113 karma · joined August 26, 2021
They also recently split the automobile UI part off from the Phone bits. That joint work was part of the problem for FOSSing everything, since they have deals with Car manufacturers which depend on their IP.
It's for this reason I like SFOS. I've tried android and ios. But they suck.
As a developer, I also appreciate the flexibility even within the limits. Gradle and co. suck.
Ah, I have to admit I don't know much about porting :) We did put up a community wiki and the HADK is up there: https://sailfishos.wiki/books/hadk/page/hadk so that you might get some idea of the process ... I'm not sure to what extent this is different from that 'halium' builds that piggz makes. It makes sense to ask him directly, since he has a stream lined process that targets, among others, the phones you have.
As to why the hardware adaption layer is needed (ie. the android one) it's where all the binary blobs are :) Among other bits.
And, yes, I often turn off wifi. I never go over my Data limits and 4G/5G is much more efficient for some reason.
I must admit, I don't do banking on the phone and keep all sensitive data off my mobiles.
From being able to do SDL2 stuff (Godot 3.5 for instance) with a wayland compositor to just not having f'ked UI, I'm happy. The challenge of getting more apps is certainly a challenge. But that's good, gives me lots to do.
Linux on Mobile has already long been here to stay.
Veritably is, as yet, a bunch of prototyping looking to become a company. The grand scheme is to become a general document validation/verification automation pipeline. Think, ISO standards compliance.
We are currently a two person team working with VisionEncoder/Decoder Models and traditional algorithms (mostly python) to assess the completeness and correctness of documents like ISO 26262 Hazard and Risk assessments.
We are looking for engineers with an interest in document processing, standards compliance, Functional Safety and the like to join us in solving some ugly real world problems. Let's face it, modern companies tend to produce documentation in a fairly random mix of 'looks about right' PDFs and excel (tm) spreadsheets. This is time consuming, tedious work that really could use the application of some intelligent machine processing.
We're looking for partners, not employees. If your curious, mail: mark.washeim@veritably.ai
Using https://github.com/ArashPartow/exprtk I have one calculator with a repl at 512kb, including a QML wrapper. I have a combined rpn calculator using sympy (based on the work of others) with exprtk as a 'programmer's calculator' on the side (https://github.com/poetaster/fibonacci). And several QML apps for symbolic CAS using sympy.
I have way too many calculators. They are so much fun to build. Now I ask myself, port the Qalculate methods to macros for exprtk or check out using libqalculate? I'm not sure. exprtk is so light? And adding macros discrete.