Yeah, I'm mildly confused by what the sell point here is. I sort of assume that any loading of remote resources automatically represents tracking information for this use-case, every email can be trivially individualized to the receiver and in turn I don't see how it's feasible to prevent resource tracking. You either load remote resources, or you don't. I guess if techniques remain relatively unsophisticated for email and few people bother then individuals might slip under the radar, though partial blocking itself represents a signal. But unlike the web individual messaging remote resources seems like an arms race users would be destined to lose.
The only thing I could see really working would depending on a really big service that processed every received email immediately and proxied all remote resources without any user interaction. Plenty of challenges there too though, and harder to make it work with regular clients. Maybe dynamically rewriting the email to point to the proxied resources? Or automatically rendering/inlining them?
>Maybe dynamically rewriting the email to point to the proxied resources?
AFAIK gmail already does that. It prevents people from knowing your IP address, but they still know whether you opened the email or not, since it's lazy loaded. Preferably there's a email service that always loads the images and stores it, so you can view images without any spying.
It does, but I guess this is for people who like images most of the time, and just want to avoid the tracking without sacrificing ux. I'm guessing it's inspired by https://hey.com/spy-trackers/, which also has a similar feature that works on a blacklist basis (presumably).
I'm curious why they wouldn't also eager-load _other_ images.
As soon as an email is received (regardless of whether the recipient account exists or not), load the image from the remote host.
This also degrades the spying capabilities of third parties. If they track image loads to translate this to "email opened", they'll immediately get pure false positives from any email they sent to @hey.com -- even if the account doesn't exist.
For anyone not using it, the great thing about Little Snitch is being able to set just the precise port numbers and remote hosts you want a given app to access while blocking everything else, rendering plug-ins like MailTrackerBlocker unnecessary. You can also set profiles for different networks, and even have them switch automatically: https://help.obdev.at/littlesnitch/adv-profiles .
> I'm not sure if these protections extend to those using third party email clients with these services
In Fastmail's case, those protections don't extend. I've tested this and Fastmail rewrites the image URLs when viewing through their webmail client but passes the message unmodified when fetching via IMAP.
I understand why they'd do this but I'd love to have a setting that says to do this privacy rewrite when using an external client.
Thanks for clarifying. Might be worth shooting them a message to ask for this feature. They've always been good about fixing bugs and at least responding to requests, even if they decide not to pursue it.
Some 3rd party antispam products also follow all the links they can find gauge their intent.
My company's marketing department has complained about this after noticing that our own clients have a 100% click rate. It's because we are reselling Barracuda.
No, it does not. By downloading the images to Gmail's/mail providers cache, they are effectively giving the impression that the user opened the mail just when it was delivered. The mail provider then shows their cache when you open the mail, the tracker will not work.
Neither provider is claiming that they download images as a matter of course. From the provided links:
Gmail: "Sometimes, senders may know whether you've opened an email that has an image."
Fastmail: "Clarified that the image server cannot see where the request came from. It may still be able to determine who the request came (ie email address validation) if the image URL has some kind of tracking data in it."
> Clarified that the image server cannot see where the request came from. It may still be able to determine who the request came (ie email address validation) if the image URL has some kind of tracking data in it.
They will know if the mail reached or not, but they will not know if the mail was opened. Because, it is possible that you never open the mail but the image is still downloaded to the fastmail server.
This plugin is awesome but one thing I've noticed lately is several financial institutions use tracking pixels to verify e-mail addresses and if you don't open an e-mail in a way that loads the pixel once every so often, they drop you from paperless statements.
I've run into this with Capital One and Discover. Every year I have to log in and click "yes I want paperless statements" and then 11 statements later I get one in the mail with a message about "we miss you, you're not opening our e-mails so here's a paper statement unless you tell us you want e-statements and you open our e-mails." Never mind that I log into the mobile apps and web-based account management systems multiple times per month...
No, they e-mail to tell me that a statement is available for viewing on their web site or mobile application. But if I don't "open" one of those e-mails every 12 months, they kick me off of paperless statements and go back to sending them via regular mail.
What's the rationale for your bank doing this? The litigation-heavy environment of its jurisdiction? Someone once sued a bank due to a loss caused by not knowing what was in their bank account, and they demonstrated that they were not 'receiving' their statements?
Just making sure but do you ever click through on an email link or do you just use it as a notification and then go and log in? I’m curious if they just need to be looking at click + open events or if that wouldn’t even help in this scenario.
To be honest, this seems like a decent check for banks to make. Based on my experience with users and the emails attached to their accounts, not all of them have an attachment to an email address that you or I might have (I haven't switched addresses in decades.) They also somehow frequently lose access to inboxes.
The banks could verify if the user is checking their mails by asking them to reply to the email or click on a link explicitly to confirm the receipt? Tracking with pixels... there's so much wrong with it that it needs a blog post of its own, and thankfully someone did just that: https://mikeindustries.com/blog/archive/2019/07/superhuman-i...
Wow! I've been trying to figure out for a year why my financial institution hasn't been able to deliver me electronic statements. Every month or so it's been auto-switching back to snail mail. I even switched email services and did extensive tweaks to my spam blockers to try to figure out what was happening. Them looking at the email image-loads never occurred to me... thanks for sharing that!
This is horrific UX. Mi bank is almost 100% paperless, and once a month they send this exact mail. I read the subject (something like "You have a message with your monthly statements") and I delete the mail without opening. They never punish me for refusing to get tracked.
Pixel tracking is awful, if you do it at least show some decency and don't blame the victim if they don't fall in your trap.
36 comments
[ 2.8 ms ] story [ 86.0 ms ] threadThe only thing I could see really working would depending on a really big service that processed every received email immediately and proxied all remote resources without any user interaction. Plenty of challenges there too though, and harder to make it work with regular clients. Maybe dynamically rewriting the email to point to the proxied resources? Or automatically rendering/inlining them?
AFAIK gmail already does that. It prevents people from knowing your IP address, but they still know whether you opened the email or not, since it's lazy loaded. Preferably there's a email service that always loads the images and stores it, so you can view images without any spying.
As soon as an email is received (regardless of whether the recipient account exists or not), load the image from the remote host.
This also degrades the spying capabilities of third parties. If they track image loads to translate this to "email opened", they'll immediately get pure false positives from any email they sent to @hey.com -- even if the account doesn't exist.
If you have it turned off it won't load remote images...
EXCEPT: if you forward an email - it will load them all, all of them. This is especially troublesome if you want to forward a phishing email to IT.
(thank you little snitch. also I am not using the latest catalina apple mail)
Your friendly email inspector who needs to explain this x times a day.
otherwise, you can just disable external image loading
See Gmail: https://support.google.com/mail/answer/145919?p=display_imag...
See Fastmail: https://fastmail.blog/2014/09/16/better-security-and-privacy...
Although, I'm not sure if these protections extend to those using third party email clients with these services.
In Fastmail's case, those protections don't extend. I've tested this and Fastmail rewrites the image URLs when viewing through their webmail client but passes the message unmodified when fetching via IMAP.
I understand why they'd do this but I'd love to have a setting that says to do this privacy rewrite when using an external client.
My company's marketing department has complained about this after noticing that our own clients have a 100% click rate. It's because we are reselling Barracuda.
If you can load all images at reception time (and put them somewhere), maybe that gives less information.
Of course it might give them a clue about your spam processing if it comes first and those don't get loaded.
Gmail: "Sometimes, senders may know whether you've opened an email that has an image."
Fastmail: "Clarified that the image server cannot see where the request came from. It may still be able to determine who the request came (ie email address validation) if the image URL has some kind of tracking data in it."
They will know if the mail reached or not, but they will not know if the mail was opened. Because, it is possible that you never open the mail but the image is still downloaded to the fastmail server.
I've run into this with Capital One and Discover. Every year I have to log in and click "yes I want paperless statements" and then 11 statements later I get one in the mail with a message about "we miss you, you're not opening our e-mails so here's a paper statement unless you tell us you want e-statements and you open our e-mails." Never mind that I log into the mobile apps and web-based account management systems multiple times per month...
I kind of wish banks didn't put links in their email.
Pixel tracking is awful, if you do it at least show some decency and don't blame the victim if they don't fall in your trap.