Crossing the Services Chasm

Sun has a large and vibrant services organization, but historically the communications channel between Services and the Systems Engineering Organization has been somewhat limited. SEs tend to think in terms of products and technologies first, and services second.

It is of course good for Sun, and good for customers, if services are included along with a product or solution sale. Unless the sale is driven by a Professional Services engagement, services attach tends to happen at the end of the process.

So how does this fit with OneStop? Most SEs utilize OneStop to get help gathering information about a product, technology, or solution – and find the links to the appropriate sales tools, communities, etc. As Sun’s roots are as a product company, that’s how people think, and that is where people start the investigative process. It’s not terribly clear to people where to find information about services, for a given product or solution.

As it turns out there is an excellent, but now well known, internal site that maps services to products. Our Eureka moment is that via a web service we are now able to dynamically populate a Services section of each OneStop with the services for that product.

Now, hopefully, the sales teams will be thinking about services earlier in the cycle, as the information will available with no friction.

Search: Separating the Wheat from the Chaff with Information Equity

Search on corporate intranets is difficult, often because algorithms based on page rank don’t work particularly well. In short PageRank is a “vote”, by all the other pages on the Web, about how important a page is. A popular document on the corporate intranet may have very few pages linking to it. Without page rank the search results are ordered largely by frequency of key words, meta data, and currency. This makes it almost impossible to find a given page with a popular or overloaded term in the title such as Solaris 10, Cloud Computing, Identity Management, etc.

SunSpace, Sun’s internal enterprise wiki, with an integrated document repository, is based around the notion of Community Equity. Each person, and each document is assigned an equity value. A document’s Information Equity is mostly based on:

  1. Hits or downloads.
  2. Updates.
  3. Number of different people accessing/updating the page. (very important in a collaborative wiki!)
  4. Currency. The equity decreases over time.

So how does this fit into search … ?

SunSpace search has a 3 tier architecture. The back end is a commercial search engine that indexes the content (wiki pages and uploaded documents) and delivers the search results. We spend a lot of time tuning this engine so that the optimal weighting is given to titles, urls, keywords, and various meta data.

The middle tier is a set of feed readers that monitor all the updates on SunSpace. The feed readers create “stubfiles” for every document that contain all the meta data, which includes equity, tags, creator, and last updated date. The feed readers run continuously and notify the back end whenever a page is created or updated. A big advantage is that new pages and updates are added to the search index immediately, no more waiting days until the crawler finds the new documents.

The front end is where most of the action is. When a user types in a search query it is submitted to the back end and 100 results, in xml format, are returned. The stubfile for each result in then read. The initial results ordering is directly from the back end search engine, but the user is given the option to Sort by Information equity, or Sort by Date. We also display other relevant information such as tag clouds and communities for these 100 results. Each document has a creator, who is generally a primary contributor to the page. We assemble a list of the creators for the results, then credit each creator with the Information equity of the result they created – and display a list sorted by sum of information equity.

A few simple use cases:

  1. Find a new document. I am constantly creating documents on SunSpace, and generally am not careful with the titles. Two hours later when I have forgotten the title and need to forward the URL to a colleague, I simple search for myself, then Sort by Date.
  2. Find a popular document. The second case refers back to the beginning of this article. Let’s say a colleague mentions a cool wiki page on Solaris 10 that’s all the buzz. I’d search for “Solaris 10”, then sort by Information Equity.
  3. Find the expert. I have a big presentation on Cloud Computing, and need to seek the advice of a knowledgeable colleague. I search for Cloud Computing, then refer to the right of the results page for the people with the most Information equity. (for that particular search.)

OneStop, A Trusted Community in Sunspace

I continue to hear that people are confused about the positioning of OneStop verses SunSpace. (for people who haven’t read this blog, OneStop is a managed community within SunSpace.)

Peter Reiser, the architect of SunSpace, recently posted the blog entry Trusted content through facilitated communities. It’s an awesome post, but I’m still struggling a bit with the word “trusted“.

We are finding that the the need for OneStop has increased, rather than decreased. With Communities and wiki sites all over the place, people don’t know where to look. Will a random community page be up to date? Will it be in a useful format? Will it contain the info SEs need? Will it be easy to find, either via browse or search?

I’m not sure the combination of the above = trusted, but I don’t have a better word! Any suggestions appreciated.

The Value of Moderation

Good news! SunSpace is a raging success. Thus far 20,000 users have logged into the system. It is full of a variety of content, great, good, bad, irrelevant – and in a myriad of formats!

OneStop is now part of SunSpace. Users find this very confusing. Is OneStop dead? Has OneStop been subsumed by SunSpace? I’m hoping that referring to OneStop as “OneStop on SunSpace” might help clarify.

The success of SunSpace has amplified the need for (Sun internal users)to have an easy way  to find information on Sun products. If you search for Identity on SunSpace you’ll find references in several dozen spaces. As we had hoped, we are seeing communities form around technologies, often based in geographical areas, many organizationally based.

Where should a user go to find information on Sun’s Identity products, perhaps OpenSSO? What if you don’t know the name of the product itself?

The answer is “OneStop on SunSpace”. Here you will find a list of menus on the homepage that are convenient for browsing, particularly useful if you don’t know the exact name of the product. You’ll be directed to a OneStop page that contains the information you are looking for, which is in a consistent and familiar format. The page has an author (or moderator) so you can count on the fact that it is reasonably up to date. The OneStop team continues to work behind the scenes to help ensure content completeness and currency.

We’re finding that useful search on SunSpace is challenging, just by virtue of the myriad of references to product and technologies in the different spaces. The implementation we are currently playing with segregates the results that link to pages in the OneStop space and displays them at the top of the results list. Google of course uses page rank to help determine relevance. SunSpace computes Information Equity per URL based on ratings, downloads, edits, etc. We’ll be enabling search users to search for keywords, then sort those results by Information Equity.

Moving forward, perhaps we should simply create new tags that will imply that a SunSpace page is moderated and in a consistent format. Search could read those tags, then group the results accordingly. With the judicious use of tags we could also generate a useful browse mechanism so that the menus would not be necessary.

At any rate, the need for page ownership (moderation), and overall OneStop ownership (to help ensure content breadth, currency, and accuracy) continues to exist.

OneStop is not dead!

I’ve been hearing that people think OneStop is dead! This is not the case. OneStop is simply better and runs on a more modern platform. I try and describe what OneStop is in the blog posting What is Onestop – next ?. In short, users can continue to go OneStop and find a set of pages that are:

  1. owned and moderated.
  2. adhere (loosely) to the OneStop format. The familiar categories are in the same order.
  3. are up to date and accurate, or at least there is framework (people and tools) in place that tries to ensure this.
  4. are there. In other words the breadth of OneStop space is complete and controlled. Products , Technologies, and Programs of interest to Sun CEs (Customer Facing Engineers) will be there.
  5. are easy to find and navigate. It’s a limited universe (~500) pages so we can make the menus, search, and a-z work well.

However, OneStop pages are now Wiki pages that support commenting and voting. There is also a WYSIWYG editor (albeit not a great one) that makes simple edits an order of magnitude easier.

SunSpace now has 200 communities and the number is growing rapidly. We’re seeing a significant need for communities to have a OneStop page for “non community” members. No content redundancy is required thanks to macros like {include} and {simplelisting}.

We will take action very soon to ensure that a page looks like a OneStop page at a glance. Currently all SunSpace pages have a similar look.

What is Onestop – next ?

Great news! We are now live  with 5 OneStop pages on CE 2.0. Anyone on the SWAN can now, optionally edit a page, in place, using a WYSIWYG editor. We’ve maintained our model of the author owning a page, with the new ability to control access.

We envision 3 general categories on CE 2.0:

  1. OneStop pages. These pages will live in the OneStop space, and each will have a primary author. The fact that it is branded a OneStop page means several things. The page:
    1. has an owner and is a moderated.
    2. adheres (loosely) to the OneStop format. The familiar categories are in the same order.
    3. is up to date and accurate, or at least there is framework (people and tools) in place that tries to ensure this.
    4. is there. In other words the breadth of OneStop space is complete and controlled. Products , Technologies, and Programs of interest to Sun CEs (Customer Facing Engineers) will be there.
  2. CEpedia. This space will be very similar to the existing CEpedia, as anything goes. It is not controlled. Users can create or update any page in the CEpedia space, in any format, at their leisure.
  3. Communities. The community area is really the thrust behind CE2.0. CE 2.0 communities have been described as a mashup — Facebook meets LinkedIn with a content-management system thrown in for good measure. Historically email has been a primary communication vehicle inside Sun, so we offer the option to build a community, with membership and access control based on an email alias.

We’re hoping for interesting synergy between OneStop and communities. A user might go to the OneStop space to find information on a product. While they are there they might notice there is a community built around that particular product or project, and actively engage in a discussion or forum. Finding the expert will turn into a non issue. We can do things like dynamically populate sections of a OneStop page based on how documents or pages are tagged in a particular forum.

Now, what do we call this new beast that we’ve been referring to as CE 2.0? (CE 2.0 has been our project name.) Onestop or Onestop2 doesn’t seem to work as the OneStop brand implies all the stuff listed in section 1, but not social – web 2.0 community as we know it today. Users are tired of the new brand, or tool of the day. Maybe we should just stick with CE 2.0.

OneStop and Community Equity

My colleague Peter Reiser has been making great strides forward with his notion of Community Equity. For a detailed write up see his blog post. He was even interviewed by Scoble on the subject!

(Heresy, heresy) but I have mixed feelings about Community Equity in the context of OneStop. I like the notion of community, and I really like the notion of encouraging participation. Having our users easily rate and comment on OneStop pages should prove invaluable. Ratings will supplement our current metrics of downloads and currency to give users a good solid indication of page value. Comments will evolve into discussions, where as we currently only offer page feedback. Discussions will then move into forums. All great stuff!

I’m a little more skeptical on the Community Equity front. I not sure our users will be motivated to participate (more) if their Personal Equity rises. Historically the lion’s share of OneStop accesses has been from people looking for answers, and secondarily browsing for information. I don’t expect that to change any time soon.

The big question is  “What’s in it for me?”. We’re asking our users to spend their limited time, rating and commenting. In my experience people need a recognizable return on investment to participate. Is a high community equity rating, and being listed in the top 10 on the homepage enough? I don’t know; I hope so. I do anticipate arguments about the formula in how CE is generated. Is it fair that a person who (perhaps without a lot of thought) rates 50 documents, gets a higher rating than someone who moderates a couple of OneStop pages that aren’t popular products, or a person that has submitted one “white paper” or Technocrat article?

OneStop Kicks Butt in User Survey!

The results are in! If you are on the Sun Network check them out.

We got 776 responses, an incredible number. Users feel very strongly about OneStop. We asked the question “How would you feel if we pulled the plug on OneStop?. (We don’t plan to do this, but the question elicits great comments.) The results were:

I’d be totally irate! 538
I’d get by 153
Don’t care 11
It’s about time 10

A typical comment was “That would treble my workload and response time”.

We contrasted OneStop to other sites the SEs generally use and found that people almost always look to OneStop first, and give very high ratings (4.3/5) when asked “How useful is OneStop to doing your job? (1 low -> 5 high)“.

A primary motivator for the survey, aside from understanding utility, is getting a feel for user priorities. Are social networking features important? Should we accelerate our move to an enterprise wiki?

Our users were very clear in expressing that the top priority, by far, is accurate, complete, and up to date content. They like the fact that content is easy to find on Onestop, and that the site is simple and consistent.

My conclusion is that moving forward with CE 2.0 we need to not break what’s working. We need to be careful about adding complexity. Social networking features that enable community are cool, and will hopefully help us deliver even better content, but not at the expense of expedient access to information.

My feeling is that the majority of the users are not planning on being direct contributers, so we need to make sure we optimize that path. The browse experience needs to be robust, search needs to work, response time needs to be fast. Information should be no more than 3 clicks away.

Make sure you watch this space (follow on posts) for a discussion on how we hope to leverage Community Equity to help raise the bar on content quality.

OneStop 2 Conundrum

We are planning on moving OneStop on to a new Confluence based platform, called OneStop2, by the end of our fiscal year. (June 30.) The Confluence platform is being extended with many new features including:

  • Contribution Equity, measured from contributing and rating content.
  • Robust, integrated document repository
  • Tagging Services
  • Expanded XML feeds (ATOM, RSS)

The Confluence platform (an enterprise wiki) is exciting in it’s own right and will provide us functionality to move OneStop into the world of Web 2.0. The integrated features I am most excited about are WYSIWYG, access control, and a commenting service. This will enable a OneStop page to be easily updated (via WYSIWYG) by  either anyone, or people specified on an access control list. Our model of primary page ownership (or perhaps moderation) will continue.

The conundrum is the mapping of OneStop pages into Confluence. Confluence supports the notion of spaces. Each space has a home page and child pages.

Should each OneStop page be a space, or should OneStop pages all exist in a OneStop space? Unfortunately Confluence doesn’t support the notion of space hierarchy, so for example, we can’t set up a hierarchy on the order of OneStop -> HPC -> ClusterTools.

We also intend to move CEpedia on to this new platform. Should there be one CEpedia space, or several?

Currently the 5 most accessed OneStop pages are

  1. Sun Secure Global Desktop Enterprise Edition
  2. Sun Fire T2000 Server
  3. Sun Ray appliances
  4. Sun Fire X4100/X4200
  5. Solaris 10

I don’t think many of these should be spaces in their own right, but perhaps should be children in other spaces. If you have thoughts on this, please comment.

Moving Forward With Confluence

This is not yet a done deal as we still have small things like licensing () and more functionality testing still to work through, but as Maverick says in Top Gun, “Things are looking good so far”.

We asked the OneStop authors to play with our Confluence test instance, read about the functionality, and provide us feedback. The response was uniformly positive, a first with this vocal group of 300!

Kemer Thomson, the guy who runs the Sun Blueprints program had some excellent feedback worth sharing:

Confluence seems to address not only the challenges in creating wiki content, but makes it a pleasure and opens new possibilities. I have played around with great success. Everything seems to work. I ran into no problems … nothing crashed. Cool features, like creating an RSS and adding tags worked intuitively.  I could recreate my entire OneStop content quickly and without any concessions to the content and structure. Indeed, the ability to add plugins and control access to pages opens up many possibilities.

I find it truly compelling that we can combine an almost seamless transition with features and functionality that will enable solutions we haven’t even considered yet.

I can’t wait to see what our smart motivated group of authors will do with this truly enabling platform.