1,946 karma · joined September 27, 2009
Unconditionally embracing nonviolence is not, properly, a goal. Nonviolence is a strategy. It must be conditional on its effectiveness.
It’s blindingly obvious that they are here merely to gum up discussion in endless loops that go nowhere under the guise of politesse and question marks, as if they’re just confused and asking clarifying questions.
The thing about assuming good faith, which I agree with in principle, is that you have to be willing to accept your own judgment of when someone is arguing in bad faith and disengage.
I think consistency is more important than perfection. If you use something like vw_ as recommended in the sibling comment, that's fine, then try to apply it to all views. (Without being overly strict; sometimes there are good reasons to defy conventions. They're conventions, not laws!)
Just keep in mind that a view is both code and a data contract just like a table definition is. All the usual best practices around versioning, automated deployments, and smooth upgrade paths apply. As soon as an application or downstream view, function, etc relies on that view, changing it risks disruption or breakage. Loud breakage if you're lucky, silent data corruption if you're not!
I also think there are unanswered questions about reliability, cost (dollar and energy), and AI business models; I don't think OpenAI can burn $2+ to make a dollar forever.
I tried using enterprise chat gpt to write a query to load some json data into a data warehouse. I was impressed with how good a job it did, but it still required several rounds of refinement and hand-holding and the end result was almost, but not quite, correct. So I'm not coming at this from the perspective of hating LLMs a priori, but I am unimpressed with the hype and over-selling of its capabilities. In the end, it was no faster than writing the query myself, but it wasn't slower either, so I can see it being somewhat helpful in limited conditions.
Unless the technology makes another quantum leap improvement at the same time the price drops like a stone, I don't see LLMs coming anywhere close to your claim.
That said, I expect to see a huge amount of snake oil and enterprise dollars wastefully burned on executive pipe dreams of "here's a pile of data now magic me a better business!" in the next few years of LLM over-hyped nonsense. There's always a quick buck to make in duping clueless execs drooling over replacing pesky, annoying, "over-paid" tech people.
They do make idiot managers happy though, because look at all those features and scrum points they “finished!” That’s why armies of them will always be there in the software ecosystem, blithely and ignorantly a net drain on whatever unlucky company is currently employing them, busily making a mess for more diligent programmers to clean up for the rest of eternity. It’s called “job security.”
Try to think of it from leadership's perspective. For the sake of simplicity, let's say there are 100 similar initiatives proposed each year, each costing $1 million. That's $100 million per year in pure costs to the business (ignoring amortization, tax shenanigans, etc). Even for a bank, that's real money, to be strategically spent and not wasted.
With that setup, making leadership feel the pain and understand at a visceral level why this particular million dollars is well spent, is all at once a good strategy, rational for leadership, rational for IT, and a healthy-ish dynamic for the business.
Aside/tangent: In a very real sense, executives are managers of the finite time, employees, expenses, etc. of the business. Now, you could, and I would, argue that they're nearly always not especially good managers, and that the purely rational choice from "the company's" perspective is to pay them more like regular employees and not so lavishly. However, "the company" doesn't make decisions, people do, with all their political games, incentives and self-interests. Since executives control the flow of information/decisions/resources/money in the company, they siphon off an out-sized share for themselves, acting as a (perhaps inevitable?) parasite on the host company. Fixing this problem is left as an exercise to the reader.
You can override the REQUESTS_CA_BUNDLE env var to point to your own certs.