And Vijay Says…

Most of my adult life was spent doing consulting work for clients around the world on topics like information management and analytics. For last few years, I have also been involved in IOT, AI etc. When I look back – one thing has been common for all “data projects”, and that is the importance of TRUST…or lack there of. Billions of dollars have been spent creating BI systems ,data warehouses, data lakes, cubes, reports, workbooks, blah blah – and yet at the end of the day, spreadsheets still prevail at EVERY SINGLE COMPANY I know !  I wrote about this in the past.

But why ?

There are good reasons why users don’t trust data they get from BI, like

1. In most cases, users cannot see how it ties back to source systems
2. Transformations and enrichments to the data are not transparent to users
3. Users do not know how well the system has been tested
4. The tools may not be as intuitive as a spreadsheet….and many more

Every single one of these problems in “Classic BI” has a solution – which either a product or a service can solve. An IT expert can probably show what transformations happened, or create some reconciliation reports for example. Or a data lineage tool can trace back from report to source. So with some additional cost, we can minimize the trust issue – though the cost may eventually become prohibitive to build trust at scale.

Past is set in stone …most of the time, at least 🙂

The reason we could do all of this is because we are essentially looking at things that already happened – which are kind of “set in stone”. And the information we got from such systems had finite values as answers like “sales in north america was $10M” which is based on basic arithmetic . If I asked the same question again tomorrow, I will get the exact same answer of $10M. If I did not – I would know right away that something bad happened. If a bad decision was made –  for the most part, it is possible to trace back and validate the data and confirm if it was indeed bad, and prevent it from happening in future.

Now lets look at the world of AI .

To avoid a religious war on terminology – please allow me to use AI in its most generic sense as an all encompassing thing that includes what we call data science, machine learning, cognitive etc. Definitions matter – but for this post humor me and pretend all the right things are covered when I say AI.

Just like with classic BI, we use a lot of data and transformations. However, the fundamental idea now is that we are not just reporting on what happened – we are trying to make the best educated guess on what will happen in future. We are not in the world of “only finite answers” here – instead of the exact sales that happened this quarter in North America, we are often trying to find what are the odds that sales will be greater than $10M next quarter, for example.

Enter the trust issue

On one hand, it is quite useful to have a system that makes such predictions for us so that we have a window to the future. On the other hand, if we do not trust the answer – it is a lot more difficult to explain how the system arrived at the answer. And if the system told me today that I had 80% chances of hitting $10M for the quarter, it could very well tell me tomorrow that I only had 50% chances of making that $10M number.  I and everyone else in my team might think the system is foolish because we can see the math to get to the $10M number we want . Lets say the quarter finished and we did exceed the $10M number – this still does not mean the system was either right or wrong. That is the beauty (and pain) of how probability works !

Can’t the creator explain the creation ?

I am often asked “Can’t you just ask the programmer or data scientist who built this to explain how the system predicted?”. Yes I can – and some times that is all it takes to get the answer. But many a time, they may not be able to give that answer with the precision you expect. AI systems are learning systems (with or without human help) – and they learn and get smarter mostly by going through a lot of data, as opposed to just crunching logic fed by a human. By the time I asked the $10M question the second time, the system might have learned something from a new pattern it detected.

AI can piss you off

A sales forecast, in the larger scheme of things, is probably not going to change the world for most of us. However, if we think of other scenarios like say salary planning  or promotions where an AI system scores everyone in a team on a complex set of parameters and makes recommendations – it is hard to accept a decision that cannot be explained in an easy to understand way. The system may be totally right – or it might have all kinds of bias built into it with the model, or the data it trained with. It might have false positives and false negatives. There are techniques to minimize all these problems – BUT If it cannot explain its results to us as users – how will we know for sure ?

Can you trust machines ?

There is another version of the trust issue – when machines need to make choices that affect us. Lets say you are a factory supervisor driving in a self driving forklift that is picking up a heavy load from one top shelf and putting it in another top shelf, while your workers are walking below its arm. The machine probably has visual recognition capabilities, and can crunch lots of parameters from data and make good decisions. Lets say one day, the machine detects the load is too much to bear and it has two options – flip on the side and injure you, who is sitting inside or drop the load and injure your workers. What should that machine do ? And if you don’t know what the machine will do – or at least know that you can over ride it – will you work with that machine ?

AI – its just like us, except it isn’t 

I also get asked “Well, AI is supposed to think like a human, so why can’t it explain its thought process like we can?”. This is an excellent question and it presents two issues. 1. We don’t all think alike – even in the fictitious forklift example, I am sure different people will choose differently.  And 2. We often take a decision, and find an explanation for the decision later if someone asks. We can’t always explain our decisions very well either to someone else except for simple cases. And finally, we make poor decisions too. So mimicking human thinking as-is perhaps is not the best way to think about AI either 🙂

AI is everywhere, and mostly harmless

I am of course not generalizing that all AI scenarios run into a trust ( or ethical or moral) issue. Many don’t – for example an AI algorithm might predict how much longer a device will work before battery runs out. I doubt I will have a trust issue if I see it work approximately well for first few times. And there are several of those kinds of “little” AI solutions all around us – and many might not ever be visible to us. We just take them for granted ! Even in the sales forecasting or promotion examples – over a period of time, we may trust what the system tells us. But the trouble is – will we give it enough time to let it work long enough to earn our trust ?

So what can we do, really ?

Just like other projects, AI projects need some basic education and expectation setting for stake holders before we embark on them. Unlike basic math, and if-then-else logic – statistics concepts needs a bit more hand holding. People tend to use terms like confidence, significance, sampling etc loosely and it is very easy to set wrong expectations with stake holders even with the right intentions. And then there is the issue of trust, and its ethical and moral considerations. It is important to discuss these thoroughly upfront, and during the projects . When done right – and transparently – AI can and does add significant value to us. Its on all of us in this industry to make sure we let AI earn trust the right way !.

 

 

 

 

There is nothing that bores me more than repetitive work ! The irony is that I don’t have this problem outside work. I can eat rice three times a day for rest of my life. I can listen to the same set of Ilayaraja songs every day. I have watched Gandhi movie 30+ times , All 7 seasons of West Wing multiple times on Netflix and probably will do it again this Christmas. I love reading the same books over and over every few years. I can spend months perfecting every last bit of synchronization between my dog and me before we compete in dog shows.

But when it comes to work – I cannot stand repetition !

I get bored very quickly. In the early stages of my career, this kind of worked in my favor – I just switched the areas I worked on every year or two and had a fun time picking up new skills and having fun, and in general my career progressed at a fair pace. In hind sight – I was probably taking way too much risk (while doing this on an H1B visa at the time) , and I had a lot of great managers who never stood in my way when I wanted to do something new. But once I was in leadership roles at work – it became abundantly clear that repetition is a good thing for abundance and it is futile to avoid it. On the bright side – I was also quite lucky that people I got to work with mostly loved optimizing repetitive work and gaining efficiencies. But the problem remained that I still could not feel excited to do encores – and since I generally won’t ask my teams to do anything I wont do myself, this started becoming a bloody pain for me.

And that is where side projects became a life saver for me ! They gave me enough variety and stopped repetitive work from distracting me.

The first thing I tried my hand was in blogging – on what was then called SAP Developer Network. And thanks entirely to Marilyn Pratt, I started enjoying it and later branched to having my own personal blog. Only a small % of my blogs talk about technology or work related stuff – and I mostly write to get things off my chest, as a venting mechanism. So I write on the wordpress app on my iPhone for the most part, never proof read or spell check. But thanks to some very loyal and generous readers , it has always been a fulfilling activity. What I realized is that it also benefited my work some how. Many a time, I have landed at airports across the world where someone recognized me and chatted, and I have had clients give me business because they googled me and found my POV on a topic of mutual interest. All this is net goodness – but 500+ blog posts (and podcasts and videos and all), I should say that it is hard to find motivation to write these days. From 2 blogs a week at one point, I barely write one post every other month now.

Then I picked up my old hobby of dog shows. I was quite successful as a competitor growing up in India and have won my fair share of top awards – and lost more than I care to count. My claim to fame was my german shepherd who could go to our corner store and buy milk by himself. The local news paper and TV channel there had that story covered when I was in college. What I realized in the second round was that my love for dogs is now significantly more than my love for dog shows. Long story short – I did not push myself or my fur kids hard enough and we mostly just hang out and cuddle.  Again – there were many positives to take to work. One is increased patience and the other is the invaluable lesson that any problem can be solved by breaking it into several smaller problems and solving each and putting it back together.  And a side “benefit” – having dog hair on your trousers are a great conversation starter with fellow dog lovers at work 🙂

My day job these days is to run a small “hyper-growth” business at IBM ( pretty sure if you name a buzz word, I have it in my portfolio – Cognitive, predictive, IOT..check, check and check) – with more people than I have ever led before, and a target with more zeros than I have ever managed before. There is something new to learn every day, to say the least. Obviously, doing such a business profitably includes a fair mix of repetitive work, and first of a kind work. Thanks to the long plane rides – my coding and math skills are somewhat kept sharpened through all this 🙂

As luck would have it – I also got the best side project I have ever had in my career. I am the executive sponsor for our consulting school for senior managers – which we lovingly school “The Bee School”. There is nothing more fulfilling than joining hands with our very passionate and skilled Learning and Education team, and a bunch of enthusiastic fellow executives to teach and learn from our next generation of leaders. It has been an eye opening experience for me to see how an education program is designed from ground up, how we train the trainers, how we execute the training and how we learn and adapt quickly. I am seriously thinking of being a full time educator now. The best part of the week we spend with a hundred senior managers is the Cognitive hackathon night where people who have no prior coding background build a virtual agent using Watson and realizing how easy modern technology is , and how much value it adds to clients ! I have to thank my “big boss”  Ismail Amla for the opportunity .

My latest side project is to build up a garden – flowers, veggies , anything that grows. The last time I did this was when I was in grade school. What makes this a fun project is that I live in AZ, where the summer temperature is a cool 100F to 120F. I have a willing co-conspirator in my mother in law who has a beautiful garden in India. My core design principle was “hyper growth” – so I had to do my research on what plants will thrive in our climate, which ones will yield flowers and fruit the fastest, and how to build for some future scale in case this actually works.

We already have some success – yesterday we ate the first egg plant and tomato from the garden . And there are about 20 different shades of hibiscus flowers in the yard. The ROI for these vegetables is now something like every $5 of veggies cost about $500 ! But the sheer satisfaction of seeing “hyper-growth” is priceless 🙂

There are some good lessons I could potentially take to my work – 1. Time spent in upfront research saves a lot of rework ( The few plants we lost in the process were all mistakes that we could have avoided if we took time to read more and ask around a bit more). 2. Automation is a good thing and something you need to put in place upfront ( everything on timed automatic irrigation now). 3. You cannot just trust process and automation if you want solid results ( Till plants take root, you need to water extra , give them support, remove weeds and generally treat them like a little baby 🙂 )

The current gardening project should keep me going for a little . But I am already thinking of my next side project – and one idea is to buy a German Shepherd puppy and train for Schutzhund/IPO. I am inspired by 7 year old Samantha – watch this video and see for yourself. Or perhaps start writing my book – The Matrix Dysfunction !

 

 

First off – thanks to SAP for having me at the event as a blogger. In my day job, I no longer work on SAP technologies closely. But most of the clients I have are SAP shops and often talk to me about SAP given my heritage. Sapphirenow keynotes and executive meetings with the board members were enlightening for me. Here is my quick commentary.

Hana – done and dusted ?

As I walked into the show floor, the first thing that caught my attention was that there were hardly any signs for Hana. After the last 6 years of Hana overdose, I was pleasantly surprised to see this. I have always maintained that success of Hana will be when SAP does not explicitly talk about it, and it just becomes a part of everything they do. To be clear – Hana is not a part of everything they do quite yet – most of their customers are still not on Hana, but SAP probably feels confident that the tipping point on adoption has happened and momentum will keep building. Congrats to all my friends who worked tirelessly on it.

Indirect Access – SAP heard the community loud and clear !

Bill McDermott needs to be commended for taking on the indirect access issue heads on in his opening keynote. Without that, the whole empathy thing would have sounded hollow. Now the proof of the pudding is about SAP following through and changing the perception in the ecosystem. I heard it loud and clear from Bill in a private meeting that he is committed to making it happen and I trust him.

All hail SAP Leonardo – just don’t ask me what it exactly is 🙂

This Sapphire was all about SAP Leonardo ! Unfortunately, I cannot quite explain what it really is under the hood. From the slides I saw, it seemed like everything but the kitchen sink was thrown into this package – analytics, block chain, cloud…. . To be honest, it seemed like SAP did not have the time to think through a crisp message before the event. I am sure this will be a lot better by TechEd. I should say – while admittedly not being a marketer myself – I doubt SAP will keep this name “Leonardo” for long. It sounds like a cool internal project name , but I doubt this will catch widespread imagination. I heard this from several people on the show floor too, including several of SAP’s largest customers.

SAP loves partners !

The keynotes were totally partner friendly – We had senior execs from Dell, intel, Google, IBM et al show up on stage at various points. But it was rather light on customers. Perhaps this was a conscious decision given the customers would have wanted to talk about their existing (boring?) – mostly on premises –  systems, where as SAP wanted to position their future with cloud and AI and all that. I trust the balance will tilt towards customers as pilot projects for Leonardo get completed. While I have no idea of commercial impact, the Hana startup program was a good thing for Hana adoption. I was surprised at not seeing it being repurposed to get Leonardo a similar lift. Perhaps more at Teched?

What is next for UI and API in the machine to machine world?

SAP did an exceptionally good job with UI with Fiori etc – and Sam Yen and team should be incredibly proud of that. It was also an expensive effort to get SAP to look cool.  Now with Bernd and Hasso painting the vision of automation, ML etc – I would love to hear what is the next evolution of SAP UI. When most of the back office and factory processes are automated – there won’t be human beings keying in things into a beautiful screen for most of the processes. And for the few screens that do remain in future, screens cannot be static either – and will need to learn user behavior and adapt. Just looking at how many screens get used today and making decisions based on it might not be fit for future. I have a similar question on APIs. Future is mostly about machine to machine interactions and AI. The way APIs are designed are not fit for this kind of future – machines for example do not need anything more than a binary protocol to talk to each other. Sapphire is a business setting – so perhaps these things will find a place at next teched.

Time to rethink demos a bit ?

I will just make two points on keynote demos. Putting sensors in drills and showing how it works in real time on a live stage is something I applaud. What I thought was less impressive was the articulation of the business case. If I am renting a drill – the renting company cannot give that drill to anyone else. So – what is the point of charging per hole drilled ? What if I rent a drill, keep it for a week and drill 2 holes ? Its not a logical business model and it detracts the people watching it. I hope SAP just finds a real life customer case to demo next time.

From the beginning of time, SAP has demonstrated dashboards. I love dashboards – and the ones showed this time were visually impactful too – and maybe had great ML behind the scenes. It just did not show that way – I think its time SAP moved to some other way of making the message hit home.

SAP makes a much needed entry into AI

Hana brought the speed and technical simplification SAP needed. Now SAP is taking the next logical step to make applications smarter. SAP leaders I met here were all realistic on the effort and skills needed to make this happen. Ariba choosing Watson as its cognitive engine, and Google Tensorflow being endorsed in a big way by SAP are all signs that SAP clearly understands that AI is an ecosystem play.  In my mind – SAP can’t get the skills and IP needed to keep up with this market without a string of acquisitions. SAP has always had a strong services ecosystem around it for decades – but they historically have been low key on role of services in creating value. The future of applications might not all be pre packaged like how ERP traditionally worked – and the significant need for domain expertise etc is readily acknowledged now. 

The one aspect where I have some pushback for SAP is the inside out view that AI sits inside SAP systems. While there are valid scenarios like invoice and payment matching where all data sits inside SAP – it is not realistic that a company can model real world with just SAP data. Again – I think its just a matter of time for the vision to mature.

It was fantastic being here – and catching up with old friends and making new ones. It was also fascinating meeting several people I have known only on social media – thanks for introducing yourselves ! Till next year !