2014 Lessons Learned: The Negative

[Alex Golec]

Originally Posted by Steven Donow

Similarly, in regards to Dean's List, I was very unhappy with the way it was handled/done in district systems/the inconsistencies between different regions in how they determined Finalists.

I'm speaking from the perspective of a Michigan volunteer here:

Being the first year of Dean's List interviews, we expected challenges along the way. Overall, the interview process went well and allowed us to add extra depth to how we assess nominated students. It also gave these students a moment to shine. I have a feeling that FIRST won't be getting rid of this process anytime soon.

For improvements, I'm only aware of how Michigan events operated and what we need to do better. I already know we need to improve how we communicate to teams about what to expect, and where their students will interview. That's something we didn't have established prior to March, but is on our radar for next year.

Scheduling will likely be a persistent challenge. So many of these extraordinary students are drive team members, or Chairman's presenters, or have other responsibilities at the event. Most events adopted a flexible sign-up schedule to work around these issues, but we're looking into how to make it better.

Providing feedback, as simple as it seems, is a rather challenging process. It takes time and thought to write good feedback, and we don't want to "sound generic." Time was a challenge this year, with the Dean's List interviews and the addition of multiple-event Chairman's presentations. Again, this is something that we'll be better prepared for next year.

If you have additional feedback specific to this process, please PM me and let me know. I'm working with our Michigan volunteers to provide feedback to FIRST about how to make the Dean's List process stronger for next year, and would like to hear perspectives from other members of the FIRST community.

[RonnieS]

Good point. Maybe time for first to make their own webcast hub.

[Amo10]

Originally Posted by dcarr

In 2014, getting a quality webcast is apparently a hard thing to do.

Except it's not. Please fix this. Baseline HD stream requirements for all events. Include equipment/connectivity/staffing in budget or find a team/volunteers who can handle it.

Multi-cam setups are ideal for the big screens at the events, but for online, a single full-field view is optimal. Perhaps offer both?

Please. Or at least make all of the streams mobile friendly not just the Nasa TV one which cut off in the middle of the most exciting part on Einstein...

[Chris_Ely]

Originally Posted by Ronnie314

In all reality, they could purchase go pros and the poles with a decent capture card and stream through twitch. You can have HD and with the fish eye we actually get a good view. Now add input to show scoring and it would be all set.

And input the event audio instead of the camera's mic. The FiM steams have good video, but the audio is just terrible.

[Koko Ed]

Originally Posted by luckof13

And input the event audio instead of the camera's mic. The FiM steams have good video, but the audio is just terrible.

Considering that lawyers troll Youtube and have had some of the videos taken down for copyright infractions this might be a good thing.

[Chris_Ely]

Originally Posted by Koko Ed

Considering that lawyers troll Youtube and have had some of the videos taken down for copyright infractions this might be a good thing.

Maybe just the MC'c mics then? So that people watching know what is going on.

[fresh_prince]

Originally Posted by Ronnie314

In all reality, they could purchase go pros and the poles with a decent capture card and stream through twitch. You can have HD and with the fish eye we actually get a good view. Now add input to show scoring and it would be all set.

This is exactly what we did at the Lone Star regional. Go Pro on a painter's pole into a capture card, overlay from scoring table into a USB adapter, and open source broadcast software.

[TheMadCADer]

Originally Posted by Pault

This game was, quite honestly, too hard. They made it so that if you wanted to put up a decent score, all three robots on an alliance had to be able to manipulate a 2ft diameter ball reasonably well. And one of the robots had to be able to launch it a good distance in order to make the match very interesting. Those are expectations which many teams just weren't ready for, and I don't think they will be for at least another few years.

Going a bit more generally, there was not enough forethought in the designing of this year's game and game manual. The examples of this have already been beaten to death, but we need to find a solution so that this doesn't happen again. Something needs to change about how the game is designed/evaluated. Maybe bringing some people with more experience on teams into the GDC, or maybe just bringing them in a couple of times when the game is in its final stages to point out the flaws. Some sort of simulation for how the game will go down would be great, but I'm not sure how they would implement it. I'm not sure what it would be, but something needs to happen.

A few people have said this, and I think it's a great idea: FIRST should bring in the various 72 hour robot groups to beta test the game. Give the GDC some very useful feedback from some very experienced people, and give the 72 hour robot teams time to produce nice polished videos, preferably about the brainstorming and prototyping process. Most teams struggle with those parts of the design process.

[StillDefective]

I agree with the rest of you on the point of game play. I am a member on an FTC team, so I am not very intimate with AA, but I went to watch some matches at the STL regional, and honestly, even the finals matches weren't all that exciting. We left after watching about six of them.

But on the other hand, we had nothing to do before the finale, so our team watched the Einstein matches, and that was probably one of the most exciting things I've ever seen. Watching 254's 3 ball autonomous work flawlessly was an inspiration within itself. I was on the edge of my seat and yelling the whole time.

This game only works at high levels where everyone is capable of doing most everything on the field, but when that happens. God it's inspirational.

[Richard Wallace]

Originally Posted by TheMadCADer

A few people have said this, and I think it's a great idea: FIRST should bring in the various 72 hour robot groups to beta test the game. Give the GDC some very useful feedback from some very experienced people, and give the 72 hour robot teams time to produce nice polished videos, preferably about the brainstorming and prototyping process. Most teams struggle with those parts of the design process.

Early in my career I had a boss who liked to come into the lab and try to break the stuff that his engineers were working on -- he would drive our motors and circuits harder than we expected, to see how robust they were. Often he would find failure modes we had not anticipated.

FIRST could find some people like that in the FRC community, independent of the GDC, to try and break the game. I think Paul Copioli has already volunteered to do that for free.

[Andy A.]

Originally Posted by Richard Wallace

FIRST could find some people like that in the FRC community, independent of the GDC, to try and break the game. I think Paul Copioli has already volunteered to do that for free.

I've thought for years that the GDC should release the game to a few groups with prior FRC experience to develop the same questions that all teams do in the first two or three days of meetings.

At least half the Q&A issues could be addressed prior to release that way, and teams could get the manual as it would normally stand after the first update or two. It just stuns me that they released this years game with wording that made it possible to do stuff like truss and catch your own ball. How was that ever not identified as something every single team would identify as a possible sneaky tactic?

I think the GDC is just too close to the game as they develop it and can't/don't think of all the ways that teams will come up with to break it.

[dodar]

Originally Posted by jimbo493

I wouldn't stream thru Twitch though, too many ads...

Twitch doesnt run ads, the streamer chooses to run the ads. Twitch is the perfect place to stream events.

[Cel Skeggs]

While sitting in the last rows of the first level of the seating on Einstein, the lights illuminating the audience pointed directly at our eyes, which was painful. I'm hoping that FIRST will choose a different method, of any, to accomplish the same task next year.

[Alan Anderson]

Crowd control could use a couple of dedicated traffic directors at specific spots in the pits. The robot paths to and from the fields were clearly marked, but the clear markings were on the floor and often couldn't be seen through people standing or walking nearby. FTC teams in particular kept wheeling their robots through the pedestrian walkway along the south wall.

[Koko Ed]

Originally Posted by Alan Anderson

Crowd control could use a couple of dedicated traffic directors at specific spots in the pits. The robot paths to and from the fields were clearly marked, but the clear markings were on the floor and often couldn't be seen through people standing or walking nearby. FTC teams in particular kept wheeling their robots through the pedestrian walkway along the south wall.

FTC won't be a traffic problem next year but additional big robots in the tunnel will cause gridlock.