Carmina Burana

It’s 21st century and needless to say we love O Fortuna. Carl Orff set it off to that tempestuous music over 100 years ago, and since then Fortuna has been everywhere. It’s in films, video games. Commercials. You name it. Recently, I decided to explore more of Burana. There are 26 poems set to music by Orff and all of them are as good as Fortuna, if not better. My favourite for now is In Taberna: Estuans interius. Burning inside, totally!

Carmina Burana

Never judge a book by its first 100 pages


The opening line of Arundhati Roy’s Ministry of Utmost Happiness is the most exquisite sentence I have read. It’s magical in a way only Roy’s sentences can be. She writes:

At magic hour, when the sun is gone but the light has not, armies of flying foxes unhinge themselves from the Banyan trees in the old graveyard and drift across the city like smoke.


But from there it kind of goes downhill. Anjum and her troupe don’t seem real enough. The old Delhi, meanwhile, feels too real, with sentences so edgy that you could feel them scrap against your hand if you tried. I find it trite, even unimaginative, which is a rather peculiar word to use for anything that Roy writes.

It lasts until around 100 pages. And then the book transforms itself. The Landowner enters the picture. And Tilo comes on board. Suddenly, Roy is talking not about old Delhi but people like Tilo and Musa, and it is with these fictional people, no doubt more familiar to her than Anjum and others, that she again writes the exquisite lines only she can write. These lines may occasionally pay the homage to that world of Macondo but they are characteristically Roy.

Just like the God of Small Things, the Ministry of Utmost Happiness is not an easy read. And it seems that it might have been quite difficult to write for Roy because it is a very political book, talking of the time and moments about which she obviously feels strongly. Yet, such is her mastery that her fury rarely breaks the surface that her words cast. She constructs her characters with a tenderness that makes this an achingly beautiful book about people, a few extraordinary and a few ordinary people. And then there are the lines like this:

The steel bubble floated on, past shanty towns and industrial swamps where the air was a pale mauve haze, past railway tracks packed thick with trash and lined with slums. Finally they arrived at their destination. The Edge. Where the countryside was trying, quickly, clumsily and tragically, to turn itself into the city.

Beautiful. And a proof that Roy, when she gets going, writes some of the best lines among contemporary writers.

Never judge a book by its first 100 pages


Kazuo Ishiguro writes in the Remains of the Day:

Perhaps it is indeed time I began to look at this whole matter of bantering more enthusiastically. After all, when one thinks about it, it is not such a foolish thing to indulge in – particularly if it is the case that in bantering lies the key to human warmth.



Comscore numbers, top tech sites in India and how to read these numbers

Marketing comscore tech sites

Important: There is no list of top tech sites in India here. I am not putting it out because I am not sure of Comscore T&C that govern the use of this data. Although some data, which I have seen publicly, has been used here.

Numbers. This one word makes the world tick. But in particular “numbers” are of great significance to the daily lives of web journalists. The numbers define a web journalist. Sad, but true. Nothing else matters. So we measure our numbers, meticulously, minute by minute. Hour by hour. Day by day. And then we compare those numbers with all sorts of other numbers that we measured earlier. We want to see where the numbers are moving. We want to see from where they are coming. We want to know what these numbers are doing when they are in front of us inside Google Analytics or Chartbeat.

The problem with the numbers that we see is they are exactly that: numbers. Once they are on the screen, calculated and arranged in tables that can be exported to Excel, they are without context. They don’t tell much, and whatever little they tell is banal at best and misleading at worst. They are not even accurate numbers. Despite all the hullabaloo about numbers, the reality of the web journalism is that there is no accurate way to measure numbers.

I can go on and on about the readership numbers, what’s wrong with them, what can be fixed and what can’t, where does news fall in all this, but that is not the point. For now, I am just talking about Comscore, the monthly numbers that it churns out for technology news sites in India and what these numbers say. Unless you are part of newsroom, chances are you will have no idea. They don’t always mean what you see and people who collect them don’t really explain what they mean.

So, let me try, although I am yet to figure it out fully.

Before we move to the number game, a couple of points:

— The list of top tech sites in India that you may see occasionally contains only India sites. There are some global sites like The Verge in the full Comscore list for India, though most of the sites here (definitely the top 10) have better numbers than the global ones among Indian readers (thank you Google).

— It is possible that the lists may have a site or two missing. There are many unknown websites with great traffic in the top sites in India. I can speculate from where these sites get the traffic. But it is often easy to miss them because no one has heard of them, or seen them anywhere on social media, in Google News or Search. There is a site called in top 15!


Now, about the numbers. On the face of it, the lists that you see look fairly standard. You have the sites, with readership numbers for them listed in a nice chronological way. The full picture is, however, not so starkly black & white. But before I talk of that, a couple of additional points:

1- On many occasions (but not always) the numbers are for a group of properties. For example:

Different properties


2- The numbers are just numbers. They don’t tell how many people are actually reading a story. For example, in July India Today Tech, according to Comscore, had marginally fewer readers compared to another tech site. But then our readers spent significantly more time reading the article. If you add to the mix the number of pages a website has, or the new number of pages/stories, you can even calculate the average time a reader is spending on each article. But that would be going too deep, probably. Losing the point in numbers you can say.

Total time spend


3- Finally, there are bugs. See this data. The right most figure — the green one — shows that an average users is spending over two days on the app/website doing something. This is most likely an anomaly.

Time bug


Several other points that needs to be factored in while you look at the list of top tech sites in India.

— There are many ways to get traffic on the web. Organic, inorganic and so on and so forth. Some tech sites in India get majority of traffic from web search. Some get direct traffic. Some from social media. Some from sources that are so mysterious that not even GA and Comscore can figure out their identity. Some pay for the traffic. Some pray for it. Some get their traffic from a mix of it all.

— For tech sites, however, the major traffic driver for unique visitors is web search. Or the paid one, either paid for directly or paid through social media. Prayers don’t help much.

— For most sites that have specs pages and automated comparisons, these pages are a major traffic driver because Google loves these pages. When you see a website on top in the list of top tech sites in India, also see check for generic content pages produced in a content farm. If they have them, you can safely assume that most — and by most I really mean most — of their traffic is coming from these pages. Similarly, some news sites, which are also part e-com and comparison sites, have sort of inflated numbers for obvious reasons.

— Although, I must add that I do find that the specs pages useful to readers. They serve a purpose, it’s just that they are not news.

— For a lot of tech sites in India, traffic is either made up of paid clicks or borrowed numbers (which is the same thing as paid clicks). “Paid clicks” is easy to understand phenomenon. The borrowed traffic is when you club your site with a number of other pages, even third party pages, in a bid to boost the numbers.

— If you ask me about the top 10 English tech sites in India, and the numbers we see in Comscore lists that float around, I feel confident about the numbers of only three sites: India Today Tech, FirstPost Tech and Indian Express Tech. Although I believe NDTV Gadgets too has significant News traffic (most likely highest among all tech sites), it also probably has a lot of other traffic in its total numbers.

— The reason is simple: I know for sure (and I am definitely a little biased here) that India Today Tech relies only on news, views, opinions, features for its traffic. We don’t do any mumbo-jumbo. We don’t have specs pages. We don’t even have galleries where each photo is a separate click (silly not having it, if you ask me).

We even made a mistake of not segregating the Google AMP traffic for the Indian Today Tech. Definitely silly, if you ask me.

All our traffic comes from actual pages with stories on them, and we see that people spend time reading our stuff. I am assuming, given whatever little I can glean from available data, that it is same for 2 or 3 other tech sites in the “top lists”.

— Comscore, given the way it tracks numbers (only local clicks, verified, volunteers and what not) can be off the mark in a significant way compared to what web journalists see in Google Analytics. You can read more about Comscore tracking here. Comscore throws up “verified” numbers that are low compared to GA. But then I also believe that Comscore is consistent enough to be a benchmarking score for the industry. It is same for everyone. And GA too has its own problems, with bot clicks (paid traffic) and ghost visits. As I said earlier, there is no way to accurately track web traffic.

— But if not Comscore then what? I honestly don’t know for you. For me, GA (which is only available internally) is good enough tool when also combined with visit time from readers. This can show me what stories people are reading and how engaged they are with that story. For people who often ask me about the site traffics (read PR people) who don’t have access to a site’s GA, it is little difficult to figure out how a news site or even respective stories are doing.

— Engagement on social media is one way to figure out real traffic to a news site. But then again, that too has a lot of this and that, including the paid stuff.

— If you ask me, the best way to see how many people a story has reached remains Google News and Search. Yes, Google can also be gamed, including for Google News and specially for Google Search. But that is a different matter.  If your story does end up on Google, it is mostly read by actual readers.

Every story that gets featured in the Google News highlights is, on average, doing manifold better than stories that are not in Google, irrespective of the site. The only exception to that could be (and this is COULD BE) the TOI stories because TOI seemingly has very impressive direct traffic. If a story also appears on top in organic Google Search, that too means it is reaching more people.

That is all, for now. Lot of words. But I wanted to put this out, given how mysterious is the world of web traffic measurement and given that fact there are top tech site lists that are floating around on the web. As I said earlier, it’s not easy tracking the web traffic numbers. And rarely they offer a full picture, although combined with a couple of other factors, they can give you a fair idea of where a site stands. But ideally don’t take the numbers on their face value. There is always more going on behind those numbers.

Comscore numbers, top tech sites in India and how to read these numbers

Homo Deus vs Homo Sapiens

Homo Deus

Homo Sapiens A Brief History was more of a pop culture summary of humans so far. But the Homo Deus by Yuval Harari is a different beast. It is more interesting, although with the same sweeping generalisations. It feels closer, more insightful, also in parts because I read it without the 20:20 hindsight that accompanies the books on history. This one is about future. And not necessarily a glorious future that awaits humans. It’s about the future where algorithms are going to rule.

In the twenty-first century we might witness the creation of a massive new unworking class: people devoid of any economic, political or even artistic value, who contribute nothing to the prosperity, power and glory of society. This ‘useless class’ will not be merely unemployed – it will be unemployable.

Homo Deus vs Homo Sapiens