Then, there is getting the VCs to even -look- at you. Want to know what VCs want for an ideal startup? The Meitu app. If it does not slurp up data/telemetry/tracking and push out ads, VCs won't even give you the time of day.

Want to have an app? Make it yourself. VCs don't want ideas, they want the business around the idea that is working and ideally "poised for growth".

GameStop used to survive when one paid $50 for a game, sold the game for $5, and the used game had the same content as the original. However, with the combination of DLC, having to have the CD and have the game registered under your account, and so on, a game might cost $250, all said and done. There isn't any real point for a used buyer to buy it, since the used media will be at $45.00, the new game will be $50, and there is still $150 of DLC that has to be purchased before the game is anywhere usable, much less playable with others.

There is also the fact that people are PC gaming more often. PCs may not have the ability to just plug and play like a console, but GOG, Steam, and MS Store games are relatively cheap, one can back them up fairly easily, and there is a wide selection. Why go to a used game store to pick up a game when you can order it and be playing at home?

Take a chunk of sodium and toss it in water, you get lye and hydrogen. The result is a very alkaline soil.

I will say that APFS is a must have update from HFS+. It has copy-on-write functionality, snapshots, and other stuff that make sense. It has a very interesting facility for encryption to allow for volume, file, and almost anything in between, with keys for everything able to be different.

However, it doesn't have the good bit-rot detection that ZFS, ReFS + Storage Spaces, and btrfs have. In fact, it doesn't have any real robust drive scrubbing type facility to find and (even better) repair ECC errors. I read that Apple is assuming that all data is stored on "premium" storage media, so they didn't add CRC checking to the code. Or, this could be not included due to performance reasons.

In any case, this is much needed upgrade. However, it still is behind everyone else, especially when it comes to bit rot.

The ironic thing is that the US is actually known as a leader in robotics. Car assembly lines are almost completely automated, for example. Chip making, pick and pull machinery is a common staple. CAD/CAM is a part of everything and anything in the US. Want to be able to design a new widget? Better know Solidworks, AutoCAD, or similar.

The talk about the US losing the robotics race is unfounded. In fact, contrary to what a lot of people believe, the US still doing manufacturing, and is definitely not going anywhere. Robotics will definitely be a part of how new plants are done, period.

People have been trying to fix E-mail so often that it became common for a pre-printed form to be copied and pasted when someone had another solution. SMTP is so entrenched that there is no real replacing it.

What might be the ideal message app is one that can use multiple channels to send a message. SMS present? Great. Signal, Telegram, or another protocol? Useful. SMTP to a specialized E-mail address with the server autodumping any spam not signed with a proof of work token or being part of a contact list? A thought. Perhaps send the same message (with a unique ID) via several different protocols, with the receiving app validate, check if any copies were damaged in transit, and dump the dumplicates?

We have a shitload of existing protocols. The ideal would be to have the messaging program use those. However, the message format should use existing standards. OpenPGP comes to mind as a good way of encoding packets that is cross platform and can be accessed on almost any platform.

Now that we have a message standard and the ability to use multiple transport protocols, from there it is making contacts, using public keys in a user friendly way without giving up security (perhaps having selectable levels of security), and doing UI work. The crypto infrastructure is the hard part that needs to be done -right- with auditors. The UI work is pretty much commodity stuff.

tl;dr, why replace existing protocols... Use multiples of them.

1Password also does something unique. It is able to store your Google Authenticator 2FA keys. That, and allow export in a text format, so you can input them into another authentication app if needed. There are other apps which can back up the 2FA keys like Authy, but the backups are only accessible to the app itself.

Yes, 1Password has had flaws, which were corrected, but it works well, and allows one to store the PW data on a cloud provider of choice.

This is going to kill AIM. Yes, it takes expense to support third party items, but being open gets more people using the service.

There have been a shitload of closed chat systems, Anyone remember "Ding!" in the 1990s? There were many "Internet phone" companies also offering chat mechanisms as well. The reason why they are not around is because never were open enough to attract third party developers.

Plus, who uses AIM these days? If I need to message someone, it will either be SMS/MMS, FB Messenger, Signal, iMessage, or even Skype. AIM isn't worth the time in keeping a client open for it.

I have wondered about different virtual machines on a phone. The retina lock might get one the VM for a workspace for personal stuff, while to access business data, it would require a fingerprint and PIN. Done right, there would be plausible deniability for this... and more importantly, it would separate business and personal stuff.

The ironic thing is that when I interviewed at one place last summer, the CTO personally started asking me questions. When I asked them about their disaster recovery plan, as the company was 100% AWS based, the reply from the CTO was, "Asking a cloud based company about 'backups' or 'uptime' is like asking a Tesla owner about the type of buggy whip they use."

Needless to say I decided to look elsewhere for work.

Not really. Solar panels are becoming as tied to a construction project as roofing materials, and other basic building supplies. Even after buildings are retofitted, there are always new things coming up, new technologies that are iffish now, but are maturing (tinted windows which may run at 1/20 the wattage a normal panel, but with the sheer square footage on a south side of a building, it might be worth doing, when the price for the tint becomes that cheap.)

Solar plants will continue to expand. With HVDC transmission methods, there is a lot of desert that can be used for solar, and with roughly 3.5% transmission loss per 1000 km, this can be a viable way to provide a few GW to a city. If the transmission loss is too great, it isn't too difficult to pull CO2 from the air and make ethanol, propane, synthetic diesel (Audi has pioneered this), or something similar as a way to fuel non-electric vehicles and stay carbon negative. Heck, with enough power and a source of water, thermal depolymerization becomes possible, which is an extremely good way to dispose of plastic and have a usable resource for fuel or manufacturing.

Solar technology will only improve as well. Panels may be near maximums of energy output, but better MPPT controllers and energy storage will be the focal point eventually as the bottleneck moves from panels.

The nice thing about solar is that it is stupidly easy to set up compared to any other energy source [1], and it is relatively maintenance free, because everything is solid state on the grid, and off the grid, the only component that wears out are batteries.

[1]: A cast off car battery, a surplus panel, a $8 PWM charger from eBay, and some 12 volt light bulbs can power the lights on a detached building indefinitely. I don't know any other energy source that can sit there and do that. The Aussies go a step further and stick refrigerators with solar panels on them in the middle of nowhere so they can get a cold one even if on the back 40. I don't know any other energy source that can do that... nuclear perhaps, but with all the fear about nuclear, you will never see a basketball-sized reactor just for powering a small building.

Torts will do little to nothing. Every IoT device has a EULA or ToS with it forcing arbitration and absolving the device maker of all blame should something happen with the item. Even with torts, the IoT company likely has a good number of lawyers who will just steamroll over anyone bringing lawsuits, or just stall the lawsuit until the plaintiff has to drop it due to lack of funds.

For the little guy, the civil system only will bankrupt them, so it is no real check.

Because IoT makers view security as having no ROI, we will keep seeing this over and over. In fact, having devices that are unable to be updated brings more money, because it forces consumers to buy the 1.1 or 2.0 version of the same IoT device.

This is why government has to step in. There is no mechanism to make IoT makers give a rat's ass about security whatsoever. The same exact thing is why we have UL listings. Companies don't make money by spending extra to have appliances that don't electrocute the end user, so government mandates UL listings in order to have a safe standard. Perhaps the same should be done with regards to security, since security mandates will not be coming from the private sector, as it does not benefit them. "A lock makes no money for anyone other than the lock maker and the locksmith."

I applaud the hotel for moving back to keys. Ideally, the system for the card readers should be a closed, air-gapped system that has zero network connectivity (almost all hotels had exactly this in the 80s and early 90s so it isn't a must have for door locks to be connected to the Internet), but moving back to a completely mechanical system isn't a bad thing either. Even with a high security locks like Abloy or Evva MKS, the cost of cutting a new key and repinning a hotel cylinder can likely far cheaper than having to maintain/update/replace a keycard system anyway.

I've seen PostgreSQL run rings around MongoDB. This being the case, there isn't a real reason to even bother with MongoDB... just stay with something tried and true that has a known good security model.

I will be happy if the phones have an unlockable bootloader, so I can install CM/LineageOS. If Google does this, that is the main thing.

Car makers need to stick to stuff that runs on the CAN, and well away from consumer electronics that get chucked in a year or two. Oddly enough one of the better audio heads I've used was the one on my Ford. It doesn't have a touch-screen display, but buttons and dials. However, Bluetooth works, and has worked with a large array of Apple, Motorola, and HTC phones. If I need navigation, that is what a suction cup holder, smartphone, and Siri can be used for. People made fun of Windows Automotive, but I found it quite usable, and has stayed usable over the years.

I am not impressed with some of the touchscreen models, just because the functionality can be obsolete or inoperable in less than six months... and unlike phones, people in the US are not going to swap cars out every year or two so they can have 128 gigs of SSD for storing MP3s compared to last year's 64 GB.