"Break free from Google ..." by purchasing Google hardware and using [software "based on"] Google software
Is it really "breaking free" from a company if the method of "breaking free" requires continued cooperation from the company
This is not to suggest using a modified version of Android isn't useful. This comment is not about GrapheneOS. (But there will be HN replies that will try to redirect focus to it anyway.) This comment is about claiming it's possible to "break free" from something while still remaining inextricably tied to it
In addition to using a custom ROM, there are methods of stopping the Pixel's attempts to "phone home" to the company that work even with the version of Android pre-installed by the company intact. However if a method requires software, e.g., drivers, or is "based on" software controlled by the company, then ultimately the company holds the cards. IMHO, this is not what it means to "break free"
Perhaps the most reliable method of stopping these connections to the company is one that does not rely on cooperation by the company. This is because if the company decides to stop cooperating, the method still works
For editing text on a screen, I prefer UNIX utilities ed, sed, ex/vi and custom filters written in C. The later can be used within ed or ex/vi via
:!filter
The slow, error-prone step is getting the text _accurately_ from the paper to bits in the computer. A personalised OCR that can recognise own handwriting might be helpful
The naivete or complacency of people who work for so-called "tech" companies that perform wanton, surreptitious data collection about computer users as their core "business model" is illustrated by the belief that what is significant for the surveillance target is how the data is used
Thus, a company performing data collection and sharing it with the government may trigger nerd rage whereas company performing data collection and using the data to help profile ad targets triggers nerd advocacy, i.e., attempts to defend the practice of data collection with "justifications" that have no limit in their level of absurdity
For the surveillance target (cf. the surveilling company), what is significant about data collection is not how the data is used, it is how the data _could_ be used, which is to say, what is significant about data collection is (a) the fact that data is collected at all, not (b) what may or may not happen after the data is collected
Moreover, despite equivocal statements of reassurance in unenforceable "privacy policies" and the like, (b) is often practically impossible for those outside the company and its partners to determine anyway
Hypothetical: Trillion-dollar public company A whose core "business" is data collection and surveillance-supported advertising services takes a nosedive due to unforseen circumstances that affect its ability to sell ad services. Meanwhile, billion-dollar public company B whose core business is data collection and surveillance services for goverments sees their business on the rise. Company A decides to acquire or compete with company B
There is nothing that limits company A's use of the data it has collected for whatever purpose the company and Wall Street deems profitable
As such, the significant issue for the surveillance target is (a) not (b)
Focusing on the fact that company B assists governments whilst company A assists advertisers is a red herring
"Currently, however, the Internet Archive does not disallow any specific crawlers through its robots.txt file, including those of major AI companies. As of January 12, [2026,] the robots.txt file for archive.org read: "Welcome to the Archive! Please crawl our files. We appreciate it if you can crawl responsibly. Stay open!" Shortly after we inquired about this language, it was changed. The file now reads, simply, "Welcome to the Internet Archive!""
"IME newpipe breaks every few weeks or so, presumably because of some youtube change / obfuscation."
Same experience. I have a commandline program I wrote to retrieve YouTube download URLs many years ago. On average it has always been more reliable than Newpipe. It continues to work when Newpipe fails and I can fix it quickly when there is a YouTube change
However I noticed recently Newpipe, the original, not SponsorBlock, old version, no updates, had been going many weeks without failure. When it eventually failed I was able to get it working again immediately by simply changing the www.youtube.com IPv4 address
reply