3 ms·
> I don't see it as a retreat... You can add the use of "retreat" in the context of a value judgement to the list of disagreements, but the reframe into a lega
by sennight 5y ago
> I don't see it as a retreat...
You can add the use of "retreat" in the context of a value judgement to the list of disagreements, but the reframe into a legal argument makes our exchange largely pointless.
> ...unless I'm aware of a convincing case that the thing causes more harm than good...
Ah, yeah - we definitely won't agree on much of anything. I'm more of a first-principles/universality sort of guy... whereas it sounds like moral relativism is more your thing.
>> I think this is a case of straw-manning...
> I just haven't heard a convincing case that ads meets any of those criteria.
See what you did there? You created a strawman. There is a difference between "ads" and "Google using data to personalize ads". I've got no beef with the former, but you are arguing in defense of the latter.
- mgraczyk 5y agoI'm not a relativist, I believe morality is a real objective thing, but that it's socially constructed. Morality is more like language than it is like physics. To your last point, you can substitute one for the other, I stand by the original point in either case. I haven't heard a convincing case from first principles as to the wrongness of "Google using data to personalize ads". Generally the principle I hear in these cases is very close to "tracking or targeting is inherently wrong", which I reject.
- sennight 5y ago> Morality is more like language than it is like physics. Not as you've described it, as a real objective thing that is a social construct. I agree that it is a social construct, and I agree that it is "like language", because I'd describe morality as, at the very least, an internally coherent social construct. "more harm than good" has no logical constraint and easily permits the violation of every prior principal while affording no solid ground upon which to build, by design. > To your last point, you can substitute one for the other... Cool, so long as you're aware that there exists the possibility of data-hoard-free advertising - and that it was common prior to recent Internet centralization. > I haven't heard a convincing case from first principles... That is because, like ad-blockers, it isn't wrong. Good luck trying to construct on internally coherent system of morals that blesses the present state of data-broker panopticon while at the same time condemning ad-blockers. If you can do that with something beyond "might makes right" then I'll be impressed.