Ross Mayfield's Weblog

Pownce: Collaboration from and at the hip

At first glance, Kevin Rose of Digg's new startup Pownce is Yet Another Status Message Service (YASMS) like Twitter, Jaiku or Plazes.  But really, its a collaboration app made for the most modern web.  It's bound for adoption because the founders can drive word of mouth and its inherent virality.  And perhaps what it does is less important than the three trends it represents.

Like others, the primary activity is messaging to your social network.  You message to all your friends or public like others, or directly like typing "D Username" in Twitter at the beginning of a message, but also lets you select a subset of friends.  Beyond messages, you can share links, files and events.  Beyond doing this on the web, there is a Windows or Mac rich client.

The digerati and diggerati will probably rant away about how it doesn't have SMS or IM integration like Twitter, how the content is mundane (same thing with blogging five years ago), how it needs APIs and microformats (which it does, and hooks into Twitter, del.icio.us, Flickr, Upcoming and Facebook are inevitable), or just complain about adding friends again (Adding friends is the new zen).  The design is slick on both the web and client and they will polish up key details like last names, comment threading like Jaiku, permalinks and need a more public space to explore. 

What does it matter how one comes by the truth so long as one pounces upon it and lives by it? -- Henry Miller

But here's the three trends:

YASMS Gets and Ad Format -- I admire startups that launch with an actual business model.  They have introduced a new Ad Format, a message broadcast into the stream with the Pownce icon (the green P in the above screenshot is an ad from PBwiki, I love wikis) that doesn't seem to persist.  I didn't mind the ad from LaughingSquid (first I saw) in my peripheral attention.  And if I did, I could pay to make it go away, but subscribing to the Pro version for $20/year and also be able to send files over 100MB.

AIR Gets a Viral App -- The client is built on the Adobe Integrated Runtime (AIR), formerly Apollo and still in Beta.  Pownce's virality will give AIR the airtime it needs for base of users to install it, making it easier for the next AIR client to come along.  This should be AIR's showcase.  That said, Pownce's model is what I call contained virality, where limits are part of the draw and when you are in you feel in (at least to share music, a hidden driver).

Consumer Collaboration Get Hip -- Anyone who follows the enterprise collaboration space will immediately see parallels with P2P collaboration apps like Groove or Shinkuro.  Or IM, Skype and more directly enterprise IM like MindAlign.  The key difference is group forming by social network and default modes of sharing more publicly.  Pownce will appeal to a very different demographic, that's already collaborating on blogs, wikis and IM, and potentially full a space in between.

There are several vectors in which Pownce could go, or others could go towards including presence, location, public IM, security, indexing and integration.  Pownce will have to open up invites soon (I'm out, please don't ask) to build its network effect before others encroach.  It isn't unique enough to gain the continuous or at least partial attention of users for yet another client.  Infrastructure costs will be greater than P2P.  At the risk of breaking the design and making it too complex, Pownce should give serious thought to the role of standards and how they could be a client for Facebook.

Please, I don't have any invites.

Going Lijit

Yesterday I happened to speak with a blog friend for the first time, Brad Feld, and he told me about Lijit.  Its a social search play that creates a widget for searching across your blog, social bookmarks and blogroll.  Today they announced $3.3M in funding from Boulder Ventures.

I've added it to the bottom right of my blog.  You can also play with it on their site, where I found Rick Klau, or on Paul Kedrosky's blog.  When setting up Ligit they have a swift utility for adding content sources such as Twitter, MySpace, LinkedIn, Del.icio.us, Stumble Upon, Flickr, Youtube and MyBlogLog. Adding the widget is simple, you gain a different kind of insight into use and you can browse implicit and explicit profiles.  Lijit reminds me of Dandelife's Lifestreaming and Sam Ruby's MeMeme for the network represented and Eurekster and MyBlogLog for the widget effect.  Its still early for Lijit, and I'll be watching the results as they come in.

CPAL Submitted to the OSI

Tonight I posted the following on the Socialtext Blog.  It may seem like a bit of inside baseball, but it could be a very big deal:

Today Socialtext submitted the Common Public Attribution License (CPAL) to the Open Source Initiative (OSI).  OSI is a community appointed body responsible for open source licensing.  OSI is the creator of the Open Source Definition (OSD), certifies licenses as OSI Certified, guards against license proliferation and gives meaning to the term open source. Socialtext is the first an only company with a Mozilla Public License (MPL) plus Attribution license (amongst 40+ commercial open source businesses) to seek OSI Certification.  We do not, however, consider ourselves open source until all our licenses are OSI Certified. We have been working through this process for some time.  Today the submitted CPAL can be found here and we look forward to the open conversation on OSI's license-discuss mailing list.  For those interested in a summary of our submission, here it is:

1. The Common Public Attribution License ("CPAL") is based on the MPL which has been approved and all of the new provisions are in Sections 14 and 15 (and Exhibit B) and  adding "Original Developer"  to certain disclaimers ("Original Developer" is a term defined in the new provisions for those who originally created the program who may be different from the "Initial Developer"). Section 14 provides for an attribution notice based on the Adaptive Public License and Section 15 provides for a network use provision based on the commonly used provision on "External Deployment" found in the Apple Public License, Real Network Public License and the Open Software License. We have used the Adaptive Public License, which is virtually the sames as the prior attribution provision which was in Exhibit B of the proposed Socialtext Public License,  as the basis for the attribution provision because it was approved after OSD 10 was adopted. We have limited the placement requirement for attribution notice to "prominent" rather than a specified size or location. We have also permitted the use of splash screens. The term "prominent" is frequently used in other OSI approved licenses such as the GPL and NASA Public License. Socialtext believes that the application software has special needs as compared
to operating systems because of the application software can be used anonymously in large distributions and can be used to provide services
through an ASP which does not provide modifications back to the community. None of the approved OSI approved licenses include both a network use provision and an attribution provision. We have limited the new provisions to those which are either the same or very close to provisions from existing licenses (see above).

2. The license can be used with any software which is licensed under the MPL and licenses compatible with the MPL. The CPAL will take precedence for combined works. Some licenses such as the GPL which are incompatible with the MPL are also incompatible to the CPAL.

The above is a summary of how CPAL is OSD Compliant, but you can also explore this points one through ten.  If the license is deemed compliant with OSD by the OSI Board, it should be OSI Certified, a mark that carries great meaning on both the community and market. Socialtext also uses other OSI-Certified Licenses within its products.  Primarily, Perl Modules licensed under the Perl Artistic License, some originated by others, some by Socialtext employees in an act of giving back to the community. While working on CPAL, we developed a new strategy for open source licensing.  Some core components will be licensed under both CPAL and the Perl Artistic License (PAL, I like to call it because of the nice ring and rhyme, although it is commonly referred to as the Artistic License).  With CPAL and PAL together we gain something in a greater community and commercial interest. Dan Bricklin highlights this aspect I'm not supposed to call triple-licensing in his post SocialCalc 1.1 released -- we now have a real Open Source project:

The SocialCalc Engine code is being released under the Artistic License 2.0. This license, written by the Perl Foundation, is basically the same as the Artistic License used for years for Perl and was apparently just approved as "Open Source" by the OSI. (Perl also has the option of being used under the GPL.) I understand that this is a pretty liberal license which allows proprietary modifications but also allows code licensed under it to be included in projects licensed under GPL. This should hopefully help this spreadsheet engine code become part of a wide variety of projects and get a large number of developers contributing to its maintenance and advancement.

Tomorrow we will post to the wiki an expanded FAQ and guide for applying the CPAL to your own project or product.  We don't recommend doing so until OSI approves it, as a disclaimer, and provide it for broad understanding of the potential of this license to serve community and commercial needs.  If not, heal a growing rift in the community overall.

I am not People Ready

Calling Socialtext Open Source

Michael Tiemann, President of the Open Source Initiative (OSI), takes a strong stand that unless a company uses an OSI-approved license they should not call themselves open source.  Technically, nobody owns the open source trademark, it is a common mark that anyone can use and modify how they see fit.  The community has a process with OSI that if followed lets a company use the OSI Certified mark.  If someone doesn't follow this process in using an approved license or getting theirs approved, they don't have the right to use the mark.  If they use the mark anyway, they will be both shamed by the community and potentially sued.  But that's technically.  Michael is suggesting that OSI and the community should defend the term open source, even through it doesn't have the legal means to do so.

Go read the whole post, here is just part of it:

So here’s what I propose: let’s all agree?vendors, press, analysts, and others who identify themselves as community members?to use the term ‘open source’ to refer to software licensed under an OSI-approved license. If no company can be successful by selling a CRM solution licensed under an OSI-approved license, then OSI (and the open source movement) should take the heat for promoting a model that is not sustainable in a free market economy. We can treat that case as a bug, and together we can work (with many eyes) to discern what it is about the existing open source definition or open source licenses made CRM a failure when so many other applications are flourishing. But just because a CEO thinks his company will be more successful by promoting proprietary software as open source doesn’t teach anything about the true value of open source. Hey?if people want to try something that’s not open source, great! But let them call it something else, as Microsoft has done with Shared Source. We should never put the customer in a position where they cannot trust the term open source to mean anything because some company and their investors would rather make a quick buck than an honest one, or because they believe more strongly in their own story than the story we’ve been creating together for the past twenty years. We are better than that. We have been successful over the past twenty years because we have been better than that. We have built a well-deserved reputation, and we shouldn’t allow others to trade the reputation we earned for a few pieces of silver...

Open Source has grown up. Now it is time for us to stand up. I believe that when we do, the vendors who ignore our norms will suddenly recognize that they really do need to make a choice: to label their software correctly and honestly, or to license it with an OSI-approved license that matches their open source label. And when they choose the latter, I'll give them a shout out, as history shows.

I applaud Michael's stand, as this kind of leadership helps move some core issues forward.  Socialtext has advocated that commercial open source companies should follow the OSI process and gain OSI Certification, as we are.  I've said that we won't use the OSI Certified mark until we have earned it, but use the term open source.  However, while this is technically correct, I'm concerned about the unfolding rift in the community. 

I am to blame for part of it.  When we submitted the Generic Attribution Provision (GAP) to OSI back in November we tried to both gain certification for an MPL+Attribution license, and do so in a non-standard way.  We sought to have attribution be a provision you could add to any of the 50+ approved licenses that allowed for extension.  While this might result in a more favorable taxonomy of available licenses and make license proliferation more future proof, we didn't follow the process correctly and submit a complete license.  We then took too long to draft and submit the Socialtext Public License (STPL) in March.  When OSI was going to vote upon it this month, we didn't have enough time to incorporate feedback, test the draft of the Common Public Attribution License (CPAL) against our own goals and have time for the OSI-discuss mailing list to vet it.  Sure, OSI itself could have moved the process forward, but ultimately I could have driven it better.  Approximately 40 companies use similar licenses without following this process and are looking to us for leadership.  I'm not certain how many will adopt CPAL if approved, but I have a role in why there is no conclusion to certifying an attribution license, yet.

Most of our developers have a long track record in open source.  Having things take so long and with less than perfect execution effects them personally.  Which brings me to the point of Michael's post, on using the term open source.  We've had a lot of conversation within our company wiki about this over the past few months.  Casey West, in particular, took me to task for us taking the position of using the term open source, but not the mark OSI Certified.  Technically, I'm not sure what we should call ourselves right now.  We are following the process and have committed to using an OSI approved license one way or another (if ours is not approved, we will adopt an approved one, even at commercial cost).  Even edits on our Wikipedia page, from friends of ours, say we are not open source. 

So what I'd like to hear from the community is what we should call ourselves.  If the consent of the community isn't open source, but to use a different term, I'll edit our site and wiki appropriately.  It should be a short term monkier for us, but I believe it is our responsibility if we want to be part of the larger open source community.

Enterprise 2.0 Keynote

Slides from my ten minute talk today...

Notes will probably be shared on the Enterprise 2.0 Wiki.

KnowNow on the launch pad

Sam Weber the VP of Technical Services is a 1.0 veteran on the Enterprise 2.0 launch pad.  No longer selling and supporting software that users hate to use.  50% of email is jumk, <20% find portals useful, 16 clicks to find information and 50% of searchers are successful.  Sharepoint is a significant problem because of the diffusion of sub-portals.  Big architecture slide time: they monitor data from multiple sources, automate relevancy, transform and deliver and manage information. We do more than RSS for integration.

Launches KnowNow Live.  Has a Netvibes like homepage where you can drag content objects, with a sidebar on the left of channels to select from.  Its an aggregator with a portal like UI you can personalize.

Stowe: I want more demo and less pitch. I see you having a hard slog battling the huge CMS guys who are going to provide this.  Then you have consumer options like Pageflakes chipping at you.  Like 9k people in your space.

David: I met these guys a year and a half ago, nice to see the launch.  Those who do the aggregation make the money.  You have a lot of good customers, but the question is it 9x better than email?

Ross: Good component in an overall aggregation play.   

LiquidTalk at Enterprise2.0

Dave Peak, on the Enterprise 2.0 launchpad, with Liquidtalk.

Right now, how many of you are checking your blackberry.  The reality is we are out of time, are out of the offfice, have few cracks in the day to get it done and few chances to connect in real time.  Difficult to collaborate when out of the office, hard to get knowledge on the go, productivity plummets when on the go.  Attracting the best talent means helping the next generation get their information immeadiacy, place value on peer input.  Roles are getting tougher while expectations are higher, with rising rep turnover rates, high dependency on top performers.  A disengaged and disconnected workforce means lost productivity and lost revenue.

Our answer is mobile workforce engagement.  Create, find, organize and push audio/video business content to mobile devices.  Leverages the most powerful means for collaboration -- the human voice.

Liquidtalk portal lets a rep gain access to video and podcast files, and synch them to their mobile device.  The inbox is about things pushed to his list by his boss to consume as well. Liquidcast lets you phone in a podcast to the Liquidlibrary.

Dave Coleman's take:  This is more interesting, would like for new vendors to explain what they do in one slide rather than five or six.  I thought this was more of DVD for conversations that you can timeshift.  You are more like Tivo than iTunes.  I find podcasts useful, but as information transfer than knowledge.

Stowe Boyd's take:  I feel to compelled to talk about the slideshow because the buildup was different.  I have the sense that this is a feature.  Creating content seems like work, but if it was part of something larger, such as being part of Salesforce.com and its activity.

Ross' take: there are a lot of podcasting companies and tools in the space.  This seems well focused for enterprise use.  Every product starts as a feature.

How to Build an Enterprise 2.0 Platform Employees Will Use

Live impressionistic transcription of How to Build an Enterprise 2.0 Platform Employees Will Use
Moderator - Rob Preston, Editor in Chief, Information Week

Speaker - Mike Fratesi, Manager, Solutions Marketing, Unified Communications, Cisco Systems, Inc.

Mike Fratesi Solutions Marketing Manager, Unified Communications Cisco Systems mfratesi@cisco.com Mike is responsible for unified communications applications solutions marketing at Cisco Systems. He joined Cisco Systems in January, 2004 upon the acquisition of Latitude Communications, a maker of conferencing solutions, where he was the Director of Product Marketing. In over 7 years at Latitude Mike helped drive the product direction and marketing of the MeetingPlace conferencing system. Prior to Latitude he was a product manager in the Pentium processor division at Intel and a consultant at Booz-Allen & Hamilton.

Speaker - Oliver Young, Analyst, Forrester Research, Inc

Oliver is an analyst at Forrester Research, Inc. covering Web 2.0 , Enterprise 2.0, and Social Computing - including blogs, wikis, RSS, social networking, and mashups - and its impact on technology vendors and the overall tech industry. As part of this coverage, Oliver also looks at broad economic trends in the tech industry. His work has been cited in major media outlets such as ComputerWorld, BusinessWeek, The Financial Times, and The Wall Street Journal as well as influential blogs such as Rough Type and Read/Write Web. Prior to his analyst role, Oliver served as a researcher at Forrester. Oliver is a graduate of Kalamazoo College with a B.A. in economics and English. He has also studied at the London School of Economics and Political Science.

Speaker - Toby Redshaw, Corporate Vice President, I.D.E.A.S., Motorola

Toby: We had a KM system where you can put information in if you had a password and could remember where stuff was and the password.  Turned on wikis 4400 blogs and 4200 wikis.  Didn't even tell anybody, just turned it on.  Plus forums and FAQs.  There is a community that has the collective knowledge of the last twenty years of the company.  This accelerates the clockspeed of the company.  We are implementing Baynote and Scuttle for folksonomy.  I thought it would be utter crap, Arlo Guthrie doing IT, but it actually works, seeing heat maps and accelerating discovery and learning.  Widgets, social networking with Visible Path.  The key is that it just has to be so easy to use for adoption.  Don't train, advertise it, did have to explain scuttle a little bit.  The higher up you go in the heirarchy the less it is used, a good thing, the real work is elsewhere.

Oliver: What I see is a bit of a dichotomy.  The marketing and communication departments, the lines of business are excited about these tools.  But IT has a lot of fear.  They look at these tools and think they are decentralized, emergent, not easy to control and my employees are doing this anyway.  The command and control thought, keeping CIOs up at night.  They learned a lot of lessons from IM, where employees used it anyway.  A lot of companies say we can't lock it out, lets figure out how to do this.  For those who are bringing it in, it is in an experimental phase.  Far less common that you find someone that says lets change things.  At Motorola you have cool tech that rolls out under the covers, something common, but it limits the potential impact.

Mike: Coming from the unified communication space, WebEx acquisition, getting into collaboration and deploying it ourselves.  We want to give people more immediate access to their experts.  A huge amount of growth in real time collaboration, moving people beyond the phone.  A ton of inefficiency needs to be driven out.  Integrating people, processes and data and making the communications a core part of the effort.  Moving from an ad hoc deployment model to a multi million dollar initiative to redo our Intranet.  Making it easier to build communities of interest, from mailing lists that overwhelm.  Chambers is talking all about collaboration, for our organization and our customers.

Toby: You have to have a small team to do this.  One manager and four people.  The actual information is driven by 250 knowledge champions, selected by the community.  A badge of honor to be a champion, shepard the wiki.  I get calls that say, did you know we have 4000 blogs, are they doing anything bad?  Yeah, probably.  Can we control it?  Yes, give me 400 people to monitor it, but it will kill it.  They schedule a meeting and I don't show up and life goes on.  I don't beat competition through better buildings, companies are human beings solving problems, driving opportunistic events by working with dumb stuff like buildings and cash.  Its like Chess.  You move once, I get to move twice.  Who do you think will win?  That's what a community platform gets you.

Oliver: To some extent the security question is paranoia, but all security is paranoia.  Stealing a laptop is a bigger risk than email or blogging.  External blogging means that with a big organization, someone is, and in regulated industries it is a serious issue.  Adoption is happening in your enterprise whether you like it or not.  In a survey, leaders estimated 20% were using these tools.  Those making no investment had about 3-8%.  Need to give guidelines about what is appropriate.  SaaS lets a business unit leader get my entire group up and running for $20 on Socialtext, for example.  Without involving IT.  I have a lot of faith in Socialtext that they are a secure service and not losing data.  But you have to have a story if it runs up the chain through IT to explain it.  The final way is frankly Sharepoint.  Have to put policies in place.

Mike: Cisco is really trying to change the way we work.  We are moving to collaborative communities at all levels towards the goal of growth.  We have people with deep expertise that are difficult to locate.  The mandate Chambers gave to IT, which has done a great job, yet there are ways to better respond to users needs, so they are investigating Enterprise 2.0 and how to integrate it with communications.  From the bottom up perspective there are wikis across the organization, they aren't integrated and you have to support it yourself.  Wiki is a much better way for managing projects, such as a product launch that is a massive cross functional effort.

Toby: I was a young guy once.   A quick comment, not to pick on Motorola because they are a competitor.  You have to build it as a platform, with freedom to use, because if it gets big it will be out of control.   Needs to be searchable, enable migration.  But to answer the question, you can't tell people not to multitask.  The old school pre web people were actually some of the biggest adopters.  Super simple to use and really useful.  These are people with greater knowledge and more to gain by using these tools.  In some places, I don't know, Kentucky, it may not be that way.  But we found the adoption curves were pretty much even across the age groups.  40+ a little more usage, but the closer you get to real work, the more it is used.

Oliver: some workspaces are earlier adopters and cultures for new technologies.  At Northwester Mutual, their employees are mostly older, 33% retiring in 5 years, and they are struggling with this.  In youth, 77% are creating content, but at 41-50 only 21% do.  Even people who are reading, there is a group of people who don't even participate in this way.  You are dead on with the knowledge champion role, the evangelists who can show people the way.  One or two people you have the propensity to adopt start using them, and to better effect, and the guy next to them sees and copies so he is not at a disadvantage.  The stealth launch needs support for the spread, for the evangelists, but having someone see how it is valuable to them in their business context is the real goal.  Asked a law firm and they said the older lawyers weren't jumping on these tools, as they were with Blackberries initially, now he can't pry them out of their cold dead hands.  Need time and familiarity, and the right to fork.

Toby: hard to measure the results.  Old school companies built campuses for a reason, discovering people by bumping into them which indirectly helps projects.  We see this happening 10k times a day.  Inside IT we are seeing the cycle time on delivering product and getting stuff done really ramp up.  A little medium is the message going on.  I see less email, which is the worst thing going through an enterprise.

Oliver: really hard to measure.  Same with email or a portal, a soft ROI.  Hard to put saving five minutes a day into dollar terms.  And it takes a leap of faith.  At the same time it depends on the installation.  One company that rolled out wikis, took a legacy database that tracked IT standards, said we can put it all in a wiki, have every IT employee update it on their own when things change.  Moved two people off of the project and on to something else. 

Toby: you can tell lots of stories like this.  Before and after with a sales wiki: borrowing materials, building a pitch, find the 12 people who worked on something like it before.  People in logistics see four red things in a dashboard, click on them to the wiki.  I say cycle times are down 12%, they say that's great how did you measure this?  I made it up, but it was based on the best information available.

Oliver: Employee surveys are a good approach.  Surface anecdotes.

Mike: (Ross: honestly, I think Cisco will have more to talk about when it comes to results, without defaulting back to unified communications, by next year)

Toby: The question how do you help people understand how to search and tag appropriately.  We have 2600 users on Scuttle, but a bunch of people who look across tags.  All of it is a simple one page description of how you do it, if you need more it is not the right solution.  I stole a rule from the CIA: if you can't feed the team in two pizzas, get someone off the team.

Twitteroll

What do you get when you cross a Blogroll with Twitter?  A Twitteroll.