I don't know if my iOS app is using GPT-4o, but asking it to translate to Cantonese gives you gibberish. It gave me the correct characters, but the Jyutping was completely unrelated. Funny thing is that the model pronounced the incorrect Jyutping plus said the numbers (for the tones) out loud.
I think cars only look good relative to other cars. But as an object themselves, they don’t add to the landscape.
Look at all the concept renderings of future apartments or idyllic European countryside’s. There’s always only a few cars or one car. Together, they are an an invasive organism.
That metric is like the megapixel wars of cameras. Just something to chase so the newer model can be “better”. It does nothing except validate your purchase when you rev it obnoxiously.
To continue your thought, what are some things that need to be regulated in order to prevent future societal decay that the technology itself brings (such as allowing cars to roam free which propagated further car-centrism?)
Also, while Dubai is flashy, I wouldn’t really consider it a first tier city in terms of actual living. The urban planning is not human scale at all and they just defaulted to highways and skyscrapers.
Genuine question, as someone who got exhausted and is out of the industry for the foreseeable future. How do you stay competitive or job hop when every post is looking for the latest web framework?
Don't choose hamster-wheel career tracks in the first place. Try to find Lindy paths [1], [2]. For example, SQL experts and DBAs. The rise and (mostly) demise of NoSQL actually cemented SQL's reputation as an irreplaceable technology. The latest Web framework is by definition non-Lindy: it's the new kid in town who think they know it all. grep is lindy, Unix is lindy, the qwerty keyboard is lindy, C++ is lindy, Computer Science is lindy, algorithms and data structures are lindy. Knowledge of the real world, whether it be finance or shipping, is lindy. Sorry, verbose reply, but I hope it helps.
Interestingly, I've been thinking it would be a good idea to brush up on my C skills, because while there are serious contestants to its throne, a gig based on Rust today often means being at the cutting edge of tech, and will probably come with a lot of buzzwords like GraphQL, React, Typescript, monorepo, AWS Lambda, daily standups — things giving me symptoms of burnout just writing down.
Whereas a C gig today probably means embedded, or subsystem far away from the frontend churn. If I'm lucky they have unit tests. The Lindy path you were talking about.
My team was asked how we should store data that is variable in structure and depth. When I suggested using mongo and not trying stuff it into a SQL db, I was looked at like I had suggested blending my firstborn for margaritas. People seem to think that NoSQL is dead and was always a bad idea.
The NoSQL movement was absolutely acrid towards SQL. The blowback is justified, IMHO. "blending my firstborn for margaritas" is very colorful language, woah :-)
Plenty of IT jobs out there that have job security to last your entire career and don't shift to the new shiny every few years. Utilities are a good example. Lots of older tech and some newer stuff, but nothing experimental.
Ex: Maintain some SQL queries, upgrade to a new database version without bricking everything, write some scripts that get fired off at a scheduled time, knowledge of the industry you're serving, update some configuration files for the company website which won't change for another decade, read some log files, create some visualization displays for operations staff, install software on certain machines, manage the company's VPN stuff, configure Linux, do the paperwork...etc
All this stuff is more generic IT and less rockstar developer. It reliably pays the bills and isn't very flashy.
There are plenty of people in IT that write code every day, some that rarely write code, and some that basically never write code, but just manage the business side of it all (there is a lot of beauracracy in large organizations).
Map territory distinction -- what they are looking for (strong engineers regardless of stack) and what the role is (a description of what you'll be doing) are two different things. Find your way to doing and articulating impactful projects that use technology to measurably create leverage and value for the business, and you will be fine. Of course, that can be easier said than done.
I'm honestly thinking about just picking up c++ or something and turning my attention to embedded systems. I'm sure there's a wealth of old tech sitting around that needs to be maintained.
I'm in embedded, there's also cutting edge stuff to work on. The nice thing is you can be mostly detached from annoying web apis and frameworks and databases. The not so nice stuff is dealing with crappy vendors, using crappy tool chains, and spending a ton of time setting up automated builds. Sometimes I think about switching to true embedded (I do embedded Linux) but that tends to pay less
I hope it humbles whoever builds the next platform. Nothing lasts forever no matter how entrenched it is in the zeitgeist.
Once something gets big enough, it becomes impossible to moderate and eventually it will fall out of fashion and lose its original intent. Look at Etsy and Amazon.
I'm finding people are rearely loyal customers to a brand long-term. Rather, they want whatever platform gets them what they want. People jump from sites like Aliexpress to Wish to Shein to whichever webiste has the most latest and marketing team.
