On 05/19/17 17:52, Zeeshan Ali (Khattak) wrote:
> On 19 May 2017 at 17:49, Zeeshan Ali (Khattak) <[email protected]> wrote:
>> Hi,
>>
>>
>> On 19 May 2017 at 15:17, Jan-Tarek Butt <[email protected]> wrote:
>>> On 05/19/17 14:01, Zeeshan Ali (Khattak) wrote:
>>
>> ..snip
>>
>>> Ok I´ll change the plan, thank you for giving me the hint of redundancy 
>>> project.
>>
>> Thanks for being so extremely flexible and cooperative. :) I rarely
>> see people being so rational so I wasn't expecting so much.
>>
>>> The old plan was:
>>>
>>> In the weak project over view I plan to divide this Project into 3 or 4 
>>> sub-projects.
>>>
>>> The first one is a restful API backend whit should include a backwards 
>>> compatibly to
>>> the old openwifi web backend. May written in Go. The New API should use MLS 
>>> as fallback
>>> if it not passible to determinate a position based on the own db entrys. As 
>>> well received
>>> WiFi information's can also redirect to MLS (if they want that).
>>>
>>> The next is a kernel module which should communicate with the restful API 
>>> and provided
>>> a device driver under /dev as a tty driver. The Linux kernel provides GPS 
>>> hardware as
>>> tty devices. This software defined tty device should print GPS format e.g. 
>>> NMEA 0183.
>>> The goal of this kernel module is that programs like gpsd and other can 
>>> easily use
>>> this standard tty device as a normal GPS hardware. The position detection 
>>> continues
>>> over Wifi. The advantaged of this module is that devices like laptops or 
>>> routers can
>>> have simulated GPS hardware.
>>>
>>> The last one a library should convert the NMEA 0183 format
>>> to long-/latitude and print that.
>>>
>>> Here is my new plan:
>>>
>>> The first one is a restful API backend whit should include a backwards 
>>> compatibly to
>>> the old openwifi web backend. May written in Go. The New API should use MLS 
>>> as fallback
>>> if it not possible to determinate a position based on the own db entrys. As 
>>> well received
>>> WiFi information's can also redirect to MLS (if they want that).
>>
>> Sounds fine to me. Is this supposed to be run on a remote server?
>>
>>> The Kernel module will be dropt because of new infromations from geoclue.
>>> We don't wanna do redundancy project. ;)
>>
>> Cool. Thanks.
>>
>>> geoclue: I discuss with Zeeshan, the maintainer of geoclue, to build 
>>> support in geoclue for
>>> standalone GPS.
>>
>> Great! Sorry I wasn't very clear about this I think. I'm already
>> slowly adding this support through an external means:
>> https://github.com/zeenix/gps-share . Geoclue itself won't need any
>> modification since it already support network NMEA sources. So if you
>> would want to help me with this goal, gps-share is what we should
>> collaboratively work on. I'm attaching my vague TODO for gps-share.
>>
>> Just to remind, gps-share is written in Rust for ensuring reliability.
>>
>>> Hopefully the new plan is OK with you or do you have so other ideas? :)
>>
>> Sure, looking forward to working with you. :)
> 
> Oh and if you need more direct interaction, IRC channel #gnome-maps on
> irc.gnome.org would probably be the best place.

Thanks, I will join there soon.

Attachment: signature.asc
Description: OpenPGP digital signature

_______________________________________________
dev-geolocation mailing list
[email protected]
https://lists.mozilla.org/listinfo/dev-geolocation

Reply via email to