[FIRST EMAIL] Stop Build Day Survey

[FrankJ]

I am sort of on the fence about bag and tag. I like the hard stop. It is more of a transition for us since we continue to work with the practice robot. We focus more on driver development and robot refinement after stop build. On the other hand stop build has a lot of baggage that already been discussed on this thread. It is probably time to change it.

One of the thus far unstated advantages that will benefit the high resource teams is they will be showing up to each event with an essentially completely rebuilt robot rather than having to do maintenance at the event. Granted other teams will have the same opportunity.

For no bag day to really benefit mid level teams, they are going to need the discipline to finish the robot around the six week mark to give them time to iterate and practice. I leave the top tier teams out of this because they already have the discipline.

Teams have to decide how much resources to put to the robot competition. Not having bag day really doesn't change this.

[Ernst]

Originally Posted by waialua359

And IMO, using VEX is a poor example of why we should get rid of the 6 week build season! Building a completely different VEX robot between events is NOT the same as bringing a somewhat different iterated FRC robot to your next event. Not even close! We spend a lot of time doing both.

This is a point I'm really stuck on. I hate the design convergence in VEX. I hate that teams, if they're driven enough, can completely scrap their robot between events and copy the most successful design they've found online. My VexU team is about to start our 2nd 9-week Release schedule for a Fall Scrimmage. We'll have another one before our Spring Qualifier. And another one before Champs. We'll have completed 4 separate build seasons. In the process we'll probably lose half of our active members again.

Obviously FRC is more complicated and it would be a lot harder to pull off copying a Week 1 robot that you see for your Week 6 or 7 event. Design convergence with some subsytems definitely currently happens by the end of the season, but those are mostly add-on subsystems, not a defining part of the robot.

[Michael Corsetto]

Originally Posted by Ernst

This is a point I'm really stuck on. I hate the design convergence in VEX. I hate that teams, if they're driven enough, can completely scrap their robot between events and copy the most successful design they've found online. My VexU team is about to start our 2nd 9-week Release schedule for a Fall Scrimmage. We'll have another one before our Spring Qualifier. And another one before Champs. We'll have completed 4 separate build seasons. In the process we'll probably lose half of our active members again.

Obviously FRC is more complicated and it would be a lot harder to pull off copying a Week 1 robot that you see for your Week 6 or 7 event. Design convergence with some subsytems definitely currently happens by the end of the season, but those are mostly add-on subsystems, not a defining part of the robot.

You must really have hated 973's 2015 and 2016 robots...

What you wrote already happens. 1678 in 2013 is another example.

-Mike

Edit: I suppose my point hinges on how you specify a "defining part of the robot"

[marshall]

Originally Posted by Michael Corsetto

You must really have hated 973's 2015 and 2016 robots...

What you wrote already happens. 1678 in 2013 is another example.

-Mike

Edit: I suppose my point hinges on how you specify a "defining part of the robot"

But design convergence doesn't happen in the real world!

Except for cell phones, computers, chip manufacturing, cars, shoes.... Kicked the ball in my own goal again!

[bobbysq]

It is worth pointing out that off-season events show that robot diversity is still maintained even when there is less of a time limit. We don't walk into CowTown and see a bunch of 254 and 118 clones every year. I think teams would still rather put out their own work, even if it's not necessarily the winning option.

[Jay O'Donnell]

Originally Posted by bobbysq

It is worth pointing out that off-season events show that robot diversity is still maintained even when there is less of a time limit. We don't walk into CowTown and see a bunch of 254 and 118 clones every year. I think teams would still rather put out their own work, even if it's not necessarily the winning option.

That's probably because teams don't want to spend a lot of money and time on a robot for the offseason. It might be a little bit different with a championship on the line.

That being said there's already a bit of design convergence with withholding allowance. See canburgulars last year.

[Deke]

One thing I haven't seen mentioned was stress on machining sponsors. Eliminating bag and tag would reduce stress on machine sponsors as well.

The team I helped with had access to manual mills and lathes and made 90-95% of the custom components needed, but some complicated parts would get cnced, laser cut/water jetted at a sponsors shop. They were willing to set aside profitable work to give us fast turn around times. Most shops in the area would give 2-4 week lead times for normal quoted work. Giving a longer time table for the sponsors to finish parts during machine down time would save them money as well. We all know how critical sponsors are to the success of FIRST, giving them some more time can only help them.

[Ty Tremblay]

Originally Posted by Ernst

This is a point I'm really stuck on. I hate the design convergence in VEX. I hate that teams, if they're driven enough, can completely scrap their robot between events and copy the most successful design they've found online. My VexU team is about to start our 2nd 9-week Release schedule for a Fall Scrimmage. We'll have another one before our Spring Qualifier. And another one before Champs. We'll have completed 4 separate build seasons. In the process we'll probably lose half of our active members again.

Obviously FRC is more complicated and it would be a lot harder to pull off copying a Week 1 robot that you see for your Week 6 or 7 event. Design convergence with some subsytems definitely currently happens by the end of the season, but those are mostly add-on subsystems, not a defining part of the robot.

Sometimes I get caught on this too, but as Dean has said time and time again, FIRST is about more than the competition. I would much rather see my students get inspired by a successful design, figure out why it works, make their own, and have it be successful at the next competition, than to have kids leave a competition saying "man, I wish we thought of that." While its true that the above can be accomplished within the confines of Stop Build Day, SBD makes this much harder.

I don't mentor in FRC solely to win competitions (319 didn't do that until 2016). I mentor in FRC because it inspired me to make a career out of robots. There's no denying that a robot that works is more inspiring to the students that built it than a robot that doesn't work. Removing bag day makes inspiration easier to achieve and I'm all for it.

[Chris is me]

FIRST design convergence will never, ever, in any circumstance, exist to the same extent that it does in VEX, for a few reasons:

• Vex uses standard hardware all around, making robots look similar and facilitating bolt for bolt copies.
• The Vex season is almost year round, allowing more time for copies and iteration.
• It takes far less time and far fewer resources to build a Vex robot than it does to build an FRC robot
• While Vex allows for some design flexibility, the COTS centric nature limits the number of highly successful and visible design variations to copy. Meanwhile, if you asked FRC teams everywhere which robot they should copy, I think you would get a lot of different answers.

This isn't to say design convergence isn't a cause for concern or anything, just that everyone's "nightmare scenario" of completely identical robots just won't happen. If it would happen, then we would see hundreds of Ri3D clones, right? Even in the peak year for clones, at worst we would see teams copy the concept with their own spin on it for the most part, with just a few bolt for bolt copies.

---

The other criticism of ending stop build I want to address is deadlines. Some have argued a hard deadline is good practice for the real world, so we need bag day to simulate that. The main problem with this logic is that, it's just completely backwards. Bag day is a soft deadline! You get to keep working on the withholding allowance. Even with no withholding, you get to use a practice robot, and plan for COTS upgrades at competition. A deadline of the actual competition day with a no-bag system would be an Actual Hard Deadline. Similarly, having everyone stop at the same time would also still happen at competitions. Everyone at the competition would have just as much time to work on the robots as everyone else at the competition!

---

I do recognize there are plenty of arguments to keep the bag, and I don't think they are all invalid. But I think once we get over our fear of change, it would do more good than harm to get rid of it. I'll post more thoughts at a later time.

[cbale2000]

Originally Posted by Hitchhiker 42

For now, I think removing stop build day will give too much of an advantage to district teams, who will be able to see how they perform before iterating it more and more, as compared to 1-regional teams who only get to play one event and can't really improve more. Because of this, removing stop build day will favor district teams even more when Champs rolls around.

Perhaps the result of this would be seeing the expansion of unofficial scrimmage events into the competition season? If you want more time to practice and iterate, there would be nothing stopping you from getting a few local teams together to practice like teams already do in Week 0 events, but in the middle of the season.

Plus, a lot of iteration can be done just by observing other events through livestreams. Not an ideal solution, but I still think having equal (unlimited) access time is a FAR better equalizer, even for teams that do only get one event.

[Jon Stratis]

Originally Posted by Chris is me

The other criticism of ending stop build I want to address is deadlines. Some have argued a hard deadline is good practice for the real world, so we need bag day to simulate that. The main problem with this logic is that, it's just completely backwards. Bag day is a soft deadline! You get to keep working on the withholding allowance. Even with no withholding, you get to use a practice robot, and plan for COTS upgrades at competition. A deadline of the actual competition day with a no-bag system would be an Actual Hard Deadline. Similarly, having everyone stop at the same time would also still happen at competitions. Everyone at the competition would have just as much time to work on the robots as everyone else at the competition!

My experience doing software design has mostly been along these soft-deadlines you mention. I worked for a medical device company for a long time. We would have a hard deadline for the *.0 releases, and have to get everything finished for them so we could send them to testing and then to the FDA on time. But once they were sent off to testing, we immediately started working on the *.1 release, fixing everything that was wrong (ie identified in testing) with the *.0 release. And ultimately, the users never saw the *.0 release, we were able to go straight to the field with the *.1 release instead.

You see this all the time with major software releases. Just look at the iPhone - version 9.0 came out Sept 16 last year, followed quickly by 9.0.1 a week later, and 9.0.2 a week after that.

So the way FIRST currently runs does align with some of what we see in the real world. It may not be applicable to every job or industry, but it is still applicable to some.

[MoistRobot]

I think getting rid of stop build day will be for the best. Teams can still set whatever build schedule works best for them and, as many have iterated, lower resource teams can practice/refine their robot up until competition time without the benefit of a practice robot.

[FrankJ]

I guess we need to look at why stop build day exists in the first place. Is it to limit resources (time) to balance the playing field? Hasn't worked. Is it intended to add an artificial dead line to increase the challenge? Hasn't worked. So it seems you either need to add rules to make stop build day more than a date on the calendar or get rid of it. Or recognize that within FRC there are teams competing at different levels. The driven teams are always put more energy into the game than less driven teams. Maybe their top level goals are different as well.

[notmattlythgoe]

Originally Posted by Michael Corsetto

This makes me sad. I can't think of a worse example to set for student than this.

Never heard of it happening before, besides on complete accident (rookie team that doesn't know to bag, etc.)

-Mike

Completely agree. This is like a coach cheating in little league baseball, what kind of example does that set?

People make me sad.

[efoote868]

I might be in the CD minority, but I appreciate stop build day. It's easier to tell members that "we expect you to participate during the 6 weeks we have to build the robot for competition" rather than "you have to participate all the way up to the first competition, and then maybe the second competition, and then maybe more until championship" etc.

Same thing for mentors, and other volunteers. It doesn't make a difference for the hard core enthusiasts, but it definitely makes it easier for the casual participant (the people not on CD in the off-season ) which is the majority of the FIRST population.