This is a topic I keep revisiting. It's constantly evolving, with new laws in different parts of the world happening pretty often. And also there's a lot of grey area with vague or incomprehensive language that hasn't yet been tested in courts.

I recognize that it's a bit of a niche topic, but I think there are a lot of us at Tildes who have to think about it. After all it potentially impacts anyone maintaining or building a non-platform web presence. It also applies to less obvious things like running an advertising campaign that involves media requested from a server you control (which can therefore potentially log requests).

For my part, I've needed to research laws relating to PII in order to come up with policies and practices in various contexts. In broad strokes it's pretty simple but as you get into details what I continue to find is that there are a lot of conflicting opinions both from professionals and lawyers. A lot of it is still open to interpretation.

I'm wondering what kinds of experience other tildenauts have around data protection and PII? Have you implemented solutions? Do you wonder about it for your own websites? Have you been involved with it at companies where you've worked? Do you have questions about it?

My iPhone 14 Pro has been at 84% battery health for almost a year now. Anecdotally the battery lasts significantly less time than it did previously, even when it was already at 84% health. I think it may just be a stale calculation, and my actual battery health is significantly lower. If I can get it to show as less than 80%, I can get AppleCare to replace it. Does anyone here know how to get the iPhone to recalculate this value?

I have Coconut batter on the Mac, and it can check the battery health for an iPhone attached with a cable. It uses a different formula, so it gives me a health of 87%. However it also shows history for when I have run it in the past, and when I tested it almost 100 charge cycles ago, it also read 87%. I don't know how any battery can go almost 100 charge cycles with zero degradation (537 to 617 cycles, so it's not like it's a fresh battery).

This topic is for the Three Cheers for Tildes mobile app.

I'll summarize the major updates at the start of each similar topic, so people can read the updates and then hit Ignore if they don't care about more frequent updates and user feedback.

Recently:

[Android] Version 1.3.6 (Feb 28, 2025): Fixed minor UI bugs.

[iOS] Version 1.3.1 (Feb 27, 2025): Fixed an annoying scroll bug when typing comments and posts.

[Android] Version 1.3.5 (Feb 19, 2025): Fixed keyboard and animation bugs.

[Android] Version 1.3.4 (Feb 12, 2025): Fixed keyboard and markdown bar bugs.

[Android] Version 1.3.3 (Feb 11, 2025): Fixed keyboard bugs. [Cancelled this release.]

[Android] Version 1.3.2 (Feb 11, 2025): Fixed bugs reported in comments.

Version 1.3.0 (Feb 9, 2025):

This is an Android-focused update. Android 15 makes apps edge-to-edge by default so it's time to move to edge-to-edge. I've enabled it on Android 11 and higher.

Edge-to-edge mostly means turning the system bars translucent, so you can see the content all the way to the edge, instead of a blank area. In practice, we still need to keep some translucent bars there, so status bar icons and the clock can still be distinguished from app content and not become a jumbled mess.

Implementing this was a gigantic pain (which is why Google received pushback from so many developers and added an opt-out). I had to redo many layouts and re-test every screen in the app multiple times, on different Android versions and different settings (portrait, landscape, single pane, dual pane). Hope it's well received by Three Cheers users! Personally it took me a day to get accustomed to it, but I've ended up liking the edge-to-edge style more. I probably won't add a setting to turn it off.

Screenshots of what it looks like on an Android 14 device as of v1.3.1:

• 3-button nav, edge-to-edge (new): https://postimg.cc/1nnc2rWG
• 3-button nav, old: https://postimg.cc/Hcq4g8Kf
• gesture nav, edge-to-edge (new): https://postimg.cc/ctf744G6
• gesture nav, edge-to-edge (v1.3.2): https://postimg.cc/TpwHPDYB
• gesture nav, old: https://postimg.cc/PNDQMKvh

Three Cheers for iOS v1.3.0 is only minor bugfixes. iPhone apps are already edge-to-edge, and this change is Google's way of copying/catching up to Apple.

Previous topic: November 2024

Where to get it

Android version on Google Play Store: https://play.google.com/store/apps/details?id=com.talklittle.android.tildes

Or sideloadable APK at https://www.talklittle.com/three-cheers/

iOS version on the App Store: https://apps.apple.com/app/three-cheers-for-tildes/id6470950557

Join TestFlight for iOS beta testing: https://testflight.apple.com/join/mpVk1qIy

Last month we started a community-maintained fork of the Tildes codebase. A lot has happened since then.

The biggest change: @Bauke and I have been added as maintainers to the official Tildes repo! As a result, we're moving the community fork to the backburner for now, as we focus on nearer-term changes that will directly improve the main website. Later on it's possible we'll pick up the fork again, where it will likely serve the purpose of self-hosting your own Tildes spinoff sites.

Deimos still has the final say on what makes it to the website. Bauke and I can't deploy changes directly. However, this arrangement is still much more streamlined than before, because we now have a lot more code review bandwidth for accepting outside contributions. Deimos has less work to do now: mostly testing out the live code on a staging server, and scanning over the code for security/privacy issues—but not full code reviews which often involve a lot of back-and-forth communication and reading and testing code.

What work have we done this past month?

It's mostly been setting up foundational stuff like configuring the GitLab repository, fixing the development environment, and writing docs.

More recently we have started fixing actual website bugs too: a bug when escaping a user mention (making sure \@talklittle doesn't turn into a link), and hiding <details> content in collapsed comments. Starting small but we've found a good rhythm and will work on more and bigger issues soon.

Big props to @Bauke for setting up a staging server! Currently at https://testing.tildes.community/ — This server will be instrumental in getting new code in a testable state in a live environment, which makes it easier to approve new features before deploying on the real Tildes site.

So we shouldn't submit code to the community fork?

No, please don't. We'll use the official Tildes repo from now on. I'll update last month's post to reflect this.

Is Docker support coming to the official repo?

Yes, very likely. Deimos has warmed up to the idea. Bauke and I have been using the Docker development environment and ironed out a lot of bugs this past month.

The official repo looks the same as before?

Our next steps are to port the community fork changes back upstream to the official repo. In addition to the master branch, we plan to add staging and develop branches. develop will be where development happens, while master will reflect what is currently deployed on Tildes.net.

How do I contribute to Tildes development?

Check this document: https://gitlab.com/tildes/tildes/-/blob/master/CONTRIBUTING.md

This was inspired by a recent post from Your Local Epidemiologist, where she lays out a set of guiding principles for the blog going forward:

• be a steady guide, trying to avoid whiplash
• providing important context - a broader perspective
• staying grounded in evidence
• being clear about what we know and what we don't know
• approaching issues with empathy

She also references the hazard + outrage framework for risk communication. I come from a safety / risk assessment background, so we usually think of risk = severity x likelihood. But as a communication framework, hazard + outrage seems pretty useful, as talking about risk to lay people is always difficult.

Thinking about one's guiding principles, writing them down, testing them in use, seems really useful to me as a way to be more proactive and less reactive in the way that I deal with the world. So then, the questions on my mind for the Tildes community are:

• what are your guiding principles?
• how did you come by them?
• how have they evolved over the years?

Since this is a text post, I'll put mine in another post below so the responses can thread under it. And since I can never resist a quote, I'll close with:

If you don't stick to your values when they're being tested, they're not values, they're hobbies.
~ Jon Stewart

Every time I get a haircut, my barber hands me a small hand mirror so that I can bounce an image off the wall mirror and see the back of my head. My hair is noticeably thinner in the back each time.

I recently was at a function and saw pictures of me standing around, some of which included the back of my head. The thinning is clearly starting to stand out in a bad way.

I feel like I’ve got two options:

1. Do something to try to save my hair (medication?).
2. Shave my head and try to rock a bald look.

I’m open to tips, tricks, and guidance on either of these (or options I’m not aware of).

I’m not very attached to my hair, so this isn’t a super emotional thing for me. I’m also not scared of going bald since, as a gay guy, I’m well aware of how compelling a bald + beard look can be on some men (my beard isn’t thinning at all, thankfully).

There’s still a question of whether it would look good for me specifically though. Also I don’t know if I’m ready to give up on my hair just yet?

One advantage I do have is summers off (I’m a teacher), so I’m going to be able to do a bald test run in a few months without too much risk. If it turns out that I’m a complete disaster without hair, I’ll just stay home and let it grow back out.

Anyway, I’m open to any and all thoughts on balding, hair loss, hair loss prevention, etc. Tell me your own experiences and what decisions you made. Let me know the tricks of the trade.

SOLVED
Bit of a long shot here because this is one of those issues where I search for the problem and you get a sea of replies like "have you checked the settings?" or "have you tried changing HDMI cables".

I just got a brand new LG OLED TV and I'm happy, but I've been watching Arcane on it and I notice jarring changes in brightness through the episode.
I'm playing through a native app (Stremio) on WebOS and it's not the source file, I've tested the same file on two different monitors and it's fine.
I went through the settings and disabled every autocorrect and "boost" capability the TV has to try and diagnose it, and the first pass did seem to improve the rate of changes, but it still happens maybe once every 5 mins of watch time.
From what I can tell it seems to be picking up particular colour/brightness changes in the source (Arcane is full of then being so vivid) and when it does, it just changes the brightness of the whole display.
I'm no expert here, I'm also colour blind, so I won't categorically claim it is definitely brightness changing, it could be contract or colour, I'm not sure, but it looks like brightness because the whole picture gets darker or lighter.

I wondered if it was actually flip flopping between SDR and HDR which honestly, it might be. If it is I have no idea how to fix that, as the TV seems to have no option to enable or disable HDR on native apps.

Any advice, thoughts, things to try would be appreciated. I'm technically orientated but I don't really know much about changing picture settings to be honest, I tend to pick the most basic/neutral setting and leave it like that.

Edit: I've dug out the old 4k firestick as suggested and don't get the flickering at all through that. Also running through the guide below helped make the picture look even better! Thanks everyone!
I might yet grab the service remote though and see if I can make the native apps work, then I can retire the firestick for good.

If you know me you probably know I hate advertising with a passion. I have blocked ads on my computer but I have limited control on my TV and phone when it comes to YouTube advertisements. And the other day I got this incredibly bizarre ad.

I only speak the tiny amount of Spanish I have learned through osmosis, but the phrase they are using is essentially “we’ll eat at home.” The scenario is a familiar one; you’ve got a kid in the car out in town and they see a restaurant they want to go and they ask if they can stop to eat. The parent says no, we’ll eat at home, because there are a lot of reasons why it’s better for them. Home cooked meals are cheaper. They can be higher quality, both in terms of taste and nutrition. They might have food that will go bad soon and so they want to go and eat it first. They don’t want to normalize eating rich unhealthy foods for their kids.

But this isn’t an ad for groceries or processed food products. This is an ad for DoorDash. A food delivery app. Literally none of the reasons you would want to eat at home apply here. Actually, using DoorDash in this particular situation is dramatically worse, because you are paying more money to get food you could have just picked up on the way only to get a worse version of it because it is no longer freshly prepared and is likely cold.

I just can’t get over this because it’s so incredibly out of touch with reality. Many people have had to have signed off on this for me to see this ad. What were they thinking? Are they so out of touch with reality that they think this is something that people do? The fact that this is clearly targeting Hispanic Americans makes it even worse. I live in Southern California and about a third of the people I know are Hispanic and all of them would laugh at this. I can’t help but wonder if the teams working on them were full of privileged white guys who are saying “yeah, this is what Mexicans are really like” or if there are also rich Hispanics on board who thinks this is something that people really do.

But this is just the most egregious example of out of touch advertising. YouTube ads are supposed to be targeted right? But why do I get ads for CRM, ERP, and accounting products when I don’t own a business? Why do I get ads in languages I don’t speak? Why is it that I can report and tell Google that an advertiser is inappropriate or against their terms and still they will show me the ads again?

And beyond that I am astonished at how many ads I see that exist to mislead you. Almost every time you see something compared or tested there is somthing they aren’t telling you. The most obvious example is Scott toilet paper; they advertise that they have rolls that are significantly longer than their leading competition. But what they neglect to tell you is that their product is single-ply while their competitors are double- or triple-ply. They actually have a very comfortable amount of toilet paper on each roll. Weirdly, all toilet paper is misleading though; what is a “mega roll”, how does it differ from an “ultra roll”, and why is one roll of it somehow equivalent to six of some unspecified other type of roll? In the meanwhile Old Spice is trying to take advantage of balding men like me by telling me that their shampoo increases the volume of hair by however many percent while there is fine print at the bottom saying that it is compared to unwashed hair, meaning their shampoo probably doesn’t do anything the cheapest generic product at the dollar store won’t.

I know this is basically just a rant, but I can’t be the only one who notices this stuff, am I?

I thought I would share this story as I've been thinking about it ever since coming out as transfem and it always makes me smile.

for every year it's been going on -- which is two... but I digress -- I've helped out at a major trade show to put some iconic industry products on display as mainly fully working examples for people to play around with.

I noticed this lady looking at one of our exhibits and struck up a conversation with her as I had done with countless other people that day. turns out she was working at the company who built that exhibit during its production run in the early eighties! we spoke a lot about her experiences with that company.

after a bit, a few more people from that same company came over and they were all reminiscing about their time working there. it was at this point I realised she was trans because she kept saying to all these old guys "you probably knew me by a different name back then"! they were all really accepting and had no issues, goes to show older people can and do respect trans people!

it really inspired me to meet not only a trans elder but a trans elder working in my industry, who had worked on an exhibit I had set up the day before -- we opened it up later and found her initials on an electrical testing label from 1983! in meeting her it feels like I saw a possible future for myself, which is not something I had properly envisioned before, not on the order of decades at the very least. I like the idea of having a future. it gives me something to strive for. I want to be the lady who goes to trade shows and regales bright-eyed students with tales of a long and fulfilling career in my industry. I want to have stories to tell and I want to be there to tell them. meeting her made all that seem that much more possible.

I hope this makes someone else smile like it does me and I'd love to hear more stories like this if anyone is willing to share!

While I consider this resolved at time of posting, this is one of my weirder, more unexpected tech issues I've run into and I thought y'all might enjoy it.

I'd been having issues with my (Ubuntu) server where its services were unavailable, I couldn't SSH in, and USB devices plugged in after the issues started weren't connecting (mouse LED would light up, couldn't toggle keyboard locks or interface computer with the devices). These issues were becoming a near daily occurrence.

So after the most recent incident I decided to watch it boot. After grub, it would just hang at a black screen. So I opted for the recovery mode and after waiting some minutes for services to time out, I was given a terminal and used that to check my devices, sensors, etc and all were reading fine.

Frustrated, I started reviewing what else could have possibly changed with my setup, and I found that change in my hand. I use my server's front USBs as a low current charger and had recently got a new vape battery that was charging at that moment.

I unplugged the battery, rebooted, everything was fine. Plugged the battery in, everything locked up again immediately.

I had been using a spare cable, it had data wired. Swapping for the included power only cable, and no issues.

While I'm curious as to what exactly is happening and why, I'm satisfied enough concluding it is a cheap vape (I mean, I'm having to charge the sucker near daily after all) that is transmitting junk over data and the OS doesn't handle it appropriately. I'll pull logs or do another couple tests if other folks are curious as well though.

Oh, and the battery is a Bear Rootz Sol. Come to think of it, my other worst battery at holding a charge was also a Bear Rootz.

Congratulations to McLaren! First Constructors' Championship win since 1998!

Great drives by Sainz and especially Leclerc. Starting from P19 and ending P3 is beyond impressive.

Hamilton came alive this race, ending P4. End of an era. I'm sure he'd've liked a podium to end his time with Mercedes, but still a hell of a drive, moving up 12 places from the start.

Felt bad for some of the guys who were or may have been in their last race: Bottas, Lawson, Colapinto, and Perez all DNF'd unceremoniously.

With that, the season is officialy over. I, for one, am glad. It's been a long season. For US folks, enjoy sleeping in on the weekends again for a few months. Looking ahead, 2025 pre-season testing is in Bahrain, Feb. 26-28, with the first race of the season in Melbourne a couple weeks later. See you all next year!

Next race:

Australian Grand Prix
Albert Park Grand Prix Circuit (Melbourne)
Sunday, March 16, 2025

Good race, fun race, but overshadowed by the baffling stewarding decisions and race direction.

What appeared to be a mirror that broke off a car is in the middle of the track, and no VSC or something to grab it? Then two cars get punctures, possibly from driving over that debris, one right after another in basically same place, but no safety car or anything called until a few laps later?

Then add on the harsh penalties. Plus Max's weird penalty for impeding-but-not-really from yesterday where he lost P1 after quali. As it's said, nothing as consistent as the inconsistent stewarding.

On a positive note, glad to see Zhou finally in the points! On his penultimate drive, no less. Only Sargeant and Bottas remain without points this year. Maybe Bottas will get some in his final race.

At least we'll get a down-to-the-wire championship showdown in Abu Dhabi. Ferrari vs McLaren for P1 in the WCC. Who are you all rooting for?

UPDATE: That was likely Ocon's last race with Alpine. Rumor, which Sky Sports appears to be confirming, is that he'll be out in Abu Dhabi, with Alpine reserve driver, Jack Doohan in for the last race. This is so Ocon can do post-season testing with Haas.

Next race:

Abu Dhabi Grand Prix
Yas Marina Circuit
Sunday, December 8

This topic is for the Three Cheers for Tildes mobile app.

I'll summarize the major updates at the start of each similar topic, so people can read the updates and then hit Ignore if they don't care about more frequent updates and user feedback.

Recently:

Version 1.2.2 (Dec 5, 2024):

• Added advanced settings to auto-vote on topics when opening them
• Added advanced setting to hide "Reply" from text long-press menu
• Fixed inserting markdown links with parentheses
• Fixed UI bugs in profiles
• Fixed crashes opening certain topics

Happy Thanksgiving! Three Cheers version 1.2 is out for Android and iOS! (Nov 27, 2024)

• Save drafts of comment replies, edits, and private message replies
• Manage drafts via Settings
• Improved some error messages

Does not yet support saving drafts of new submissions

I know some of you were hoping for topic submission drafts, but I wanted to release what I've implemented so far with comment drafts. This version took a decent amount of time to develop, and I didn't want to delay it further by adding submission drafts in this release. Thanks for your patience!

Previous topic: September 2024

Where to get it

Android version on Google Play Store: https://play.google.com/store/apps/details?id=com.talklittle.android.tildes

Or sideloadable APK at https://www.talklittle.com/three-cheers/

iOS version on the App Store: https://apps.apple.com/app/three-cheers-for-tildes/id6470950557

Join TestFlight for iOS beta testing: https://testflight.apple.com/join/mpVk1qIy

I watched the video "Dirt-Cheap Solar Generator Setup - What Can It Power?" by Alex Beale yesterday. In the video, the presenter purchases a cheap solar panel and power bank for a total of $250 and tests household devices to see what kind of wattage it can take.

Turns out the cheap device can power most household objects. I'm curious if it's possible to try a similar setup to meaningfully cover my electricity usage on a day-to-day basis. Unlike the creator of this video, I don't have a yard. I'm exploring whether a solar solution could work with my constraints.

I wanted to share this idea with Tildes and hear any ideas/feedback/experiences from people who have tried something similar.

My goal

• Investigate whether it's feasible to save money with this setup
• If so, find a setup that I can use to power some or all of my household devices off-grid
• Pay less in electricity costs
• Be able to rely on my power bank when the building's power goes out

My current situation

• I live in an apartment in an urban area in the US northeast (southeastern Pennsylvania).
• I do NOT have a balcony (sadly).
• I have two south-facing windows and four west-facing windows (two of which get a LOT of light, two of which get a decent amount but not as much). However, my south-facing windows don't get direct sunlight most of the day since there is a building in the way (just a bit in the late afternoon).
• The dimensions of the windowsills are about 16 inches in depth and about 55 inches in width. That's enough for a small or medium-sized solar panel, but not most panels designed for outdoor use.
• I can open the windows all the way and it would be physically possible, in theory, to hang something outside rather than keeping it on the windowsill (we'll get into this...)
• I pay for all my electricity. The heat/air is central and inaccessible, as are a couple of the overhead lights (there are only a couple), but everything else could theoretically be replaced with energy from a power bank, I think.
• I already have a power purchase agreement with a renewable energy supplier. All my energy is "green" in the sense that I'm making it slightly more economically feasible to produce green energy.
• In a less expensive month, the bill might come to around $75. In an expensive month, it might come to around $350. If I heated the unit as much as I would like, it would be higher.
  • I'm pretty sure most of my electricity use is the heat/AC.
  • I'm pretty sure most of the remainder is the washer and dryer dishwasher, followed by the washer and dryer.
  • I spend some amount heating my water otherwise (shower, sinks), but I can't address that with a solar panel.
  • I wouldn't mind saving money on the rest if that's all that's feasible.

My power needs (I think)

• I don't know much power my oven uses. It's electric, which the internet says averages around 2500 W.
• My dishwasher runs at 120 V and 18.7 A. That is... uh... 2244 W. Also a lot.
• My dryer says 120 V on the machine, and if I look up the model number, it seems to be at 15 amps. So that's 1800 watts.
• My washing machine, uses....... IDK how many watts. No information on this thing exists. But it uses a 120 V outlet. The internet says an average washing machine could use up to 1400 W (sounds high, but okay).
• My computer has a 520 W power supply, but I doubt it ever uses that much. It's old and mostly runs Solitaire. Maybe one day I will get a fancy new machine that does like 1000 W.
• My fridge uses 115 V at 4.5 A. Let's pretend that's 120 V and say 540 W.
• I can't tell how many watts my TV uses, but the internet says it could be up to 200 W.

I would want a bit of breathing room with the rated wattage so that the power bank doesn't explode.

No way I can run all these devices simultaneously off the power bank, but perhaps I could time them. Even partial coverage could be adequate if it makes financial sense.

Possible spatial configurations

I see a few possibilities here:

1. Stick a bunch of solar panels in the windowsills, propped up at a suitable angle.
2. Stick the panel directly in the window like an air conditioner. In this case I would want to use a two-sided panel, since there would be some ambient light coming from the inside. This would inevitably reduce my home's energy efficiency, which I would have to account for.
3. Weld some sort of counterweight to a mounting system that I can use outside the window to perch the panel at an ideal angle. This would also reduce my home's energy efficiency, but maybe by less since the gap could be minimized. It is also probably illegal. And dangerous.
4. Just fill my living room with solar panels. I will need to have a conversation about that. But it could be done.

I think #1 is the only one that is actually reasonable. #2 would cost me too much in heating bills to make any sense. I think #3 would get me evicted or sued by the city. And #4 is unreasonable considering that I like to use my house to live in.

My budget

I would maybe spend $500-1000 on this if the payback period was a few years. Not more than 5 because panels by then will be a lot more efficient.

I guess I could spend more than that if it makes sense. I have good credit and a high credit limit. I don't make that much money but I could save for a bit or carry a balance for a bit. I think I would draw a hard line around $3000.

The power doesn't go out here too often, so that isn't the main selling point, more of a nice-to-have.

My research

I'm just looking on Amazon. I know I could maybe get something a little cheaper elsewhere. I'm searching for "power bank with solar panel".

• 4000+ W range: OUPES Mega 5 for $3300+ with panels (I think?). That's a lot of money for something that might not even work. And I don't believe it would be $3300, most other models I'm seeing are closer to $5000.
• 2200 W range: Jackery Explorer 2000 for $2500 with the 2 panels. Maybe this could cover my energy-intensive appliances, one at a time. Still expensive. The solar panels will... maybe fit in the window with about 1/2 inch to spare? I could angle them slightly if needed?
• 1800 W range: (maybe) Anker SOLIX C1000 for $800 with 1 panel. It also says 2400 W so idk what it is really. Can't determine the width of the solar panel.
• 1000 W range: EF ECOFlow River 2 for $700 with 1 panel. I am also not sure how big the panel is.
• 600 W range: SinKeu Portable Power Station for $230 with 1 panel.
• 300 W range: Marbero Solar Generator for $230 with 1 panel.

I'm discovering that these companies never specify the dimensions of the solar panels because they assume no one would ever use them indoors. Probably for good reason. If they don't fit horizontally, I guess I could stand them up lengthwise.

I'm sure it would be possible to find a cheaper solar panel separate from the power bank. But as a ballpark, those are my initial numbers.

Math

Imagine I spend $2500 on the 2200 W Jackery model, with 2 panels. I could put them in the two bright west-facing windows. Let's just pretend they can get their maximum output and that charging time isn't an issue.

Let's say I use this every time I run my oven, dryer, washer, or dishwasher (not at the same time), and otherwise I connect it to my fridge and most other electronics. (I know I said my oven could be 2500 W, but I usually just use one burner, and I bet that's more like 1500 W.) Let's pretend the hassle of running all those cables has been solved somehow.

I really don't know exactly what my electricity breakdown is. If I have the heat/AC mostly off and am not in the unit a whole lot (e.g. traveling for some of the month), that's when I end up with a $75/mo bill. But if I'm out of the unit, that means I'm also not using my dryer/washer and other devices as much. I guess the fridge was still on. The $350 side of things is because it's hot and I'm running the AC more, and maybe slightly higher dryer usage for some laundry-related reason. So maybe I'll say the minimum I'm spending on non-heat/AC electricity per month is $50-100, and at most maybe something like $150.

So in this super optimistic scenario, the portion of my monthly electricity bill I suspect I'd be able to theoretically save would be $50-150. Let's just say $100. That means it would take 25 months to pay back if all goes well. Honestly that's not so bad.

If I instead bought the 1000 W ECOFlow setup for $700, I could cover all my non-intensive electrical uses. The main beneficiary would be the refrigerator, the computer, and maybe the television. I think the savings there would be like $25-50/mo at most, being pretty generous. If I say $35, that would be about a 20-month payoff. That's also pretty good, but if it'll take about the same amount of time to pay for itself, I'd rather get the higher-wattage one.

Numerous caveats:

• It's not always sunny in Philadelphia.
• The panels will not be at 100% efficiency because the window is cloaked in shadow half the day, and there is no way I get the angle perfect, and maybe the glass affects how much energy they can absorb or something (would it help? hurt?). This isn't inherently an issue, except...
• ...that the charging time for the power station might be long. Some of the reviews say 6 hours in maximum sunlight, so for my use-case there's no way it would be faster than 12 hours for a full charge, and probably more. I use my appliances kind of a lot. The fridge is always running, and either the washer, dryer, oven, or dishwasher are running a pretty good chunk of the time. I think that I could get by with this... but I probably wouldn't be able to use it as much as I'd like (there are often moments where more than one is running simultaneously), so...
• ...the payoff time wouldn't actually be 25 months. I suspect there are some factors I'm not thinking about that would reduce the realistic amount of energy I'd save here. Let's say it is half as effective in my setup as a normal person's, because the sun is half as visible or something, and the charging time is too slow to use often, or whatever. At a 50-month payoff (4 years), I would start to get skeptical. But not inherently opposed.
• I use my windowsill for other things that like sunlight, like plants. I would have to move the plants that really love the sun somewhere slightly dimmer. Is it worth it???????
• I also like to look out my window, and if I had a huge solar panel there (especially if I had to stand it up vertically for space reasons), that would be a bummer perhaps.
• Constantly charging and discharging this thing probably reduces its effective wattage output fast. I.D.K. by how much. But there might be a point where I can't use it for my high-wattage devices, which would largely defeat the purpose of buying an expensive model. Not clear to me what the timeline for that would be.
• I still can't figure out if the panels would actually fit in my windowsill.
• Since it wouldn't be able to run more than one energy-intensive device simultaneously, I'd have to be careful not to keep it plugged in to too much stuff at once. It would overall be a bit of a hassle to plug in the panels and the charger and everything. Maybe the panels could stay plugged in, but even so.
• Running extension cords all over the house and putting a giant solar panel in even just one or two of the windows would result in a catastrophic domestic state of affairs. I can just imagine the eye-popping that would happen if I set that up unilaterally.

Takeaway

There is almost definitely something I am not considering. My math is so napkin-based, so imaginary and so optimistic.

But even if the practical efficiency isn't great, I feel like this could maybe possibly actually work, and could save me a bit of money. I would just have to spend it all up front. I mean I don't have $2500 on hand. But I could scrounge up part of it and put the rest on credit. And it would make for living room conversation with guests?

Realistically this seems like an amusing but too-annoying to actually do sort of project. If you know anyone who's done something silly like this and had it work, let me know! I'd be curious what their experience was.

Solved

I was idiotically installing a legacy driver instead of the current driver. I installed the current one and it is now working.

Background on Tildes

For background I posted about distro recommendations a few weeks ago: https://tildes.net/~tech/1ji6/switching_to_linux_looking_for_distro_recommendations. I settled on Fedora KDE Plasma.

Steps taken before the problem

I installed Fedora onto an unallocated space on my SSD, alongside Windows 10. I have tested that Windows 10 is currently working, and that Fedora starts normally. I have also used the boot media to install onto my laptop with no issues. However, on my desktop before installing GPU drivers for my GTX 970, I found Fedora to be quite buggy. This would include the session freezing and needing to be restarted , or my screen would go black with text saying something along the lines "desktop session cannot be unlocked, press ctrl + alt + f3 and login and run a command (I cannot remember the command)" and then I could switch back to my previous session. Overall, it was a buggy mess. My thought process was that it may be a GPU driver issue, and so I started the process of installing RPM fusion and installing the driver from there. I have a GTX 970, so I followed the section titled "Legacy GeForce 8/9/200/300" on this page: https://rpmfusion.org/Howto/NVIDIA?highlight=%28%5CbCategoryHowto%5Cb%29. The commands I ran after installing RPM were: sudo dnf update -y and sudo dnf install xorg-x11-drv-nvidia-340xx akmod-nvidia-340xx and followed all the prompts provided as necessary. I would then leave my computer for about 30min, before returning and restarting (thinking that a restart may be necessary to be fully using the drivers).

Current Situation

After going through the process listed above, I turned my computer on and booted into Fedora. My screen would hang on a black screen with a flashing white underscore at the top left corner. The computer would not get past this point, no matter how long it was left. Hitting "ctrl + alt + f2" would give me a working terminal that I could interact with, showing the basic CLI of Linux working. I did not play around much with the commands, but I do know the basic shutdown now command was working.

Troubleshooting Steps Tried

I was thinking that it may be due to the driver being X11 and I am running Wayland. I searched online and results seemed to indicate that this would not cause an issue
(I will add more steps here when people recommend them)

Okay, so this is getting really long, I'll put the ask up front: I have a strategy, I think it is reasonable. Now is a point where I can easily change things, and it won't be so easily later. So I'm looking to see if anyone has trod this road before and can recommend any of:

1. a different build system that will be easier to manage for this use case
2. a different strategy for using cmake that will be easier to manage
3. any gotchas I should be aware of, even if you don't have better solutions.

Background

I have a project I'm working on where the ultimate deliverable will be a hardware device with 3-4 different microcontrollers coordinating with each other and interacting with a PC-ish platform. This is a clean rewrite of a C++ codebase. Due to the microcontroller (and some of the PC APIs) being C++, the language of choice for most of it is likely to remain C/C++.

I'm succeeded in setting up a build system for embedded code. The old code was arduino, so it relies a lot on those libraries, but I've managed to set up enough custom cmake to get off of the ardunio tools altogether, even if I am borrowing their libraries and some of the "smarts" built into the system about setting build flags, etc. So far, I have a dockerized toolchain (cmake + make + gcc-arm-none-eabi) that can successfully build ARM binaries for the target platform.

The thing that I'm up against now is that I'd like to have a robust off-target unit testing infrastructure. My ideal case is that everything in the embedded system will be broken down into libraries that have clear interfaces, then to use unit tests with mocks to get high coverage of test cases. I'll still need some HIL tests, but because those are harder to set up and run, I want to use those for integration and validation.

In terms of OSes available, we're mostly working on Windows systems using WSL for linux. I'd like things to be as linux-based as possible to support CI on github, etc.

Goals and Cmake limitations

I started out using cmake because I hate it least of the tools I've used, and I am at least pretty far up the learning curve with it. But a limitation I'm hitting is that you can't do a mixed compile with two different toolchains in one build. The reasons why cmake has this limitation seem reasonable to me, even if it is annoying. You can easily change the toolchain that your code is built with, but that seems to be largely targeted at cross-compiling the same binaries for different systems. What I want to do is:

• build my code libraries with embedded settings for linking to the embedded binaries and build those embedded binaries (the end product)
• build my code libraries with linux-ish tools and link them against unit tests to have a nice CI test process
• (eventually) also be able to build windows binaries for the PC components -- when I get to that point, I'd like to get away from the MSVC compilers, but will use them if I have to

Current strategy

My current plan is to configure a library build like this (pseudocode):

add_library(mylib sources)
if (BUILD_TYPE STREQUAL BUILD_TYPE_EMBEDDED)
<embedded config>
elseif (BUILD_TYPE STREQUAL BUILD_TYPE_LINUX)
<linux config, if any>
endif()

#unit tests are built for each library
if (BUILD_TYPE STREQUAL BUILD_TYPE_LINUX)
add_executable(mylib_test sources test_sources)
target_link_libraries(mylib gtest etc.)
endif()

For the rollup binaries, I make the whole target conditional

if (BUILD_TYPE STREQUAL BUILD_TYPE_EMBEDDED)
add_executable(myembedap sources)
target_link_libraries(mylib)
endif()

Then the build script (outside cmake) is something like

cd build/embedded
cmake <path to src> <set embedded toolchain> -DBUILD_TYPE=embedded
make
cd ../../build/linux
cmake <path to src> -DBUILD_TYPE=linux
make

Things I like about this strategy:

• It's relatively simple to do all the builds or just one of the builds (that control would go in the shell script)
• I have one source tree for the whole build
• It lets configuration be near code
• It lets tests be near code.
• I think it's extensible to cover the PC component builds in the future

Things that worry me:

• It feels like a hack
• Support for off-target tests feels like it should be solved problem and I'm worried I'm missing something

Thanks for reading. If you made it this far, you have my gratitude. Here's a video with funny out of office messages that I enjoyed.