In this case I operate on the assumption car drivers are inherently good people. So when I am cycling in the middle of the lane (when the lane is too narrow for safe sharing), and they are behind me hitting their horn, I give them the benefit of the doubt as to whether they are being a malicious prick.

So I stop the bicycle, get off look around, check my pockets, and if their window is open I ask “did I drop something?” Because surely they would not use their horn to demand that I move to the side so they can pass me unsafely. Surely they are kindly signalling to me that my backpack is open, or that something fell from my bicycle.

Every single time, I never manage to discover that anything was lost or out of place.. but I continue to give drivers the benefit of the doubt every time.

It doesn’t take long for mold to grow on empty beer bottles. Considering beer bottles get returned for a refund, you have to assume that the brewery will make an effort to reuse as many as possible.

I toured a brewery once and they showed us the big industrial bottle washing machine. They said the bottles get scanned for cracks using a laser, and rejects obviously get tossed. The question is: what about mold, which adheres quite well to the corners of the glass? I wonder if the laser also detects bottles that didn’t get clean. Or if they just figure the temps would kill everything and just be considered safe enough from there.

My SOHO router has a some “green” configuration features to reduce energy consumption:

• (wi-fi) A scheduler for the Wi-Fi radio to turn it off automatically during times it won’t be used. (dd-wrt has this)
• (wi-fi) A power level throttle (10%, 20% 50%, 100%) so you are not amplifying the signal beyond the range that you need. (dd-wrt has this)
• (wi-fi) A bezel button on the chassis so you can easily turn Wi-Fi on and off without entering the configs.
• (wi-fi) choose an SSID that does not feed an oil partner (details).
• (ethernet) A per-port choice of 1 Gbits/sec or 100 Mbits/sec. Apparently capping it to 100 Mbits/sec saves energy because they’re calling it a green setting. I’m a bit surprised the savings would be notable enough to justify the option. But I doubt my uplink has more than 100 Mbit/s anyway so I capped my ports.

Beyond the router:

• (uplink) Since GSM radios use 30 times more energy than a wire, obviously getting your internet over cable, dsl, or fiber are more energy efficient than GSM (and probably any wireless uplink for that matter).
• (web browsing) Disable image loading in the browser because images are much heavier than text. Most images are junk anyway.

reverse tethering

I’ve started reverse-tethering my phone over USB, so I can keep Wi-Fi disabled on both my router and devices most of the time. This option is threatened though, because the Android tool Gnirehtet is no longer maintained.

When Gnirehtet eventually dies, theoretically it’s possible to use openVPN for reverse tethering. But the ovpn project has decided to scrap the clearnet option under the naive view that there is no use-case for an unencrypted tunnel. If you can’t get cipher compatibility between your mobile device and your PC, it will not work.

• broken: Ungoogled Chromium ver. 90.0.4430.212-1.sid1
• works: Ungoogled Chromium ver. 112.0.5615.165-1

If anyone has problems getting Ungoogled Chromium (and likely Google’s Chromium as well) to work on Lemmy, notice the versions above. The Lemmy webclient is a dysfunctional disaster in the old version but they fixed whatever the problem was in recent versions.

The GPS receiver in Androids takes a long time to acquire satellites. I find that suspicious because back in the days of TomTom and Garmin, satellite acquisition was only slow on the first bootup or after being powered on far from home. But those dedicated satnavs seemed better at remembering satellite data. It seems like Android might be deliberately slow in order to incentivize users turning on “Google Location Services” (GLS). I also notice OSMand sometimes thinks I’m moving along a few meters away, sometimes on another parallel street. I did not see that degree of inaccuracy on TomTom or Garmin.

I will not agree to GLS because I will not feed Google. So how can I improve the speed of getting a fix and the accuracy?

I know there is an app that uses the phone’s other sensors to track position from an origin that you specify. It claims to not even need GPS. I still have to try that. But it might be useful if it would use GPS to periodically recalibrate.

Is there any free-world way to fetch a db of SSIDs and GSM towers in a city, and bypass GLS?

update

Thanks for the replies. Just now (14 days after my post) I happened to discover replies here when visiting the instance directly while logged out. That’s really screwy that I apparently need to be subscribed to the whole community to get notifcations of replies to my single post. Now that I’m logged in and viewing the slrpnk mirror, I can’t interact with the other comments.

The problem I have is on long trips (via bicycle or on foot) my phone’s battery hits 15% remaining and screen dims mid-trip, which is essentially blank in daylight when navigating. I’m in airplane mode with wifi also disabled. So the only power consumers are the screen and the GPS receiver. Yet I’m still forced to power down, swap batteries, lose the clock time (which GPS strangely fails to correct), and wait to reacquire a GPS signal. Then OSMand remembers the route parameters but forgets the route (a bug). And because the phone’s time is 1am, I have to either update the time or force OSMand into daytime mode.

Big hassle and unwelcome interruption. I see 3 fixes:

1. Repurpose an old phone to receive the GPS signal and feed the lat/long over bluetooth to your navigation phone. Since a bluetooth radio in receive mode consumes around ⅒ the energy of a GPS receiver, the main phone battery will last much longer. The GPS phone need not power a screen, so it can obviously run quite long if it’s only powering GPS chips and bluetooth in tx mode. (refs: GPS uses 13-38%, bluetooth uses ~1.8% / 17.9mA on one chip; math-intensive research I didn’t read because it would make my brain explode)

2. Attach an external USB battery. I reject this because I don’t want to strap another box to my arm and run a cable into my water resistant phone strap.

3. Get an Android-compatible phone with a dual mode LCD, so a low-power e-Ink mode can be used in daylight. I reject this because I boycott Russia and IIRC only Russia has phones with dual mode displays. I would perhaps be open to buying just a raw dual mode screen (not from Russia or Israel) and then use it to replace a cracked screen on a 2nd hand phone.

I guess it’s debatable relevance to solorpunk travel. Two phones in case 1 consumes a little more power overall but it keeps a phone out of the landfill and makes it useful.

update

Found an f-droid app that looks good for this. It will even run on AOS 2 which means quite old phones can be used to feed GPS coords over BT. This app could be useful as well.

Question: I always disagree to “Google’s location service” nag -- (using towers and/or wifi APs) to supplement navigation (no idea what gets shared with Google and also don’t want wifi or GSM eating battery).. but if a separate phone is feeding the fix, then the power problem goes away. But there’s still the sharing problem. Is there a way to harvest the tower info before a trip anonymously and use it without feeding Google?

update 2

I tried using an external bluetooth GPS device -- one that is dedicated to that purpose from the palm pilot days. I was able to pair to it over bluetooth but after pairing it would not connect to it for any kind of session. It’s as if the android does not know what to do with a GPS server.

Some instructions out in the wild say: “In the Android playstore fetch ‘bluetooth GPS’ or ‘bluetooth GNSS’ App.” Well, I don’t do Playstore.

One step is to go into settings → “Developer options” → Debugging → Allow Mock location → enable. That makes no difference for me.

The instructions also say: “Before you launch your GPS software, launch ‘bluetooth GPS/GNSS’, click “connect” and check “Enable Mock GPS Provider” -- which is a non-starter for those not inside Google’s walled garden. Guess I need a free-world variation of this app which apparently uses the external GPS device to feed a mock location. I found these two apps:

• GPSTest - this is an apparently useful test app but seems unable to use external devices
• RtkGps (abandoned¹) - claims to make a connection over bluetooth to an external GPS, but does not work for me. Mentions SiRF IV but not SiRF III, which may be my problem. IIRC, RTK was a SiRF III competitor.

¹ This repository has been archived by the owner on Mar 28, 2023. It is now read-only.

Banks and credit unions spam me with ads of their services, some of which are quite high volume. I never gave my express informed consent. I may have consented to this buried in some fine print, but certainly was not asked for consent in a manner that would make me consciously aware that my inbox will be attacked with ads.

My problem is not really the annoyance. I can probably go through some opt-out hoops. The problem with banks specifically is security. Every time the bank e-mails you for any reason, both the metadata and the payload data are sent in-the-clear, thus enabling all handlers of the email to know where you bank. This info is valuable to both debt collectors and thieves.

So, new rule for non-GDPR regions of the world:

Banks that e-mail customers must very loudly obtain your unquestionably informed consent. The bank must give you a separate doc that says:

“Bank X will certainly send a flood of spam, and that flood of spam will disclose where you bank to all email providers and potentially ISPs and e-mail forwarding providers. All recipients free to sell that data to debt collectors. Show that you wholly agree to this abuse below by hand-writing out ‘please feel free to abuse my e-mail address’ and signing that statement.”

There must be a picture of a big eye or a zorro mask or cyber criminal with a hoodie next to that agreement (inspired by cigarrette box rules).

I believe if that level of transparency were in play, people would not agree and banks would either have to offer an email-free option or they would lose business.

#fuckBanks

I think it was the prime minister (or spokesperson) who made this very clever argument: (paraphrasing) “we are not taking away choice… cigarettes are designed to inherently take away your choice by trapping you in an addiction.”

I’m not picking sides here, just pointing out a great piece of rhetoric to spin the policy as taking away something that takes away your choice. Effectively putting forward the idea that you don’t have choice to begin with.

(sorry to say this rhetoric was not mentioned in the linked article; I just heard it on BBC World Service)

cross-posted from: https://slrpnk.net/post/4687232

I’ve boycotted Coca-Cola & Pepsi products for over a decade. All the non-alcoholic drinks at my workplace cafeteria were Coke products. Even the orange juice (Minute Maid is Coke).

So I complained… saying directly¹ to the outsourced catering company that Coke is contrary to local values and that we should have at least one ethical option, while at the same time stressing that bringing in Pepsi products would not solve the problem. I said I’m currently limited to water, beer, and wine. And obviously when I choose tap water they make nothing on that then they have to wash my glass.

They replied to say they’ve decided to bring in more drink options. Couple weeks later they had Arizona iced tea and various coconut water kinds with aloe vera. And I noticed lots of people buying them. There’s still the problem of plastic waste from the containers but getting some people off Coke was a bigger stride to make IMO.

Coke’s wrong-doings are only fractionally environmental, but I wanted to mention it here because the story demonstrates how a simple 1-person action can sometimes scale beyond just one individual. AFAIK, I was the only one to complain about the Coke monopoly.

Note that only the few colleagues I mentioned this to know it was boycott-driven. People buying non-Coke drinks were simply taking what they wanted with no idea that an anti-Coke boycott action lead to more options. The ease of it is notable. I did not have to undertake the big effort of rallying a crowd.

1. indeed I took the liberty to contact the catering company directly, bypassing my employer who actually had the contract with the catering company. It caused no issue. I guess it was clear enough that I was just an employee and not acting on behalf of the employer.

I ran this command to see if the PDF menu was small enough for my capped internet connection:

$ torsocks curl -LI 'https://cafevanbommel.nl/wp-content/uploads/2023/11/Van-Bommel-Menukaart-November-2023-FOOD.pdf'
HTTP/2 200 
date: Tue, 09 Apr 2024 16:01:40 GMT
content-length: 1480
cache-control: no-cache, no-store, must-revalidate, max-age=0
cache-control: no-store, max-age=0
server: imunify360-webshield/1.21

PDF was only 1k, so of course I have no objections. Fetched it using wget, and it was just ASCII text in the form of HTML-wrapped javascript. WTF?

<!doctype html>
<html lang="en">
<head>
    <meta charset="utf-8">
    <meta name="robots" content="noindex, nofollow">
    <title>One moment, please...</title>
    <style>
    body {
        background: #F6F7F8;
        color: #303131;
        font-family: sans-serif;
        margin-top: 45vh;
        text-align: center;
    }   
    </style>
    </head>
<body>
    <h1>Please wait while your request is being verified...</h1>
    <form id="wsidchk-form" style="display:none;" action="/z0f76a1d14fd21a8fb5fd0d03e0fdc3d3cedae52f" method="GET">
    <input type="hidden" id="wsidchk" name="wsidchk"/>
    </form>
    <script>
    (function(){
        var west=+((+!+[])+(+!+[]+!![]+!![]+[])+(+!+[]+!![]+!![]+!![]+!![]+!![]+!![])+(+!+[]+!![]+[])+(+!+[])+(+!+[]+!![]+[])+(+!+[]+!![]+!!
[]+!![])+(+!+[]+!![]+!![]+!![]+!![]+[])),  
            east=+((+!+[]+!![]+!![]+!![]+!![]+!![])+(+!+[]+!![]+!![]+!![]+!![]+!![]+[])+(+!+[]+!![]+!![]+!![]+!![]+!![]+!![]+!![])+(+!+[]+!!
[]+!![]+!![]+!![]+!![]+!![]+[])+(+!+[])+(+!+[]+!![]+!![]+[])+(+!+[]+!![]+!![]+!![]+!![]+!![]+!![]+!![]+!![])),
            x=function(){try{return !!window.addEventListener;}catch(e){return !!0;} },
            y=function(y,z){x() ? document.addEventListener('DOMContentLoaded',y,z) : document.attachEvent('onreadystatechange',y);};
        y(function(){
            document.getElementById('wsidchk').value = west + east;
            document.getElementById('wsidchk-form').submit();
        }, false);
    })();
    </script>
</body>
</html>

To troubleshoot, I loaded the same link in a GUI browser. PDF.js fetched a proper PDF that turned out to be 1.6mb. Fuck this shit. It’s not as bad as some restaurants (~20mb menus loaded with pics), but still, it could have sucked my credit dry because the asshole web dev pulled this shit. The content-length header exists for a reason.

I wonder to what extent the restaurant’s web admin is just naive about what’s happening, considering the “imunify360” in the header, which suggest some shitty MitM might have done this without the Wordpress user really knowing.

But what’s driving the protectionism? I should be able to, for example, have a scraper bot harvest all the PDF restaurant menus before visiting a region. They should want my business.

The linked article states:

“You always have the right to a minimum 2-year guarantee if the digital content or service turns out to be faulty, not as advertised or not working as expected.”

IIUC, this means if a service is paired with software, and the API + software employs #forcedObsolescence mid-contract, they must fix or refund. Thus two example scenarios come to mind:

• If you were to pay ProtonVPN for premium service in the year leading up to June 2021 and you ran AOS 5, you would have lost service after less than an annual subscription period. ProtonVPN would have to remedy it under EU law.
• If your bank charges annual fees and they push a forced upgrade at any time that obsoletes your platform (so you cannot use the forced upgrade), the bank might be in violation of this EU consumer protection law.

Images can be fully embedded inline directly in the HTML. Tor Browser displays them unconditionally, regardless of the permissions.default.image setting, which if set to “2” indicates images should not be loaded.

An example is demonstrated by the privacy-respecting search service called “dogs”:

• clearnet
• onion

If you search for a specific object like “sweet peppers”, embedded images appear in the results. This feature could easily be abused by advertisers. I’m surprised that it’s currently relatively rare.

It’s perhaps impossible to prevent embedded images from being fetched because the HTML standard does not include the length of the base64 blob ahead of it. Thus no way for the browser to know which position in the file to continue fetching from.

Nonetheless, the browser does not know /why/ the user disables images. Some people do it because they are on measured rate connections and need to keep their consumption low, like myself, and we are fucked in this case. But some people disable images just to keep garbage off the screen. In that case, the browser can (and should) respect their choice whether the images are embedded or not.

There should really be two config booleans:

• fetch non-local images
• render images that have been obtained The first controls whether the browser makes requests for images over the WAN. The second would just control whether the images are displayed.