> Can anyone give a coherent explanation of why the intention behind Grokipedia is... bad?
If one takes Musk at his word that his only intention is to provide an alternative Wikipedia, there is no problem; it is simply an alternative Wikipedia with a different editorial bias. If one believes that Musk is the modern day Joseph Goebbels, then his stated intention is likely false; Goebbels' strategy was to provide 60% truths and 40% lies, which is expressly facilitated with something like Wikipedia (Grokipedia in Musk's case). There are many possibilities in between the two which seem to suggest the likely fallibility of Grokipedia. All things considered, especially Musk's track record with the truth, it is reasonably assumed that this is a propaganda tool more than an education tool.
Please remember that you asked for "a coherent explanation" rather than something you will necessarily agree with. Put yourself in the shoes of someone who believes the opinions in this explanation.
"an alternative Wikipedia with a different editorial bias"
I think the stated goal is to reduce bias, not to just rotate it to another view. Of course some things are difficult to reduce to facts, but plenty of others are not.
There is an objective truth, but many people know it by the alternative name "left-wing bias". You see the problem? (Or, if you prefer, pretend I wrote "right-wing bias")
Flat earth may be a better example, since everyone HN user is hopefully a round-earther. Suppose you are talking to someone who thinks the earth is flat. Flat-earth belief used to be a joke, one or two decades ago, but now these debates actually happen with some regularity and the round-earther never wins. Do you think you could do better than every other round-earther? The proverb about playing chess with a pigeon is relevant here. What you call "objective reality", they all "round-earth bias".
> You're saying there is no objective truth, and there is not point in trying for it?
Not exactly but kinda. I'd like to think it's less nihilistic than this makes it sound, at least. Here's a definition which I think is a good working one:
> A preference or an inclination, especially one that inhibits impartial judgment.
To attempt to answer the question directly, there is still agreed truth, such as in the practice of law where a judge decides what facts are relevant to the case. In that context, the legal framework is, barring appeals, simply to accept the judge's "truth"; at least, once they've made their decision. To prevent that from devolving into madness, we train lawyers and judges to understand the meaning of certain legal terms so that they all have a similar understanding and can argue using terms which have an agreed-upon meaning. Still, there is much bias in this. If there wasn't, there would be no need for lawyers. The closest thing we get to objective truth is something like 1 + 1 = 2, but it's still agreed truth.
There are facts, but even the reporting of facts can be biased. Which facts you report, the context in which they're reported, the details that are included, the order in which they're reported. These choices have an effect on how the facts are interpreted; they are a bias. If I want to say, "The sky is blue and the Blue Man Group are entertainers," but you want to say, "Mars is called the red planet and Mars Attacks! is a movie," who is being less biased? Is a news article that reports, "Mars is called the red planet and the sky is blue," more or less biased than either of the other statements? Is one more objective than the other? How come the author didn't mention the tall smurfs if they're unbiased? Surely reporters aren't required to report all of the facts in every instance. How do they, unbiased, decide which facts are relevant to their report?
With the idea of "reducing bias", it is obvious that in order to verify that bias has been reduced, one must accurately measure "how much" bias there is. So how do you do that? Further, how do you do it in an unbiased way? How do you even know if you're doing it in an unbiased way without an unbiased measurement?
Another way to think about it is in terms of opinions. How do you express an opinion without bias? How do you express the same opinion with less or more bias? Is, "I don't care," less biased than, "I care a lot."? Is, "This makes me angry," more biased than, "I'm indifferent to this."? I would say they're equally biased. At least, I can't come up with a good reason to claim that indifference is not a bias. So, neutrality is a bias. I'm not sure what isn't a bias. I guess a bare fact. But, again, that doesn't mean the presentation of a fact is not without bias.
To give an example, it is a fact that Donald Trump is serving his second term as President. Do people often present that fact without bias? It's also a fact that Donald Trump has not served two consecutive terms as President. Is that presented without bias? Both of those decisions are biased: minimally, why would one present either of these facts in a given context?
Or an example of indifference: if someone says, "I don't care about Biden's capability to serve as President," does their indifference suggest a lack of bias? Is it more or less biased than, "I think it's bad that Biden is President given his apparent senility."? Is it more or less biased than, "I think it's good that Biden is President given his apparent senility."?
Biases are simply different or the same, neither lesser nor greater. I don't know how I could even go about measuring bias without introducing bias.
Can anyone give a coherent explanation of why the intention behind Grokipedia is... bad?