diff --git a/Ghostery.md b/Ghostery.md
index 95b02eb..b1a4271 100644
--- a/Ghostery.md
+++ b/Ghostery.md
@@ -5,7 +5,7 @@ Last updated: 8 August 2020 | [download](https://github.com/ghostery/ghostery-ex
Ghostery helps you browse smarter by giving you control over ads and tracking technologies to speed up page loads, eliminate clutter, and protect your data.
-### Why Betterfox Loves Ghostery ❤️
+### ❤️ Why Betterfox Loves Ghostery
* Blocks ads and trackers effectively. Annoyances too!
* Deploys AI and heuristics to catch trackers that are **not** on popular blocklists. (And if a tracker isn't blocked, then the request is modified and anonymized to prevent site breakage.)
@@ -100,16 +100,16 @@ Privacy Badger is, to the best of my knowledge, the only other popular extension
For example, Privacy Badger will try to learn over time which domains are "tracking" you then block these in the future, based on locally analyzed information. In contrast, Ghostery's anti-tracking is working out-of-the-box and privacy protection is derived from a global knowledge of trackers on the Web.
-Another big difference is that Privacy Badger will block requests, but Ghostery's anti-tracking is able to drop unsafe data-points from requests without having to block them completely (e.g. fingerprints, tracking cookies, unique ids, etc.); this in turn leads to much lower breakage of websites. To be clear, Ghostery also employs a traditional content blocker, but the privacy of our users does not depend on us blocking all requests (which is impossible by the way, unless you want to break lots of websites). So the anti-tracking will sanitize any request which was not already blocked, to ensure privacy is not at risk.
-
-[](https://postimg.cc/Q9gtLZY2)
-
https://s3.amazonaws.com/cdncliqz/wp-content/uploads/2016/07/08101643/cliqz_whitepaper_tracking1.pdf
+Another big difference is that Privacy Badger will block requests, but **Ghostery's anti-tracking is able to drop unsafe data-points from requests without having to block them completely** (e.g. fingerprints, tracking cookies, unique ids, etc.); this in turn leads to much lower breakage of websites. To be clear, Ghostery also employs a traditional content blocker, but the privacy of our users does not depend on us blocking all requests (which is impossible by the way, unless you want to break lots of websites). So the anti-tracking will sanitize any request which was not already blocked, to ensure privacy is not at risk.
Also, a heuristic approach like Privacy Badger is limited by just having local knowledge. In many cases it will not know if data sent is unique to a user (this can only be tested by opening another browser and checking if a different value would be sent). Thus some kind of collaboration is required between users to determine what data is safe, and what is not—and this is the method Ghostery's anti-tracking uses.
We do identify potential user-identifiers (i.e. any value which would allow to identify a user uniquely over time) if only one user is sending such data. The assessment is done as a quorum, where only data that a lot of users are sending is considered safe, since it could not be used as a way to link records by a third-party, hence track. To do this only with local information is impossible, and while it can offer a good degree of protection, the collaborative effort implemented as part of Ghostery is much stronger.
-Also, as explained in my original answer, all messages are anonymized and no record linkage can be done on the server side (we have no way to know if two messages come from the same users). We wrote extensively how this is possible in our two blog posts about Human Web and our anonymization network layer. Of course, this means that there is no unique identifier attached to messages.
+[](https://postimg.cc/Q9gtLZY2)
+
https://s3.amazonaws.com/cdncliqz/wp-content/uploads/2016/07/08101643/cliqz_whitepaper_tracking1.pdf
+
+Also, all messages are anonymized and no record linkage can be done on the server side (we have no way to know if two messages come from the same users). We wrote extensively how this is possible in our two blog posts about Human Web and our anonymization network layer. Of course, this means that there is no unique identifier attached to messages.
Last but not least, yes the data you mention is useful for building features that are yet to come. For example, we needed data before we could launch the tracking protection feature few years back. There is a chicken and egg problem. Some seem to be very focused on the data part, without attempting to evaluate if that data compromises the privacy of the users in any way. Sorry, but it is not always the case that data implies lack of privacy, we wrote about it [here](https://0x65.dev/blog/2019-12-02/is-data-collection-evil.html). [^6]