I just spent 2 hours debating and testing and arguing in circles and bitching about library catalogs with two colleagues and I just want to say
AO3’s website is really, really, really impressive, functional and ergonomic and cohesive. the tag system is INCREDIBLE and AMAZINGLY maintained. this is my professional librarian appraisal.
I’ve found 1 library catalog that meets my standards. even the national library of France’s catalog is shitty in comparison to ao3.
praise.
It’s awesome! As a total ignorant, can I ask what AO3 does and library catalogs don’t?
i might actually type out a longer answer but what it really boils down to is: YOU ACTUALLY FIND WHAT YOU’RE LOOKING FOR
ok so here’s the long unreadable (and probs uninteresting to anyone else than me) version:
– the site design and overall look. it’s easy to read, easy to navigate, and easy to notice what you can click on. Makes good use of fonts and text sizes and styles to make important things stand out and be easily found at a glance, and is just overall very readable. The icons with hovertext! The tags! the amount of info that’s readable at a single glance and actually fits on the same page!
this is BASIC STUFF and it is not a given on a LOT of professional library websites i run into regularly and that drives me INSANE. (Mostly bc one of the very popular, cheap, and easy French-language library catalog softwares has a default online catalog design that sucks and which librarians generally don’t tinker with much.)
– again this seems obvious, but the filters when you’re inside a fandom/tag are SO VISIBLE and SO EXPLICIT. The filters menu makes it instantly clear what it’s for, is easy to navigate and understand and use, intelligently suggests the most popular tags first (which also immediately gives you a lot of information).
My library’s online catalog (which uses the default website set-up I mentioned above) has exactly the same thing, but stupidly executed, unreadable and incomprehensible, and somehow completely unnoticeable despite being exactly in the same place on the page. The site design makes very bad use of the space on the page and basically you just don’t even look over there because it’s so far away from where the rest of the information is and it simply never catches your eye, and even when it does, the vocabulary used is so obtuse you don’t realize what it’s for.
IT’S SO… STUPID AND EASILY FIXABLE… but apparently no public library in the french language can afford a website designer, or they’re all horrifyingly bad
– and finally: THE TAGS. One of the biggest issues we have in catalogs is that people use different words for the same thing. In order for you to find books relevant to your search, we have to apply topic keywords to them (basically: tags), but of course there are Norms so that all libraries, or at least all employees in the same library, use the same keywords. Except despite the norm that still doesn’t happen. I don’t know how it goes in the English-language world but for French language it’s all horribly complicated and surprisingly non-functional, despite how easy it seems in theory, and leads me to complain about the Bibliothèque Nationale de France about once a week at least.
Easy example that I’ve complained about today (for the 6th time this year): ADHD. The term used by the BNF, that we are supposed to use, is “Trouble de l’hyperactivité avec déficit de l’attention” (“hyperactivity disorder with attention deficit”). That’s… not only outdated but flat-out inaccurate (according to French’s current stance on it) — the term people actually use nowadays is the opposite way around, “trouble de l’attention avec ou sans hyperactivité” ( “ADD with or without hyperactivity”), commonly abbreviated to “TDA/H”. The BNF’s system does accommodate for various synonyms, but it appears unaware of this one, so if you search “TDA/H” in the keywords, you won’t find anything. You’d have to look in the title, and if none of our books have it in their title, you’ll find nothing at all, and won’t even be redirected anywhere if we strictly follow the BNF system. (WHAT IS THE POINT OF KEYWORDS THEN, one might ask.)
Tl;dr: you look for the word you and most people actually informed about a topic use, and find nothing at all because some rando has decided that’s not the word you should be using. (Unsurprisingly, this problem pops up a looot for keywords related to minorities, mental illnesses and LGBT+ topics.)
It’s like if you tried to search a site for “fluff” and didn’t find anything because the site has decided to continue using “WAFF” instead. Also, the site has decided that hurt-comfort and guro fic are the same thing, makes no distinction between levels of romance and eroticism so there’s no way to tell cute handholding from smut, and believes that the word “furry” means they get a dog.
=> The system of letting people use their words and linking them — making them synonyms — with what other people have used for the same meaning completely blows my mind. I am in awe of the fact that it works, and that it’s still happening, even though iirc tag-wranglers are unpaid volunteers. I couldn’t imagine doing something like that in just our catalog, and AO3 is massive.
The result is: not only do you find what you’re looking for, but if your search accidentally picks up other things too, you know what it’s actually about because you get it in the author’s words.
AO3′s tag system is an incredibly clever and simple solution to a very real and thorny problem that I run into almost every day.
tl;dr AO3 is just generally a perfectly functional and user-friendly site, instantly easy to use in order to tailor your search to exactly what you want (and even more so with the addition of the exclusion operator to the filters sidebar), and on a technical library-science viewpoint, it’s fascinating.
This is taking me back to when AO3 was first born, and I was having a conversation with someone (@icarusancalion, I think?) about how I didn’t think the tagging system was ever really gonna be useful.
I knew the kind of top-down tagging system that libraries use was often useless for the same reasons you’re describing here: academics like the idea of a priori systems and exclusive classification schemata, but AO3 tagging is useful precisely because tags can be messy and overlapping rather than strict hierarchies. You’ll never get all fandoms everywhere to agree on a common tag family, I said c. 2008. It’ll be outdated before it’s even implemented. But relying entirely on user-generated tags will be a logistical nightmare, past!Maud also argued, because there would be no way to manage synonyms and near-synonyms and typos that would rapidly bloat the system to uselessness.
Well, 2008!me was right about top-down schemata but wrong about user-submitted tags, thanks almost entirely to the work of the tag wranglers: human curators who take the time to link and nest related tags as they come up, without relying on a pristine (and utterly dysfunctional) a priori system to do so.
Would real-world academic libraries benefit from tag wranglers? Absofuckinglutely, but I really don’t think most of them would ever implement them for the same reason past!me was skeptical of them. Maybe if they were shown how well it works on AO3 (where the wranglers are all volunteers!) they might be persuaded to hire some workstudies or under-employed PhDs to wrangle for them. And then the world would be a better place.
I have given talks about our “curated folksonomy,” which is what it’s called, to librarians and archivists! And @cfiesler has done great work re: tags and such!
Can confirm that AO3 is fascinating from a library perspective, and that librarians think it’s rad. I gave the keynote (talking about AO3) at a library conference on open repositories last summer and got such amazing reception!
My work about this: http://cmci.colorado.edu/~cafi5706/CHI2016_AO3_Fiesler.pdf
But also, an academic friend did her dissertation in library science on tag wrangling in particular. Here’s one piece of that; I can’t immediately find an open access copy but I’m sure if anyone wants to read her work they can contact her and she’d be happy to share: https://dl.acm.org/citation.cfm?id=2957295
Tag: fandom
HEY, ROY/ED FANDOM
This is a long post, but please bear with me! ♥ And PLEASE REBLOG to help to spread the word! ♥♥♥
By now, probably all of you have heard about the fact that this garbagefire of a website has finally started to collapse. In a lot of ways, I think it’s a good thing, but Tumblr had two things really had going for it:
- it was one platform that different kinds of contributors could use for different kinds of content;
- the vast majority of the fandom population used it as a home base.
This brings me to my primary fear: our ship’s amazing community may get scattered to the winds if everybody jumps ship in different directions.
This brings me to my suggested solution: I’m going to make a spreadsheet of alternate usernames and sites and accounts and destinations for members of the Roy/Ed fandom. This way, we can at least all see where our longtime fandom friends are sharing their work!
HOW DO YOU GET IN ON THIS? Easy peasy lemon thing!
Just copy and paste and then fill out the form below, with as few or as many accounts as you feel comfortable sharing. Message it to me here, or on Twitter, or wherever you feel like, and I’ll add you to a Google Spreadsheet that will be publicly accessible, to help the Roy/Ed fandom to find each other on other platforms.
You, posting your work to AO3: These tags are excellent, they are super-clear about exactly what I mean, and no one will ever be confused about my intent.
Me, a Tag Wrangler receiving your tags through the wrangulator: ACTUally-
The Tumblr tags specify using characters’ full names.
This is honestly never something I considered before, but I’m guessing that – a lot of the time – tag wranglers probably see tags to organise/sort independent of the work itself, so it could prove confusing to see say “Human Peter” or “Mutant Keith” … the names are so common that it could belong to dozens of fandoms, with no guarantee the wrangler is a part of those fandoms either.
When writing character tags, it’s worth bearing in mind.
Not only will it make things easier for volunteers, but it’ll likely have the added benefit of increasing your audience, too, as people will have a more specific tag to search and use in place of one that could apply to several characters. After all, nothing more irritating than searching for x and finding y!
This is correct, except it’s not “a lot of the time” – it’s every time! We have to make an extra effort to go to an individual work to see the tag in context. Otherwise, all tags arrive to us without context. And if we’re wrangling hundreds of tags a day, we’re just never going to have time to check the works on all tags.
Even worse, “Human Peter” can (and has) been used by multiple people to mean different Peters. We can’t edit tags, so we can’t fix that. The tag will simply become unuseable. Human Peter Hale or Human Peter Quill, those are wrangleable tags! Human Peter, not so much.
This post explains the Peter problem in detail. (This one is funnier and shows the problem visually.) This post shows how the wrangulator works on the back end. And this post suggests how to make life easier for tag wranglers in general.
If anyone has questions about how tags work or needs help with deciding how to tag, my asks are open!
no but how much audacity and sheer entitlement do you have to have to tell people they need to stop posting their darkfic and porn fic and any other fic you don’t like to ao3 so you can have a safe space when ao3 was literally created as a safe space for writers to post their content without fear of it being randomly wiped out by pro-censorship assholes with an agenda like what has happened to plenty of other fic archives before?
“but a lot of us see ao3 as a safe space to get away from that kind of nasty content” – lol you can see the middle of a busy interstate as a safe space all you want too but that doesn’t mean that you get to walk into the road and scream at all the cars going by that they’re the ones infringing on your safe space either
ao3 is not, has never been, and will never be a site meant for nothing but children’s stories. you can “see it” like that as much as you want but there’s a difference between fiction and reality and that view of what ao3 is like is as fictional as the stories posted on it.
This is why AO3 has tags! AND FUCKING CONTENT WARNINGS. you don’t want to read it? Or see it? You don’t have to! Use the filters, that’s what it’s for
Yahoo reports big loss, writes down Tumblr value
I FUCKING KNEW IT.
SO. IF YOU KNOW YOUR FANDOM HISTORY, YOU CAN SEE THE WRITING ON THE WALL RIGHT NOW.
AND IN CASE YOU DON’T, I will tell you a story.
I don’t know if Yahoo as a corporate entity hates fandom, or if it LOVES fandom in the way a flame longs to wrap its embrace around a forest. Or maybe it’s just that fandom is an enticingly big and active userbase; but just by the nature of our enterprise, we are extremely difficult to monetize.
It doesn’t matter.
Once upon a time – in the era before anyone had heard of google – if you wanted to post fandom (or really, ANY) content, you made your own webpage out of nested frames and midi files. And you hosted it on GeoCities.
GeoCities was free and… there. If the internet of today is facebook and tumblr and twitter, the internet of the late 90s WAS GeoCities.
And then Yahoo bought GeoCities for way too much money and immediately made some, let’s say, User Outreach Errors. And anyway, the internet was getting more varied all the time, fandom mostly moved on – it wasn’t painful. GeoCities was free hosting, not a community space – but the 90s/early 00s internet was still there, preserved as if in amber, at GeoCities.com.
Until 2009, when Yahoo killed it. 15 years of early-internet history – a monument to humanity’s masses first testing the potential of the internet, and realizing they could build anything they wanted… And what they wanted to build was shines to Angel from BtVS with 20 pages of pictures that were too big to wait for on a 56k modem, interspersed with MS Word clipart and paragraphs of REALLY BIG flashing fushia letters that scrolled L to R across the page. And also your cursor would become a different MS Word clipart, with sparkles.
(So basically nothing has changed, except you don’t have to personally hardcode every entry in your tumblr anymore. Progress!)
And it was all wiped out, just like that. Gone. (except on the wayback machine, an important project, but they didn’t get everything) The weight of that loss still hurts. The sheer magnitude…
Imagine a library stocked with hundreds of thousands of personal journals, letters, family photographs, eulogies, novels, etc. dated from a revolutionary period in history, and each one its only copy. And then one day, its librarians become tired of maintaining it, so they set the library and all its contents on fire.
And watch as the flames take everything.
Brush the ash from their hands.
Walk away.
Once upon a time – in the era after everyone had heard of google, but still mostly believed them about “Don’t be evil” – fandom had a pretty great collective memory. If someone posted a good fic, or meta, or art, or conversation relevant to your interests? Anywhere? (This was before the AO3, after all.) You could know p much as soon – or as many years late – as you wanted to.
Because there was a tagging site – del.icio.us – that fandom-as-a-whole used; it was simple, functional, free, and there. Yahoo bought it in 2005. Yahoo announced they were closing it in 2010.
They ended up selling it instead, but not all the data went with it – many users didn’t opt to the migration. And even then, the new version was busted. Basically unusable for fannish searching or tagging purposes. This is the lure and the danger of centralization, I guess.
It is like fandom suffered – collectively – a brain injury. Memories are irrevocably lost, or else they are not retrievable without struggle. New ones aren’t getting formed. There is no consensus replacement.
We have never yet recovered.
Once upon a time… Yahoo bought tumblr.
I don’t know how you celebrated the event, but I spent it backing up as much as I could, because Yahoo’s hobby is collecting the platforms that fandom relies on and destroying them.
I do not think Yahoo is “bad” – I am criticizing them on their own site, after all, and I don’t expect any retribution. I genuinely hope they sort out their difficulties.
But they are, historically, bad for US.
And right now is a good time to look at what you’ve accumulated during your career on this platform, and start deciding what you want to pack and what can be left behind to become ruins. And ash.
…On a cheerier note, wherever we settle next will probably be much better! This was never a good place to build a city.
Other important fandom history events:
Fanfiction dot net? Used to host all ratings of fanfic. Then some conservative anti-pr0n organization was like, “No, you can’t do that.” And FF.net was like, “You’re right, we can’t do that.” And anything explicit was taken down with extreme prejudice. Fans created Adult Fanfiction dot net, but it never really reached the same level of popularity as the original FF.net.
I’m not entirely sure on the details, but I’m pretty sure fandom central moved more-or-less to LiveJournal at that point, or possibly even slightly before. Eventually, LJ was sold to a Russian company, Sixapart (IIRC). Then came the ads, and Strikethrough, and Boldthrough, and finally a massive change to interface that caused a lot of users migraines, and fandom collectively packed up and moved again.
Some people stayed; it was their internet home, and I certainly don’t blame them for staying, but LJ was never the same after. Fandom sort-of splintered over the course of these events, with the RP related splinter largely moving to Dreamwidth (a site very similar to LJ, but started and run by fandom in much the same way AO3 is), and other groups moving to Tumblr or Plurk.
Dreamwidth isn’t nearly as active as current Tumblr, but I trust the hosts and staff not to screw fandom over, which is more than I can say about almost any site besides AO3. Dreamwidth is run entirely on user-purchases, which has almost certainly been impacted by the fact that PayPal won’t work with them – because Dreamwidth refuses to censor explicit content. That’s right: Unlike LJ before it, and quite probably FF.net as well, Dreamwidth decided it valued its users more than a possible bump in revenue.
Dreamwidth refuses to censor explicit content, and its format allows much more interaction, discussion, comment threads, communities.So, is it possible to bk tumblr to dw?
I know you’ve all heard me yell about Pillowfort before, but since this post is pre-PF I thought I’d just add it on.
Pillowfort.io is a new social networking site that’s still in private beta, that is trying to combine the best of LJ/DW and Tumblr. There are communities, threaded comments, and privacy control like LJ, with reblogging like Tumblr. Nothing explicit is against the TOS unless it’s posted repeatedly in the wrong place, but Nazi bullshit IS against the TOS. You can currently lock posts to followers only, soon they’re adding mutuals-only. If you edit or delete a post, all reblogs of it are edited or deleted, too. Lots of cool stuff!
They recently reopened their 4th beta wave signups – if you’d like an invite, sign up now, because the next wave will only be available through their crowdfunding campaign!
Now, it is definitely still in beta – there are lots of bugs, and lots of features they’re still working on adding. And since they’ve been letting in users a little at a time, there’s not a huge crowd of people there yet, but at this point it’s up to several thousand users so activity is starting to pick up.
If you’d like to see a little of what it’s like, here is a baking community I created that’s doing fairly well. Here is the Check Please community where Ngozi is cross-posting a lot of the new stuff so you can comment on it with real threaded discussions! And here is my blog, which is 90% me yakking about personal stuff because I’m keeping most (but not all) fandom stuff in comms. (Though a lot of my blog is also followers-only.)
Note related to OP: It’s not totally clear what their business model will be. They’ve done one crowdfunding campaign and will soon do another, but beyond that I’m not sure. They do NOT want to have advertising, and I’m pretty sure they’ve talked about a model much like LJ/DW, where there are paid accounts with extra features, or a la carte extras you can buy, etc. Honestly, that was a really successful model for LJ and I’d really love for another site to use it. I gladly paid $20/year to LJ for like 5-6 years and I’d rather do that again for a site I love than be inundated with ads or have it go under.
Pillowfort is amazing! It needs some more features, yes, and some more content, sure, but the developers are engaged with the users! If you were an LJ user and miss that, you’ll love pillowfort. If you’re exclusively tumblr, it’s got a lot more content control so come on over!
I’m colebaltblue over there!!!
You can’t import Tumblr to DW directly. (You can’t import Tumblr to anything directly; Tumblr is very, very different from every other blog-ish platform site.) You can import specific Tumblr entries to AO3 via their import tool, if those entries are fanworks.
Pillowfort is new and shiny and I hope it lasts longer than Imzy, but I am not expecting that – it’s not backed with serious funding, and I think the staff severely underestimates the hassles of managing fandom drama. (Currently, there is no drama. It’s tiny. But if people start using it to import their too-explicit-for-FFN works, it’ll collect the same “discourse” drama that Tumblr gets.) Also, PF hasn’t mentioned how they intend to make enough money to keep things going, nor how they’re going to cope with GDPR requirements.
So: Mildly hopeful, but there are several red flags that speak against the chances for long-term survival.
But definitely: BACK UP anything and everything on Tumblr that you care about. Make copies at the Wayback machine. Copy things out to Word docs. Save those files, because as soon as Yahoo realizes that there is no amount of ads that will bring money flowing like rivers, they will kill Tumblr like they killed Geocities and Delicious.
Survey Results: Fan Platform Use over Time
Particularly for those who were kind enough to participate in our survey last week, or to share it even after we halted data collection (because we received so many responses so quickly!), I wanted to give you something interesting right away. As you know, the academic writing and publishing process can be lengthy, so who knows when you might get a full paper from us! But in the meantime, this was the analysis I did this weekend.
The survey asked for participants to indicate what platforms they use/used from a given list, and also to indicate a date range (e.g., Tumblr 2006-2018). I parsed those date ranges in order to determine for a given platform how many of our participants were active in a given year. (This actually gave me an excuse to write some code for the first time in years. Jupyter Notebooks are super cool.)
(Click on the image above for full resolution!)
The Y axis is number of survey participants who indicated using the platform during a given time, and the X axis is year. (This starts at 1990, though I’ll note there were 10-ish participants who indicated using usenet, email lists, and/or messageboards in the 1980s.)
Some interesting things to note:
(1) See how fanfiction.net has a spike where there was a big drop off but then it stabilized? That’s around the time that they cracked down on adult content.
(2) I expected to see Livejournal decline drastically sooner, but it actually continued to climb a bit after Strikethrough and related things, until Tumblr and AO3 both started getting very popular. Based on what I’ve seen qualitatively so far, I do think that people were starting to leave, but that there had to be critical mass elsewhere in order for that leaving to start going en masse. There were also a lot of people who continued using Livejournal while they picked up other platforms as well.
(3) As my PhD student collaborator Brianna said, we have “a beautiful arc of AO3 and Tumblr being besties forever.” (This makes sense to me based on some findings from my previous work about AO3, and how Tumblr filled in the gap of social interaction left by Livejournal.)In the “other” category of fan platforms used, the most popular was Discord. This doesn’t surprise me! For the most part, participants had only been active in it for the past couple of years, which is why it didn’t show up specifically in the survey (which was constructed based on interview data we already had). We also saw less frequent mentions of Facebook, reddit, delicious/pinboard, and IRC.
Digging into the qualitative data will give this data much more explanatory power, but I think this is very interesting!
We also asked participants what their primary fandom was for each platform they used. Based on a pretty simple analysis (most popular words!), here are the top five fandoms from each platform:
Usenet: Star Trek, Buffy, X-Files, Star Wars, Sailor Moon
Email Lists: Harry Potter, Star Trek, Buffy, X-Files, Gundam Wing
Messageboards: Harry Potter, Buffy, Star Wars, Lord of the Rings, Sailor Moon
Fandom-Specific Archives: Harry Potter, Buffy, Stargate, X-Files, Doctor Who
Fanfiction.net: Harry Potter, Naruto, Buffy, Star Wars, Gundam Wing
Livejournal: Harry Potter, Supernatural, Stargate, Doctor Who, Merlin
DeviantArt: Harry Potter, Naruto, Kingdom Hearts, Supernatural, Final Fantasy
Dreamwidth: Harry Potter, Supernatural, Marvel, Stargate, RPF
Archive of Our Own: Marvel, Star Wars, Harry Potter, Supernatural, Teen Wolf
Tumblr: Marvel, Star Wars, Supernatural, Harry Potter, Teen Wolf
Twitter: Star Wars, Supernatural, Marvel, RPF, Yuri on Ice
Note that this is NOT necessarily representative of the overall popularity of certain fandoms on these platforms. Our survey, because it was targeting research questions about fandom migration, asked for participants who had been in fandom for 10+ years. This means that our results skewed older (mean 31; median 30; SD 8.6). And of course, most of the participants are currently in fandom, which means that it also misses people who have left fandom.
It is interesting to see the change across platforms and over time though! My favorite tidbit is how Star Wars was popular, dropped off, and then came back with gusto.
This is only the tip of the iceberg on this data analysis! If there’s anything else that is easily shared as we do this analysis, I’ll continue to do so. Otherwise, wish us luck and I’ll eventually share a completed analysis if/when (fingers crossed!) we publish on this.
I have a list of emails from everyone who participated and wanted to give us that info to share the results. If you’d like to be added to that list, send me an email at casey.fiesler@colorado.edu. Or just feel free to follow me here, or myself and Brianna on Twitter.
tora42's nonsensne 