Privacy, copyright, resolution, and control issues aside, it seems like Facebook currently has the best software for collecting, cataloging, and tagging images. The facial recognition in Facebook is even excellent, automatically suggesting to tag friends accurately in poorly lit, blurry shots. The timeline, album, and geographical features are great, with the biggest weakness is usually bandwidth to sync a large bulk of photos in one go.

I am certainly not suggesting he use Facebook, quite the opposite. I'm saying that commercial standalone software should try to be as good and easy to use as Facebook and similar like Google+.

Android is an objectively bad OS. It's designed for mobile devices and commits the great sin of failing to be parsimonious with computing resources.

I am a current Android user (Galaxy S4) and have always championed it over iPhone due to the greater device control and options. I'm getting off that train with my next phone purchase. The last nail in the coffin was getting to see a heads up comparison of battery life of HTC One M8 Android vs WP 8. Previously it was easy to dismiss WP 8's battery life on underpowered CPUs and lots of crazy tweaks by Nokia engineers. Now the truth is out, that Android is just a sluggish OS due to poor optimization and the ignorant insistence of using scripting language / virtualized code everywhere instead of compiling for the target.

I have an earlier model G35, and as much as I would like to have the push button start option, I fear exactly that. So far I think only a couple members of a local club have had to deal with it, though I hear it is more prevalent with some of the SUVs.

This does coincide with the BMW and similar thefts due to remote hacking. It's unacceptable and either the manufactures should be footing the bill directly or paying an underwriter to take responsibility for security mistakes.

I should clarify that although all of my examples stem from the perspective of consumer devices in the home, they should equally apply to the implementation of industrial, infrastructure, and similar commercial systems.

This is all fine and dandy as long as a few simple requirements are well understood by the technology implementers.
1) Legal liability for appliances and their cock ups is handled in much the same way as it is now.
If my toaster starts a fire and burns down my house, the insurance cuts me a check immediately while they handle the legal battle with GE (toaster manufacturer) and UL (Underwriters Laboratory who signed off on the device safety). The same legal protections for technology in appliances should be in effect. If my dishwasher gets malware (or just includes it by default) and causes my refrigerator to stop working and spoil my food, the insurance cuts me a check for my losses and to replace the defective appliances, and handles the legal issues.
2) Device interoperability needs to be as easy as antiquated analog systems.
You could hook up a TV with RCA or coax connections and be watching off a VCR or DVD player in minutes. It's ridiculous that HDMI is rarely so straightforward (it occasionally works this simply). Resolution, aspect ratio, audio stream selection, and DRM phone home setup is retarded.
3) Pick a device class to be the central hub, management, and gatekeeper. I suggest home wifi routers or a cheap, simple network bridge device.
4) Full control of permissions and actions by devices with simple to understand language.
If I don't want my washing machine downloading ads for Tide and Purex, it better f---ing not download ads. Same goes for usage tracking uploads.
5) Power usage should not significantly go up. If anything, connected devices should be able to lend clues as to how little changes can make my home greener and/or lower utility bills. Example: washing machine suggests running wash cycle at 8 PM to get out of peak usage billing. It sends me an SMS if I select, to let me know to put the clothes into the dryer.
6) Device electronics / control should be isolated to prevent the additional complication from increasing failure rate.
It's stupid, bad engineering that the more features a home appliance has, i.e. the more premium it is, the more consumers see failures. If the toaster can't get online, it better still make my toast when I press the button.

Not to mention the frequency where customers realized they are / were getting way overcharged for the amount of code they essentially purchased.

It is interesting to see Agile + managed hosting / development / support become popular for software products in niche markets where a new version every 5 years would cover every use case. At the same time those niches are severely disrupted as soon as a competing product enters the marketplace.

One model which would work functionally but massively reduce the headcount most FOSS projects like to tout would be the Habitat for Humanity setup.

Tell people that to contribute, these are the days and times for which they can sign up. Tell them what scope knowledgeable leader they will be reporting to. Let the group leaders track down who showed up (physically or virtually) and hand out assignments.

It's hard enough to get software developers to fix a problem in a product for which one is paying. Nearly every issue I have run into with open source is a driver or compatibility issue which was previously documented years prior to my own stumbling upon it. The developers weren't interested enough back at that time to fix it, and it leaves me with little faith that it is worth my time to chime in with a "me too", not to mention the hate for resurrecting old threads or bug reports.

Part of the problem with open source is freedom. Not enough people sat down at their desk and told to fix it instead of working on what interests them.

I was less concerned with the issues of stereo depth perception. I am comfortable with people driving with only one eye. I have a pretty good understanding of how the fovea and periphery vision systems work and interact. One eye massively increases the peripheral blind spot on one side, but that is easy to compensate for with mirrors and moving your eyes / head.

Double vision where the two images don't overlap concerns me quite a bit more. I am sure her dominant eye takes over to the point that she might as well have just one eye, but there is a greater range for misinterpretation of perception with two images which don't agree.

Something about her driving sounds irresponsible.

Usually, I have found the culprit in large organizations with strong granular security to be the <proprietary single purpose business application> developers and support. The number of times I have watched a new person get onboard and have proper, restricted intranet access, and then the application support people have to open everything up to them to get that one proprietary app to work is astounding.

It is really great that AirBNB is being a responsible civic citizen and charging / paying taxes which apply to their business.

Can we go ahead and explain to Uber and Lyft that they need taxi licenses and to pay their share or gtfo.

Wouldn't it just be easier to aggregate information from social media sites using a weighted system. Just put 4Chan at the top of the weighting, with Facebook next and use separate weighting scales for positive versus negative mention counts. Both are valid predictors, so it should work and get closer.

I'm glad one of my side jobs is setting up IPS / IDP and similar security on firewalls. I'll never be thirsting for work.

Horribly ineffective, but that hasn't stopped the iOS fandom from embracing the iPad as a poor substitute for a content creation device, including software development.

Embracing the device world and abandoning the desktop means exclusive large sources of cash and interested parties who would take control of the kernel from Linus. That's why he needs the desktop to stay alive as a dream. Red Hat, IBM, and Apache are probably the only ones with any clout keeping Samsung and Google from digging deep and streamlining the kernel for ARM / mobile device use exclusively.

I wouldn't even blame Google, Samsung, et. al. if they did decide to just flat out fork the kernel completely and drop the cruft. Hopefully they would bring an ABI driver framework with them.