Datamining and Ruining Surprises

While I was out on vacation, Gnomecore posted something interesting, discussing the relative merits (or lack thereof, in this case) of datamining for World of Warcraft.

Since Wrath of the Lich King, when I got back into the game after a brief Burning Crusade hiatus, datamining has been a fairly popular thing for fansites to do. MMO-Champion rose to prominence on the back of datamining, and Wowhead claimed the top fansite throne once they got into the game. These days, datamining is simply a fact of life for WoW fans – new content hits a public server, and immediately the content pours forth.

I want to start with how I feel about datamining and then progress forward from that to discuss the effects it has, because I want my viewpoint to be clear as it may color what I say – I generally think datamining itself is okay and eagerly read most of it. Part of my content strategy (how fancy!) for posting the analysis of lore and systems-focused expansions ahead of this piece is to reiterate something I’ve said casually in a few posts – for as much as I talk about the lore of the game, I don’t really care about it that much. When it does well (the Jaina finale in 8.0), I really enjoy it and get invested, but the majority of my interactions with WoW lore just aren’t that major to me. I skip most quest text, I auto-skip cinematics in-game past the first play, and generally, I don’t really concern myself much with the plight of the characters in game. The game makes an effort to make me care, and I definitely do follow lore recaps and attempt to mentally catalog all of it, but it just isn’t what I play the game for. One thing that Classic definitely did better for someone like me is invest the lore effort into worldbuilding rather than a steady plotline, and I think that is why some people do tend to gravitate to that experience – it creates a lively canvas backdrop onto which you can project your character and tales.

So datamining of plot points doesn’t bother me, and I am eager to consume them. For me, as I said in a comment on Gnomecore’s post, the act of reading spoilers is something I can separate from the experience of playing the content that includes that story, and I think it is possible to enjoy both. But I also understand that it is a weird wiring that I have, because in all other forms of media, I generally also don’t mind spoilers. I read the plot of Avengers Endgame nearly 7 months ago and only finally saw the movie on my return flight from Tokyo to Los Angeles last week! Did it mean I had a sense of where things were going? Yes, but it also didn’t tarnish the experience the movie put together for me to enjoy.

What I do find awful, however, is how open the datamining sites (especially Wowhead) are with their spoilers in titles and headlines. Putting a spoiler alert at the beginning of an article titled “Patch 8.4 Spoiler – Arthas’ Ghost and Jaina Proudmoore Have Child in Shadowlands – Broadcast Text” is meaningless and awful. For one, how did a ghost get a human pregnant? Isn’t Jaina more of a Thrall fan these days – again? Where in Shadowlands did they have this weird family? How long does a ghost baby need in the womb? (okay, confession time – this isn’t a real spoiler.)

I think spoilers are a healthy part of any entertainment ecosystem because they mean people are excited to know and engaging with the content. I think that a part of that is that the spoilers need to be kept under wraps and safe from those who don’t want to see them, however.

Plus, selfishly, speculating based on datapoints from the actual game data is fun to write and tend to be my most trafficked posts (save for Blizzcon news and general editorial content!), so that makes my choice a smidge easier.

So let’s discuss the perspective of all three sides, trying to get to neutral viewpoints for each!

The Fansites: Their viewpoint is the easiest to understand – Wowhead and MMO-Champion could just be news sites, like dozens of others. There is a market for the news around WoW, just as presented by Blizzard or witnessed in the live game and PTR. However, what has made Wowhead in particular a juggernaut in fansites has been their datamining and coverage of those spoilers. They have the greatest range of tools and utilities hooked into their datamining, from model viewers that allow you to pre-customize upcoming Allied Race characters and gear, to comprehensive gameplay change listings including pre/post update spell comparisons and the like, and the site’s whole function in the community is to be the nexus of all things spoiled. I’m not privy to what amount of their traffic comes from datamining compared to their database or raw news articles, but I have to imagine it is the majority, given how quickly datamining goes up upon new build deployments. To me, these sites have the clearest-cut cases of anyone in the datamining argument – like it or not, datamining gives them views, which pays bills, which keeps the people at the site employed. Both examples provided can do a much better job hiding spoilers, but overall, I get why they do what they do. As someone who likes spoilers and uses them as content fuel, I also appreciate their presence. Maybe, just, y’know, try to keep giant headlines and front-page photo accompaniments spoiler-free?

Blizzard: Blizzard has, believe it or not, a pretty big stake in datamining too. The implicit relationship they have with the fansites has grown more apparent in recent times, and we now know of another twist.

Blizzard has, for nearly a decade now, established publicly through words that they don’t particularly like datamining, but live with it as a consequence of WoW’s relative popularity and the openness of their testing process via PTR. However, via the 8.2.5 story quests, we also know that Blizzard is capable of hiding data rather easily from dataminers – while many encrypted files (cinematics, voice-over lines, models) don’t fully hide, the story quests and their related broadcast text were hidden until patch day, as players completed the content.

So then the question is – why aren’t they encrypting more lore bits? Already we know the ending of the Ny’alotha raid, a heap of dialogue around the events leading up to and after the raid, and that is on top of all the other things revealed already! The answer is pretty simple, really – Blizzard benefits from the conversation around the game. If you’re the WoW team, in a moment when people are tuning out of the live game and the Classic numbers seem to be plateauing and declining very slightly, the best thing you can have is people talking excitedly about the game. Spoilers in datamining gets that conversation started, and allows people to evaluate a possible return to the game. Is it a good strategy? I don’t really know for sure – I’d be curious if Blizzard has numbers to suggest that it works. I do imagine it makes their lives somewhat easier though – datamining and the comments provoked by it are easy feedback farming, and having the datamining out there also simplifies the process of reveals like the one at Blizzcon last weekend – if enough people already know the ending, then it becomes easy to gloss over.

Is encryption too new of a tech for Blizzard to use more regularly? I don’t really think so. As a technology, some form of encryption has been around for a long time, and with Blizzard specifically, the seeming ease with which they implemented it into the PTR files for 8.2.5 suggests a degree of comfort with it. That leaves me with one fairly reasonable conclusion – Blizzard wants the hype from datamining, and has some unknown boundary at which they will then lock things down.

Players: So what role do we have in datamining? Well, not much of one – we are consumers of an end product, either the datamined content in blog posts on Wowhead or elsewhere, or the finished game. Well, okay, that’s not quite true – despite the “either” framing, I think it is very possible to both consume datamining and the game itself. However, the truth is that we don’t have much of a say one way or another. If Blizzard wants content tested publicly, then it bears some risk of being unable to untangle the content needing playtesting from the accompanying lore content. Even expansion betas often run into this problem, as they won’t show a cinematic but will have a placeholder that tells you a cinematic is due to be inserted, which can be a spoiler in its own right. Square Enix manages to keep FFXIV content secret in one real way – they have internal QA playtest the content, so nothing leaks out ahead of time (bar a few exceptions).

The challenge is that our choice is limited to the only power most of us get to assert in reality – consumer choice. You’re often pushing uphill in a Sisyphean task, however, the weight of those making the opposite choice pushing you down the hill into submission. So long as there are more people clicking than not, Wowhead and MMO-Champion are going to post the spoilers for all to see and smaller fansites, bloggers, and writers will keep writing about them. I don’t say that to be defeatist about it – but in a capitalist society, that is the choice you have, limited though it is. You can try to convince others to join in, but that has limited possibility as well unless you are a convincing orator or there is an inciting event that really pushes people forward. I mean, I like datamining just fine and read it all, but if they posted the correct ending of Shadowlands tomorrow before a release date even exists, I’d probably be irritated enough to join in and never visit Wowhead again.

There is also a value of sorts to be had for players from datamining. I know a lot of lapsed fans of the game who often gauge their interest in subscribing or returning by the things they read from datamining, or use that content to keep a rough idea of the current plotlines in the game so they can simply return with each new expansion and play that story live. Is that a way to consume content? Sure – I used to think it was weird, but then again, I also “keep up” with WWE’s wrestling product primarily this way, since watching the whole thing is dreadful. If I felt the same about WoW, I’d probably be perfectly fine with datamining spoilers as my sole engagement with the game. It certainly isn’t the same as playing, but to some people, that is enough – to have a rough lay of the land such that you could drop back in at a moment’s notice. When I last left the game for a reasonable length of time, in 2013, I was still sort of aware of what was going on. I followed some discussion of the patch 5.3 content and watched the release trailer for 5.4, which led to me resubscribing. If I didn’t have a sense of what was going on, would I have resubscribed? Maybe – it was probably more my friends than anything else that led to that resub, but at the same time, logging back in after 6-ish months away can be daunting and push someone away from trying.

Overall, I suppose it is a tough topic – I do think datamining sites have a need to try and keep spoilers from being exposed to those not wanting them, but at the same time, I do think that datamining has some value to fans and to Blizzard, and until something dislodges one side of that argument or the other, we’re going to be seeing spoilers from datamining as a trend into the future. While I personally like them, I do empathize with those that do not, and hopefully sites and blogs (mine included!) keep that front of mind when creating content that reaches an audience comprised of both camps.

2 thoughts on “Datamining and Ruining Surprises

  1. I always find this topic interesting since other games show that a different approach is absolutely possible: SWTOR for example does use a PTS but always cuts out the story content. If Bioware can do it, then Blizzard could definitely do so as well – if they were so inclined.

    It simply comes down to the attitude of both devs and players though. Datamining, spoilers and the people who consume them do exist for SWTOR as well, but it’s very much a minority pursuit as most players prefer to be surprised by story developments as they play through them, so even on Reddit mods will come down hard on you if you post any spoilers (even for old content) without appropriate tagging.

    The problem for WoW lore enthusiasts is that they are very much in the minority. One thing you didn’t mention is that spoilers can also turn players away – I make no claim to represent a large number or anything, but being exposed to spoilers was why I didn’t come back to retail for WoD despite of initially being quite excited about it. The more you spoil, the more players who are mainly there for the story are driven away, the less the community cares about avoiding spoilers – it seems a vicious circle to me.

    Also, I gotta disagree with spoilers being what made Wowhead big… mainly because I remember its rise to prominence happening during BC, caused by having a much cleaner interface than the competition at the time, and a database that updated through a lot of automation instead of requiring tedious manual updates for everything. I’m not saying they don’t get traffic from their datamining posts, but I’ll be very surprised if that isn’t a tiny percentage compared to people simply using their database. I know I use Classic Wowhead ten times a day to look things up but never look at their news unless someone links me to a specific post.

    Liked by 1 person

    1. Those are great points – and a good SWTOR perspective I don’t have!

      It’s weird, because I am the opposite about FFXIV – the newest patch launched while I was overseas and I went to great lengths to ensure I didn’t see story spoilers. Likewise, for some forms of entertainment, I will do the same – but something has to really have my imagination to get that treatment, and I think WoW just doesn’t have it in that way.

      Also a great point on Wowhead – to be honest, I blocked it out of my head how bad sites like Thottbot and Allakhazam were back in the day, and didn’t really use Wowhead as a database until mid-Wrath. It is still a cleaner database, although their ad scourge is something of a problem if you are sans adblocker.

      The point on being turned off by the lore reveals is an interesting one I hadn’t considered, but equally valid as a point against the idea of datamining. I think Blizzard has decided that any coverage is good coverage and they’ve allowed datamining to continue on despite them being able to encrypt content or even do something more like Bioware with SWTOR or Square Enix with FFXIV.

      I do wonder how Wowhead’s traffic breaks down – I haven’t needed to use the database as much since Blizzard built the functionality of quest tracking and objective locations into the game, but I suppose when I interact with the more granular details of their datamining, that is in the database side of things. Given the paid role I tried to secure there earlier this year, my suspicion is that their guides get enough traffic to be worthwhile, because that is what they wanted to pay for, although it may just be the case that they maintain the database as-is and don’t need new staff for it.

      Either way, with the SWTOR example, I do think there is sufficient evidence there to state more clearly that Blizzard sees some benefit in datamining, enough to outweigh the negatives, but I do find myself curious how much data they really have on that topic.

      Liked by 1 person

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.