|
Post by Youngster Joey on May 30, 2019 18:16:01 GMT -5
Yeah, idk, I hit panic mode yesterday. Doing better today. Finally pushed the first task I was working on out the door, and picked up another task which seems pretty straightforward.
I'd be totally lost on a compiler. I guess that's where the university background comes in handy. I'm assuming you'd have a better foundation with a university degree than someone who taught themselves how to do abstracted for loops by themselves. Meanwhile, I learned about two's complement once.
I think that's a weird question, frankly. I can't think of any answers I could give that wouldn't make me sound like a smarmy not a very nice person >(, unless I chose something deliberately self-deprecating. Plus, while I'm definitely guilty of thinking to myself I'm better than X person at something"*, how can you be certain you're definitely better than everyone else in that room? Maybe you suck, relatively or absolutely speaking. I remember reading one study where about 70% of people reported they believed their sense of humor was "above average." People aren't good at judging themselves against the general public.
*tbh one of my flaws is that I actually really get impatient/annoyed by people who don't do things up to my standards. I'm cool with someone doing shoddy work because they're new to it, or if they're genuinely trying hard; that's totally fine. But if I think you should know better and you're just settling, then it really irritates me, and I will silently judge.
|
|
|
Post by Friendly Person :) on May 31, 2019 4:47:53 GMT -5
I want to respond to you Joey, and I will tomorrow. For now I just finished Lain, and I want to jot down some quick thoughts before bed.
It's a great 'cerebral' show, which is an extreme rarity in anime. The problem is that it tends to falter a bit in the characters, which Derman pointed out (and for his sake, I won't be posting any spoilers). As a result, it's difficult to call it a great show overall. The only character to really undergo any kind of arc is Lain. The thing is, for all the psychological components explored, she doesn't quite come close enough to breaching the barrier between 'real person' and 'representational of a type of individual'.
I've never thought about it much, but I think what separates the barrier is defined by the number of unique emotions one feels towards a character. I'm not referring to a 'layered' or 'multi-dimensional' character. Those I would quantify by the number of emotions one feels simultaniously (quick example: think of a family member. Chances are you feel several (likely conflicting) emotions towards them). By 'unique' emotions I mean, do I laugh/cry/cheer for/with this character throughout the story? Say what I might about Tony Stark, he was a guy that you liked. When things went badly for him, you felt bad. When things went well, you were happy. When he was in danger, you felt tense. He might not have been the most layered/dimensional character, but he did feel -distinct-.
In contrast, I didn't really feel a heck of a lot toward Lain as a character. Which isn't to say there isn't a character there to like, we just aren't *shown* anything beyond those elements strictly related to the plot. There isn't a single moment of humour or levity in the entire show. So much time is spent with awkward stares that you just kinda wonder if anything is going on back there. Yes, there are reasons for many of these awkward moments -- reasons that tie into the plot which is very cerebral and clever and what-not -- but the problem is that reality is not comprised entirely of awkward moments. By opting not to show us those counter-balancing slices of reality, you limit the character to a few select personality traits: traits we recognise and relate to, but which are ultimately too few in number to create an individual (which is effing ironic considering how much of a role individualism plays).
Anyway I don't want to be super down on it. I think it's good. If anything, it's due to the rarity of anime attempting to tackle such ambitious issues with this kind of weight that results in the need to be critical. Lain is taking itself seriously. Kudos. As such, I will do it the courtesy of also taking it seriously.
|
|
|
Post by Friendly Person :) on Jun 1, 2019 0:50:59 GMT -5
Okay, now I can respond.
From what I've gathered from my friend who *does* have a university background, university is nothing compared to actual skills. Granted, despite being supposedly well-rated, his university sounded like a festering sh*t-hole that actually taught very little. Overall I think there's more to be said for someone who can learn and adapt. It might take you more time to get up and running, but I think in the long run that level of self-motivation is likely going to keep you more relevant in a field where things change on a fairly regular basis. My mother also lacks a college background despite being in a coding position. It's been a lot of work, but not only has she managed to keep her job, but she was in on the ground work for a lot of their present systems, which gives a level of job security beyond a simple college degree. It's going to be challenging, but isn't that basically what you wanted?
Also I think it's fairly standard for people to expect others to put in at least the same amount of effort as they are, regardless of the situation. Even in a video game, if everyone else wants to win and there's the one guy who just wants to have fun d*cking around, resentment breeds pretty quickly. Generally it's better to just ask if it's worth getting upset over, because excess frustration can only end in two ways: egocentric assh*le, or stressed-out nut-job.
|
|
|
Post by Youngster Joey on Jun 1, 2019 13:57:46 GMT -5
Oh, don't get me wrong--it's not universal. Something like compilers and memory management is indeed probably better suited to a grad than someone self-taught, as it's closer to both the metal and theory. Degrees are quite heavy on theory. I have zero theoretical background.
However, to your point, there's a reason why, despite the huge need for developers, it's actually quite difficult to get a job as a recent grad--you likely don't know anything practical. The industry moves too fast for university courses to possibly keep up, and some practical issues you have to deal with in the real world sort of can't come up in academia. e.g., a lot of software complexity comes from the fact that you have hundreds or thousands of people using your application. How do you teach how to deal with that without, you know, an application that attracts that sort of traffic? Sure, you could teach about the technologies people use, I guess, but still (also, see point about industry moving fast). Moreover, just because something is "technically" the right solution doesn't mean it will actually work in the real world. Oftentimes you'll find companies will have their own custom offshoots of popular technologies to meet extremely specific needs that you just couldn't have dreamt up in academia.
Obviously it's quite reasonable to get annoyed if other people's work impacts yours. I get annoyed if it doesn't impact me. For instance, we have this one guy at work who was complaining about the horsepower of our technology tools and that his code was taking hours to run. He insisted he needed more computing power, so my manager asked me to look at his code to see if I could help. So I did. Among other issues...
1) He repeated code everywhere. This is like the biggest sin in programming. If you need to reuse code, you design it so that it is defined once and called with a one-line-long function; you don't copy and paste 30 lines of code 10 times all over the place. This is a programming 101 concept. 2) The framework we use has a lot of pre-built features that are highly optimized. He didn't use them in lieu of his own hand-rolled versions. This is like refusing to buy tires for your car and making your own tires, except you made them from scratch in 10 minutes with some sh*tty material and they go flat constantly. 3) He was using programming techniques that are a genuinely terrible fit for the framework we are using. Spark hates for-loops and using them actually makes Spark, a very performant framework, extremely non-performant. He had for-loops everywhere. 4) Said for-loops were unnecessarily long, nested, and convoluted. For Derman, consider something like O(N^4) when O(N) would do.
Obviously, none of these things impacted me since I have never worked on his project. But it still offended me because this dude has been programming for 10+ years and it genuinely looks like beginner's code, when you just don't know any better and are still learning the ropes. Worse, when my other coworker and I pointed out where things could be done better and more efficiently, he got offended and told us, "Well, it works." Yeah, but you're complaining that your code is slow and that you need more compute power? That's not what I call "working," and even if it did work okay, don't you want to write code that works better? I know I do. Why in the world would you aim for "good enough"? I'm not talking about squeezing the last 0.1% out of a thing that's already 99.9% perfect. I'm not pedantic and anal-retentive like that. I'm just talking about being a generally better programmer. In the end, he incorporated none of the changes my coworker and I suggested, and ended up getting more horsepower for his sh*tty code. Zero interest in actually diagnosing and fixing the problem at hand.
No skin off my back that his code still took hours to run, but I just really hate that attitude. I personally want people to tell me how I can write code better, and it baffles me that some people have no interest and/or actively get offended.
|
|
|
Post by Youngster Joey on Jun 1, 2019 15:30:16 GMT -5
I'm never going to stop loving this video.
I'm simultaneously concerned that someone made this--how do you even think up something like this?--and also so glad they did. If it were me and I made this video, I'd have felt like I'd provided my contribution to humanity.
|
|
|
Post by Youngster Joey on Jun 1, 2019 20:30:42 GMT -5
Write New York Times bestseller, huh. That either says something about the quality of garbage people will read, or this is just wildly off-base. I much sooner see AI automating a mathematician's job than writing quality books. I'm skeptical of these graphs anyway. Some things will happen much sooner than we think; others will take much longer. Seventy years ago, people probably imagined us with flying cars. Few, I imagine, could have predicted the internet. (Also, I kept wanting to write fifty years ago, but that's because for some reason in my head, "fifty years ago" is the 1950s. I guess I'm just stuck in the 2000s. The Internet was a reality fifty years ago, albeit a not-very-known one. Also, fact-checking on the internet tells me that the US military began research into what turned into ARPANET in the 60s. But anyway. Shut up. My point is made.) I like to couch any predictions I make about technology by noting that I thought the iPhone was stupid when it came out, and didn't see the point in owning a smartphone. In my feeble defense, I reasoned that the only use case I could see for the iPhone would be listening to music, but we already had iPods for that and an iPhone's battery would just drain faster. (I still feel this way about smarthingyches. I'm waiting for smart watches a la Xenosaga I where they project holographic screens that have tactile input.) edit: My last stream of consciousness--is anyone else getting constant bad gateway errors from nginx on this site? Or is that just me? edit2: seriously? it's going to censor me on smart watches without the space because of tw@t?
|
|
|
Post by Friendly Person :) on Jun 2, 2019 4:14:30 GMT -5
Fifty Shades of Grey. Twilight. John Green. F*ck, a computer writing a NYT bestseller sounds pretty plausible to me. Just saying.
|
|
|
Post by Youngster Joey on Jun 2, 2019 10:59:51 GMT -5
Where I could see a bestseller coming in is when the first somewhat coherent book comes out. There might be a rush in sales because, well, it was written by AI! Subsequent books wouldn't see such a bump once the novelty wore off, assuming the quality didn't otherwise warrant it.
Although, I've seen people train models on Hallmark movie scripts. They're somewhat nonsensical, but also somewhat accurate, in large part because Hallmark movies are so formulaic. I suppose formulaic genres like romance or pop crime could see themselves replicated sooner than others. Good riddance to them. I can't stand boilerplate books.
|
|
Derman
Oracle Knight
I still don't have a knife tag on my golden birth knife
Posts: 194
|
Post by Derman on Jun 2, 2019 11:27:05 GMT -5
I've seen people feed harry potter to an AI that generates stories based on that. The result was entertaining, although occasionally it didn't make any sense. I could see a book based on only that concept being a NYT bestseller. EDIT: hah, Joey posted about the same thing as I was writing my post.
I can see what you mean though, and I don't see AI replacing human writers any time soon. Definitely not in the next 20-30 years. I'm slightly annoyed by the fact that "Retail salesperson" and Go are both at around the same level, despite both being very different problems. Go could be solved with just massive amounts of processing power, or a good algorithm. Either way, it's mostly a mathematical/abstract problem, with a clear "solved" state. Retail salesperson is a task that needs the ability to adapt and serve people. It isn't a problem that's solved by raw processing power or genius algorithm, and it's not easy to say when the problem is 'solved'. Honestly, most of those examples seem a bit out of place.
Also, no gateway errors for me.
That guy sounds like a perfect example of a person who shouldn't be writing code in the first place. "It just works" is not good enough, we don't have infinite processing power in every case. A lot of tasks are not that hard if the goal is to just get it to work, but the problem is to solve it without spending time in endless nested for-loops. My first two tasks here have both been about optimizing a current implementation. First task was to refactor some old, messy code, and this windows binary thing does have a working implementation, but it's WAY too heavy to be used in the actual product, and it's the definition of "it just works". So I have to figure out how it can be done without searching through the entire binary for a specific series of bytes.
I agree that compiler/binary stuff is probably better suited for someone who has some theoretical background. As I was reading through the documentation trying to understand the structure of the binary file, I ran into a lot of stuff that was mentioned in some of the courses. Still a lot of stuff I understood nothing about, but after some digging I could connect it back to the basic stuff we were taught at the university (stuff I thought I'd never need to use, but thought were interesting regardless).
|
|
|
Post by Friendly Person :) on Jun 2, 2019 14:07:38 GMT -5
I have no doubt that a computer will someday be able to generate an entertaining story. In fact, all of your Marvel and Star Wars movies could all probably be replaced with computers at some point. Where I think the breakdown will occur is when you start getting into the more introspective, observational, or just to put it in a one-lump pretentious word: 'literary' side of things. Engaging with art is like engaging in a conversation. People read more 'literary' books to try and reveal or understand or explore some kind of nature of humanity/society. I just can't see engaging with a computer in that type of method. It might be interesting to see what a computer might "observe" about humanity, but ultimately when it comes to discussions regarding humanity, I'd rather engage with someone who possesses first-hand experience.
Of course, that's assuming things advance to a point where AI is significantly ahead of where it is now. You've all pointed out those auto-written books and how janky they are. While you aren't wrong, consider also the earlier conversation regarding recommendations. If a system cannot look at what I've enjoyed and generate a solid, unique recommendation based off of that (as I've said, I have never found a generated recommendation to be of any value), how can it look at humanity as a whole and find something insightful to say, much less a compelling way of saying it?
Maybe I'm just in denial, but I never see computers completely taking over art. Hell, if deckle edges suggest anything, it's that there will always be a market for stupid, flawed sh*t.
|
|
|
Post by Youngster Joey on Jun 2, 2019 20:55:25 GMT -5
Well, I got solicited on that same block near my apartment again. Also apparently because I looked sad again.
This time it was a couple, and they offered me hugs to make me feel better. And when I said I don't like hugs, they asked me if I wanted to make out instead. Needless to say I did not take them up on their generous offer.
|
|
|
Post by liaraeyne on Jun 3, 2019 0:28:44 GMT -5
*chimes in*
Do people ever talk about the Tales Of series in here?
Sorry for the interruption.
|
|
|
Post by Friendly Person :) on Jun 3, 2019 1:00:38 GMT -5
From time to time. There's not been anything to talk about since Berseria. We've all played several of the games, so if there was something specific you wanted to ask, by all means ask. I'm gonna be honest though, if you're looking for constant discussion on Tales, we might not be the best people to talk to. Not only have we been the only three people consistently on this site for years, none of are what you might call "active" in the Tales community.
Also it's a random thought thread: you aren't interrupting anything. If you have a thought, or question, or anything really, by all means post.
To Joey: You're so lucky, getting hit on twice. People tell me I look angry all the time, but nobody ever offers to hug me, much less make out! I tell ya, women get a free f*cken ride...
|
|
|
Post by Youngster Joey on Jun 3, 2019 11:24:57 GMT -5
I've been meaning to play Tales of the Abyss at some point again, at least.
Although I'm not keen on the voiceless skits... I enjoy the skits nonetheless, but I don't like reading the dialogue 5 times over before it allows me to advance.
I could play a ripped Japanese voices-English text DVD on my PS2, but this would require me dealing with the load times... eugh. I usually recommend people the 3DS version on this count alone.
I strongly dislike physical contact with anyone other than cats (people who freely hug acquaintances are just... confusing), so I'm not sure this is a blessing, quite. If I ever change my mind, though, I know what street to hang out...
|
|
Derman
Oracle Knight
I still don't have a knife tag on my golden birth knife
Posts: 194
|
Post by Derman on Jun 3, 2019 15:05:27 GMT -5
What's wrong with people over there? Other than people trying to sell me stuff, nobody has talked a single word to me when I walk around the city. I guess we are way too good at minding our own business. The only thing people have mentioned about my face is that I seem to be smiling a lot. Not sure what kind of smile they are talking about, or if it's even a positive thing. But I guess it doesn't matter, people won't be stopping me on the streets because I look too happy and need to be pulled back to ground.
|
|