November 2006 - Posts - Willy-Peter Schaub's Cave of Chamomile Simplicity

November 2006 - Posts

Let us have a poker tournament in our next Scrum planning session…

Core of the agile scrum methodology is the empowerment of the “team” and this empowerment is never more important than during the scrum planning meeting. We do not have dictatorship in scrum driven projects, i.e. a military general style project manager dictating who does what, when and more importantly how. However, we also do not have Swiss democracy in which every team member must cast their vote before the project can advance to the next beachhead. It is important, and often difficult, to select a core set of team members who are able to estimate effort and most importantly represent the “team”.

 

So what happens during these planning sessions?

 

An average duration of such a planning meeting is 8 hours, broken into a 4 hour product backlog selection and a 4-hour sprint preparation meeting. In the first half the product backlog items, relevant for the next 14-day sprint are selected, prioritized and estimated by the team using the Poker cards … more on these cards in a minute.

 

The second half of the planning meeting is dedicated to selecting, prioritizing and estimating sprint backlog items for the next sprint, again using Poker cards.

 

Why play poker, the Project Manager is asking in the back row?

 

Using the card values Ace=1, 2-10, Jack=11, Queen=12 and King=13, each team member in the planning session gets a set of poker cards.

 

Each item is discussed and then everyone get’s an opportunity to raise the card that is the most accurate estimation in terms of humanoid-days. At first there will be a wide variation of estimates, which requires a few more rounds of discussions and “poker play” until the team feels comfortable of the average estimate.

 

So what are the rules of the game?

 

The most important facet of Scrum we have learnt is that it is not a prescribing methodology and no-one should ever convince you otherwise. When creating the project and the project team you should have a ceasefire meeting to discuss and agree on the rules of engagement for the specific instance of a project, which includes agreeing on the power game rules. The scrum master must blend into the team and adopt the team’s understanding of scrum and modus-operandi into the scrum methodology for that project.

 

Scrum is supposed to be agile…

Posted by willy with no comments
Filed under: ,

Considering TFS Proxy Server? If yes, check the licensing.

The Team Foundation Server (TFS) licensing whitepaper on http://www.microsoft.com/downloads/details.aspx?familyid=1FA86E00-F0A3-4290-9DA9-6E0378A3A3C5&displaylang=en defines the licensing for the various bits and pieces of TFS and is a worthwile read. Note that TFS Proxy requires a full license is licensed on per-device basis.

Posted by willy with no comments
Filed under:

OmShanti Scrum … looking for input.

We are in the process of designing a new version of the Scrum Methodology process template for Team Foundation Server and urgently need your input on what “would be cool”, “what is essential” and “what is hhhmmm in current templates” so that we can create a community process template that will add as much value as it is creating excitement at the moment.

 

Visit us on http://www.codeplex.com/Wiki/View.aspx?ProjectName=VSTSScrum and keep that input flowing.

Posted by willy with no comments

http://www.drp.co.za website emigrating to the US.

Please note that we are in the process of moving our www.drp.co.za website to the US, where we are going to be able to offer better performance.

We have been receiving numerous incident reports that the download of the quick reference posters and snippets are failing due to timeouts, which is really attributed to the limited bandwidth we currently have available.

We will, however, probably have ad-hoc downtimes of the site while we are transferring the nuts and bolts and wish to apologies for any inconvenience caused.

We will advise when the site is running in its new home.

Posted by willy with no comments
Filed under: ,

.NET 3 RTM Installation ... hello gremlins and caution early adopters

Post take 2 ... sorry I had to fix some of the sequencing and layout, which I failed to spot last night. Appologies.

 I have tried to install the .NET3 SDK RTM product on a number of our R&D VMs, which had previous versions, bits and pieces of .NET 3 installed. Even running the uninstall and following the readme instructions lead to installation implosions with "weird" errors. In utter despair and with the support from our support team, I installed the .NET 3 Runtime and .NET 3 SDK successfully onto a naked Windows XP machine, after which I installed Team Suite and the other .NET 3 pieces.

A - The journey that was troublesome and which I have not resolved ... mainly due to lack of time needed for further investigations:
1. Uninstall .NET3 BETA pieces from development machine hosting Visual Studio 2005 Team Suite, using the Uninstall tools.
    http://www.microsoft.com/downloads/details.aspx?FamilyId=AAE7FC63-D405-4E13-909F-E85AA9E66146
2. Install .NET3 Runtime
3. Install .NET 3 SDK ... "implosion".

B - The journey that was troublesome and which I have not resolved ... mainly due to lack of energy:
1. Take development machine hosting Visual Studio 2005 Team Suite, which has never seen any .NET3 BETA pieces
2. Install .NET3 Runtime
3. Install .NET 3 SDK ... "implosion".

C - The journey that lead to Rome:
1. Create new XP VM image.
2. Install .NET3 Runtime
3. Install .NET3 SDK
4. Install Visual Studio 2005 Team Suite
5. Install other .NET3  pieces, i.e. VS extensions

I hope that the release version will cater for the A and B environments, because few of our developers will be able to reinstall their development machines and start from scratch. Also it is not logical to install the SDK before the IDE ... maybe I am just sensitive at the moment.

Posted by willy with 1 comment(s)
Filed under: , ,

Book 3 - Software Engineers on their way to Pluto (Snippets 1, 2)

For years ... or rather decades ... I have been interested in the space program and fascinated at how the space agency manages to communicate with Voyager-2 which is now "roughly" a mere 65AUs from Earth (that's 9,548,500,000km), how the space agency manages to patch the software on the Cassini probe which is even closer at 8.2AUs (i.e. 1,204,580,000km) and direct the probe through the rings of Saturn. It makes you wonder how the software industry back on Earth could be designing, developing and deploying solutions ... when we are typically a mere few meters of walking distance away from the solution.

 

This is where my collegue Geoff and I decided to embark on the exciting journey of book 3. We will start publishing snippets of the book, which is expected to go into final editing during December, to give you an idea of what we are trying to achieve and an insight into the exciting content.

 

Snippet # 1 from Defining the Context section: "What we can learn from the space agencies"


Although Pioneer is the furthest man-made object in space at a proud 91.865 Aus, Voyager 1 is the furthest operational man-made object in space at 81AUs … that’s 12,116.6 million kilometres from Earth … and is expected to continue transmission of data until 2020 … that’s 42 years of operational time. The average IT solution age is a mere 2 years, with continuous handholding.


 

The Huygens probe successfully landed on the Titan moon on Jan. 14, 2005, 7 years after launch, 4,500 million kilometres from Earth. The average IT solution is implemented on-site, with 75% of solutions failing or being de-scoped in terms of features or postponed before eventually seeing a “happy” business user.

 

The success stories of the space missions, the ability of the space probe engineers to design, develop and deploy solutions at vast distances, without the chance to stroll over to a computer and apply a quick patch, is in our eyes a fascinating achievement and one that the information technology software engineers should take note of, analyse and learn.

 

This book takes a critical look and investigates four solution teams and four vastly different approaches of solution life cycle that we have been exposed to during our 23 years in the information technology industry. To make it slightly more interesting we have transposed the four teams and their ways of operation to four teams of software engineers who are receiving the instruction to build and travel to Pluto and back, using their own space ship. We often make the statement that we do not believe that any software engineer would ever consider crossing a suspension bridge across a 500m gorge, built the way the industry builds the majority of information technology solutions.

 

There is no doubt that many projects are complex, based on unclear specifications, unknown and changing requirements, with perplexing human interactions and rapidly evolving and changing technology. However, it seems strange that the space agency is capable of achieving miracles, while the information technology industry normally ends each solution life cycle with bug storms, long hours, stressed team members and compromises on quality, release schedule and/or functionality to meet artificial milestones.

 

We believe that with effective management, clear specifications, open communication, usage of technology as a tool and not as a complexity challenge, keep-it-simple designs and continuous inspection, analysis and reporting, we can join the success stories of some of the previously mentioned space probes. We can avoid the high failure rate of information technology projects and we can finally deliver the return on investment that business deserves.

 

Read the remainder of this book like a story based on four enthusiastic solution teams, referred to as T1, T2, T3 and T4, and come to your own conclusion as to which approach you believe is the correct approach. Although we are intentionally not making any reference to life cycle tools or methodologies you should recognise some of the methodologies used by the teams and recognise tools and technologies. As mentioned previously each team has been transposed to this hypothetical space program from the information technology age during the past 23 years. Some teams may even recognise themselves and if that is the case, please realise that this book has been written to critically look at the way we work, not at the good, the bad and/or the evil of individual project teams. There are many, many more projects and environments, but due to commercial constraints, we had to pick four main candidates.

 

To the space agencies and true rocket scientists we extend an apology for probably breaking a gazillion space and technology rules to achieve the near impossible. We are merely trying to use a more spectacular analogy to highlight the challenges of the information technology projects of yesterday, today and probably tomorrow.

Let us return to the year 2004 and start at the beginning … enjoy the journey.

 

Snippet #2 from Mission Leadup section: "Moment of Truth"


Morpheus tapped on the open door of Thor, manager of Quality Assurance. He asked, "Got a moment?"


Thor was obviously deeply involved in something, but he looked up and saw the strain on his face said, "Of course. What's on your mind?"

 

Silently Morpheus went in and fell into the visitor's chair and his tension increased visibly. "I have a meeting of all the trainees chosen for the Pluto project. This is the decisive moment. After today there will be no turning back."

"Morpheus, my friend, this whole thing has been your dream. You sold it to our director's here at CC&E, they sold it to Space Agency, who then sold it to the President and the budget committees, and now you need to sell it once again. It may be a moment of truth, but it's more your moment of triumph."


"Yes, yes, but I need for these 'software engineers' to look at the reality from the outset. I will ask them to lay it all on the line. It's the old bacon and eggs thing; the pig is committed, but the chicken is just involved."

 

"And you want them to realise that they are the pigs?"

 

"Yes… No…"

 

"No not pigs. They are heroes. They had all he odds spelt out to them at their interviews, right?"

 

"Yes, but …"

 

"No buts. You get out there and sell your heart out. Of course, I know you want them to have a complete picture, you will give it to them and at the end of it, and they will all willingly give you their best, no matter the cost … or the risk. We are all counting on you."

 

Morpheus stood and Thor followed suit and moved out from behind his desk. He grabbed Morpheus's hand and they shook hands with an exchange of friendship and support greater than any words could have said.

 

... watch this space for more snippets.

Posted by willy with 2 comment(s)
Filed under: ,

TFS and Software Factories, by Marcel de Vries

Marcel de Vries, one of our TFS/VSTS MVPs, has written an excellent article "Measuring Success with Software Factories and Visual Studio Team System" looking at software factories and TFS ... have a peek at it on http://msdn2.microsoft.com/en-us/library/aa925157.aspx.
Posted by willy with no comments
Filed under:

MSCert: Analyzing Requirements and Defining .NET Solutions Architectures

Together with a bunch of analysts and developers we are preparing for the MSCert Analyzing Requirements and Defining .NET Solutions Architectures exam.

 

Our findings to date include …

 

What seems important?

§        Although poorly covered in the MSPress book, ORM diagrams appear important.

§        If you are after solid coding, consider enterprise templates.

§        If you are after solid business logic, consider a comprehensive test plan,

§        Denormalisation for replacement of runtime computation is OK.

§        Child implies many and parent 1 in terms of relationships.

 

What is controversial and seems to be the Achilles heel of this exam:

§        Logical business entity recognition, for example we believe that in the Employment Services case study by Transcender that ESOfficer is a valid entity.

§        Physical database design, the allocation of attributes (especially with price and quantity) can result in a number of valid data schemas … yet only one will be correct in the exam.

 

Has anyone written the exam? If yes, are we on track with our preparations and has anyone found that the data modeling answers presented by the exam and Transcender are controversial, i.e. there is more than one feasible answer, yet only one counts in the exam?

Posted by willy with 2 comment(s)
Filed under:

Recovering Team Foundation Server (DR Exercise)

JP is busy sweating after an attempted recovery of our TFS server in a simulated disaster recovery. It appears that the recovery of team project sites is not as trivial as it appears … we will keep you posted on our success/challenges of recovering using the latest and greatest administration guide.

 

In the interim JP is grinding along …

Posted by willy with no comments
Filed under:

Book 2 goes public and book 3 gets a mention ...

Just spotted http://www.itweb.co.za/office/bbd/0610240808.htm which highlights the release of our second book "Interoperebility for the Connoisseur" and a mention of book 3.

 Geoff and I are having immense fun with book 3 and we believe that anyone with a slight interest in space and/or scifi will really enjoy the book "Software Engineers on their way to Pluto>

Posted by willy with no comments
Filed under: ,

TFS and WCF on the move ...

I am refocusing on the latest bits and pieces of .NET3, in particular WCF, and I really like what I see. The technology is really coming together and together with tools such as MEXExplorer from Juval and the LiveTrace utilty from the gurus, i.e. Craig, the communication and interoperability toolbox will be an amazing arsenal technologies. We will be covering both WF and WCF in our "Thor" readiness initiative ... watch www.drp.co.za for details.

In parallel my colleagues have taken my concerns regarding TFS recovery and a move to a new server to heart and are planning a simulated disaster recovery .... i.e. switch off production server and recover it on another server. 

We will be reporting on both our WCF findings and the TFS disaster recovery as soon as the "dust settles".

Posted by willy with no comments
Filed under: ,