The Meaning of Search
by Kathleen Gilroy
I spent time at the FastForward 07 conference in San Diego asking people if search had changed their lives. Here’s what they said….
Author Archive
Interview with Tim O’Reilly
by Kathleen Gilroy
Tim O’Reilly on web 2.0:
- Web 2.0 is defined by building systems that get better as much people use them. This means asymmetric competition in the information business. But there are opportunities to work in the global information commons. O’Reilly hosted a panel where he interviewed the search person from Reed and the head of business development for Fast. They discussed producing more contextual search and looking at federated search where the data coming from multiple customers was combined and made available.
Interview with Andrew McAfee
by Kathleen Gilroy
Andew McAfee on enterprise 2.0:
- Enterprise 2.0 is about new forms of collaboration and unlike previous enterprise computing efforts, e20 enables the expression and capture of judgement.
- E20 will not happen just by building new technologies and expecting people to use them. It is hard to get e20 to become part of the DNA of a company and it will require sustained management and leadership through coaching, rewards and incentives, leadership, and building a culture that is attuned to the benefits of working in this new way.
- E20 is very different from groupware (Notes, Sharepoint) in that it is very unstructured. Groupware often failed because it demanded too many rules and the terms of interaction were defined from the start.
Ray Lane’s interpersonal enterprise
by Kathleen Gilroy
Here is a quick interview I did with Ray Lane last night after he opened up the conference.
[Tags: ff07}
“Why can’t we build this on Sharepoint?”
by Kathleen Gilroy
I have just started to demo our new Otter Networks service – it takes the best of web 2.0 tools and makes them available to communities inside and outside the enterprise. (You can see the slides that go along with the demo here.) After having done this for four groups, the first question that comes up is why not do this on Sharepoint? I have my own reasons which I will outline below. But I would love some additional help on articulating why or why not just do this in Sharepoint? Here are my responses:
1. Social Networking/Digital Identiy – Sharepoint is designed for internal collaboration via email and document sharing. The newest version does have blogging, wikis and RSS capabilities:
(From the Microsoft web site) MOSS 2007 provides a number of new features while improving the existing collaboration platform by providing built-in abilities for document management and alerting, adding new capabilities like blogs , wikis, and a Web Part for publishing and receiving RSS feeds.
But Sharepoint does not have built in social networking and rich profiling which we think is key for finding the right people to help you work. Each person in an Otter Learning Network is automatically given a weblog and a dynamic profile page that includes tags and aggregates all of the contributors you have made to the network. People, contributions and ideas can be searched for and tracked using RSS feeds.
2. User Experience – Web 2.0 consumer applications have been successful because they are open, easy, and fun. Sharepoint has a notoriously complex user interface which along with IBM, Jerry Bowles characterizes as “top-heavy style that aims to please IT by building in a lot of expensive control functions that frustrate end users and ensures that social software experiments are likely to fail in large organizations.” Otter Networks are designed to “‘consumerize” the experience of moving seamlessly back and forth between the desktop and web applications.” (Again Jerry Bowles.) We use a web service (iUpload) to customize how information is published and shared and integrate the best web 2.0 consumer tools for managing podcasting (Hipcast) and video (Brightcove) and the desktop (Netvibes). Otter Networks are open and simple in their structure thereby making it easier to build contributions and participation.
3. Cross network collaboration. Sharepoint is designed for work group collaboration inside the enterprise. But increasingly work is done inside AND outside the firewall. So as part of our service, we build and maintain a library of public materials (podcasts, videos, slide decks, pdf files, and blogs) that can be syndicated into each network and used by contributors.
4. Tagging. While Sharepoint does incorporate search, it lacks tagging—a web 2.0 feature that we feel is critical to collaboration. Tags enable personal categorization of information inside a network. Tags make it easier to find and track both information and people. Tags make collaboration simple and seamless.
5. AJAX Desktop. Sharepoint now incorporates an AJAX-like desktop in its MySite feature:
My Site that fully utilizes the functionality of a SharePoint site. Users can create lists, document libraries, or subsites under their My Site and customize the viewing of these sites for both their personal and workgroup views.
While I have not yet tested this, it doesn’t look like it has the flexibility or the personalization of the AJAX desktop. (It would be great to hear from someone who has used My Site and who can compare and contrast with an AJAX Desktop.) For demo purposes, we have integrated our first Otter Network with Netvibes. Netvibes is very much like Google desktop and Pageflakes and Microsoft Live. We have selected it because it supports secure RSS feeds and it has a good API for developing modules that integrate with our network. We can set up a nice podcast player in Netvibes, for example, that draws in all of the podcasts in your network. And we are working on enabling publishing from Netvibes to the network. All of these things make the user experience simple and fun.
6. Me-First design. Sharepoint and other enterprise collaboration systems like Lotus notes are missing a key ingredient in today’s social networking systems: a focus on the individual rather than on the work group. Stowe Boyd described this as “Me-First design”:
The basic model of 90’s era collaboration, a la Lotus Notes, is all about the group. Information was managed in group-based repositories, then passed around for review, or published to intranet portals via customized apps. Information era workflows where people are first and foremost occupiers of roles, not individuals, and the materials being created are more closely aligned with groups than individuals. Web 2.0 social tools—largely—work around a different model. Social networks—explicit ones like MySpace and Facebook, or implicit ones in social media—are really organized around individuals and their networked self-expression.
We have built Otter Networks from the ground up with a focus on the individual. Our design incorporates the structure described by Boyd here:
I start with a profile of myself, since I am the center of my network. I characterize my interests, history, job, whatever. This could include feeds, queries, and all manner of dynamic information, not just static text. I tag myself, to make it easier for others to discover me. The buddylist is the center of the universe, so I next start to link to those people and sources most important to me. Their traffic—flow of insights, recommendations, and presence—is the most important thing forming my world. And of course, I want to share my traffic with my network: links, recommendations, posts, presence. All my downstream buddies, those who want to read my traffic, can access it. But we don’t need groups to do so. Instead of groups, we need groupings: tagging the elements of network traffic is sufficient.
This design philosophy of building around the individual rather than the group profoundly changes the user experience. In the new web the individual contributor is at the center and everything emanates out from there. In an application like Sharepoint, the individual exists to serve the group and is empowered simply to publish and read inside a shared private space. Our experience with bringing web 2.0 inside the firewall is that blogging and rich profiles give people a new kind of reputation and expertise. They make it easier to present yourself and to be found. This is really transformational in terms of engaging participation and contribution. In the Otter Network, the individual is given recognition and her expertise is made manifest.
So please post your thoughts on this pro and con….
Adoption Issues
by Kathleen Gilroy
One of the reasons I think you are not seeing good discussion about adoption is that there are very few good examples of adoption of enterprise 2.0 services. If our experience is any indicator of what is going on generally, there are a couple of good reasons for this. With each new client we spend a lot of time struggling to get the approvals of senior execs and IT before we can undertake even modest tests of web 2.0 technologies inside the firewall. Once that hurdle is overcome, then we face an even bigger adoption hurdle: what Andrew McAfee calls, the 9x better than email problem. The perceived value of a new web service has to be 9x greater than email for it to gain any adoption.
I think there are ways to get people to adopt these services but the technology vendors are not offering them (nor should they be expected to). We are trying a few things that I will continue to document in my blog but I believe we need more of a user’s community where people who are working to implement enterprise 2.0 inside their businesses are sharing their experiences. Right now the main forums for writing about this subject are more theoretical than real. That is not bad because theory can precede practice and can give us good mental models for how to proceed. But more is needed.
We have had some success with adoption. Here are some concrete things we have done that may be of help:
1. Make the web service very simple to use – so simple that there is virtually no learning curve. For our new learning network web service we are integrating everything into Netvibes. Users set up a Netvibes account and then go to a start page where feeds can be added with one click. Once the feeds are added to netvibes, everything can bee seen in one consolidated view.
2. Tie the service to a critical business process or need and customize it to that need. Our first learning network is being built around the process of conflict resolution for a customer where conflict is an endemic problem and any performance improvements will make a huge difference to people and to the company.
3. Model best practices. As part of our service we include a Learning Director who models how to work in the next and coaches people on how to work in the world of enterprise 2.0. In my podcast interview Andrew McAfee, we talked about the issue of management in the world of enterprise 2.0. On public web, the peer production services are not managed. They grow and benefit from network effects. How then do you manage “network effects?” This is a big open question that needs work. Because the long tail doesn’t scale down to the enterprise, management does come into play: incentives, rewards, recognition are all critical ingredients to successfully managing networks.
4. Train. Train. Train. People need to be shown not only how to do things but why. To do this, we are using a new low-cost web conferencing service called Adobe Connect and we will be offering weekly web seminars where we make compelling arguments for participation.
These are just baby steps into this new world. To move things to the next level, I plan to do two things: write a white paper on this subject (if you have stories to share, please leave them in the comments section here or email me kathleen@ottergroup.com) and help organize a community of enterprise 2.0 users.
An interview with Tom Mandel
by Kathleen Gilroy
I spoke with Tom Mandel of Connectbeam. Tom has been involved in developing social software for a number of years and during our 30 minute conversation we talked about how web 2.0 technologies like tagging and digital identities are migrating inside the enterprise. Tom has some powerful insights into how enterprises begin to adopt these new technologies and how they are integrated intothe existing “IT stack.” I think you’ll find our discussion very interesting. I did. Here’s my interview with Tom Mandel of Connectbeam.
Where oh where does my profile reside?
by Kathleen Gilroy
[cross posted on and Future of Communities]
In the past couple of days I have had conversations with two people representing different enterprise 2.0 services (iUpload and ConnectBeam) about managing the online profile. In thinking about building second generation web communities, I have come to believe that the online profile is at the heart the new web. In the search economy, you need a dynamic digital identity. It is the means by which the right people find you and then connect with you. But in the world of web services, where oh where does my profile live? I know have mini profiles all over the place: I started with my blog and added a profile on Linked In. I thought Linked In might provide me with a good home for my profile but when the fee for service elements were added to Linked In, the really useful aspects of hosting my profile there were lost — people can’t find me unless they pay for a higher level of service. I’ve got mini profiles on Flickr and del.icio.us but they can’t act as profile central. And I’ve just added a new profile to our new Otter Networks service which is built on iUpload and Netvibes.
What I need is a place for my profile that can be plugged into any web service I join. And by plugged in I mean can dynamically draw text, bookmarks, images, and videos from all of these services and build them into a dynamic view of what’s going on now. And I’m not dealing with the issue of residing “inside” an enterprise.
I’ve been interviewing people for a paper and podcast series on new communities and when I ask them about profiles, this is what I hear:
We connect into Active Directory but that is primarily for authority and authentication. We recommend that you build a rich profile on our service and at some point in the future we will integrate this rich profiling back into the enterprise directories.
Is this really how it is going to go? When and how will these things be integrated in a way that really does provide me with a rich, dynamic digital identity that can cross web services as I move in and out of communities of interest and practice?
Interview with Andrew McAfee - Harvard Business School
by Kathleen Gilroy
I spent some time last week with Andrew McAfee at HBS discussing the subject of enterprise 2.0. You can listen to my podcast interview with him below.
A transcript of our talk is attached here. 
The most interesting aspect of our interview was our discussion of management and leadership in the 2.0 enterprise.
We talked about network effects:
Kathleen
Once things are up and running, success has to do with network effects. My next question is can you activate network effects inside the enterprise? Can you manage this process and what does management mean?
Andrew
This is a really fundamental question because there are not managers on the web and web 2.0 has grown and flourished without this constituency that we have inside the corporation called management. The idea of network effects gets tied closely to the long tail. Even if a small number of wikipedia readers ever make an edit, there are still enough of them that we get the result that we observe, this huge, growing online encyclopedia. The web 2.0 long tail will not scale down to the enterprise. The .001 percent of people contribute to a wiki. That’s essentially no one contributing to it. The biggest concern people have when they think about network effects: do these network effects still happen when they are not enough actors to activate them? This is where management comes in to create a culture where people are encouraged to contribute. We have a higher ambient percentage of people contributing to these platforms as opposed to just passively consuming them. You don’t need that many contributors before good things start happening.
And incentives:
Andrew
There are a variety of things. Most of the smart managers I’ve talked to tend to prefer the soft over the hard when it comes to incentives and motivations. They do coaching; they encourage their people to contribute to these platforms. But they don’t say you must make 20 blog posts or you won’t get any bonus. I did it in my class I’ve got some very busy students. They’ve got too much else going on. Inside a company what they’ve got going on is their jobs. We don’t need to put in place super hard incentives to encourage participation. We do need to do managerial work to create a culture and spirit that this is how we are going to collaborate and share information and knowledge.
And leadership:
kathleen
What does leadership look like in the 2.0 enterprise? Does it change?
Andrew
Yes it does. I’m not sure how much it changes. I can see a couple of things. One is that these technologies really put to the test the leadership boilerplate that if you want to gain control overcomes you have to let go of control over people and events. That’s very easy to say. These technologies force you to put that philosophy into practice. When you deploy these things you by definition can’t look over the shoulders of your people and make sure they are doing the right thing. You can’t piegon hole them into participating you want them to, they are going to do it they way they want to — at least at first. It is a test of whether or not the leadership of a company means what they say when they put that in the annual report. The other challenge to leadership here is it is going to make leaders more as coaches or as shapers of an online culture as opposed to people who get to dictate the online culture or dictate how IT gets used. If you deploy an internal blogosphere you can’t specify in advance what that is going to look like, but there are many things you can do to shape it over time so that it heads in a direction you find productive.
kathleen
Is it a modeling behavior. Do the leaders have to be bloggers?
Andrew
It is not the case that the leadership of a company is at the forefront of everything they want employees to do. I don’t think they have to be bloggers but it falls back on coaching and this vague word of culture which turns out to be incredibly important, and on signaling to the organization the behaviors they want to see. Formal and informal rewards of highlighting the great job someone did and just showing that you consider this new stuff important and that you don’t think all we ahve to do is deploy it and walk away and think the employee base will do it all on its own. That’s a very bad strategy.
Enterprise 2.0: Getting started
by Kathleen Gilroy
One of the problems with the discussion about Enterprise 2.0 is that it remains abstract and conceptual. Until organizations see real improvements in performance that outweigh the risks in implementing web. 2.0 solutions behind the firewall, they will not move forward. So how do you begin? What are the specific steps that you need to take to get an enterprise 2.0 project underway?
I have been involved in implementing several of these projects this year. In my workshops on this topic, I suggest that groups find a deep need that can be addressed by a distributed architecture and start there. These needs are often about improving performance that can be directly tied to improved revenues or profits.
Once this need has been identified I recommend prototyping with consumer web 2.0 tools. Most of the features of the enterprise 2.0 toolkit can be demonstrated on a small scale with readily available tools: blogs, wikis, podcasting, and rss aggregation. It is very simple to build a platform for your prototype out of these tools.
Next step: get going. Give prototypers some basic training in how to use the tools and what is expected from them. Emphasize how using the new tools is going to save them time and expand their access to critical information that will improve performance. Get people solving an important problem using the web 2.0 toolkit.
Bring in experts to provide foundational content for the project and stimulate and guide thinking. Model behavior for how you work in a 2.0 network.
Run your project for a sufficient length of time (120 days) so that the group really gets the experience and benefits of working in this new way.
I have used this “recipe” with quite varied groups of people and found it to work in every case. There is a reason that 60 million people have started blogs and thousands are podcasting. These tools really work. They offer a powerful new channel for creative expression and efficient new means of getting the right information. But they are a new language and like a new language, the best way to learn them is immersion.
And in every case our prototypes have propagated beyond the initial small project. People tell their friends and colleagues about their experiences. They won’t abandon their blogs and aggregators once the prototype is over. Innovators in the group start their own prototypes.
Enterprise 2.0 is emergent. So let is begin simply and emerge naturally.
