 three minutes max cap beyond three minutes will drag you off the stage screaming and yelling all right that's the original way lighting talks were created at the python conference and we're going to do it uh the original way so given that context who wants to come first you don't have a choice thank you so i'm going to be a timer and uh people could sit in these chairs to line up so we know how many people are coming next please those are the line-up chairs and i'm going to do the quick timer it's three minutes and uh it would be great if you can quickly introduce yourself and the topic and we will get started and that part is not coming up no that's not coming okay summa here infosys i'm an agile trainer and coach um so today uh my uh topic of discussion is how to help people uh learn planning and estimation in agile using flower points and i was kind of discussing with richard and savita about it so that's the idea of where it comes from okay um so planning and estimation is something that we think of in user story i mean story points and that's kind of a difference mindset so what i do as part of my trainings is uh take people through a journey of how is the six layers of agile planning within that release management and actually take them away from it i give them a real life situation and say okay the theme for the day of planning and estimation is probably beautifying this room or beautifying this entire hotel okay and then from there i derive a few uh backlog items one of them i put it purposefully as creating a flower garland a garland of flowers paper flowers okay and then i move that into their release goal okay so now they have a release goal then we do a hackathon actually and i help them actually make up our paper flower so they actually get a hands-on with a paper flower they know and then we do a sprint when they do a sprint they realize how many paper flowers they're able to do in one single sprint and that's where they derive flower point so they start relating to the whole concept of doing flower point which is like your story point and from there they derive what is their velocity and from there they derive how they would probably how much time they would take to do the entire garland which is a release goal so starting from what is your flower point to your um what is your uh velocity to you what is your release goal sir hi i'm puja vandile i'm from persistent systems my in fact it's not a lightning talk it's more of a debate we know that project manager is not needed in agile project or a scrum project uh we have also started thinking about do we need a dedicated scrum master also we have you know reached your situation where the scrum masters are asking us you know puja i don't have any work do i really need to be on the project maybe that person was needed when the project was new when the processes were needed to be set up now once the team has settled down there is a rhythm which is there they're producing a good velocity everything is there scrum master actually doesn't have any work for the all those 10 days or 15 days at least for those eight hours so we are in a situation where do you really need a dedicated scrum master you know that's how it starts that you shouldn't have part time scrum master but full time scrum masters but i don't have work from my scrum masters so maybe it will really help if you tell me what are the things that the scrum masters can do in i'm not talking about the meeting setups or the matrix dashboard maybe those are the things that are initially needed maybe put things in jira derive those metrics dashboard that's all you know but after that from the second day until the 14th day if it's a three-week sprint actually all technical things are decided by the teams they you know handle those impediments dependencies requirement dependencies are handled by the bs even there he is not needed so what he or she is supposed to do you know estimation again there is not much that person is contributing because these the ba and you know technical people they are deciding what should be the story points and all those things so we are we are in a situation where we are trying to find out work for the scrum masters no so you have both the jrs right either you make somebody from the team so maybe when the project starts you have a dedicated scrum master who will set up things but as the yeah but both sites are there right you can have a dedicated scrum master you can have somebody from the team playing both the roles you know either as a developer tester and the second part is the scrum master also but that's what the idea we started with and we are kind of reaching to a situation where we say that you don't need a dedicated scrum master you don't need a full-time scrum master maybe somebody from the team can do that yeah yeah so we are but that's what the situation of our scrum masters is you know they are getting into okay yeah no believe me that's okay that's all we are into that situation they they're not adding too much of a value you know even the teams they are deciding you know in terms of process improvement they know if the build is failing you know the configuration management person is there or that developer is there why the build has failed that scrum master is not adding any value why the build failed it is the developer and the you know the configuration management that's what i'm saying those 13 days not much work okay thanks okay um hi everyone my name is yashasri and i'm from tcs data consultancy services working as an architect or come portfolio lead come developer of in scrum project since 2008 now so i wanted to share i wanted to share story of a campaign that we did in in our teams now we started in 2008 and we started learning in the first couple of years we are more of learning and things like that and then we found like even though teams were doing individual retrospectives and trying to improve there were overall there was a stagnation kind of you know that enthusiasm and energy about practicing agile and doing something more and giving value had subsided so we did a campaign which was known as reduce waste campaign and it was basically an idea storm where we encouraged teams to give their ideas as to you know how can we reduce the waste in the current process and it actually went very well you know we did a lot of campaigning and some of the folks give ideas that we actually took ahead and implemented so some of the ideas that we implemented was one one challenge so earlier we were doing a two week sprint and a three month release and from there we actually moved on to one week sprint and one month release it took time but we have got used to it and that's the norm for the new teams as well now the second thing that came up was a lot of processes in terms of deployments i mean because this has been an enterprise there were a lot of processes to push things across environments you know a lot of security concerns infrastructure related concerns and so on so people you know people loved the way Hiroko works right everybody in the who's family with ruby rails would know what is Hiroko and it's just a one command deployment and people wanted to go there so that was another idea which we actually went ahead and tried to implement so it actually took us a lot of time this was through a DevOps initiative that we did so we actually had to talk to a lot of people you know try and understand the enterprise level constraints what is it that the security team wants what is it that you know the infrastructure team wants and so on and what we felt is maybe we can go ahead and build a product ourselves or custom application to be frank not to market to the people that can actually take in all of these enterprise level constraints and give the developers the flexibility like Hiroko does and we actually were successful in so we have created a custom app which is deployed which is actually a one click deployment across environments so basically as a takeaway what I wanted to share was that any teams who start adopting agile will become you know stagnant at some time and may become standard stagnant at some time but then we shouldn't stop and the whole spirit and principle behind adopting agile is continuous improvement so we should definitely go ahead and do that and it will keep that energy and enthusiasm sustained within the teams thank you very much I picked up only 14 out of 32 are connected because it's going to be on the 19 top so first misconception is that agile is not new but it's not new because all of us know that the agile manifesto was prepared in 2001 February 2001 so it's not new and in fact this crumb or pattern language was introduced way back in 1995 so that's not new so actually all of us know that it's prevalent and it has a better success rate this are we so many other traditional technologies do you have any data on that sorry do you have any data on that not really but I have not that's a myth that's a myth that's a myth that agile projects work better than what is called projects I have I have come across a few statistics a few days back I can I can send you the link of course there are challenges even in the agile projects but I'm only giving the comparative here I'm not saying that they are the best in every situation he made a statement saying it's more successful than I challenge that more successful but of course there are doesn't mean that 100% I don't like it because I have seen about this what I have seen I can share that all I'm trying to say is it's a myth and let's move on okay other thing is people feel that agile is not plan oriented just because you know there is no upfront big detailed plan unlike in traditional projects so it's not the case it's continuous it's incremental I know it's very clearly explained in the onion diagram that you know the planning happens at high level in the right from the US plan and it goes on up to the daily planning through the daily meetings okay next agile means no documentation it's not true this miscut self-perception comes from one of the values saying that working software is more important compared to the comprehensive documentation but it doesn't mean that it doesn't require a documentary it requires but it should be just enough documentation in fact that can be one of the daily the team must take preparation from and support this master it's not so this is where the people go around because they come from the traditional dogmas that we take orders from the product manager here in fact you're in daily meeting you shouldn't go look at the master you're addressing the team and you're working as a team next you have agile is a bullet solution for software engineering problems they think how can we one more minute okay thanks so it's not a bullet solution it is it cannot be suitable for all types of projects as I said then you have four pair programming the other myth is that pair programming is I mean is very advantageous again there are cases that many times if the pairing is not properly done it can actually work the other way around so it's very important that the properly synced pair of programmers are pairing and then there is a better productivity and better agile produces poor quality output this is a bit I have heard sometimes but actually because of the frequent inspect and adopt cycles through various scum ceremonies like daily stand-up sprint reviews or sprint retrospectives there are many chances of improving upon and this is not true agile only works for trivial projects no there have been projects of you know highly scientific projects where agile has been implemented successfully next agile is always a better option then compared to the other alternatives no it's not it's not a case that can be actually a blend depending on the type of project that we have developers get to do all right thank you sir big fact program and how we use that practices I'll tell you my name and where I come from at the end when he rings the bell I want to give you the story that's more important than me I'm talking about this program which runs for 10 years and the scope that we have also managing a data center in fact we are managing this our customer so everything right up to even sending documents my career to the customer to the end customer it's just a key project so the what does the customer do they give requirements and then go oversight that's all and they pay the money on they just you know penalize us that's all they do and of course impose a whole lot of regulations for every single part of it whether it's a security you know certification or a health certification or security or disability related certification for frontier everything that's just regulations and regulations for almost every part of this program and so it's about 300 people and across many many different locations so as you can see there's a huge amount of collaboration required stakeholders so a lot of churn a lot of rework a lot of communication gaps and it was very tough very tough and we really struggled and so I happened to work in that program which was very it was initially it was a nightmare and then what happened after about two hours about 11 months or so what we decided to do you know it started going into it went for a release and then we also had production support help desk everything coming into our own responsibility so that latter part of the life cycle is also ours because we have to run it for 10 years and then what did we do so we came up on this actually it's not about somebody sitting there and saying do this people started as they always do they just improved right up bottom up and what did they do you know the testers they started giving feedback to the product teams and the development team saying that these are the areas they came up with this estimation model if you're going to touch this piece I think I'm going to take at least three cycles this piece I think I'm going to take only one cycle so then they started giving that feedback the developers immediately got to know that okay this is where these guys have a very good idea of where we make mistakes so even if the people change in that team those are inherently complex pieces of the product and the customization so they started working on it then the ui teams they started giving feedback and security so each place so the security you know people on the performance testing side they knew that security and we are going to have non-conflict conflicting requirements because three minutes up so similarly then the bpo back office front office they said if you're going to design it this way we are going to take longer to answer people we cannot meet this answer within so many seconds so what happened was a fantastic beautiful collaborative effort of people coming together during the requirements phase straightening out right from production support up to the domain business analyst all different parts of the team they got together and they did a whole lot of mistake-proofing and it came to a stage where it's running beautifully we've got down time to market quality is just shot up and we've got lots of very good appreciations and service credits and other things have really come down okay so thank you my name is radhika i'm from tcs i work for this program i'm also in the smack area agile in evangelist i'm going to talk about velocity skilling agility right we heard some beautiful lightning talks before this about how you do story point estimation and how beautiful story point estimation is i think we completely missed the point uh the guys who actually came up with story points discarded it right away because they said that's a stupid idea uh and we seem to be running crazy behind it right uh anyone's aware who came up with story points idea no but yeah we can dance around fire and god will give stuff from the top right because uh it was originally uh what cunning ham's idea uh of basically trying to create a level of in direction uh and then you know you can talk to what cunning ham he seems to have discarded that idea long back uh let's talk a little bit about what is the problem with story points and then we'll come to velocity because without story points you really cannot have an effective velocity calculation at least the way it's done these days uh so if you look at story points the most common story point mechanism these days is to use Fibonacci series uh right everyone agrees with that so uh what was the reason behind choosing Fibonacci series why didn't they just pick one two three four five the reason behind picking Fibonacci series was non-linear right so that's very important it's non-linear uh what does that mean that means if a story point is you know so you do one two three uh and five and seven so five is not you know five times more effort than one right and it's not also about effort estimation it was about relative sizing or relative complexity estimate right and the idea was to try and bucket things into different categories so it'll help us get an indication of how we can incorporate these things into the upcoming sprint uh but what ended up happening was basically uh we took story points we totaled them up and we said okay this was the velocity that I achieved last sprint right so let's say you did a three-pointer story and a five-pointer story and a ten-pointer story so you said okay my estimate is now my velocity was 15 story points because that's what I achieved last time next sprint if you took 15 one-pointer stories what would happen you will never be able to finish 15 story point estimates there is no doubt about that you will never be able to finish 15 one-pointer stories uh in in the same duration that you did three story points because it's non-linear if it was linear then yes you could take 15 story points and you could easily do them 15 one-pointer stories but the whole point was that it's non-linear and we completely missed that right the the rational behind it was we didn't want to apply arithmetic and statistics on it but we went back to applying arithmetic and statistics on it and then teams getting measured by velocity and teams you know management going crazy with it so agility is killing velocity what's the goal of your scrum team to produce more software faster i really think there's lots of behavioral indicators which i would say you know i think one of the things a lot of people talk about is output versus outcome so you could focus a lot on output which is velocity or capacity planning capacity utilization or you could focus on outcome which is little more than throughput right which is delivering value which is actually seeing users doing someone walked up to me and said our management is really pushing hard to deliver more and more stuff and you know agile is the way to do that right because agile is about faster delivery and it's like no agile was about doing less right you you you deliver faster because you do less not because you do more so i think you should focus on how little you're delivering and still adding value that would be the kpi i would look at not how much crap i can generate how fast in fact we have started focusing more on the business value delivered than velocity in one of the project we had a very good velocity but the business value delivered also poor that the product actually did not succeed in the market and then later realized we have not really focused on the values which you know the feature that we wanted to do the velocity was good but the but the product didn't do well so what's the use of that velocity so now we just focus more it's not like we don't look at velocity but the focus is more on the business value all right time out thank you