6

Softwares like Internet Explorer 3 and Photoshop 1.0.1 are on-topic. They are considered retro and their codebases only partially resemble the modern versions. However, a consensus about software that has not been removed from the codebase has not been reached.

Case study: Windows Explorer Shell Command Files

A feature added to Windows with Internet Explorer 4 (95 / NT 4). .scf files behave like shortcuts in the shell - they can't be opened in programs using the built-in Windows open dialog's GUI - only by typing in the full file path. These files were used in the now-defunct Quick Launch bar to show the desktop, open IE4's Channels feature and open Windows Explorer(?). These files have not been created by Windows since XP.

Case study: Macintosh QuickDraw API

Any Mac OS GUI developer will remember QuickDraw. This revolutionary graphics API allowed people to not only draw lines, rectangles, circles, images, text and more, but also allowed them to use Apple's patented "regions" technology. Apple implemented it into Mac OS X for backwards-compatibility; until Mac OS X v10.8 one could still compile programs using QuickDraw, and in v10.8 it was only removed from the SDK; QuickDraw programs would still run.

Are such software features on-topic?

  • 1
    What kind of question do you have in mind regarding these features? – Stephen Kitt Mar 2 '17 at 14:28
7

I say we try it.

We already have the for questions about adapting retrocompuing hardware on modern platforms, so why couldn't the same apply to old software or SDKs on modern software platforms? I guess we would have to restrict questions to using legacy APIs only, or maybe mixing modern and legacy APIs in a single program.

  • 3
    Definitely agree, but I think it should be more historical interest type questions than practical "how do I maintain this software" type questions from people working on real-life codebases. We don't want to just become a "stackoverflow for legacy" type thing. – Muzer Mar 1 '17 at 11:17
5

My two cents:

I tend to think the answer should be yes. The commercial need for backward compatibility can keep this software around modern code bases for quite a while. It's also hard to enforce... without access to modern software source code, who's to know what's in it or not? Even with access, to what extent would code have to be changed to be considered novel/replaced.

For me, I guess 'Retrocomputing' means 'Computing as it was conducted in the past'. That has very little to do with how that past manifests itself in current software.

1

Some technologies essential to modern systems might have been considered "new" when they were introduced, their success was in attracting much attention, and the time when that was the case had distinct practices of historical interest. These technologies are now "mature", and their success is in not attracting attention.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .