Відео

I was writing COBOL between 1982 and 2000, when they decided they didn't need so many of us any more. Honestly I don't know if I would have preferred the last 24 years as a fully employed static corporate coder or the poor but happy free soul I am now. Maybe I would have died of boredom and bad air conditioning.🤔

My 2c take is that it’s actually dangerous and will lead to a trainwreck of massive proportions It’s too easy for management to use Rust “safety” as a crutch - hire cheap and inexperienced web devs to cobble together a thousand dependencies, use LLMs to generate code, and trust that “if it compiles it works” Everything that Rust “achieves” in terms of safety it does by removing control from the programmer. What could possibly go wrong ?

This guy clearly doesn't know anything about how the web should work... And as anyone like him, he farts his own opinion thinking it's infused gold. "I don't like CSS zen garden because that's just not how i want to write web pages" lmao is that your argument? Really? Not exactly rock solid. He's missing the main point of this which is that there should be a separation of concerns between the presentation on the content. But he completely ignores that, and shits his own opinion thinking he's edgy or something... I'm puking!

Full interview? :)

Ummm Flutter??!

So first time watching you guys talk, have to say, you guys sound younger than you look, but take that as a compliment because I am either the same age or slightly older :D

Brandow's explanation is rather "Meh".

I agree - more tooling doesn't get you good software - but all the security vulnerabilities out in the wild do deserve some consideration - if it can make not-great programmers unable to create huge exploits that affect the entire internet - maybe that's a big thing. At least it's one layer or piece of the problem.

I agree, one big MEH.

I retired 2 years ago, but would consider 250K !!!

ua-cam.com/video/uKG0IEyeZb0/v-deo.html

Completey agree music discovery is still broken, maybe more so with Spotify. But today is your lucky day. Search for Cold Chisel and the singers Jimmy Barnes and Ian Moss. Start with their self titled album from 1978.

After implementing a small Signals library in TS I describe Signals as just the Observer Pattern with automatic dependency acquirement. How does this automatic behavior happens? Simple. To use the value a signal holds, you are actually calling a function that returns the value, (in solid.js this is explicit, other frameworks might hide this fact behind a proxy) this function call does not simply return the value it also first checks to see if the signal is being used inside an "effect" (createEffect) or "computed value" (createMemo) and if that's the case then it adds this effect/computed value to its observers list. Whenever you setState, or setSignal, thereby mutating the signal's value, it has that list of observers that it acquired in the previous step and it notifies all its observers about the mutation. The observers in turn will do as they please with this notification, eg. effects will run an effect, computed values will simply recalculate their own value based on the signals new value. Computed values (createMemo) are special because they are both an effect and a signal themselves, meaning when they get notified of a change, they will change their own internal value AND notify other computed values or effects of this change causing a cascade of updates. That's it. Nothing more. It boils down to the Observer Pattern, but since signals automatically register themselves as subjects to their observers, the dependency acquirement process seems "automatic" to the user.

This isn't an unpopular opinion, it's just a fact.

I have been waiting for the COBOL developer shortage for a over a decade now. I worked in non mainframe COBOL from 1986 to 1998. I transitioned to Oracle PL/SQL in 1999 and that's where I've been ever since.

Some people just like the sound of their voice.

0:06 rare occasion nerds and normies both relate

Pepperidge farm remembers.

we need new browser engine period. very impressed with what Andreas and chris have been doing. thank you bringing these legends for interview.

Pretty neat

started out with COBOL in 1979. ended up with .net. just retired at 62. it was a fun ride.

Yeah, your opinion sucks. It scales fine. Seriously, Google uses it with a massive monorepo. If you're doing something that doesn't work, you're the one making a mistake. Dude, GitHub runs off of libgit2... an open source library implementation of Git. Your opinion is wrong because you have proven that you're uninformed. Repos are cheap, as well.... who cares if they have tons of repos.

It's an unpopular opinions because it's a wrong interpretation of what people mean when they say easy. Everything labelled as "easy" or "hard" is relative. Difficulty is not an intrinsic property. Python (or Go, w.l.o.g.) is "easy" only because it is easier than many other languages. If you take a large enough random sample of people from the earth (including non-programmers), and let one group learn python and another group learn rust (or even assembly), then you will see very quickly that the people learning python will produce software orders of magnitude more functional or elaborate and the people learning a "difficult" language. Nothing is intrinsically easy. And everything that is indeed "easy" can present itself to be endlessly difficult the deeper you go. The generalization that every programming language is difficult is an empty statement with no relation to reality, you could take anything in this world that is unanimously labelled "easy" and elaborate on how this thing could be very hard. Humans are lazy, so they say "easy" instead of "easier than most other languages". But it's the latter that is implicitly meant most of the time

Wrong. Learning Go is easy. TEACHING GO is different and you can't approach it the same way as other languages. We don't have bad Go students, we have bad Go teachers. And people who don't understand the first thing about programming language education as well, apparently.

Video starts at 2:25

Yeah this is wrong. Q/ZMK are huge productivity boosters

You can convert all of your Mainframe pieces to a server now with no loss of function or speed. In fact, it will run faster and cheaper than dealing with IBM. COBOL is NEVER going away.

Sorry but didn't make any clear sense to me

i don't understand the hate comments? Git is very a old technology ,people just become used to git existing problem and its still isn't going to change anytime soon.

Love the show!! Please add chapters though

Folks like to spend money on stupid shit. We just see the same thing on GitHub. Trust me, we’ll end up where Idiocracy projected us.

Isn’t it a matter of diminishing return? Basic data analysis is really useful for understanding your business, but you might not be able to extract that much more knowledge by crunching even more data?

Was in Cobol from 1977 to 2000. After Y2K the job market dried up, Oh we don't need you anymore! So like a lot of Cobol programmers I went and did something else, I ended up as a manager of a big box store.

This isn't an unpopular opinion 🥲

I realized what was behind that laugh...

No way, I love me some Bobiverse! Thanks for all the great books Dennis Taylor, can't wait for the next one to drop soon! Oh yeah, first!

I am a Canadian citizen, with 5 years of experience in Micro Focus COBOL programming, and I need to find a job, can someone help with that

This really helped me!

The only people who think this is unpopular opinion are frontend devs 🤣

I recently started to work on a Tauri app with Next, and setting up the frontend was a breeze, everything just works! But then I wanted to set up a simple local database and realised I would probably have to spend the better part of a decade to learn arguably the most difficult language around to make anything useful on the backend (maybe a slight overstatement, but not by much!) If you happen to be a frontend developer that also knows Rust, Tauri is for you! But the other 99.9 % of developers should probably stick with Electron

I have an idea and observation I would like to share with you. It's about developing a particular type of neural network to be used in large language models. In many neural networks each input is mapped to each node in the hidden layer and each hidden layer is mapped to another adjacent hidden layer until the output layer. I propose that the hidden layers be partitioned into groups and that in between input layer and the hidden layers there are placed logical circuits or the equivalent thereof. And that their output would be sent to the adjacent hidden layer, etc.. By training on a network with this type of configuration each partition of the neural network would take on qualities that are specific to a category of sorts. So if for example an XOR circuit came between two partitioned layers it would be able to prevent both layers from operating at the same time, because of course, XOR is equivalent to saying statement A or statement B but not both. Of course you would want to consider using any logical circuit such as , NAND, OR, NOT, AND. As you well know any logical circuit can be built from any of these logical components so it would not necessarily be useful to mix all the components, however, it would probably be a considerable help to be able to mix them so that when the analysis of the neural network was done you would have some clue as to what section was working and choosing to produce the output and why it may have done so. This concept is very similar to a famous NSAT problem, although it is strictly different as you are aware. However, it does have some of that flavor. As an extremely simple visual aid I ask that you picture a neural network divided into sections and separated by a NOT circuit. If I were training the circuit with an input, "a cartoon of a dog is not an actual dog", the NOT circuit would prevent a section of the neural network from outputting something like, "a cartoon is an actual object". Note: Emailed to various parties developing neural networks including, Open AI, Google, Meta, Anthropic, Amazon and various comment sections on the internet. Second note. Many of these companies make it difficult to contact any person. If for some reason this idea should prove useful but did not reach your company, I hate to be so straightforward and bold, but I say it's all on you.

Hi can u pls provide example of providing summaries to clients considering they can use chargpt

Did COBOL for 17 years then newer languages for the last 15 years. I've only been on two projects that converted COBOL to Java or .NET. I was the spelunker diving into the COBOL code and wrote out the specs for Java developers or did it myself. I keep hearing there is a shortage. I offered to do COBOL part time remote. But stubborn clients won't compromise. I do prefer Java, C#.NET and Python but so long as I am having fun and being challenged I'm good. OK so I don't want to use ISPF editors. Hope that is not still around. We were doing Micro Focus COBOL last time I worked with it.

I didn't quite get the point (even though I have used and like Elixir). If the point is, that the BEAM can utilize all cores, so can optimized low level frameworks. And since Elixir isn't that good at number crunching, shouldn't the Machine Learning part use native code anyways? And if that is the case, what prevents me from using this native code in other environments outside of Elixir? I do understand the point about the distrubuted GPU machines. But I think that doesn't make Elixir "a great match", I could just use a simple API to communicate with a GPU machine in environments outside of Elixir.

"unless you have a reason to" - the reason varies depending on if it's your personal project, community project, working at a startup, enterprise or operating in a regulated industry. As always, the answer is: "It depends.."

I’m one at my company.

Neovim + Kickstart.nvim 10/10

I do because if API:s change gradually it tends to be much easier to adapt to each smaller change than to port from version 1 to version 10.

I retired from the grind almost five years ago. I really enjoyed developing mainframe applications, which I would lie to do again. But, any positions I find are 24/7 support. There is no way I want to clean up after the kids. And management is another problem. I've had hot-shots tell me that I've forgotten my skills ad can't perform. You don't forget 30+ years of mainframe. Why do I want to get back into the field?

Legend ❤