This article is part of the On Tech newsletter. You can sign up here to receive it on weekdays.

A remarkable clash between an American internet company and the world’s largest democracy takes place over the reasonable limits of freedom of speech.

The background to this are ongoing protests by farmers in India against new agricultural laws. The Indian government, citing its laws against subversion or threats to public order, requested that Twitter delete or hide more than 1,100 accounts that it says promoted violence or disseminated misinformation.

Twitter has fulfilled some of the Indian orders. However, Twitter has refused to remove accounts of journalists, activists, and anyone else whom the company says are adequately exercising their right to criticize the government.

Prime Minister Narendra Modi’s government says Twitter is breaking the law. Twitter says India is breaking its own laws. And democracy activists say technology companies like Twitter shouldn’t play along when governments pass laws that are effective in restricting freedom of expression.

There are regular disputes between internet companies and governments – both democratic and non-democratic – about whether or not agencies are breaking a country’s law. What’s unusual here is how public and high-profile the disagreement is and that India has threatened to lock up Twitter employees.

I spoke to David Kaye, a law professor at the University of California at Irvine and a former UN Special Rapporteur on Freedom of Expression, about Twitter’s decisions in India, how to reverberate, and the consequences of some tech companies setting the rules of global discourse.

Shira: Do you think Twitter is making the right call?

Kaye: Yes. In essence, Twitter is saying that it is not following instructions it deems inconsistent with Indian law and that it violates people’s human right to freedom of expression.

Under the Modi government, India did not democratically crack down on people’s right to speak out against their government. I’m not sure why Twitter picked this moment to take a stand and not two or three years ago when the company took government pressure on people posting about Kashmir.

In my role at the United Nations, I asked Twitter to explain what happened. The company didn’t answer. In a way, this week was Twitter’s response.

But Twitter defies a democratically elected government.

People shouldn’t get the impression that these companies see themselves as above the law. An important difference in India is that the order came from a government ministry – not a court. Twitter says India’s demands to freeze accounts or remove posts have not come about through the regular rule of law.

What other questions does the stalemate raise for you?

I have the same question people asked after Trump was banned from Facebook and Twitter: what about all the other countries? Will Twitter hold its own against governments in Turkey, Egypt or Saudi Arabia? And how far is Twitter ready to go? Would it risk being blocked in India?

(Twitter doesn’t automatically respond when a government – including the US – requests the company to fetch content or share user information. Here is information from Twitter on how often it responds to such requests from authorities in Saudi Arabia, Turkey. Egypt , India and the United States.)

How should we feel that some internet companies have the power to shape citizen engagement for their governments and set the limits of appropriate expression?

It is a problem. These companies have massive and largely inexplicable power. The fundamental question is: who decides what is legitimate language on these platforms?

Both internet companies and governments deserve the blame. The companies have not provided transparency about their business operations, rules and enforcement. Instead, we have perpetual cycles of decisions that appear in response to public pressure. And governments, for the most part, haven’t done the hard work to create smart regulation.

What does intelligent regulation look like?

The challenge for democratic governments is to improve the transparency of social media and to put it in a legal framework – but not to impose content rules that are abused and impair the freedom of speech of users or the rights of companies to enjoy an environment they want create for users. That is the lingering tension.

The law on digital services proposed by the European Union is a very sophisticated piece of legislation in this regard. The US is still screwing this up.

(See also Tom Friedman, the columnist for the New York Times Opinion, who writes that he advocates the European strategy for regulating the internet.)

Facebook is starting to experiment to reduce the number of political posts and materials in its newsfeed.

The reason, Mark Zuckerberg explained recently, is because people were telling Facebook that they “don’t want politics and struggles to take over their experiences”. But have you seen Facebook?

As my colleague Kevin Roose has tirelessly reported – and as an account he tweets every day – the Facebook posts with the links that generate the most reactions, shares, and comments are overt political festivals of anger. What is Facebook doing? Kevin and I talked about it:

Shira: Your analyzes haven’t shown that humans to do Do you want politics and anger on their news feeds?

Kevin: People contain a multitude, and their stated preferences often do not match their stated preferences. If a nutritionist asked me about my ideal diet, I would list healthy foods. But if you put a Big Mac in front of me, I’ll eat it. I think it’s credible that Facebook users say they don’t want politics and anger, but when their boyfriend posts a great meme by Bernie Sanders …

I also suspect that a relatively small number of people are responsible for a wide variety of interactions on Facebook – and that these super-sharers really care about politics. According to Facebook, only 6 percent of the content in the US is political. So most of the Facebook content might be instant pot recipes and baby photos.

Is the silent majority of Facebook the people who don’t want all of politics?

Possibly! Or people just aren’t honest (or don’t know) what they really want. I think we’ll find out from this Facebook review.

Should Facebook give us more about what we actually click on or what we do say we want to click?

Facebook, like basically all social media apps, is designed to give us more of what we like. It’s very lucrative, but it didn’t go so well for democracy.

What if a social network was designed to feed our aspired selves rather than our lizard brain impulses? Would we like it more Or would we miss the drama and the fighting?

  • America’s unofficial unemployment hotline: During the pandemic, more Americans turned to a Reddit forum for advice on navigating the confusing unemployment insurance systems, writes my colleague Ella Koeze. It is also a place to feel sorry for others who are going through the same difficult circumstances.

  • The algorithm includes void: Companies that make specialty clothing for people with disabilities state that Facebook’s automated systems routinely reject advertising and listings for their products. The problem, writes my colleague Vanessa Friedman, is that computers are nuanced and Facebook’s systems often label adaptive clothing as advertising for medical devices or as “adult content,” which is against company rules.

  • The digital divide in the church: Wired writes about the churches that thrived as worship and mostly went online during the pandemic – and the struggles of others who didn’t have the resources to go virtual.

Eight-year-old Leo wrote a stern letter to his NPR station because he had no further broadcasts on dinosaurs. So NPR asked Leo to interview a dinosaur expert. It was wonderful.

We want to hear from you. Tell us what you think of this newsletter and what else you would like us to explore. You can reach us at ontech@nytimes.com.

If you do not have this newsletter in your inbox yet, please register here.