<Snipped quote by FinalPalladium>

I want to point out that RPG has existed a lot longer than since 2014. In early 2014 the website was entirely deleted and by the time Mahz rebuilt it on a new server it lost a massive amount of users and activity. It was actually originally made in 2007, I myself having joined it in 2008. I don't know how easy it would be to get the data from before the Guildfall, or if its even possible, but going off my memory alone the activity RPG gained once it was back up was a shadow of its former self. Don't know how I would compare it to the activity data you presented, but I'd reckon that "usual" RPG activity was probably comparable or greater than the 2017 spike you observed.

It's already been mentioned about the bot invasion that year, but also worth noting is that in 2017 that Mahz returned from a long break and continued working on the site as well as the creation of the site's Discord. Both of these could've contributed to the elevated levels of activity, as the bot invasion wouldn't explain why there wasn't an immediate drop in activity after it was resolved. I think that year a few rival sites changed their policies about accepting new users to make it "invite only" and that may have contributed as well.

It's definitely partially this, although I can't really think of a way to back this up with data. I've been here since long before guildfall, and the site was more active before guildfall than after. So, I think calling this pace 'the normal pace' is a little deceiving. RPG used to be way more active than right after it was recreated. Big reason for that was probably that a lot of people had no idea what happened and no idea when it would be back up. If at all. It'd not be super surprising if a lot of people just straight up never came back after Guildfall.

Interesting as this data is, you're basically always going to be working on 50% of the "real" data, tainting any real semblance of a conclusion. A counter argument is always going to be that you're missing information -- and while I think it's cool we have this info, using it to disarm arguments about the site's activity levels is.. not entirely fair and constructive, and I think wholly incorrect. If 2017 is the standard we use to measure activity, then any drop is a drop. If 2017 was when the site functioned at the best it could (lots of people, lots of RP's to choose from) then surely we should aspire to be that again, and not just some shitty baseline forum where it's a group of 100 people writing about the same thing over and over. That group is gonna die eventually, and the intake of new members isn't enough to sustain RPG.

Other than that, I also have another issue with the data as an argument, RP's created over time is not really a suitable metric to measure activity by at all. Like, if I create 100 topics in a day (and suppose I wouldn't get banned for this), is that truly how we measure activity?

Either way the data is still interesting and it's fun to see people engage with the site, but I'm not sure that this outcome is really as meaningful as I'd like it to be in a sense of explaining why things are the way they are now.

The way things are now and how they got to be that way is a subject which never ceases to entertain.

To be frank, I find the distribution of tags more interesting than the amount of posts or activity over time. There's perhaps minute differences you could find in the latter between each subforum whilst the former speaks to the overall enviroment of each subforum, trends and so forth.

@FinalPalladium I hope you don't mind me dropping this into your thread but out of curiosity, I bootlegged your code and replaced all of the necessary variables relevant for the Advanced Section to see if there were any statistical differences. Props to you. I recommend everyone take a look at their code if they are more curious about other aspects of the site in general. Really adaptable and easy to modify.

Few things to note. I am a dabbler in code and thus, my IDE was different to what the OP was using. I had to use html.parser instead of lmxl parser due to difficulties with my software. I also had difficulties with the matlibplot module that the OP used to visualize their data. Therefore, I used the painful process of manually shoving it all into Excel. Could have done a R Plot instead but eh. Anyway, here is the result.



There are a few differences and similarities of note. To begin.....

- We see a parallel scenario with Casual/Advanced Tags in the Advanced Section. Casual tags are commonplace whilst Advanced Tags are barely non-existent whilst the opposite exists in the data for the Casual section. This may imply that GMs use these tags as a way of quantifying themselves relative to the rest of the section.
- More even split of proportions between small group and large group in Advanced Section relative to Casual.
- Sci-Fi more prominent than Anime/Manga for Advanced.
- Top 4 genres are Fantasy, Sci-Fi, Fandom and Modern.

Might apply this code for Free if OP doesn't. Anyway, thought I do this and I was both mildly surprised and expectant of the results. Sorry if OP was in the process of doing one for Advanced.

Feel free to use the code as you wish. I'm glad that someone thought that a simple project like that was worth anything.

Also, it's interesting to see the conclusions you took from the advanced forum, however, the first one (marked in red) is based on the same lack of context I had when I first wrote this.

To understand what's truly happening with the section tags just refer to @stone's post (quoted below). They found out what was truly happening.

<Snipped quote by FinalPalladium>

Yeah, I also took a look at your code (it checks out). Since it (seemingly) had no problems with the other tags, the Casual tag not showing up is definitely a Guild thing, not your program. Cheers!

Thanks again for the appreciation. I'm glad people are still having fun with this idea.