> This throws me back to the good ol' days of fitting all software in a few dozens/hundreds of KBs on Speccy and similar systems.
And yet very few developers would want to go back to the days when we had no syntax highlighting and editors couldn't hold more than 64kb of text at a time and the OS could only run one thing at a time. Times change (thankfully). The 64kb machines now live in the domain of hobby projects, not machines for "getting things done." (There's a rather active group of folks who still hack on the C64, C128, and similar machines. Search for the "MEGA65" to find a recent (late 2021/early 2022) machine which was built solely for that crowd.)
> And yet very few developers would want to go back to the days when we had no syntax highlighting and editors couldn't hold more than 64kb of text at a time and the OS could only run one thing at a time.
Syntax highlighting, multi-megabyte file editing, and multitasking were all standard on ordinary software development PCs twenty years ago; perhaps closer to twenty-five.
And those ordinary software development PCs had eight, sixteen or thirty-two MEGAbytes of RAM -- not gigabytes, like today. If you're saying syntax highlighting, multi-megabyte file editing, and multitasking are the reason we need three orders of magnitude more RAM to get the machine to do the same thing (only often more slowly) now as then, you're either selling something or you just don't know what you're talking about.
> you're either selling something or you just don't know what you're talking about.
There's no need to be combative. i've been programming since the 80s (mid-90s professionally), and in _no way_ long to return to those days. It was fun while it lasted, but the dev tools, environments, and targets we have nowadays are a lot of fun in their own ways too, and many (perhaps most) of them wouldn't be possible/feasible with 16MB of memory on a 120mhz PC.
i'm selling nothing but the idea (for $0) that people expect far more from software now than then, and there's no way to run a modern app in such a limited environment.
I'm giving away my ideas for free, so I'm morally superior to you! ;-D
Yeah, sorry, didn't mean to be more combative than that.
The idea I'm giving away is that 99.9% of the stuff people expect[1] from software nowadays is just plain unnecessary. Actually detrimental, in many (most?) cases: It just complicates things.
___
[1]: Or perhaps, I suspect, that software developers nowadays just think people expect from software, so they provide it without even reflecting whether it's really necessary at all.
> Or perhaps, I suspect, that software developers nowadays just think people expect from software, so they provide it without even reflecting whether it's really necessary at all.
For the most part i blame that on the marketing folks rather than developers, but FOSS projects like this one have no marketing folks who i can push the blame off to ;).
The marketroids are the ones that make devs think that, yes, but whichever way they get it, the devs end up with that exaggerated impression. Affects FOSS devs too; they can't make themselves selectively deaf only to marketroid-speak. Well, I guess in part one can, but given how pervasive it is throughout the industry, not entirely.
And yet very few developers would want to go back to the days when we had no syntax highlighting and editors couldn't hold more than 64kb of text at a time and the OS could only run one thing at a time. Times change (thankfully). The 64kb machines now live in the domain of hobby projects, not machines for "getting things done." (There's a rather active group of folks who still hack on the C64, C128, and similar machines. Search for the "MEGA65" to find a recent (late 2021/early 2022) machine which was built solely for that crowd.)