EARLIER THIS YEAR, The Intercept got a new visual identity, the first step in a multi-phase redesign. Later this month, we’ll be making an important change that’s invisible to our readers, adding a new analytics system to help us better understand how our stories spread and how they can be promoted to a larger audience. We thought it was important to describe this system — and its privacy implications and safeguards — to you in a transparent fashion.
The biggest challenge we faced in adopting a new audience measurement system was preserving reader privacy; modern analytics tools virtually always come from outside vendors who become intimate third parties in the relationship between publishers and readers. It was important to us to try and rebalance this relationship in favor of the reader. Since launching a little over a year and a half ago, The Intercept has always coupled its drive to expose information closely held by the powerful with efforts to protect data that rightfully belongs to our readers. That’s why we serve all our content over well-encrypted “HTTPS” web connections and why in April we became only the third internet service, behind Facebook and Blockchain.info, to allow people to contact us over HTTPS-encrypted connections to the anonymity network Tor.
Working with the online analytics firm Parse.ly, we’ve arrived at an analytics architecture we believe is compatible with The Intercept’s core values. Parse.ly’s software allows publishers to monitor and analyze traffic to their stories, to figure out how readers come to the site and what they do when they get there. It also provides information about how stories are shared on social platforms like Twitter and Facebook and how search terms drive incoming traffic. Normally, this functionality is provided by causing readers’ web browsers to directly ping Parse.ly’s servers.
Together with Parse.ly, we’ve arrived at a system whereby readers of The Intercept will not directly ping Parse.ly. Instead, they will continue to send web requests to our own servers, which will, in turn, forward some of those requests on to Parse.ly, after stripping out readers’ internet protocol, or IP, addresses. Parse.ly will use these requests to track our readers via random unique identifiers that we generate. It will not be possible for Parse.ly to correlate readers’ visits to The Intercept with their visits to other Parse.ly-enabled sites.
Beyond these architectural safeguards, Parse.ly has also agreed not to log IP addresses of visitors to The Intercept, not to infer or store geolocation data of visitors to The Intercept, and not to set or receive network-wide Parse.ly tracking identifiers from visitors to The Intercept — even though it should not have the technical capability to do any of these things in the first place.
Readers who do not wish to be tracked even via the anonymized system outlined above may opt out simply by activating the “Do Not Track” feature on their web browsers; this will deactivate both the new Parse.ly system and our old system, based on software called Piwik.
While noting our work with Parse.ly, it’s also important to disclose that there are other service providers we continue to use who do have access to information like your IP address, which articles on The Intercept you visit, and which computer operating system and web browser you use. Such partners include our content delivery network, CloudFlare; our video provider, JW Player; and platforms we use from time to time to embed rich media on our site, like SoundCloud and Vimeo. You can read more about these third parties on our newly updated privacy policy.
We expect to activate our new audience measurement system in one week or shortly thereafter. If you have additional questions or concerns, you may post them in the comments below, or contact us directly.
This looks like a CIA plot to ruin us all at one time.
It may turn out that that draws most ‘regular readers’ to this site is not so much the articles but the comments. Certainly I read all the articles that interest me but I think the reason I come here ‘regularly’ (which is what Big Data wants to know — couldn’t they just, y’know, ASK?) is to sift through the comments for the nuggets there often are. I appreciate that ‘The Intercept’ provides a platform for discussion as well as publishing very good articles.
It’s hard to separate the two motivations, but I’m pretty certain that I check the site more often because of comments, since I could come just once a week for the articles themselves. So, there you go, you didn’t need to suck up my data and analyze it, thereby introducing a little resentment into the mix, I can just tell you!
Although I sometimes wonder why some comments disappear into a black hole and I suspect you will probably end up alienating me and I will write you off as just another ruined fruit. Perhaps the rot has already set in. This is the kind of thing you can’t discover from click-counters and referral pings.
Big data is a philosophy, attempting to apply the scientific method to decision making in business or politics or war. Given the success of the scientific method, this is highly appealing to rationalists.
The difference to the traditional scientific method is that the data collection is no longer directed to solve a specific problem. A scientist may offer a hypothesis, then think which data would support or refute this hypothesis, and then attempt to collect that specific data. With Big data, one starts with collecting data first, and then tries to see which answers it may hold. The philosophy also adds the assertion that with more data, you will make better decisions.
So Big Data looks like the scientific method, but is not: With Big Data, the easily accessible or observable data will bias your decisions. Other data that might be harder to get will be ignored, and the Big Data analyst will not try to find such blind spots: he is paid to crunch terra bytes of data, not think about confounding evidence he might be missing.
Media practicing Big Data analysis of their readers will measure the “impact” of their work in clicks and comments. Thus, editors will drive their staff to write what hogs attention and creates controversy. Even though they may hold high the banner of the Fourth Estate, Big Data analytics will subtly drive their output towards that of tabloids, simply because the overall social impact or importance of a story is very hard to measure, and if the data is not collected, Big Data will ignore it. Sure, editors may let authors write an story they deem important from time to time, after all, one needs to keep the staff happy. But once the collected metrics point towards a decline, those features will be seen as less crucial: After all, business-first and the data does not lie.
The scientific method would suggest that decision makers should consider which data they really need to make a specific decision, and then collect that data specifically, and evaluate it with respect to the specific decision they need to make. It is sad that the Big Data movement has convinced many modern decision makers that it is more important to make decisions fast and cheaply, rather than well.
Hard to figure out if you are talking in a sarcastic way, but “Big data” doesn’t “hold” any answers. It may, however, articulate self-serving illusion you may be entertaining.
RCL
I’m enjoying reading Zeus’s comments on this site. Pity ‘The Intercept’ doesn’t ask for our website addresses rather than our email as then we could give away our privacy willingly and happily in exchange for spamming it with our creative daubs, as I’m sure he has a website tucked away somewhere.
Zeus asks how many people read these pages anyway, but as everyone knows even after you have gone to great lengths to extract your reader’s data, and here it must be like pulling a tooth for them, it is bound to be entirely misleading, especially here, since most people must surely use Tor to read and comment, unless they like being on watchlists, so for all they know it could be five people appearing to be a massive crowd through frequently changed Tor identities. So what use is the data anyway, except to put a few curious cat-lovers on watchlists.
I myself copy GCHQ into all my emails. I hope they enjoy reading them. Especially now as almost every secure email provider has been DDoSed out of existence by JTRIG over the past few days and we’re all having to spill our secrets in Gmail once again.
Thanks.
I don’t have a website, but I was writing a book / game thing that I hoped to take to GenCon in San Diego one day, but I fear I will not be welcomed by the US government no more!!! Oh well, I will probably publish it all on a wiki for free and run it as a hobby. I have also plotted pretty much the entire medieval / ancient world on Google Earth – it is kinda nuts, but then so am I!
I was diagnosed belatedly with a nasy disease about 5 years ago. For a whole year I couldn’t sleep and was in constant agony. I would pass out for an hour or two each day and then awaken, on fire and feeling like every joint had seized up. Not sleeping REALLY screws you up, and for about 3 years I was a zombified mess trying to get some semblance of life back into myself and nearly dying a couple of times. All the time I stayed positive and did not wallow in the misery. Now I feel utterly depressed and disappointed with America! It’s always been a bullying jackass, but this is just insanity what it is doing.
I started writing as something to do. I researched the Greek Roman and Celtic myths and “found” my own take on them, that there were some good gods and a whole bunch of conniving and scheming gods trying to usurp power and return the mysterious gods worshipped on the island of Samothrace and near Thebes, called the Kabeiroi. It was real fun to do and I have written a lot – still nowhere near finished!!! – and then the Ed Snowden story broke, just as I was getting my senses back together. It seemed like the wicked shenanigans in my story had come to life, this master plan to rule the world.
One of the major themes is the concepts of morality in Ancient Greece and Rome, how they differed, and how Rome is a sort of twisted take on it all. To the Ancient Greeks, the greatest moral sins were Hubris – excessive arrogance and vanity to a degree where it insults the gods and all men, and Xenopatia – breaches of the code of hospitality afforded by hosts to strangers, and reciprocated by the guests. Many of the Greek myths revolve around these themes. It seems to me America is guilty of these sins: hubris in believing it can rule the world, and Xenopatia in trying to do that.
In the story the god Pan – I call him Ogmios, a Celtic god akin to Hercules, but with a silver tongue that enslaves men – is one of the real bad guys, and his party piece is causing Panic and “sowing the dragon’s teeth” – creating discord and division amongst people. He is also Dispater, Plouton, the god of the Wealth of the Earth. Sound familiar?
I grew up with the first generation of computer programmers. I started my own web company in 1995 – it was too early and failed – so the web is like a friend to me and the EFF has been there since the beginning, warning of this impending trouble and mass surveillance risks.
It made sound a cliche, but we have nothing to fear except fear itself. These people are nothing special, they just sit at the top of a greedy pile of selfish idiots willing to abuse and lie and kill for a few dollars. I say FUCK THEM.
By the way, I am mad about the Epithets of the Greek gods, they are very telling about the nature of those gods, and also very ambiguous and open to my own interpretation! Hence my ZeusThis and ZeusThat. There is a massive Greek Lexicon written by Liddell and Scott and amended by Jones, known as the LSJ – it is my Bible! It’s great. It contains nearly every Greek word from about 700BC to 300AD, and the striking thing about it is how familiar and normal much of the language is. Life really wasn’t that different, they had tyrrants (a Greek word) and democracies (a Greek word) and morons (a Greek word) and Satire (a Greek word) and Comedy (yep, you guessed it!) and Tragedy – the root of which is “Goat Play”! The Greeks had a rather bizarre infatuation with Goats. That’s Pan for you…
I have also read a fair bit about Iran, Persia as it was, particularly under the Zoroastrian Sasanids, I have read some of the Avesta and researched Zoroastrianism. It is an interesting religion, arguably the first montheistic religion. It’s central premise is that there is a struggle between Oromasde – Ahura Mazda, the Good God – and Arimanius – Ahriman, the Evil God.
It seems it was a very simple and peaceful religion, not given to exuberance or pompous display. It’s temples were mostly very simple structures containing sacred fires – Atar – often out in the wilds where travellers could find comfort.
It is often forgotten that Persia is one of the world’s great cultures and that the Iranians are arguably the oldest intact culture on Earth.
Throughout history they have been attacked by nearly every great empire – the Scythians, Greeks, Macedonians, Romans, Huns, Mongols, Timurids, Russians, British, Soviets, Americans – and survived as a recognisable people.
I hope they remain that way.
My reading has made me both a Hellenophile and an Iranophile.
Zeus Xenios grants hospitality to ALL men.
I come from England. It is an amazing country, truly a magical corner of the world despite what its critics think. There is nowhere else like it. It defies concise description. It is where Marx was inspired to write the Communist Manifesto. Where the Magna Carter was written. Where the great modern trade empire was forged. It is the epitome of a monarchy. It has great Socialist traditions. It is still hugely class-ridden. It is divided by every town and accent, yet it is united as England, yet it is divided as the United Kingdom.
It is the birthplace of the Industrial Revolution and the IT Revolution, inventing computers and programming and the internet. It also invented TV and penicillin. It is a pioneer and leader of literature and music. It is the inventor of the world’s most favourite sports. It led the world in education and healthcare and social services.
Its gently swells into life in Spring, gently basks in Summer, glows golden in Autumn, and dies in Winter. It is the greenest place on Earth, in every nook and crack plants grow. Despite being renowned for awful cooking it has some of the finest ingredients in the world – milk and meats and cheeses and fruits and grains and seafood and drinks. Its parliament has been a model for many as has its legal system, its police and its armed forces.
At its heart sits London, the greatest city on Earth, with Roman ruins, a medieval fortress, great green parks, amazing nightlife, the cuisine of the world, theatres, free museums, unrivalled sports, great pubs, enormous festivals and carnivals, an unrivalled and generally happily coexisting ethnic diversity, amazing business opportunities, great universities, unrivalled shopping, a brilliant public transport system, and it still keeps a cozy and easy-going feel to it where everyone feels welcome.
Yet now the UK is a depressing shithole run by Yankie and Saudi-loving cocksuckers. Get out of my country, shitheads, you’ve ruined everything. I fucking hate them.
But the biggest point to the story is the ambiguity of all things, the idea that one man’s drink is another man’s poison. That Bravery and Honour and Decency to one man is Hatefulness and Violence and Immorality to another.
When we cross each other’s boundaries we inflict ourselves upon others, and so we must understand the nature of Guest and Host, of Xenia. Imperialists do not think of such things and our moral compass no longer points in that direction.
I live in a tourist resort and tourists are very often arseholes. They completely forget that people live here, that it is rude to gawp and treat the world as if it is there for their entertainment, a cutesy escape from their “more serious and important” lives.
It is not, and it is often a “paradise” BECAUSE it is not subjected to the shitty apologistic lifestyle of the West, that works blindly and ignores all its sins and sufferings in the name of earning a living.
It is Paradise because that’s the way we fucking well want it to be.
My story is about tat we can only exist through diversity, no homogenisation. Not through One god, One world, One dream. This is modern fascist bullshit dressed up in a cozy liberal coating of “being nice and understanding”.
The amount of elitist, racist, nasty Americans I have met who think they are “nice” because they vote Democrat is untrue.
And America’s attempts to play Imperial Master are as hateful to me as anything, and should be hateful to all peoples. What the fuck does America know about hill farming in Etheopia? Or kindergarten schooling in Burma? Or football matches in Croatia? Or shee farming in Patagonia?
KEEP YOUR FUCKING NOSES OUT AMERICA. WE DON’T WANT YOU.
I am all for making money, I am all for America being strong. But not at my expense or that of my country’s.
Why is ‘The Intercept’ hot on the heels of stories like this:
https://theintercept.com/2015/11/06/how-law-enforcement-can-use-google-timeline-to-track-your-every-move/
but oblivious to the dangers of Gravatar? And what is the point anyway of having Gravatar on this site when only about two people purposefully use it for commenting here and on various cat blogs?
Is ‘The Intercept’ actually getting anything out of the deal? Or is it just giving away this prime personal tracking data out of the goodness of its heart, because it wants to support one of the most invidious trackers on the web.
If this was a recipe site, perhaps I could understand that people might want to have their own pretty little avatars on their comments, but here why is it being used at all? This hasn’t been answered, even if the powers-that-be at ‘The Intercept’ seem ignorant of what Gravatar is and how it operates and how the data it collects would be priceless for a state actor to hack or demand.
The best thing you will learn at ‘The Intercept’ today is that you should look after your own privacy, because if you don’t someone else will strip you of it, knowingly or because they don’t know any better.
If I were the NSA, I think I could come up with the concept of Gravatar in my lunch-hour. Well, we just don’t know, do we? And the people here don’t seem to have a clue.
So tell us, The Intercept dearest: how many hits do you actually get per article?
Because I am thinking it is just us who write on here and a few people from the NSA and GCHQ with clearance enough not to get thrown into Tartarus for daring to look upon The Gorgon, Medusa. so around about 100.
I hope a few world leaders too, just to see my fucking angry RANTS. :)
Less than 100,000 – a very popular glossy magazine – or more than 1,000,000 – a respectable newspaper?
It’s purely vanity calling, but I am sure EVERYONE is curious to know how popular the world’s most “rad rag” is doing on the click-o-meter.
After reading the Privacy Policy for this site:
https://theintercept.com/privacy-policy
and the comments about Gravatar below, I wonder why ‘The Intercept’ is even bothering with semi-anonymizing data sent to pars.ly, given that the site is leaking people’s personal data all over the place like a old rust-bucket ship.
I suppose I expected more from ‘The Intercept’. Lesson learned.
Changed my mind, one last question for Ms. Reed and Mr. Tate, since there are rarely open postings available to do so: Have you guys done FOIA requests (and their equivalents in other countries) on the site regularly?Have you guys given any thought to doing so, if not?
If a site/editor/writer proclaims to be one thing while obviously espousing values and providing a danger to its users, under whatever cover, one should probably question why and how these decisions get made.
Not planning on coming back to comment (certainly after this newest ‘improvement’) but I just wanted to say… STILL no warrant canary?
(FWIW I’m not stating the two thoughts are at all related, I’m just saying that because I’m suprised that, after all this time, one still doesn’t exist). While I’m also broaching unrelated subjects, it might behoove anybody who might want to comment or leak to you to understand how this might affect their profiles with your new tracking system. Surely heavy users of all kinds, with a persistent tracking system, are going to stand out, so you’re already selling out your users before they even say anything. These times can be tracked via the backend. Tracking these via the backend removes the need for parse.ly to have anything like what you’re talking about. Which is why I said I don’t think you understand how the internet works. It doesn’t stop and start with your website and go somewhere else then come back unrelated.
Last comment. Good luck with the site, and have a good weekend.
It’s a shame.
One last point: Obviously all supplied email addresses on comments have to be sent to gravatar.com in order to check whether you have a personalized gravatar image associated with that address.
If you don’t then you will get the default image and the MD5 hash in its URL of whatever real or fake email address you supplied when commenting, and this image is tied to your IP via that hash.
There is no safeguard in always using invented email addresses to post because your IP is inextricably linked to that MD5 hash because its first occurrence in their logs for a particular page stains your IP address indelibly as the author of the post with that supplied image URL. And if you use your actual email address, the MD5 hash is open to cracking and linking by anyone looking at the page.
This was pointed out by a few people months ago; it was met with disdain and a lack of caring. So we obviously shouldn’t. Privacy wasn’t put into the thought of this site at all. The only way to deal with it is to not visit, or to assume that you have no privacy (at least, or especially, if you comment, and especially if you have scripting on or aren’t intimately familiar with your browser (generally all of the above).
I’m not making any negative comments about this site at all. I’d never do such a thing. But I do think it’s important that people know that this isn’t the first time this topic has been discussed (at length), in order to prevent themselves from thinking that spending an hour or so trying to explain privacy issues is generally met with no real changes, just ‘explanations’.
The rss feed is useful for a list of the articles, fwiw, but I wish it listed more articles.
If you don’t want Gravatar tracking your reading, which I don’t (but, oops, didn’t think of before your note), set your browser to block images from there, or even put a bunch of 1.gravatar.com etc. DNS names in your hosts file with bogus addresses.
That’s like hiding your head under the blanket because if you can’t see the monsters then the monsters can’t see you. Maybe you need to look into how Gravatar works.
To summarise, then. A URL like this:
https://1.gravatar.com/avatar/700396edd9127db23d363085a78d567d?d=mm&s=40
loading onto this page along with dozens of URLs like it records the IP address of everyone loading this page to read it at gravatar.com, and, the first time it occurs as a request in Gravatar’s logs the IP address associated with that request is the author of the comment. They have the author’s IP, and the IPs of those who have loaded the page it is on.
And if you go elsewhere where Gravatar images are loaded, they can tie the IPs of people looking at that page with the IPs of those looking at this page. If you comment, they will have your IP and the first occurrence of an MD5 hash on that site will identify you as the same author of the first occurrence of an MD5 hash on the other site. The IP address of whoever first loaded the image will be the one who created it.
If you commented on both sites from the same IP address, Gravatar will know. If that IP address can be traced back to your own door in your own street, because you’re not using Tor or a VPN, then you may have just handed a government a 10,000-comment-long dossier of your opinions.
Yet people here are discussing the fine points of parse.ly. If you have discussed Gravatar here before, as I gather you have, then I can only surmise that no-one knew very much about it.
If this is still not clear, let’s say I host a small avatar image on my website and you decide you would like to use it for your own avatar on a forum. If you’re sensible, you’ll download it and upload it to somewhere you control yourself, but if you’re lazy or not too skilled you will use the URL to the image on my site (which is exactly what all those grey and white images by people’s names are, images downloaded from gravatar.com). Say you use this avatar image plundered from my site on lots of sites. Well you have now provided me with a list of all the sites where you make comments and I can find all those comments, because every time that image is loaded onto a page with one of your comments on it this is recorded in my own logs. I could even change that image to something else with the same filename and embarrass you on 10,000 posts, which I did once actually, replacing a bird with an arse. Of course he’d forgotten that I was hosting his image and he was stealing my bandwidth. So, you see all these useless grey circle images that are forced on you by your name, like this:
https://1.gravatar.com/avatar/700396edd9127db23d363085a78d567d?d=mm&s=40
are recording your IP address and supplied email in the logs of Gravatar even though you have never had anything to do with Gravatar. And whenever you take your IP address to another site loading these useless Gravatar images, there’s your IP again in their logs, and if you comment, there’s your MD5 hash identifying your comment as belonging to that IP. Clear now?
Even for people who do not comment, anyone loading a page with Gravatar images on it (default or individual, doesn’t matter) can be tracked across the web by their IP address simply because Gravatar is used on many sites and all they have to do if they want to see what a reader of this site also likes to read is search for an IP address in their database. This is also open to hackers and states to find out. Now you have a browsing history of an IP address so far as Gravatar-enabled sites go.
And that’s just by reading this site, not even commenting. If you also comment then your browsing history and political opinions from all those sites you have commented on can be tied to a single IP address: yours.
If people don’t know this, they need to wise up. I use Tor, so I couldn’t care less, but if you don’t use Tor these Gravatar-enabled websites ‘dedicated to privacy’ are actually fucking you bigtime out of their own ignorance.
Ryan Tate said: “Right now, nothing in our comment system encourages people to use Gravatar or points them at that site. If you have a Gravatar, it’s because you created an account there.”
— The point is that those who have chosen to have nothing to do with Gravatar are still allocated a unique default Gravatar image ID, and the MD5 hash of this is identifying. Gravatar has the IP address and supplied email address of every specific comment. You say you’ve talked of getting rid of Gravatar, but you still have it and from this comment of yours it seems you don’t really understand exactly how it works for those who have chosen not to use it with an image of their own, who, as you say, are opting in. The plain truth of the matter is that everyone who comments here is actually opting in anyway without realising it. If you do not know how Gravatar actually works, and what it is doing, what hope for someone just coming here and making a comment?
Ryan,
I wouldn’t worry about the paranoid crowd too much. It’s understandable that a site focused on national security and technology issues would draw the literal tin-foil hat types but I don’t think they understand very much about the technical details and so get easily upset about such a reasonable change like this one. As far as I can see this won’t infringe anyone’s privacy in any meaningful way (given that they choose to read on the Internet in the first place). Keep up the good work.
There’s no such thing as a little suspect. I think they are selling out to make a point. Trust no one. Roger that, SAM. How many times have they stuck Honey Badger in the can and I still get loose? Now they’re bringing the noose so Snaggle Purse is otter here!
Hay, is this whole system running on Linux’ suspect kernal? Servers us right.
This isn’t written in any language I can comprehend. I think it proves my point.
Does this mean you’re leaving the commenting section, abbadabba?
“jackrousseau” <— dumb NSA cunt.
hello—
I went back to Do Not Track. Added a list. But when I look in that right corner, I see: your browser does not support do not track — with a red x
then you have enabled do not track with a green check
Am I now opted out?
I might note also that if anyone here is foolish enough when commenting to give a real email address (why are you even asking from for them?) then it is feasible that it could be discovered from the gravatar image MD5 hash that is very publicly available to anyone just by viewing the image:
http://arstechnica.com/security/2013/07/got-an-account-on-a-site-like-github-hackers-may-know-your-e-mail-address/
It just gets better. Frankly astonishing that you spent nearly year on your parse.ly adventure and let gravatar.com have the keys to the back door. How can you be trusted??
LOL – The “webmasters” here seem to be worse than amateurs…
While this is a longstanding part of the site, and nothing to do with what we’re now rolling out, we have discussed removing the option to use Gravatar.
Right now, nothing in our comment system encourages people to use Gravatar or points them at that site. If you have a Gravatar, it’s because you created an account there. If you’ve already chosen to set up a Gravatar account, something you’d do entirely apart from surfing The Intercept, it does get included in our comments, since they are WordPress hosted. But if you set up a Gravatar, that’s exactly what you’d expect to happen, since that’s the whole point of setting up a Gravatar.
Saying they “have the keys to the back door” is not even a little bit true, by the way.
Anyway, we appreciate the feedback on Gravatar. It’s something we continue to discuss.
” If you have a Gravatar, it’s because you created an account there”
You don’t seem to be getting the point at all? I don’t have a gravatar account. All the ‘anonymous’ posts (like this one) automatically get an image from gravatar. Your own comment system is linking post to gravatar images that are tagged with some kind of ID.
is linking posts*
In fact, is gravatar.com not only capturing the IP address of every specific comment, but every reader of this page whether they comment or not? I would think it obviously is, knowing what I know about looking an analytics logs. Everyone is downloading unique image IDs from gravatar.com just looking at this page, therefore their IP address is in gravatar’s logs. Is that not true?
Tut tut.
Gravatar clearly seems to get IPs if you load the comments. As I said elsewhere, we’ve discussed removing them. I’d be all for it, actually. We’ve had gravatars and other places where IP addresses leak for a long time. I mention some of them in the post:
“While noting our work with Parse.ly, it’s also important to disclose that there are other service providers we continue to use who do have access to information like your IP address, which articles on The Intercept you visit, and which computer operating system and web browser you use. Such partners include our content delivery network, CloudFlare; our video provider, JW Player; and platforms we use from time to time to embed rich media on our site, like SoundCloud and Vimeo. You can read more about these third parties on our newly updated privacy policy.”
Well, you could have mentioned gravatar too. The question remains, what’s the point of your allegedly less intrusive parse.ly tracking when you are leaking all sorts of information about your readers.
And actually, the only feedback you need from readers is how many times an article is read.
The question is, Ryan, is gravatar.com capturing the IP address of every comment made here?
Sometimes the detail is in the little things, and all the fuss about anonymization elsewhere has missed the obvious elephant in the room.
Oh, you are Ryan, you are using a nice little gravatar. Except yours doesn’t come from gravatar.com, yours comes from here:
https://firstlook.org/wp-uploads/sites/1/2014/08/Ryan-Tate_avatar_1409293833-96×96.jpg?d=mm&s=40
So why aren’t all the rest, these blank-faced avatars, also coming from firstlook.org?
Why is gravatar being used at all on a site like this, is anyone here using their own gravatar image? I don’t see anyone importing their own personalized tracking beacon, which is all gravatar is.
And, while I think about it, it is interesting that though all the gravatar images on these comments are the same blank-faced image, being loaded from gravatar.com, they all have different IDs.
Here are the last two, for instance:
https://1.gravatar.com/avatar/a487fbcccb3e8b3e68e50650e556581a?d=mm&s=40
https://2.gravatar.com/avatar/5e3bb56e5b70e5c015ebef0687e4522c?d=mm&s=40
Is this tracking anything?
You see how easy it is to lose trust and have people start sniffing around everywhere when you become a tracker.
Here’s a nightmare scenario for you to consider, when bringing in this new system:
Say I was a commenter here who occasionally had a whistleblower streak, but I otherwise didn’t identify myself. Now say I said something very revealing one time about something very secret, foolishly admittedly, but there you go, and, thankfully, I hadn’t given away anything personally identifying. Wouldn’t a persistent ‘anonymized’ identity make it easy to go back to all my other comments to find something I may have said some other time that was more personally identifying? Now you have outed a whistleblower. Think on it.
Revealing information as a whistleblower in our comments is a bad idea, period. This has been the case since our comments first existed. I do not believe Parse.ly makes it any worse, but other things make it a bad idea.
The right way to leak information is described by us here. This article is also linked from our contact page. https://theintercept.com/2015/01/28/how-to-leak-to-the-intercept
When you leave a comment, you leave an email address and your IP address is recorded. If either of these are traceable and still in our records, and are able to be obtained by some law enforcement agency, you lose anonymity. Our privacy policy discusses our retention of emails and IP addresses and our stance on law enforcement requests.
Why do you even need them? I am happily and brutally writing because I do not give a shit anymore, but the majority won’t even COME to this website over security fears, the very thing you claim to be fighting.
It is a FUCKING FARSE, Mr Tate. A lazy and cheapskate farse, particularly given that eBay was one of the pioneering companies regarding internet security for transactions and Ed Snowden is the inspiration for all of this. Your owner has amazing tech boys and website designers at his disposal, yet we get this mobile-friendly stack of crap on which to challenge the possibly worst breakdown of law and security the world has ever seen.
Thanks, but you know, no thanks really.
“eBay was one of the pioneering companies regarding internet security”
translation : ebay was one of the pioneering companies acting as a spy for the criminales known as “US government”.
It does seem a strange idea to so miserably lose the reader trust you have built up just for the sake of knowing whether your CSS layout is catering sufficiently for all screen sizes, and whether someone has come here with referrer information from a PRISM unobtainable internal NSA URL (as the hidden ‘dogshit’ .onion site revealed so brilliantly), or from the BBC, or from a WordPress blog. People come from somewhere, that’s always been the case, and if they come from their bookmark there is no referrer anyway, so whoopy-doo that analytic.
Ryan Tate said: “They ALSO have a contract obligation to discard any IP addresses they receive. This is a failsafe, a bit of redundancy. This is indicated at the end of the sentence, the part you cut off in your quote: “— even though it should not have the technical capability to do any of these things in the first place.””
— A failsafe for what does not happen just in case it does? That’s a curious idea. Either IPs are anonymized or not before passing on. If there’s even a chance they might not be, then should data be being passed on at all? In any case, the IP data is being collected in the first place in order to anonymize it.
I would have thought most people read this site in Tor anyway. How is your analytics going to be at all useful in that regard, since your persistent identification is of a Tor user, not a persistent Tor user who is always the same Tor user. Or is there something you’re not telling us?
“Either IPs are anonymized or not before passing on”
No no, they are not anonymized, they are simply omitted. I have sometimes said the system as a whole is “anonymized.” IP address info is simply not passed to Parse.ly.
” failsafe for what does not happen just in case it does? ”
That’s what “failsafe” literally means :-)
Tor Browser users could be anonymously tagged as repeat visitors during a given session just like any other user. The difference is that Tor Browser automatically deletes all cookies at the end of the session, so on a subsequent session, they would appear as a new visitor. Obviously, if they turn on Do Not Track in Tor Browser, they won’t ever be anonymously tagged at all.
Don’t trust anyone. Don’t you know that? Ask Snowden, Greenwald, Poitras. Ask Miranda if he trusts GCHQ. Ask Micah Lee – better yet, read his posts.
BTW, where does GG stand on this?
You could have asked readers about this before ten months ago. If it’s so important to do this (with a billionaire’s money available and editorial independence), why not do it in-house?
It is my understanding that compliance with DNT is voluntary. So say other websites.
What is the importance of and difference between mobile vs. desktop? Do you not publish the same on both?
If you want to know about social media, look at it.
I’m a lawyer. Did you ever hear of breach of contract? And what if USG enables or requires more tracking and info from parse.ly?
I am flummoxed. I thank you for letting us know about this and welcoming robust exchanges.
P.S. Per Disconnect, 31 trackers from Automatic on this page; 0 trackers on the Schneier article linked below.
I have 0 trackers reading this article and comments.
I am not a techie enough to give you a good response to this. If it matters, I am running Mac OS X El Capitan and Chrome browser.
Hello Jerry, We do not rely on trust or contracts. EVERYTHING Parse.ly agrees not to retain is, on a basic technical level, UNAVAILABLE to them in the first place due to our proxy setup. The agreement with them is a failsafe backstop against any kind of error or bug down the line in which data somehow leaks through this design. This setup required a large amount of work by our engineering team and is explained repeatedly in the blog post.
On the Automatic Inc. trackers, those are the gravatar images attached to some commenter names. We’ve had those for a long long time but have discussed removing them. I’d actually be all for it.
I hope this weeds out some of the paranoiacs from the commenting sections. :) I sincerely hope all threatening to leave forever do. (They won’t.)
As a web software developer i agree there is a certain requirement for figures like: how many guys read a certain article and where does most of this traffic come from. From my perspective these technologies have little in common with evil spying performed by secret services. In contrast to the greater part of the audience i feel very much respected that TI is transparently communicating their plans to learn something about how their content is read. Ask yourselves: How many other publishers do that? Just because secret services use computers for evil stuff does not mean each and every computer is a satanic tool. I still would like to know: what was TI missing when you were PIWIK? I’d personally rather trust an open source software than a proprietary(?) solution.
We did give Piwik a go. The Piwik interface is rough, and does not include any data on how our articles are shared on social media.
That said, the plan, I believe, is to keep it running alongside Parse.ly. If we find there’s no real edge to Parse.ly, I suppose we could go back.
Journalism is dead. The Internet is dead. Freedom is dead. Our paid servants have become a living horror story and most people moronically sit by and ignore it. I am sadly glad I have a chronic and deadly illness as I really don’t want to be around in 30 years’ time. I have been arguing with my friends and family all my life about this eventuality, and no one ever listened and they all thought I was paranoid – and here it is, and I have no desire to say I told you so, just a deep sorrow at how revolting our leaders are and how pathetic everyday folk have become. Just glad I have a loving wife, fuck everything else.
“Readers who do not wish to be tracked even via the anonymized system outlined above may opt out simply by activating the “Do Not Track” feature on their web browsers”
Thank you, that makes all the difference, but you should make that clear to new readers perhaps through a ‘sticky’ at the foot of each page.
For what it’s worth, some activist Indymedia sites pioneered HTTPS over TOR, and it was useful to tell which individual collective had been compromised by state infiltrators – ie the ones that claimed to allow that but actually didn’t were dangerous.
Happy to update the section on Tor HTTPS certs on the indy media sites, if they got an actual cert! We definitely were early adopters on that, not the people who invented or pioneered it.
Hi Ryan,
I believe those anti-authoritarian IM sites required us to accept insecure certs regularly for each article, although the same verifiable fake certs each time, but I’ll leave their admins to explain that if they care to claim credit as each site was different. It was something to do with them having no Server Name Indication, but this isn’t my specialist subject so I’m guessing. This site is certainly still an innovator, no doubts from me, I just don’t have all the history to satisfy your journalistic curiosity.
My main point was tracking is less useful to your employer as many readers here will already have their ‘shields up’ by default. And any reader that doesn’t know the basic of internet security will be easily scared by what they don’t understand, as the comments here show. Tracking is very scary for them, even if it’s benign, anonymous tracking to improve the site.
Timing our reading of articles isn’t that useful anyway because some of us cut and paste your text to read offline, and others admit they open the articles that interest them all at once and read them at leisure. Sometimes I just open an article I don’t have time to read to show approval, which indicates how how surveillance can be misunderstood. You are not a normal website so you won’t get normal data from the tracking, tell your employer to bear that in mind. Just know that we love you! And at least most of trust you. We just don’t trust our APT states and get a little paranoid.
And maybe add a preview button to stop me making errors like that?
and even quick post editing, maybe? Before other people reply to you?
RCL
The entire farce of Govt domestic spying -to me, has been: What business is it of theirs, what sites I am visiting? It just makes those nosy goons more nosy and puts everyone who checks to see if the Govt is lying on a Govt shitlist!
Now TI is starting to follow that same path.
Oh, ” Parse.ly has also agreed not to log IP addresses of visitors to The Intercept, not to infer or store geolocation data of visitors to The Intercept…”
We all know now that Yahoo (and other internet interests) sold us all down the river by (secretly) being in bed with the NSA and violating our privacy when they could have blown the entire surveillance picture out of the water by going public and telling the Govt to go fuck themselves! However this did NOT happen.
I feel oh SO much better now that Parse.ly (or is it oregano?) has “agreed” not to fuck us!
Ryan and Betsy, your site is starting to smell like a pigsty.
No offence, but if this measure places you at any extra risk then you are certainly at more risk today than you realise. You need to take responsibility for your own internet security. Sadly that means becoming a bit of a techie and learning more about the devil in the details. I’d suggest reading the comments section of technical publications like The Register, or asking advice from more technical posters here.
Of course everyone deserves internet security, not just nerds like me, but at the end of the day we face Advanced Persistent Threats like the NSA and GCHQ and so we all should up our own game. The author of this article includes advice on turning off this tracking if you wish – it’s only your fault now if you ignore that simple advice. And if you didn’t already know that simple measure, you won’t know a dozen other basic steps you should already have taken.
Danny –
Why should one have to be a techie or nerd to have any degree of privacy? You even admitted such in your post. You said: “The author of this article includes advice on turning off this tracking if you wish – it’s only your fault now if you ignore that simple advice.” Again, I also feel why should the onus be on us to opt out of something that, especially a site like TI shouldn’t be doing in the first place.
You shouldn’t ‘have to’ be a techie to keep your privacy, but the main reason you do need to take responsibility is our governments abuse.
Now, how much you want to learn to protect yourself is up to you. The reason every website does some surveillance, even if they claim not to, is because otherwise they’d be taken offline by script-kiddies never mind APTs like the NSA. Think of it as necessary maintenance, and appreciate the sites like this that are at least being honest.
It says in the article how to turn off tracking. It shouldn’t have to, you should know that, everyone should know that. It’s akin to the warning on a packet of peanuts “May contain nuts’. Why should any peanut manufacturer have to warn us that peanuts ‘may contain peanuts’? They just do because some folk are that stupid yet are litigious.
Hey, TI still gives its users the possibility to opt out of it altogether!
This is bs
There you go with a self-serving metaphor that doesn’t answer feline16’s very basic question. Other users have asked important questions here and on that “new visual identity” post, which have been at most rationalized.
That metaphor would be better worded along the lines of almond farmers including peanuts in their closed boxes (as they do with olive oil) and one farmer making part of his very business critisizing those practices as “dishonest”, “unmoral”, “inconstitutional”, … and then pretending to be honest because it is telling you, “you can still open the boxes and take out the peanuts in our boxes” …
RCL
… we are “better” because we are letting you know and, look, we even make it easy for you to rip the box open so you can more easily take the peanuts out …
RCL
Parse.ly does not have access to the IP addresses. We proxy all the connections to them. They could not share the info with the NSA if they wanted to.
They ALSO have a contract obligation to discard any IP addresses they receive. This is a failsafe, a bit of redundancy. This is indicated at the end of the sentence, the part you cut off in your quote: “— even though it should not have the technical capability to do any of these things in the first place.”
If I assumed too much, allow me to apologize Ryan. My error stems from the statement that “Parse.ly has agreed not to log IP addresses”. That statement highly excites my other (paranoid) personality.
Excuse me! My bad?
Any reason you didn’t just use http://piwik.org?
We used Piwik for about a year. We’d have preferred to only use that but we weren’t getting the data we wanted out of it. The Piwik interface is rough, and does not include any data on how our articles are shared on social media. But it’s an evolving project, and a good one, and something we’ll keep tabs and check back on.
hi thelastnamechosen
Even with a billionaire benefactor and no pressure for advertising dollars and you guys still can’t help but track people. It’s like a compulsion. I’ve got an idea, why don’t you write what you believe in instead of writing what will get you the most facebook “likes” or “re-tweets.”
Spying defined: Listening to someone without actually wanting to hear what they have to say.
We get more analytics while this comment section festers. You want to listen, but you still don’t want to hear us.
they could have asked me and I would have answered, now this site has lost a fan,
Dear Intercept,
What a disappointment!!!! Think of it this way… there was a time when one bought a paper from a news stand and read that paper in private. No one knew what pages were turned to or how long one stayed on a page. Just because all this monitoring can be done, does not mean it is necessary or even right. You will lose many privacy advocates by doing this. What a mud splash in the face of privacy for your thirst of the need to know by joining the ranks of the very ones you have been writing about to expose. I shake my head with shock you are doing this. Good bye!
Sorry to hear that. We spent about 10 months working with Parse.ly and internally with our tech team to create this anonymized proxying system, and do feel it offers a high degree of privacy.
Hi Ryan
It could have taken 12 years but it still offers less privacy (privacy=freedom) than what we had yesterday.
There is already enough of a chilling effect without this
Ryan,
You and the Intercept are missing the point. It is not about how anonymzing your new system is. You have joined the ranks of the mass surveillance vacuum system. Please bring your staff back into meetings with a more enlightened conscious awareness. You are in violation of the very things you have brought to light to the world. Unbelievable!
LYB –
I fell EXACTLY the same way. TI is losing/has lost the trust of the readers it said it would serve.
How it came about that Snowden brought us to be tracked.
Pure corporate speak. Look, you guys have been dragged kicking and screaming by your readers on this issue since day one. Revisionist history is not a great way to begin a conversation that depends on trust.
If you actually cared about privacy, you would make tracking opt-in instead of opt-out. Opt-out is your confession.
Do you really believe that you can protect people’s privacy while using twitter and facebook to advertise for you? You are encouraging people to participate in a system that is directly opposed to protecting the data that rightfully belongs to your readers. You can’t have this both ways.
The tracking of individuals across the internet is the perfect example of how the media keeps falling into irrelevancy, and how easily corruptible they are. The second biggest privacy story of the decade is not only ignored by the media–because the media is a co-conspirator–but even after Snowden reveals how corporate internet tracking and government surveillance go hand in hand you still have the Intercept using analytics to track twitter and facebook advertising.
Even with a billionaire benefactor and no pressure for advertising dollars and you guys still can’t help but track people. It’s like a compulsion. I’ve got an idea, why don’t you write what you believe in instead of writing what will get you the most facebook “likes” or “re-tweets.”
Spying defined: Listening to someone without actually wanting to hear what they have to say.
We get more analytics while this comment section festers. You want to listen, but you still don’t want to hear us.
And the best part…
You have actually come up with a system that circumvents anti-tracking software–and you call this pro privacy!
Fuck you.
Thanks for the comment.
To answer your question, the reason we made it opt-out instead of opt-in is that we felt we’d redacted enough private data before proxying to Parse.ly for this to be the default. As the post indicates, we strip out IP addresses, and Parse.ly (obviously) doesn’t get any parse.ly-set cookies, like their cross-site tracker, only the cookies WE set — which means your cookies only identify you by a random ID that we generate and that can’t be correlated with any internet activity outside of theintercept.com.
In short, we strained to remove any data that could identify you uniquely to parse.ly OTHER than a randomly generated ID. This includes any data that parse.ly could even hypothetically use to cross tabulate your intercept visit with any other visit. Opt-out then seemed reasonable.
Further, the “opt-out” mechanism is the standard Do-Not-Track header, so we’re not asking anyone to jump through special weird hoops to opt out.
That said, obviously we’re paying attention to the feedback on this decision.
I’m confused on your meaning on this: “Do you really believe that you can protect people’s privacy while using twitter and facebook to advertise for you?”
To be clear, this new Parse.ly system in no way links your Twitter or Facebook identities to your Intercept visits, and we’ve added nothing to our site to give Twitter or Facebook direct access to our users.
(We have, historically, embedded tweets from time to time, which does cause your browser to ping Twitter, but that is not connected to anything in this post/decision/change).
What Parse.ly does with Twitter and Facebook is uses their APIs to look for people sharing our content. So if you make a public post linking to us, we will see this, just like anyone on the internet could see it. But we won’t be able to make any connection between your social shares and your visits (if any).
(And we do not run any advertising FROM Twitter and Facebook. I believe we may have run some ads ON those platforms, but never using any data about visits to the Intercept, because that data does not exist in any way that could be linked to identities on those platforms.)
(To be clear, the word “direct” can be cut from this sentence, it was not an intentional hedge: “we’ve added nothing to our site to give Twitter or Facebook direct access to our users”)
Thanks for the reply.
You have always been opt-out. The idea that this new system is the reason you chose opt-out strains credibility. You hold the same position you have always had.
I love that you are protecting my anonymity except for the number you have assigned in order to uniquely and permanently identify me.
This should dissuade any illusions, perceived or projected, of scrubbing identifying information:
https://panopticlick.eff.org/
Only a geek or an advertiser (hard to tell the difference these days) could say this with a straight face.
I want to talk about this proxy. Haven’t you just come up with a system that circumvents anti-tracking software that people use to prevent being tracked? I knew this was coming at some point, but I never thought the Intercept would be leading the way. At least with the current system, I can block tracking by IP. Your system hides all that.
You have invented the Tor of internet tracking. A system to hide, protect, and anonymize the corporations who are tracking us. What the fuck were you thinking?
Facebook and twitter offer the potential of free advertising for your product if you are willing to allow your customers to be tracked. This business model is called the “publicity whore” because it relies upon advertisers, fame seekers, and publicity whores to advertise and provide publicity for twitter/facebook/etc, while seeking publicity for themselves. It is as close to a perpetual motion machine as we have ever come, but ultimately the bills get paid because people are relentlessly spied upon using this model.
When you participate in the twitter/facebook/etc publicity machine by trading you customer’s privacy for a PR lottery ticket, for example with links to facebook and twitter, you are making the conscious decision to trade your reader’s privacy for the chance that this may provide “free” advertising for you.
Ask yourself why you treat your readers different than you treat whistle blowers. Your readers are your whistle blowers.
Where is the SecureDrop for us?
“You have always been opt-out. ”
This is a bit disingenuous. Our prior “opt out” was “OK, you won’t even go in our regularly purged server logs, cataloged entirely in house by Piwik open source software.” In other words, it respected the most extreme definition of “tracked”. The vast, vast majority of sites wouldn’t even consider this something to opt in or out of, but we respected people’s right to make a choice. You’re saying that makes us sleazy, for keeping basic web logs by default?
“I love that you are protecting my anonymity except for the number you have assigned in order to uniquely and permanently identify me.”
I don’t think anyone disputes that the model is a persistent, anonymous identity. Are you saying persistent identity and anonymity are mutually exclusive? Pairing the two is an old model on the internet. I mean, we respect if people don’t want that, but it’s not a new concept, and I do think it’s reasonable to make opt-out the default when the tracking identity is anonymized.
The browser fingerprinting point you make is certainly a valid concern (https://panopticlick.eff.org/). It’s something we’re discussing with out tech team, to see how much we can strip out of the user-agent header to prevent this. Mostly we just want to be able to see how many mobile vs desktop visitors we have.
You deride D-N-T as the opt out mechanism, but it has the virtue of not requiring site-by-site opt outs, which suck. It’s also got a decent UI in virtually all browsers. Do you really think it would be better if we made people fill out a little web form, instead of just respecting DNT? Honest question.
“Facebook and twitter offer the potential of free advertising for your product if you are willing to allow your customers to be tracked. This business model is called the “publicity whore” because it relies upon advertisers, fame seekers, and publicity whores to advertise and provide publicity for twitter/facebook/etc, while seeking publicity for themselves.”
I’m genuinely confused by what you’re talking about here. To be clear, we do nothing to allow Facebook or Twitter to in any way track our users, and we get no free actual advertising from them. Yes, we have readers who choose to share our posts on Twitter and Facebook. We have accounts on those services. We have links you can click that will give you a form to post to those services. Is this what you’re referring to? If people want to share publicly, we’re not going to stop them, nor is that our place. It’s not “allowing [our] customers to be tracked” by Twitter and Facebook to have accounts on those services. But maybe I misunderstand you.
You previously used google analytics. You were shamed into moving off that platform, and now you want to go back to a platform that allows you to track what your readers do and see across other websites besides the Intercept. That’s what makes you sleazy.
Look, “do not track” is like telling someone that they aren’t wearing their “do not grope” bracelet. This is the taunting of a psychopath, not a serious proposal for discussion. I really shouldn’t have to opt out of being groped.
It’s called opt-in. Simple and honest. If I want you touching my ass I will tell you.
For now, I think the facebook/twitter thread has become a distraction, but I will say that I find it hard to believe you don’t understand how the internet tracking economy works, and how we ended up elbow deep in ass groping. There are reasons why you are promoting facebook and twitter, and the lack of facebook/twitter javascript hasn’t always been the case here.
Educating people about tracking, the tracking economy, and how to avoid it is exactly your place. You are journalists for God’s sake! This is what you should be doing, not tricking people into being tracked by facebook and twitter just so you can get some “free” publicity.
“now you want to go back to a platform that allows you to track what your readers do and see across other websites besides the Intercept”
This is completely false. Neither we not parse.ly can track you across the web as with Google Analytics. This is exactly why we spent months and months setting up a proxy system that strips IPs, generates its own random IDs, and by design prevents parse.ly from accessing any of its cookies, like the ones it uses to track people across sites. Neither we not parse.ly have the capability to figure out what other sites you visit. Obviously if you click a link to get to an article we get the same HTTP referrer field anyone else gets, but this is nothing like what you’re talking about where Google Analytics or stock Parse.ly can track you across sites.
“tricking people into being tracked by facebook and twitter”
Again, this is totally false. Literally our only connection to those sites is 1> we have corporate identities there 2> we have our own self hosted links that you can click to make a post for those sites, if you want. If you don’t click the link, they have no idea you were ever here.
Then why in the world would you need to send any information, much less unique personal identifiers, to a third party? What exact information is parse.ly providing to you? Did you really spend 10 months and presumably a nonzero amount of money so that parse.ly could draw pretty graphs of your server log files? Your comments are contradictory to say the least.
Again, what exact information is parse.ly providing to you and how does this allow you to “understand how our stories spread and how they can be promoted to a larger audience.”
As far as facebook and twitter goes–as I suspected, you are being purposefully obtuse. In my first post, I said “The tracking of individuals across the internet is the perfect example of how the media keeps falling into irrelevancy, and how easily corruptible they are”–your responses on this topic demonstrate this idea perfectly.
Ryan, understand I am criticizing you as an employee of the Intercept, not as an individual. Nobody chooses to work ten months coming up with ways to track people. Your sins show why you are easily forgiven–this is not your first choice.
I think we are getting closer to the disconnect here. Yes, we did choose Parse.ly and do all this work in part to get a better look at data that is entirely contained in server log files. We have an editorial team of dozens of people, including writers, editors, researchers, fact checkers, designers, etc. You and I might be happy grepping through server logs, but it is not a widely held skill, and it frankly also does not tell me or anyone else what we need to know in a rich or rapid manner.
You may deride “pretty graphs,” but those graphs can be very useful.
A big part of what Parse.ly buys us is, in theory, what Piwik should do, but in practice does not.
Another part of what Parse.ly does it entirely separate from looking at visits. It also runs searchers on Twitter, Facebook, etc, for people sharing our articles, which helps show us social engagement. As I indicated, we do NOT attempt to link visits with social media accounts, nor do we have a way to do so. But we do like to know about people who choose to share publicly our content in aggregate.
On Twitter and Facebook more broadly, we seem to be talking past one another. I see no incompatibility between criticizing Twitter and Facebook editorially — which I am trying to build a team to do — and acknowledging that people will share our content there and making the best of that. Our founders and editors are not exactly reticent to use social platforms, so it should not come as a surprise that we use them too as an institution. That does not preclude criticizing them as platforms and as companies.
Jesus Fucking Christ. Writers, researchers, fact checkers, and designers have no business looking at these kinds of analytics, much less in a “rich” and “rapid” manner. Analytics is not farmville. This is not a game to make the workday go faster, this is ultimately private information that belongs to your readers. Why would you let some fact checker rummage around your reader’s private data in real time. Incredible. The privacy nightmare continues. You guys really need to get your shit together.
Your grep comment is bullshit and you know it. It doesn’t come down to grep vs graphs, but calling one graph prettier than the other. Also, I’m not deriding graphs, pretty or otherwise, but you knew that. I’m having a hard time trying to imagine how anyone could possibly be against graphs. But what do I know–maybe there is a big anti-graph contingent.
There is no disconnect. What you guys are doing is so stupid and embarrassing, that to assume you were doing what you now admit to doing would have been a grave insult. I understand this is not your decision, but holy smokes how in the world did that take ten months to do something so dumb?
This has nothing to do with sending personal unique identifiers to Parse.ly. You know that. Quit trying to conflate them.
Did all this really come down to not wanting to use two different analytics packages? Jesus Fucking Christ.
Let me speak clearly about facebook, twitter, “social engagement” and what you guys are doing here.
You have advertisements for facebook and twitter constantly at the top of every article on this website. Most importantly, these advertisements have been purposefully designed so that they are confused with the Intercept itself. This is not just a mixing of advertising and journalism, but advertising and publishing.
In that 4×4 box at the perpetual header of every article on this website, you have ads for facebook and twitter, along with a mailto link and a link to the comments section. You have purposefully mixed advertising with website navigation. It’s just like a “free” porn site.
For the moment, let’s skip over the shaky ethical foundations of this decision, and get on to the hardcore capitalism. This kind of exclusive, relentless and ethically dubious advertising of two huge internet companies is at least bringing in the big bucks. So how much is it that you are being paid for all this advertising? Nothing? Seriously? Nothing? The really important part of selling out is the selling part. It’s right there in the name. They even made it first. If you are selling out for free, you are really missing the point of the exercise.
Admittedly money isn’t everything, so maybe facebook and twitter are bringing you benefits in some other way. So let’s dissect the transaction. You recruit and send to facebook and twitter unsuspecting people, including children and the elderly, to be tracked, spied upon and toil in the facebook fields and twitter mines providing free content in the hope that some of these same people will also promote your website while they are there.
And this seemed like a good idea to you? From a pure capitalism perspective, the privacy of your readers is worth much more than what you are getting in return. You are trading dollars for fractions of fractions of a penny. It’s stupid deals like this that makes evil much less effective and bad ass. The reason to be evil is to get cool stuff. Not just to be evil.
For the moment, let’s ignore ethics, let’s ignore capitalism, and just talk about how completely corrosive this practice has been to journalism and the internet itself.
Tracking and spying is killing journalism and the internet. It really is that simple.
Ask yourself: What do you want the world to be?
Heya, TLNC, been a while. Just wanted to say, save your breath; breadth here is only an invitation to hell, as a friend of mine found out the hard way (and is still paying for it).
Give up on the idea of security if you visit here. That’s all there is to it. Their ‘web designer’ is a clueless stubborn fool and despite months and tons of comments it’s only gotten worse. Look at the comment counts from last year and compare them to this year. A thousand — several hundred per article. Now, it’s… human interest and maybe the biggest articles pull just above 100 entries.
If I could make a SINGLE request — ONE request — it’d be stop making the FRONT PAGE require js to get a list of all the articles. One can turn off js and the other scripting unless one wants to reply to a comment, but one can’t load more than a handful of articles to open without js.
No offense, web dev person back there but you need to get an education. And a lot of people need to learn the meaning of irony, here (and by that I mean behind the scenes, too, apparently, if this is the newest grand idea).
Other than that, I don’t really care. I’m not a regular visitor, nor do I intend to be. Whatever reasons I had to come here are gone now, anyway.
I just wanted to comment how utterly ridiculous it is to even put the word ‘privacy’ in the same sentence/headline as ‘audience measurement system’ is.
And that’s shameful. The only good thing I can say is at least this time you said you added it… well, except it seems you said so after you added it.
Good job.
PS Just to be clear, I appreciate all your comments. This kind of informed pushback is a big part of the feedback we were hoping for.
Feedback means people can change things. ‘Informed pushback’ generally is NOT something someone should ever want as a web dev — it means you’re not listening to your users at all — it means YOU aren’t informed and that THEIR pushback implies you never were listening or asking in the first place.
I’m not getting involved here, nor do I have any interest in commenting further, but if you’re confusing those two terms, then it’s no wonder users here are flustered and flabbergasted. I’m letting you know this because it seems there is a disconnect and you’re driving users away more and more. Have you considered that maybe this is BECAUSE you’re adding things like tracking and the like and muddling up the site with a lot of third party stuff — and not letting feedback have anything to do with it?
Pushback means people don’t like something. If you have no intention of acting on something, that’s not feedback: that’s forcing a user to go along or go away. If your numbers go down, it may behoove you to question if this might be why.
I really appreciate the heads up!
This is the first time I have read what is done with information attached to my browsing that I understand.
It’s very rare to know without a doubt a business has Honor and Integrity and its Word is its bond. A very beautiful thing!
Presumably most readers are here because they value knowledge? That said, some of the comments puzzle me. It should already be self-evident that we have to take control of our own Internet privacy and security. This doesn’t have to be either complicated or costly. There are safe, reliable and free add-ons for browsers which limit the amount of identifiable information being spewed out and tracked. This is a responsibility of the individual, not of the WWW at large!
I’m a little surprised at the comments left here.
You’ve been given disclosure of what’s going on, being done, etc. which is far more than I can say for any other site. I see nothing wrong with The Intercept wanting to raise their profile by determining how their stories are disseminated. I’m wondering how many of you freely give up your information to Google/Chrome without a moment’s hesitation but are balking at being told what’s transpiring here.
You can opt out. You can use tracking blocker plugins like Ghostery and other browser plugins. You can utilize the numerous browser plugins for Firefox to obsfucate what you’re doing, or use TOR. Take some time to educate yourself on these things.
Last, you can use a paid VPN service that allows you to connect from different servers around the world and hide your actual location.
It’s rather foolish to complain about this and throw out a “I’ll not come here again” while ignoring the fact that you’re relying on a website to take care of your own OPSEC.
Vukovar – Maybe you don’t get it…
TI has exposed the breadth of spying. To find they’re essentially joining the ranks of the spies and trackers is really a betrayal of what we thought they stood for. To say we give info to other sites without blinking, well, man y of us shudder and hate the thought of doing that, but find alternatives limited for one thing. Why should one have to be a techhie to have any sort of internet privacy? Why should we HAVE to put up defenses against tracking that shouldn ‘t happen in the first place? And I’ll echo tlnc (sorry to abbreviate, but I’m a lousy typist) WHY should OPT-OUT be the default?
In short, yes, we “put up” up with tracking at other sites, unfortunately. That doesn’t make it acceptable, and certainly not acceptable for a site like TI who should be more aware of the effect of tracking on users and not give in to it.
I use ghostery to decline the use of trackers on websites. Some have nearly 20 trackers, now TIC has one…it’s no big deal too me!
This reminds me of this post by Bruce Schneier: https://www.schneier.com/blog/archives/2015/02/everyone_wants_.html.
Everybody wants you to be secure… except when it comes to their own business practices.
This is very sad. I come on this site everyday to read articles and I won’t be doing that anymore. Glenn, you need to write something up about this. Until then, ciao!
It is kind of sad but not unexpected that TIC instead of leading in the privacy field after so many posts about devastating threats to our fundamental right to privacy, for commercial and financial reasons betrays all those principles it claims to hold dear.
Do TIC editors really need to know whether 10 or 10 millions readers it has? Or these are accountants who need it.
It is even more regrettable that TIC was launched on the heels of Snowden disclosure of illegal mass surveillance violating the US constitution and since became public almost it was completely legalized by acts of congress. All of that brings to the fore crucial questions of trust and motivations that must be asked.
Some answers can be found here:
https://contrarianopinion.wordpress.com/2015/01/06/snowden-affair-unasked-questions/
I must admit I like keeping a few The Intercept articles open throughout the day, no idea if the page is reloaded over the hours, the browser doesn’t really give me such a option to control it. So I feel a little bad for having a few articles around that I haven’t yet gotten around to read. :|
What a stunning piece of bullshit. “Audience measurement system”? Very Newspeak of you. In plain language that means spying on your readers & finding out a lot about them you don’t need to know.
How did print newspapers manage for centuries to be profitable without having to look in the window as their readers opened the morning paper? Why do you think you need to know how someone came to your online publication?
Why do you not simply measure your success by number of page hits & understand that the better articles you write and publish the more readers you will get?
What’s that old rock song? Meet the new boss, same as the old boss. Except the old boss had the decency not to be hiding behind your sofa watching to see if you went for the sports or business sections first.
For shame, The Intercept. For shame.
Thanks for the feedback. To clarify one point:
” In plain language that means spying on your readers & finding out a lot about them you don’t need to know.”
The only new piece of information we get from this setup is how long a reader spends on each article. We, like any site on the web, have historically received basic information like IP address and information sent as HTTP headers, including a User-Agent string describing your web browser, and in some cases a Referrer header, indicating what web page sent you to us. We have no information about your identity beyond this, like social media accounts, your web surfing history (beyond the Referrer header), etc, as many other sites collect.
Parse.ly, via a proxy we control, gets an anonymized, persistent random ID, set via cookie, that it can use to show us various sorts of traffic and engagement charts, and the information listed in the paragraph above, except for the IP address. It gets the random ID via a cookie no cookies it can use for any other purpose.
Anyway, thanks for the feedback.
Avidus –
I totally agree.
Thanks for the description. However, it seems to leave the most pressing questions unanswered. It would be interesting to know what data you actually send to parse.ly (I got that you do not send the IP address but identifying people by IP address is a bit 90s anyway…). It would also be nice to know what parse.ly then does with the data.
Assuming Do-Not-Track is OFF in your browser, Parse.ly gets a random ID we generate, and, associated with that random id, your user-agent string (what browser you’re using on what platform), in some cases the HTTP referrer field (what link you clicked to get here), and information on how long the randomly-IDed visitor spends on the page. The random ID persists across visits.
So basically Parse.ly knows what stories these random IDs are visiting and how long they spend on them and where they came from.
In terms of what they do, you can look at the parsely site to see how they present it back to us. Parsely does have their own privacy policy, which you can read, but you have to note we’re very different from their other customers — they worked with us for almost a year to create this anonymized setup, so information they might otherwise be able to share, e.g. what other sites visitors to website A also visit, is NOT available to them in our case. https://www.parsely.com/tour/
Heya Ryan, This isn’t feedback, this is information: “The random ID persists across visits.” This is TRACKING — PERSISTENT TRACKING. I don’t think you understand how it works, but I’ll at least credit you for replying to people here (with canned speech, but still).
Thank you very much for listening as we all tell you this is PERSISTENT TRACKING (which by the way can be correlated in all sorts of ways) and you try to assuage our ‘fears’ after ten months of hard work and planning to make sure that this is something that you would not possibly consider changing your mind on.
I won’t be commenting again, so no worries as to replying. Just wanted to explain my confusion (and yours). Please don’t get upset with me.
Explain why.
Pretty simple that.
Why?
Why have trackers at all, except to track.
The Intercept tracks its viewers.
Why?
Ryan and Betsy why?
Come on the question is easy.
For you, the answer will be hard.
You track.
You want to know why someone comes to the Intercept.
You want to know how.
You want to know from where.
Why?
Why?
Why do you want to know?
Who else will know?
Who else will ask why, how, from where.
You have shown your hand here and its bloody.
Out damp spot.
I see I’m not the only one troubled by this. I have to agree I don’t want to be analyzed. Getting analytics on readers strikes me as a very slippery slope for TI to go down. We’ve all been wondering about the business model… this isn’t reassuring. Hadn’t really thought much about cloudflare or vimeo, but based on some comments below, I’m wondering.
What I find especially troubling is: “Beyond these architectural safeguards, Parse.ly has also agreed not to log IP addresses of visitors to The Intercept, not to infer or store geolocation data of visitors to The Intercept, and not to set or receive network-wide Parse.ly tracking identifiers from visitors to The Intercept — even though it should not have the technical capability to do any of these things in the first place.” Do you REALLY trust them to HONOR that promise?
I think TI should be way more careful about reader privacy.
Can someone give me some bit of clarity here. I clicked on that do not track me link, but I’m not sure of the “results” I was seeing… It seemed to be saying my browser wouldn’t support it, yet on the other hand, that I had enabled it. Can someone help here?
Hey there, You need to locate and follow the instructions for your particular browser in the top right corner of that site: http://donottrack.us/
Once you have it set up, you can go back to the site and it will tell you if it’s enabled.
The site does not include instructions for iPhone/iPad Mobile Safari, you can find instructions for those here: http://www.cultofmac.com/256690/enable-do-not-track-block-cookies-for-better-mobile-safari-privacy-ios-tips/
For Android Chrome: http://www.techrepublic.com/article/how-to-enable-chrome-do-not-track-for-better-mobile-privacy/
Thanks, Ryan Tate –
I will definitely go back there and check everything out.
It would have been better if you had spent your time reverting to the much better original look and commenting system rather than breaking the site even further by adding spyware and specious justifications.
Agree 100% !!!
While what you are doing sounds admirable to what many sites do with their analytics, I think it’s time to take a moment to consider why you want any information about your visitors at all, where they have come from, and what pages they have read, anonymous or not.
When I set up a site I naturally put code on the pages from a free stats provider. This was years ago, when my thinking was different and I felt I should know how many readers I get and where they came from. But with my interest in privacy in more recent times I find I have become much less interested in this information.
Now, I regard it as a kind of insult to my readers, even if I have only that one tracker showing up in Ghostery it is a reminder to myself that I am not upholding the values I espouse. So I decided to remove the tracking. I haven’t done all the pages yet, but I’ll get it done.
That’s why I’m surprised to read this on ‘The Intercept’ site. I think you should be above wanting to analyze your readership and just rest assured that you’re getting plenty of readers because your content is good. Far more than me, that’s for certain, and my content is probably also good, maybe even better in some ways, but hey, one guy has more luck than another, it’s the way of the world.
I see your point here. Part of our mission is to have an impact on society. Sometimes knowing that a lot of people read a particular post, or knowing if people really read the post all the way through or just bounced, or knowing if people get “hooked” on the site and come back, can help provide 1> feedback on whether we could package our content better, or if we are doing fine on packaging, 2> encouragement, particularly in the face of criticism on PR and social channels and in other media outlets, often from powers that be, 3> leverage with, and rewards to, our sources and other contributors, in terms of emphasizing the value in spending time talking to or working with us.
Being able to refine, repackage, have greater confidence in, and derive greater leverage from, our journalism helps us have a larger impact on society and reach more people. It gives us the potential to take our message farther.
If it were impossible to do this in a mannner that was compatible with our values, we would not do it. But it’s not so binary. Assigning you a random persistent ID, and stripping out information that could identify you uniquely in the real world, allows us to track you without knowing or storing any information on who you actually are, or information that could be used to derive who you actually are. That’s the goal, at least.
“But it’s not so binary.”
That’s the problem: you’re talking plausibly, openly, rationally. The ones going cray-cray see the world as mutually exclusive binaries.
Would it be accurate to state that this analytic system is not designed to raise profits in any direct way? You’re certainly not selling data. Seems like an internal system that has some “marketing” value in terms of encouraging further leaks, frankly — well, not just that, but participation with the site altogether. Why risk all if fourteen people are going to read about it. Etc. Not every leaker will be Snowden-level; some may get no platform other than this.
At least that seems part of the rationale.
Would it be too much, though, to just post the contract and tech stuff you have with the vendor just to shut up the paranoid? No idea if that’s a reasonable suggestion, mind you.
Anyway, my two cents is: seems kosher to me. Don’t be lying to us and keep it open and it’s all good. :)
“Would it be accurate to state that this analytic system is not designed to raise profits in any direct way? ”
Absolutely. We do not have advertising and no immediate plans to get any. I can’t definitively say we’d never have advertising, ever, or that knowing our audience size wouldn’t help get ads, but I can say the motivation was 100% editorial.
“Seems like an internal system that has some “marketing” value in terms of encouraging further leaks, frankly — well, not just that, but participation with the site altogether. Why risk all if fourteen people are going to read about it. ”
Correct :-) Or at least potential marketing value, once we learn more about the traffic.
“Would it be too much, though, to just post the contract and tech stuff you have with the vendor just to shut up the paranoid? ”
Two things on this: 1> Hopefully this was clear in the post, but the contract provisions are a failsafe in case the technical safeguards ever fail unexpectedly; at this point, they don’t have access to the things they’ve agreed not to retain :) But 2> this is a decent point and I’ll look into it.
Thanks for the comment.
It took two “journalists” to write this deceptive piece of propaganda.
Hang your heads in shame you sold out pieces of shite.
I am reminded how I was banned at Salon by Mr Greenwald after complaining (crapflooding) on Mr Greenwald’s articles about Ron Paul and the 39 trackers associated with that article.
Nothing has changed.
The First Look’s Intercept, proposes more not less Surveillance.
You do not sell anything, or do you?
https://en.wikipedia.org/wiki/Parse.ly
That’s who the intercept is in bed with
Dreamit.com
What business of yours is it, who reads what, how or why?
You have to be the biggest conners in history you wankers.
Shove your transparency where the sun dont shine.
Good morning, don’t you think you have overdone your night shift? Why than you use java script to comment here when you are so paranoid. Please Tom, go some place else with your rude crap.
The Intercept’s visual identity is almost identical to the ISIS flag if you place them both at a distance where you can’t read the text. Their anti-US stance are not of the same kind of malevolent intent, so it will be good if either one of them changes its visual identity in order to appear more individualistic.
Cloudflare is a BIG problem for a LOT of people. I genreally DO NOT visit sites hosted by them. Cloudflare also seems to take a very special interest in tor use… Because the feds are interested. When you say cludflare, I say “Feds”
Why why do I say that? Cloudflare seems to alway be in exactly the right place to spy on people of interest to the US government. They are known collaborators WITH the US government’s spying operations as well? See:
“What do imprisoned journalist Barrett Brown’s “Project PM” domestic spying research site, the supposedly secure email service “Hushmail”, the “New Pirate Bay” torrent site, a Ferguson organizing site, an ‘apology letter to Vladimir Putin’ site, and ISIS’ most popular chat rooms all have in common? They’re all being, or have been stalked by the Domestic Spying organization “CloudFlare”
http://auntieimperial.tumblr.com/post/119691050059
The problem with Cloudflare is that the second any company on the Internet becomes successful, it is beautifully poised to become a spy agency. Doesn’t matter what it is, what it does … if it has a chance to watch most of the user sessions on the web, everybody is going to want it, and when I say “everybody” I mean the kind of people who don’t take no for an answer.
This is a systemic problem.
I appreciate your attempt to balance privacy with your business interest of gathering metric data on visitors. In particular, I approve of the way you proxy visit tracks through your own servers and deliberately render them anonymous prior to sending them to Parse.ly.
Please keep up the good work.