As part of our series giving an insight into what goes on behind the scenes at the Archive of Our Own, Support staffer Sam has written up a day in the life of an AO3 Support staffer! Sam started out volunteering with the AO3 as a tag wrangler, and joined the Support team at the beginning of the July Ticket Blitz. He has degrees in journalism, English literature, and cognitive and discourse linguistics. He's taught skills-based computer classes for near a decade, so Support was a natural fit. Sam tends to jump on tickets involving CSS code and the skins; downloads, especially ePUBs; and all things Tag, as he's currently the liaison (read: troublemaker) between Support and Tag Wrangling.

By and large, Support is all about answering all the tickets that come in. To do this involves a whole lot of trying to break the Archive in new and creative ways, keeping a light eye on what the Coders are up to under the hood, and generally trying to divine what the users need and want.

There are a whole bunch of resources we use to do so (some of these resources are only accessible to staffers - we've linked the public ones):

• 16bugs - This is our main ticket tracker where we keep information on each ticket that comes in, and any communication with the user or other committees. This will soon (hopefully) be replaced with the Support Board.
• Campfire - The chat interface all AO3 committees use. Support has its own room where we'll request betas and comment on tickets, life, and fic; and we're often lurking in the Coders room, checking for surprise!bugs and fixes.
• Both the "Beta" Archive (the real one) and the Test Archive (where we test new code) - Sometimes we have to track down to see what a reported bug is doing and possibly see if it's already been fixed on the next release.
• The wiki "Knowledge Base" - One of our Support reps has been collating the answers to commonly answered questions into a massive internal reference source. This is awesome, because it helps ensure that knowledge gets passed on and we don't have to duplicate work.
• Google code otwarchive issues - The list of things-to-do for the Coders, to see if a bug is known or a feature planned (and occasionally provide all the information).
• Squee! - This internal squee page is where we keep all records that we're doing something right (700+ and counting). It helps us track what's working and also provides a nice place for people to go when they need a boost!

I pop open my email to see if anyone sent beta requests to the list or if any tickets have come in. (Most responses to users are beta-read by a second support member, for accuracy, clarity, and something resembling UK/US/CA/AU English.) I'll also log into Campfire and check the Support room, since several staff will leave their beta requests in there.

Both 16bugs and the Support form send an email to each Support staff when a ticket comes in. I tend to not read the emails themselves, but use them as a sign to go check 16bugs and see what new tickets are there.

Certain tickets we immediately assign to another committee (Legal, Abuse, Tag Wranglers) and wait for a response from that committee before contacting the user. Some committees will contact the user directly, some don't.

Since every ticket is different, I'm going to give examples of two recent tickets. (All confidential details are removed, but hi, users, if you recognize your questions!)

A ticket comes in from a user asking about a problem logging in with hir OpenID. I open the ticket in 16Bugs, and in the ticket set my name for "Assigned to" and "Status" to "Solving".

I've heard some talk about the discontinuance of OpenID, so I poke my head into the Coders chatroom in Campfire and ask if anyone has more specific information. I luck out and one of our Coders knows of two open code tickets regarding OpenID, which saves me the time. I open the tickets in GCode and skim them, seeing the current development status for OpenID (we're planning to phase it out).

To make sure the code is still working as intended, I use my own account as a guinea pig, setting up my own OpenID login, logging out and testing it. It all works, so I assume that the problem comes from improper configuration. I step back through the process involved and make detailed notes to set up OpenID. I add those to our Knowledge Base on the wiki so next time we have a question like this, the info is easy to find.

I then compose a reply to the user as a comment to the ticket in 16bugs. I also copy in the links to the GCode tickets into 16bugs for reference. After coming up with the response, I poke my head into Support chat in Campfire. Fortunately, one of the other Support staffers is on, so I ask hir to beta my response. Sie reads it over, we discuss and revise a few lines, and sie comments in the thread that it looks good. I copy the response from 16bugs.

In my email, already set to forward through the official email, I search and find the ticket email that came in and reply to it, using the copied text from 16bugs.

After sending the email, in 16bugs I set the status to "Solved". If the user responds, I can find the ticket in 16Bugs and reopen it as needed. When the user responds that sie doesn't actually have an account, I send back a betaed response on how to get an invite, either through the queue or through a friend.

Another ticket has come in regarding a tag that's misfiled - in this case, a fictional football team that has somehow been wrangled into the "Football RPF" fandom. Since this relates to wrangling policy, I'll mark the ticket to watch it and assign it to the "Tag Wranglers" and wait for a response from one of the Wrangling committee members before I send a response to the user. In this case, it's an easy fix by the wranglers, and I'm able to quickly notify the user that the tag has been re-wrangled.

There. My two tickets for the day - with the new influx of staff we've had, and a fairly slow inflow of tickets lately, sometimes I don't even get the option to do that many! (However, different times of year or new lots of code can produce a sudden uptick, so we take the rest while we can!)

Sometimes tickets aren't nearly as straightforward. Sometimes it takes time to track down the bug - while I'm doing so, I'll set the status to "Testing". If the response requests additional information from the user, I'll leave it as "Solving" until I can get a response from the user. If I find other bugs in 16Bugs or code issues in GCode, I'll leave links in the comment leading to them, as well as leave links to the active ticket elsewhere. If the ticket contains a feature request, I'll make a note on our wiki's Feature Requests page and if it continues praise, I make a note on the Squee page.

Let it be said: us Support minions are human. There are tickets that have us staring at our monitors in awe, wondering "how did they do that?" There are tickets where we realize we've answered the same thing frequently, and therefore need better documentation and/or to prioritize a bug fix. There are times that we look at a ticket and mentally draw straws about who gets to tell the Coders that the recently-fixed feature isn't so fixed.

All that said? It's all worth it. It's worth it, helping the users better interact with the Archive. It's worth it, seeing the feature requests and ideas. It's worth it, feeling like I'm contributing to the development of the Archive. It's worth it.

I've now knocked out a couple tickets, updated a page on the wiki, updated a bug on GCode, and tripped over a work I want to read. Never let it be said I can't take a sign! Off to read!

Mirrored from an original post on the Archive of Our Own.

Today we'd like to shine the spotlight on the Translation committee. While you have probably seen the fruits of the translator teams' work on the OTW website (and coming soon, the Archive of Our Own!), the Translation committee is the one pulling the strings in the background. Its three members hele, Tonje and Julia are responsible for organising and aiding the translator teams.

"Language nerdery" is part of the committee's everyday operations. The committee members help translator teams solve language puzzles (quick, what’s the equivalent of a “3rd District Court of Appeals” in Finnish? Wikipedia to the rescue!) and liaise with other OTW committees, such as asking Legal to clarify US-American concepts or discussing the intricacies of our multilingual website with Webmasters. We'll expand on the creative aspect of actual translation work in an upcoming spotlight — although all committee members share a fascination with language, actual translating is not part of their job description. Rather, the priority for the committee is keeping the teams up to date and organised.

The committee calls their project management style "creative improvisation". From the start they've had to fit processes and tools to a unique project that's ever-growing in both scale and scope (eleven languages on our website, and the committee has bigger plans yet). It’s definitely exciting, but it comes with a certain amount of trial and error, which the translator teams have been graciously putting up with.

A simplified graphic of the committee's modus operandi. Please note: this figure doesn’t accurately represent number or size of teams, committee relationships, etc.

If this work sounds interesting to you — in short, if you are interested in language-related project management — the committee is looking for a few additional members. Please contact either the Volunteers and Recruiting committee, or, for additional information, the Translation committee.

Written by Franzeska Dickson

Tag wranglers are our biggest pool of volunteers, with over 100 wranglers around the world working to organise the Archive's 100,000+ tags. The wranglers put in a varying amount of work, each wrangler deciding for themselves how much time they can dedicate to wrangling. We now profile a week in the life of one of our... more dedicated wranglers...

Monday: Look at all those fandoms without wranglers. How is InuYasha still missing a wrangler?! I’d better do some emergency wrangling before it gets any more out of control. That’s funny: these character names aren’t in Japanese order. Or are they? Maybe the original manga uses names in a weird order. I’d better go look at Wikipedia...

5 hours later:

Damnit.

Tuesday: 20 minutes spent making a google docs spreadsheet of all the Japanese names. 5 minutes spent wrangling all the pairing and character tags now that I’ve decided on a standard format. 5 hours spent catching up on all the Tuesday tv shows. (What’s up with that, US networks? Why is everything on Tuesday?!)

Wednesday: Hey, I know that obscure 19thC novel! I don’t remember this character though. Maybe there’s a copy on Project Gutenberg...

Thursday: Robin Hood doesn’t have a wrangler. That should be easy to wrangle... Huh. There’s a megafandom page on the internal wiki.

...

Just how many Robin Hood adaptations are there anyway?! And they all spell the names differently! At least the megafandom team has a spreadsheet.

Friday: Characters and pairings seem pretty well under control, but my fandoms are full of freeform tags. Hey, ‘Bob’ isn’t a character in my fandom. Who’s Bob? I wonder if this is on Fanlore...

Saturday: Dreamwidth’s content search is awesome! I can search for my name, and my new fandom, and that weird word from word a day that I’m pretty sure no one has actually used in any posts. Look, someone posted about tag wrangling. And they have a question. That I know the answer to!

Sunday: This series isn’t even on Wikipedia. Maybe if I google it, I can find something. Hey, a bunch of the characters are listed on the TV Tropes page...

...
...

What do you mean it’s Monday already?!

Text by Benny

The Tag Wrangling committee is where we are shining the spotlight next! The Wranglers are vital to Archive of Our Own – they make sure that all the diverse tags associated with a fandom are grouped and align as the individual writers intended. This means that when different people use different tags to indicate the same thing (for example, referring to the character from Harry Potter variously as ‘Snape’, ‘Severus Snape’, ‘His Greasy-Haired Half-Blooded-ness’, 'Sexy Sev', etc.), the Tag Wranglers make sure all those tags are connected invisibly to lead to the same results. That allows fannish diversity without sacrificing usability for readers!

Like we said, the Tag Wranglers are crucial to the function of the AO3. Without them, finding the works you want would be far more challenging, and labeling the works you post would be far less flexible. It’s not an easy job, though – Tag Wranglers must be very familiar with the fandom they are wrangling for, and sensitive to the needs of both current and future users from a variety of traditions.

For example, what happens when there’s a fandom – let's use Teletubbies as an example – that none of the available wranglers are familiar with? Generally that means the wranglers must do research (one of the rare examples where you can justify marathoning episodes of Teletubbies, we suppose, though the wrangler would also consult sources like Fanlore and Wikipedia), but how much of the canon and fannish history should they research before they are certain they are familiar enough to wrangle its tags?

Another common situation is when Wranglers encounter tags that are ambiguous in their intent. For example, if someone has an intentionally OOC Dumbledore who they tag as ‘Headmaster Albus MacDaddy-D’ and who stars in a number of their stories and only their stories, should his tag link to all the Dumbledore tags, or does he deserve his own unique status? These are the sort of dilemmas tag wranglers encounter every day.

We talked about how tags work before in Wuzzles Explained, if you're interested in reading more. Tag Wranglers account for about half of our non-Staff volunteer force, so we are going to spotlight several other aspects of their work in the next few weeks.