28 ms·
Part of the issue is that there are still things many people - even those of us who haven't used an Amiga in years - miss from AmigaOS. The tight integration o
by vidarh 6y ago
Part of the issue is that there are still things many people - even those of us who haven't used an Amiga in years - miss from AmigaOS.
The tight integration of everything via AREXX, for example (though AREXX is an awful language, the expectation of a tightly integrated AREXX port in applications was important).
The device model that led to a plethora of device drivers and filesystem drivers doing interesting things (e.g. FrexxEd, an editor co-written by Daniel Stenberg of Curl fame, exposes a filesystem where you can find all open files, so anything that can read or write a file can manipulate the editor buffer directly). Datatypes. Widespread "standard" libraries for pluggable compression. A culture of tightly controlled APIs to the point that e.g. the default file requester etc. was easily replaced across huge swathes of applications just by dropping in new libraries.
It's especially notable that none of it is down to technical limitations of modern OS's, but cultural differences. You could do the above trivially in Linux (or any other OS) if you wanted to, for example, but it only matters if it gets wide traction. Amiga had a culture of quickly coalescing around standards that promoted reusability and modularity and sticking to them.
And that is also part of the reason why some of the hardcore fans stick to it: Their 30+ year old applications can open modern file types if someone writes a datatype for it (and they often do). Their 30+ year old apps can transparently compress and decompress new formats if someone writes a suitable library for it, and so on.
I'm not going back to AmigaOS, but I do still wish we'd see a resurrection of that culture, because in the ~25 years since I used AmigaOS as my primary OS, while we've surpassed it in a whole lot of areas, there are still areas where it was more pleasant to use an Amiga.
- richardjdare 6y agoYes, I think the high degree of integration, the ease with which you could work with the full capability of the hardware, and the lack of cruft made the Amiga a highly immersive environment for the enthusiast user. I think about this a lot. I really want to know what the "Amiga feeling" was made of so I can reproduce or encourage it today. It was one of the main things I enjoyed about computers in the first place. I also think its important for Amiga people to pass these notions along somehow, in a similar way to people who've worked with Symbolics Lisp machines or other innovative technologies whose core ideas were not taken up by a mainstream that surpassed them for reasons unrelated to the actual value of those technologies.
- vidarh 6y agoIt's tricky, because it's difficult to separate the things we remember with a feeling of nostalgia but that do not really matter from the things that were genuinely useful. I keep wanting to build some libraries and tools to experiment with this, but a lot of it is about culture, as I mentioned, which is often harder than the tech. One simple example is the AmigaOS ReadArgs() call. It provides a standard mechanism for parsing command lines. An important consequence of it is that auto-complete for Amiga terminals that supports it can do typed auto-complete for any application that uses it without anyone having to write auto-complete script or specifications. It's not a hard thing to do. The ReadArgs templates are really basic. Probably too basic. There are tons of good open-source command line parsers. All that's really needed is for people to settle on one with a suitable static spec format, and a mechanism for making it easy to extract that format from an executable. But it serves very little purpose unless you convince a lot of people to do it. There's a chicken and egg scenario there, in that it's a lot harder to find people prepared to spend time creating "standards" like that and pushing them than it is to find people to work on solving an issue with a more immediate payoff. Especially when you really need a critical mass of applications that supports these specs before the benefits starts becoming obvious.