Thrills Over Time

[Funt Solo]

Yeah - it's mostly just borne of curiosity about how I could build something bullet proof to tackle the data - and I appreciate it is overly complex.

The supertitle/title/subtitle maze I got into was trying to tackle various odd situations...

In the Karyn example I posted up, there are four separate titles for the strip (even when she's not guest-starring in another title):
- Karyn, Psi, Division
- Karyn
- Karyn, Psi
- Visions

Or, with Abelard Snazz, there's actually no strip with that title (because it's in a Robo-Tale, or a Future Shock, or it's called "Genius is Pain: An Abelard Snazz Misadventure"). Although that's not true now since prog 2206, but you get the idea.

So, I thought of the concept of a supertitle (that, in most cases, is identical to the title, so we probably don't need to do that).

And then Trifecta...

[I, Cosh]

The supertitle/title/subtitle maze I got into was trying to tackle various odd situations...

In the Karyn example I posted up, there are four separate titles for the strip (even when she's not guest-starring in another title):
- Karyn, Psi, Division
- Karyn
- Karyn, Psi
- Visions

Didn't realise that about those Karyn stories (I've flicked through that era of the Meg once at best) but this is a good example.

Obviously, we can safely ignore the ludicrous idea that a bunch of stories published around the same time featuring the same character are intended to be considered as one group irrespective of whether artists stuck in different logos or editors didn't bother about standardising the strip name on the front page. With that out of the way we can get on with trying to solve the problem!

Now I see what the idea of the Supertitle is but, to me, that's introducing an imaginary element you don't really need. Like you say, in most cases it is the same as the strip title.

A couple of thoughts on how you could handle this.

[tbLinkedThrills]. A simple mapping table where you record the main strip (lets assume thats the first appearance) and any others which are a continuation. Pretty simple and you can easily query in either direction (or both.)  For Karyn, it looks like this would solve the problem once you combine it with your [featuring] table.

Starts to get a bit more complicated with spin-offs and split appearance though and you might also need a [tblLinkedStories] for certain situations. Is Pussyfoot 5 really a Dredd spinoff?

[tblThrillCharacters]. Mapping table which links each strip with its central characters (various other options for relationship type, e.g. main, supporting, villain) who we can then assume are in the majority of stories and thus don't need to be explicitly in [featured]. This way you add [Character].[Karyn] for each of those Thrills and union with [featuring].[Karyn] to get all relevant stories.

This one has the benefit of allowing individual characters to spin out of one story into another. For example the structure below would be the easiest way to let you fully track Dirty Frank without confusing it with other Lowlife characters. You can easily see how the same thing will work for Shimura/Inaba/Hondo City Justice/etc.

tblThrillCharacter
LOWLIFE.NIXON
LOWLIFE.FRANK

HERSHEY.HERSHEY
HERSHEY.FRANK

featuring
TITAN.NIXON
TITAN.GERHARDT
SMALLHOUSE.FRANK
SMALLHOUSE.GERHARDT

Barney has a separate category for one-off thrills, many of which later spun out into their own series with the first installment marooned there. Not sure if that story Visions fits in there by your definition but not every story will have or need its own strip title either so there needs to be a way to handle that.

Bonus points for figuring out a way to handle a group of characters like The Dark Judges.

[Funt Solo]

Obviously, we can safely ignore the ludicrous idea that a bunch of stories published around the same time featuring the same character are intended to be considered as one group irrespective of whether artists stuck in different logos or editors didn't bother about standardising the strip name on the front page. With that out of the way we can get on with trying to solve the problem!

*applause*

[Now I'm going to read the rest of your post.]

[Bolt-01]

I know this is merely supposition, but if you were to try and actually populate this RDB the sheer amount of manual data entry would be staggering.

[Funt Solo]

I know this is merely supposition, but if you were to try and actually populate this RDB the sheer amount of manual data entry would be staggering.

Yes. Although, if (the purely theoretical) we were smart about it, we might be able to trawl Barney OR trawl my magic spreadsheet to automate part of that process.

But, yeah - there are approximately 8000 things (stories, text stories, posters, articles etc.) to pop in there, each of which has lots of data associated with it (my table has 12 columns) - so about 96,000 things to add.

One of the many reasons that I often find myself stopping at the start. The design would have to be shiny-perfection in order to move forward.

[TordelBack]

This thread is by far the best porn on the internet. Have you lads thought about starting an OnlyFans? You could give up the day jobs and just talk tuples all day long for an audience of thirsty nerds...

[Funt Solo]

Is Pussyfoot 5 really a Dredd spinoff?

Yes! Sort of.

I have it listed as "Dreddverse: Devlin Waugh".

[Not really answering any database conundrums with this information, I realize. But I am distracting myself with a Dreddverse Map of Linked Thrills that's been in the works for a while.]

[sintec]

FWIW I've also thought about this quite a bit. Off and on I've been messing around trying to create a little application to create my own bespoke digital GNs using a combination of the cbz files I've bought, Barney and an index of content.

Say I want to generate my own collection of every Sinister Dexter story since they arrive in Generica. I have all the files. I can get the relevant prog numbers from Barney. I can use that list to filter my files. The biggest problem I have is getting the right pages from those files into my new one. (The second biggest is handling correct page orientation in case there are splashes.) Keeps me occupied.

I considered this over the christmas break (having not written any code for a whole week my brain clearly decided it needed to think about code again). My data doesn't have page numbers in though and adding them felt like a pretty mammoth data entry task.

Yes. Although, if (the purely theoretical) we were smart about it, we might be able to trawl Barney OR trawl my magic spreadsheet to automate part of that process.

This is basically my process for building the graphs I've stuck in the Ultimate Collection thread. I've got a bunch of different spreadsheets containing different data sets and some scripts that trawl them and spit out new csvs with the data munged together. It'd be simple enough to have something like that call APIs on a database to populate it instead of writing to a csv. Happy to help with that kind of thing if people want to take this further.

One of the many reasons that I often find myself stopping at the start. The design would have to be shiny-perfection in order to move forward.

"Perfection is the enemy of progress" as my boss is often heard to quote as we all get bogged down in overdesigning a thing he wants finished next week.

[Funt Solo]

I've been playing with a uniform format for publishing my data online. Is this nice / useful?

[Colin YNWA]

Looks shiny - I like it.

Also reminds me we've had two year gaps between stories before, so I just need to be more patient!

[sintec]

If you're looking a uniform format then you might want to include a Writer column for stories where authorship hasn't been consistent (Durham Red, Indigo Prime, VCs as a few examples).  Your heading would then be the original creator credits for the series.

[Lorenzo]

It pleases and sparkles - I'd love to see a resource as clear and as useful as this.

[AlexF]

Yes, a thing of great clarity and usefulness!

[Funt Solo]

If you're looking a uniform format then you might want to include a Writer column for stories where authorship hasn't been consistent (Durham Red, Indigo Prime, VCs as a few examples).  Your heading would then be the original creator credits for the series.

Thanks, sintec - I hadn't thought of original creator credits - I'll do that.

Currently running some test on compressing some of the column headers to make room for wider tables, and also getting some hover-text to show up for who did the covers. (Trying to get an implementation that works on touch screens as well as mouse-y things.)

[sheridan]

I know this is merely supposition, but if you were to try and actually populate this RDB the sheer amount of manual data entry would be staggering.

Yes. Although, if (the purely theoretical) we were smart about it, we might be able to trawl Barney OR trawl my magic spreadsheet to automate part of that process.

But, yeah - there are approximately 8000 things (stories, text stories, posters, articles etc.) to pop in there, each of which has lots of data associated with it (my table has 12 columns) - so about 96,000 things to add.

One of the many reasons that I often find myself stopping at the start. The design would have to be shiny-perfection in order to move forward.

Regarding the data entry task - you could crowdsource it.  Say export to a shared online spreadsheet (google sheets, for example) get some trusted people with the requisite progs and megs (us lot) to tackle a series or year each, then import it all back and hope that transfers haven't introduced any formatting issues along the way.