends are up to when convenient, and posting my own state knowing that I won't be disturbing others (unless they have explicitly asked to be alerted).

2. Small-group sociality - Jaiku is not about celebrity. I'm interested in sharing state with a small group I'm nearly always in contact with, what Mimi Ito has called full-time intimate community.

3. Mobile sociality - Jaiku was designed with the mobile "living phonebook" interface in mind. SMS alerts crowding the inbox of one of the few working personal and functional communication channels is not my idea of improving communication. I use the SMS-in posting to Jaiku when I'm using my Nokia 8800 and with my N70 I use the Jaiku phonebook.

4. Background sociality - Jaiku allows me to integrate other online identities and feeds (including delicious, flickr and any RSS) into my single jaiku presence feed. This is done in a way that doesn't confuse these background posts with my explicit state messages.

Rafe posts about adding your Twitter to Jaiku.  And I wanted to add one last thing.  If you aren't trying it with a mobile client, you can't get the real experience.

Attention Abundance Companion

Since my last post is really a question, let me provide some questions I would ask if it was the topic of an open conversation:

What are examples of giving attention transparently that save time?

Is the skill on the rise focus or flow?

How do we teach flow?

Which is of more value to vendors: transactions or engagement to co-produce and co-promote?

Which is of value to PR: exclusives or inclusives?

When you give attention to a co-worker, is the benefit usually short-term or long-term?

Which would your employer rather you give more of?

If attention isn't scarce in the attention economy, what is?

Will Twitter fail because of overload?  Why not? Social network as filter? Ability to downshift to asynchronous?

If social dynamics are positive in a community, does the role of community management shift to directing attention by fiat? Is that expensive?

How does memory change with abundant attention?

What corporation executes best at giving attention?

Is a shift from push to pull as the dominant model of attention management do we discover more or is it the proverbial echo-chamber?

If we shift from push models of attention management, to pull, are we more productive?

There, I've put you back in middle school with a lazy teacher.  That or a professor with the case method.

Attention Abundance

Read/Write Web has a good overview the attention economy the other week,written by Alex Iskold and edited by Richard MacManus.  They suggest information overload will be solved through personalization and explain the trust issues of information metadata.  Unfortunately, the analysis is flawed on both the cause and scope of the solution.

Pointing to the meteoric rise of weblogs as measured by Technorati, they argue the more information there is, the more difficult it is to manage attention.  Attention economy is a relatively underdeveloped theory, and Wikipedia page for that matter -- from which they ground this assumption:

Herbert Simon was perhaps the first person to articulate the concept of attention economics when he wrote:

"...in an information-rich world, the wealth of information means a dearth of something else: a scarcity of whatever it is that information consumes. What information consumes is rather obvious: it consumes the attention of its recipients. Hence a wealth of information creates a poverty of attention and a need to allocate that attention efficiently among the overabundance of information sources that might consume it" (Simon 1971, p. 40-41).

However, merely having more information available in the world does not mean there is a dearth of attention.  This supply/demand view is the economics of scarcity.   This this is information we are talking about, which is driven by the economics of abundance.  Turn the equation around from consuming information to giving attention.  Then we can proceed.

Just because there is more information doesn't mean you have to consume more.  Ignoring is not ignorance.  Further, the medium in which individuals gained the means to publish is the same one that provides the means to manage attention.  This isn't broadcasting, where Sarnoff's Law says the value of the network is the number of nodes that consume.  It is a network.  Where Metcalfe's Law measures value by N squared and Reed's Law of Group Forming holds we have an exponential potential through communities to process information.

After Simon, with the rise of the web people thought it would be crushed by the Babel problem -- if everyone can be heard, how do we choose who to listen to amongst the cacophony?  Yochai Benkler addressed this in the Wealth of Networks.  He pointed out flaws in the theory that money would be the only driver for a solution, given the abundant means to produce filters through commons-based peer production.  And how within hypertext, when producers are filters, using the blogosphere as an example, we can scale without being overwhelmed. The social network is the filter.

This is, in part, because of transparency.  Oddly enough, transparency begets memory. We don't have to consume and process every bit of information that crosses our desk, but can fall back upon search and social discover.

Benkler, however, underpinned his analysis by submitting that the one scarcity is time and attention of users.  Classical training makes this an obvious insight, perhaps too obvious.  As users, and anthropological observers, it is hard to see how the overload could be anything else.  But I think networks are still optimized for the days of Sarnoff and encroaching upon Metcalfe.  Media is increasingly networked, but the role of groups in filtering is undervalued by your average user.

Attention economy collapsed two attractive words and tried to give meaning to the former.  The best part of it was we sought different rights when enjoining attention into markets.  The right to own what we were giving.  The right to not have it abused, especially at a time when our attention metadata was by direct marketing, or ouright data theft in security failures that dominated(s) the headlines.  Identity and attention rights, are still only valued by people who have experienced market failure and operational risk.  This may change, but more catastrophe awaits and will drive it.

Some of the attention markets interest, and Steve Gillmor or Seth Goldstein could correct me on this, is on how much a given consumer could be paid to pay attention to a pitch (e.g. for mortgage advertising) that results in attention.  This could be modeled and treat attention as fungible, and even enable trading of attention futures.  Especially if there was trust in the market for consumer privacy.  But I can't swallow how transactional this is.  It provides incentives for offers and advertising of low quality and prematurely commoditizes. Do you just want consumers to consumer, or could they collaborate?  Could a better measure in most cases be engagement?

Coping with attention overload by the individual is full of all kinds of GTD and life hacking tips.  I personally don't believe you can engineer a solution as a former consumer of information, and the most important productivity and innovation boost you can give yourself is to let go of the stress.  Stowe Boyd summarizes that the opportunity is shifting from focus to flow.

On the supply side, Steve Rubel remarks upon some better practices:

To cope, we've developed a defense mechanism - what Linda Stone calls Continuous Partial Attention. The content industry has responded by chunking things down for us into snacks that complement the meal. That's smart.  I told marketers to do the same. However, something at some point has to give. The only way out is perhaps with tools that make things easier for us.

However, much of the conversation around Attention is framed upon scarcity-economics and psychology.  Assume that attention is finite and the cognitive emotional overload of allocating it is overwhelming.  However, what if Attention is framed upon abundance-economics and sociology?  We have an abundant desire to give attention, and while time is short, when we give to others in groups, what we produce saves time.  Especially compared to going it alone.

Here is my hypothesis, that i am too stupid or distracted to pull in the existing research around to truly prove -- that:

• attention is abundant
• when you give attention, others give it to you
• the artifacts of your attention snowball
• when in a state of flow, you can participate in snowballs at a low cost
• you are more efficient and innovative producing through collaboration
• collaboration happens at a lower cost through openness, transparency and social discovery

In summary, the gift of of attention begets not only attention, but time and attention.  Attention is not a zero-sum game.  So treat this snowball for what it is, and let's see what sticks to it.