More testing, more asking for ideas, and a final prototype! Elly and I adjusted the design to a point where I really like how it is laid out. The plan is to start selling through gfxhax(.com) soon. Hopefully everything runs smoothly! I have to wait for parts to arrive and then figure out shipping. Other than that everything has been settled.
The progression of prototypes. Version to be sold is the bottom one.
I love the logo Elly designed for this project! (Well she designed it all and I love the overall design too!)
Prototypes for ATX Box
The laser arrived at the hive on thursday (6/2) and it is just as awesome as I thought it would be! I finally got to make some real prototypes of the box and what it will look like when I get to sell it.
There are a few small modifications that I need to work out. It’s mostly down to adding a website and logo, moving the binding post labels, add vents to the side, etc. The only major thing that needs to be fixed is how the atx plug is mounted. It isn’t nearly solid enough to actually use right now. I think I will have to add some grooves a couple of mm deep so it can rest deeper and then epoxy it in for a solid plug.
ATX Power Supply breakout
Getting further on my atx power supply breakout kit. Elly designed a 3d model for the case so it could be 3d printed for prototyping. I tweaked the design a bit and had it printed twice on a makerbot, but unfortunately both times the print failed about halfway through. This was still enough to provide a general idea for what it will look like and what I need to improve.
This is a really in depth way to split programming knowledge and experience into different levels.
Proper password security.
After being locked out of the online part of my banking I have been thinking that many corporate websites have password security wrong. The main threat that these websites are protecting against are dictionary attacks. While this is definitely necessary they are combining two different methods of protection that make it much harder for the user to login. They require a mix of symbols: numbers, letters, etc and also require that you get the correct password within a certain number of attempts (often five). The issue I have with this is that it is hard to remember a password that isn’t one you would normally use and isn’t natural to remember and a limited number of attempts.
There are a few options here that can greatly reduce the burden of remembering the correct password for the user. One is that requiring symbols could be removed, but when registering a user make sure they are not using an extremely common password (reverse dictionary attack?), or increase the number of login attempts to something like 20-30. Limiting login attempts for programs like ssh to a small number like 5 is fine, as it normally doesn’t completely lock the user out. The program normally just adds a wait period. However, with bank accounts they fully lock you out and you must go to the bank to sort everything out.
In my opinion they are adding a lot of burden on the user that doesn’t necessarily need to be there.
Why the makerbot sucks.
Sure the makerbot may be a cool tool to have, but in my opinion it sucks! The design of it wasn’t 100% thought out and the software driving it is wretched. It appears that the makerbot guys wrote the minimum line for what the software needed and then just left it up to the community to improve it from there. I think someone could make a decent amount of money by fixing up the software and then selling it to disgruntled makerbot customers.
The reason I gripe about this is I was using the makerbot at Hive13 (local hackerspace) and mid print the thing got caught or something and started printing into the middle of the air! So I have a half printed prototype that took an hour and twenty minutes.
Another thing I recently read about that blows my mind: They didn’t have PID control on the extruder until a while back when someone sent in a patch. This meant the temp of the extruder could vary by 30C+. That’s absolutely ridiculous and unprofessional in my opinion.
There is definitely a lot that could be improved with the makerbot, but it’s still a cool tool to play around with.
Adafruit likes my idea! Yay!
This is a tumblr for anything that passes my mind. (That’s the whole point of a tumblr, right?)