Hacker News Reader

Comment by chongli | Hacker News Reader

This article is a bit strange. It seems to completely ignore the existence of Frotz [1], which has been around since the 1990’s. You can run any Z machine games you like on Frotz and I’m not aware of any compatibility issues. So I find it puzzling why the opening paragraph of the article would say something like this:

The source code for many of Infocom's foundational text-parsing adventure games, including Zork, has been available since 2019. But that code doesn't do anything for modern computers, nor even computers of the era, when it comes to actually running the games.

[1] https://davidgriffith.gitlab.io/frotz/

View on HN

It doesn't ignore it; it explicitly says "Lots of work has been done in open source realms to create modern, and improved, versions of these interpreters for pretty much every device imaginable."

That opening sentence may be a bit clumsy, but what it's trying to say is that the games don't run directly on hardware on any of these machines, but instead runs in an interpreter.

Right, but those interpreters have been available for decades. That opening paragraph is an attempt to generate hype based on a false premise. That a sentence tacked on at the end of the second paragraph completely obviates the problem set up by the introduction only underlines the point: this article is much ado about nothing (blogspam) so we’d be better off linking to and discussing the original source.

The details about the original interpreter are what is most interesting for the Hacker News audience, and that’s been buried by that clumsy attempt at sensationalist journalism. For anyone here interested in playing the old games, my original comment gives them the details directly.

But you said the article completely ignores it.

The point isn't that interpreters weren't available. It's that the original source wasn't available. The blog post talks about comments and variable names, that's what is valuable here.

It sound like they found the original compiler/interpreter/toolchain for it.

Frotz seems to be a modern-ish re-implementation?

Yes, that’s all correct. My point is that if you’ve wanted to play Zork or other Infocom games on your modern computer, you’ve been able to do so for decades using Frotz. The article makes it sound like running the games had been impossible until now, which is not true.

The main reason the discovery of the original interpreter source code is interesting has nothing to do with running the games. It’s for historical interest and for anyone looking for some insight into the development process for an early example of a portable virtual machine. Unfortunately, the source code is light on such details.

> The article makes it sound like running the games had been impossible until now, which is not true.

Good luck getting Frotz running on a C64.

Not Frotz, but ozmoo, I could run some Z5 games with patience.

Now, for v8 games, even with a REU they're really slow.

It also confuses the difference between the Z-Machine and the Zork Implementation Language (ZIL). The Z-Machine code is the part that is platform independent. ZIL is a Lisp-like language that Infocom used to write their games, but it just ran on their in-house system (although recently people have recreated the ZIL compiler for hobbyists) -- it was complied into Z-Machine code. But there really isn't a fixed connection between the two. Today, most Z-Machine interactive fiction is written in Inform rather than ZIL. But Z-machine interpreters don't need to know how the Z-Machine code was generated, much as the JVM and .NET systems don't have to know where their code came from.

If you can force yourself to the end of the second paragraph:

Lots of work has been done in open source realms to create modern, and improved, versions of these interpreters for pretty much every device imaginable.

The InfoTaskForce interpreter was one of the earliest 3rd party z machines. That dates to 1987 and was originally developed on the Macintosh but was ported to DOS, Unix, Amiga, Atari, ... Zip (Mark Howell) was also highly influential and ported everywhere. The only definitive date I can find is for it is version 2 (late 1993) but there are usenet references to it in 1992. Frotz was a re-write of zip and dates to 1995 or so.

Reply on news.ycombinator.com