The Twitter Paradox: How A Platform Designed For Free Speech Enables Internet Trolls

TERRY GROSS, HOST:

This is FRESH AIR. I'm Terry Gross. Twitter is one of the stars of this election cycle. Everyone is tweeting their reactions to debates and linking to articles. Donald Trump tweets provocative messages even in the middle of the night. The paradox of Twitter is that it's become a great platform for free speech, enabling the voices of individuals and pro-democracy movements to be heard and to connect with others.

But it's also given a platform to trolls, people who write tweets that insult, harass and threaten their targets. Last week, the Anti-Defamation League issued a report detailing the widespread use of Twitter to harass journalists with anti-Semitic messages during the 2016 campaign. Twitter executives are wrestling with how to deal with the trolls without undermining Twitter's commitment to free speech.

A little later in the show I'll talk about this with BuzzFeed tech reporter Charlie Warzel, who's written a series of articles about harassment on Twitter. But first, we're going to hear from a writer who's received countless harassing and threatening tweets. David French writes for the conservative publication National Review and is a senior fellow in its think tank arm, the National Review Institute. He has been and is still being trolled for his critical comments about Ann Coulter, the alt-right right and Donald Trump.

He's part of the Never Trump movement. And Bill Kristol even suggested that French consider running as a conservative alternative to Trump early in the campaign.

David French, welcome to FRESH AIR. Tell us first what set off the attack by trolls.

DAVID FRENCH: Well, first, thanks very much for having me. I've been listening to FRESH AIR for many years. It's a privilege to be here, although not under these circumstances. What set off the attack by trolls was really something pretty simple.

In September or mid-September of last year, I had noticed that Ann Coulter, who is a very prominent supporter of Donald Trump, was tweeting out a lot of thoughts that are common to this part of the right - even though I hate to even call them part of the right, they call themselves part of the right - called the alt-right that were explicitly white nationalist in their tone and tenor. And so I wrote in our group blog on National Review called The Corner that Ann Coulter was deliberately appealing to these people. And I - and basically and politely said this is something that's inexcusable and it has no place in the conservative movement.

And then I had no idea what was about to happen next. My Twitter feed basically exploded. I have - did not have that many followers - in the thousands, certainly not like the more prominent folks in politics, but it was unbelievable. I began to see images, for example, of my youngest daughter, who we adopted from Ethiopia many years ago, who at the time was 7 years old - images of her in a gas chamber with a - Donald Trump in an SS uniform about to push the button to kill her. I saw images of her Photoshopped or, you know, artist's rendering of her face in slave fields.

I was called all manner of unbelievable names, which is kind of par for the course for Twitter, but among them was this term that has gained currency in recent years called cuckservatives. Cuckservative is somebody who's been cuckolded by the establishment, by the liberal elite. And then people began to refer to my wife as having sex with black men when I was deployed to Iraq in 2007, 2008. And it just descended from there. And that's a side of Twitter I know that others had experienced, but I had certainly never experienced it before. And then it just got worse.

GROSS: So this was all because you criticized Ann Coulter?

FRENCH: Yes, it was because I - not just criticized Ann Coulter. I mean, that happens all the time. I mean, she's a frequent target of criticism. It's because I criticized this group called the alt-right. It's - and for, you know, those who don't know what the alt-right is, it's a collection of mostly younger people who are rebelling against mainstream conservatism, rebelling against progressive liberalism and have really began to adopt white nationalism, white identity politics. They will forward around Nazi memes. They're very definitely anti-Semitic. They have launched an avalanche of anti-Semitic hate against a number of Jewish journalists. And so this small group is very, very, very vocal online. It has a disproportionate impact online that far outweighs its actual numbers.

GROSS: So you start getting, you know, these trolling tweets. Did they start multiplying? Did it start small and keep getting bigger?

FRENCH: Yeah, it really did multiply. So the next thing was, you know, I'm a writer, my wife is a writer. So we both wrote a follow-up piece - or I wrote a follow-up piece; she wrote her own piece. She put it in The Washington Post. I put mine in National Review that highlighted just some of these terrible tweets and basically said, look, we're not going to be intimidated by this, but we want to expose it. We want to show people what's going on out there if you criticize the people - some of the people who are supporting Donald Trump or criticize some of Donald Trump's allies, particularly the alt-right.

Well, then it just blew up after that. The number of tweets multiplied exponentially. The same images kept coming again and again, including pornographic images. My wife's blog, somebody - several people found her blog on Patheos, which is a religious website, and filled the comments section of her blog with images of African-American men mainly being murdered, committing suicide. It was horrible. It was horrible.

So I was able to - fortunately, she was out of town and offline when those images started to roll in, and I was able to scrub them before she could see them. But it would be horribly traumatizing to even see these images. And then - but some of my neighbors saw them who read what I write and what my wife writes, and they were traumatized by it. They were fearful for their own safety. And that was really not the beginning of the end of the harassment we received, but the end of the beginning after we began scrubbing and blocking and reporting all of these images.

GROSS: So I'm just wondering, do you think that all the trolling was a result of you criticizing the alt-right, or do you think it was also because you're critical of Donald Trump? You became part of the Never Trump movement.

FRENCH: Well, it's both-and. I mean, it initially began - I'd always criticized Donald Trump throughout the primary, but I had never received that. It was when you began to make the linkage between Donald Trump and the alt-right that they really began to get stronger and stronger. And then when you - the more you are against Donald Trump and the more that people feel like people are reading you or listening to what you have to say, then the alt-right comes down on you even harder because they really are sort of like the online shock troops of the Trump movement. I mean, they are the most aggressive, the most vicious, the most threatening, the most targeted.

And again, you know, I wouldn't write about this if this had only happened to me. You know, I would have dealt with this in private. I would have dealt with this with law enforcement, tried to deal with this with Twitter. But then I began to hear story after story after story, I mean, multiple people who've taken on Donald Trump, in particular his appeals to white nationalism, in particular his racially-coded appeals. Any time you call that out, any time you criticized that, that part of Trumpism that intersects with the alt-right, they come at you viciously, especially if you're a journalist, especially if you're a writer.

GROSS: Did the viciousness that you experienced on Twitter cross over into the larger world?

FRENCH: Oh, yeah, absolutely. So, you know, it crossed over into my wife's blog, as I just mentioned. And then there was a brief time where I considered mounting an independent presidential run against Donald Trump. I was talking to Bill Kristol about that, and that leaked out into the press. So there was a whole new level of Twitter attacks that occurred then. But then you - we got an emailed threat that was very serious.

And then most recently - and this is alarming and jolting - my wife was on a phone call with her father. They were talking about something unrelated to the alt-right. They're just having a phone conversation, and someone literally broke into the line and began talking about Donald Trump, spewing profanity.

We still don't know how this happened. We've contacted law enforcement, they're stumped as to how to how it happened. But it did lead to a pretty frantic and worried search in my father-in-law's home as to whether someone had been in his house and had literally picked up the other end of the phone. So it has crossed over well beyond Twitter. It has crossed over well beyond just comments on blogs and intruded on to, you know, what people would call real life - your phone calls, your emails. So my wife, you know, she carries a handgun now at all times. She's gotten more handgun training. We're much more careful about the security around our home. And again, we're not the only ones. There are other journalists who've spent tons of money on security systems for the homes. They have gotten handguns themselves after never carrying them before in their lives.

It really is a campaign of fear and intimidation against journalists across the political spectrum who go against Donald Trump, and in particular go after Donald Trump's connection to the alt-right.

GROSS: How did you explain this to your daughter, who I think you said is 7 years old?

FRENCH: (Laughter) You know, we - she's now 8, so - she was 7 when this started. So this has been going for more than a year now. So we have sheltered her from this completely. She's an African-American girl in a white family with - other two siblings are white and her parents are white. And she's grown up in the South, and there's going to be a whole lot of sensitivities over the years.

And already our sensitivities and explaining and her understanding how she - how and why she's different from her siblings, why her skin color is different, how people interact with us as a family differently, these are things that all kinds of multi-racial adoptive families work through. And it's a very delicate and sensitive thing. So the last thing that we want to do is to introduce this element into her life. So while my older two children, who are 17 and 15, know that their younger sister has been subjected to this, she does not. And we will try to keep this from her as long as we possibly can. I mean, you know, she's still too young to listen to FRESH AIR podcasts.

(LAUGHTER)

FRENCH: So we're keeping this from her. But there will come a time when we have to introduce this to her and we have to explain that this happened. And this is going to be a very, very, very difficult and painful conversation. And I can't even imagine how she's going to feel about it. And it's going to be a real challenge for our whole family and something that we're kind of pushing down the road because right now, her older siblings have borne the brunt of concern about this. Her parents have borne the brunt of concern about this. And thank God she's been immunized from it and protected from it, but that can't last forever.

GROSS: If you're just joining us, my guest is David French. He writes for the conservative publication National Review, and he's also a senior fellow at the National Review's think tank arm, the National Review Institute. We're going to take a short break, then we'll be back. This is FRESH AIR.

(SOUNDBITE OF MUSIC)

GROSS: This is FRESH AIR. And if you're just joining us, my guest is David French, who writes for the conservative publication National Review and is a senior fellow at the National Review Institute. And we're talking about his experiences being trolled. He started getting really harassing, threatening, vicious tweets after he started speaking out against the alt-right and became a part of the Never Trump movement.

So sticking with Twitter now, when you started to be trolled on Twitter, what were your options? What could you do?

FRENCH: Well, you know, your options are relatively limited. I mean, you can block or mute the Twitter account, which is - you know, we began blocking Twitter accounts by the hundreds. You can report Twitter accounts for banning, which Twitter did ban a few of them. My wife took the lead in that and spent better part of a day reporting account after account after account. And many of them did get banned.

And then, you know, if something on Twitter does cross the line into what you think is a true threat - in other words, an actionable - a threat that implies an immediate potential for violence, you can refer that to law enforcement. You know, it's pretty tough to identify any given tweet as a true threat. Normally, it's threatening or intimidating, but it's not a true threat and it's probably protected by the First Amendment, even as bad as some of those images were.

But Twitter, of course, is not a government entity. It does not have to allow any and all constitutionally-protected expression on its site.

GROSS: So you blocked a lot of - blocked a lot of people. And blocking means their messages will no longer be received by you, and they know that you've blocked them.

FRENCH: Correct.

GROSS: You muted some of them, which means that their messages will no longer be seen by you, but they don't know it.

FRENCH: Right.

GROSS: So you're still officially getting their messages, but you're actually not.

FRENCH: Right, right.

GROSS: So there's no revenge that can be sought for muting someone because they don't know, whereas if they block you - if you block them, they might want to get even with you in some way.

FRENCH: Right.

GROSS: Were you concerned about that, that...

FRENCH: Well, you know, I'll be honest with you, Terry, I was not - I'm not a Twitter expert by any means. I mean, I began to reflexively block before a friend said, hey, mute them because it has the same effect as blocking in the sense that they can't - you can't see what they write. But it's kind of a - you get a bit of satisfaction out of it because they will spend time busily working on their little posts and putting together little - their little images to no effect because you're not seeing it. So the mute button I think is more effective in some ways than the block button because again, they don't know when they've been muted. And I think that's a more effective way, all things considered, that - the more effective tool that you, as the Twitter user, have.

But, you know, Twitter itself, I think, has a real problem on its hands. I mean, this is a platform that right now I use only because professionally I feel like I must. Every journalist is on Twitter. Following journalists on Twitter is a great way to see breaking news as it's unfolding and emerging. And so for right now, it's indispensable for our profession for a very limited purpose. But I know journalist after journalist, writer after writer, public figure after public figure who literally dreads opening their Twitter app right now.

So good on Twitter for creating something that for a time is indispensable. But they haven't fixed this troll problem, and that means they're replaceable as soon as somebody gets a better platform because you don't want a product that people dread to use.

GROSS: I'm wondering how time consuming this process has been for you, both in terms of trying to mute or block people who are trolling you and in terms of trying to get Twitter to ban people who are trolling you.

FRENCH: It's been very time consuming. I would say that between the muting and the blocking and the attempts to ban over the last year, you're talking hours and hours, multiplying into days of work because you're reviewing your timelines, you are going through the process with - of reporting the bans to Twitter. Now, the bans we mainly focused on early on, and we were able to get the worst-of-the-worst accounts banned. And since then, we have focused much more on blocking and muting, and that's much less - that's a very quick process.

GROSS: When you reported people who were harassing you on Twitter, what's your understanding of the criteria Twitter used to decide whether the people you reported to them were worthy of being banned?

FRENCH: Well, you know, they're looking for hateful speech. They're looking for threatening speech. All of these things are subjective. So there's going to be some give in the joints. However, the speech that we were dealing with - so far over the line. Our situation was black and white. I mean, you just had to show them what the tweets were.

There is no circumstance under which Twitter would want its platform to be used to show pictures of 7-year-old girls in gas chambers or pictures of 7-year-old African-American girls as slaves. There's no circumstance where that is what they want their product to be used for. So all we had to do was show them the evidence, and, as I said, they responded reasonably.

The problem was that these trolls use a system that was created with, I think, perhaps, an excessively idealistic view of human nature, a system that was used to connect everyone with everyone else who's on Twitter. They used the system itself in a way that it was not designed to be used for but was easily exploited. And that exploitation is where Twitter has a real problem. They can respond, I think, with some decent speed once they're aware of an issue and once they're aware that people are receiving real harassment. The deeper problem that they have, as I said, is they set up a system that, from the ground up, is going to give basically everyone one clear shot at you first.

GROSS: Is one of the problems that you ran into that, like, if somebody has a lot of followers and that somebody sent you a harassing tweet, then all their followers are following that harassing tweet and they start sending you harassing tweets, too, even though they have no idea who you are?

FRENCH: Yeah, absolutely. And this is actually a pretty common tactic you will see. You will see someone with a large following fix a target, and they will tweet something at that target that is cruel but not over the line to receive a ban or a block - or to receive a ban. But they know - they know they have thousands and thousands and thousands of followers who have no conscience and no scruples. So it's almost like sending out the signal flare or sounding the bugle charge.

And you will see this online. And a number of the more quote, unquote, "respectable" atl-righters have used this tactic frequently. And that is, they're going to fix that target with their own tweet, knowing full well and intending full well for the resulting consequence. And then they go back to Twitter and they say, well, it's not my fault. Look at my Twitter timeline. And I'm not responsible for these people.

And it's a very hard thing for Twitter to pin down and respond to because then - when these same people go back to the conservative public and will say look, Twitter is discriminating against me. My tweets aren't nearly as bad as anyone else's. And so that's a real difficulty. And again, it's a difficulty that's inherent in the platform. It's the way people have taken this platform and figured out diabolical ways to use it. And it's one of the things that makes - I think puts Twitter on precarious ground right now.

GROSS: What would you like to see Twitter do? How - how do you wish they had responded to you?

FRENCH: You know, I - boy, I'm going to - I'll put on my entrepreneurial tech titan hat right now (laughter) and try to figure out a...

GROSS: Which I'm sure you own, yeah.

FRENCH: (Laughter) Right. You know, I think that the one thing that Twitter has to get a handle on, which is so unfortunate because it's so key to its appeal as a platform when done well, is that ability to immediately target and attack - and put whatever you want to put on their timeline just by hitting their handle with no filter, no pre-clearance. And then that - that's what exposes people to trolling.

And when it works out well - when people like someone and they're popular, it can be a joy. You know, the other day - I'll tell a story on myself - I'm a big fan of Dwayne The Rock Johnson, America's greatest celebrity. And I had a highlight when I tweeted something at him, and he tweeted right back at me. That's a cool thing about Twitter. I had to screenshot that, showed it to my kids, that means I'm somebody. The Rock tweeted at me. You know, that's a fun thing. Who doesn't like that? But then that very thing that people use for joy and for fun, when it's in the wrong hands becomes a weapon for intimidation and creating fear. And I honestly - boy, I tell you, if I knew the answer to that I would be doing it.

GROSS: My guest is David French, a writer for National Review and a senior fellow at the National Review Institute. We'll talk more after a break. And I'll talk with Charlie Warzel, who's written a series of articles for BuzzFeed about how Twitter is dealing with trolls. I'm Terry Gross, and this is FRESH AIR.

(SOUNDBITE OF MUSIC)

GROSS: This is FRESH AIR. I'm Terry Gross, back with writer David French. We're talking about how he's been trolled, harassed on Twitter ever since he started writing critical comments about Ann Coulter and the alt-right and became part of the Never Trump movement. French is a writer for the conservative publication National Review, which published his article about being harassed on Twitter titled "The Price I've Paid For Opposing Donald Trump." So you referred earlier to almost being afraid to open Twitter because you don't know what you're going to see and it might be very offensive. So are you still concerned about that? Are you still getting trolled?

FRENCH: (Laughter) Oh, my goodness, I - we dread it. I mean, so the other day I was - I was at dinner - at lunch after church with some of our friends, teachers in the kin - in the little school that my kids attend. And they said, well, have things calmed down for you? And so I said, well, let's see.

And I just opened my Twitter timeline and scrolled through the first three or so. And the first one was just a profanity-laden tirade against me. And I kind of laughed at that. But my wife was looking over my shoulder, and she saw the next ones which I hadn't read yet, and she just burst into tears.

And, look, I don't like to admit this stuff because the trolls feed on your anguish and your pain. They - they love it when they can get to you. But, you know, people need to know this is the reality of what it's like right now when you're out there on these issues. And my wife is a survivor of sex abuse and assault. And here was Twitter...

GROSS: Which I should say she's written about, you know...

FRENCH: Exactly.

GROSS: ...So she's comfortable with you saying that.

FRENCH: Yes, yes. I would not just spill that. She's written about it in no less prominent of a publication than The Washington Post, so I'm not telling stories on my family. And so she's a survivor of a very - horribly traumatic experiences. And, you know, there's people boasting about having sex with her one after the other after the other on Twitter, these - these men who are bombarding the Twitter feed with boasts because again, going back to earlier in the interview when I noted that people had claimed that one of the ways for harassing us is because we have an African-American daughter was to say that my wife had been sleeping with black men while I was deployed.

And so you just have this bombarding your Twitter feed, she saw it. And it's - and she's tough. I mean, she is tough. But, you know, the first hundred times, it's one thing. When it starts to get to the next hundred and the next hundred, it's - it wears down even the strongest person.

GROSS: So, I mean, one of your concerns is that the alt-right has attached itself to the Republican Party nominee, Donald Trump. So this is obviously a connection you'd like to see undone.

FRENCH: Oh, absolutely.

GROSS: What do you think will happen? Do you think that the alt-right is now just kind of a part of the Republican Party?

FRENCH: Well, you know, I will tell you this, if the alt-right is part of the Republican Party, the Republican Party's days are numbered because any movement that attaches itself to that level of hatred and viciousness is - it's going to end up being just a small, bitter, vicious rump of a party because it has no electoral future at all. Now, the problem that we have is that before that happens, before that extinction of it happens, an awful lot of damage can be done. And one of the things that's most disturbing and, again, why I wrote my piece, it's not so much that there is an alt-right that exists. We've always known that there is a trolling culture out there and there's a culture that glories in hurting and hating and intimidating other people. That's been there.

What is new here is this notion that not just the GOP nominee, who has retweeted alt-right accounts and whose team is very, very well aware of this dynamic out there, but also major outlets of mainstream formerly what I would call mainstream conservative media, like a Breitbart, like a Drudge, directly and intentionally and purposefully stoke this alt-right nonsense. They publish long, extended pieces that essentially function as rationalizations of the alt-right, excuse-making for the alt-right. They advance alt-right memes and alt-right themes into the culture. They do all of this very, very deliberately. There's nothing conspiratorial about it. It's just right there out in the open.

And so one of the most disappointing things - I would say the most disappointing thing about this election cycle has not been that this alt-right exists. We've always known, as I said, that there is this dark underbelly out there of the internet. We might not have known how big it was or how focused it was. But what has been so profoundly disappointing is the outright indifference and unconcern to its existence shown by so many more mainstream Republicans. And that's one thing that's been incredibly and deeply disappointing.

GROSS: David French, thank you very much for talking with us openly about your experiences. Thank you.

FRENCH: Well, thank you so much for having me.

GROSS: David French is a writer for National Review and a senior fellow at its think tank arm, the National Review Institute.

TERRY GROSS, HOST:

Our next guest, Charlie Warzel, has written a series of articles about harassment on Twitter and how the company is trying to deal with it. Warzel is a technology reporter for BuzzFeed.

Charlie Warzel, welcome to FRESH AIR. As you point out in your articles, one of Twitter's greatest strengths is also one of its vulnerabilities, free speech. The founders of Twitter were strong free speech advocates. It's the free speech approach to Twitter that has enabled Twitter to be such an important platform for pro-democracy movements, for the Arab Spring. Can you talk about that paradox that the free speech that Twitter embodies is also Twitter's vulnerability?

CHARLIE WARZEL: I think that this is one of the fundamental issues of the internet, this issue of free speech right now. And what we're sort of seeing is the idealistic understanding of what the internet could be, this utopian idea that so many entrepreneurs and people who have created these enormous social platforms, that they believe at their core that the internet can sort of raise all voices and really be an amazing tool.

And to have that anonymity tends to be something that these platforms favor. In Twitter's case, its core to their idea of free speech, and free speech is one of the founding principles that Twitter is built upon and this understanding that to truly connect the world, to truly be the pulse of the world, you have to give people the option to be able to be free of persecution. And that's why you saw so much of what happened in the early days of Twitter with the Iranian revolution and the Arab Spring, where Twitter played such an important role for political dissidents. It really sort of protected and allowed them to have a voice and elevated the platform.

GROSS: Would you compare Twitter's policy with Facebook and Instagram in terms of what you can say and how - and how you have to identify yourself?

WARZEL: Instagram and Facebook have adopted a real identity-centric approach. You have to give a version of yourself. You can't choose a pseudonym. You have to project some version of the person who you really are. And that is a very powerful thing, and it's a reason why Facebook is sort of one of the primary ways we authenticate ourselves across the internet.

And as a result, Facebook has its own problems with abuse and harassment but not nearly to the same degree because there's no way for people to sort of hide behind an anonymous account name or an anonymous avatar. On Facebook, you have to project that image of who you are. And Facebook has really doubled down on that. They have a lot of strict community standards as well as Instagram. One of those is no nudity, and it is something that has - that strong stand has inured the platform a little bit more to the kind of abuse that we're seeing grow so rapidly on Twitter.

GROSS: Yeah, you mentioned that Twitter is one of the few social media platforms used in the adult entertainment industry because it allows nudity.

WARZEL: Absolutely. And for that, it's been an incredibly useful platform for adult entertainers. And it is an example of giving a voice and being a home for people that don't necessarily have a voice. And I think that you see that actually working - not to make too much of a jump, but it's really one of the same principles that's at play with a lot of activist movements in the country.

It is a place where you can broadcast your raw opinion. You can get the news out that, you know, maybe some platforms are wary to broadcast. And that's been an incredibly successful tool for all kinds of movements like Black Lives Matter and the Arab Spring.

GROSS: So one of the kinds of videos that's a real issue on social media is beheading videos. Sometimes a beheading video, as gruesome as it is, is news because it proves that a hostage has been murdered. Sometimes it's purely harassment. People have been getting beheading videos just as a way of upsetting them, of harassing them, of threatening them. Can you compare, for instance, how Facebook and Twitter deal with beheading videos?

WARZEL: I think this is something that my reporting showed Twitter struggled internally with a lot, especially in in 2014, when this rash of ISIS beheading videos really started to flood the internet. There were internal meetings that we reported that showed that Twitter's executives were truly concerned that the platform would be overrun by this kind of content that may be newsworthy but is also broadcasting a very distinct message and is also incredibly disturbing.

But Twitter has become the place where news happens, where you get that raw eyewitness account and access. And Twitter had to figure out a way to harness the best of that. And I think that is something that they're still struggling with. They've created a newsworthy clause which allows them to allow certain images based off of their relevancy to public information and to the news. But there was a worry inside the company that if Twitter were to be overrun with these grisly, just very disturbing videos that there really wouldn't be anyone who would want to sign on.

GROSS: If you're just joining us, my guest is journalist Charlie Warzel. He covers tech for BuzzFeed, and he's written a series of articles about Twitter and trolling. We're going to take a short break, then we'll be back. This is FRESH AIR.

(SOUNDBITE OF MUSIC)

GROSS: This is FRESH AIR. And if you're just joining us, my guest is journalist Charlie Warzel. He's a tech reporter for BuzzFeed. He's been writing a series of articles about Twitter and trolling and what Twitter is and isn't doing to try to stop trolling.

You write that there's really a discussion, a debate within Twitter about what is Twitter. Is it more of a communication utility where it just - you know, it opens up the lines and you do what you will with it, kind of like the phone company, or is it a mediator of content, where content has to have some oversight? Would you describe more about that debate?

WARZEL: Anyone who's been following Twitter for the past decade has watched this evolution. Twitter started out as a very sort of quick, short-burst messaging platform, just the 140 characters, no images, no video, really sort of like a status update, what you're doing that day or at that moment and in a sense evolved just incredibly to be this sort of media-rich platform that has content partnerships with the NFL and media outlets like Bloomberg. As a result, Twitter really sort of is this media company. It is a place where news happens. It is a vibrant source of news for so many people.

And yet Twitter is also a utility in many ways. Twitter is this communication method, this digital way of reaching somebody, of having a conversation? It provides that infrastructure. And the real problem here is - seems to be that Twitter doesn't really want to put itself in any kind of box like that. They're very reluctant to, and they keep redefining, you know, who they are.

And the problem with that redefinition is that a utility is not subject to nearly the same kind of moderation as a media company. I can send you or anyone almost anything over a text message and AT&T or Verizon aren't going to moderate that and have no requirement to moderate that, whereas if I use a blogging platform to, you know, smear somebody or say something awful about somebody, there is sort of a standard on the internet that has been created that that should be regulated.

GROSS: What kind of effort is Twitter making to come up with a solution, a product that will both protect people on Twitter and protect free speech?

WARZEL: This is the fundamental problem, the free speech element really, really hampers Twitter. It's very important to them that no voices be silenced. And yet the task of moderating is to silence certain voices to some extent. Twitter introduced a quality filter not too long ago that they have rolled out to everyone. It used to be only for special verified users, so lots of celebrities and journalists. But this filter has proven - it's driven by an algorithm, and it's proven to be generally poor. It's also an opt-in filter, meaning everyday users are everyday users are going to have to go through their settings and change that. And that's something that I think plenty of everyday Twitter users who aren't sort of in the weeds don't necessarily even know they have that option.

GROSS: What does it filter?

WARZEL: The quality filter will ostensibly favor tweets that are created by verified users. I think that there is some effort to filter out certain search terms perhaps that are particularly violent or racially insensitive or tagged to hate speech in some way. But again, this is all very proprietary information that Twitter doesn't really let anyone in on, especially journalists. And this is one of the difficulties in covering Twitter from this angle of harassment is that there's so little knowledge as to what Twitter is really trying to do and so little effort on their part to disclose any of it, an unwillingness to disclose any of it that makes it difficult to see how, if at all, they are earnestly trying to fix this problem.

GROSS: BuzzFeed conducted a survey of Twitter users. There were 2,700 users who responded to the survey. I would say right at the jump here this is a very unscientific (laughter) survey.

WARZEL: Yes.

GROSS: This is representative of people who knew about the survey and decided to participate in it. That said, what were some of your takeaways from these responders?

WARZEL: I'll also just stress that this is an unscientific survey. But nonetheless we wanted to hear from users themselves and understand exactly what happens when they do go out there and experience harassment and report it. And what we found was that roughly 46 percent of respondents told us that the last time they reported an abusive tweet to the company, the company took no action on the request that they were aware of.

Another 29 percent said when they reported abusive tweets, they never heard anything back at all. It was effectively radio silence. And 18 percent said that when they did report an abusive tweet, they were told that the tweet did not violate Twitter's rules of being either a violent threat or hateful conduct. Only 56 instances out of roughly 2,700 people surveyed showed that Twitter deleted an offending account or a tweet that violated these rules. And so I think what - what the survey really showed was that regardless of what Twitter is doing behind the scenes, Twitter is doing a poor job of communicating exactly what's going on once you hit that report button.

GROSS: The New York Times this week ran a double-page spread of all the people, places and things Trump has insulted on Twitter since declaring his presidency. And there were one, two, three, four, five, six, seven, eight columns of really small print covering two pages of these tweets. Have you been following his tweets? And I'm just wondering how you think Donald Trump's use of Twitter is affecting perceptions of Twitter.

WARZEL: I think this is one of the most fascinating things about Twitter is just how integral it has been in this election. Donald Trump has been able to really leverage Twitter to get his message out to the base, really sort of skirting the media. And then also using Twitter as a way to pick up a lot of free media.

He can send out a string of incendiary tweets at 3 in the morning and by 7 a.m., they're dominating all of the morning shows. Twitter has been just central in this. And yet so much of that message lately has been so negative.

And if you look at the way that Donald Trump tweets and sort of what that New York Times spread can kind of show is that Donald Trump is himself a very effective troll with regard to Twitter. He says incendiary things that may or may not be based at all in fact. He sort of is looking for the reaction more than he's looking for any sort of substance. The fact that you have engaged with it, that you are outraged by it is just as important as whether or not you believe in it.

And I think that, you know, that behavior is again sort of being normalized in that sense to have somebody who exhibits a lot of this trollish behavior be elevated to the most covered human being in America or maybe the world for this past 18 months, I think that that has a profound effect on how other people, you know, choose to use the internet.

GROSS: If you're just joining us, my guest is journalist Charlie Warzel. He covers tech for BuzzFeed. And he's written a series of articles about Twitter and trolling. We're going to take a short break, then we'll be back. This is FRESH AIR.

(SOUNDBITE OF MUSIC)

GROSS: This is FRESH AIR. And if you're just joining us, my guest is journalist Charlie Warzel. He's a tech reporter for BuzzFeed. He's been writing a series of articles about Twitter and trolling and what Twitter is and isn't doing to try to stop trolling.

What kind of response have you gotten from Twitter to your requests to interview the CEO or get more information about what they're trying to do to deal with people who harass other people?

WARZEL: A major fundamental issue of my reporting on Twitter has been the lack of transparency. Twitter has not allowed us to speak with Jack Dorsey on this issue. And Twitter has not made any executives available to talk about this issue yet. If you speak with Twitter about this issue, they will say that theyre working on it and that this is something that they take very seriously now and have always taken very seriously and that they are actively working towards putting out some tools that will that will stop this. What those tools are is yet to be determined, and they have hinted publicly that we might see some of those things soon.

But there's not a lot for Twitter right now to gain perhaps by acknowledging this problem without putting forth a solution. That seems to be sort of the company line. I would argue, however, that so many of the people that I've spoken with love Twitter but are so frustrated and sort of feel that the company isn't angry enough about this, about this failing that, you know, the people at Twitter surely want to see this problem go away as much as anyone else.

And what people would like to see from Twitter - the people I have interviewed, the people experiencing this abuse on a daily basis - is a little bit more outrage. In 2014, Twitter's former CEO Dick Costolo released a memo that said we suck at dealing with abuse. And that memo was greeted by people who experience harassment with a lot of kudos. People were sort of thrilled to know that the company saw it, was frustrated and was going to deal with it. Since then not much has been done, and there's this sort of growing frustration as Twitter stays silent that maybe it doesn't understand just how bad this problem is.

GROSS: So why did Disney and the company sales force decide against buying Twitter?

WARZEL: The reports showed that, among a number of reasons, investors in both of the companies were troubled by a lot of the issues of harassment that are currently plaguing Twitter and all the bad press that that sort of entails. You have a lot of very high-profile celebrities who have quit the platform, like the "Saturday Night Live" actor Leslie Jones.

And when those sort of things happen, they create sort of this PR disaster for Twitter. So the harassment issue has sort of for the first time truly started to impact Twitter's bottom line. In past years, harassment has been something that Twitter can sort of point to as a small cordoned-off problem. It's something that's happening, the company regrets that it's happening, it is trying to fix the problem, but the rest of Twitter is out here spreading great information, you know, being the place where celebrities can interact with each other and with normal people. And it is billed as this wonderful community.

This sort of shows, though, that the harassment issue and the fact that abuse is increasing on the platform at a pretty alarming rate, it's finally affecting Twitter's bottom line. It's finally affecting how the company is performing, how the company is viewed in Silicon Valley and in the eyes of plenty of its competitors. And I think that that could be a moment for Twitter. It could be a real reckoning where Twitter finally says we have to get this problem under control or risk the future.

GROSS: Charlie Warzel, thank you so much for talking with us.

WARZEL: Thank you so much.

GROSS: Charlie Warzel is a technology reporter for BuzzFeed.

(SOUNDBITE OF MUSIC)

GROSS: Tomorrow on FRESH AIR, our guest will be chef Anthony Bourdain. His book "Kitchen Confidential" was a best-selling behind-the-scenes tell all about the restaurant business. In his Peabody Award-winning CNN series "Parts Unknown," he travels the globe sampling foods from diverse cultures. But his new cookbook, "Appetites," focuses on the food he makes for his family at home. I hope you'll join us.

(SOUNDBITE OF MUSIC)

GROSS: FRESH AIR's executive producer is Danny Miller. Our senior producer is Roberta Shorrock. Our interviews and reviews are produced and edited by Amy Salit, Phyllis Myers, Ann Marie Baldonado, Sam Briger, Lauren Krenzel, John Sheehan, Heidi Saman, Mooj Zadie and Thea Chaloner. Therese Madden directed today's show. I'm Terry Gross.

Transcripts are created on a rush deadline, and accuracy and availability may vary. This text may not be in its final form and may be updated or revised in the future. Please be aware that the authoritative record of Fresh Air interviews and reviews are the audio recordings of each segment.