How did Windows 95 decide that a setup program ran? | Hacker News Reader

How did Windows 95 decide that a setup program ran?

devblogs.microsoft.com

58 pointssoheilpro12 comments

That is one of the reasons why desktop Linux conquest is slow. Linux people won't approve such kludges, which means broken applications and back compatibility. I am not sure which strategy is correct, but I certainly have a compassion for users who just want their pc to work, and not search for library versions or manuals how to compile code in a new system.

I sometimes feel like Hal changing a lightbulb[1] when I'm trying to install an app on Linux, either I don't have the correct distro, I need to enter km long command line to add some repository to apt, I need to download 346 packages from I don't know where to the build the app, and finally find out that it's not what I wanted.

Yes, I know that's how Linux works and I know all the advantages of that, but when you see now that many install are just 'curl mywebsite.com/install.sh | bash', you start to wonder which is a "good" solution.

1. https://youtu.be/AbSehcT19u0

I wonder why specifically those languages. Were only those markets using localized file names and everyone else sticking to English?

Or was it just random chance, i.e. some setup programs from those locales happened to cause issues and got escalated, so those names got on the list?

In 95 localization was scarce. Also, the partial "inst" would have matched "install" translated to several languages (about 15 by my quick count, could be more).

Maybe many other languages were already covered by "inst" due to common language roots or just straight up borrow "setup" or "installer" from English because there was no established localized term.

"Ayarla" is very confusing... Setup can't be translated to Turkish as "ayarla". The correct word would be "Yükle/Kur" (verb) and the correct noun would be "Kurulum". Ayarla means "configure" and "ayarlar" means "settings".

Also, why just Turkish and Hungarian?

Same goes for the Italian "imposta". It's a direct translation of "set up", but not a verb anyone would use to mean "setting up a program", and I've never seen an installer named "Imposta.exe". Maybe they had an example of such a poorly translated program, though, which warranted adding a special case to the list. The correct verb would be "installa", which was already covered by "inst".

DiscMaster[0] is useful for this!

It seems to agree with you. There's nothing I could find named imposta.exe or aryala.exe, but I did find installa.exe.

[0] https://discmaster.textfiles.com/

Windows 95, for all its UX niceties, was a hack job under the hood. Reading this is just another example. When people make comparisons to clunky Model T cars in the 20’s and Win95, they aren’t far off.

> When people make comparisons to clunky Model T cars in the 20’s and Win95, they aren’t far off.

I think they are. Model T’s were clunky because they were minimalistic; Windows 95 was clunky because it tried to do impossible things.

This is a fine example. In the Windows 95 architecture, it’s impossible to detect whether a program is an installer, but they tried anyways.

Another example is Plug and Play. The original IBM PC parallel ports were designed for printers. That interface got extended (a few times. See https://computer.rip/2023-01-29-the-parallel-port.html) and reused to support tape drives, CD-ROM drives, scanners and whatever. Plug and play was intended to detect these devices, but with the original port about anything you sent on it to a printer resulted in characters being printed.

Also, there was no way for a printer to tell the PC it was a printer. All it could indicate were a few bits to signal they could receive data, were out of paper, etc.

So, how do you figure out what, if anything, is connected to a parallel port?

The answer is “very carefully”, and I don’t think it ever worked flawlessly. I definitely remember a tape drive that, whenever Windows started, flashed a few LEDs and got in a weird state that required some manual intervention.

>> When people make comparisons to clunky Model T cars in the 20’s and Win95, they aren’t far off.

>I think they are. Model T’s were clunky because they were minimalistic; Windows 95 was clunky because it tried to do impossible things.

Either way it's more like a 1966 Corvette compared to a modern Prius when it comes to speed. And technology.

I think I understand your point. But I could argue that a Model T was trying to do an impossible thing too. Like, contain a gasoline explosion to reliably create forward thrust in an affordable design. And make them quickly and reasonably safely. That was “impossible” before.

So I guess the lesson here is any first technology that is trying to do the impossible will be clunky. Fortunately there was a lot of car competition so we got much better cars. Assuming you exclude that industry’s successful lobbying to bend US society into car culture.

After thinking about this I think my analogy was correct but for a spectrum of different reasons.

Reply on news.ycombinator.com