Scrum Unplugged

What is Scrum?

A variation on Sashimi, an "all-at-once" approach to software engineering. Both Scrum and Sashimi are suited best to new product development rather than extended development. Sashimi originated with the Japanese and their experiences with the Waterfall model. They had the same problems with the Waterfall model as everybody else, so they adapted it to suit their own style. Realizing that speed and flexibility are as important as high quality and low cost they reduced the number of phases to four -- requirements, design, prototype, and acceptance -- without removing any activities, which resulted in overlap of the Waterfall phases. Then they made the four phases overlap. (Sashimi is a way of presenting sliced raw fish where each slice rests partially on the slice before it). Other companies took Sashimi one step further, reducing
the phases to one and calling it Scrum. (A scrum is a team pack in Rugby, everybody in the pack acts together with everyone else to move the ball down the field).

Applying Scrum

For each Waterfall phase there are a pool of experienced people available, form a team by selecting one person from each pool. Call a team meeting and tell them that they have been selected to do an important project. Describe the project, include how long it's estimated to take, how much it is estimated to cost, how it is expected to perform, etc. Now tell them that their job is to do it in half the time, with half the cost, twice the performance, etc. Tell them how it's done is up to them and explain that your job is to support them with resources. Now leave.
Stand by, give advice if it's requested, and wait. Don't be surprised if a team member thinks the whole thing is insane and leaves. You'll get regular reports, but mostly you'll just wait. At somewhere around the expected time, the team will produce the system with the expected performance and cost.

How does Scrum work?

The first thing that happens is the initial leader will become primarily a reporter. The leadership role will bounce around within the team based on the task at hand. Soon QA developers will be learning how requirements are done and will be actively contributing, and requirements people will be seeing things from a QA point of view. As work is done in each of the phases, all the team learns and contributes, no work is done alone, the team is behind everything. From the initial meeting, the finished product is being developed. Someone can be writing code, working on functional specifications, and designing during the same day, i.e. "all-at-once". Don't be surprised if the team cleans the slate numerous times, many new ways will be picked up and many old ways discarded. The team will become autonomous, and will tend to transcend the initial goals, striving for excellence. The people on the team will become committed to accomplish the goal and some members may experience emotional pain when the project is completed.

Why does Scrum Work?

The basic premise is that if you are committed to the team and the project, and if your boss really trusts you, then you can spend time being productive instead of justifying your work. This reduces the need for meetings, reporting and authorization. There is control, but it is subtle and mostly indirect. It is exercised by selecting the right people, creating an open work environment, encouraging feedback, establishing an evaluation and reward program based on group performance, managing the tendency to go off in different directions early on, and tolerating mistakes. Every person on the team starts with an understanding of the problem, associates it with a range of solutions experienced and studied, then using skill, intelligence, and experience, will narrow the range to one or a few options.

Keep in mind that it can be difficult to give up the control that it takes to support the Scrum methodology. The approach is risky, there is no guarantee that the team will not run up against real limits, which could kill the project. The disappointment of the failure could adversely affect the team members because of the high levels of personal commitment involved. Each person on the team is required to understand all of the problem and all of the steps in developing a system to solve it, this may limit the size of the system developed using the methodology.

Planning Poker (Time Estimations)

Time estimating using planning poker


Estimating is a team activity - every team member is usually involved in estimating every story. Why?
  • As per Agile at the time of planning, we normally don’t know exactly who will be implementing which parts of which stories. Stories normally involve several people and different types of expertise (user interface design, coding, testing, etc).
  • In order to provide an estimate, a team member needs some kind of understanding of what the story is about. By asking everybody to estimate each item, we make sure that each team member understands what each item is about. This increases the likelihood that team members will help each other out during the sprint. This also increases the likelihood that important questions about the story come up early.
  • When asking everybody to estimate a story we often discover discrepancies where two different team members have wildly different estimates for the same story. That kind of stuff is better to discover and discuss earlier than later.

If you ask the team to provide an estimate, normally the person who understands the story best will be the first one to blurt one out. Unfortunately, this will strongly affect everybody else’s estimates.

There is an excellent technique to avoid this – it is called planning poker (coined by Mike Cohn I think).


Each team member gets a deck of 13 cards as shown above. Whenever a story is to be estimated, each team member selects a card that represents his time estimate (in story points) and places it face-down on the table. When all team members are done the cards on the table are revealed simultaneously. That way each team member is forced to think for himself rather than lean on somebody else’s estimate.

If there is a large discrepancy between two estimates, the team discusses the differences and tries to build a common picture of what work is involved in the story. They might do some kind of task breakdown. Afterwards, the team estimates again. This loop is repeated until the time estimates converge, i.e. all estimates are approximately the same for that story.

It is important to remind team members that they are to estimate the total amount of work involved in the story. Not just “their” part of the work.The tester should not just estimate the amount of testing work.

Note that the number sequence is non-linear. For example there is nothing between 40 and 100. Why?

This is to avoid a false sense of accuracy for large time estimates. If a story is estimated at approximately 20 story points, it is not relevant to discuss whether it should be 20 or 18 or 21. All we know is that it is a large story and that it is hard to estimate. So 20 is our ballpark guess.Want more detailed estimates? Split the story into smaller stories and estimate the smaller stories instead!

And NO, you can’t cheat by combining a 5 and a 2 to make a 7. You have to choose either 5 or 8, there is no 7.

Some special cards to note:



0 = “this story is already done” or “this story is pretty much nothing, just a few minutes of work”.

? = “I have absolutely no idea at all. None.”

å = “I’m too tired to think. Let’s take a short break.”

Extreme Programming Core Practices

The 12 “XP Xtudes” (Xtude is XP means ‘Attitude’) of Extreme Programming (XP) grouped into four categories

1. Fine Scale feedback

XP thrives on providing feedback at smaller intervals with higher frequency. This allows controlling deviation at the right time, since in software or any othe industry for that matter, once deviation starts happening it is dificult to control at the later stages.

  • Test Driven Development via Programmer Tests (Unit Tests) and Customer Tests (Acceptance Tests/Automation Tests)
  • Planning Game (Definition Iteration Objectives and playfield etc)
  • Whole Team (Onsite Customer + Programmer + Quality Team + Customer Team + Scrum Master + Product Owner)
  • Pair Programming ( two engineers participate in one development effort at one workstation)

2. Continuous Process rather than Batch

3. Shared understanding

4. Programmer welfare

YAGNI

YAGNI perfectly sums up XP (Xtreme Programming). Here is why and how.

"You Arent Gonna Need It" (often abbreviated YAGNI) is an Extreme Programming (XP)practice which states:

"Always implement things when you actually need them, never when you just foresee that you need them."
Even if you're totally, totally, totally sure that you'll need a feature later on, don't implement it now. Usually, it'll turn out either
  1. You don't need it after all, or
  2. What you actually need is quite different from what you foresaw needing earlier.

This doesn't mean you should avoid building flexibility into your code. It means you shouldn't overengineer something based on what you think you might need later on. There are two main reasons to practise YagNi:

  • You save time, because you avoid writing code that you turn out not to need.
  • Your code is better, because you avoid polluting it with 'guesses' that turn out to be more or less wrong but stick around anyway.

A scenario that explains the practices:




You're working on some class. You have just added some functionality that you need. You realize that you are going to need some other bit of functionality. If you don't need it now, don't add it now. Why not?

"OK, Mohan, why do you want to add it now?"
"Well, Rahul, it will save time later."
But unless your universe is very different from mine, you can't 'save' time by doing the work now, unless it will take more time to do it later than it will to do now. So you are saying:

"We will be able to do less work overall, at the cost of doing more work now."
But unless your project is very different from mine, you already have too much to do right now. Doing more now is a very bad thing when you already have too much to do.

And unless your mind is very different from mine, there is a high chance that you won't need it after all, or that you'll need to rewrite or fix it once you do need it. If either of these happens, not only will you waste time overall, you will prevent yourself from adding things that you do need right now.

"But Rahul, I know how to do it right now, and later I might not."
"So, Mohan, you're telling me that this class you're writing is so complex that even you won't be able to maintain it?"
Keep it simple. If you need it, you can put it in later. If you don't need it, you won't have to do the work at all. Take that day off.

YAGNI in the context of the other Extreme Programming practices

You have a Release Plan: each User Story has been assigned to an Iteration where it will be done. Under the current Iteration Plan, you are working on an Engineering Task that you signed up for, in support of one of the Iteration's User Stories. As always, you have signed up for as much Ideal Programming Time as your Load Factor indicates you can accomplish.

You are evolving the system to have the new functionality required by the User Story, defined in the Engineering Task. You add capability to any class we need to, directly growing from the requirement. If you find yourself writing duplicate code, you refactor to eliminate it, even (perhaps) adding an abstract class, or making a subclass, etc. You and your co-programmers always keep the code clean.

You're building a class, and suddenly you get an idea for a feature you could add to it. You don't need it right now, but "Someday we're gonna need ...", you say to yourself.

Keep in mind that you are employing other Extreme Programming practices that allow you to deal with the future when it happens. Collective Code Ownership allows you to change anybody else's code to give it the functionality you want. Refactor Mercilessly and Once And Only Once make it easier to understand the best way to add your functionality. Unit Tests help ensure that your added functionality won't break any past functionality. So if you do need to implement this feature in the future, it probably won't be much harder than it would be to implement now.

At this moment, you have a choice: continue working on what you signed up to do, or begin working on something you didn't sign up to do, and that isn't needed in this Iteration.

Therefore, tell yourself YAGNI. Set aside your thoughts and fears about tomorrow and get back to work on today. Without a clear use for the feature, you don't know enough about what is really needed. Spending time on it is speculative at best.

One Responsibility Rule

From Bertrand Meyer's “Object Oriented Software Construction”, there was the statement:

A class has a single responsibility: it does it all, does it well, and does it only.
Classes, interfaces, functions, etc. all become large and bloated when they're trying to do too many things.

When a function has too many responsibilities, it becomes buried deep in "Special Formatting", which has a "Code Smell".
To avoid bloat and confusion, and ensure that code is truly simple (not just quick to hack out) we have to practice "Code Normalization", which seems to be a variation on “Once And Only Once” and also “Do The Simplest Thing That Could Possibly Work”.
This is part of “Responsibility Driven Design”.

Special Formatting :- Sometimes in a method you wind up wanting special formatting. We take it as a sign, instead, that you should rewrite the method. It could refer to the situations where you would want to indent the statements to focus / highlight some important logic.
Code Smell :- A code smell is a hint that something has gone wrong somewhere in your code.
Code Normalization :- The attributes and operations of a class should depend on the responsibility, the whole responsibility, and nothing but the responsibility. If this can't be made to work, the class has lower cohesion, and might be a candidate for re factoring.
Once and Only Once :- One of the main goals (if not the main goal) when Re factoring code is that each and every declaration of behaviour should appear “Once And Only Once”.
Do The Simplest Thing That Could Possibly Work :- "All that is complex is not useful. All that is useful is simple." – Mikhail Kalashnikov
Responsibility Driven Design :- Design based on the principle that some class has to be responsible for each task that the system will carry out. The responsible class may collaborate with other classes to carry out its task. (It's probably a good idea if each responsibility occurs Once And Only Once.)
Lazy Programmer

This idiom is related to the Albert Einstein's Principle “A scientific theory should be as simple as possible, but no simpler”.

It can be worded as "Do as little work as possible to get the task completed, but no less."
So, complete the task by trying to write the minimum amount of code that satisfies the requirements and passes all the tests. That means not falling prey to the “Not Invented Here”* syndrome and instead embracing the work of others to lessen the amount of work you have to do. It also means embracing standards so that there is more potential to leverage others work.


In a different vein, this can mean a programmer that practices “Copy And Paste Programming” and “Someone Else’s Example” programming. The programmer does not take the time to understand the code, but simply forces pieces together with minimum effort.

Anything more than cursory testing can be ruled out altogether.

* Any code Not Invented Here is not as good as code that I (or we) wrote
How to best use product backlogs
Product backlog is normally the starting point, it is where AGILE's heart lie. It is the first step in starting an AGILE based software development. It is nothing but a sorted list of requirements/feature requests/user stories or what ever term you use for high level tasks based on importance of the items in it. The best part is that the items are described in Customer's language which he understand basically the one which reflect the business benefit for him. Lets call these items user stories or backlog items.

Fields associated with each story

SID – Story Identification Number, a unique identification may be you can use an auto-incremented number. This is to keep track of stories even if we change their name.
Name – It is a short descriptive name of the story. Length should not be longer than 4-12 words. e.g. “Generate the employee attendance report”. It should be small, crisp, distinguishable and communicative so that stake holders knows what we want system to do. Stake holders include developers, testers, product owner.
Importance – It is the importance rating given by the product owner. There are many theories circulate around what should we use as the rating number. some suggest using P1 to P5, other says 1 to 10. I have read an interesting article which says we can use some large numbers like 21 or 135 etc, where higher the number more it has the importance. The motive behind this is to have unique importance number between stories. Second logic which is supporting it says that lets say you rated one story as P1 or 1 what will you do if a new story comes up between the iteration which is higher in importance that this one. Would you rate it as -1 or 0.5? Secondly with using high numbers we should leave some gap between importance numbers like story 1 may have importance rating of 10 and story 2 may have 9. This way is a new story comes up which is high in importance than story 1 but less in importance than story 2 we can accommodate it in between without using making sheet look ugly by making it as 2.5 or fractions. Nice thought :).
estimate – the first assessment of how much work is needed to complete this story. The unit of storage can be of your choice, normally we use a terminology called story points. The story points are usually corresponds roughly to “ideal man-days”.

  • You can ask the team with a optimum number of people usually 2, in how many days you can finish a complete demonstrable & releasable version of this story. If the answer comes like "with 2 guys we will take approximately 3 working days" then the initial estimate is 3 X 2 = 6 story points.
  • The concentration is not on getting every things absolutely right in the first place but the concentration is to check the relative estimates that means that a 4-point story takes about double the effort as compared to a 2-point story.
How to demo – This is basically a high level description of how this story will be demonstrated at the Sprint Demo. This is nothing but a simple test spec. “If you do this, and then do that, then you can expect this to happen”.

(Note: If you practice TDD (test case-driven development) this description can be used as pseudo-code for your acceptance test code. )

Notes – any other info, clarifications, references to other sources of info, etc. Normally very brief.

PRODUCT BACKLOG (example)



SIDNameImp.Story PointsHow To DemoNotes
1001Allocate Employee to Project407Log In, Open Employee Detail Page, Change Employee Department from Bench to "Project X", Check the Employee Status in the Employee Detail PageNeed to create the class hierarchy
1002Check Employee Allocation History253Log In, Go to "My Detail" page, click on Project HistoryNeed to modify the css for this page.

As the author of this sample excel sheet said and I agree that these are the basic 6 fields we need at the end of the day. You are free to add up whatever fields you want but a easy going product can survive in these fields. I myself is fan of using excel as Product sheet, it gives you lot more freedom and great control. You can allow product sheet to be in control of the "Product Owner" or you can feel free to share in on a machine with all team members, allowing them to add new stories as and when they arrive but always remember that only "Product Owner" has the right to set importance or write estimates.

HOW WE DO PRODUCT "BACKLOGS"



The last thing you want to do is to place the Product Sheet in the version repository, a good idea would be to rather place it on the shared drive. This way you can allow multiple people to access the same without causing locks, merge conflicts etc.

Additional story fields



Sometimes you may want to use few more additional fields in the product backlog based on your project's priorities, I will try to list few which are pretty common.

Category– It is a rough categorization of the story, for example “UI” or “performance”. This way the product owner can easily filter out all “UI” items and change their priority , etc.
Components - Best way is to realize it by “check boxes” in the Excel document, for example “client, middle ware, server”. Team / Product Owner has to identify which technical components will be required in completing this story. This will be typically useful when you have more than one Scrum teams.
Requester – Here as part of reporting, product owner may like to keep track who has requested for the feature to keep him updated on the progress.

Bug tracking ID – It will be useful to keep track of any direct relation between a story and reported bug(s). Specially if you have a separate bug tracking system, like we have Bugzilla.

How we retain the product backlog focus at a "business level "



It is worse to have a Product Owner coming from Technical Background, but in our case we couldn't help it. We didn't found anyone else suitable. The nightmares it caused initially was adding stories like this one "Add composite primary key to the crystal_soak tables”. Why does he want this? The real underlying goal was definitely something like “to disallow user from adding same soak twice for a particular crystal”. It may turn out that there was no primary key since crystal can be associated to more than one soak and thus users were sometimes by mistake associating the same soak more than once. It was something completely different. The team is normally better suited to figure out how to solve something, so the product owner should focus on business goals.
When you see technically oriented stories like this, you should ask the product owner a series of “but why” questions until we find the underlying goal. Then we rephrase the story in terms of the underlying goal (“disallow users to associate duplicate soak to same crystal”). The original technical description ends up in the note column (“Creating a composite primary key might solve this problem”).

Once Upon a Time....

A Chicken and a Pig lived on a farm। The farmer was very good to them and they both wanted to do something good for him।
One day the chicken approached the pig and said, "I have a great idea for something we can do for the farmer! Would you like to help?"
The pig, quite intrigued by this, said, "of course! What is it that you propose?"
The chicken knew how much the farmer enjoyed a good healthy breakfast। He also knew how little time the farmer had to make a good breakfast. "I think the farmer would be very happy if we made him breakfast."

The pig thought about this. While not as close to the farmer, he too knew of the farmer's love for a good breakfast. "I'd be happy to help you make breakfast for the farmer! What do you suggest we make?"
The chicken, understanding that he had little else to offer suggested, "I could provide some eggs."
The pig knew the farmer might want more, "That's a fine start. What else should we make?"
The chicken looked around...scratched his head...then said, "ham? The farmer loves ham and eggs!"
The pig, very mindful of what this implied, said, "that's fine, but while you're making a contribution I'm making a real commitment!"
Chickens and PigsOn Agile projects the term Pig has come to describe all the developers, designers and testers who commit to the actual work। The term Chicken is applied to everyone else who make intellectual contributions but do not commit to any work।

Note: Images are borrowed from other sites*