How Haystack Works

In recent days there has been a great bruhaha over Haystack, the anti-censorship software aimed to help Iranians inside of Iran.

On September 2, 2010, Evgeny Morozov, a journalist colleague of mine, and online columnist for Foreign Policy magazine, wrote a thought-provoking piece about Haystack. In it, he called into question how Haystack works and argued that in fact, Haystack may be dangerous to Iranians, given that no one knows precisely how Haystack works except its two creators Austin Heap and Dan Colascione. 

I’m afraid that what this has now turned into is people feeling personally attacked rather than discussing the merits of Haystack.

In gChats in recent days, I told Heap that Morozov raises a lot of good points concerning Haystack and indeed, Heap has responded to the charges on his blog. In fact, the two have since been in direct contact by email.

But Morozov isn’t the first to bring up these points. In fact, his questions are ones that have been raised months earlier privately by other smart tech folks that I know and respect, like Danny O’Brien (Center to Project Journalists, former EFF, Jacob Appelbaum (Tor) and Ethan Zuckerman (Berkman Center, Harvard ; Global Voices).

Essentially their question to Heap has been: why should people trust Haystack, when you won’t open up how it works? 

Heap has responded by essentially saying: no one should trust Haystack any more than you trust Psiphon, Freegate, Tor, or any other web anonymizer. 

In short, this is a question that has plagued Haystack since its inception. 

And frankly, this is a larger problem with reporters like me reporting on highly technical issues that we fundamentally don’t understand. I am not a cryptographer, nor a network engineer. That being said, I was present at one of the earliest demonstrations of Haystack held in San Francisco in the summer of 2009. Other programmers in the room, many of whom work for major Silicon Valley corporations, expressed no concern that this was an unbelievable or ridiculous project.

Now, when I’ve spoken with Heap, and reported on Haystack, I’m essentially taking Heap’s word that Haystack does what he says it does. I have no means of proving that it doesn’t, nor that it does. Even when Heap has demonstrated the software for me, I have no real means of confirming his claims.

The best I can do, as a journalist, is to try to temper my interest and enthusiasm for a project like Haystack with other voices. I’ve reported on the project a few times for PRI’s The World, for my forthcoming book, “The Internet of Elsewhere,” and most recently for Popular Science magazine. 

In two pieces for The World, I countered Heap’s claims with skepticism raised by Ali Reza Eshraghi, an Iranian journalist living in Washington DC, who said in my July 9 2009 report:

“I am not 100 percent sure that by using all these technologies, all of these softwares, that means that ok, I can be safe and secure in Internet. But yes, it will definitely be helpful for me, but also they are also trying to find out, you know, again new softwares, new technology, how to monitor again the browsers?”

More recently on April 13, 2010, I included comments from Prof. Nader Entessar of the University of South Alabama who said this about Haystack

“We shouldn’t look at it in terms of a major tool, even a very effective tool to pressure Iran to change its policies, let’s say in the nuclear arena or other areas.”

Now, in addition, I need to come clean about a personal connection that I’ve had to Haystack since its inception. 

In getting involved in this discussion and presumably having read a lot of what has been written about Haystack in the press so far, Morozov came across a piece that I penned for Popular Science magazine earlier this year. 

In a private email, Morozov wrote to me: “I feel that I need to ask you the following: what’s your relationship – if any – to Babak Siavoshy? If there is relationship, why wasn’t it pointed anywhere in the piece as a potential conflict of interest?”

Morozov is completely correct. I should have come forward a lot sooner regarding my personal connection to Haystack and sincerely apologize for not doing so earlier and in a more transparent way. 

Babak Siavoshy, Haystack’s managing director, is my first cousin and in fact I introduced him and Austin Heap back in the summer of 2009. I also introduced Heap to my first-cousin-once-removed, Karim Sadjadpour, who is a well-known Iran analyst at the Carnegie Endowment for International Peace. Sadjadpour is also a member of CRC’s Board of Advisors. 

I definitely should have made this connection more public and probably even refrained from reporting directly on Haystack without disclosing this connection, or perhaps even refrained reporting on Haystack at all.

I have no excuse for not disclosing this potential conflict of interest earlier other than to say that I somehow justified it to myself that a member of my extended family, who was on the board of this organization that I was reporting about, whom I didn’t ever interview, was far enough away that it was ok. 

But upon further reflection, Morozov is completely right, I should have made this more clear to my listeners, readers, and editors for not coming forward sooner. I promise that such an oversight and mistake will never happen again. This potential conflict of interest, or even appearance of a conflict of interest, is something that violates journalistic ethics that I hold dear.

All of this said, I would like to close by making two final points:

1) Heap’s blog post, in which he criticized Morozov for not having spoken with him directly, was also spot on. I think that much of Morozov’s questions and concerns about Haystack could have been addressed directly to Heap rather than just being a blanket shot across the bow of Haystack. If I am at fault for not disclosing my connection to Siavoshy, Sadjadpour and Haystack, then Morozov is also at fault for not contacting directly Heap initially to address his questions about the project.

2) I have great respect both for Morozov and Heap, and consider them both as peers and also as friends. I’ve met socially with both of them multiple times on separate occasions and they are both very intelligent and nice guys.

While I am very interested in Haystack and its goals, I also think that Morozov has provided thoughtful skepticism that has influenced my own thinking about the role of technology and the Internet in promoting democracy around the world. When I first began work on my book, “The Internet of Elsewhere,” I initially had thought that I would write something about the “liberating effects of a wired world.” But both Zuckerman and Morozov have provided me with a healthy dose of skepticism.

I look forward to reading Morozov’s forthcoming book about the role of the Internet and democracy, which is due to come out later this year.

As I wrote on my blog back in May

As much as I love the Internet, it is no more capable of causing revolution than the telegraph was, as Tom Standage showed in his great book, The Victorian Internet.

The fact of the matter is that for all the talk of ‘Twitter Revolution’ in Iran — the status quo has been preserved. Khamenei is still doing his thing and Ahmedinejad is still doing his. There’s no evidence to suggest that the Islamic Republic is in danger of collapse anytime soon.

I generally agree with Evgeny, although I may not be as cynical as he is. The bottom line though, is that I feel like Fox Mulder on the X-Files: I want to believe that the Internet helps to build democracies, but as of now, I simply cannot.


  1. I’m glad you went public about your family connections with the project. But I think your friendship with Austin could have been more grounds for conflict of interest comparing to the distant cousins involved in the project. (The fact that you’re not mentioning there are Iranian internet savvy people who are questioning the authenticity of this project due to not knowing any Iranian who is actually using it in Iran might say something about this.) On the side note, I agree with yours and others’ skepticism on the role of technology in promoting democracy. (part of my master’s thesis on the use of Internet by the women’s movement in Iran has touched upon this). I just think you came to the conclusion for the wrong reasons. (getting disappointed at how people are treating each other online and thus concluding that Internet is not the liberating tool you once thought it was?)

  2. Sanam,

    Thanks for your comment. I presume that you’re referring to your friend who emailed me. Just because he hasn’t seen a copy of Haystack and doesn’t know anyone who has doesn’t mean that it doesn’t exist. Austin has told me that there are less than 100 copies inside of Iran currently in use and again, I have no reason to disbelieve him. I wonder, would your friend also say that there are no Linux users in Mashad just because he hasn’t met any of them?

    Regarding my skepticism, perhaps I was not clear. It was forged long before Haystack even existed and well before Morozov and Heap were disagreeing with each other.

  3. Pingback: ?????? ??????? ?? ??? ???? ????? ? ?? ??? ??????! | ???? ?????

  4. Pingback: Jillian C. York » Haystack and Media Irresponsibility

  5. Pingback: …My heart’s in Accra » Evaluating censorship circumvention tools and “irresponsibility”

  6. I know a bunch of Sharif good-level IT alumni, working on top companies and projects and active opposition supporters. Some of them hadn’t even HEARD of this software. I mailed Heap in the summer and didn’t even get a response. I hope there will be a serious investigation on Hay-Stack.

    Jaso Jashak
  7. Pingback: The Internet Freedom Fallacy and the Arab Digital activism « sami ben gharbia


  9. Pingback: Les cyber-activistes arabes face à la liberté sur Internet made in USA (2) » Article » OWNI, Digital Journalism

  10. Pingback: The Internet Freedom Fallacy and the Arab Digital activism (2) » Article », Digital Journalism

Comments are closed.