Hey everyone, I've been doing AI related development, training and building at a mid-to-large enterprise company for the last year. I wrote this article to offer my perspective on the ever changing landscape in AI currently. I'm interested in having more discussions about where we're headed as an industry and how this is impacting everyone.
>No, we wouldn't magically have developed modern computer-controlled battery packs of lithium ion batteries in 1920 if we just wanted it hard enough.
That's not what the commenter said. Don't put your interpretation of the words into theirs.
It is very feasible that the investment of 100-some-odd years of battery research and a marked non-future invested as deeply into oil and gas as we have now would have rendered our entire world vastly different. This is not a claim that the future would have happened sooner, but rather the events that unfolded and the research would have been different.
You can't be a one man band, but if the organisation supports it, you can achieve both. In my experience, it doesn't need to be a trade off. We can get close to the users and program to your hearts content. We need the right company and environment to want to invest in that for us. There's a great loop we can get into by doing things like offering demos of new features to clients weekly or bi-weekly. What I've seen happen is that we don't need to code as much because we know the specific problems that need to be solved, rather than loads of guesswork and trying to write blanket solutions for misunderstood problems.
Can you explain any of the technical details around this perchance? I'm super curious. I know that SO_REUSEPORT[1] exists but is that the only little trick to make this work? From what I've read with SO_REUSEPORT it can open up that port to hijacking by rogue processes, so is that fine to rely on?
This article doesn't add up the points to anything that solves for the given problem. Owning identity isn't solved by saying "don't trust ${third party}! Come trust ${my preferred third party}, it's better!" Any blockchain is still a third party that all parties involved with need to place trust in. It isn't somehow more or less trustworthy just because it exists.
> Many people, including myself, believe that the individual should be able to own their own identity.
Yes, this is nice wishful thinking, but on a global scale it's not really possible or feasible.
> OAuth2 should be used for what it was intended to, which is for a web service to provide another web service with a user’s data given that user’s consent. It should not be used as a global digital identifier because that’s too important to be owned by anyone but the individual themselves.
So, instead of OAuth being in the hands of FAANG[1] it's in the hands of ${blockchain-of-the-year}? How does moving the trust from a centralized company to a centralized blockchain change MY ownership? If I move everything away from FAANG to someone's blockchain, I have no assurance that chain will continue existing. If there's a flaw found in it and everyone moves to another chain, now what? Sure, we can make the same claim about FAANG not continuing to exist, but the point is there's no inherent advantage here, they're equal. FAANG are supported by millions of individuals and companies that are all, together invested in their success. There's no unilateral agreement on blockchains and I doubt there ever will be.
>With social recovery, instead of having to trust Google, you can choose who you trust, and instead trust a given set of friends, family, and services.
Again with the trust this and not that. All of my friends, family and other services need to then agree that they're all going to trust ${chain} instead of FAANG. It doesn't fix the problem. "the blockchain" isn't just one thing. Who's chain do we all shift trust to and from and based on what security? At least with Google I can rely on their security because if they end up with a breach of trust it's going to have a massive, real impact on share prices and consumer trust around the globe. That's incentive enough for me to rely on it day-to-day.
This article has some interesting tidbits but overall seems like just a baseless rally against FAANG by someone who knows very little about complex authentication or trust and security in the real world.
I don't mind the article, but one of the main points links a study from 2017 on school safety. Schools have become less safe since 2017. Since 2017, school shootings in the USA have skyrocketed [1].