I would caution against assuming that legal analysis is required to match technical analysis.
At the least, any legal proceedings would involve a lot of argument over exactly what the words in the statute mean, what the intent of the legislature was, and whether reasonable attempts were made to meet them. And these things will often be decided by decidedly non-technical judges. And that's even assuming that the legal system tries to deal with things in an unbiased manner, which is not a safe assumption now, if it ever was. Given that there is also a large number of other legislation targeting the likely involved companies, not to mention other legal proceedings, there is certainly the potential for spill over effects as well. I would not want to rely on an argument in court that "label" only means label in the Gmail sense, and doesn't apply to a header, subject prefix, category, keyword or folder.... or color or whatever. Or that junk didn't mean spam. Brandon On Sat, Jul 30, 2022 at 9:19 AM Jaroslaw Rafa via mailop <[email protected]> wrote: > I'm not an American, so it's basically "not my fairy-tale" (as we say in > our > country), but I can't stop wondering at the use of the word "label" in the > proposed regulation. > > I already asked (in a bit sarcastic tone) in one of the previous emails, > what > is a "label" in context of email in general. Because there's simply no such > thing: if you look at email protocols, or operation of server software, > there's no such thing as "label". Spam filters may add a *header* > indicating > that a message is spam, but is a header a "label" or not? And if the spam > filter does not add a header, but just directly moves the message (using > sieve for example) to spam *folder* on the server? Is a *physical folder in > the filesystem* a "label" or not? > > "Labels" per se exist only in some implementations of MUAs, most notably in > Gmail web interface. So either the regulation is targeted particularly at > Google, or it's authors never saw any other email system than Gmail and > imagine that a "label" is some universal thing (which wonders me, because > don't they have their internal email systems at Congress or governmental > institutions?) > > Another question is, how are the operators supposed to distinguish > political > messages from non-political ones? The only reasonable method that comes to > mind is submitting by political senders in advance to the operators a list > of sender addresses that shouldn't be filtered. Operators can then > whitelist > them. > > But can't compiling a list of such sender be considered some form of > "applying a label"? In that cse the regulation becomes self-contradictory: > in order to comply with the regulation and "not apply a label" to political > messages, you have first to "apply a label" to senders of those messages, a > label that says "don't apply any label to messages from this sender". > > Just some doubts that - at least for me - show that this entire proposal > doesn't make any sense. > -- > Regards, > Jaroslaw Rafa > [email protected] > -- > "In a million years, when kids go to school, they're gonna know: once there > was a Hushpuppy, and she lived with her daddy in the Bathtub." > _______________________________________________ > mailop mailing list > [email protected] > https://list.mailop.org/listinfo/mailop >
_______________________________________________ mailop mailing list [email protected] https://list.mailop.org/listinfo/mailop
