Agile for Embedded Part I

I have used a number of development techniques in my career and, most recently, have taken up Agile Development as a means of rapidly building systems. However, Agile's heritage is in the development of web sites and I am currently interested in the development of embedded systems. The differences between these two require some modifications to the basic Agile techniques. I plan to publish a series of posts that describe the changes I would use.
This post is concerned with User Stories.
A User Story is a simple way of capturing the requirements of a system. It has many similarities to the Use Case as defined in UML. Both define requirements based on a consistent outside-in perspective from a User to the System.

The User Story is a single sentence in the form
As a [user role] , I can [feature] ,so that [benefit] .
Instead of the generic "user", the user role tries to identify different classes of users with different requirements. User is too generic. But Tom, Sally, Harry are too specific. When we see that Tom and Sally are customers and Harry is the administrator then the roles of Customer and Administrator emerge with separate interfaces, transactions, and requirements.
The feature  is what I want to accomplish with the system as the User Role.
And benefit; is why I want to use that feature. What value do I get from the story.

Suppose we are building Amazon.com. Then we might see user stories like:

As a Customer, I can search for something, so that I find the item I am looking for.
As a Customer, I can select an item from a results list to get more details, so I can decide if I want to buy it.
As a Merchant, I can register my business as an Amazon storefront, so I can sell my products.

So the catch with an embedded system is that the rich interaction with Users is not always present. In a "headless" system with no user interface, the embedded system depends on sensors to modify its behavior and will often send outputs to actuators. Let us imagine a thermostat as the system. I may have a simple user interface so that the homeowner can:

As a homeowner, I can change the mode of the HVAC between Off, Heat, and Cool, so that the house is maintained at the temperature I want.
As a homeowner, I can change the temperature limits, so that the house is maintained at the temperature I want.

A simple context diagram for the thermostat is shown below

The "Commands" shown on the diagram are for mode control and temp control, and map cleanly to the two user stories. But what about the "Display"? Unlike a web site where the user takes an explicit action of clicking on a link to see a new page, the display on a thermostat is always there. Or is it? That is what the Battery is doing on a diagram. Remember what the display looks like without the battery inserted? Blank. Insert the battery and the display comes on (usually showing default information) and stays on until the battery runs out of power. One could imagine a user story like

As the homeowner, I can insert a battery, so my thermostat will turn on and maintain my house temperature.

But what about the Thermocouple that provides the "Current Temp"? We use that to compare to the Mode and Set Point so that the Thermostat can tell the HVAC to turn on or turn off.

There is no natural user story involving the homeowner and the thermocouple.

My idea is to add event modeling as defined by Paul Ward and Stephen Mellor. An event is something that occurs at some point in time in the environment of a system and, for which the system should have a response.
This could also apply to the users but I think the user story format is fine for that. Below is a table that captures some of the events for this system

Battery inserted
Battery runs low
HVAC begins heating
HVAC begins cooling
HVAC stops heating
HVAC stops cooling
Temperature fluctuates

When the battery is inserted and when the battery runs low can both be determined by measuring the voltage.
We are not interested in the current voltage at anytime just these two events.

The diagram below shows the modes of the HVAC

This is a basic state diagram. If the behavior of our system will be different depending on what mode (i.e. state) it is in then using such a diagram helps separate the behavior and to better understand what would trigger the system to change states. For example the event of HVAC begins heating could occur when the system is Off. It should not occur when the HVAC is already cooling. Many events are important because they help us organize these modalities.

Finally, the Current Temperature that fluctuates is a special kind of event. The temperature always exists and has a value that the system wants to measure to determine if it should be heating or cooling and it wants to display the current temperature to the homeowner.

Events (like user stories) are intentionally abstract. As the response to the event is being analyzed one must answer how we will determine that the event has occurred. We must focus on the system boundary and look at the interfaces. Sometimes these interfaces are constrained. For example, suppose the HVAC manufacturers have gotten together and decided on a specific interface standard for status/controls. They might dictate the HVAC Status would be conveyed by a twisted pair wire with +5 volts indicating Heating, -5 volts indicating Cooling, and 0 volts indicating Off. So our thermostat could monitor the voltage and detect when the four events occurred.

So lets put this together in a user story / event table as shown below.

Conclusions
By adding events to user stories, I have introduced a means to capture requirements for embedded systems that do not have significant behavior at the human interface. This is consistent with the general approach of agile development because:

• The event/response/benefit complexity is similar to the user story and can be treated in a similar fashion in SCRUM
• The events have an outside-in perspective similar to the user story. This avoids "the system shall" type of requirement.

Stay tuned for Part II (Can Hardware be designed from User Stories / Events?)

Jeff Sutherland's Pearls

Last week I attend an Agile RTP meetup hosted at the offices of Relevance Software . I heard Jeff Sutherland speak about the "Five Problems solved for PatientKeeper". If you are into Agile then for me to repeat what he said would be redundant, and if you are new to Agile then you probably need some basic education before jumping into his talk.

However, I captured the following pithy quotations from Jeff:


1. "World domination always gets people excited". When the PatientKeeper management team was forming the company, they needed a vision of what PatientKeeper would be when grown-up. Jeff speculated that Bill Gates had a vision of owning the PC desktop. Sooo... the PatientKeeper team decided to own the mobile devices used by clinical professionals. They wanted to create a framework that would be considered the gold standard for third party medical application developers.

2. "If it is impossible show me the impediment list". Attributed to the CEO who took over PatientKeeper in 2003 and wanted the code deployed to hospitals after each Sprint. When told that the installation time was a bottleneck to this tempo, he formed the "Live by Five" program, which brought in PatientKeeper expert services to work with the hospital IT staff to make sure the installation took place by 5:00 the day of install.

3. "If Lean is all about removing waste, then SCRUM is all about removing impediments". Jeff likes to teach Lean Manufacturing to executives trying to adopt agile development because he believes that they can relate to the examples from the world of manufacturing.

4. "Is the Tango a methodology? Neither is agile". Someone in the audience asked Jeff to compare the light-weight "method" of agile with a bureaucratically heavy approach like CMMI. Using the analogy to the Tango, Jeff said that like that dance, the Agilista does not plan so much but responds to the circumstances of the project.

a mobile device dilemma Part II

Since last December when I shared the small family crisis caused by the way the cellular industry has conspired to keep me off of Android a couple things have changed:

• AT&T has begun to offer some good Android phones. I would seriously consider the Captivate for my use.
• Rumors of Verizon taking on the iPhone persist. If that happened I could move the family over and my wife and daughter could stay on the iPhone while I could pick and choose from a wide variety of Android devices.

However, my decision continues to be complicated by the fluid nature of the service provider landscape. Now it is the emergence of higher bandwidth networks. For most of what I do on the mobile device a solid 3G connection gets the job done. But if I could get 2-10x the bandwidth via HSPA+ or LTE then should I wait until the right Provider/Device/Price comes along on one of those networks?
Because of the financial impact of disengaging my family from 3 two year contracts on AT&T my easiest option would be to stay with them and face a simpler choice:

1. Wait until this September when I am eligible for an upgrade discount and jump on the Captivate and use 3G
2. Wait until AT&T roles out HSPA+ in early 2011 and hope there is a nice Android device able to take advantage of that network.
3. Wait until AT&T roles out LTE in 2011-12 and hope there is a nice Android device (by then it will have a 2GHz processor and an mega-mega pixel camera) able to take advantage of that network.

Obviously these three choices represent the constant agony of the early adopter... whatever is in my hand is obsolete and the nice new shiny gadget is always over the horizon. BUT that feeling is tempered by my frugal side so that I tend to upgrade about every two years. I am tending towards option 2 with some caveats:

• Since AT&T has not announced any concrete steps or devices towards HSPA+ (unlike T-mobile) I am afraid early 2011 could easily become late 2011.
• With AT&T moving off of an unlimited data plan, will HSPA+ carry a double premium? First a higher $/byte charge then 3G, and second, my natural tendency to play with the technology and do video conferencing on the device until my first monthly bill comes in (Ouch!).

Decisions Decisions.
What would you do?

Agile Team Dynamics

Last night at the Agile RTP meeting we had Don Gray speak to us on Agile Team Dynamics.

The meeting was broken down to 5 teams with each team having approximately 10 people. One person was designated as the Product Owner. Each team was given a bag of supplies and an instruction sheet for the Product Owner to give requirements to the team. My team got a bag of colored paper, tape, and pipe cleaners.

We were told to construct something that was:

• Artistic
• Tall
• Sturdy

We were given 20 minutes for the exercise. The picture shows what we created.

Then Don asked each of use to take some post-its and write down what we had contributed to the exercise.

I think my set was:

• Contributed vision of "Eiffel Tower"
• Started rolling paper into tubes
• Worked in team of two specialists Roller + Taper to create legs
• Asked other team to give us balloons

Then Don told us about the Kantor Four Player Model.

During the exercise, each of the team members was contributing in one or more of the four interactions. Don asked us to classify each of our contributions. I had thought most of mine fell into the "move" category. In fact, among all the participants, the vast majority of actions were classified as either move or follow. A few Bystand and almost no Oppose.

Don mentioned that in an Agile team setup the Scrum Master should be a bystander, allowing a self managing team to do most of the Mover type activities. I made the observation that a classic problem with many agile teams is that the Scrum Master (often being an old project manager) can not help but try to take over and "move" the sprint along. Don agreed and gave some war stories from his own experience on how this can be a problem.

Other comments from the audience pointed out that the team dynamics in our 20 minute exercise did not realistically compare with a real project having team members familiar with each other and with the political / cultural context of the company surrounding them.

My main take away for the night was that in any healthy development teams these four players can each contribute something of value and that all team members should accept the presence of these roles in the team dynamics.

Meeting Fred Brooks

Last night I met Fred Brooks.

I was moderating a panel discussion of our local chapter of the IEEE Computer Society. It was about "Practical Software Development".

Afterwards, an older gentleman walked up to sign our registration sheet and apologized for joining late. I glanced down and the name was "Fred Brooks".

I had this flashback to 1977 when I was a young software developer. Two years out of school with a shiny new Masters in Computer Science. I was just starting to call myself a Software Engineer and I was in this meeting with about ten older, wiser, non-programming, engineers who did not understand why my software was on the critical path and why I had just announced a slip in schedule. The Project Manager ( a terrifyingly gruff Sargent type as I remember) suggested that we add a couple of other programmers to the team to help get me out of the ditch. I can remember holding up my copy of The Mythical Man Month and saying "I think that adding more people to a late project will only make it latter. Let me tell you why".

This was one of those pivotal moments in my career. The ideas in that book helped me earn the respect of the Electrical Engineers, Mechanical Engineers, Industrial Engineers, etc who surrounded me but did not understand my profession. Along with books like "Software Engineering Economics", and "Structured Analysis and System Specification", the "Mythical Man Month", became the core of my Software Engineering.

So, thanks Dr. Brooks.

a Sojourn in the Clouds

I have been learning IBM's Mashup Center, demoing to clients, and weaving it into my Web2.0 briefing as an emerging technology. I am able to do this in the Lotus Greenhouse as a free service from IBM. Recently IBM released version 2.0 of the product and after a product webinar I was ready to try the new/improved product. Alas, the greenhouse only had version 1 (that has just been remedied). Anxious to get going I fired notes off to people in IBM asking when I would get the new shiny version. I then stumbled on the fact that IBM had v2.0 available on Amazon's EC2 and if one was only using it for developer investigation (compared to a mashup for multiple users) then the product was FREE.

IBM provided a getting started video and pdf so off I went. I would listen to the video for a bit and then try to repeat the steps. There are a lot of things to do up in Amazon Web Services to get started and several freeware programs one needs to download in order to get things to work. I got stuck several times and finally resorted to reading the pdf line by line and carefully doing EXACTLY what was written. I never got my Ultra VNC to work but was able to use my Firefox browser to link directly to the Mashup Center Instance running on EC2.

For the configuration I had requested (Linux with 4GB storage) I was paying Amazon about $1.75 for each 24 hour period. I left it on for a few days while I developed and demoed for a client and was then able to terminate my instance.

If I did not have access to Lotus Greenhouse, and I was someone wanting to play around with a program like Mashup Center then using a infrastructure cloud is a good way to do it. And how do I like IBM's Mashup Center? Let me build up some more examples and then I will post a blog with my experiences.

a mobile device dilemma

The exclusive deals that device manufacturers are making with service providers will ultimately cause a family crisis for me.

Today my family of three is using an AT&T unlimited family plan. On that plan we have three devices:

Blackberry Bold - me

iPhone 3GS - wife

iPhone - daughter

I picked the Blackberry last year when my old one died. It was not my ideal choice. I really wanted to jump on the Android bandwagon. I am an avid Google user and wanted to have a state-of-the-art experience of integration. Alas, I did not do so at the time because:

1. AT&T did not have an Android device (I suspect they may never have one).

2. The T-mobile HTC G1 was not a game changing device

However, at that time my wife/daughter were using dumber devices mainly with voice/texting capabilities and I probably could have migrated the family over to T-mobile.

BUT since I got my Blackberry (which has OK Google apps), my wife/daughter became iFans. At this point, I would have to pry the iPhone from my wife's cold dead hand.

So you see my dilemma. At some point there will be this wonderful, amazing device (like an HTC HD2 running Android 3.x) on a service provider like Verizon and I will either have to ignore it, convince my family to abandon their iPhones, or split the family unit into separate accounts ($$$).

I had hopes that Google's push for an open network would have worked out. Being able to purchase any device and activate on any network (radio compatibility assumed) would make this little family crisis in the making disappear.

What is the perfect Agile Tool? - It Depends.

Within the Agile/Scrum/XP community there is a love/hate relationship with tools that support the development process. I used to call these Computer Aided Software Engineering (CASE) tools but that term has fallen out of favor. CASE is more associated with the waterfall wold the the Agile Manefesto revolted against.  There is one camp of Agilistas that will only use 3x5 index cards posted on a board in a war room. And even when considering tooling, the complexity of the tool is a major discussion point.
Last night, I attended an agile tools shootout hosted by the aRTP group. We looked at the following tools:
Zen
Cucumber
PivotalTracker
Rally
ScrumWorks Pro
Microsoft Team System
IBM Rational Team Concert
Jira/Greenhopper

Actually, the demos were given in two separate rooms and I was only able to personally see the Zen, PivotalTracker, ScrumWorks Pro, and Rational demos. From the demos and what I could see from their web sites I would broadly seperate the Microsoft and IBM tools from the rest and put them into the more complex category. However, this is because with both of these tools the vendors are attempting to cover the full development cycle and making sure at detailed design and coding they have things covered. For example, Team Concert was demoed as an Eclipse plugin with source code control, build management, real time notifications, project management features all enabled
Other tools, such as Zen and PivotalTracker tended to be more like electronic 3x5 cards with electronic boards. They did offer the advantage over a manual system of being able to automatically calculate burn down and other statistics.
In the middle of the complexity spectrum were Rally and ScrumWorks Pro because they added more project management features and the ability to integrate with other tools.

So which would I pick? Like any good consultant the answer is "Depends".

It depends on the size and complexity of the organization using the tool.
It depends on the target architecture and technologies used (e.g. one tool I did not classify above is JIRA/Grasshopper which is specifically used for Ruby development)
It depends on the level of contol over tool content needed (e.g. several tools were SaaS with concerns over security)
It depends on the sophistication of the developers.
It depends on the level of formality required of the process (e.g. if federal certification of the software is required then more traceability and reporting will be needed)
It depends on the risk accommodation of the users (Want to go with a small flexible rapidly changing tool/company OR stick with a slowly moving but more stable large vendor)

I did enjoy the exposure to the tools and we will be holding another shootout in the future.

Virtual Conferences

In a previous post I mentioned attending the Enterprise 2.0 conference being held in Boston MA...only I was not able to be in Boston... I was sitting in my home office in Cary NC.

Today, I upped my game and simultaneously attended Enterprise 2.0 from San Francisco CA and the Internet Summit from Raleigh NC. Plus I did some billable work for a client and was writing this post. Can I have a productive day? OR am I evolving to the attention span of a squirrel. Anyway, here is how I would compare the two conferences from a virtual attendance perspective:

Enterprise 2.0

This conference offers a quality virtual experience. There are several remote mechanisms offered to gather information:

Twitter - #e2conf 2719 followers, On a typical full conference day there were 17 tweets from the conference. However, approximately 1500 tweets related to the conference occurred in the same time period. For example,

RT @ terrigriffith: Linden Labs brilliant. Gifted laser pointers allowed responses to ?s shown on screen http://yfrog.com/j8kepzj

BTW I thought the use of the lasers for audience participation is cool but they need to give me a virtual laser as well.

Facebook - 881 fans, There are a posts for each of the sessions, posts by exhibitors, and posts by participants.

LinkedIn - 50 members, 1 discussion posted. This is a weaker presence for this conference.

Video on Demand - David Berlind hosts demo presentations from the exhibit floor. Each session is about 10 minutes and what I like is that David asks some penetrating questions that expose weaknesses as well as strengths of the vendor's product.

Streaming Video - The A/V team uses multiple camera shots with some good closeups of the speaker and interweave slide images into the stream. Well managed scheduling.

Conference Blog - The blog is written by approximately 8 people and includes immediate summary of current conference presentations (e.g. Integrating Google Wave into the Enterprise) but also has entries between conferences (e.g. 10 posts in September).

Internet Summit -

This conference also focuses on the emerging web technologies and business models with content that covers both enterprise as well as consumer use of the web. They support the following virtual attendance tools:

Twitter - #isum09, 777 followers, all conference status posts. There were approximately 225 tweets per hour from individuals during the conference at mid day. Lots of commentary of speaker points that correlate directly with the video stream.

Facebook - 155 fans, 22 posts mainly status announcements.

Video on Demand - Replay of the main tent sessions.

Streaming Video - The A/V work is a little rough in a few areas 1) Listening to Richard Jalichandra of Technorati and was not able to see the slides he was showing. 2) Just before a session was to start the A/V person turned on the video of the stage including mikes that picked up some amusing "open mike" comments from the speakers. 3) They have a background music channel that they sometimes forget to turn off when the presentation starts.

However, during one panel session I experienced a first for me... the speaker was referencing a good book "Naked Conversations" by Scoble and Israel... while continuing to listen to the panel, I reached over to my Kindle, found the book online, and purchased / downloaded it.

I enjoyed both conferences but think the better virtual conference goes to Enterprise 2.0 put on by TechWeb.

So what is coming over the horizon? Recently, Ruven Cohen hosted a beta CloudCamp in the Cloud. This was a completely virtual event. A replay is available on UStream

So why is this not just a webinar? Well during the unpanel session the participating audience could ask questions which would be displayed and then people could raise their hands (virtually) to discuss the question raised.

I think that using a better social networking platform for real-time sharing of data / voice / video will give a better result.

There have also been several virtual conferences held in Second Life. The New Media Consortium (NMC) has thousands of members who participate in virtual conferences. The nice thing about a second life conference is the immerse experience... your avatar walks around the conference, talking with speakers and other participants, strolling through an exhibitor floor and even participating in pre/post conference social events. Unlike the remote participation in conferences like the two discussed above, the second life conference gives me the opportunity to actively contribute. After a virtual conference session I can walk up to the presenter and introduce myself and ask a question. What is lacking, is the ultimate realism of being at a real life conference or even watching a presenter over a good video stream. A lot of subtle communications is non verbal through facial expressions and hand gestures. Perhaps someday we will have technology that captures real time video of our faces and renders that onto an avatar.

Given the cost of travel and time away from work I imagine that more and more conferences will be moving to remote access it not complete virtual participation.

BarCamp 2009

This Saturday I attended my first BarCamp. The BarCamp RDU 2009 was hosted by Red Hat at their HQ on the NC State University campus in lovely Raleigh. Approximately 200 people attended and from that audience 36 people were able to present on a diverse set of topics.

Similarly to the CloudCamp I participated in last fall, the BarCamp is run in an "unconference format". At the beginning of the day people that thought they had a good idea would put the name of their topic on a sheet of paper, queue up for the podium, and when your turn came, pitch your concept to everyone for a minute or two. Then you paste your paper into an available room/time slot and after everyone is done pitching the voting begins. People walk up to the wall (see picture) and if they like the concept they put a mark on the paper. This allows the organizers to shuffle presentations between small and large rooms based on interest and if no one voted for your topic you can gracefully cancel.

Here are the topics that were presented this year:

• HTML 5 Discussion
• Power Present in 15 Minutes
• Learn how to Juggle
• Secrets of Effective Nomading
• Recommender Systems: Lessons Learned
• Free: Profit Killer, Inevitable, Necessary, or all of the above
• Static on the Line: How to handle feedback
• Server side Javascript with Dojo
• Palm Pre: Development for noobs
• Potpourri for $500
• Bughouse (a chess variant with two boards and four people)
• Polyphasic Sleep Q&A
• The Intersection of Usability, Accessibility, and SEO
• Building your A Team
• Rapid Return on Investment: Achieve 12 month break even using emerging technologies
• CALEA: Lawful Intercept
• GeekDads
• Soft Appliances
• How to do Social Networking when there is no "Network"
• What's up with OpenSocial
• WTF is Biz Dev
• Intro to jQuery
• Alternate JVM Language overview
• Polka! - Triangle Vintage Dance
• When things go horribly wrong
• Which Languages and Technologies will be around in 10 years?
• Productivity of a Submariner
• Google Wave
• Managing the performance of servers in a large network... on the cheap
• Webkit Debugger
• How Smart Startups Win
• The Small Business Web
• Zombie
• Self Publishing Roundtable
• Query optimization in PostgenSQL
• Twitter Roundtable

As you can see this is not your typical technical conference. For example, I was exposed to the community of Polyphasic Sleepers for the first time at this conference.

So here are the talks that I went to:

Free - presented by Martin Smith

Marty led a discussion on the aspects of marketing concerned with the niche markets (ala The Long Tail) and with new business models where content/services are offered for free to the consumer and revenue is generated via ads or with premium services offered to the free subscribers.

From a long tail perspective we discussed some of the benefits of business that operates in that market:

Distribution of Risk - Palm is betting the business on the Pre. If it does not succeed the company will probably not survive. If instead of a single product, a company was able to offer a large number of products to niche markets the risk would be distributed. One commenter mentioned that some products (like cell phones or pharma) require a large production to offset the development expense.

Marty mentioned the increasing complexity of the Internet and recommended NonZero by Robin Wright as good background on how our society is evolving to deal with increasing complexity.

Someone else in the room said that the Long Tail principle applied to more than commercial products. She thought that ideas were also finding small niche groups of people. And those people tended to be more passionate about the idea and more likely to take action in the small group.

Rapid Return on Investment - John Baker

That’s right; I was able to get the new material that Chris Hanebeck and I have been working on in front of this audience as a beta test of concepts.

The basic premise that Chris and I have is that one can find projects in a company that can achieve break even ROI within twelve months. We use a combination of out-of-the-box thinking, emerging technologies, and discovery of analogies solutions from other industries to achieve the results. To get a copy of the material I presented go to my website. A few people drifted out of the room during the presentation and afterwards a participant mentioned to me that the examples I used in the presentation (RFID used in the Supply Chain) was probably not familiar to the audience of BarCamp. I am planning on developing a version that does focus on emerging Internet technologies and will be ready for next year.

What’s up with OpenSocial - Dave Johnson

This one was definitely more technical. Dave presented the basics of OpenSocial and the progress some companies like LinkedIn, Google, Ning, and Yahoo are making using the standard to share data and gadgets associated with social networking. The official site has a wealth of information. And Dave has his own personal blog where he covers OpenSocial and other efforts like BarCamp.

With as many companies investing in OpenSocial it would seem that current problems (e.g. poor security) will be solved.

What Languages and Technologies will be around in 10 Years? - Jeff Terrell

Jeff is graduating from UNC Chapel Hill and wanted to speculate with the audience on what languages/technologies it might make sense to invest time in learning. For example, will Ruby on Rails be around for a long time?

This lead to a diverse discussion on a wide range of topics:

• The language/technology will depend on the solution being developed. COBOL is still being maintained on mainframes in banks while C is common on embedded systems.
• The "browser" based interface is likely to continue grow in ability to support more and more applications.
• The browser based rendering engine is complimented by the continued penetration of always available high bandwidth wireless networks.
• The current keyboard I/O may be replaced by gestures or by voice recognition.
• Augmented Reality will become more common (see Layar )

Google Wave - Joe Gregorio

Joe demoed the Google Wave using a couple other members of the audience to mutate the wavlets being created. He also showed how Robots worked (very cool implications) and how Gadgets (using a semi-OpenSocial structure) can be dropped into a wave.

Google wants Wave to be a true replacement for email (and I suspect a lot more) and therefore is opening up the control of its future to Open Source.

The audience (this was the most popular session I attended during the day) asked a ton of questions. For example, how will Wave work for the user on an airplane (assuming they are disconnected) who continues to work for four hours mutating the wave they left the ground with. What happens when they land and resync? Joe explained that the Operational Transforms would be processed and that the wave would be left in a correct state for all users. Joe said Google does not promise that the results are meaningful, just consistent. So there will need to be some common sense applied to the approach. One idea I have for that is since most documents being created by a team effort would have division of labor it might make sense to add an optional check in / checkout protocol. So just before leaving on my trip, I check out Chapter 4 in the wave and when somebody else tries to touch it that person is told to wait till the material is checked back in for common use.

Google Wave is really cool.

So my Saturday at BarCamp was well worth the time and if you have not experienced one yourself I encourage you to jump on the web and see if one will be happening in your location soon.

Learning from Mistakes

I attended a brown bag conference call hosted by the Industrial Research Institute on the topic of "Learn from New Product Failures". Our speaker was Jim Hlavacek, one of the authors of the paper by that name published in Research-Technology-Management. They have had experience in reviewing failed projects at manufacturing companies based on techniques used in the medical profession. In many teaching hospitals, when there is an adverse patient outcome to treatment, the clinical team undergoes a Mortality and Morbidity Conference, where the course of treatment is reviewed by an objective team and mistakes identified.

Hlavacek reported experiences at companies like Intuit, Toyota, and 3M where a regular process of review is built into the engineering/marketing culture.

Under Hlavacek's approch the Failed Product Review (FPR) would consist of the following:

Background

Name of the failed venture

Dates project began and was terminated or shelved

New Venture leader and cross-functional team members

Objective and qualified principal investigators

Inputs

Face-to-face interviews with people who were particpants on the project

Face-to-face interviews with OEM and end-use customers who were involved

Face-to-face interviews with distributors/dealers and/or key suppliers

Obtain all e-mails, business plans, documents, trials, and project presentations

Methodology

Develop timelines and milestones of critical events or decisions

Doucment the unfavorable outcomes with data

Develop fishbone diagrams for the project and processes

Develop root-cause analysis of the fishbone diagrams

Recommendations

What went well for the project

What went wrong for the project
Lessions learned and corrective actions

This approach to learning from mistakes reminded me of some similar approaches I am familiar with. Those of you with a military background may have participated in an After Action Review. This is used primarily during training exercises to understand what happened, evaluate everyones performance, and discuss what could be done better. From the USArmy manual comes a similar outline for an AAR:

Introduction and rules

Review of objectives and intent

Training objectives

Commanders mission/intent

OPFOR commander's mission/intent

Relevant doctrine, tactics, techniques, and procedures

Summary of recent events (what happened)

Discussion of key issues

Chronological order of events

Battlefield operating system

Key events/themes/issues

Discussion of optional issues

Soldier/Leader skills

Tasks to sustain/improve

Statistics

Others

Discussion of force protection (safety)

Closing Comments

The final example is from my experience as a Certified Scrum Master. Under SCRUM a small team produces a product release using a succession of short time boxed miniprojects called Sprints. Sprints typically take anywhere from 2-4 weeks for the team to produce a functional version of the product. After every Sprint the team gets together for a Restrospective Meeting. In this meeting the team discusses the following:

What worked well last Sprint that we should continue doing?

The practices that worked well during the previous Sprint should be identified and continued in the coming Sprint.

What didn’t work well last Sprint that we should stop doing?

The team or customers should identify practices that worked against the team during the last Sprint and focus on stopping those things during the next Sprint.

What should we start doing?

The team identifies practices that should be implemented during the coming Sprint that will help them work better together.

Out of this discussion comes a list of actions that the Scrum Master captures and and is responsible for implementing during the next Sprint.

So lets compare/contrast these three approaches:

1. The FPR is conducted when things go wrong. The AAR and Retrospective occur for all outcomes.
2. Both the FPR and AAR require the participation of one or more objective reviewers. The Retrospective depends on the team and, sometimes, invited guests.
3. The FPR and AAR both use detailed analysis to find root causes. The Retrospectice is more adhoc.
4. The AAR assumes that particpants will change behavior based on issues being surfaced, the FPR has a deliverable of lessons learned but no clear followup for change, and the Retrospective has a set of actions with the Scrum Master responsible for seeing they are implemented immediately.

In my opinion, the key to making any of these techniques work is to have a culture of trust surounding the proceedings where everyone understands that people make mistakes and that for the majority of participants mistakes that are surfaced will not be used to punish them. I say majority, because if the same individual is constantly exposed as making repeated mistakes and not correcting behavior then it may result in termination.

Over my career, I have participated in and led a lot of project/product reviews. Almost always, people are defensive and guarded about what happened. To set the right tone in establishing a review program executives should lead by example, being willing to have their actions reviewed and also demonstrating with their subordinates that mistakes they make are not used to influence annual appraisals.

Pampered Pooch 2

I am a sucker for a "59 Minute Scrum" hosted by Bob Galen. I had participated in one last November as documented in this Blog. Last night the IIBA hosted Bob and I was again a member of a six person team producing a brochure for the Pampered Pooch Day Care. I wanted to get another feel for the team dynamics during the sprints and to compare the results with the last session.

Here are my observations:

1. While the aRTP session was comprised mainly of programmers and the IIBA was comprised of business analysts (duh), there was little difference in the results of producing a brochure. I guess if Bob ran a session at the Pet Care Services Association the outcome might be different.

2. Before we started the Day 2 Sprint, Bob pulled the four Scrum Masters aside and told two of them to go back and emphasize the quality and completeness of the brochure, and told the remain two Scrum Masters to tell there teams to push for as much content as possible in the time remaining. The results were telling. The two teams pushing for Quantity delivered 13 and 8 user stories respectively, and the two focused on Quality delivered 6 and 4 user stories. So teams will listen to the direction of the Scrum Master. Ultimately to have a released product both the Quantity and Quality need to be good enough. So is it better to get lots of 60% quality content in early sprints and then tighten it up all at once towards the end of the iteration? OR do you push for 80% quality content and achieve less content per sprint. A real trade off that the team needs to decide based on coupling/cohesion of the user stories. If the stories have few dependencies then push for the higher quality per sprint. With lots of dependencies you need the total content present to debug and refactor.

Enterprise 2.0 Conference

I had some local commitments so I could not fly up to Boston again (See earlier post on Johnny goes to Harvard) and thought I would miss the Enterprise 2.0 Conference Bummer. Little did I know that these people are trying to practice what they preach. I joined the Twitter stream #e2conf and am getting real-time tweets from all over the conference. Also there is a blog where I can read material from most of the presenters and see the opinions from other attendees.

Finally there is a e2TV video stream of the general sessions and vendor demos.

This morning I participated in the Launchpad contest. Four vendors who were finalists got a chance to do a quick demo for the audience and then the audience voted for a winner via SMS.

The four finalists were:

Bantam Networks - they have a enterprise project workspace that allow team mates to communicate, share info and manage relationships.

Brainpark - they also create a social network for projects. The difference seems to be an engine that suggests to the user people who have skills that might help, docs with info that might help, feeds/links with info that might help.

Manymoon - another social network for developers. This one has a good integration with Google apps and with external participants.

YouCalc - this is a different one... an analytics environment that can pull info from a lot of different sources and present graphic analysis. Also it is a product based on "wikinomics". The apps are created by the user community. If you create an app it must be made publicly available for others to use modify. The data that is analyzed remains private.

So the voting took place (I thought Manymoon was really good) and the winner was YouCalc with 53% of the vote.

As I am finishing up this post I am listening to a demo of Lotus Live from IBM. Last night they won the big Buyers Choice Award for best product of the show. This was a vote by attendees.

This type of virtual conference experience still lacks the level of deal making / networking that can happen in a f2f environment but you can't beat the price (FREE) and not having to sit on an airplane and then that ride in from Logan airport (Ugh).

It's Raining, It's Pouring

I attended a Webinar today where IBM discussed their Cloud Computing initiative including their "Cloudburst" offering. David Dworkin of the Tivoli business unit took the audience through

justifications for going to cloud which included a survey conducted last year of companies that had implemented a cloud application. The top three reasons for going to cloud for these companies was:

1. Innovation
2. Time to Profitability
3. Reduced costs

IBM is recommending that companies first move to internal clouds that reside safely inside the corporate firewall but consolidate various departmental applications. They claim such a move will have following benefits:

• Can reduce IT Labor costs by 50%
• Can improve capital utilization by 75%
• Reduce provisioning cycle times from weeks to minutes
• Can reduce end user IT support costs by 40%

In my opinion it seems Amazon EC2 is more SMB start-ups with quick roll out and low up front costs while IBM is aiming at Fortune 500 with large IT budgets under pressure.

During the webinar the host asked the audience (I did not see how many were attending) a couple of survey questions...

Which best describes your organization's level of adoption of Cloud Computing services? None, but not evaluating. 33.8% None, but currently evaluating one or more services. 27.4% Currently getting ready to trial a Cloud Computing service. 9.6% Limited trial adoption of one Cloud Computing service. 14.5% Currently running one or more crucial set of business tasks through the Cloud. 14.5%

David thought this was a little surprising compared with survey results IBM sponsored last year. He speculated that companies may be using clouds without being aware of it.

What are the biggest reasons your organization has yet to migrate any services off to the Cloud? Concerns over security 38.0% Need to "own" and manage the data center 19.0% Regulatory obstacles 12.0% Management does not see the potential for quick ROI 13.0% No skepticism, just looking for the right solution 42.0%

Of course remember that this population had already self selected to having enough of an interest in cloud to invest time in attending the webinar so these answers do not represent the general market.

Web2.0 revisited

On June 4th I presented my Web2.0 briefing to 21 participants representing 17 companies at an event hosted by Matrix Resources. Matrix offers briefings to its customers as a complimentary service and from the comments I heard as people were settling in for my talk it seemed to be a much appreciated program.

During the event I captured some informal statistics from the audience on specific Web2.0 usage patterns.

As a percentage of participants how many:

Use Wikipedia?- 100% as reader. 0% as author.

Author a blog? - 0%

Use Digg? - 10%

Have a LinkedIn Account? - 76%

Have a Facebook Account? - 76%

Have used Craigslist? - 76%

Participate in Second Life? - 0%

Use Twitter? - 24%

Use Ajax to develop apps? - 15%

So what does this mean? Like most statistics with small samples... not much. But I like to ask people and see if any trends are appearing that are different then the official surveys.

Also during the briefing we had a lot of discussion about how companies developing web2.0 apps are realizing revenue. The table below is my analysis of some of the more popular companies I mentioned in the talk.

I have added this table to my Web2.0 presentation page 32. In general I found that the strategy for most of these companies is to give the functionality away for free and rapidly grow a large user base. As the application matures and more users are locked in they obtain revenue streams through a combination of advertising and premium services.

Johnny goes to Harvard

Last weekend I attended the Deep Agile 2009 conference at Harvard University. This two day conference was put on by Agile Bazzaar, an ACM Chapter dedicated to the improvement of all things agile. Approximately 90 attendees participated in the conference and I thought that this was a very well run production. Kudos to the Agile Bazzaar volunteers and especially Nancy Van Schooenderwoert who chaired the program. 

The links above give the highlights from the events and I only want to add my own observations:

Jack Ganssle represented the non-agile development community and gave several presentations on his approach to embedded systems development. He favors object oriented development and follows Bertrand Meyer's Design by Contract process.

James Grenning was one of the original signers of the Agile Manefesto  He says he attended that event for the skiing but I suspect he had more involvement... He is a strong advocate of Test Driven Development, Pair Programming, and SCRUM. He showed all these elements to the audience but in some cases they were the regular versions without a significant embedded twist.

Russell Hill is a development manager at Key Technologies. He is an advocate of Test Driven Development for embedded and thoughout his presentations talked about his experiences at Key building a reusable set of boards for the various products Key manufactures. This system includes hard real-time behavior developed for an FPGA connected to a Motorolla micro processor handeling the UI and control logic. He brought a valuable perspective to the conference on what is achievable for embedded development using agile.

Here are my notes from a panel session:

In practice how can a HW based system be delivered incrementaly?

James - I don't do a lot of HW dev BUT "is it working?" is a good test.

Russel - Our HW is based on FPGA and has some flexibility. But boards would be developed incrementaly.

Jack - There is a lot of religion in agile.... HW tends to be late and broken.... and we don't anticipate that for SW

Do you expect to modify an embedded system on a 2-3 week sprint cycle?

James - would like visable progress without neccessarily being deliverable

Russel  - Key Technology can turn around a HW change very rapidly (2 days!)

When an embedded system is used for a consumer product who is the product owner?

Russel -  We have thousands of customers, Marketing dept, Field Serices represents the interests of the customer. Nancy - any problems with that. Russel - Sometimes. Only in last couple years has marketing been strong participant.

James - this is an organizational problem (of getting participation)

How to reconcile "HW requires long lead times" vs "Agile is incremental" ?

Russel - Board was not available for a couple years. SW was written to spec and was able to deliver 2 weeks after HW release.

Jack - Up front commitment to HW features and to the schedule is different for embedded.

James - At the beginning of an agile project one needs a vision of both HW and SW.

How do you handle scheule constraint.

James - in agile nothing is negotiable until it is late.

Russel - when we are late our internal customers now about it real soon. Visibility to decision makers is important.

Jack - This is not unique to agile. Quality, Schedule, Features pick any two. Jack thinks features should be unconstrained while keeping Quality and Schedule fixed.

Any experience including HW engineers in the SCRUM.

Russel - intersted but not active daily 

James - integrate early and often. Some teams use HW engineers as customer. One client in Finland does overnight board turns.

For an embedded system can a product backlog include User Stories that target either HW or SW? If so, what does a HW User Story look like.

Russel - We have not done that.

James - if you try it write a paper.

Jack - I think it would be very difficult.

I am concerned about the lack of upfront design. What if a User Story requires a big change?

Russel - we just experienced this with a project that required a substantial redesign. The requirement was given a year ago on an eight year project.

In classic agile the key to development is decomposition of the epic story into slices. How do you decompose in embedded?

Russel - Frankly we struggle getting our HW engineers to think agile but it is getting better. We had one example where a story to eliminate a spurious image required both sensor, hw platform, and sw.

What role if any does architecture have in agile?

James - we work on it every day. Every time a sprint is completed there is an architecture that supports SW completed to that date.

Russel - Some of our architecture is harder to change. 

Jack - Architecture and Design in embedded world is a lot less maliable. Up front design is very important.

Nancy - Some companies have a culture of detail which is driven by politics.

My question was the one about HW User Stories.... I am still looking for a way to weave the EE participation into a high tempo devlivery using agile approaches. I will take James up on his suggestion to write a paper when I have it all figured out.