Stop adding email tracking links to phone numbers
shkspr.mobi
shkspr.mobi
https://addons.mozilla.org/en-GB/firefox/addon/remove-fbclid...
Too bad ssl makes it hard to this at the router level (similarly to Pi Hole, but for HTTPS, not DNS).
As frustrating as tracking on the internet may be, you'll be a lot more resentful of your bank when they outright refuse to refund you any of your money, when it becomes clear that you opted to MITM your traffic, and your MITM host got compromised, resulting in the theft of your banking credentials.
There are many reasons why you should MITM your hosts, and there are many reasons why companies Bluecoat make large sums of money providing solutions to this problem. It is not a trivial problem and solution. In a much simpler space, "just run an FTP server" does not encompass what Dropbox does.
IMO it changes very little unless you’re living with other people.
From the perspective of dealing with my bank, my desktop being compromised by malware that results in bad things on my bank account, is something that could happen to a large number of their other customers. There will be policies in place, and there is a stronger defence that the money was stolen, and not that I was negligent in my security practices.
In the event that I opted to break the end to end encryption between my browser and their server with my own middleware, it will be argued that I am technically competent to manage my IT security properly, and that the bank cannot be faulted for my failures. They could try this defence on the first instance too, but given that one scenario impacts everyone, and the other only impacts people who have declared that they know better, I don't expect much sympathy for the latter group.
Which is why I do it, although I didn't start until DoH came around. Your comments are very well taken, though -- doing this certainly does bring an entirely new collection of problems with it.
But for some people (at least, for me), the cost/benefit of properly implementing a MITM is lower than the cost/benefit of remaining vulnerable to marketers and other tracking ne'er-do-wells.
https://addons.mozilla.org/en-GB/firefox/addon/clearurls/?sr...
"Recommended extensions are curated extensions that meet the highest standards of security, functionality, and user experience. Firefox staff thoroughly evaluate each extension before it receives Recommended status."
https://support.mozilla.org/en-US/kb/recommended-extensions-...
(presumably any extension like this has access to every URL you visit, which is enough to give me pause)
So you need to rewrite it before it gets clicked.
Hooks into chrome.webRequest.onBeforeRequest.addListener(...) to rewrite the request url at request time.
https://gitlab.com/KevinRoebert/ClearUrls
Edit: Looks like it's available for Chrome as well.
Have a button that says, "Call Me". If you have the user's number place a call to them and connect it to the destination number. If you don't have the user's number then collect it before placing the call.
There are much better tracking benefits and data collection for phone numbers.
If one is doing marketing and wants a person to call, there usually isn't much of a difference in conversion rate between a call to action like "Call 800-555-1212" and a form that will call you.
Source: An employer powered a metric shit-ton of advocacy call campaigns - like the STOP SOPA & PIPA stuff that took over the internet a few years ago.
With a typical bug code works not as author intended it to work. Numbers of such bugs can be reduced by testing and better tooling. But software development as an industry creates incentive to be overconfident and not to question own assumptions: developers who don't stop to read documentation/standards/gather information in anther way deliver more features in less time.
So if they're going through the motions they were told to, but don't fully understand why (and don't QA properly) this is what you end up with.
Could also be a system using the same logic of surfacing every <a> tag link for tagging before deployment (based on the utm names, there's at least some type of semantic logic to the naming that indicates a person was involved at some level and knew it was a phone number they were tagging).
[1] https://docs.google.com/spreadsheets/d/1xDl9GdpuWB9kzXzWxKj3...
Most WYSIWYG email builders built into these tools are pretty painful to begin with, and especially so when content with links is copied from one tool to another. (Think of the poor marketing person being given various blurbs in Word, Gmail, etc. and told to make a functional email out of them while having to use a shitty WYSIWYG.)
Beyond that, most marketers I know have absolutely zero input into what email platform they use. Many organizations are stuck using antiquated, broken tools (I'm looking squarely at you, Eloqua) because of complex existing integrations they have set up.
I don't see this changing any time soon. Few marketing orgs have the IT budget and/or on-staff knowhow to fix this problem. Nor are they likely to be aware of it. Nor, for that matter, are they likely to prioritize it over the many more pressing things they could be fixing that more directly affect revenue.
Also take a moment to consider if you to track what links people are clicking on at all.
Like if the first thing everyone does is open their "messages" ... maybe give them an option to just do that automatically.
Send that info to a third party ... no I'm not down with that.
I was speaking more generally about what folks think of as "tracking" being a really wide topic that can involve totally innocuous "tracking" that is actually to their benefit, and far worse that is more of a dark pattern.
utm parameters are generally for any time you control the url on a foreign site. You want to know that this variant of this ad on this website got this many clicks.
There are plenty of use-cases where having tracking metadata is useful. I can't think of a single one where having a dialer interpret a `?` is.