Design Team Fedora Activity Day (FAD) Event Report

Fedora Design Team Logo

design team fad attendees portrait
From left to right: Mo Duffy, Marie Nordin, Masha Leonova, Chris Roberts, Radhika Kolathumani, Sirko Kemter (photo credit: Sirko Kemter)

Two weekends ago now, we had a 2-day Fedora Activity Day (heh, a 2-day day) for the Fedora Design Team. We had three main goals for this FAD, although one of them we didn’t cover (:-() :

  • Hold a one-day badges hackfest – the full event report is available for this event – we have wanted to do an outreach activity for some time so this was a great start.
  • Work out design team logistics – some of our members have changed location causing some meeting time issues despite a few different attempts to work around them. We had a few other issues to tackle too (list to come later in this post.) We were able to work through all points and come up with solutions except for one (we ran out of time.)
  • Usability test / brainstorm on the Design Team Hub on Fedora Hubs – so the plan was that the Design Team Hub would be nearly ready for the Flock demo the next week, but this wasn’t exactly the case so we couldn’t test it. With all of the last-minute prep for the workshop event, we didn’t have any time to have much discussion on hubs, either. We did, however, discuss some related hub needs in going through our own workflow in our team logistics discussion, so we did hit on this briefly.

So I’m going to cover the topics discussed aside from the workshop (which already has a full event report), but first I want to talk a little bit about the logistics of planning a FAD and how that worked out first since I totally nerded out on that aspect and learned a lot I want to share. Then, I’ll talk about our design team discussion, the conclusions we reached, and the loose ends we need to tie up still.

Logistics

I had already planned an earlier Design Team FAD for January 2015, so I wasn’t totally new to the process. There were definitely challenges though.

Budget

First, we requested funding from the Fedora Council in late March. We found out 6 weeks later (early May, a little less than 3 months before the event) that we had funding approval, although the details about how that would work weren’t solidified until less than 4 weeks before the event.
Happily, I assumed it’d be approved, filed a request to use the Red Hat Westford facility for the event. There were two types of tickets I had to file for this – a GWS Special Event Request and a GWS Meeting Support Request. The special event request was the first one – I filed that on June 1 (2 months ahead) and it was approved June 21 (took about 3 weeks.) Then, on 7/25 the week before the event, I filed the meeting support request to have the room arranged in classroom style as well as open up the wall between the two medium-sized conference rooms so we had one big room for the community event. I also set up a meeting with the A/V support guy, Malcolm, to get a quick run through of how to get that working. It was good I went ahead and filed the initial request since it took 3 weeks to go through.
The reason it took a while to work out the details on the budget was because we scheduled the event for right before Flock, which meant coordinating / sharing budgets. We did this both to save money and also to make sure we could discuss design-team related Flock stuff before heading to Flock. While this saved some money ultimately, IMHO the complications weren’t worth it:

  • We had to wait for the Flock talk proposals to be reviewed and processed before we knew which FAD attendees would also be funded for Flock, which delayed things.
  • Since things were delayed from that, we ended up missing on some great flight pricing, which meant Ryan Lerch wasn’t able to come 🙁
  • To be able to afford the attendees we had with less than 4 weeks to go, we had to do this weird flight nesting trick jzb figured out. Basically, we booked home<=>BOS round trip tickets, then BOS<=>KRK round trip tickets. This meant Sirko had to fly to Boston after Flock before he could head home to PNH, but it saved a *ton* of money.

fad budget spreadsheet screenshot
behold, our budget

Another complication: we maxed out my corporate card limit before everything was booked. 🙂 I now have a credit increase, so hopefully next event this won’t happen!
The biggest positive budget-wise for this event was the venue cost – free. 🙂 Red Hat Westford kindly hosted us.
I filed the expense reports for the event this past week, and although the entire event was under budget, we had some unanticipated costs as well as a small overage in food budget:

  • Our original food budget was $660. We spent $685.28. We were $25.28 over. (Pretty good IMHO. I used an online pizza calculator to figure out budget for the community event and was overly generous in how much pizza people would consume. 🙂 )
  • We spent $185.83 in unanticipated costs. This included tolls (it costs $3.50 to leave Logan Airport), parking fees, gas, and hotel taxes ($90 in hotel taxes!)

Lessons Learned:

  • Sharing budget with other events slows your timeline down – proceed with caution!
  • Co-location with another event is a better way to share costs logistically.
  • Pizza calculators are a good tool for figuring out food budget. 🙂
  • Budget in a tank of gas if you’ve got a rental.
  • Figure out what tolls you’ll encounter. Oh and PAY CASH, in the US EzPass with a rental car is a ripoff.
  • Ask the hotel for price estimates including taxes/fees.

Transportation

I rented a minivan to get folks between Westford and the airport as well as between the hotel and the office. I carpool with my husband to work, so I picked it up near the Red Hat Westford office and set up the booking so I was able to leave it at Logan Airport after the last airport run.

Our chariot. I cropped him out of the portrait. Sorry, Toyota Sienna! It has nice pickup. I still am never buying a minivan ever, even if I have more kids. Never minivan, never!
Our chariot. I cropped him out of the portrait. Sorry, Toyota Sienna! It has nice pickup. I still am never buying a minivan ever, even if I have more kids. Never minivan, never!

With international flights and folks coming in on different nights, and the fact I actually live much closer to the airport than the hotel up in Westford (1 hour apart) – by the time the FAD started, I was really worn down as I had 3 nights in a row leading up to the FAD where I wasn’t getting home until midnight at the earliest and I had logged many hours driving, particularly in brutal Boston rush hour traffic. For dropoffs, it was not as bad as everybody left on the same day and there were only 2 airport trips then. Still – not getting home before my kids went to bed and my lack of sleep was a definite strain on my family.
So we had a free venue, but at a cost. For future FAD event planners, I would recommend either trying to get flights coming in on the same day as much as possible and/or sharing the load of airport pickups. Even better, would be to hold the event closer to the airport, but this wasn’t an option for us because of the cost that would entail and the fact we have such a geographically-distributed team.
The transportation situation - those time estimates aren't rush hour yet!
The transportation situation – those time estimates aren’t rush hour yet!

One thing that went very well that is common sense but bears repeating anyway – if you’re picking folks up from the airport, get their phone #’s ahead of time. Having folks phone numbers made pickup logistics waaaaay easier. If you have international numbers, look up how to dial them ahead of time. 🙂

Lessons Learned:

  • Try hard to cluster flights when possible to make for less pickups if the distance between airport / venue is great.
  • If possible, share responsibility for driving with someone to spread the load.
  • Closer to the airport logistically means spending less time in a car and less road trips, leaving more time for hacking.
  • Don’t burn yourself out before the event even starts. 🙂
  • Collect the phone numbers of everyone you’re picking up, or provide them some way of contacting you just in case you can’t find each other.
We're dispersed...
We’re dispersed… (original list of attendees’ locations or origin)

Food

This one went pretty smoothly. Westford has a lot of restaurants; actually, we have a lot more restaurants in Westford with vegetarian options than we did less than 2 years ago at the last Design Team FAD.
For the community event, the invite mentioned that we’d be providing pizzas. We had some special dietary requests from that, so I looked up pizza place that could accommodate, would deliver, and had good ratings. There were two that met the criteria so I went with the one that had the best ratings.
Since the Fedora design team FAD participants were leading / teaching the session, I went over the menu with them the day before the community event, took their orders for non-pizza sandwiches/salads, and called the order in right then and there. (I worried placing the order too far in advance would mean it’d get lost in the shuffle. Lesson learned from the 2015 FAD where Panera forgot our order!) Delivery was a must, because of the ease of not having to go and pick it up.
For snacks, we stopped by a local supermarket either before or after lunch on the first day and grabbed whatever appealed to us. Total bill: $30, and we had tons of drinks and yummy snacks (including fresh blueberries) that kept us tided over the whole weekend and were gone by the end.
We were pretty casual with other meals. Folks at the hotel had breakfast at the hotel, which meant less receipts to track for me. We just drove to places close by for lunch and dinner, and being a local + vegetarian meant we had options for everybody. I agonized way too much about lunch and dinner last FAD (although there were less options then.) Keeping it casual worked well this time; one night we tried to have dinner at a local Indian place and found out they had recently been evicted! (Luckily, there was a good Indian place right down the road.)

Lessons Learned:

  • For large orders, call in the day before and not so far in advance that the restaurant forgets your order.
  • Supermarkets are a cheap way to get a snack supply. Making it a group run ensures everyone has something they can enjoy.
  • Having a local with dietary restrictions can help make sure food options are available for everyone.

Okay, enough for logistics nerdery. Let’s move on to the meat here!

Design Team Planning

We spent most of the first day on Fedora Design team planning with a bit of logistics work for the workshop the following day. First, we started by opening up an Inkscape session up on the projector and calling out the stuff we wanted to work on. It ended up like this:

Screenshot of FAD brainstorming session from Inkscape
Screenshot of FAD brainstorming session from Inkscape

But let’s break it down because I suspect you had to be there to get this. Our high-level list of things to discuss broke down like this:

Discussion Topics

  • Newcomers
    – how can we better welcome newcomers to the team?
  • Pagure migration
    Fedora Trac is going to be sunset in favor of Pagure. How will we manage this transition?
  • Meeting times
    – we’ve been struggling to find a meeting time that works for everyone because we are so dispersed. What to do?
  • Status of our ticket queue
    – namely, our ticket system doesn’t have enough tickets for newbies to take!
  • Badges
    – conversely, we have SO MANY badge tickets needing artwork. How to manage?
  • Distro-related design
    – we need to create release artwork every release, but there’s no tickets for it so we end up forgetting about it. What to do?
  • Commops Thread
    – this point refers to Justin’s design-team list post about ambassadors working with the design team – how can we better work with ambassadors to get nice swag out without compromising the Fedora brand?

Let’s dive into each one.

Newcomers

This is the only topic I don’t think we fully explored. We did have some ideas here though:

  • Fedora Hubs will definitely help provide a single landing page for new comers to see what we’re working on in one place to get a feel for the projects we have going on – right now our work is scattered. Having a badge mission for joining the design team should make for a better onboarding experience – we need to work out what badges would be on that path though. One of the pain points we talked about was how incoming newbies go straight to design team members instead of looking at the ticket queue, which makes the process more manual and thus slower. We’re hoping Hubs can make it more self-service.
  • We had the idea to have something like whatcanidoforfedora.org, but specifically for the design team. One of the things we talked about is having it serve up tickets tagged with a ‘newbie’ tag from both the design-team and badges ticket systems, and have the tickets displayed by category. (E.g., are you interested in UX? Here’s a UX ticket.) The tricky part – our data wouldn’t be static as whatcanidoforfedora.org’s is – we wouldn’t want to present people with a ticket that was already assigned, for example. We’d only want to present tickets that were open and unassigned. Chris did quite a bit of investigation into this and seems to think it might be possible to modify asknot-ng to support this.
  • A Fedora Hubs widget that integrated with team-specific asknot instances was a natural idea that came out of this.
  • We do regular ticket triage during meetings. We decided as part of that effort, we should tag tickets with a difficulty level so it’s easier to find tickets for newbies, and maybe even try to have regular contributors avoid the easy ones to leave them open for newbies. We had some discussion about ticket difficulty level scales that we didn’t get to finish – at one point we were thinking:
    • Easy (1 point) (e.g., a simple text-replacement badge.)
    • Moderate (3 points) (e.g., a fresh badge concept with new illustration work.)
    • Difficult / Complex (10 points) (e.g., a minor UX project or a full badge series of 4-5 badges with original artwork.)

    Or something like this, and have a required number of points. This is a discussion we really need to finish.

  • Membership aspects we talked about – what level of work do we want to require for team emembership? Once a member, how much work do we want to require (if any) to stay “current?” How long should a membership be inactive before we retire? (Not to take anything away from someone – but it’s handy to have a list of active members and a handle on how many active folks there are to try to delegate tasks and plan things like this FAD or meetups at Flock.) No answers, but a lot of hard questions. This came up naturally thinking about membership from the beginning to the end.
  • We talked about potentially clearing inactive accounts out of the design-team group and doing this regularly. (By inactive, we mean FAS account has not been logged into from any Fedora service for ~1 year.)
  • Have a formal mentor process, so as folks sign up to join the team, they are assigned a mentor, similar to the ambassador process. Right now, we’re a bit all over the place. It’d be nice for incoming folks to have one person to contact (and this has worked well in the past, e.g., Mo mentoring interns, and Marie mentoring new badgers.)

Pagure migration

We talked about what features we really needed to be able to migrate:

  • The ability to export the data, since we use our trac tickets for design asset storage. We found out this is being worked on, so this concern is somewhat allayed.
  • The ability to generate reports for ticket review in meetings. (We rely on the custom reports Chris and Paul Frields created for us at the last FAD.) We talked through this and decided we wanted a few things:
    • We’d like to be able to do an “anti-tag” in pagure. So we’d want to view a list of tickets that did not have the “triage” tag on them, so we could go through them and triage them, and add a ‘triage’ tag as we completed triage. That would help us keep track of what new tickets needed to be assessed and which had already been evaluated.
    • We’d like some time-based automation of tag application, but don’t know how that would work. For example, right now if a reporter hasn’t responded for 4 weeks, we classify that ticket as “stalled.” So we’d want tickets where the reporter hasn’t responded in 4 weeks to be marked as “stalled.” Similarly, tickets that haven’t had activity for 2 weeks or more are considered “aging”, so we’d like an “aging” tag applied to them. So on and so forth.
    • We need attachment support for tickets – we discovered this was being worked on too. Currently pagure supports PNG image attachments but we have a wider range of asset types we need to attach – PDFs, Scribus SLAs, SVGs, etc. We tested these out in pagure and they didn’t work.

We agreed we need to follow up with pingou on our needs and our ideas here to see if any of these RFEs (or other solutions) could be worked out in Pagure. We were pretty excited that work was already happening on some of the items we thought would help meet our needs in being able to migrate over.

We don’t have enough tickets! (AKA we are too awesome)

We tend to grab tickets and finish them (or at least hold on to them) pretty quickly on the design team these days. This makes it harder for newbies to find things to work on to meet our membership requirement. We talked about a couple of things here, in addition to related topics already covered in the newbie discussion summary:

  • We need to be more strict about removing assignees from tickets with inactivity. If we’ve pinged the ticket owner twice (which should happen in at least a 4 week period of inactivity from the assignee) and had no response, we should unapologetically just reopen up the ticket for others to take. No hard feelings! Would be even better if we could automate this….
  • We should fill out the ticket queue with our regular release tasks. Which leads to another topic…

Distro-related design (Release Artwork)

Our meetings are very ticket-driven, so we don’t end up covering release artwork during them. Which leads to a scramble… we’ve been getting it done, but it’d be nice for it to involve less stress!
Ideally, we’d like some kind of solution that would automatically create tickets in our system for each work item per release once a new release cycle begins… but we don’t want to create a new system for trac since we’ll be migrating to pagure anyway. So we’ll create these tickets manually now, and hope to automate this once we’ve migrated to pagure.
We also reviewed our release deliverables and talked through each. A to-do item that came up here: We should talk to Jan Kurik and have him remove the splash tasks (we don’t create those splash screens anymore) and add social media banner tasks (we’ve started getting requests for these.) We should also drop CD, DVD, and DVD for multi, and DVD for workstation (transcribing this now I wonder if it’s right.) We also should talk to bproffitt about which social media Fedora users the most and what kind of banners we should create for those accounts for each release. So in summary: we need to drop some unnecessary items from the release schedule that we don’t create anymore, and we should do more research about social media banners and have them added to the schedule.
Another thing I forgot when I initially posted this – we need some kind of entropy / inspiration to keep our default wallpapers going. For the past few releases, we’ve gotten a lot of positive feedback and very few complaints, but we need more inspiration. An idea we came up with was to have a design-team internal ‘theme scheme’ where we go through the letters of the alphabet and draw some inspiration from an innovator related to that letter. We haven’t picked one for F25 yet and need to soon!
Finally, we talked about wallpapers. We’d like for the Fedora supplemental wallpapers to be installed by default – they tend to be popular but many users also don’t know they are there. We thought a good solution might be to propose an internship (maybe Outreachy, maybe GSoC?) to revive Nuancier could serve them up.

Badges

We never seem to have time to talk through the badges tickets during our meetings, and there are an awful lot of them. We talked about starting to hold a monthly badge meeting to see if this will address it, with the same kind of ticket triage approach we use for the main design team meetings. Overall, Marie and Maria have been doing a great job mentoring baby badgers!

Commops Thread

We also covered Justin’s design-team list post about ambassadors working with the design team, particularly about swag as that tends to be a hot-button issue. For reasons inexplicable to me except for perhaps that I am spaz, I stopped taking notes in Inkscape and started using the whiteboard on this one:

photo of whiteboard (contents described below)
Swag discussion whiteboard (with wifi password scrubbed 🙂 )

We had a few issues we were looking to address here:

  • Sometimes swag is produced too cheaply and doesn’t come out correctly. For example, recently Fedora DVDs were produced with sleeves where Fedora blue came out… black. (For visuals of some good examples compared to bad examples with these sorts of mistakes, check this out.)
  • Sometimes ambassadors don’t understand which types of files to send to printers – they grab a small size bitmap off of the wiki without asking for the print-ready version and things come out pixelated or distorted.
  • Sometimes files are used that don’t have a layer for die cutting – which results in sticker sheets with no cuts that you have to manually cut out with scissors (a waste!)
  • Sometimes files are sent to the printer with no bleeds – and the printer ends up going into the file and manipulating it, sometimes with disastrous results. If a design team member had been involved, they would have known to set the bleeds before sending to the printer.
  • Generally, printers sometimes have no clue, and without a designer working with them they make guesses that are oftentimes wrong and result in poor output.
  • Different regions have different price points and quality per type of item. For example, DVD production in Cambodia is very, very expensive – but print and embroidery items are high-quality and cheap.

Overall, we had concerns about money getting wasted on swag when – with a little coordination – we could produce higher-quality products and save money.
We brainstormed some ideas that we thought might help:

  • Swag quality oversight – Goods produced too cheaply hurt our brand. Could we come up with an approved vendor list, so we have some assurances of a base level of quality? This can be an open process, so we can add additional vendors at any time, but we’ll need some samples of work before they can be approved, and keep logs of our experience with them.
  • Swag design oversight – Ambassadors enjoy their autonomy. We recognize that’s important, but at a certain point sometimes overenthusiastic folks without design knowledge can end up spending a lot of money on items that don’t reflect our brand too well. We thought about setting some kind of cap – if you’re spending more than say $100 on swag, you need design team signoff – a designer will work with you to produce print-ready files and talk to the vendor to make sure everything comes out with a base quality level.
  • Control regional differences – Could we suggest one base swag producer per ambassador region, and indicate what types of products we use them for by default? Per product, we should have a base quality level requirement – e.g., DVDs cannot be burnt – they must be pressed.
  • Okay, I hope this is a fair summary of the discussion. I feel like we could have an entire FAD that focused just on swag. I think we had a lot of ideas here, and it could use more discussion too.

    Meeting Times

    We talked about meeting times. There is no way to get a meeting time that works for everybody, so we decided to split into North America / EMEA / LATAM, and APAC regions. Sirko, Ryan Lerch, and Yogi will lead the APAC time (as of yet to be determined.) And the North America / LATAM / EMEA time will be the traditional design team time – Thursdays at 10 AM ET. Each region will meet on a rotating basis, so one week it’ll be region #1, the next region #2. Each region will meet at least 2x a month then.
    How do we stay coordinated? We came up with a cool idea – the first item of each meeting will be to review the meetbot logs from the other region’s last meeting. That way, we’ll be able to keep up with what the other region is doing, and any questions/concerns we have, they’ll see when they review our minutes the next week. We haven’t had a chance to test this out yet, but I’m curious to see how it works in practice!

    Fun

    Chris’ flight left on Sunday morning, but everybody else had flights over to Poland which left in the evening, so before we went to the airport, we spent some time exploring Boston. First we went to the Boloco, a cool Boston burrito-chain, then the sun decided to come out so we found a parking spot by Long Wharf and I gave everybody a walking tour of Quincy Market and the North End. Then we headed to the airport and said our goodbyes. 🙂

    From left to right: Mo, Masha, Marie, Radhika
    From left to right: Mo, Masha, Marie, Radhika

    What’s Next?

    There’s a lot of little action items embedded here. We covered a lot of ground, but we have a lot more work to do! OK, it’s taken me two weeks to get to this point and I don’t want this blog post delayed anymore, so I’m just going for it and posting now. 🙂 Enjoy!

    13 Comments

    1. Pingback: Links 17/8/2016: GNOME and Debian Anniversaries | Techrights

    Leave a Reply

    Your email address will not be published. Required fields are marked *

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