Dates for ZxZvm "disks"

[8BitAG]

There are a lot of games in the database that run using Z-machine, the virtual machine created by Infocom, and later targeted by modern adventure writing systems such as Inform. They've been put onto disks that use the excellent ZxZvm to run them on the +3.

For example...
https://spectrumcomputing.co.uk/x-list?label_id=15855
...are games originally written by Infocom.

There is an issue with the "publication dates", though, which in this case are pretty non-sensical.

For example, we've got games like Infidel being listed as published in "1983"... for the Spectrum +3. Which is just nonsense, on many levels, if you're just reading the entry without any prior knowledge... this "release" wasn't published then, the +3 didn't even exist then, etc.

I'm not sure what to suggest to resolve this issue, but it doesn't feel quite right and logical as things are currently. Obviously there is an additional wrinkle in that the CP/M versions (published in the UK for the Amstrad CPC & PCW, but perfectly playable on the +3 with Locomotive's CP/M) had slightly different (later) release dates than the original Infocom releases. Some of these ZxZvm entries reference both ZxZcm and CP/M.

(I really have almost zero interest in Infocom & Z-machine... but as it came up on a Facebook group as being just odd, I have to actually agree with the poster there....)

[catmeows]

Well, it Is virtual machine at all. When I load Chuckie Egg into emulator, I don't care when was the emulator built.

[8BitAG]

Well, it Is virtual machine at all. When I load Chuckie Egg into emulator, I don't care when was the emulator built.

The way it's presented in the database at the moment makes it appear that these are official releases by Infocom with those dates. It is confusing. (I know it's confusing, because an Infocom expert has just pointed out to me on Facebook that it's confusing). At the very least these "bundles" of "emulator" & game files, should be flagged and labelled as being later, unofficially sanctioned productions.

ZxZVM is a great tool and I can see how these ready-built disks could be really useful (even if we really generally don't include unofficial releases in the database). As it stands, they're just not that well documented and the context is missing.

For example, we've got stuff like this...
CIA Adventure (ZxZvm)
https://spectrumcomputing.co.uk/entry/3 ... ture_ZxZvm
...in the database, listed as being written by Kevin Thomas in 1996... (actually J. Kevin Thomas, btw)

That's actually a z-machine port of a very early 1980 text adventure by Hugh Lampert that was written in BASIC for the TRS-80.

[8BitAG]

The fact is, if I search "text adventures" published in "1983" then I get Infidel by Infocom sitting in the middle of that list.

That *is* a nonsensical search result, without context.

[8BitAG]

Putting aside the dates issue for one moment, and the fact that the "original publishers" never published these ZxZvm versions, there are also major issues with some of the other publishers listed.

For example...

The Mysterious Adventures...
https://spectrumcomputing.co.uk/entry/3 ... es_1_ZxZvm
https://spectrumcomputing.co.uk/entry/3 ... es_2_ZxZvm
...For some reason these are listed as being published by Adventure International (UK). I'm not sure why this is? This original publisher of most of the games on the Spectrum were Digital Fantasia (although that's a whole thread in itself). You couldn't credit Digital Fantasia for any of these games, though, as they are (sometimes very poor and buggy) ports of the original games to z-machine. So they are basically fan-productions, not authorised versions.

The same goes for the Scott Adams games...
https://spectrumcomputing.co.uk/entry/3 ... es_1_ZxZvm
https://spectrumcomputing.co.uk/entry/3 ... es_2_ZxZvm
https://spectrumcomputing.co.uk/entry/3 ... es_3_ZxZvm
https://spectrumcomputing.co.uk/entry/3 ... es_4_ZxZvm
...They're all unauthorised fan ports (as far as I'm aware) of varying quality. They're definitely not the "Adventure International (UK)" versions.

All those games exist already in the database as proper, authorised Spectrum releases.

There are other games in the ZxZvm collections that were never made as z-machine versions back in the day. I've already mentioned CIA, then there's ones like Softporn Adventure....
https://spectrumcomputing.co.uk/entry/3 ... ssic_ZxZvm
...which was a 1981 BASIC adventure, and probably shouldn't be visible in Spectrum Computing anyway, following the "adult" rules.

Strictly, following the site policy on Activision games, the Infocom titles shouldn't be included. (Although they are obviously freely available everywhere)

[Einar Saukas]

For example, we've got games like Infidel being listed as published in "1983"... for the Spectrum +3. Which is just nonsense, on many levels, if you're just reading the entry without any prior knowledge... this "release" wasn't published then, the +3 didn't even exist then, etc.

I also think this information looks weird... although it's accurate!

The Spectrum +3 is really the minimum hardware required to run these Z-machine adventures, despite the fact they were created before the Spectrum +3.

I'm not sure what to suggest to resolve this issue[...]

That's the problem

The only option I can imagine is to move these titles into a new hardware category called "ZX Spectrum +3 (ZxZvm)", to distinguish them from regular "ZX Spectrum +3" titles. This way, you would still find them when looking for Spectrum +3 titles, but their pages would make it more clear that these are not regular titles developed for the ZX Spectrum +3.

What do you think?

[Einar Saukas]

Putting aside the dates issue for one moment, and the fact that the "original publishers" never published these ZxZvm versions, there are also major issues with some of the other publishers listed.

IMHO we should credit as original publisher whoever originally published these Z-machine versions. Even if they never predicted a Spectrum +3 user would be able to play them.

If you can help track down more accurate publishing information about these titles, I will gladly adjust these entries. Feel free to send me a PM!

[8BitAG]

I also think this information looks weird... although it's accurate!

I can't agree with the fact it's accurate. It is not. These homemade third-party "compilation" disks were not published in the year indicated. And they were not, in most cases, officially published by the organisations or individuals listed, e.g. Infocom. Homemade compilations are not official releases.

I can understand that it's really useful to have these pre-built disk images of the ZxZvm emulator together with sample z-machine game files, but the way they are being catalogued gives them the status of official releases. They are not and should be clearly flagged appropriately or they will unknowingly cause confusion in the future.

[Einar Saukas]

Again, I understand your point. So how can we do better?

[8BitAG]

Again, I understand your point. So how can we do better?

If it were me, I would list them as downloads on the ZxZvm entry; rather than individual entries in the database. They aren't official releases of games, they're just timesaving disks, put together by a third-party, of files to use with ZxZvm.

There are the odd exception, such as some of John Wilson's games, where he officially (and deliberately) released versions of his PunyInform games for +3 that used ZxZvm. I'm sure there are other Inform authors that would quite happily sanction a "Spectrum +3" release. Again, I'd have no issue with those being included... but I would suggest that they were put in the database with the actual date the +3 version was produced, not the original date the game was written.

There are thousands of z-machine games that people could create ZxZvm compilation disks of, if they were so inclined. Dropping them into the database, with their original dates and publisher attached and with zero context, would be a nightmare. It'd be like dumping the entire Commodore PET library into the ZXDB, with the original publication details of each game, simply because someone made a ZX Spectrum NEXT emulator for that machine. It might sound really cool and useful initially, but you'd end up with a database flooded with non-Spectrum stuff totally drowning out the "real" Spectrum titles.

[AndyC]

Yeah, I'd be inclined to say the "publish date" should reflect the time the +3 version was released. Not entirely sure the best way to handle the actual publisher though as it feels like they should appear published as homebrew rather than the original publisher of the z-machine version.

In the same way that if I made a SCUMM interpreter and "released" a version of Maniac Mansion, I'd expect it to be released this year and "published" by me, rather than indicate it was released back in the day by LucasArts.

Including some info about who the source files were originally produced by might also be useful, I'm not sure how best to represent that in the database though.

[Alessandro]

I did some of those +3 DSK files myself at Einar's request some time ago. I employed CPCFS v.0.9.0 by Derik van Zuetphen and Kevin Thacker.

All of them are in fact homebrew adaptations of Infocom's 'story' files made because Infocom did never release anything for the Spectrum.

Those with a DivIDE or similar interface and ESXDOS may run a 'story' file directly - provided it's encoded with Z-Machine versions 3, 4, 5 or 8 - with Bob Fossil' .zxzvm dot command - see here for more information: https://www.spectrumcomputing.co.uk/for ... php?t=6169

I believe the year should be that of original release, and a note should specify something along the lines of "Spectrum adaptation for ZXZVM".

[StooB]

Yeah, I'd be inclined to say the "publish date" should reflect the time the +3 version was released. Not entirely sure the best way to handle the actual publisher though as it feels like they should appear published as homebrew rather than the original publisher of the z-machine version.

In the same way that if I made a SCUMM interpreter and "released" a version of Maniac Mansion, I'd expect it to be released this year and "published" by me, rather than indicate it was released back in the day by LucasArts.

Including some info about who the source files were originally produced by might also be useful, I'm not sure how best to represent that in the database though.

The "Licence Tie-Ins / Inspiration" section is where this kind of information usually goes. Although it could do with a new type adding like "unofficial conversion" as it's more than simply "inspired by".

[Einar Saukas]

If it were me, I would list them as downloads on the ZxZvm entry; rather than individual entries in the database.

If they are not entries anymore, it won't be possible to find any of them when searching for title, author, company, etc.

I don't believe making games inaccessible from searches is a database improvement. There must be a better way to improve accuracy without sacrificing usability.

There are the odd exception, such as some of John Wilson's games, where he officially (and deliberately) released versions of his PunyInform games for +3 that used ZxZvm.

Could you please provide a list of the titles that have official ZxZvm releases? Do we currently have the correct publishers and release dates for them?

About the others, let's fix their information. I don't mind removing their release dates if that will improve accuracy. And perhaps removing their publishers too, referencing them instead as publishers of the original games/licenses. This way, these publishers will be still searchable, but they won't appear as publishers of the ZxZvm version.

Agreed?