the 3 stages of me discovering a new tech:
- anti-hype: ew it's new must be bad
- engagement: oh so actually cool I'll use it for everything
- desillusion: ew I'm disappointed I was right it's shit
- acceptance: oh it's actually good when used properly
Now I like Go and Docker. It took some time, mostly for me to know why I should hate some examples and appreciate its few good uses.
Some are still mostly hated, like Node. (which I just can't really like when there's Go for instance)
Some I always liked, like Rust (which didn't really have this hype phase I felt a need to oppose, it's just slowly getting better)
Node... I just feel like it's a bad language and a bad ecosystem. And we took that bad base and pushed it hard enough to work in production for some reason and now it's mildly working and it's just sad we spent all this effort there.
Just imagine what we could have now if all that work and resource were put into... Rust?
Probably a very unpopular opinion, but I don't like the fact that there are so many languages and frameworks. It's extremely suboptimal. Diversity is great but we surely could profit from having less redundancy.
Is there anything good coming from having Ruby/Python/Perl/JS? They are overlapping on most point and solving the same problems in the same ways. We could reasonably just keep one or two.
@jkb I don't mean One True Langugage, but... One okay syntax common to everything. I can't think of anything Ruby does that Python doesn't for instance, we could write ruby-style python and enjoy not having to translate modules. Same for JS, there are interesting paradigms but not strongly linked to the language itself.
@CobaltVelvet I entirely concur.
.@CobaltVelvet No two people will agree on which are good. I'd keep Python and JS/Node.
Python's elegant, fairly fast, and perfect for solving logic/statistics problems.
JS is the language of the web on front and back, and Dylan/self-style objects are awesome.
Perl's an obsolete hack of awk, forget it.
Ruby's an abomination, all the elegance of Perl with the speed of BASIC.
@mdhughes At least two people already agree. :p
@CobaltVelvet You just know someone will complain that they miss the functionality of an obscure function from language X when it is dropped in favour of Y.
@CobaltVelvet yes, but *which* one or two? the software world is notoriously bad at coming to a decision on this sort of stuff.
Vim or EMACS? thirty years, and we're still at it.
One thing I *think* you're implying is some languages/platforms are pushed just for fashion, as I think is now the case with having anything Docker because reasons.
There are instances in which Docker (or a language/platform) is perfect for the job, and others when it's merely being used because fashion/habit/show.
@CobaltVelvet I think the One True Language Per Purpose approach has been tried before and failed miserably, yes they are overlapping but each one has its own set of strengths and its own approach to solving the same problem. Diversity may not be a bad thing here, especially since we have the Internet so we can have many people working on each project despite them not working in the same company or living in the same area.