Don’t ask questions if you don’t want to know the answer.
The other day I posted Back to the Drawing Board, describing changing how I create prerequisite diagrams for the Echelon Reference Series. I think I’ll be able to do a lot of the basic stuff through scripts, leaving only the harder (but much more interesting) ones for manual intervention.
I could even have distinct diagrams for different data sets. That is, the PRD-Only books could have different diagrams than the 3pp+PRD books, which have different diagrams than the PRD+PZO books (I’m working on them!) and the Everything-I’ve-Got books (ditto… both the last two are waiting on cleansing the Product Identity from PZO sources).
For giggles, I decided to count how many diagrams this could mean. I kind of wish I hadn’t looked.
| Data Set | PRD | PZO | 3pp | EGD | TOTAL |
|---|---|---|---|---|---|
| PRD-Only | 3,589 | 102 | 3,691 | ||
| PRD+PZO | 4,070 | 3,036 | 139 | 7,246 | |
| 3pp+PRD | 4,101 | 8,456 | 237 | 12,798 | |
| Everything | 4,506 | 3,053 | 8,459 | 257 | 16,275 |
| TOTAL | 16,266 | 6,089 | 16,915 | 735 | 40,005 |
This is only a potential volume of diagrams to create, and assumes that every single object that either has prerequisites or is a prerequisite of another object gets a distinct diagram.
The reason the different data sets have different counts is because as more source sets are included, more otherwise unlinked objects get connected. For instance, in the PRD-Only set I have 3,589 potential diagrams for PRD-sourced objects, but in the PRD+PZO set there are 4,070 potential diagrams for PRD-source objects because 481 more are prerequisites for objects in the PZO source set… and the PZO source set could have many of the 3,589 PRD-Only objects as prerequisites.
Similar logic follows when adding the 3pp source set (512 PRD source objects are now prerequisites that previously weren’t). At first I was a bit puzzled that the ‘Everything’ set says I’ve got 4,506 potential diagrams in the PRD source set, until realized it means there are about 400 that are used by objects in only one of the PZO and 3pp source sets.
The ‘EGD’ source set is a bit of an oddity, in that the objects in that source set sort of don’t exist. They are all ‘stub’ objects created to give child objects (such as class features) something a prerequisite could anchor to. Consider, many things might have the evasion class feature as a prerequisite. Because so many (and it only takes two, but there are many more than that) classes and archetypes have the evasion class feature, ‘evasion class feature’ cannot be uniquely resolved. In building the index files, any time I found a situation like this I added an empty object such as ‘evasion class feature’ (no parent class or archetype) for searches to latch onto.
… 40,000 potential diagrams. Many can fall away if they would be only a single node (I rarely show nodes for things like skill ranks or class levels or ability scores), many others are likely to be trivial (two-node diagrams), but it’s still a very large number.
I can reduce the number by simply doing one per object (based on the Everything data set) and let the reader see all the things that aren’t in the data set for the book they’re reading. This doesn’t feel good to me, but it would reduce the total potential volume by quite close to 60%.
I think I’ll see where automation takes me. Automation is good. Though even if it solves 90% of the load it still leaves me with more than 4,000 diagrams to fix manually.
Why do I get myself into these situations?
Update
I’d forgotten magic item requirements use the same code as selectable options (with a different field name).
| Data Set | PRD | PZO | 3pp | EGD | TOTAL |
|---|---|---|---|---|---|
| PRD-Only | 2,574 | 102 | 2,680 | ||
| PRD+PZO | 2,998 | 2,484 | 137 | 5,625 | |
| 3pp+PRD | 3,086 | 8,051 | 237 | 11,378 | |
| Everything | 3,434 | 2,501 | 8,054 | 257 | 14,250 |
| TOTAL | 12,092 | 4,985 | 16,105 | 733 | 33,933 |
If I subtract these new numbers from the original set above, it’s looking somewhat more favorable.
| Data Set | PRD | PZO | 3pp | EGD | TOTAL |
|---|---|---|---|---|---|
| PRD-Only | 1,015 | 0 | 1,011 | ||
| PRD+PZO | 1,072 | 552 | 2 | 1,621 | |
| 3pp+PRD | 1,015 | 405 | 0 | 1,420 | |
| Everything | 1,072 | 552 | 405 | 0 | 2,025 |
| TOTAL | 4,174 | 1,104 | 810 | 2 | 6,072 |
| % Change | 26% | 18% | 5% | 0% | 15% |
Still quite a lot, but significantly fewer.