The article links to Mozilla’s press release / blog entry about the acquisition of Anonym [0]. It’s pretty dystopian reading. The last three paragraphs and the summary of Anonym are more worrying than anything else I’ve read on this so far:
> This acquisition marks a significant step in addressing the urgent need for privacy-preserving advertising solutions. By combining Mozilla’s scale and trusted reputation with Anonym’s cutting-edge technology, we can enhance user privacy and advertising effectiveness, leveling the playing field for all stakeholders.
I can only interpret this as the urgent need is money, and wants to sell its "scale and trusted reputation". Mozilla has been down this road before. It was not good for them.
> Anonym was founded with two core beliefs: First, that people have a fundamental right to privacy in online interactions and second, that digital advertising is critical for the sustainability of free content, services and experiences. Mozilla and Anonym share the belief that advanced technologies can enable relevant and measurable advertising while still preserving user privacy.
This is some pretty weak wording for a press release. The economics of the situation are that advertising will always trump privacy. Researchers have successfully de-anonymized anonymised data sets, including medical records. Why would these data be any different?
> As we integrate Anonym into the Mozilla family, we are excited about the possibilities this partnership brings. While Anonym will continue to serve its customer base, together, we are poised to lead the industry toward a future where privacy and effective advertising go hand in hand, supporting a free and open internet.
Anonym’s customers are advertisers, right? The same people who for decades poured money into eroding that free and open internet that we had…
> About Anonym: Anonym was founded in 2022 by former Meta executives Brad Smallwood and Graham Mudd. The company was backed by Griffin Gaming Partners, Norwest Venture Partners, Heracles Capital as well as a number of strategic individual investors.
Well, it seems Anonym, Smallwood and Mudd had a nice piece about them written in the Wall Street Journal [1]. From the second paragraph:
> Graham Mudd and Brad Smallwood each spent more than a decade building Meta’s advertising system, which allowed the company to offer granular data about how ad campaigns worked with individual users, often by tracking their web and mobile activity.
For those interested in using AES with reduced rounds as a PRNG, it is covered in the paper "Parallel Random Numbers: As Easy as 1, 2, 3" by John Salmon et al.
If you look up a data sheet for any 7483 variant, it is immediately notable: the first lines are something like "4-bit binary full adder" and the description says "…accept two 4-bit binary words" If you skip to the next paragraph, you’ll see they are onto transistors. Ideally the author could show a data sheet, but there may be reasons they can’t.
I found the material that followed to be a clear exposition of the fundamentals – this is as somebody who has tried to learn electronics over the years but nothing ever stuck. This did a better job than anything else at a pace that is about right for me.
But YMMV and I personally relish being in a room full of people who are smarter than me.
> We designed our initial solution as a distributed system using serverless components (for example, AWS Step Functions or AWS Lambda), which was a good choice for building the service quickly.
The message seems more that they outgrew AWS lambda but that lambda was a good choice at first.
I’ve been asking questions about F# after a while of using OCaml. I’ve found GPT-4 to hallucinate syntax that doesn’t exist, such as being able to do open <modulename> in a function. It’s actually something I assumed too, because you can do this in OCaml with "let open <modulename> in", but you cannot do this in F#.
Despite this, it’s still been a lot more helpful than searching, is mostly correct, and is my go-to resource.
Had to admit I came to the same conclusion when I read this part of the readme:
> The NC license permits "use by any charitable organization, educational institution, public research organization, public safety or health organization, environmental protection organization, or government institution is use for a permitted purpose regardless of the source of funding or obligations resulting from the funding."
But on reading the license, personal noncommercial use is fine too. It would have helped me if that info had been in the readme!
> A time field is an unsigned 4-octet number containing the number of seconds elapsed since midnight, 1 January 1970 UTC.