489 karma · joined May 22, 2014
It works in a very different - and very keyboard-centric - way. Having to touch a mouse slows users down dramatically, and so they keep closing our "pretty" app and opening up another terminal session. Can you imagine using web apps without touching the mouse at all? Or even most GUI apps?
And the speed... the time it takes to navigate from one screen to the next is milliseconds. They'll be typing three or four screens in advance. And they'll have four to six terminal sessions open. There's no GUI app that comes close, and I'm not sure it's even possible.
So yes. Web as desktop is bad. Really bad. Backwards. Unusable. Broken. And the people who work on it can't even understand the problem - that user efficiency and speed are REALLY IMPORTANT. Until "modern" devs understand that sometimes the front-end of choice is "ncurses", we're going to continue to have lots and lots of problems.
That's fine, though. My plan for retirement is to make millions supporting desktop apps that can't be replaced because the "web for desktop" programmers have a ceiling on the level of app complexity they can handle. Line of business apps require 1 to 2 orders of magnitude more effort and so the things written before that paradigm are not economically feasible to replace.
If you thought knowing COBOL in 1999 was profitable, just wait 20 years until only a few old-timers know how to maintain the millions of custom WinForms or VB apps that make businesses work.
The bad idea is those companies then sharing information about children that should be private and parents should have a say in. Blaming parents when companies have outrageous privacy policies regarding child accounts is ridiculous. We have to teach our children to survive in this world.
My 9 year old triplets all have greenlight cards. This article is very disturbing. But they are learning fiscal agency, which is something my generation wasn't taught at all. When I finally got a job (at 15) it was way too late for me. I spent it all on coin-op video games. That might be reasonable at 9, but not at 15.
We aren't here to discuss how great your parenting skills are. My ex has a narcissistic personality disorder and ... well. Any time somebody brags about how great a parent they are - especially when it's in the context of controlling children - I automatically think they aren't a great parent. That's probably just me and you probably are a really great parent, but I promise you I'm not the only one here thinking it.
Honestly that's the best thing you can do for on-the-job sanity. I understand that people feel they have a right to abuse others, but at some point you really don't have to take it.
Switching from the "hot" framework we started using last year to the "hot" framework that everybody loves this year is a step backwards from actually solving any problems. And two steps back from being able to maintain the resulting system. The world of startups is so very "now" focused that I think a lot of people (especially on Hacker News) forget that most programmers are still dealing with code written 20, 30, 40 and in extreme cases 50 years ago.
Better is great. Easier is great. But most old code can't be refactored, only replaced. Which means everything that interoperates with that old code has to either be replaced or it has to work exactly the same way. Sometimes that makes sense. Sometimes that's an absolute disaster.
Something that looks "simple" to you really can be too complex in a greater context. Rant all you want, this is just a fact of life. The software development profession has been so fascinated with shiny new toys for the past 25 years that we tend to forget that the hidden (and dangerous) part of the iceberg is not only in the past, but what we're creating now will just make it all that much bigger and more dangerous unless we're very careful.
But. I know places that are so stuck in the past that they are paralyzed and can't do anything at all. There certainly is plenty of "we've always done it this way" that needs to be thrown out the window. Introducing some new workflows or tools could really simplify things moving forward. The problem is knowing which is which. In order to know which is which you need someone who understands the system(s) at a "heroic" level.
I see us moving that way with things like "full stack" or "devops", but those are mostly only considering newer software. Making 30+ years of IT infrastructure work effectively and efficiently really is an epic task. And not one most CIOs I know of are at all capable of.
Sorry, I guess you triggered a flood of things I've been thinking about this past year. I guess I would sum it up by saying that your own rant is very personal and doesn't necessarily take the context that software is developed within into account.
B: OSX. The OS400 of personal computer operating systems.