3 ms·
Id say that its the wrong way to think about it. As long as you are working in the software engineering field change is a near certainty. You will not hit on
by ThinkBeat 3y ago
Id say that its the wrong way to think about it.
As long as you are working in the software engineering field
change is a near certainty.
You will not hit on a technology / framework / product
that will keep you going for 40 years. [1]
Working with Flutter you have learned a lot that is valuable going forward.
Learning to adopt new tools is part of the job.
- nailer 3y agoTime is a limited resource. Choosing to invest in the right technologies at the right time is the important part of a career in tech.
- mgkimsal 3y agoC, Java, PHP, SQL, Python, Ruby have kept a lot of people going for more than 20 years, and will likely continue to be relevant in some capacity for at least another decade (if only maintenance, not even including new projects started with those tech stacks). I've been working with PHP and some variation of SQL for the past 28 years, and it will go on for a while longer. I've also worked in the Java world, and have done a smattering of other tech along the way as well. Not everything lasts 40 years, but many things last more than 3-5.
- ThinkBeat 3y agotrue. But Java 2023 is quite different from Java1. The tools you use now are different (unless vi) You have had to learn a lot of new technologies, frameworks, syntax to keep up. PHP has also gone through so many changes that it is near a different beast these days than when it first appeared. So if you had relied on your skills you learn as PHP 1.0 and Java 1.0 today you would not be very productive.
- mgkimsal 3y agoMuch/most of Java 1 will still compile, at least (FWIR). The evolutionary progress of java 1... to whatever it is now (current one I'm using is 17, but I know there's more past that) has largely been adding of new libraries/methods, etc, and the majority of your java skills from 10-15 years ago are still valid (although might not be the most efficient any more). I've watched people go through Flash/Silverlight/Actionscript, multiple MS tech stacks, and many many other promising tech. Many have had to large scale reinventions of themselves and cram 'new to them' tech as the old stuff has gone away. It's definitely easier to evolve with a tech stack over the years. You get a better sense of idioms, best practices, etc. that is hard to replicate overnight. And yeah, some tooling is different over the years too, but yeah, vi still works :) Shiny new 'hey look at this' ideas can pop up in newish langs, but diving deep in to them isn't often the best path for most people. The truly good and impactful ideas tend to get recognized and adopted in the more mainstream ecosystems over time, giving you most of the benefits without needing to throw out much of your experience.
- pjmlp 3y ago> Much/most of Java 1 will still compile, at least (FWIR). ... Actually it won't, past Java 9, as after the modules introduction it was also decided to take action on deprecation, now things get really removed after a couple of releases.
- mgkimsal 3y agoInteresting... I have a project that was originally done in J8, and... it's still working/compiling as of a couple weeks ago, but it's not necessarily as advanced as it could have been. I think we moved to J11 a few years ago, and IIRC J17 in 2022, and didn't hit any issues other than some dependency versions.
- itpragmatik 3y agoI started doing Java (and sql/Oracle) development in late 1990s and even today I am deploying REST APIs written in Java, MySQL to AWS. So, yeah, it has been about 23+ years so far!