SkyBelow

23 ms·

181.
by SkyBelow 2y ago
>I'm not claiming that we were smarter, but by sheer coincidence, most people, even folks like me who skipped school, had very solid fundamentals. Higher barrier of entry should statistically lead to less people making it past and t
182.
by SkyBelow 2y ago
Is a single order of magnitude within a reasonable range of error for the calculation? Some cases that is a massive miss, but other cases that might be within the expected statistical range. In the former cases it implies something entire
183.
by SkyBelow 2y ago
Based on our understanding of human psychology, despite it being as incomplete as it is, it would seem reasonable to argue that this trend, if it does exist, would not be a binary rule but one whose strength depends upon numerous factors th
184.
by SkyBelow 2y ago
The cynic in me is wondering if this is them attempting to swap to another tactic to establish a different relationship with the same end goal.
185.
by SkyBelow 2y ago
>It's like people who get mad at customer service people Except the customer service people willingly accept a job to act as an in-between so that any appropriate anger at a company is directed at an employee. To what extent do com
186.
by SkyBelow 2y ago
>But you trust scanlation groups? Neither will give you perfect, professional-level translations. I often prefer fan level over professional level because they are targeting different audience. As far as quality goes, there is a range a
187.
by SkyBelow 2y ago
Have we hit the limit of the analogy, or have we hit a limit in our understanding? Both neural networks and actual brains have behaviors that emerge from the interactions of smaller components. Neural networks have trivial connections com
188.
by SkyBelow 2y ago
>Why not see if it can assemble a copy of itself -- if you find a genie in a lamp why not ask it for more wishes? Robots that can build things already exist and can be purchased second hand, but I haven't see this level of hacking t
189.
by SkyBelow 2y ago
Increasing the rent price makes it much easier for new rentals to come online, as long as the government isn't effectively stopping new competitors. Much the reason why grocery stores, car mechanics, etc. aren't going to raise th
190.
by SkyBelow 2y ago
AI native like recent digital natives, who have more time using software but far less time exploring how it works and less overall success at using digital tools? AI reminds me of calculators. For someone who is proficient in math, they bo
191.
by SkyBelow 2y ago
Outside of math and computational science, nothing is proven to not work because scientific research doesn't work in proofs. Even in math and computational science, there are fields dedicated to researching known proven wrong logic be
192.
by SkyBelow 2y ago
Best for an individual or for society? Consider a simplified example. There is some area of scientific research. Working within the framework gives you a 1 in 4 chance of making some minor improvement. Working outside the framework gives
193.
by SkyBelow 2y ago
AI does best in places where there is plenty of data for training and where messing up when used is only a mild problem (and where the problem is quick and low cost to identify). Art fits both of these categories. Healthcare data does not.
194.
by SkyBelow 2y ago
Such actions, even just the act of deleting text, conveys a message you were ordered to not convey and the government is not likely to take too kindly to that.
195.
by SkyBelow 2y ago
When I hear Paper Apps, I'm expecting something with a decent bit of complexity. Likely some number crunching, maybe needing an external source of RNG, using multiple pages. It says to me that this is something significantly more tha
196.
by SkyBelow 2y ago
52 pages is a bit of an investment to pick up and read a random item, and I suspect the reading speed will be much lower than a fictional novel given the different density of the material. Think of it like an abstract for a research paper,
197.
by SkyBelow 2y ago
I spent a day fighting with date-fns trying to get some date calculations working and was to the point I was questioning if my entire approach was flawed because there was no reason I should be spending that much time figuring our some simp
198.
by SkyBelow 2y ago
I don't ever use it for fire and forget, but I've been wondering how well that might work in small side projects where hidden bugs aren't a big concern. Like using a fire and forget to spin up a small javascript game. But n
199.
by SkyBelow 2y ago
I was thinking about the authenticity of my writing earlier this week and wondering why I have no problem accepting code from an AI and committing it, but I find the idea of passing off an AI's writing as my own feels not just wrong, b
200.
by SkyBelow 2y ago
Depends upon the user. Are you fast (and confident) at evaluating the output and discarding the bad suggestions? This is why I think using AI hurts some developers and helps others, and the usefulness is best for those who already have a
201.
by SkyBelow 2y ago
One of my earliest jobs was supposedly programming but was actually a slow descent into tech support for in house applications under the hood and I glad I took that as a hint to move elsewhere. Since then, it has always been a balance betw
202.
by SkyBelow 2y ago
Also, keep yourself employable. What you get hired to do and what you'll find yourself doing 6 months later, 2 years later, etc. aren't going to be the same. Whatever you are doing, keep in mind how much of it is really a market
203.
by SkyBelow 2y ago
That isn't guaranteed, because if so you can always do the opposite of what you think and you will overperform. Even if you think you know something, you are, at best, still being random. Anything perceived decrease in return from ta
204.
by SkyBelow 2y ago
Even compiling code isn't deterministic given different compilers and different items installed on a machine can influence the final resulting code, right? Ideally they shouldn't have any noticeable impact, but in edge cases it m
205.
by SkyBelow 2y ago
Maybe a deeper truth that is harder to put into words but which feeds into both of them. Something captured in much higher dimensional concept space that, when forced into our 3D world (and our <whatever>D political discussion space)
206.
by SkyBelow 2y ago
I find it an interesting commentary on child protection laws (and standards organizations that don't reach the level of legal enforcement) and related to an idea I've had for years, that people are interested in appearing like the
207.
by SkyBelow 2y ago
>but in most cases that's not going to be believed by a jury. Is there are least some argument of reasonability? I have an old Runescape account I would love to be able to get back into, but I don't even remember the email it
208.
by SkyBelow 2y ago
Pickup but silence might end up being better than letting one's voicemail grab it. Would make for an interesting study.
209.
by SkyBelow 2y ago
>The biggest downside, however, is the rapid accumulation of technical debt. Is that really related to the LLM? Even in pre-LLM times, anytime I've scrapped together some code to solve some small immediate problem it grows tech debt
210.
by SkyBelow 2y ago
How you use the LLM matters. Having an LLM do something for you that you don't know how to do is asking for trouble. An expert likely can off load a few things they aren't all that important, but any junior is going to dig themse

More ›