*coughs* *taps microphone* Is this thing on?
Ahem, after something of a pause, this is (hopefully) the return of our regular AO3 newsletter! Your friendly AO3 news poster (that’s me, Lucy!) started off the year with good intentions and a shiny new format. However, between writing various other special edition posts, co-chairing the Communications committee, and dealing with real life, somehow the newsletter kept slipping down the to-do list. The good news is that new Communications staffer Camden has volunteered to take the baton, so hopefully these newsletters will be making a more frequent appearance in the future. So, without further ado, a few updates on the major happenings in recent times.
Major doings: performance, performance, performance
For the last few months, all the AO3 teams have been working really hard dealing with the various performance issues on the Archive. These started to appear in around May and got really acute in June, before we got them under control with a lot of hard work and some emergency measures (which involved 5 code deploys in the space of a few weeks). There’s been some heroic work from Coders, Testers and Systems to get the situation under control quickly: we’re really grateful to everyone who pitched in and helped out. We’re continuing to work on performance, most recently with a RAM upgrade and upgrades to our server software, and we’ll keep users posted as new developments arise.
Our performance crunch was largely down to a gigantic increase in traffic: we had over 1.4 MILLION visitors in the month of June alone! Our invitations queue also continues to grow apace: we recently increased the number of invitations being issued each day to 300, but this is still not keeping up with demand. We’re doing our best to keep expanding, but we can only increase accounts so fast while preserving site stability, so we’re afraid the queue will probably remain long for a while. 🙁
Tag filters are coming back – we promise!
As an emergency performance measure, we disabled our tag filters. We know they’re much missed, but their absence has enabled the site to run without 502 errors. We’d hoped to have the replacement filters finished by the end of July, but unfortunately they’re taking a bit longer than we’d hoped (turns out we have to let coders leave the house now and again). Everyone is working super hard to get them coded, tested, and up on the site as soon as possible – we apologise for the inconvenience in the meantime.
Tags and media categories
Speaking of tags, the new Category Change workgroup recently got off the ground with their discussion of how we organise the media categories on our Fandoms page. This is an issue which we’ve been chewing over in-house for a really, really long time – we know lots of users find the current categories confusing and/or problematic, and we’d like to come up with a better way of handling them. However, this is a really big can of worms: for example, we know that “Anime & Manga” is problematic as a catch-all category including manhua, manhwa, etc., but we also know that lots of fans are used to looking for anime and manga and would be confused if we scrapped that terminology altogether. So, we’re looking to find ways of handling the different categories which are understandable but also less problematic.
The Category Change workgroup includes members from lots of relevant committees: Internationalization and Outreach, Tag Wranglers, Support, and Accessibility, Design and Technology. They will be discussing the various issues and a range of possible solutions.
As part of this process they’ll be seeking input from various groups inside and outside the organisation and using that feedback to help identify some possible solutions. They’ll also be looking into the technical aspects – for example, improvements in searching and browsing on the Archive in general will open up more possibilities for how we handle categories specifically – so we don’t wind up with a great theoretical solution that’s impossible to implement. Tag wranglers have already given some great ideas and thoughts on this issue – they’ve been discussing it on and off for at least two years – and the workgroup is really looking forward to moving ahead with it. Stay posted for more news on this!
Tag wrangling – a new view!
We’ve had lots of feedback from users who’ve said that they’d love to be able to get more information about how tags are wrangled and how they all relate to one another. This has also been one of the most frequent feature requests from tag wranglers, who put lots of work into making the tags link up in a meaningful way and would like their work to be more useful to users. Our awesome coder sarken took up the challenge and did some work to make the tags pages which are currently only visible to tag wranglers visible to any user. This will be an ‘alpha’ feature when it goes live – we expect we’ll get lots of feedback about how it could be better – but we hope this news will be welcome to lots of people who have been wondering what’s going on behind the scenes!
For those who are really curious about tag wrangling, the Tag Wrangling Committee are also currently preparing a public version of the tag wrangling guidelines. We hope to have these posted shortly after the tags pages are made public.
What else is up in the world of tags?
The Archive welcomes all types of fanworks! We have plans to make it much more useful for different types of fanworks by adding some more browsing and posting features. However, right now, people use the “Additional Tags” field to make it clear what type of creative endeavour their particular work consists of. Tag Wranglers have wrangled the more common tags like podfic, fanart, fanvids and AMV, and even things like knitting and crochet patterns, and we’re always excited to see people tagging for a diverse range of fanworks.
But you might not be aware that the tags we’ve wrangled go further than just describing the media type and content. Once the filters are reenabled, you’ll be able to use the additional tags we’ve made canonical to further filter those media types by format or length – click on one of the fanwork types linked above, and browse the list, or use the advanced search.
For Fanart, there are tags that describe medium, such as Crayons, or Digital Art. For podfic and video, there are tags which describe the audio and video formats that it is available in, such as Video Format: AVI or Audio Format: MP3. For podfic specifically, there are tags which indicate the length of the recorded work, such as Podfic Length: 0-10 Minutes, through to Podfic Length: 15-20 hours.
Wranglers will add more to the canonical tag lists as we find them, but if there’s a media related tag that you’d find useful to be filterable, let us know through a support request or by tweeting us at @ao3_wranglers!
We hit a snag recently with our podfic provision when Google disabled their Audio Player, which we were using to enable people to stream podfic. The good news is that we’ve snagged a copy of the audio player code to host on our own servers, so this will return in our next code deploy. When the deploy has taken place, you’ll once more be able to embed podfic, using the following embed code:
<embed type="application/x-shockwave-flash" flashvars="audioUrl=MP3_FILE_URL" src="http://archiveofourown.org/static/audio-player.swf" width="400" height="270" quality="best" > </embed>
(Essentially, if you were previously using the Google player, you’ll just be replacing the Google url with http://archiveofourown.org/static/audio-player.swf.) Apologies for the inconvenience in the meantime!
The recent increase in users on the site has resulted in a corresponding increase in Support tickets. The fantastic Support team have been amazing at keeping up with all the tickets and giving users the help they need. Their turnaround is really quick and we know that people appreciate their hard work – thanks Support!
Support staffer Yshyn wrote a great post on Support ticket stats which gives an insight into what kinds of questions people ask and how many tickets we receive. One of the things which has been particularly nice for Support recently is the number of tickets which have included some positive feedback for the site, even when it’s been experiencing problems. They always pass this onto the other teams and it makes everyone’s day brighter – thank you lovely users!
Roadmap, feature requests, and plans
We try to keep users informed on what’s going on with the site, although it can be lots of hard work! You can see the details of all our code updates to date in our release notes. Awesome AD&T staffer mumble is currently working really hard to update our very, very outdated Roadmap: this is important as a way of helping us focus our work, and a useful way of giving users an idea of what we have planned. We’ve also recently started using a new tool, Trello, to manage feature requests from users. Like our codebase on Github and our issues list on Google Code, this is open to the public, so you can now see what’s already been requested and vote for the things you’d like! You can also see proposals that were rejected, and why. Support staffer Sam wrote up a handy introduction to our internal tools, and there’s also a new AO3 Internal Tools FAQ.
We have limited resources, so we can’t always introduce requested features as fast as we’d like (right now, we’re focusing on site stability and trying to avoid too many new things). However, we hope that these tools will help people see what’s in the works, and we’re excited to have found a tool (Trello) which makes it easier for people to let us know which features they’re most interested in.
We welcome feedback from users! If you have questions or comments, feel free to leave them in the comments of the latest news post, or send in a Support request (if you’re reporting a bug, please send that to Support, as they’re super efficient – comments on our news posts sometimes get overlooked).