141–150 of 731 posts
AFAIK, any modern development is so complicated. Setup a web server: * VM from Cloud provider * Installing a bunch of staff * Throwing in a k8s cluster * Service mesh * Yaml files * Various network configuration * DNS * Firewall * Security etc.
Alternatively: provision a Heroku dyno, add a single-line Procfile, git push heroku and you're done.
Earlier quoted context omitted.
The trouble is that "don't build shit software" isn't actionable advice on how to avoid building shit software. For someone seeking to be a more skillful web software engineer, there's no information there.
Just don't build software and then we won't have shit software
I think it should be a more widespread idea that people should aim to leave a company with a net-negative `cloc` contribution.
Earlier quoted context omitted.
It means the underlying problem is not the tooling, but the use of the tooling. In other words, "I can choose the correct tool but still fail from poor approach of problem."
If the tool allows you to shoot yourself in the face without warning, and has a reputation/history of doing it to a lot of other people, then the tooling is indeed the problem.
Webdevs tend to suffer from an inferiority complex, that they are not considered “real” developers by the rest of the community. So they have responded by making the simple things they do monstrously complicated to try to prove something to the C++ guys, who simply don’t care, until they try to use the dumpster fire that is any modern website, that is. Then it has the opposite of the intended effect! Someone else men…
I mean, you only need check out all the vulnerabilities in said C++ software for verification of that fact :)
I think what you're trying to say is that you don't consider 'webdevs' to be "real" developers.
When he was born I used to dream of having him look at a button on a web page and him asking me "what actually happens when I click that button"
I shudder to think where I would even begin to answer that today.
Earlier quoted context omitted.
Did you consider that the next developer might well come along and post a similar comment: > It replaced a site that was pretty much a home-grown PHP framework written by a single developer. Random mixed logic in templates, hard to update CSS and a poorly designed 'not-an-ORM' ORM. Not to say that your rewrite is like that, but one man's garbage is another man's treasure.
And here's the trick to not building shit software: don't build shit software. It's a crude way to put it, but the reality is that you can build bad websites in any framework (including old PHP stuff and new JavaScript hype). You can also build good websites in any framework. It depends way more on the organisation or team developing it than on the libraries you use.
The point of (well-designed) frameworks is that they act as “rails” and make it difficult to make bad decisions.
They also make it easier for people on a team (or people who maintain the project after you leave) reason about your code.
Therefore, as far as I’m concerned, not using frameworks is a huge red flag.
edit: bring on the downvotes!
It isn’t. With modern frameworks (e.g. React + Typescript + some CSS framework), it is surprisingly straightforward to build the front ends for highly complex applications.
It’s true that you can create a MVVM app feeling with a MVC framework and a little Ajax, but really, I’ve never been more productive than I have with react.
I don’t think it’s really that much more complicated, but I do think 99% of the learning resources are outright terrible, and that was certainly a hurdle I had to overcome. Which is hard, because learning new things while you have a full time job is already hard.
Earlier quoted context omitted.
Who is considered a "real" developer these days?
Developers who don’t care if anyone thinks they’re real developers, I guess!
Earlier quoted context omitted.
Everyone(well almost) wants a SPA or at the very least non reloading pages. No one wants to reload the page with every action they perform or if only a part of the area on screen needed to be updated and the page was reloaded anyways.
If it's a traditional website and not an app, then I'm totally fine with it reloading every time you navigate to a new page or submit a form. Modern servers and browsers are plenty fast for full page reloads, as long as there isn't anything extra slowing down the load or rendering.
Earlier quoted context omitted.
"Everyone wants a SPA" is such bullshit. GitHub used to just load for a hundred milliseconds and then show me everything I needed; now it loads for a hundred milliseconds and shows me a spinner while it does another roundtrip for some javascript, then parses and executes that giant truckload of code, then does _another_ roundtrip to get the JSON or whatever describing the data I actually want to see, then executes it…
GitHub works without JS just fine. Switch it off. In fact, I'm pretty sure Github is rendered server side for the first load. You shouldn't be seeing any spinners. I'd check if I wasn't on my phone.
Sure, the list of my repositories is server rendered, but I already know my list of repositories. The thing I'm waiting for 99% of the time when loading github.com is to see if there's any new activity.
Sorry about the graphical glitches, blame nvidia or xorg or gnome or something.