Small tech companies tend to be design or engineering led. Organizations grow and evolve, and as they become large they become marketing or finance led.
I too have tried to use a USB power bank as a kind of DIY UPS to power a Raspberry Pi SBC. What I learned - which may be unrelated to your experience - is that all of the several power banks I tried can be either charging, or powering a device, but not both at the same time. So my Raspberry Pi would run until the power bank got low, then the power bank would cut power to the Pi and charge itself, then power the Raspberry Pi up again. I had to buy a power bank specifically advertised as a USB UPS; that worked fine. The normal power bank seems to be designed to charge a battery powered device like a mobile phone (whose own battery can be thought of as a kind of UPS for the device), but not a device that requires continuous power (like the Pi).
Yes, lots of them. I hate reading on the screen, enough that I own a color laser printer more or less just to print articles that I want to read that are online. But for the most part, I find articles in print magazines to be far superior over what is online: better moderated, more interesting, better written, etc. This probably isn't a complete list, but I pay for: THE NEW YORKER, BUSINESS WEEK, THE ATLANTIC, TIME, WIRED, MIT TECHNOLOGY REVIEW, HARVARD BUSINESS REVIEW. I also take advantage of online content that my subscriptions provide. You'll note that none of those are computer related stuff; that I do get online, but that's in part because all of the print computer magazines I subscribed to went out of business.
Being able to get feedback can be both a pro and a con, depending on the nature of the feedback.
Sometimes you can use information asymmetry to your advantage; thinking in public may work against that.
For sure thinking in public can be a form of marketing, especially if you are self-employed, like I am.
Having said all that, I've been writing a public blog since 2006, and have 39 repositories in GitHub, most of which are public, and that is a form of thinking in public as well. So I guess I cast my vote on the issue a while ago. I try to be authentic, honest, true to my values, open to change, and be the best person I can be, when I share my opinions.
I also have a MacBook Pro and a GPD Micro PC. The latter is ideal for taking into the field (in the case of my GPS work, literally a field) to connect to the various pieces of hardware I deal with. It's really a pocket-sized industrial PC, in the sense it has a lot of physical ports, e.g. Ethernet, a DB9 serial port, etc. taking it quite useful for hardware hackers.
Historically, this is what people have used public libraries and university libraries for. They're typically not internet-free spaces, but a little discipline on the user's part solves that.
I'm an embedded/real-time/systems software developer. The realms I work in tend to be either the very small or the very large. When I wanted to learn Rust and Go, I coded the same application in both of them, one that I had previously coded in Java, C++, and C: I implemented the Generic Cell Rate Algorithm using a virtual scheduler. The GCRA is a traffic shaping algorithm that takes as parameters a desired peak rate, sustained rate, and maximum burst size, to meter or police traffic. I coded it as a library function, and provided a unit test to convince myself that it worked. I can only learn by doing, and that's what I did. One of the reasons I like this approach is it requires a lot of stuff: threads, synchronization, modularization, etc. https://coverclock.blogspot.com/2018/11/vamoose-rustler.html