4 ms·

Retro-Simplicity

17 points by ingve 3y ago

eternityforest 3y ago
The main thing that changed is expectations. I could make a simple utility far easier now than ever. But I'm less likely to, because I expect multi device sync, edge case handling, a long history of successful use, mobile support, support for common file formats, etc. Personal sites are still really cool, but I appreciate them more as creative expression more than anything, and the tech still has a long way to go in the convenience department, now that everything is SSL(Although neocities seems to have it down for some kinds of site!). The challenge of simple tech is no longer the tech, it's finding the applications, or I suppose for some in learning to appreciate the simplicity enough that you don't mind the extra work. It's why I always think knot tying is so unique and special. It hasn't lost any value at all. It's not a point of controversy, or a retro thing people go out of their way to do. If you know knots, you'll probably find a scenario where it's the obvious first choice. I don't encounter that kind of thing with simple tech nearly as often. The existing general purpose stuff is easier than any custom app I could make, everyone already knows it, there's usually some foundation developing it, and the path of least resistance and least chance of major hassles, at least to me usually seems to be adapting the task to fit the most popular well established tool.
RetroTechie 3y ago
Came across that site before, really nice! Problem with modern systems is mostly feature creep, I think. A crazy-hungry need to support any innovation or refinement introduced elsewhere. By now we could have ultra-responsive, dirt cheap devices that last a month+ on a battery charge, with a full suite of productivity apps & development tools built in, booting in 1s from flash. But of course, monochrome LCD character displays won't do. In comes color bitmap graphics. 2D won't do, software 3D is slow, so hardware 3D acceleration must be included. A couple of programming languages won't do. 30+ of them are needed as minimum. Text-menu based UIs are handy, but ehm... ugly? In comes GUIs. Simple, responsive, reliable GUIs could be a thing. But of course we need resizable scrollbars, user options for where to put a taskbar, themes, resizable fonts rendered to subpixel accuracy, user settings for keyboard repeat delays or mouse double-click speeds, widescreen & rotateable monitors, and beautifully animated drop-down menus. Picking some standard cpu architecture would be nice. But of course there's supercomputers, uC's, internet servers, car electronics & many more. Many of which use cpus optimized for their market. Then from screen+keyboard we go mobile. Much of which on devices with tiny screens, using touchscreen & thumb-scrolling for input. This demands new UIs. And with built-in camera's, video capabilities are important. Having all that, of course most software should run on as many of those systems as possible. With a consistent look & feel everywhere. And it's logical that code-once, press a button & run everywhere, is the norm. But all our data lives in the cloud these days! So having a full-featured browser (preferably a selection of them) is a must. Almost forgot - security! Software systems used to be based on open & friendly interaction. Not anymore! Networks are hostile. To fix, use process isolation, sandboxes, VMs, etc etc. Not saying all the above isn't nice to have, useful or good-looking. But all those features & checkbox items DO come at a cost. Worth it? You decide...

17 points by ingve 3y ago

The main thing that changed is expectations. I could make a simple utility far easier now than ever. But I'm less likely to, because I expect multi device sync, edge case handling, a long history of successful use, mobile support, support for common file formats, etc. Personal sites are still really cool, but I appreciate them more as creative expression more than anything, and the tech still has a long way to go in the convenience department, now that everything is SSL(Although neocities seems to have it down for some kinds of site!). The challenge of simple tech is no longer the tech, it's finding the applications, or I suppose for some in learning to appreciate the simplicity enough that you don't mind the extra work. It's why I always think knot tying is so unique and special. It hasn't lost any value at all. It's not a point of controversy, or a retro thing people go out of their way to do. If you know knots, you'll probably find a scenario where it's the obvious first choice. I don't encounter that kind of thing with simple tech nearly as often. The existing general purpose stuff is easier than any custom app I could make, everyone already knows it, there's usually some foundation developing it, and the path of least resistance and least chance of major hassles, at least to me usually seems to be adapting the task to fit the most popular well established tool.

RetroTechie 3y ago

Came across that site before, really nice! Problem with modern systems is mostly feature creep, I think. A crazy-hungry need to support any innovation or refinement introduced elsewhere. By now we could have ultra-responsive, dirt cheap devices that last a month+ on a battery charge, with a full suite of productivity apps & development tools built in, booting in 1s from flash. But of course, monochrome LCD character displays won't do. In comes color bitmap graphics. 2D won't do, software 3D is slow, so hardware 3D acceleration must be included. A couple of programming languages won't do. 30+ of them are needed as minimum. Text-menu based UIs are handy, but ehm... ugly? In comes GUIs. Simple, responsive, reliable GUIs could be a thing. But of course we need resizable scrollbars, user options for where to put a taskbar, themes, resizable fonts rendered to subpixel accuracy, user settings for keyboard repeat delays or mouse double-click speeds, widescreen & rotateable monitors, and beautifully animated drop-down menus. Picking some standard cpu architecture would be nice. But of course there's supercomputers, uC's, internet servers, car electronics & many more. Many of which use cpus optimized for their market. Then from screen+keyboard we go mobile. Much of which on devices with tiny screens, using touchscreen & thumb-scrolling for input. This demands new UIs. And with built-in camera's, video capabilities are important. Having all that, of course most software should run on as many of those systems as possible. With a consistent look & feel everywhere. And it's logical that code-once, press a button & run everywhere, is the norm. But all our data lives in the cloud these days! So having a full-featured browser (preferably a selection of them) is a must. Almost forgot - security! Software systems used to be based on open & friendly interaction. Not anymore! Networks are hostile. To fix, use process isolation, sandboxes, VMs, etc etc. Not saying all the above isn't nice to have, useful or good-looking. But all those features & checkbox items DO come at a cost. Worth it? You decide...