As humans we have a tendency to categorise things, we like to organise our houses, offices, cities etc. We like to put things in cabinets or drawers or shelves. But this habit transcends physical space to people - we like to categorise people; into Christians, Hindus, Muslims, Jews ... Into Leftists, Liberals, Rightists, Fascists, Apologists, Colonialists, Rascist ... Into Indians, Russians, Americans (whatever race they represent!), Blacks, Whites, Caucasians, Asians ... Well even men, women and others!

And then, we categorise ourselves also - even if we don't like what we think we are - we'd like to think of ourselves also as belonging to one tribe or sect or thought.

But the facts stare us in the face - biologicaly you are not the same person you were a moment ago, cells in your body are constantly changing, thoughts in your mind are constantly evolving, your beliefs are constantly changing!

People too cannot be categorised - someone who is liberal about political thought can be a conservative on personal relations, a politician who propounds capitalism may prefer Banks to be nationalised ...

Gandhi taught us not to think of people as masses - as Hindus or Muslims or British or Indian. But the more fundamental learning is not to categorise ourselves, not to box our own identity into one or the other. Great souls can transcend what they themselves are or were - Valmiki gained sainthood from a past as a dacoit!

True freedom starts from thyself, stop bracketing yourself, then stop yourself from bracketing others, then stop others from bracketing others - true freedom comes by freeing oneself from our own limitations.

Most multinational companies seem to have a problem with the data localisation requirements, may be because most of these companies are incorporated overseas and would prefer to follow the legal jurisdictions of countries where they are incorporated. Countries such as the United States or even the European Union have personal data processing laws which may necessitate that data processed by companies incorporated in their territories should preferably be stored within their own territories; while the HIPAA or GDPR does not explicitly require that such data is kept in Europe but some of the other clauses of GDPR will be more convenient to comply with if the data was co-located within Europe.

However, even for MNCs who have a significant local presence with local subsidiaries incorporated/registered in India, the data localisation regulation presents a cost implication. Take for example payment processors such as MasterCard or Visa - currently, they host their servers overseas and with the data localisation requirement they will have to set up additional infrastructure in India store a copy of data here. Depending on how voluminous this data is, the additional data store would require significant capital expenditure and maintaining it would incur significant operational costs as well. Further, this would also make this local data store come under the jurisdiction of Indian authorities who may request for access to the data under various other laws such as the IPC [Indian Penal Code] and regulations by RBI and SEBI - complying with such requests will require setting up a team to manage and service the requests - all adding up to the Operational expenditure.

However, as argued by Sachin Bansal and Manish Sabharwal in this piece, these costs notwithstanding, MNCs should fall in line and comply with the regulations given that the regulations are in favour of Indian general public i.e. the MNCs' customers. They also counter argue, that by not mandating that this data should be kept only in India and by allowing this data to be maintained overseas, the regulations have been fair and they do not present significant operational challenges to MNCs or even Indian companies doing businesses overseas.

But if you looked that this clause as a dispassionate external observer, maybe Norwegian national for example, you would realise that this clause which allows the primary data to be kept overseas and only a copy to be maintained within the Indian territory, is actually a very convenient way for the governments and regulators to say that "I don't care what happens to the data of my citizens wherever else it is kept, all I want is to be able to snoop into the personal information of my citizens as and when I want according to the laws defined in my country".

By allowing data to be stored overseas, without any restrictions, the regulators are providing no cover to snooping or leakage of this data from an overseas territory. For example, Indian citizens' data kept on servers of MasterCard, Visa or American Express in their US data centres is open to being snooped by the US government. At the same time, the same data can be accessed by the Indian government or regulators from the local data stores. In effect, there is absolutely no protection for citizens and all provisions of these regulations only facilitate the regulator or government while allowing (or coercing?) businesses to be complicit participants in this game of personal data espionage.

To be fair, I must also add that it isn't trivial for personal data to be accessed by government agencies anywhere in the world. Most personal data such as credit card numbers is kept in an encrypted form and is not visible in plain sight. A hacking attempt on a credit card database, for example, is unlikely to result in stealing of this data. Nevertheless, a government agency with sufficient privileges can get access to this data by using local laws and forcing the company to decrypt the data and provide them a copy. So the risk of data snooping by government is still very real, even though, the actual mechanisms may be complex.

A decade ago one would have called such allegations as baseless but after the revelations by Edward Snowden about the Prism project by the US government, this is a reality and can no longer be swept under the carpet as a figment of someone's imagination.

So what would be the solution one would ask, clearly for a developing nation like India it will not serve well to close its own data economy by mandating that any service provider should store data ONLY within the Indian territory. In the larger interest of a globalised world also, this would not be a welcome restriction. But opening up data to all world governments to facilitate global economy is taking a very narrow view of how data protection can be implemented for citizens of India. Solutions can be sought if one has the will to protect citizen data and not an intention of only allowing access to their information for purpose of government or national interest.

A simple solution would be that storing data is allowed in any part of the world, as long as all personal data is encrypted 'at source' [i.e. when it is collected*] and the encryption key used to encrypt data should be stored exclusively in Indian territory. This way if any government - foreign for Indian - wanted to get access to this data they would need to have access to the key which is kept in India. To access this key they would need to file proper claims under local Indian laws and establish their need to access this data for valid purposes. This mechanism can not only protect personal information from being snooped into by foreign governments but can also be a very effective way of preventing unauthorised or unscrupulous elements within the Indian government machinery from being able to access this data at will. Police or Income Tax officers will not be able to access personal information of citizens just because they happen to represent the government unless they have proper authorization.

The challenge, however, seems to be that regulators and governments are only looking at one side of the picture which is how do they ensure national or government interest rather than focus on protection of citizen personal data.

While I am guilty of not having represented this view to the Justice Srikrishna Committee, I plan to submit this piece of text to the 'Ministry Of Electronics And Information Technology' which will be presenting this bill in the Indian Parliament. I do hope that the government reconsiders data localisation requirements to ensure that the law is front-loaded with considerations of citizens interest rather than only protecting the interests of the government or the nation-state.

* An example of data encryption at source is passwords - when you choose a new password, the actual password is not sent to the website, but rather your password is encrypted by your browser itself and an encrypted 'hash' of the password is sent to the website and the same hash is stored in the servers. Your actual password is not stored by the website.  

I just read this news article in reaction to the Government's decision to allow courts to award the death penalty to those convicted of raping children up to 12 years of age. The government's decision comes in the wake of nationwide protests against child rape.

But as the article states, and I quote - "fear that with the death penalty, most people will not report child rapes, as in most cases the accused are family members. The conviction rate will come down further".

I have been myself thinking about this ever since I have heard of the law being passed. The problem in India has been the implementation of a law and not having a stringent penalty. I quote again - "We already have the death penalty for several offences and that has not led to any deterrence. If we are looking to create a deterrent, then we have to create it where it works."

The issue is not that child rape or any crime doesn't have stringent punishment but that conviction is low and justice takes so long to deliver that it's often akin to being denied. This is evident in several cases involving politicians and celebrities - Salman Khan was accused of killing people sleeping on a footpath by running their car over them. The proceedings took so long that Ravindra Patil, the key witness (who was the Police protector for Salman on that fateful night), died of severe ailments [and under mysterious circumstances] before the decision was arrived at. The case was won by Khan, but not without the fact that the key witness was NOT present in the courtroom to present facts. The case against Salman for the killing of the black buck, a protected wildlife species, was also decided 20 years after the incident.

I am not relating the case of Salman here as an unrelated rant, but rather to prove how judicial system is faced with severe delays and any form of stringent punishment will not help matters anyway if the sorry state of awarding justice continues.

We need to think of rather strengthening the system, making sure that police officers can discharge their duties without political pressure, the judicial system is also kept corruption free (which is otherwise ensured by the Constitutional provisions of judicial independence) but more importantly speedy and efficient in discharging justice.

On this count, the less noticed aspect of the government's decision is laudable - in addition to allowing capital punishment, the government has set the time limit for investigation as well as completion of the trial of all cases of rape as two months and prescribed the limit for disposal of appeals to six months.

As citizens, we also need to become more responsible - when you bribe the traffic constable with a hundred rupee note instead of paying the fine, you circumvent the law and strengthen the loop which makes the police system more amenable for manipulation than astute enforcement of the law. You do the same when you put up that Facebook post of Salman 'Bhai' being the kind-hearted human being and not an undertrial accused.

The judicial system does not work in isolation - as several cases in past have shown, public opinion and general behaviour of public play a major unwritten role in how judiciary looks at cases. If the public, in general, is content with delayed and subverted justice, leniency to politicians and celebrities - the judicial system will mould itself in the same general frame. But if we stop being malleable about justice based on our opinion and infatuations, the judicial system will gain strength to deliver justice faster and more efficiently.

Let us not just rant for stringent punishment, but rather a more efficient judicial system!
.

Image Credits: Michael Coghlan via Flickr

There have been a lot of controversies of late - a movie which was claimed to have twisted historical facts, a caste-based protest, a politically influenced crime or the issue of sharing river water between two states. In all these cases, facts elude the public and the vote always remains is pretty divided and vitriolic because there are always two sides of every story.

It is no-one's guess that a filmmaker may have apparently taken a lot of liberties in historical depiction or that a particular law may have been misused or water or financial devolution could have been fairer or anything else. However, protestors have often gone wild with threats to kill or even cause damage to public and private property.

Sane voices argue that these non-issues keep people and media away from discussing real issues like the state of the economy, the rise of crony capitalism, deteriorating finances of the middle class or farmers committing suicide. So why is it that that large hordes of people keep thronging the streets. Why is it that the public today is paying attention to non-issues, why is it that "youth leaders" are spewing venom or announcing bounties on actors and directors or threatening bloodshed in the name of a river - when there are many other major issues they could pick up even if it were to win elections.

The answer probably lies in the role models which the society has conjured up to the masses - especially young people's minds. [Sidenote: It is usually the Young who control what gets discussed in public sphere, but due to the majority of the youth demographic, this is especially true of India].

Those of us who were (little) early to the Internet party would find the current forwarding and fake news culture on Whatsapp / Facebook similar to what ranting & flaming on newsgroups (e-groups) used to be in the early 2000s.

However, newsgroups, because their access was to a limited few, were also places where ideas were born, partnerships formed & organisations were created. But I find Whatsapp (or other forms of messaging apps like Discord or Telegram) not being so.

One probable reason for this, I feel, is the lack of long-form prose which email as a medium supported and which messaging as a medium shuns. Long form writing forces you to think deeper, engage in self-correction between various coordinated (or uncoordinated) parts of your own thinking and also help the reader absorb the context and message both.

To be sure, the long form thinking was not mandatory in an email, and so, there was reactionism, flaming and shaming on newsgroups as well, but that was also accompanied on the side by deep thinking and logical reasoning. Certain people or certain e-groups built a culture of not reacting by moderation which was a feature in these e-group systems.

Messaging apps of today, on the other hand, do not support any such features, you can't moderate even if you wanted to, the interface does not encourage writing or reading the long-form and it stresses reacting, sharing or instant response rather than careful, well thought and calibrated response.

Just to take a measure, how many of you reading this long-form post of mine have had the patience to reach this point of this text. Instead, if I would have sent this as an email, you would have.

An ex-Facebook exec, Chamath Palihapitiya, felt guilty of creating a time-wasting monster in Facebook. Mark Zuckerberg also feels guilty at times of fake news, and fake news I argue is not a phenomenon of its own, but rather its spread is a result of this instant reaction that messaging and Social Media apps encourage.

Its high time technology companies realised that their design decisions can't just be slaves to what 'users want' or what will make the platform more 'viral'. Facebook is viral enough, WhatsApp is ubiquitous enough. They need to now design for the more conscious audience or to improve the conscience of and thinking abilities of its audience. I hope Facebook and Whatsapp teams are paying attention to this.

Epilogue - results of survey

I had first posted this spiel on my Facebook wall and instructed people who read till the end to click the 'Love' icon and those who stopped after the first paragraph (which is previewed on their Fb timeline) to click the 'Sad' sign. The quick one day survey got 10 reactions - 6 loves, 4 Likes but no 'Sad' reactions. Mostly those who hit the like, just browsed past the first few lines and hit the 'Like' without even referring to the instructions - which all the more illustrates that we really don't pay 'attention' to the content of Social Media posts and react based on very flimsy reading of the content.

Title Image Credits Álvaro Ibáñez

Stepahnie Hurlburt, a Graphics Engineer & Entrepreneur has tweeted a very useful checklist for software developers releasing Open Source products and tools to help in their adoption. I think the checklist can be used by all product developers / marketers to improve adoption.


Here goes the list, you can also read it on this Tweet Thread here or go to Stephanie's Twitter handle here.

“Why isn’t someone using my software product or open source tool? It’s good!”

A checklist for you:

1. Have you described what it is and what the benefits are in a way a non-developer can understand?
2. If someone Googles to try to learn more about it, is this description easy to find?
3. Is this description easily skimmable? If someone looks at it for 6 seconds can they be convinced?
4. Do you compare your tool to other similar tools so people feel educated about pros/cons of yours?
5. If performance matters, do you have easy-to-skim benchmarks that include comparing it to other tools?
6. Do you have a demo? If it’s open source, are there well documented and easy to build samples/demos?
7. Have you spoken or written about your tool? Twitter, a tech blog, online communities, meetups, conferences, podcasts, etc
8. Have you pitched it in 1:1 conversations with developers who might be interested?
9. Have you talked with developers who may be interested to learn about why they haven’t tried it or don’t use it?

Key takeaways:

• Describe your tool in an easy to understand way
• Do not underestimate the power of informational interviews

Follow up Tweets:

• And if the tool is meant to be used by devs, still describe your tool in a way a non-developer can understand
• To learn more, read up on marketing/sales and talk to professionals. It’s a fascinating field and I have a huge amount of respect for it.

Checklist items added by others

1. Is there a list of active users so I can see if co’s I might trust have decided to use it...
2. Does it look like it’s got a viable community of contributors that might continue to develop it if main dev moves on.
3. Is the main dev succeeding at turning it into a viable biz so they don’t burn out.
4. Is a me-too library... too many options in a space means that many will turn into abandoned code. Image loading c to other lang binding sys

Jim Collins has been researching and writing about the Flywheel as a value of enterprises. For those who may not know, here his most famous illustration - the Amazon Flywheel.

Lower prices led to more customer visits. More customers increased the volume of sales and attracted more commission-paying third-party sellers to the site. That allowed Amazon to get more out of fixed costs like the fulfillment centers and the servers needed to run the website. This greater efficiency then enabled it to lower prices further.

In another article, Jim writes - "those who drive companies into decline often abandon the big thing they already have, grasping instead for a new big thing, then another and another, falling into a doom loop of chronic inconsistency"

While not written in the article above, but my experience says that 'Consulting / System Integrator Services' is plagued by this particular phenomenon. Consulting firms are forced to move from one "next big thing" to the other constantly. This is not a fault of the company or the management but just the way services business is - customers call consultants or integrators when they don't have skills to do something internally. And since talent is usually available for most 'ingrained' work, consultants end up doing only - the 'next big thing'.

Some service businesses like actuarial, legal or accounting seem to be insulated from the 'skills' issue, but they too more often than not get pulled into managing change for their clients. [Ex. accountants need to help manage changes in taxation on an almost constant basis, lawyers have to help manage changing laws, insurance advisors often changing valuation methods and so on.]

Hence (professional) services firms never get a chance to scale "without" adding more resources to the revenue engine. Key is "without adding resources": resources here being people and their skills and hence paraphernalia like a management structure to manage people, office spaces, managing people churn etc

On the contrary, a product business has its flywheel - it can keep running faster and faster every week, month, year with the same effort [read: resources] you put in. What traction you build in the first week just gets reinforced with the second week's effort and so on. Not so in services where each relationship, each skill, and then each project has to gain momentum from scratch - they don't feed into a flywheel.

And so if you have 1 'Engagement Manager' managing 3 clients, you will need 3 to manage 9 relationships. In contrast, a team of engineers or designers can build version 1 of a product in year 1 and then add many more features to it in year 2 without losing the value of work done in year 1.

The last point - in "almost" all services, loss of human resources means 100% loss of IP. If a consultant leaves you, you have to recruit another as good or train another to become as good. But with products, the IP is (partially) captured in the product and so if a designer or engineer leaves, s/he can't take away the IP that got locked in the product. Though there will still be some loss of IP (because they'd take expertise in architecture, understanding the product etc), the loss would be less than 100% - how much less would vary from product to product.

This is one reason why it is harder to find too many billionaires who own services companies - the exceptions being the founders of IT Services Giants viz. TCS, Infosys, Wipro, Cognizant, Accenture etc. Few of them have hacked the services model to productise their service; for example Michael Dell who converted 'assembly of computers' into 'manufacturing computers'.

Another class of services business which have bucked this challenge are restaurant chains where the service of providing 'food' is hacked into providing experiences. Branding and consistent ambience play a big role in helping such businesses make the transition from 'pure services' to 'branded services'. Big 4 consulting firms also have done this in professional services as have some legal firms and 'real estate services' firms like JLL and CBRE. The existence of these exceptions, however, goes on to prove the rule - most competitors of these services 'biggies' are much smaller and local players. They play on cost and not value and hence have another reason why they can never scale.

On the contrary, there are many 'product' players who have scaled in past with or without VC funding. This could be one of the major filters that wannabe entrepreneurs can use to select business ideas on which to stake their lives.

I have mentioned in past that "If the Indian constitution is a copy of the British and American systems, it’s the most intelligent and well customised copy ever made"

STV/IRV used in single-seat elections with more than two candidates. Instead of voting only for a single candidate, voters in IRV elections can rank the candidates in order of preference. Ballots are initially counted for each elector's top choice. If a candidate secures more than half of these votes, that candidate wins. Otherwise, the candidate in last place is eliminated and removed from consideration. The top remaining choices on all the ballots are then counted again.

This process repeats until one candidate is the top remaining choice of a majority of the voters. When the field is reduced to two, it has become an "instant runoff" that allows a comparison of the top two candidates head-to-head. IRV has the effect of avoiding split votes when multiple candidates earn support from like-minded voters

... the solution [to legislative gridlock] is not to scrap the checks and balances ... but rather to adopt new voting rules (like instant-runoff) [to] facilitate greater flexibility within political parties and hopefully the creation of viable new ones

Influence from Medieval Indian governance