The future isn’t all what it used to be anymore

Posted on by • Posted in Uncategorized • 2 Comments

This is really difficult for me to say since optimism about technology making our future better is what has kept me going all my adult life. It’s why after a degree in mechanical engineering and an MBA , I chose to be a programmer . It’s also why despite multiple leadership opportunities on sales and general management, I continue to be a hands on technologist .

It’s not that I have become pessimistic suddenly about technology’s power to transform society today – nothing could be farther from truth . It’s just that I have a lot more pessimism about the humans who use and control the technologies that will impact us .

I have been quite an active participant on social media – especially thanks to the easy access via my iPhone . Between Twitter , LinkedIn and Facebook – I have more than twenty three thousand or so connections (including some duplication for sure). On twitter – I only follow about 150 people, mostly because I can’t keep up with a larger feed . I strongly believed that this network has given me mostly net goodness.

I have thought a lot about what is my primary principle for social media . I think the honest answer is convenience !

I do 90% or more of all my social media activities on the apps on my phone . At some point, I started accepting vast majority of connection requests without too much due diligence – clearly not a smart idea and I am slowly cleaning it up now. I haven’t fiddled with ALL the privacy controls on each platform . It’s not that I was fully ignorant of what these platforms did with my data – just that I didn’t think of it more than as a nuisance with a bunch of merchants trying to sell me stuff non stop . I have often discussed with friends from my line of work how some of these targeting algorithms could be optimized to make it less annoying .

Then this Cambridge Analytica thing came out ( and the continuing conversations about the Russian influence on elections) , and yesterday night I read Zuckerberg’s response on Facebook . It’s extremely depressing to say the least .

The irony is that yesterday night is when I reinstalled the FB app back on my phone after a month away from it – and the first thing I noticed was Mark Z response ! I did go and tighten privacy controls as soon as I read it !

I work in analytics and AI – and have a special interest in getting insights from unstructured data . That means I do know how easy it is for FB and others to gain a very deep level of understanding of our lives . I also don’t think that privacy controls by themselves are of significant benefit . When you have a lot of data from a lot of people – you don’t need every last bit from every individual to get the deep insights. I will spare the tech aspects here – but suffice to say , these platforms have disproportionate power even if we assume they are all angels . We also know by now that they and the people they give access to our data are not exactly angel like .

I do value the ability to stay connected with friends and family . I also enjoy the vacation pictures and puppy videos . So the only solution I can think of is to significantly reduce what I discuss on FB etc. I didn’t miss Facebook when I stayed away for a month . So I also wonder if I could just get out of it for good and be done with it . I know I am not alone on these thoughts .

There is an interesting cross cultural aspect to consider too . I have spent a lot of time in Europe thanks to my work . There is no comparison between US and EU when it comes to privacy . If I lived in Europe for longer , I seriously wonder if I would have traded privacy for convenience . Plus the government wouldn’t have allowed a lot of what FB etc has gotten away with in US . Given its global reach , I do expect FB to get hauled up in EU at some point soon .

Then there was the poor woman cyclist who was killed by an autonomous Uber car in Tempe , AZ . It’s not very far from where I live – so this hit home harder than usual . Tempe police has released a preliminary report and video (Its disturbing – so not linking it here) . I really wish the lady was way more careful about crossing the road at night . Such a tragic end ! I am not at all a legal expert – but it’s quite possible in my view that law might blame the lady and not hold Uber responsible for this accident .

I have a big interest in the topic of man and machine working together, and have written and spoken about it a lot . A critical question here is whether a machine should be held to a significantly higher standard than a human in similar situation . Several of my friends think a machine should be held only to the same standards as humans.

For at least two reasons , I actually think machines should be held to significantly higher standards than humans

1. A machine is more efficient than humans and can keep getting even more efficient in lesser time than humans by comparison . So the flaws in those machines are also amplified several fold more thanks to mass production of machines . We can’t risk the world being full of half baked machines , irrespective of benefits in cost and convenience . No price is too high when it comes to protecting human life

2. A Machine can make faster decisions than humans and use more sources of information than a human can to make those decisions . At the same poor visibility , a human driver probably will have made the same mistake the autonomous car did – and that’s a fair argument. But vision is not the only sensory option for the car – motion detection , heat detection etc are all options and there are plenty of sensors/actuators/radar/lidar on such cars. And the cost is also declining pretty fast. So I think it’s a false equivalency to say a human driver would have made the same error and hence the machine should get a pass .

And in the video – it looks like the driver sitting there didn’t notice anything till last second , arguably because of the trust in the machine to do a good job . This trust is what worries me . In the situation where there is a passenger in front of car is straightforward – the car should break . It could get much worse in cases where the decision is a choice between two bad options like hitting one person or hitting another via swerving. If the straightforward option itself is not reliable , how would we expect the machine to react in more complex situations ?

I think Uber did the responsible thing by pulling the self driving cars off the street . They are also apparently fully cooperating with the investigation . I also think AZ authorities are correct in not making any snap judgements on tightening regulations.

This should wake us all up – testing autonomous systems is quite hard to begin with . And it needs a lot of inter disciplinary research investment to get better and more consistent . We are not exactly short on money or talent to get it done – we just need to put safety as a bigger priority than it is now . I love capitalism as much as the next person – but commercial greed just cannot be allowed to over rule safety under the branding of capitalism

I absolutely think our future is still about technology doing good things to improve our quality of life, including social media and self driving vehicles . But it’s high time we took a long and hard look at what are the top priorities in our quest to get there . Better , faster , cheaper is not enough – we need to add SAFER as a first rate citizen into the value proposition and it should not be negotiable !

Advertisements

12 questions that I get asked the most about social media

Posted on by • Posted in Uncategorized • 7 Comments

A lot of people know me primarily through social media – this blog ( and SAP blogs in past) , LinkedIn , Facebook and Twitter . This has generally been a net positive experience for me .

I have been stopped for conversations at airports around the world by random folks. I have walked into meetings with clients where they knew my point of view from my social media posts and hence felt like they knew me already . I was quite active in training and showing dogs till few years ago – I keep in touch with them on Facebook . I have hired quality talent , and made great friends over LinkedIn and Twitter . Just last week a guy who went to college with me and stayed at the same hostel in India – he was in Phoenix for a conference and looked me up on LinkedIn . We had lunch at my house and caught up after a few decades . The list goes on and on – and it more than makes up for the handful of terrible experiences I have had on social media .

I routinely get asked a bunch of questions about my views on social media and how I use it – mostly by people who want to learn , but occasionally also to just imply that they don’t think very highly of it

So I thought why not just address the common questions here .

1. You seem to have a lot of time to be active on social media . Don’t you have a real job ? I can barely keep up with my day job

Contrary to popular belief , I do have a real job and that needs my full attention 🙂 . However , over time I have figured out how to integrate social media into my daily routine . I generally use social media only on my phone – including for long form blogging. That saves a lot of time .

A lot of what I write about is of course regarding my day job . The conversations I have with my clients and colleagues and people I meet on my travels are usually the triggers for my posts .

We always find time for what we like and prioritize . I like to eat Indian food and don’t mind driving 10 miles extra to find an Indian restaurant . Another person might like working out daily and will spend that time running or finding a gym. “I don’t have time” translates usually to “it’s not important for me”.

2. Aren’t you worried about what you say openly in public domain ? Does your manager yell at you for writing something ?

I only talk about my personal views – and usually only about things where I have a strong point of view. So I don’t feel any worry . If I suspect people will mistake it as my employer’s opinion – I state upfront that it is just my views . It has worked well for me so far . I also have no problem standing corrected in public when proven wrong .

Thankfully none of my managers have ever yelled at me for blogging (they do so occasionally for others things) – and have only encouraged me to do so . And I have been their mentor a few times to ease them into social media.

3. Do you have a time table for social media ?

No I don’t blog or tweet at set times . I say something whenever I feel inspired to do that . I don’t check what times I should post to make biggest impact . I don’t aim to get a certain minimum number of page views. I do check statistics on my blog when WordPress alerts me and it’s usually something like “20% of what I write get 80% of views” . Time tables and fine tuned analytics work for many bloggers I know personally – I just haven’t taken that route yet .

4. How do we get started on blogging , twitter etc ?

I can only say how I progressed . My primary use then and now is as a source of curated information . I read at least 10x what I write . Then I started commenting on blogs that I started following . It took me a couple of years to have the confidence to write a blog myself – that was on SAP SCN site . A few years later I figured I have things to say that have nothing to do with SAP – so I started this blog on WordPress .

Twitter use has evolved over time . I use it less and less these days to say something . But it’s still one of the best places to get information . I only follow 150 people there and I think close to 9000 people follow me . Some folks think it’s rude to not follow back . I ignore that policy – I can’t stay on top of a feed any bigger than what I have now . I keep changing the list every now and then .

5. Do you only talk about work ?

Hell no ! I write about anything around me – politics , religion , economics , cricket, dogs … whatever inspires me is a fair topic . Some folks aim for depth on any one topic like Technology . That’s not my approach – all topics that interest me are fair game .

I try to not deviate from work related stuff on LinkedIn . But the truth is all popular social media platforms now have content overlap . You can use the controls they provide to tune your experience to some degree but it’s hard to completely set hard boundaries . I do have a handful of work friends on Facebook for example – and it has not stopped me from expressing what I have to say .

6. Do you take a break from social media ?

Yes I do ! Throughout the year , I take a few months off each channel I use . At the moment, I am out of Facebook . I delete the app on my phone to prevent the temptation to take “a quick look”. These breaks are one reason I continue using these platforms over the years . It also gives me time to read more books !

And there are a few months every year when I just can’t find any inspiration to blog . I have considered just giving it up altogether . But then magically one day I feel like it again and I open the app. The opposite also happens occasionally – I will post multiple blogs back to back .

7. What does success look like ?

For me , engagement is the big prize . When people comment and share and email me about something I wrote – I feel I did something worthwhile . Page views are a feel good metric too – and over the years a lot of page views happen for reasons I can’t always explain rationally . Some of the blogs that I thought I wrote terribly have had the most views – and sometimes even the most comments .

I push all my blogs to LinkedIn and twitter . And if I see someone tweeting or sharing , I will retweet or like. That’s the extent to which I promote . I don’t do any active SEO etc . As the networks have increased over time for me in various channels , I do get more people reading and commenting .

8. How long should a blog be ?

I am a minimalist by nature and quite lazy . So I try to use the minimum words to state my point of view. English is not my first language – so I don’t always succeed . Since I read most blogs on my phone, I tend to favor shorter blogs . There are amazing bloggers who do 5000+ words on average – I don’t have it in me to do that . 1000 to 2000 words is the most I can do.

9. Does someone review what you write ?

Not at all . I am kind of ashamed to admit this – but I also don’t do any proof reading or spell checks . When I finish typing, I hit publish . My wife usually spots errors when she sees it on her feed and routinely criticizes me on not being more careful . Generally this has worked for me so far so I don’t think I am going to change it . On the flip side – this is my biggest fear about writing a book . The process of professional editing scares the crap out of me 🙂

Strangely – I am fanatic about proof reading every document I create at work .

10. Who is your target audience ?

This may not come across very well – but the truth is – I write for myself . I like to believe someone other than me and my mother will find what I write to be useful – but as I mentioned before , only a relatively small proportion of what I write evoke any reaction from my readers . I am cool with that . When I do get feedback , I incorporate it as much as I can the next time I write .

11. Has blogging helped your career ?

Oh yes , without a doubt . While it’s hard to make a specific attribution – blogging and tweeting has generally been quite helpful in making me stand out at work. And as I said before – knowing my POV through my writing has generally been a help in my interactions with clients and business partners . To the best of my recollection – I have never been promoted or given a raise because of what I do in social media . I have been asked this a lot over the last decade . Career progression has historically been generally all tied to traditional metrics likes sales and delivery numbers and people management and so on . But indirectly , social media has only helped my case .

12. Do you use your blog to evangelize your employers products and services ?

In general, I don’t . I don’t think I sound credible pimping what we sell . However – I am very proud of where I work and what we do . So that might occasionally reflect in what I say . There are a few notable exceptions to this . Last year, an analyst criticized Watson in his report and I felt strongly that it was unfair – so I posted a rebuttal in my blog . I did explicitly state though that I am an executive at IBM and I own stock in the company . I have also written about SAP HANA when I worked at SAP Labs in the engineering team .

Math and Science in daily life – Part 2

Posted on by • Posted in Uncategorized • 2 Comments

Last time, we discussed how Pythagorean theorem helps us in every day life. Today, I would like to invite your attention to Pi , a beautiful constant (yet one we cannot fully measure) that is in practically everything we see and use.

pie

Image courtesy https://firstnewnan.com/casual-pi-day/

So, what is Pi ?

The common definition is that it is the ratio of the circumference of a circle to its diameter. How cool is that ? ALL circles have the same ratio of circumference to diameter – and thus all circles are similar figures. In Math we represent it as the greek letter  π .  There is a lot of interesting trivia to know about pi – so lets start there

Pi goes on and on

π is an irrational number ( a geeky way of saying it cannot be represented as a regular fraction like 1/2 ). It goes 3.1415926535897932384626433… with no repeating pattern emerging in its decimals.  It can be approximated to 22/7 for use in daily math. Its a standard test for supercomputers to see how many digits after decimal point of pi can be computed. Practically only a few hundred are needed even for the most complex applications though.

Pi is the reason you can’t square a circle !

Ď€ is not just irrational – its also transcendental ( yet another geeky way of saying it cannot be expressed as the root of a polynomial ). Not all irrational numbers are transcendental – like for example the square root of 2 is irrational, but not transcendental. Its a very special class of numbers – mostly because it is very hard to prove a given number is transcendental. You may have heard of “Squaring the circle” as a way to say “trying the impossible”. It is because of pi being transcendental that you can’t create a square with the exact same area as a circle with a compass and straight edges. 

Pi has Indian connections 

While the origin of pi is attributed to the Greek mathematician Euclid, several Indian mathematicians like Aryabhatta, Madhava led the effort in providing proof and to calculate its value – usually based on infinite series of numbers. Its also interesting how the algorithms used to calculate pi has changed over time. It started with infinite series, then moved to iterative algorithms – and then switched back to infinite series again thanks to the use of an equation developed by the genius mathematician Srinivasa Ramanujan

OK, so now about the use of Pi in daily life

Finding the area, volume, circumference of anything that has a curve

Anything that is curved has some association to a circle in math – and circles all have pi as the ratio for their circumference to diameter. Consequently calculations for all those curved things like circles, spheres, cones, bell curves and so on ALL have pi in it.

Pi is also useful with straight lines and angles

While we commonly talk about measuring angles in degrees, math geeks measure it in radians. Quite simply a full circle is 360 degrees and in radians it is 2Ď€ radians. In other words 1 degree is π/180 radians. Thanks to pi finding its way into measurement of angles – it features in most of trigonometry as well – even though all you see is straight lines and angles between them. This extends to Calculus and other branches of math which we will explore later. Pi is the one constant that is everywhere !

Pi in the sky…and elsewhere

Every aspect of transportation uses Pi in its calculations. This one is personal for me – being a mechanical engineer by trade. Think of a plane or a car – everything from calculating surface area to wind resistance includes pi (remember all curves lead to pi !). Not only that – pi is integral to the calculation of all navigation be it in air, water or land – including finding the distance between stars and planets etc ! This is especially critical for planes that always have to fly in arcs, and where fuel consumption calculation can be a life or death issue.

Pi and the length of rivers

Prof Stolum of Cambridge University showed using Fractals that the ratio between the actual length of rivers from source to mouth and their direct length as the crow flies can be approximated to pi !

Pi is a friend of random numbers

We saw how Pi helps with curves and angles between lines. But it also shows up in seemingly unrelated domains. Here is an example – among a collection of random  numbers, the probability of two numbers with no common factor is 6/Ď€2 . If you are a math geek – you can find many other examples like this – like Cesaro’s theorem and Buffon’s needle problem where pi comes up unexpectedly.

Pi as a safety net 

Whether we are designing a beam which should not break apart from vibrations ( My mind is racing back to my machine design classes) , or a video game (much more pleasant memories from creating games on BASIC) that should not crash when users do random things – the engineers factor in some “randomness” into their model to account for real life. No surprise – all those models typically include pi ! This is just a result of something we spoke above – such probability distributions typically use “area under a curve”, and that by definition needs pi

Math and Science in daily life – Part 1

Posted on by • Posted in Uncategorized • 14 Comments

In my job, I spend a lot of time explaining technology topics to my clients in simple english. Off late, a lot of such conversations are about AI, Data science, Quantum Computing etc . Those topics are rooted in maths and physics (amongst other things), and I often find there is some fear about math and science that exists in the minds of some people listening to me, and it gets in the way of their appreciation of technology.

I am also the dad of a teenage daughter who loves math and science. From my daughter and her friends and teachers – I got the feeling that many a time students get education in math and science in quite an abstract way, and it leads them to think “Why should I really learn any of this? I am not going to use this in real life”. In a few weeks, I am going to give a talk to high schoolers on how the math they learn in school manifests in real life in ways they may not have realized .

So I am going to try a couple of posts here to see if I can explain how simple math leads to powerful and beautiful things we see and use in our daily lives. I would really love your feedback on each topic including alternate/better explanations, and also your suggestions on what else would make good examples. If anyone wants to post a guest blog – we can consider that too.

Let me ease into it with simple math and see how it goes . We can build up from there . Here we go !

Pythagorean theorem

Lets start with the very simple and quite powerful Pythagorean theorem . It states that the square of the hypotenuse (the side opposite the right angle) is equal to the sum of the squares of the other two sides. We remember it via this elegant equation

 A2  + B2 = C2 Screen Shot 2018-03-02 at 4.33.04 PM

This theorem has been proved in several ways , so we will skip that part and get to the fun aspects . If there are three positive numbers A, B and C where A2  + B2 = C2    is true, then it means that there exists a right angled triangle with A and B as the short sides and C as hypotenuse. Some of the fun is based on this converse property of this simple theorem.

How to make square edges on your garden bed

Any three numbers which make true is called a pythagorean triple. The numbers 3,4,5 form one such triple ( 3 squared is 9, 4 squared is 16. the sum of 9 and 16 is 25 which also happens to be the square of 5 ).

Here is a simple problem – lets say you are making a garden bed to plant vegetables.  How do we make the edges square if all you have is a tape measure ? Easy – just make a a triangle with sides 3, 4 and 5 units using the tape and mark the end points on the ground. Voila – You have edges at perfect square !

How big of a ladder do you need ?

An extension of this is also true for finding the length of a ladder required to paint a wall – probably a question kids get in school exams. If the point where the ladder needs to touch the wall is 4 meters, and the bottom of the ladder is 3 meters away from the wall on the floor – the length of the ladder needed is 5 meters !

Or let’s say you want to convert stairs into a ramp on your back porch – the same concept applies .

Real engineers use this information frequently – the one I remember the most from my college days is in surveying land, finding length of trusses etc .

Planning your next painting project

Lets say we draw three similar figures using the sides of the right angled triangle forming one side of those figures. Similar figures just mean the lengths of sides have same ratios, the included angle between sides are the same.  Easy way to think about it is zooming a picture on your phone . All parts increase or decrease in size – but their ratio to each other stays the same . It does not matter what the figure is – it could be a square, a pentagon or another triangle. The area of the largest figure (the one using hypotenuse as one of its sides) is ALWAYS the same as the sum of areas of the two smaller figures. How cool is that ! See this picture from wikipedia to see what I mean

520px-Pythagorean.svg

Lets say these three squares are walls you need to paint . For the same quantity of paint you will need to paint the largest square – you can paint both the smaller squares !

How much marinara do you need ?

It’s not just about squares either – this area computation via Pythagorean triples also applies to circles . So for the the quantity of marinara you need for a 5 inch pizza can be used to coat both a 3 inch and 4 inch pizza ! If your mom is like mine – I wouldn’t advice trying this math on her while she is making you pizza . Theory of “mom is always right” over rules every other rule 🙂

If you are a social media marketer

You probably are a fan of Metcalfe’s law which says the value of a network is proportional to the square of the number of connected users in it . Let’s say you are given access to three potential networks . A has 3 connections , B has 4 connections and C has 5 connections . Since The squares of A and B add to the square of C – you get as much out of a network with 5 connections as you would get from A and B which together add to 7 connections . Spend your time and money wisely !

PS : I have my own misgivings on this – but would like to hear your feedback anyway . I didn’t come up with this myself and can’t remember anymore who told me this originally . But I have teased many a marketer friend with this at social events 🙂

What’s next

I think I would like to example the wonderful pi next . There is pi in pie !

Looking ahead – what jobs will technology take away ?

Posted on by • Posted in Uncategorized • Tagged , , , , , , • 2 Comments

b8VJ4I6wR0yEt8Lxn1M3hg

As with a lot of things like politics, religion and so on – the world is sharply divided between people who believe AI and Robots (or automation in general) will take more jobs away than it creates.  I was drawn into this debate yet again by a few friends couple of weeks ago – so let me jot down while I still have it fresh in my mind. My crystal ball is not any more effective than yours when it comes to looking into the future – but there are a few scenarios where I do think jobs will be taken away. If your job is in one of these categories, the smart thing to do is to gain additional skills. Just to be clear – I also think there won’t be any net job losses. As always – this is all strictly my personal views on the topic, and not that of my employer.

Another way to look at this is – many companies will automate tasks and eliminate labor where they can to save costs. If you have skills that can make them earn more revenue directly or indirectly, you get to stay employed. Otherwise instead of reinvesting the savings, the company will probably treat it as profit, or keep the cash for future. The job itself might stay in many cases – but there just won’t be a need for as many people to do it. Granted – there will always be exceptions. Technology will also create a bunch of new jobs – which I will write about in another post.

I think there are at least four categories of jobs that will get disrupted soon.

1.If most of what you know is public knowledge

This is especially true for my own profession, which is consulting. In the 90s when I got out of college – there was no google. If I knew something special ( from books , professional magazines, training etc ) – a client would pay some money for me to tell them what I knew that they did not. That does not happen much any more – there is no premium for internet access . Clients and consultants both have access to similar information – so you need to know more than what is available on internet to fetch a premium. It might sound ridiculously obvious – but this is a bigger threat to (especially junior) consultants than almost anything else .

You absolutely need to stay couple of steps ahead of market to add value to a client today. Having a logically defensible point of view, knowing what others in the industry are up to, what disruptions are on the horizon, what untapped opportunities exist are all still things a client will pay a premium for.

Consultants are not the only ones at risk either. As an example – A hotel concierge function could also fall in this category. You don’t need a human to get you a restaurant reservation, check weather, know the local tourist spots and  so on. However, it will be hard to replace a human who can help you score a last minute Hamilton ticket in Broadway, or one who can answer questions from four different customers in parallel and make them all feel special.

2.If your work is all about short tail questions from a customer 

A lot of systems we use were not designed with end users in mind. Thanks to that, a lot of human intervention is still needed for people to use things they bought. A good part of customer service calls are about answering questions like “whats my account balance”, “can you reset my password”, “Can I set up a payment plan”, “Can I use a different credit card” etc. Automation is mature enough already to do those things without human intervention . If that is all your skill is – your job probably will be taken away soon.

But there are lot of things automation cannot do in this scenario  – at least not yet. For example , talking a customer out of canceling a service is not something AI can do very effectively like a trained retention specialist. From a customer’s point of view – an automated way of resetting a password, or making a routine payment would be easier/faster than needing to talk to someone. But when you are upset with poor service , or want to talk through multiple options – there is nothing worse than listening to a machine with a long menu. Also think of this – as tech (and laws) improves all around, in most categories customers will have zero or low switching costs.

So if you are skilled at higher value service – you should be in hot demand. The money an employer saves by automating the short tail responses is what lets them invest more in higher value services. Of course we can also take a cynical view that some companies will just add it to bottom line and not bother re-investing. While that is a short term possibility, I doubt they can do it in long term without risking their whole business.

3.If you are in a job where process trumps thinking 

There are several jobs where the job needs very little original thinking. The critical thinking is done by few people who designed the workflow, and not by people executing. This would include things like preparing fast food, paying invoices, checking totals, scanning documents etc. These jobs are generally at risk given they are easily automated – and probably the only reason they are still around is because of the one time cost of implementing new technology. Given all tech will eventually commoditize, this is only a temporary safety net.

The human intervention will be limited to exception processing in these workflows, especially those that involve safety, brand issues, downtime issues etc like – what if the lettuce delivered is rotten and you need to run to local grocery to buy some ? What happens if the scanner stops working the last day of the fiscal period ? Do you want to harass a customer on collecting $100, when you know in 3 months they are due for a $1000 renewal?

4.If your job is only about answering questions, and not about asking questions

Computers – and all the advances in AI and Quantum computing and whatever comes next – will keep getting better at answering more and more complex questions. There are questions a computer can answer faster and more frequently than humans today – like who was the 44th President of US? What planet is closest to earth in the solar system ? . There are questions that are really hard for computers too, where a human can often answer effortlessly – like who was the quarterback of the super bowl winning team the year the 44th President of US took office ?  But over time, we should expect computers to generally be able to answer most questions we ask.

But humans are way better than computers when it comes to asking questions. At some point, computers probably can interpret a medical image better , and compare it against a million other images faster than any trained human medical expert. However, that is only a starting point – human experts are way better at asking better/unique/complex questions and explore any body of knowledge and expand on it. This is why I think no expert system will eliminate doctors – they will just make the quality of medical service a doctor can provide a lot better, and reduce mistakes. In short – We get to ask the smart questions, and mostly leave finding the answers to machines.

In various forms – this phenomena will play out in every job . People who have access to smart machines that can find better answers get to make decisions faster and cheaper than others, and that is how competitive advantages will be created in the market.

So in a nutshell – differentiation in future will be based on humans who can ask better questions than they can ask today, and machines which can answer better, faster and cheaper than they can today.

Sounds pretty straightforward, but we will of course fight this every step of the way. When horse drawn fire engines were first introduced, humans used to race them on foot to prove their superiority. We know what happened after that. For many reasons – political, legal, social and economic – just because technology can be used to effectively solve a problem does not mean that it will happen fast. So in my view, there is practically very low risk of massive unemployment any time soon. But without a doubt , every job around us will evolve in a way that human value add will become all about asking better questions and technology’s value add will be about giving better answers.

Microservices – What have we learned ?

Posted on by • Posted in Uncategorized • Tagged , , , • 3 Comments

Yesterday, I shared some of my thoughts on serverless architecture  and ended up getting a lot of feedback and a lot of it went back to SOA and then logically on to Microservices. So I thought it may be worth penning some thoughts on this topic as well. Microservices are not a fad – they get used quite a bit today across big important companies, although perhaps not very consistently. Also, I think we have enough experience by now to calibrate our expectations compared to a few years ago.

pexels-photo-302895.jpeg

What is microservices architecture?

I am sure there is an official definition out there somewhere. But a simple way to describe it is as a way to design a solution as a collection of services, each of which does one thing quite well independently. Each service is its own thing – own process, implementation, language etc . The collection of such services that solve an end-to-end problem is then orchestrated with a lot of automation capabilities to form an application.

Why is this a good thing compared to a “monolith” architecture ?

Separation of concerns is not a new idea in designing systems. Microservices architecture is built on this principle. The traditional way of building an application includes a front end ( html/js) , a database ( SQL/NoSQL/File/…) and App server to handle logic. When we hear people criticizing “monolith” apps – they are usually referring to the serverside logic built as a logical whole.

Monoliths are not bad per se – they can be designed and implemented in a modular way, and can scale with the help of good design using load balancers etc. Just that when it comes to scaling, testing etc – you have to deal with the whole even though only a small part needs to change. As cloud becomes more and more the default deployment option, the flexibility to scale and change quickly becomes a bigger need than in the past. Microservices is a very good way to deal with it. Many monolith systems will co-exist with the world of Microservices .

How micro is a microservice ?

This is one area where the wisdom from actual projects tend to throw cold water on the theory and philosophy of microservices. The tendency for many engineers is to go super granular in service definition . Almost without exception, everyone I know who started with this approach has regretted it and agreed to start with fewer services and then break them into smaller chunks over time. The operational overhead is quite significant as you play with tens of services – you now have to maintain and monitor several services, and at some point there is a performance penalty for too much communication across a bunch of services that all do one little thing.

Another interesting aspect is whether your system needs to behave more in a synchronous fashion or an asynchronous fashion. When you break the system into smaller chunks, essentially you are favoring asynchronous communication between them. Then if you need to make it work in a synchronous fashion – you may question your granularity decision quickly.

What about the product/project team?

I have seen several ways in which teams are organized , and have spoken to folks who worked in such teams where I had no direct involvement. There are a few consistent themes

  1. The need to communicate frequently and freely is a make or break criteria, way more than traditional approaches. With great flexibility comes great responsibility !
  2. One big advantage that comes with microservices is that each service can be implemented with a different fit for purpose language. And each service might choose a different database for persistence. While that is all great in theory, just because you can should not translate to you should. For large projects – too many technology choices leads to diminishing returns. Choose wisely !
  3. There is practically no good way to hand off to an ops team when dev is over. Microservices forces a DevOps culture – or at least DevOps tooling for sure. Its probably a good idea to get EVERYONE in the team some training in tooling. You need different muscles for this world than dealing with a Tomcat cluster. The promise of CI/CD needs a highly trained, high performing team. I may even venture to say that best practice is to have the same team that builds the team to continue to support and enhance systems built on microservices. There are just too many moving parts to effectively transition to a completely new team.
  4. There is no substitute for experience. There are not enough highly skilled folks around , so the ones you get need to carry the weight of mentoring their less experienced colleagues. Written standards might not be enough to overcome this. A common observation is two services looking at the same business object – like a vendor object that is of interest to an accounts payables service and a compliance service – and interpreting the semantics differently. Only with experience can you catch this early and converge.

Is it truly easy to make changes compared to monoliths ?

If you are a microservices fanatic, you probably are well versed in all backward compatibility tips and tricks, and hence your answer has to be YES. I will just say that there are some cases where you wish you were working in a Monolith, especially when faced with pressing timelines. A good example is the changes many apps will need due to GDPR  . When multiple services need new functionality – you need to wrestle with the best approach to get this done. Would you create a new service that others can call ? Maybe a common library? Maybe change each service and make local changes? Each has obvious penalties. No silver bullets – decisions taken in designing the app will dictate whether you buy aspirin from Walgreens sized box or Costco sized box 🙂

What about monitoring, testing, debugging etc ?

All the overheads on these topics that comes from distributed computing are in full force here. This is one area where the difference is significantly more noticeable than in the monolith world. Many of us are fans of doing Canary releases . You should have some consistent philosophy agreed on upfront for release management. Whether we admit it explicitly or not, lean and fast deployment has a tradeoff with testing effectiveness. Essentially you are relying more on your ability of montiring your app ( via all the services and messaging frameworks and redundancies) and making quick changes vs trusting impeccable test results . This is a significant change management issue for most technology teams and especially their managers.

So is microservices architecture a safe bet for future ?

There are plenty of public success stories today of microservices implementations – and conferences and tech magazine articles and youtube videos and all that. All Major SIs have expertise in implementing them. So in general, I think the answer is YES. However, I am not sure if microservices over time will be any less frustrating than monoliths in how they evolve. I probably will get some heat from my purist friends for saying this – but perhaps one way to smoothen the journey is to start with a monolith as we have done in the past, then as it evolves – perhaps have services that call the monolith’s APIs. And as you learn more, break down the monolith to a full set of services. I am not saying this because I am a non believer – I am basing it strictly on the talent available to do full justice to a pure microservices based architecture in a mainstream way. Just because Netflix did it does not mean everyone can. In any case – the mainstream pattern in large companies any ways is to start with their old monoliths and roughly follow the approach I mentioned.

Is Serverless for you ?

Posted on by • Posted in Uncategorized • Tagged , • 2 Comments

One of the more recent architecture choices we can play with is the idea of serverless aka FaaS (Function as a Service). Thankfully, it is not hyped like say Machine Learning is. But nevertheless, it is widely misunderstood – often leading to bad design choices. I am just going to list a few questions I have been asked often (or I have asked fellow techies) , and give my point of view on those. I will try to keep this at a level where it makes sense for people who are not full time technologists .

pexels-photo-132037.jpeg

Are there really no servers?

To begin with the name serverless itself is quite misleading. Your code does not execute in vapor – it still needs servers. From a developer point of view – you choose a provider to handle a lot of things servers do (but not everything) , and you can focus on your application tasks. That is not the same as there being no servers. Its one of those things where my developer friends smile and wink, and my Ops friends roll their eyes 🙂

Is it really much simpler than other options ?

A very hard question to answer with a YES or NO. If we look back 10 years or so, it was all about service oriented architecture (SOA). Now think how many well designed services were created in the time since then ? I personally have seen way more badly designed/implemented services than good ones. My point is when you try to deconstruct an existing application into smaller ones – it often (not always) becomes more complex, not more simple. I know it is counter intuitive till you think through it, or work on an actual project. The simplicity argument is strongest in favor of FaaS when you eliminate server management from what a developer has to worry about – but even there, you need to be careful about where the server logic goes. Sometimes you implement it in client, sometimes you move it to the function, and some times you need dirty hacks to keep everything working. Simplicity is in the eye of the beholder.

It is cheaper ?

When used for the right scenarios, it is indeed cheaper. The obvious case is bursting – where once in a while you get a lot of traffic to handle. If you are not careful about designing – and especially if you don’t test with production level data, its quite possible that you may end up with a more expensive solution than having to deal with full time server management. That is hardly unique for serverless though. Poor choices have a price to pay now and/or in future.

What does the developer not have to worry about ?

For the code implemented as a function, the provider ( like AWS Lambda) takes care of making sure it will get executed when triggered. So things like horizontal scaling, fault tolerance and high availability are things you don’t need to worry about. Needless to say, you have to make sure your code is friendly for parallel execution. And it is still on your plate how the application as a whole works – Lambda etc only controls what you passed on to it. So developers can enjoy more breaks ( or officially , they can work on making their app so much better ) 🙂

Also, if your server code is in java – which is true for many cases – you save a lot of time because you don’t have to redo the whole code. You can lift and shift with low effort. Another thing to gain more coffee breaks !

Also, API gateways become a good friend for developers in mapping parameters to and from functions. This makes development and maintenance more efficient in many cases. API Gateways themselves are fairly new too – so there is that. I guess you can also consider authentication to be implemented via the gateway – but my security expert friends may not like this idea. I need to think through this one more.

What is the big difficulty for developers in FaaS world ?

If you are like me, you spend more time debugging than actually writing code. As with all distributed computing scenarios – you have a tradeoff to make here. As we introduced these paradigms in quick order, the monitoring and debugging tooling has not kept pace. So the first movers typically spend more time debugging across all the layers without useful tools and it can be quite frustrating and inefficient.

How about testing ?

Since computing is distributed, you should plan for higher quality of testing in general. Its also about setting expectations with testers and users. Since the server is not always on waiting for a request – it needs to be switched on every time, and then it stays on only for a few minutes. If you have a lot of near real time needs, implementing it as FaaS is perhaps not the first option to cross your mind. Also – a lot of dirty hacks get introduced while testing if you are not careful with designing. A common one is to keep pinging the service to keep it awake since you realized some tasks take longer. You really need a close approximation of production peaks and valleys in testing to make sure you don’t get a midnight call to debug.

isn’t FaaS stateless ?

Short answer is yes , of course. But often we need some hack to hold state – usually by using a cache or database . Some logic on session management could be in client side too.

Is AWS Lambda the only option ?

Lambda is definitely the most popular and have been in the market the longest. But a lot of big players like IBM (OpenWhisk), Miscrosoft ( Azure functions) and Google ( Google cloud functions) . So you do have choices – they all have different things they support, but probably will converge over time. I will resist the temptation to talk about standardization 🙂

So what is a good place to start ?

Serverless is a newbie in the world of architecture – so proceed with sufficient caution. Since my playground is in large enterprise space, what I have seen the most is large existing apps offloading small parts of their functionality to functions. Those companies who have embraced DevOps also consider serverless when they create new apps . At the moment, I don’t expect to see a lot of pure serverless architecture options in large enterprises. Some kind of hybrid approach is probably where we are headed. Once the tooling gets strong, I am sure we will see definite patterns emerge.