Catholic Info
Traditional Catholic Faith => Catholic Living in the Modern World => Topic started by: Ladislaus on March 12, 2023, 04:40:15 PM
-
Someone has inquired about learning computer programming in C#? I'd be willing to teach some classes, but there would have to be some serious interest. IMO, it would take about 2 years of part-time work (without being too aggressive in terms of schedule and time commitment) to absorb the concepts. There are these "boot camps" out there, but IMO they're a scam, as the pace is too fast for someone to truly absorb what needs to be absorbed. At the same time, 4 year colleges and universities are a total scam, where 3/4 of your coursework is in areas outside your "Major" field of studies, calculated to keep throngs of Leftist art history professors employed. AND the professors of Computer Science have either never or only decades ago actually written software for a living. Often they teach abstract concepts using antiquated and barely used computer languages.
I have had thoughts to start up a "school" that would offer Bacherlor Degrees ... although they'd be "unaccredited". If you put on your resume that you have a Bachelor's from this institution, nowhere are you claiming that it's from an "accredited university". I've never had a prospective employer check with my university about whether I had actually ever even gone there. But if they do, I'd be prepared to send information explaining why this degree is superior to that of a 4-year university. I've had to deal with new-hire "Computer Science" grads before, and they knew next to nothing, and left me wondering why they had paid $100,000 in college tuition.
I live in Ravenna, OH, and I would start up an institution, "Ravenna Institute of Technology". I would do this based on a tuition-by-honor system plan. No tuition at all unless and until you get a job in the field, and then you would pay back over time what you felt it was worth to you. Of course, I would only do that for Traditional Catholics.
This wouldn't be for everybody. Not everyone has the aptitude to do well in computer programming. If you did well in math and like math, that might be a hint that you'd have the aptitude for it and interest in it. This course would assume no prior knowledge of computer programming, however, though some kind of aptitude is essential. I started doing this type of course just for my son, but at some point he decided he didn't really like it, and I agreed that it probably didn't suit his temperament. He was also the type who would get frustrated if he didn't 100% understand something immediately. Even after I had gained a lot of experience, some new concepts took much repetition before there was an "aha" moment and I completely understood it. So perseverence is important. I think that it would take about 2 years to get familiar enough with programming to become effective enough to be employed as a programmer. One could go at a more aggressive pace, but I think that a slower pace that allows concepts to "sink in" is much more effective, and then of course you don't have to dedicate your entire life to it as well, if you have other duties in life.
I think that the course would consist of about an hour of lecture per week, which could be recorded and made available for streaming later and for subsequent review, and at the end of each lecture, there would be a significant amount of "homework". Students in the class would then be free to e-mail me if they run into issue or have problems. I've found that you don't REALLY learn something until you get in there and work with it hands on. If you don't have a 10:1 ratio between hands-on work and lecture, then IMO you're not going to really learn the subject matter. You just HAVE to struggle and make mistakes and run into bugs and get frustrated. Then in addition to the initial lecture, we could have group code reviews session of the completed assignments. So between the initial lecture (which could involve either a video or leaning some concept in some other way, including from videos by others), lots of hands' on work, and then review of and feedback on the work, I think you could become a very effective programmer and run circles around the 4-year-college Comp Sci degree graduates in less than 2 years.
Just send me a PM if you're interested.
I'd do this mostly to help out fellow Traditional Catholcs, to keep them from having to go to the cesspools that are modern universities in order to get a job that pays well and can support a family ... but also because I'd be rebelling at the utter and absolute scam that is the 4-year university. So many young people are being brainwashed into thinking that a 4-year-university degree is necessary. In some fields, you just can't escape it, as there are various certification boards involved, but in many fields it's a huge waste of money and time ... and subjects young people to potential corruption at these universities.
When I was in positions to hire people, I often opted for someone without the degree who seemed brighter and had more concrete actual skills than for the degreed Computer Science major. In fact, I got to a point where I would not even look at a Comp Sci major who didn't have a least a couple years of experience, since I came to believe that the degree is utterly worthless.
I think that there is a drawback and risk to not charging any tuition. I could put a lot of effort into this only for individuals to drop out due to lack of perseverance, getting distracted, etc. So I'm not quite sure how I would approach it. I put a lot of time in with my son, and it went for naught.
-
Oh, one thing that I've also learned is that you should be reasonably skilled at typing to have any hope of success at programming. I knew this one guy who seemed like he knew all the concepts and could talk about programming, and then he sat down to his keyboard and began to hunt and peck for one character at a time. He could have been the most brilliant programmer in the world conceptually, but this defect would make him ineffective and very unproductive in the real world.
You don't have to be able to type 120 words per minute, but you also can't be saddled with hunt-and-peck typing skills.
-
Practice makes perfect. I used to hunt and peck, now I no longer waste time hunting. Learning to play a musical keyboard helped me naturally start using more fingers while typing.
-
I speak from ignorance, but. Do you think that a large part of programmer jobs may disappear in the next few years with the advancement of AI? GPT can now make entire software just by asking it. And it's only a few months old.
https://twitter.com/ammaar/status/1635754631228952576
-
Honestly I don't really believe in AI. I've never seen any AI that did anything I could verify that was all that impressive. Are you able to recreate what the person claims to have done in this video?
-
I speak from ignorance, but. Do you think that a large part of programmer jobs may disappear in the next few years with the advancement of AI? GPT can now make entire software just by asking it. And it's only a few months old.
https://twitter.com/ammaar/status/1635754631228952576
It's interesting that this is happening, because a decade ago I had thought of many of the things AI is being used for. I used to think that anything the human mind can think, a computer can be made to do. It's just logic, though complex and difficult to make a computer do. But, once a computer has been made to do the things the human brain can think, then the development of such technology can rapidly increase.
I used to think how Windows shouldn't tell me there's such and such problem, it should fix it. Or, Windows shouldn't say it doesn't know how to open a file, it should try displaying it in different ways, figure it out, do the human thing if only it was programmed to do such. I had it do that recently, so I just rewrote the extension/file type, and tried opening the file, and it worked. Once the initial problems are worked out with AI, I should be able to tell my computer what to do, and have it do that, even if it means writing new software with the same logic based imagination a human would. Think Star Trek: "Computer, make an audio & video editing program that can convert any common file type into any other, includes a 35 band equalizer ranging from 4hz to 25kHZ, and a tool that increases the resolution of both audio and video." "And give it a very dark gray scale color scheme."
-
I speak from ignorance, but. Do you think that a large part of programmer jobs may disappear in the next few years with the advancement of AI? GPT can now make entire software just by asking it. And it's only a few months old.
Just from my experience, probably not. They might attempt to use this to fill in some of the holes, but there is a severe shortage of skilled programmers already. There is a surplus of "qualified" individuals, especially from non-Western countries. That said, the spam potential of GPT does provide an excuse for TPTB to implement digital ID in the future.
-
I speak from ignorance, but. Do you think that a large part of programmer jobs may disappear in the next few years with the advancement of AI? GPT can now make entire software just by asking it. And it's only a few months old.
https://twitter.com/ammaar/status/1635754631228952576
Anyone can write spaghetti code. Making good quality, secure, and efficient software requires a lot of foresight and skill
-
I speak from ignorance, but. Do you think that a large part of programmer jobs may disappear in the next few years with the advancement of AI? GPT can now make entire software just by asking it. And it's only a few months old.
https://twitter.com/ammaar/status/1635754631228952576
Perhaps eventually ... no sooner than a decade, AI might cause a drastic drop in the number of programmers needed. But it's still pretty far off. GPT-produced software is still not particularly good; it can do relatively simple things, but nothing particularly complicated. But even for GPT, it's the old saying of "garbage in, garbage out". If the requirements are not specified adequately (much of the challenge with programming), there will be all kinds of issues. If it gets to that point, a lot of programmers would convert to being solution architects or business analysts to actually feed the requirements properly into AI systems.
I know that the banks, IRS, fed, NASA, and many others still rely on sometimes 40- or 50- year old systems that need to be integrated with, and there are all kinds of architectural considerations required to determine the best integration patterns based on the required system "quality attributes".
-
Anyone can write spaghetti code. Making good quality, secure, and efficient software requires a lot of foresight and skill
Right. These are the "quality attributes" to which I refer. There have long been various product out there to allow non-programmers to, say, build websites, or other basic software tasks. These are OK for simple things. But I don't see "AI" becoming very good at architecting, designing, and implementing ideal solutions taking these types of concerns into consideration.
-
Just from my experience, probably not. They might attempt to use this to fill in some of the holes, but there is a severe shortage of skilled programmers already. There is a surplus of "qualified" individuals, especially from non-Western countries. That said, the spam potential of GPT does provide an excuse for TPTB to implement digital ID in the future.
Indeed, if GPT could replace some of these "offshore" teams from India that are a horrific nightmare for me (as we speak, at my current company), I would be grateful, but I don't think it'll replace skilled programmer and architects anytime soon. That's probably at least a decade off.
-
It's interesting that this is happening, because a decade ago I had thought of many of the things AI is being used for. I used to think that anything the human mind can think, a computer can be made to do. It's just logic, though complex and difficult to make a computer do. But, once a computer has been made to do the things the human brain can think, then the development of such technology can rapidly increase.
I used to think how Windows shouldn't tell me there's such and such problem, it should fix it. Or, Windows shouldn't say it doesn't know how to open a file, it should try displaying it in different ways, figure it out, do the human thing if only it was programmed to do such. I had it do that recently, so I just rewrote the extension/file type, and tried opening the file, and it worked. Once the initial problems are worked out with AI, I should be able to tell my computer what to do, and have it do that, even if it means writing new software with the same logic based imagination a human would. Think Star Trek: "Computer, make an audio & video editing program that can convert any common file type into any other, includes a 35 band equalizer ranging from 4hz to 25kHZ, and a tool that increases the resolution of both audio and video." "And give it a very dark gray scale color scheme."
Right, there's no reason this can't happen for basic tasks, but the types of software solutions I've had to implement over the years, some of which have been very challenging even for skilled architects and programmers, they won't be doable by AI for at least a decade or more.
-
Honestly I don't really believe in AI. I've never seen any AI that did anything I could verify that was all that impressive. Are you able to recreate what the person claims to have done in this video?
Agreed. There's no such thing as true "AI". They're promoting the notion of AI for philosophical reasons, in their attempt to claim that the human mind and human consciousness can be reduced to material causes. It feeds into the entire God-less evolution agenda. I could see demons taking possession of such computers to give the appearance of being sentient, but actual "AI" is not possible.
-
It would be as simple for a angel (fallen or unfallen) to inject/manipulate signals or values into a computer's memory, passing through an ethernet cable, etc. as it would be for us to impress a 2 year old with a hand-puppet. "Oooh, it's alive! Wheeee!" An angel could look at an ethernet cable, see the small fluctuating magnetic fields around the cable, and know instantly what is being sent over it. Even if the data were 2048-bit encrypted! Even if the data were being sent at 1 GB/second.
To an angel, the most advanced math in existence is as boring and simple as "2+2=4" to the average human. But more than just understanding, their "brains" work at lightning speed compared to our meat brains.
Think of how easily we can deceive, manipulate, and impress a 2 year old. Only we're much smarter than a 2 year old, and, well, the angels are that much smarter than us.
Angels don't reason per se. Their intellects don't use step-by-step reasoning. They don't ratiocinate. They instantly intuit all the conclusions. Their intellect doesn't require meat -- they have no brain to get tired.
An angel could instantly grasp a whole complex software program -- even without the source code! Something no human could do. They have an intuitive-level understanding of the whole natural world -- the way we completely understand and wrap our minds around, say, a child's hand-puppet.
-
They've been predicting programmer-free software design since the 1980's. Remember Visual Basic?
It's a fundamental truth: You can't have design without a designer.
As long as they want things designed, they're going to need human engineers to design it. Computers will never be able to take over, because they're dumb as rocks. Fast, but stupid. Without a human intellect to program them, they can't do anything useful.
But of course some modern idiots think that something as complex and wonderful as LIVING ANIMALS came about by random chance -- no designer required. 1 million monkeys typing at 1 million typewriters for 1 million years could eventually write the works of Shakespeare by accident. Yeah right!
Humans aren't good at mind-boggling numbers. It DOES make sense, yes. But what they don't grasp is this: before those monkeys came up with Shakespeare, it would take 10 to the 3000th power years, and the universe would die a heat death before then. All the stars in the universe would burn out first. So effectively it IS impossible.
Some creationist ran the numbers. They first pointed out that according to mathematicians and scientists, anything more than 10 to the 40th or so is considered "impossible". Then they gave us the "bad news" on how long it would take to create one strand of DNA... It was glorious. I don't have the video handy, but I think I own it somewhere.
-
I found it!
https://www.youtube.com/watch?v=W1_KEVaCyaA
-
It would be as simple for a angel (fallen or unfallen) to inject/manipulate signals or values into a computer's memory, passing through an ethernet cable, etc. as it would be for us to impress a 2 year old with a hand-puppet.
Yes, that's more along the lines of what I mean by a devil taking "possession" of an AI system, basically controlling it in order to manipulate it (so using the term a bit loosely).
-
They've been predicting programmer-free software design since the 1980's. Remember Visual Basic?
It's a fundamental truth: You can't have design without a designer.
Yeah, the idea behind Visual Basic was that business folks would be able to write software. Didn't work out that way at all, since you still needed pretty serious programming skills ... certainly not at the level where you had to write C++ on Windows to build and position every little button and label by hand, but there was still a lot of business logic and algorithm required.
In MS Access, they had similar "forms", so that there truly anyone could write a data reading and data entry type of app without programming. But it you wanted to do something special with the data beside read, edit, and save ... there was programming there too.
I've seen Visual Basic programs that consisted of nearly 100,000 lines of code. I was one of those VB programmers back in the day, from about 1999 - 2001, and I wrote all kinds of stuff, including image processing applications that got pretty sophisticated. Of course, many programmers felt that VB programmers weren't "real" programmers, but I transitioned to both C# and Java without skipping a beat.
-
I found it!
https://www.youtube.com/watch?v=W1_KEVaCyaA
Yeah, this is riduclous. They probably wouldn't be able to even quantify the building of a cell, since the probabiliy is ZERO. So even this one protein is almost impossible, but the smallest cell requires dozens of proteins and other things, and they would have to all happen in the same tiny space all at the same time, and then organize themselves.
Protein 1: "Hey, Protein 2. Come over here. Let's hold hands and form part of a cell memebrane."
Protein 2: "OK."
Protein 3: "You two come over here and we'll build a bigger membrane."
Protein 2: "No, you come over here."
Protein 1: "Now we need to figure out how to reproduce. Where are proteins 4-20?"
Protein 4: "Now we need proteins 21-30 to work out an excretory function."
... as if proteins had any kind of intelligence or awareness to begin with.
This is so ridiculous that only a bad-willed atheist or utter moron could believe it, a moron who's orders of magnitude more stupid than these proteins would have to be to organize together to build a functional living organism.
I've always considered this. When an embryo is fertilized and a new human life is created, the cells start dividing. They are all identical cells. At what point do they start differentiating and then organizing. Who's doing the organizing and who's taking the lead on it? "Hey, you, go over there and start forming bone." "Now, you, become heart cells." "You become a blood cells." "We need more of you now." No given cell can organize and orchestrate the entire exercise. Of all the identical cells that form when the fertilized egg begins dividing, who takes charge? And how do they communicate complex instructions with each other?
-
I don't know why more Americans don't just go abroad for college. Very expensive, yes, but probably still a hell of a lot cheaper than going to college stateside, especially if your college is far from home and you have to rent a dorm, etc.
-
Of all the identical cells that form when the fertilized egg begins dividing, who takes charge? And how do they communicate complex instructions with each other?
I am not convinced it is miraculous, but some complex system of communication between cells, so they know where they are based on the DNA code. It would be very interesting to know how it works.
-
They've been predicting programmer-free software design since the 1980's. Remember Visual Basic?
It's a fundamental truth: You can't have design without a designer.
As long as they want things designed, they're going to need human engineers to design it. Computers will never be able to take over, because they're dumb as rocks. Fast, but stupid. Without a human intellect to program them, they can't do anything useful.
But of course some modern idiots think that something as complex and wonderful as LIVING ANIMALS came about by random chance -- no designer required. 1 million monkeys typing at 1 million typewriters for 1 million years could eventually write the works of Shakespeare by accident. Yeah right!
Humans aren't good at mind-boggling numbers. It DOES make sense, yes. But what they don't grasp is this: before those monkeys came up with Shakespeare, it would take 10 to the 3000th power years, and the universe would die a heat death before then. All the stars in the universe would burn out first. So effectively it IS impossible.
Some creationist ran the numbers. They first pointed out that according to mathematicians and scientists, anything more than 10 to the 40th or so is considered "impossible". Then they gave us the "bad news" on how long it would take to create one strand of DNA... It was glorious. I don't have the video handy, but I think I own it somewhere.
It's the fast detail that is the computer's advantage. With this extreme speed it can consider various designs based on information it already knows, and with any adaptability programs it can design new things, do all the math to test and compare designs to see what fits the intended parameters, and even learn from those results to invent and experiment with new designs. It's just what people would do, but much faster. So, although there may be some lack in true natural artistic imagination, a very complex computer should still be able to design things that we may think are amazing or genius solutions, if programmed with the right inventing and problem solving processes.
There's a program that has been in use for many years now that experiments and calulates until it finds the strongest and lightest solution for a part for whatever piece of engineering needs it.
I really think computers can be made to think and do just about everything a human can short of receive divine inspiration, or pray, or something like that.
-
It's the fast detail that is the computer's advantage. With this extreme speed it can consider various designs based on information it already knows, and with any adaptability programs it can design new things, do all the math to test and compare designs to see what fits the intended parameters, and even learn from those results to invent and experiment with new designs. It's just what people would do, but much faster. So, although there may be some lack in true natural artistic imagination, a very complex computer should still be able to design things that we may think are amazing or genius solutions, if programmed with the right inventing and problem solving processes.
There's a program that has been in use for many years now that experiments and calulates until it finds the strongest and lightest solution for a part for whatever piece of engineering needs it.
I really think computers can be made to think and do just about everything a human can short of receive divine inspiration, or pray, or something like that.
I've been using ChatGPT to help me with a coding project and, while it is astounding (even just a couple years ago I couldn't imagine this), it won't be a replacement for even a single developer any time soon. It's code is only useable if there's someone giving it detailed prompts (and I don't mean in a business-requirements sense, but the prompt-giver must also understand that software architecture and implementation required) and who can read over the code to check for and point out errors, and then take these code snippets and adapt them to fit in wherever they're needed in the codebase. At the moment it's only use-case is to give you some inspiration, like how a StackOverflow question might show you how something may be designed, and information (because of course you can ask it "what functions are available to do X in Java?").
It's possible that in a few years you'll be able to have some AI look at your entire codebase and give it some prompts to get it to produce code that actually has that in mind. Then we might have something that could most the work of a graduate developer. But, you need graduate devs to get senior devs, so it's not like companies will stop hiring grads.
Where it'd become a real concern is when an AI service could do the work of dev with 5+ years of experience, but that's still a whole paradigm shift away IMO, and even when we do get to that point, it'll just mean devs switching over to being prompt-engineers. You would gradually need a lot less devs, but the software sector continues to grow, so I don't know if we'll ever see mass lay-offs or redundancies (at least in the foreseeable future). It could just happen that the decline in the number of devs needed to do produce X lines of code gets counteracted by an increase in the amount of lines needing to be written.
-
I can foresee the possibility of AI becoming so complex that there will be job openings for people who can continuously update it to correct and prevent errors or inefficiencies, which could be called AI mental illnesses.
-
I am not convinced it is miraculous, but some complex system of communication between cells, so they know where they are based on the DNA code. It would be very interesting to know how it works.
Not, it's not miraculous, but it's also not due a material cause. You conflate miracle and supernatural with material cause. This organization and marshalling of the cells comes from the animal soul in any living organism.
-
There's a program that has been in use for many years now that experiments and calulates until it finds the strongest and lightest solution for a part for whatever piece of engineering needs it.
I really think computers can be made to think and do just about everything a human can short of receive divine inspiration, or pray, or something like that.
.
This is a physical process, though. Computers are great at doing anything physical or mathematical. What they can't do is think in an abstract manner.
What you're describing sounds like going through either a series of trial-and-error designs and testing all of them using a software simulation, or maybe implementing principles of efficiency and design that were programmed into it by a human being. There's no real difficulty with saying a machine can do those things. But it would take a human being to invent a completely new technology, for example, or implement an already-existing technology in a completely new way.
In general I think AI is way overhyped. The examples of how it's used that I've seen do not impress me at all. They've been talking about self-driving cars using AI to detect objects for many years, and it was just a few years ago one of these wretched things broadsided a fire truck, killing the person inside (the driver), because its AI didn't detect a fire truck directly in its path.
-
Interestingly, just the other day, I had a VP at our company ask me to look into developing software with AI. Of course, what he has in mind is to replace a bunch of us programmers with automatic software development. It's a total pipe dream. What is he going to do, pick up a microphone and say, "build some software to analyze our engineering data"? :laugh1: That's about as close as these guys would get to being able to produce requirements. Our systems are extremely complex with some very specialized business rules.
-
AI is the latest hype. Remember when people thought search engines were omniscient? Well, let me tell you Google has PLENTY of blindspots, empty spots, or holes. Google does NOT know everything. Google lets you down plenty of times. Especially those of us interested in OLD things. And lately, if you want anything other than the Official Story or the "mainstream consensus" opinion on an issue, forget it!
-
At the same time, 4 year colleges and universities are a total scam, where 3/4 of your coursework is in areas outside your "Major" field of studies, calculated to keep throngs of Leftist art history professors employed. AND the professors of Computer Science have either never or only decades ago actually written software for a living. Often they teach abstract concepts using antiquated and barely used computer languages.
I have had thoughts to start up a "school" that would offer Bacherlor Degrees ... although they'd be "unaccredited". If you put on your resume that you have a Bachelor's from this institution, nowhere are you claiming that it's from an "accredited university". I've never had a prospective employer check with my university about whether I had actually ever even gone there. But if they do, I'd be prepared to send information explaining why this degree is superior to that of a 4-year university. I've had to deal with new-hire "Computer Science" grads before, and they knew next to nothing, and left me wondering why they had paid $100,000 in college tuition.
I'm thinking the degree is only useful to get past the HR department, the goal being to actually get a technical interview.
-
I'm thinking the degree is only useful to get past the HR department, the goal being to actually get a technical interview.
100% agreed ... in IT at any rate. There are some fields where you definitely need the degree ... because it's equivalent to being licensed.
-
Interestingly, just the other day, I had a VP at our company ask me to look into developing software with AI. Of course, what he has in mind is to replace a bunch of us programmers with automatic software development. It's a total pipe dream. What is he going to do, pick up a microphone and say, "build some software to analyze our engineering data"? :laugh1: That's about as close as these guys would get to being able to produce requirements. Our systems are extremely complex with some very specialized business rules.
This reminds me of one of my friends recently asking me would I be willing to do some contract work for him that involved basically doing exactly that for a 3rd party (so they'd be contracting him who'd be contracting me). When I tried pressing him for even the vaguest of requirements, I got nothing. I tried explaining how the complexity of the task could vary wildly and that I needed to know what I'd be dealing with to judge if I could actually do it for him, or how long it could take, but still the answer was to basically figure it out as I went along--which he didn't seem to realise would cause massive issues with the guys hiring him once I inevitably entered into development hell on a wildly out-of-scope project.
When I turned down the offer, he seemed pretty gobsmacked that I'd turn down easy money. In reality what I was turning down was months of stress and a inevitable end to our friendship.