Little bugs in the database 4

[StooB]

Many of these comments are from the WoS database I believe @StooB, but I'm sure Einar would have more details. The comments can just not be displayed if the majority of users were to be in agreement.

I can do that easily if wanted.

Unless someone is willing to volunteer to sort through them (based on previous similar requests that seems unlikely so we tend to move on and assume it's not that crucial), it would just be a display or not display decision.

They're all from ZXSR, not WoS. Having scanned through them I can't see much that can be salvaged apart from the awards listings, and even then I'm not convinced of their accuracy.

[PeterJ]

There are many notes from WoS too (via Jim Grimwood), not just the ZXSR ones. Do you want those to go too?

[Alessandro]

I remember WoS did not display such stuff. If you look at the entries, they all originate from ZXSR in fact. As for their accuracy, they are mostly rants from the same person, judging by their "style", and completely irrelevant like that comment about Light Force.

Database entries should not host opinions anyway. There's Spectrum 2.0 for that.

[PeterJ]

I'm sure I remember WoS displaying SPOT & SPEX comments [mention]Alessandro[/mention]?

[Alessandro]

I'm sure I remember WoS displaying SPOT & SPEX comments @Alessandro?

Yes but those were limited to information like the passwords for SU covertape games or alternative releases, not rants about this or that being crap or played in 1984 etc.

[PeterJ]

I'm glad I wasn't mistaken afterall. There are comments from SPOT & SPEX. These are considered useful though?

If we are talking about the ZXSR ones (that wasn't clear from the easier posts, but is niw) we can just remove them from the pages if this is the consensus.

Just let me know either way.

Many thanks.

[Juan F. Ramirez]

Both ACE 2 and Zythum have separate releases for Zafiro (tape without inlay) and Z Cobra (inlay without tape or release date). I guess they actually refer to the same release?

I couldn't find more info about both games. They have clearly Z-Cobra inlay & instructions but the TZX file says "Zafiro" with no additional info. They might be either Zafiro or Z-Cobra.

I had a search in SPA2, (spanish archive) and saw:

- Zythum: only Z-Cobra release, no mention of Zafiro.
- ACE 2: Both labels included, but every inlay has the Z-Cobra label. Different TZXs but I can't see which label they belong to.

I 'suspect' there's only a Z-Cobra release, but I can't prove it with the existing info.

[PeterJ]

Having scanned through them I can't see much that can be salvaged apart from the awards listings

It would be an on it off option if no-one wants to review them. Let me know and I can hide the appropriate fields over the weekend. It's quite straightforward.

[Einar Saukas]

The entry for Light Force contains a curious remark:

What the hell is that voting in the Crash Readers Awards for 1987 all about. As far as I remember, Lightforce didn't have any music and secondly it was released in 1986.

However if you take a look at Crash #38, March 1987, page 35, Light Force wasn't even nominated for best music - the winner was Ping Pong instead, as it is also remarked in its entry.

I'm afraid you are looking at the wrong issue. The "1987 Crash Readers Award" was published in April 1988:

https://archive.org/stream/crash-magazi ... 6/mode/1up

This comment from ZXSR is accurate. It doesn't make sense that it was elected both "2nd Best Shot-em-up of 1986" and "2nd best Music of 1987". Unless there was a new 128K version with music that we don't know about???

But this is only the beginning. The Light Force ZXSR remarks section contains many errors.

I fixed "1st Best Music" and removed duplicates in the ZXDB version that I released today. Please wait until SC gets updated tomorrow.

I didn't notice any other errors.

Probably they were added by different people and in different times

No, all these new remarks came from ZXSR. This information is visible in your screenshot.

without any care about inaccurate

I strongly disagree!

Let's keep in mind that users were reporting errors in Martijn's WoS database for 2 decades, afterwards errors in ZXDB database for another half a decade. However nobody was revising and reporting errors in ZXSR until now... All things considered, I'm actually impressed about ZXSR accuracy.

and/or duplicated entries

I did my best to automatically remove duplicated information when merging ZXDB and ZXSR (see script "104_zxdb_UTF8.sql" for instance). Now I'm gradually removing the remaining duplicates manually in every new ZXDB update, but it's going to take time...

[Alessandro]

I'm afraid you are looking at the wrong issue

That's because the only reference to Crash Readers Awards in Light Force's entry is:

Crash #38 - 1987/3 35 (PDF) (VIEW) Feature - Crash Readers Awards

so that reference should be amended as well. I see I was misguided by that

Unless there was a new 128K version with music that we don't know about???

To my knowledge, this is the only place the existence of such a version could be ever inferred from. I am inclined to think it was a printing error or composition mistake...

No, all these new remarks came from ZXSR. This information is visible in your screenshot.

I didn't make myself clear, it seems. I meant that probably they were added to the ZXSR database by different people and in different times.

Let's keep in mind that users were reporting errors in Martijn's WoS database for 2 decades, afterwards errors in ZXDB database for another half a decade. However nobody was revising and reporting errors in ZXSR until now...

Maybe because, differently from ZXDB and the WoS database, the ZXSR is much more recent and could take advantage of previously gathered data. The WoS database was compiled mainly from magazine references, and some of these references, especially those about soon to be released titles, have been found to be inaccurate during the following years. There was little or no way 20 years ago to have so many instruments and references at your disposal like those we can take advantage of in this enlightened era

What about my suggestion of adding Lightforce as an alternative title for the game?

[Alessandro]

Zia Rosa should be marked as authored with The Quill, as stated by CASA and confirmed by examination with Unquill.

[Einar Saukas]

I do think it would be good practise to quote sources and have a visible changelog for entries on the database; but I'm not sure that's something that is possible with the system.

We have already adopted the practice of quoting sources (providing links to the corresponding discussion in this forum) whenever some information is corrected in ZXDB. These links are stored in the "auditing" source files that are available for download here. Here's an example from latest file "106_zxdb_UTF8.sql":

Code: Select all

-- https://spectrumcomputing.co.uk/forums/viewtopic.php?p=26872#p26872

-- [druellan] Changed old Label(id=6458, name=J. Eyre, comments=null, country=Country(id=GB, text=UK), country2=null, from=null, owner=null, wasRenamed=false, deceased=null, linkWikipedia=null, linkSite=null, labeltype=Labeltype(id=+, text=Person))
UPDATE labels SET name='John Eyre' WHERE id=6458;

In this example, the forum link was added by [mention]druellan[/mention], everything else was automatically generated by ZX-Ed. This way, if there's any doubt about any change that was made, we can find the exact change in these "auditing" files and see the link to the original discussion.

In theory, it would be even better to have a visible changelog providing a detailed explanation for every change. In practice, this extra work would increase too much the burden on anyone making these changes, I doubt anyone would do it properly. For instance new WoS spent half a decade implementing a new system to keep track of changes, so there's now a field called "Reason" to write a rationale for each change. You can see it here. Take a look and see if you can find any change where this field provides any reason for making any of the changes... The best you will find is a short sentence like "added intro screen" or "updated comments" just mentioned what changed, without any explanations for the reason it changed.

At least our "auditing" files show exactly what changed (for people familiar with SQL) and a link to the proper discussion about the reason for this change. It's not ideal, but it's the best we can do to keep track of all changes without any extra work. At least this information has been good enough for me, whenever I had to look back at previous changes.

At times it feels like we are just being regarded as a nuisance contributing to this thread and that we're not regarded as proper contributors because we're not directly editing the database.

On the contrary, I deeply appreciate your (and everybody else's) work on this! Thank you very much!!!

[Einar Saukas]

There are duplicates all over the place

I have now removed most duplicates within ZXSR. But there are still many cases where similar information was stored in both ZXDB and ZXSR, so these appears twice after the merge. Now I'm gradually solving these duplicates.

Personally I think the risk of providing a relevant information twice is still better than missing it. Thus the reason I added everything first, to later identity and remove duplicates afterwards, instead of not adding anything and then hoping to find out later if we lost anything relevant. Although I can understand that people may find it annoying to see duplicated data for now...

Perhaps the alternative is to display ZXSR comments at SC in a separate section, to keep them apart from everything else?

and lots of opinions

Some of these opinions are relevant (the comment about Lightforce music award is a good example), others perhaps not. I will just need more time to sort them out.

None of this should have been just dumped into the database!

They were not "dumped". Every comment from ZXSR is clearly identified as such. The same way ZXDB also contains conflicting information from SPOT/SPEX that's displayed at SC (for instance here although nobody considered them to be a problem...

This makes me wonder if the actual problem is not really ZXSR data, but the way it's presented mixed with ZXDB comments. I can try to change SC so it's displayed separately, if there are no objections.

[Einar Saukas]

What about my suggestion of adding Lightforce as an alternative title for the game?

Actually I think it should be renamed to "Lightforce".

I see no reason to keep calling it "Light Force", it seems to be spelled "Lightforce" everywhere.

[Einar Saukas]

According to old WoS, Equinox was authored by Raffaele Cecco, Chris Hinsley, and Nick Jones.

According to SPOT/SPEX, it was authored by Chris Hinsley, Dave Perry, and Nick Jones.

According to this Crash review, it was all of them.

[Einar Saukas]

Captain Dynamo is classified as "Spectrum 48K/128K", but there are comments from both old WoS and ZXSR saying "128k only". Which one is correct?

[druellan]

I think most people submitting corrections (I dislike the term "little bugs") include references for their reasons so if you want sources then they are there to be added. I apologies if I've become a little bristly when it comes to your comments. At times it feels like we are just being regarded as a nuisance contributing to this thread and that we're not regarded as proper contributors because we're not directly editing the database.

Oh, please don't think that. I'm personally having a blast doing this, not only for contributing to the database and to perpetuate this information as accurate as possible, but also for the things I learn on each report. If I take my time is because I try to read carefully each report and follow all links provided, this also helps me to better understand the correction and sometimes complement the report with new data. I'm also human, so, mistakes might happen and I appreciate anyone checking now and then to ensure the corrections are there and in shape.

[StooB]

and lots of opinions

Some of these opinions are relevant (the comment about Lightforce music award is a good example), others perhaps not. I will just need more time to sort them out.

They might be relevant, but everyone's got opinions. Why should anyone care what this person thinks?

None of this should have been just dumped into the database!

They were not "dumped". Every comment from ZXSR is clearly identified as such. The same way ZXDB also contains conflicting information from SPOT/SPEX that's displayed at SC (for instance here although nobody considered them to be a problem...

This makes me wonder if the actual problem is not really ZXSR data, but the way it's presented mixed with ZXDB comments. I can try to change SC so it's displayed separately, if there are no objections.

If you put all the "data" in and then have to take out 99% of the cruft then it's "dumping". It just makes a mockery of having the little bugs posts, where we have provide evidence for every little change and wait for up to 2 years to have it implemented when you come along and add hundreds of completely unchecked comments without any scrutiny.

The problem with comments like these, and the SPOT ones, are that they aren't big enough bugs to put the effort into reporting so they just continue to exist. It would have been simpler to just take the 1% of potentially useful content and put it through the existing system where they would be subject to the same quality control as every other change.

[Mort]

Hi Bought this now ,so will get it scanned hopefully this week when it arrives

[PeterJ]

But who sorts through and finds the useful 1%?

I do want an open discussion about what classes as evidence for a change, but that's for another day.

As mentioned numerous times [mention]StooB[/mention], one person looks at these bugs. It's a volunteer effort, no one takes any money. We do it at our pace and have other commitments and interests in our lives. Unfortunately you need to accept that things get done when they get done. That's how it is, and that's how it always will be. We all do the best we can, and constantly criticising isn't helpful. Maybe a thank you might be nice sometimes.

Just remember that if ZXDB hadn't happened you would still be stuck with something not updated since 2013 and a closed database.

If someone wants to pay someone's salary then they can do it full time and wizz through them.

We do of course appreciate the work that goes into reporting bugs.

[StooB]

But who sorts through and finds the useful 1%?

If somebody actually looked at the data they'd see that the comments have a 'section' entry which indicates what the comment is about. We already know what was released on ROM cartridges, compilations and covertapes so all the entries marked as "Compilation", "Covergame" or "ROM Cartridge" can be ignored. If you just want awards then just keep the entries marked "Awards".

It's your site. If you think comments like this (and it's not even on the right entry), are appropriate then fair enough:

Your Sinclair cover game. This had become quite a (in?)famous game on the Spectrum scene. Worth looking at to see what a pile of sh*te it really is!

[PeterJ]

It's my site [mention]StooB[/mention], but not my database. It's just a front end to it. They are two very different things.

I'm proud of all the work Chris has done on ZXSR for the wider community and the work done by Einar to integrate it. We need more like this.

You could be constructive and submit a list to [mention]Einar Saukas[/mention] (or make a pull request on the GitHub page) of ones you feel should be removed which would be a positive contribution to the community who use ZXDB.

You could also use or fork ZXDB and make your own front end (as anyone can). That's the joy of open source. Just make sure you credit ZXDB.

[Stefan]

According to old WoS, Equinox was authored by Raffaele Cecco, Chris Hinsley, and Nick Jones.

According to SPOT/SPEX, it was authored by Chris Hinsley, Dave Perry, and Nick Jones.

According to this Crash review, it was all of them.

The four were Mikro-Gen - http://www.crashonline.org.uk/32/mikrogen.htm

Cecco: definitely
Hinsley: design - https://www.retrogamer.net/profiles/dev ... ele-cecco/
Jones: music - https://zxart.ee/eng/authors/n/nick-jones1/equinox/
Perry: graphics probably - https://www.retrogamer.net/retro_games90/arcade-pool/ - https://www.retrogamer.net/profiles/company/mikro-gen/

✓ Reviewed

[Stefan]

What about my suggestion of adding Lightforce as an alternative title for the game?

Actually I think it should be renamed to "Lightforce".

I see no reason to keep calling it "Light Force", it seems to be spelled "Lightforce" everywhere.

Agreed - I guess the confusion comes from the logo and loading screen which split the title over two lines.

[Stu]

On the subject of Crash Readers' Awards -- Elite are missing a reference to their "Best Software House" win in the 1986 awards (the magazine page mistakenly references the game Elite instead).

✓ Reviewed