Competition Super Dongle

[Kyle Fenton]

Originally Posted by sanddrag

Full auxiliary channel access

I agree with you. There is a lot of crud on the 900mhz band, and to be able to switch to whatever channel works is sometimes invaluable.

[JVN]

No.
The communication protocol between the Arena Controller and the OI is proprietary and confidential. Knowledge of this would enable a team to manually switch to competition channels. This would COMPLETELY compromise the integrity of the FIRST competition.

The pins that have already been released are the only ones that perform any useful functionality. Everything else is controlled during the communication between the 2 processors via a serial link, and as such, is the protocol NOT for release.

Yes.
Someday, some FRC participant with too much time on his hands will “crack” the code, and reverse engineer the communication protocol. From that point forward, no matter how much we try, we may not be able to protect the integrity of the communication link between you and your robot. Although we continue to monitor communication with our scanners to protect competition integrity; we keep things proprietary for your protection.


JVN

[TonzOFun]

Originally Posted by Billfred

Besides, the dongle pinout's enough for me. Disable, autonomous. What more do you need?

We were hoping to completely duplicate competition control as much as possible. All you practically need is the pin-out, but we also want to provide power, radio and change team color. In other words, they would plug in competition and be provided everything that happens in a competition. It would be very useful for training new drivers.

[JamesBrown]

I still dont see why it would be necessary to have a "super dongle" the dongle we use is a box with three labled switches, Auto, Practice and Power. Team color an easily be done just go to radioshack and buy a couple of blue and a couple of red led's. The power would not be any more compact if it was a power supply to the competition port than if it was going to the regular pin for the power. It is impractical for IFI to release anything more than they have done.

If you want to simulate problema and you use less than 4 joysticks then get the pinout for these joysticks and make a switch box that would attach to the pins for yaxis x axis and all of the buttons. Then in the code p4_y could do something like cut power to the motors or rest the circuit breakers. If I remember there are 12 different inputs for each joystick, 4 for the hat 2 each for the x and y axis 2 buttons on top, one button by the thumb and the trigger. by wiring these to switches and assigning a binary number to each possible thing to go wrong you could have 2^12 or 4096 different problems. The software would see what switches were open and what was closed and then simulate the problem. Doing this now would also give you a nice set of switches to use as preprogrammed arm posistions or to use to reset pots in competition. This may be a more practicle use of your time than building a "super dongle"

[Mike]

Security through obscurity is never a good tactic. Look at a commonly used program that some of you may know, Windows. It's source code is not available to the public. Because of this, it is harder to write virii (because you can't see evident holes, you must find them.) Now look at an open-source operating system, Linux. You could see the code and find holes, but for some reason there are less virii for Linux than for Windows. (Yes, I know it may be because of less Linux users... but my point is still valid.)

If somebody wanted to go through the trouble of making a killer robot, finding the funding, etc. they will go through the trouble of figuring out IFI's pinout positions. Right now they are just penalizing teams that may want to use it for legitimate, helpful, purposes... such as TonzOfun.

Open the protocol/pin diagrams.

[Tristan Lall]

Originally Posted by Mike

Security through obscurity is never a good tactic. Look at a commonly used program that some of you may know, Windows. It's source code is not available to the public. Because of this, it is harder to write virii (because you can't see evident holes, you must find them.) Now look at an open-source operating system, Linux. You could see the code and find holes, but for some reason there are less virii for Linux than for Windows. (Yes, I know it may be because of less Linux users... but my point is still valid.)

Actually, it's very frequently a good tactic, so long as it's not the only tactic. Why make it easier for someone to compromise the system? Just make sure that if it's guarding something important, there are other measures to prevent unauthorized access.

As for the Windows-Linux analogy, I don't think that that really helps your point. You've pointed to a correlation, but haven't established causation—there are many more facets to that issue that should be considered.

By the way, it's not "virii", it's "viruses", both in English and Latin. Here's why. (Scroll down...the article tackles other, related issues first.)