It's uniquely hard to get rid of Android (when you don't consider iOS as an option, because possibly the remedy would be worse than the issue). A degoogled Android is already a good step.
I'm as little reliant on proprietary services as possible and I'll jump to Linux mobile (having used a pinephone for one year as my main phone, I know it'll be comfortable for me) when a convincing phone that's not very expensive given the outdated hardware it has and that doesn't rely on an Android base can run it smoothly enough, but that phone doesn't exist yet.
Alternatives include not having a smartphone but the network effect and the expectations on you having one are quite strong.
The Fairphone 5 I have is very close to run PostmarketOS well, I have hope but for now few but critical features are missing.
Or you could self host (of course that costs time and you need to have a strong backup strategy... but on the other hand you should also backup what you host at the gafam, you never know when they decide to lock you out for some reason).
It's kinda sad to flee from one gafam just to trap yourself in another one. However, congrats for your successful degoogling!
I got an iPad 2 found in a drawer that would be perfectly good if the software could at least be easily unlocked so you could only hope to install an alternative OS since they don't want to update it themselves. Or if they allowed installing stuff from outside their store.
But no, the thing runs an ancient WebKit and that's pretty much your only way to run custom code on this stuff.
I can't even create an apple account from this stuff anymore, and apple won't let me create an apple account from any other device I have. Which could possibly help with jail breaking it. And it would already be a pain to be able to create an apple account because they require a name, a birth date, a phone number, an email address... All this to access a store from which nothing would work anyway for an old device like this.
So, fuck them.
And android increasingly becomes like this too.
Meanwhile, you can still run an up to date Linux on a laptop from the 90s.
I'm quite fine with no comments but correctly named variables and functions. This can't become out of sync contrarily to "out of band" comments. I take this over commented code with poorly named stuff any day.
I've also seen a lot of comments that restate what the code already says and that's just noise, more work to keep in sync, an additional thing that can fail, and more cognitive load because you have to read twice the same thing (best case, if code and comment are still in sync). That's the result you risk when you think you must comment your code.
I appreciate the occasional comment that explains why something seems overly tricky or weird or not immediately intuitive. Once, I had left such a comment that saved myself years later from making a mistake. Of course, this should be kept at a minimal level. It leads to me liking clear code with few comments the most. (Some guidelines, even if it's not perfect, to limit complexity and spaghetti code help a lot).
Function, class, module documentation is also useful so you don't have to read the whole thing and you know what it's intended to provide (which is slightly different than simply what it provides, and this differences is important).
>> the fact you didn't know about the existing private open source offline mapping solution also rang an alarm in me
> why? that's a bit of an overreaction from your alarm bells honestly.
Maybe. Let me explain myself. In a former life, I worked in academia and you kinda triggered my research mode (which is, believe it or not, quite positive for you). There, your work is easily more solid and convincing if you reaserched a bit the state of the art. By working on an alternative to GPS navigation, in such a setting, I would expect you to know the field well.
Now, I understand that you didn't particularly look into apps pretty much focused on GPS positioning since they are not supposed to do what you want, and I'm aware that you aren't doing academic research nor pretending to.
> let me add it today haha. thank you.
No pressure, I'll try to think of coming back to check once in a while. Thank you as well.
> I am not sure if you looked at the repo I linked
I did skim it to get a grasp of your approach. But I'm answering statements made here too.
I think your project is worth pursuing. I'm genuinely curious what you come up with. This is actually why I spent quite some time writing my comments: I hope they helped you in some way. I do believe your project could use clearer presentation of the motivation (here or in the readme). In particular, the "GPS is not private" statement and the fact you didn't know about the existing private open source offline mapping solution also rang an alarm in me (a comparison with the "state of the art" would be nice). "Imagine GPS is jammed / is not available at this time at your place with your hardware" is actually sufficient as far as I am concerned. Relying on a privacy-preserving solution out of principles even on a spy phone comes with caveats that need to be exposed but I believe in this approach.
I'm not pushing back. If you have something like a blog that I could follow, I would subscribe to its RSS feed because I'm quite interested actually. If not, well it's already quite nice you share your curiosity work with the world.
I occasionally contribute to OSM. You just send your edit and they are directly live. I don't think there's prior moderation. On these apps you log in to OSM using an OSM account IIRC (though I mostly use EveryDoor or directly osm.org for this, occasionally StreetComplete, I don't contribute using Comaps)
If by AGPS you are referring to the use of nearby WiFi and cellular networks and querying a service for that (there are ways to have that info purely offline, but I do believe such databases rely on many people leaking their positions to exist at all), I'm not convinced your approach helps much:
1) if you run stock iOS or Android with the Google services, I believe you are already screwed anyway: I wouldn't trust them not to constantly leak stuff about you in the background even if you don't actively use the GPS, at least to build their traffic info and to keep their access network-based location database updated. If you are concerned about the privacy aspect of this stuff, you should be getting rid of the Google services anyway. That leads to the next case:
2) If you are running a phone without the Google services, then you don't have that kind of AGPS. Unless you go the extra mile of explicitly configuring microG to enable network-based location. I used to do this, but on my current phone, getting the position purely from GPS has been so reliable and quick that I didn't bother setting this up (I suppose it could provide some battery saving, but there are privacy concerns indeed, and not relying on such databases that basically depend on people giving up bits of their privacy is always nice). Your main concern there would be the shady blobs most phones run that could be doing you don't know what behind your back, but that's regardless how you use your phone.
If by AGPS you mean the download of the data about the positions of the GPS satellites (MSA) then yes, indeed, I agree, you will leak to your service provider (but not more than if you are not already fully offline - so basically if you use your phone like a phone you already leak as much) and to the server providing the data (which is probably Qualcomm (XTRA) or Google, or possibly your service provider). Then the solution is fully offline, and your current options are non assisted GPS, which can take 10 minutes in good conditions to get a fix if you haven't used it in a while, and reading the map. From what I understand, your current strategy is to help with the latter, which needs to be better than eventually getting a GPS fix (which feels somewhat niche, but I can see some advantages).
(There's also AGPS MSB where you basically ask a remote server to compute your location from what your GPS receiver gets and nearby cellular networks, the leak here is obvious but from what I've read this is quite rare nowadays, it was more useful when GPS receivers didn't have enough compute power to do the compilations themselves, so that would be mostly irrelevant.)
All in all, if I'm not wrong (happy to be proven wrong): as far as privacy is a concern, depending on what you mean by AGPS, your approach only applies to, respectively, devices without the Google Play Services or similar stuff, or to fully offline devices. It needs to be better than, respectively, GPS assisted with the satelite positioning data (which can be good for up to 7 days and then updated through regular GPS if I'm not wrong), or regular GPS. I believe your angle of attack can be battery usage, or better than the time to get a GPS fix. That seems quite tough especially that your approach as I understand it is intrinsically scoped (works mostly in cities where it's easy to ask about what's around).
I believe a stronger angle of your approach is resilience: less dependency on the GPS system which can indeed be jammed or which can fail (although there's redundancy: we do have several constellations of satellites from different countries, so a general failure seems unlikely, and a GPS failure would be catastrophic because many critical things depend on it).
Now, there are devices or places (especially in dense cities with high buildings) where the GPS simply doesn't work well, so a convenient, privacy-preserving alternatives to the GPS would be great in those situations6!
> It's not fully private if it requires GPS connection though!!
How so? GPS is like FM radio: you send nothing, you only receive.
Apps like organic maps or comaps let you use the maps fully offline and you can compute itineraries without GPS when your need this (from point A to point B, with as many stops as you wish).
I strongly recommend you to seriously look into comaps or organic maps if you don't know them.
Now, "GPS isn't working or depletes my battery, what do I do?" is an interesting topic worth looking into. It seems you are trying to automate what we all do when GPS doesn't work well. I find that relatively easy in a city, not so much in a road on the countryside.
In France we just turned off nuclear plants due to the heatwave. Heatwave probably in part caused by climate change, itself caused by us being already too power hungry.
We could do without AI for a bit maybe and focus on actual needs and on reducing our power hungriness instead.
For ASML and Carl Zeiss (which I didn't know about), it seems like a stretch from what I can read about them.
But for Spotify, why not?
> It’s a very thin and a political line between being a gatekeeper and a very successful company.
Of course.
If you are a tech company that becomes as successful as to be a monopoly or a participant in an oligopoly with a strong network effect, why wouldn't you be recognized as a gatekeeper?
> But this isn't about the iPad or the iPod -- it's about product design.
Obviously GenAI. The author time-traveled to us, stole that sentence and put it in his article. He got encoding issues on blogspot so he typed the dash himself.