Been doing a little gaming…
The Kindle worked perfectly fine but with an awful UI. The main use case is YouTube for a 2 y/o, and the stock launcher doesn’t do that well. Root and a custom launcher gets you there, so I did that.
The specific version of Fire I have was supposed to be tricky to root. There were forums of people bricking them, and the general workaround seemed to be reverting to an older firmware and then trying the root exploit…which seemed to be generally available as a windows executable. Not happening. I decided to give rooting a shot by pushing the su binary and Google apk set over via ADB on my Mac, which worked perfectly. Not sure what the issue had been. Play Store achieved…Nova Launcher installed. Happy 2 y/o.
The Asus was a different story…and the toughest root project I’ve had to tackle to date. The Transformer was a refurb from Woot, I think. I could look that up but I’m not going to. The combination of JellyBean and my particular firmware made any exploit attempts moot. I could not push via ADB, but the device acted like the files were invalid. Same method as pretty much anything else, but simply failing. I tried several methods I found online, but they all yielded the same outcome…except one really sketchy one.
I managed to use a really, really iffy Chinese root manager called 授权管理. The download and exploit were all in Chinese. Awesome. It worked, and I immediately replaced that with su. There was a good 10 minute window running who-knows-what. Sort of disconcerting. Bloatware wiped, and the tablet runs crazy fast finally. I’ve been checking my logs, and nothing fishy has popped up…looks like the hack worked.
Anyhow, two devices, each more usable. Go root.
The small lab assistant and I tackled the robot arm project last weekend. Fifty percent of team swan tron embarked upon the assembly phase wearing bunny slippers…
We did this in two big chunks, due to attention span limitations and nap purposes.
Testing Phase 1)
Testing Phase 2)
Emma figured out the remote pretty quickly, especially considering she has interfaced nearly exclusively with touch screen UIs. That may be next, in a round-about way. I’d like to stick some rPi brains in this and build a web UI, so the small assistant can run this with a phone or tablet. We’ll see where we end up with this.
At some point, I volunteered to make us some end tables for the house. It turns out that I really don’t like end tables…everything I came across was either way to big or too flimsy to take any kid abuse. The main level in our house is basically wide open, so keeping foot traffic lanes open is key. Long story short, I ended up in the copper area of Lowes and came up with a solution…
Maybe this is brilliant, or maybe my brain automatically looks for reasons to use my torch and Dremel. Probably the later.
Emma was willing to help with the stress test, but unwilling to put down the GBA…
This actually turned out fairly decent, especially considering the fact I did all of the woodworking on that chunk of poplar with a Dremel Multi-Max. My soldering skills are more tuned to electronics, so I burned the shit out of the bulkier joints due to lack of flux. Adds character, I guess.
I have another stick of 3/4″ copper, so plan on doing round two as a winter project. I keep thinking that some sort of electrolysis setup would be awesome for the copper cleaning phase, which could be a project in itself. This project would be a lot more intense with a car battery and a whole bunch of wire.
I stumbled across something potentially amazing…cheap Chinese LED modules:
For $35USD, I now have everything I need to assemble a 64×64 LED matrix, provided I stay off the coffee long enough to knock out a billion solder joints.
This should be fun, and potentially challenging. I’m guessing these were initially spec-ed out for a consumer with something pretty specific in mind, but going about this DIY is going in blind. There are no libraries to leverage, so the software will have to be written from the ground up.
Time to blow the dust off of the Arduino…
I snapped a picture of a coworker’s “script” he is working on utilizing the object libraries we develop. Apparently, I’ve been doing it all wrong.
How to code
1) PrtScn a bunch of stuff
2) Paste into MSPaint
3) Add giant headers for reference
5) Highlight the shit out of stuff
6) Strike-out any redundancies stemming from poor MSPaint-ing
7) Break out the pink highlighter
8) Locate someone to do your job for you
9) Ship it
It has been a busy several days. I decided to upgrade the fleet to Ubuntu 13.4…and everything that a non-long-term-supported distro involves.
Step 1_: backup data
Step 1a: reevaluate backup regimine
Step 1b: realize backup system is non-sustainable
Step 1c: implement cloud storage solution
step 1d: place eggs in cloud basket
Step 2_: upgrade machines
Step 3_: fix a whole bunch of junk
Step 3*: junk fixing
Step three got ugly. Canonical blacklisted a whole bunch of stuff that ran and runs well. Thumbs down.
The upgrades to my machines were simple, but significant issues arose in updating from 12.04 to 12.10 and from 12.10 to 13.4. The former was mainly in ‘not compiling video card drivers’ and the later was primarily pertaining to neglecting Dell’s primary Wi-Fi drivers. I went manual and had boxes that were mostly functional, but decided to roll the dice on fresh installs…which, predictably, worked well. That gut feeling in doing something of that nature is very Window-y, and I don’t like it one bit.
In short, poorly played by Ubuntu. It might be time to get some real QA going on these upgrades…I spent enough time researching this stuff to be upset, and I debug issues similar to this professionally. There was a ton of noise on forums, so it is safe to assume the user base was significantly disheartened by this process.
On the flip side, 13.04 is slick…Unity is improved a ton…all of that. The marketing is not cool, and I’d be better off with pure Debian if this wasn’t one of my hobbies. Worst upgrade of all time, but still something I look forward to and will likely do again.