Friday, January 31, 2014

Day 27: More Power + Collection Testing

We received our 4 MiniCIMs from West Coast Products today. VexPro was out of stock but R.C. over at WCP was able to get us what we needed. They also have great calculators on their resources page to help you figure out belt and gear spacing or calculate your drive trains final speed.

We mounted the 4 MiniCIMs and added 4 talons to our practice bot so we now have 8 motors on our drive train. We didn't do much driving, but from the little testing we did the acceleration is much nicer now and feels a lot more like our 6 CIM drive from last season.

Our collector testing is still underway; we found some bugs last night that needed to be corrected and we worked on that for most of the day. Here are some clips from tonight's tests. We still don't have all the variables right but it's collecting much better now.


We found a problem with our pneumatic wheel switching at the end of the meeting. For some reason the modules aren't keeping themselves off the ground like they should be in traction mode. We're using 4 1.0625" bore pneumatic cylinders at 60psi, that should be enough to hold up to about 200lbs but that math doesn't seem to be fitting reality. It probably means we didn't account for something in our math. We'll investigate more tomorrow.

Parts were dropped off at Royalty Metal to be powder coated and should be back sometime next week.

We have also been working on getting an IMU setup to give us orientation of robot at all times. We had the first successful serial transfer today of quality orientation data. The IMU has 9 axis of measurement so it is able to reduce the amount of error that you would see from just using a standard gyro.

- Spectrum

“It's not what you don't know that kills you, it's what you know for sure that ain't true." - Mark Twain

Wednesday, January 29, 2014

Day 26: It looks like a robot

Lots of robot pictures today. We got the back panel mostly constructed and the collector mounted. The robot in this configuration weights 117lbs with the battery. That means we should still have weight for MiniCIMs and the catching wings on the competition robot.







We did get a little bit of testing done on the collector and shooter but nothing really video-worthy. Hopefully over the next few days we'll have full system integration and testing video to upload. We also got the competition parts cleaned up and ready to be taken of to Royalty Metals for powder coat tomorrow.

- Spectrum

“I'm a greater believer in luck, and I find the harder I work the more I have of it” - Thomas Jefferson

Day 25: Sheetmetal Everywhere

We just got our second run of laser cut parts back from our sponsor's facility. One of our students was there till almost midnight getting everything cut and bent for the team. In total of 125 individual parts were produced last night.


We had another "ice" day today so there was no school, but since the roads were clear and the storm advisory was over at noon we decided to have an optional meeting.

We were able to get a lot accomplished since we got the sheet metal back and had things to assemble.



Above are pictures of our right collector and our launcher + back panel for our robot. These were both completely CADed and have a lot of details in them.

The collector is mounted on the practice robot and we were able to do a few stationary tests, but the real test will be tomorrow when we can drive and collect at the same time into our launcher. The launcher isn't complete because we are missing a few parts that will be in later this week and a few 3D printed parts that we need to make, but it is in a state that we can test fire it and we should have some video of it mounted and firing tomorrow.

We cut all the parts for the competition robot yesterday as well, and we are confident in the launcher design and we know that we can make adjustments to the collector quickly. We will be taking all the competition parts to Royalty Metal on Thursday for powder coating.

- Spectrum


“True happiness comes from the joy of deeds well done, the zest of creating things new.” - Antoine de Saint-Exupery”

Monday, January 27, 2014

Day 24: Octocanum on the ground

Sorry for not having a post yesterday, we were hanging out with a few young teams at the Houston Bootcamp build day round 2.We were working on getting their electrical systems worked out and finalizing their build plan for the last half of the season. We'll have some of them over to our shop this weekend to work on bumpers and to continue improving their robots.

The Mecanum wheels finally came in today. We were able to get them on our chassis and we had them driving around.  The code still isn't 100 percent and we don't have driver/field-centric control yet, but once we do we think this drive train is going to be very fun to drive.



We also had one of our senior students, Collin, working on laser cutting and bending our next round of parts. Here is a video of him bending one of the pieces for our collector.


Tomorrow we should have parts back, and we have already talked to our powder coat sponsor about getting the parts in to them this week. If all goes to plan we might have two running robots at the end of week 5. (Though by saying that out aloud we've all but guaranteed that everything will not go to plan.)

- Spectrum

“Spoon feeding in the long run teaches us nothing but the shape of the spoon.” - E.M. Forster

Sunday, January 26, 2014

Day 22: Zoom Zoom

Here is the the video from our Butterfly drive test. This looks really good and is just a lot of fun to drive. If the omnidirectional drive doesn't work out we will very quickly switch back to butterfly.


We finished up the drawings for our launcher and our collector. Every part that we cut at Solarcraft has to have a drawing that goes with it so that you know how to bend the part after it has been cut. Here is what one of the drawings looks like. They outline how long the flanges are on the sheet metal so that we can program the press to bend the part at exactly the right spot.


Another very important system was prototyped today: handles. We got this idea from team 610 in this CD post. All it takes is some steel cable, a couple cable crimps, and pneumatic tubing. A lot teams probably already have this stuff in their shop. Crimp a loop in the cable then slide over the tubing and crimp a loop in the other end. Rivet/bolt this to your chassis and now you have convenient light handles that can just fold into your frame perimeter. This should work really well on the AM14U Chassis, put one bolt on the outside plate and one on the inside plate. Make one for each corner of your robot. Your drive team will thank you a lot for having handles, and you lower the risk of someone doping your robot.


Speaking of team 610, one of their students, Ryan Tam, put together an awesome design tutorial and parts library. If you ever wondered how the World Champions train their team to design high-quality, reliable robots, now's your chance to find out. We highly recommend taking some time to read over this; our team will be using this to help train new members in the future.

- Spectrum

“You can't use up creativity. The more you use, the more you have.”  - Maya Angelou

Saturday, January 25, 2014

Day 21: Ice Day low of 28 degrees in Houston

We had an ice day today so no school. Many of the Houston overpasses were iced over this morning. This only happens every 2 or 3 years, but when it does we overreact a bit, especially compared to what most people from the northern half of the country deal with for several months out of the year. We still met after school since by then the roads had thawed out and much of the traffic had gone down.

We learned that our Mecanum wheels wouldn't be getting to us till Monday so we decided to put the drive train together using Omni wheels instead for a Butterfly type drive train. We definitively understand why some of the powerhouse teams (148, 217, etc) have used this drive train in the past. It is so smooth and practically anyone can drive it well. Hopefully when we get drive centric control working with the mecanum wheels we will like that even more.

We'll have video up tomorrow of the robot driving around. Here's the latest CAD file, and you can see the collection mechanism CAD is coming along nicely.



- Spectrum

"When given a choice, choose the thing that scares you a little. If it's 100% safe, it is holding you back." - Jeff Atwood




Thursday, January 23, 2014

Day 20: Strategy and Tactics

Lots of CAD work today and we worked out some dimensions for our wings that will help us catch the ball and also maintain position of the ball. Not to much to show but instead we wrote down some of the strategy and tactical ideas that the team has been kicking around the past couple weeks.

Questions we're asking ourselves
What percentage of teams will be able to catch a pass over the truss?
How can we make it easier for teams to catch our passes?
How can we avoid defense on our passes?
How can we speed up the truss/catch process?
What killer features will the best robots have?
How can we be better at rebounding? / How can our alliance be better at rebounding?
How can we play smart defense when we don't have the ball?
How can our alliance best utilize its time?
What's the best way to communicate within our alliance?

Some possible alliances in Aerial Assist

Standard FMI
Forward/Attacker/Striker - Last with the ball. Scores in the high goal. Possess in the alliance colored zone. Hopefully can catch a pass from the midfielder.
Midfielder - The ball handler. Takes the ball from the inbounder and is normally the robot that trusses. Should have consistent and accurate passing options and the ability to quickly acquire the ball from another robot. Possess the ball in the white zone.
Inbounder/Defender/Back - Takes the inbound pass from the human player. Hopefully can catch the pass directly but also could take the ball off the carpet. Extremely important to cycle times, must be able to secure the ball quickly and get it to the next robot as fast as possible. Should play smart defense against the opposing teams. Forward once they don't have the ball. Possess the ball in the opposing alliances zone.

This alliance would play the game in a very smooth motion, with all three robots playing critical roles in amassing the most points possible for the team. These positions are possibly playable by many of the same robots so teams can be flexible in which position they play in a given match. The midfielder robot has a lot of options in this alliance; once they pass the ball they can either help the Forward by playing Full Back or retreat and help the Back slow up the opposition alliance.

Super Goalie + 2 Forwards
Super Goalie - This goalie robot is able to play lock down defense against the opposing alliance while also taking the inbound passes and goalie kicking them over the truss to one of its two alliance forwards. 

This requires a very unique robot, but it could give a big advantage the alliance that can pull off this formation. The super goalie is able to move the ball half way down the field in the air and has the potential to throw to two different robots making it much harder for the defending alliance to predict the location of the throw. The forwards will have to learn to pass to each other if they want to maximize their cycle points. 
A similar system could be played by the standard alliance if their Back robot has the ability to truss the ball at all.

Super Star + Double Back
Super Star- Take the inbounder's pass, throws to itself over the truss, & scores in the high goal.

This alliance will definitely exist at events, especially in matches where alliances might be shorthanded and only playing with two robots. If the Super Star robot can truss and high goal 3 times as fast as the opposing alliance can cycle they are guaranteed to at least keep pace with the other alliance. That's no easy task but when the other alliance has to deal with two defending robots it might be possible.

There will be plenty of other alliance structures, post your ideas in the comments.

- Spectrum

"Tactics mean doing what you can with what you have." - Saul Alinsky

Day 19: Launcher Video and More CAD

We finally have some good video of our launcher prototype. It's performing as we expected, with a wide range of potential shots, some of them we didn't even think about as we were designing it. Again this was inspired by Boom Done and we want to thank them for such a great idea.




Our original chassis that will be used for our practice bot weighed in at 19.84lbs. It was made from 1/8" aluminum and didn't have much weight reduction.


The new chassis will be made from .09" thick Aluminum and will weight in at 11.84lbs, It is sliced and diced to cut weight. We know this will add a lot of chassis flex but since we are using Mecanum wheels we're alright with that.

We're also working on the lightening pattern for the back of the robot. Put your opinion in the comments.

 Option 1: Standard Diamond Cutouts

 Option 2: Spectrum Cutouts Standard

Option 3: Spectrum Cutouts with Reverse

We should be driving this weekend, with or without Mecanum wheels, as the practice robot is just about finished being wired. Our goal is to finish the CAD for all the mechanisms over the next two days and have drawings ready to go to Solarcraft to be cut and bent early next week.

- Spectrum

"Whenever I tackled the impossible or the miraculous, I remembered the magician Rene Lavand, who had only one arm. Poet and extraordinary card manipulator, he baffled fellow illusionists by concluding his brilliant demonstrations with, "What I just showed you can also be done with two hands!" - Philippe Petit, Man on Wire, 2002, p. 236.

Wednesday, January 22, 2014

Day 18: Parts Arrive

After the long weekend we got many of our orders from week two. We had a large amounts of screws, bolts, and nuts from McMaster-Carr and Bolt Depot. We received our encoders and IMU from Sparkfun, and we also got an AndyMark order which contained part of our PDV--it consisted of Talons and stealth wheels.

We also received several VEXpro packages. We finally received our hex bearings that we have been waiting on, and we also received our VEXpro PDV items along with two 2x1 VersaFrame rails. The biggest surprise was the VEXpro order that we placed late Sunday night arrived today; we ordered parts for the real robot already since we expected a week or long lead time, but they got them to us in well under 48 hours. VEX is really stepping up to make everything right with their customers. We're still waiting on 4" Mecanum wheels, but that's our last back ordered item.

This might look like the same picture from yesterday but several hours have gone into improving it since then. Many of the items are very subtle but will make a huge difference when we actually start assembling it. We started the process of changing our CAD model for the drive train from .125 to .09, it takes some time to make sure all the spacing stays the same. We also worked on the second revision of our collector, which hasn't been produced yet, but we are already finding ways to improve it.

The launcher had a very successful test and we should have more to post after we get some video during tomorrow's meeting. (Hint: 2 CIMs and 2 BB775s is 1200W)

- Spectrum

"We are all salesmen every day of our lives. We are selling our ideas, our plans, our enthusiasms to those with whom we come in contact." - Charles Schwab

Tuesday, January 21, 2014

Day 17: Another Long Work Day

We were off from school for MLK so we had another long work day.

We improved our pneumatic launcher to where it give a nice chest pass. We aren't trying to make this prototype score in the goal or go over the truss, we are going for extremely compact.


Along with that we continued the CAD for our launcher and pickup mechanism and continued development of the prototype launcher to better fine tune it.


We heard from VEX that our items should be shipping this week so hopefully we will be driving this weekend.

- Spectrum

“Action expresses priorities.” - Mahatma Gandhi


Monday, January 20, 2014

Day 16: Mechanism CADing

We worked more on CADing up the game specific mechanism for our robot. Most of it is still in draft form but we're pretty happy with the direction it is going.
Note: The flappy things have changed drastically in the last few days, and will continue to change. 
Note 2: There is also a second collector pole, it is just not shown in the picture.

We still have some prototype work to do on both the collector and launcher before we are comfortable sending them to the laser, but we think we are getting pretty close.

We're building this with the possibility of up to 14-16 motors. There aren't that many functions, but since the current rules allow for so much power, if you build it in it allows your motors to work in a much more efficient part of their power curve.

Hopefully we will be driving our practice chassis by this weekend.

- Spectrum

“One option is to struggle to be heard whenever you're in the room... Another is to be the sort of person who is missed when you're not. The first involves making noise. The second involves making a difference.” - Seth Godin

Sunday, January 19, 2014

Day 15: Robots and Rookies

 We put together as much of the chassis as we could. This one ways about 50lbs as shown here. The real one will be lighter as some of the main components will be .09 aluminum instead of .125 and we will have more lightening holes.

We pressed in PEM nuts using a hammer and a punch. We're not sure how long these will hold but we're hoping for minor things like speed controllers this should be enough.


Today we hosted team 5287 from Community Christian School in Orange, TX. We helped them put together a model of Ri3D Boom Done's El Toro Collector using scrap material that we had. There students will be able to build their own version now that they've better seen how it's done.


We talked a lot more about our launcher mechanism and how we should build supports for it. We're pretty happy with the concept we have going and should be able to get CAD versions of it put together pretty quickly.

- Spectrum

“The thing I remember best about successful people I've met all through the years is their obvious delight in what they're doing and it seems to have very little to do with worldly success. They just love what they're doing, and they love it in front of others.”  - Fred Rogers

Saturday, January 18, 2014

Day 14: Two Weeks Down

After the sheet metal came in today, we started construction on the practice robot.

We set up a test version of our octocanum modules.


The drive train will hopefully be finished early next week once our Mecanum wheels and hex bearings come in from VEXpro.



This weekend will be a big weekend for us. We should finalize our collector CAD and also get much closer to a final design for our launcher.
\
- Spectrum
\
“You do not really understand something unless you can explain it to your grandmother.” - Albert Einstein

Thursday, January 16, 2014

Day 13: Sheet Metal

We finished the CAD for our chassis and a few other parts yesterday and one of our seniors who interned with our sponsor Solarcraft was able to go to their facility and cut and bend our chassis tonight.


Collin with the chassis he just cut and bent


We worked on a few more prototypes including a pneumatic catapult and a wheeled shooter. The motor-driven catapult is still the most promising but we have some ideas for the others that may make them win out in the end.

We will have most of the items we need to make our drive train tomorrow afternoon except for a few pieces from VEXpro. This weekend should be full of assembly and more prototype testing; we are working on gathering as much information about the prototypes as we can so that when we build the real version we can know it will work well.

- Spectrum

“For me, I am driven by two main philosophies: know more today about the world than I knew yesterday and lessen the suffering of others. You'd be surprised how far that gets you.” - Neil deGrasse Tyson

Wednesday, January 15, 2014

Day 12: Trussing

Tonight we finally started having success with one of our shooting mechanisms. We were able to score in the high goal and truss with this shooter.

This shooter is heavily inspired by Dr. Joe Johnson and the Boom Done Robot in 3 Days team. It currently is running 3 CIM motors in a VEX single speed (5.33:1), a 15t to 30t sprocket reduction, and a final 12t to 22t sprocket reduction. These are just the items we had on hand, and we will hopefully be optimizing this over the coming weeks.

We have been working on this for over a week now, and we strongly believe that having a shooter with multiple shot trajectories will be key to making good clean passes over the truss.

Our chassis is going to be laser cut and bent tomorrow. Our sponsors had a big project that they had to get finished tonight so we couldn't get any time on the laser, so tomorrow one of our students will go to Solarcraft and run the laser and press brake to make all of our parts.

We are finalizing the "Sipping Bird" collector. It's based off this version, but now all of the motors will be contained inside the robot and both roller assemblies are independent.

We still have a few other shooter prototypes to test before we make our final decision.

- Spectrum

"It's always too early to quit." - Norman Vincent Peale

Tuesday, January 14, 2014

Day 11: Drive Train CAD Complete

We finalized the CAD for the drive train today. We are cutting a heavy version for our practice bot that doesn't have a lot of holes that will be in the real version, which will save us some time. Here is the belly pan. It has mounting point for 16 victors/talons, and although we won't be using all of them, it's nice to know we have them.

The prototype shooters still need some work but we have started designing the final version of our intake. It's heavily based on the intake rollers from VEX robots. We also will be able to outtake the ball from its stored position using only the rollers so we can score in the one point goal.

We started construction on our bumpers yesterday. This year we are going to make our bumpers early and we are spending time to make sure they are done very well. Last year we had some issues with the pool noodles sagging a bit after a few competitions, but we don't think that will be a problem this year.

- Spectrum

“Little by little we human beings are confronted with situations that give us more and more clues that we are not perfect. ” - Fred Rogers

Monday, January 13, 2014

Day 10: Back to Blogging

Sorry we missed a couple days. We had a Bootcamp build day yesterday where we helped 5 teams get their AM14U up and working.

We were back to work today, we have several prototypes still in development and we haven't finalized our robot design.

We worked through the electrical and pneumatic design, and we should be laser cutting on Wednesday after we do all the drawings tomorrow night.


- Spectrum

"Even a mistake may turn out to be the one thing necessary to a worthwhile achievement." - Henry Ford

Friday, January 10, 2014

Day 7: It's been a week

We continued improving our prototypes today after our design meeting yesterday. We modified our collector prototype to a version inspired by one of our VEX robots from this year.






          We also continued to improve our shooter/thrower prototypes. Lately we have been working on a thrower that runs off two mini CIMs and throws the ball over the back of our robot. So far we have been trying to improve the range.


-Spectrum

"Success consists of going from failure to failure without loss of enthusiasm." Winston Churchill

Day 6: Need, Want, Wish...

Today was our first design review. We have seen the 72 hour builds, prototyped a little, and  played a human version of Aerial Assist. We think we have a good understanding of the game. Here is how we break it down.

Need:
  • Drive
  • Speed
  • Roll ball out of robot on ground or into low goal
  • Sturdy/Rugged
  • Catch from inbounder / Collect from Human Player / Receive Inbound
  • Hold the ball - securely enough to drive around


Want
  • Pushing Power / Torque
  • Agility
  • Bouncing Ball Control (rebound)
  • Catch well from other robots
  • Shoot a Pass/Lob Shot
  • Variable Power Shooter
  • Ground pickup
  • Lob to Human Player
  • High Goal Score


Wish
  • Strafe
  • Shoot across a zone (lob from back over truss)
  • Shoot from zone line (18' shot)
  • Shoot from the wall
  • 2 ball auton
  • Eject ball on power off


Why bother?

  • Brake (mechanical)
  • >5' extension...
  • Jump
  • Turret
  • Scissor lift...
  • Ground pickup from sides
  • 3 ball auton...

There might be more aspects to the game but these are all the ones we covered. We'll be dealing with the programming side of things a bit later.

Today's update is a picture of the current state of our CAD Model. It's still rough, but the drive train is coming along very nicely. We are working on the electronics layout, as well as fitting everything else into the robot. (All the CIMs are just place holders, there will only be 6 on the real robot.)


- Spectrum

"Start by doing what's necessary, then what's possible, and suddenly you are doing the impossible." – St. Francis of Assisi

Wednesday, January 8, 2014

Day 5: Prototyping doesn't stop & Square Kit Chassis

Quick update today, we still have a lot of things to prototype before we finalize a design. We're hoping to have a chassis built by next week.



We have a Bootcamp Build day for some of the young teams in the area on Sunday. While working on the CAD files for the AM14U, we mocked up a version that is square 27.5 x 27.5. This was mentioned on Chiefdelphi, but we weren't sure how easy it would be for teams to make. We think it's pretty simple--it's really just making a half-wide/half-narrow robot.


The front and back pieces  are both cut to 20.25".

This square version has a non-symmetrical wheel base but it should still work very well. It also allows the 25" ball to easily fit inside of it. Add a plywood base to the inside and you'll have a very nice chassis for Aerial Assist.

- Spectrum

"My attitude is never to be satisfied, never enough, never". - Duke Ellington


Tuesday, January 7, 2014

Day 4: Inspiration From Other Teams

We have been putting in some serious hours CADing away at the drive train. It may look simple from the screenshot below but there is going to be a lot going on once it's all finished. The modules are heavily inspired by Aren Hill and Team Neutrino's Butterfly Drive from 2013. They were our pit neighbors at IRI and gave us an awesome tour of the robot.


We broke off into a few groups to design some new prototypes today. Started working on a CAM powered launcher. We wanted a CAM profile and weren't sure where to get a quick one to test with, but then we remembered our friends 148 used  CAM on their 2010 robot, Armadillo. The CAD for that robot can be found on FRCDesigns.com. We printed out a 1:1 version of it and cut it out of plywood on the bandsaw. Hopefully we will get something going in the next day or two.


While looking for other collector options to test in parallel with our roller claw development, we worked on a version of 118's Ballacuda pickup system from their 2008 robot.


We don't plan to use any of these systems verbatim, but starting from where someone else has already been allows us to develop solutions much faster. Everything still has to be tested, modified, and tuned for this years game requirements.

- Spectrum

“Stop waiting for something to happen or for someone to give you permission to do things.” - Kate Kendall

Day 3 of Aerial Assist

We continued our prototyping today.
The improved roller claw is working very nicely. We're not convinced that we will even use a roller on our final design but we are still working on improving it as much as possible. We are tuning it to find out as much about the different variables as we can. We want to know what is the optimal roller height, roller distance past the bumper, height of the back bar, downward force to press on the ball, etc. All of these little details will make the final design that much better.


We are pretty far along in CADing our drive train for this year. Here is a quick screenshot from our progress.
We have posted some of the details behind this design in our 2014 Design Notebook. Our goals is to make it pretty flexible. We believe that catching may turn into an important factor in this game and having the ability to move omni-directionally will make that task much simpler. We also know that there will be times that we need to stand our ground or attempt to move another robot for this reason we are planning to use an Octacanum drive train that can shift between mecanum and traction wheels. We are building in contingency for allowing us to switch from octacanum to a butterfly drive with standard omniwheels replacing the mecanum wheels and also a standard six wheel skid steer drive if we desire.

Several of our other prototypes are still under construction and not ready for even simple tests, hopefully by the end of the week we will have a solid idea of where we want to go with the robot.

- Spectrum

“Do not follow where the path may lead. Go instead where there is no path and leave a trail.” - Ralph Waldo Emerson

Monday, January 6, 2014

Day 2: Prototyping really begins

Today we continued our prototyping in earnest. We had members working on several different launching and collection mechanisms.

Here is our most well developed ball pickup device. It's a simple over the top roller claw using VEXpro and Banebots parts.





Another of our prototypes is the ball puncher. It uses surgical tubing to pull back a shaft and punch the ball. The goal is to use a mechanism like this to punch the ball over the truss and also into the high goal.


As you you can see this isn't working exactly as we would like but we learned a lot from it. We'll be working on a more traditional catapult tomorrow afternoon.

Here is a sneak peak at another prototype that we should be finishing up tomorrow.

Hopefully everyone is watching all of the great content coming out of RobotIn3Days.com and BuildBlitz.com. They are putting out great content to help teams figure out what they want to build, but remember that watching videos and seeing pictures can't replace good hard prototyping, plus it's fun.

- Spectrum

“All great achievements require time.” - Maya Angelou


Sunday, January 5, 2014

Aerial Assist Day 1

Day 1 of Aerial Assist went very well. Here is the animation for anyone that missed it this morning.


We worked all day learning the rules and getting familiar with all the ways to play the game. Here is our document with some of our notes about the game.

After we got some quick mock ups of the field elements built, we headed to the gym for a brief little session of Human Aerial Assist. 


We tested low goal scoring as well, and it seems like all robots should be able to score in the low goal.



We started prototyping briefly, but not much to see yet. It's going to be the start of a ball launcher built using several pieces from the kit of parts.


Lots more to do before the first weekend is over.

- Spectrum

“I have nothing but applause for those brave enough to fail, and fail again.” - Seth Godin