To be clear, it is possible to defend yourself, but ofc no defense is perfect. For illegal activities that might attract federal attention? best to avoid them at all costs. But for general privacy in your day to day life that isn't practical for most people, and "every cellphone is snitching on you" is way too reductive, even if it's true in a sense.
For example the person you are replying to mentions Linux phones, many of which publish schematics and make hardware kill switches for the discrete modem a big selling point. That still isn't perfect protection, but it would protect you from a stingray... and if you have a thought out threat model, you can make the choice whether or not that device is appropriate for your situation.
Let me be clear first: If you want to get rid of advertising, then yes your advise is OK. If you want to defend against the Surveillance system, it's not close to adequate. This is the fundamental gap I'm trying to address.
I understand where your heart is at, but you are making a mistake. Free/Libre software is about Freedom, and from that guarantee we can build other guarantees about security and privacy. However Freedom itself does not guarantee security nor privacy. Freedom is also the freedom to shoot yourself in the foot.
To be clear, it is possible to defend yourself, but ofc no defense is perfect.
There is a perfect defense: Don't use technology. Much of this advise is trying to use technology to fight technology. It's a rabbit hole that has no bottom, and the best defense is to not play. The problem is attack surface. Technology is incredibly complex and is chattier than your extroverted :LIB: friend at brunch, and boy howdy do people love to listen! You can reduce this attack surface, but it never goes away as long as you are using technology.
But for general privacy in your day to day life that isn't practical, and "every cellphone is snitching on you" is way too reductive, even if it's true in a sense.
Here's the trail crumbs you might make on the Web as you browse each and every website:
DNS request - sends URL domain/hostname (www.hexbear.net, for example), collects IP and timestamp. Your ISP is often the default DNS, so they are collecting this information. Google (8.8.8.8) and Cloudflare (1.1.1.1) as well.
1st party HTTP(S) request - Encrypts body but sends URL domain/hostname in the clear across the network, collects IP and timestamp.
3rd party request - Usually advertising, but also could be security (Sign In With Google, Okta, etc), collects IP and timestamp
3rd party cookies - Sent and updated with every request to that domain (Amazon cookie to Amazon.com, FB cookie to Facebook.com, etc), collects IP and timestamp
1st party advertising - Think Amazon's "Customers also bought...", has full access to your request, collects IP and timestamp and User-Agent.
Logs - Usage data about what you do on the website, both front-end and back-end, collects IP and timestamp
Telemetry - Usage data about what you do with your app, collects IP and timestamp
You can use custom software for #3 and #4 on the device (most of the advise here), but do you block google.com? You can use a network DNS blocker (e.g. Pi-Hole) for #1, #3, #4, and some of #7, but that only works on networks you control. VPNs advertise as solving #2, but that's pure ideology; it only moves where the routing traffic goes and still can log information in transit.
This also ignores data brokers who buy all of this information and compile it together.
And this is just the advertising/surveillance defense against tech companies. I haven't even touched or defense.
if your threat is state-level actors your computer security is approximately moot and maybe you should spend your money on laywers and having a discreet way out of the country
Our research for this report involved interviewing experts on this issue and reviewing approximately 150 publicly available documents covering awards, solicitations, requests for proposals, and related information on contracts. We found significant evidence of agencies exploiting loopholes in existing law by purchasing data from private data brokers. The practice has prompted scrutiny from government watchdogs as well as members of Congress (Tau, 2021a; Wyden, 2021).
The problem is a byproduct of the lucrative private market for personal data, where many companies that offer online services collect, analyze, and sell data about individuals using those services. This data is aggregated by companies called ‘data brokers’ that typically lack any direct relationship with the individuals whose data they collect and sell, but may accumulate personal data from multiple sources with varying degrees of granularity, ranging from anonymized trends to the specific locations of individuals at specific times. Advertisers, retailers, and other companies may then seek access to data for varied commercial purposes.
As our research demonstrates, law enforcement and intelligence agencies are among the customers of some data brokers, spending millions of dollars to gain access to private sector databases which often contain very sensitive and very personal information on individuals.
To be clear, it is possible to defend yourself, but ofc no defense is perfect. For illegal activities that might attract federal attention? best to avoid them at all costs. But for general privacy in your day to day life that isn't practical for most people, and "every cellphone is snitching on you" is way too reductive, even if it's true in a sense.
For example the person you are replying to mentions Linux phones, many of which publish schematics and make hardware kill switches for the discrete modem a big selling point. That still isn't perfect protection, but it would protect you from a stingray... and if you have a thought out threat model, you can make the choice whether or not that device is appropriate for your situation.
Let me be clear first: If you want to get rid of advertising, then yes your advise is OK. If you want to defend against the Surveillance system, it's not close to adequate. This is the fundamental gap I'm trying to address.
I understand where your heart is at, but you are making a mistake. Free/Libre software is about Freedom, and from that guarantee we can build other guarantees about security and privacy. However Freedom itself does not guarantee security nor privacy. Freedom is also the freedom to shoot yourself in the foot.
There is a perfect defense: Don't use technology. Much of this advise is trying to use technology to fight technology. It's a rabbit hole that has no bottom, and the best defense is to not play. The problem is attack surface. Technology is incredibly complex and is chattier than your extroverted :LIB: friend at brunch, and boy howdy
do people love to listen! You can reduce this attack surface, but it never goes away as long as you are using technology.
Here's the trail crumbs you might make on the Web as you browse each and every website:
You can use custom software for #3 and #4 on the device (most of the advise here), but do you block google.com? You can use a network DNS blocker (e.g. Pi-Hole) for #1, #3, #4, and some of #7, but that only works on networks you control. VPNs advertise as solving #2, but that's pure ideology; it only moves where the routing traffic goes and still can log information in transit.
This also ignores data brokers who buy all of this information and compile it together.
And this is just the advertising/surveillance defense against tech companies. I haven't even touched
or 
defense.
if your threat is state-level actors your computer security is approximately moot and maybe you should spend your money on laywers and having a discreet way out of the country
That's not necessarily true. Police are purchasers of this data from data brokers. It's state surveillance without any need for a warrant.