Serval Adoption

24 views
Skip to first unread message

Klaus Wuestefeld

unread,
Sep 7, 2019, 7:51:19 PM9/7/19
to serval-proje...@googlegroups.com
Hi, I'm the author of the dormant Sneer (Sovereign Computing) project. http://sneer.me/project.html

I met Dr Paul in Barcelona in 2012.

I wonder: are you guys happy with Serval adoption or would you like for it to be more widely adopted?

Cheers, Klaus


Paul Gardner-Stephen

unread,
Sep 7, 2019, 10:56:06 PM9/7/19
to Serval Project Developers
Hi Klaus,

We would obviously like to get Serval to the point where it will be more widely used.  There is a bit of software development work that we need to do to enable that, however.

Paul.

--
You received this message because you are subscribed to the Google Groups "Serval Project Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to serval-project-dev...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/serval-project-developers/CAMAooZE9%3DcKzLbsr3rC6VDzKoPpdyuRzu9yO%2Ba5b-ceQooUk1w%40mail.gmail.com.

Klaus Wuestefeld

unread,
Sep 8, 2019, 1:23:39 AM9/8/19
to serval-proje...@googlegroups.com
Hi, What do you think is the bit that's missing?
> To view this discussion on the web visit https://groups.google.com/d/msgid/serval-project-developers/CA%2B_T8-Cb-YcUtTyLDCjGZ--L8khVMkHkYQeO7gx7n6%3Dq3wUAJQ%40mail.gmail.com.

Paul Gardner-Stephen

unread,
Sep 8, 2019, 2:29:17 AM9/8/19
to Serval Project Developers
Hello,

We have some known bugs in the Rhizome LBARD synchronisation code and radio handling, as well as needing to update the app on the Play store.  The current plan is to do a fair bit of work on this early in the new year, if not sooner.

Paul.

muebau

unread,
Sep 8, 2019, 6:31:55 AM9/8/19
to Serval Project Developers
Hello,
I found the alpha version of the new Serval version. Unfortunately it crashes strait away (Android 9). Is there a way to help with some logs or something?

Ulf

On Sunday, 8 September 2019 08:29:17 UTC+2, Paul Gardner-Stephen wrote:
Hello,

We have some known bugs in the Rhizome LBARD synchronisation code and radio handling, as well as needing to update the app on the Play store.  The current plan is to do a fair bit of work on this early in the new year, if not sooner.

Paul.

On Sun, 8 Sep 2019 at 14:53, Klaus Wuestefeld <klauswu...@gmail.com> wrote:
Hi, What do you think is the bit that's missing?

On Sat, Sep 7, 2019 at 11:56 PM Paul Gardner-Stephen
<pa...@servalproject.org> wrote:
>
> Hi Klaus,
>
> We would obviously like to get Serval to the point where it will be more widely used.  There is a bit of software development work that we need to do to enable that, however.
>
> Paul.
>
> On Sun, 8 Sep 2019 at 09:21, Klaus Wuestefeld <klauswu...@gmail.com> wrote:
>>
>> Hi, I'm the author of the dormant Sneer (Sovereign Computing) project. http://sneer.me/project.html
>>
>> I met Dr Paul in Barcelona in 2012.
>>
>> I wonder: are you guys happy with Serval adoption or would you like for it to be more widely adopted?
>>
>> Cheers, Klaus
>>
>>
>> --
>> You received this message because you are subscribed to the Google Groups "Serval Project Developers" group.
>> To unsubscribe from this group and stop receiving emails from it, send an email to serval-project-developers+unsub...@googlegroups.com.

>> To view this discussion on the web visit https://groups.google.com/d/msgid/serval-project-developers/CAMAooZE9%3DcKzLbsr3rC6VDzKoPpdyuRzu9yO%2Ba5b-ceQooUk1w%40mail.gmail.com.
>
> --
> You received this message because you are subscribed to the Google Groups "Serval Project Developers" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to serval-project-developers+unsub...@googlegroups.com.

> To view this discussion on the web visit https://groups.google.com/d/msgid/serval-project-developers/CA%2B_T8-Cb-YcUtTyLDCjGZ--L8khVMkHkYQeO7gx7n6%3Dq3wUAJQ%40mail.gmail.com.

--
You received this message because you are subscribed to the Google Groups "Serval Project Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to serval-project-developers+unsub...@googlegroups.com.

yanosz

unread,
Sep 8, 2019, 6:42:55 AM9/8/19
to serval-proje...@googlegroups.com
Heiho,


Am 08/09/2019 um 08.29 schrieb Paul Gardner-Stephen:
> Hello,
>
> We have some known bugs in the Rhizome LBARD synchronisation code and radio
> handling, as well as needing to update the app on the Play store. The
> current plan is to do a fair bit of work on this early in the new year, if
> not sooner.

Sounds cool :-).

I'm also thinking of using serval messaging in fieldtracks.org - having
not found a suitable messaging application, yet.

IMHO creating an unstructured (aka mesh) wide area network is not such a
big thing nowadays - a lot of stuff exists. (directed wireless links,
sub-Ghz radios, satellite uplinks, DSL lines, etc.)

But .. an Android App providing an acceptable user experience (UX) on
different device-models is very challenging:
This includes a lot of maintenance work (Android changes frequently) and
a solid understanding of Android's limitations (e.g. energy consumption
& saving, disconnect prompt if a wifi-ap has no internet vs. wifi-p2p on
different devices).

Is there a roadmap, yet?

Greetz, yanosz

--
There's a ripped off cord
To my TV screen
With a note saying:
"Im not afraid to dream"
-- Donkey Boy, Crazy Something Normal

Paul Gardner-Stephen

unread,
Sep 8, 2019, 8:18:55 AM9/8/19
to Serval Project Developers
Hello,

Yes, please provide logcat output so that we can figure out what is going wrong.

Paul.

On Sun, 8 Sep 2019 at 20:01, muebau <mue...@gmail.com> wrote:
Hello,
I found the alpha version of the new Serval version. Unfortunately it crashes strait away (Android 9). Is there a way to help with some logs or something?

Ulf
On Sunday, 8 September 2019 08:29:17 UTC+2, Paul Gardner-Stephen wrote:
Hello,

We have some known bugs in the Rhizome LBARD synchronisation code and radio handling, as well as needing to update the app on the Play store.  The current plan is to do a fair bit of work on this early in the new year, if not sooner.

Paul.

On Sun, 8 Sep 2019 at 14:53, Klaus Wuestefeld <klauswu...@gmail.com> wrote:
Hi, What do you think is the bit that's missing?

On Sat, Sep 7, 2019 at 11:56 PM Paul Gardner-Stephen
<pa...@servalproject.org> wrote:
>
> Hi Klaus,
>
> We would obviously like to get Serval to the point where it will be more widely used.  There is a bit of software development work that we need to do to enable that, however.
>
> Paul.
>
> On Sun, 8 Sep 2019 at 09:21, Klaus Wuestefeld <klauswu...@gmail.com> wrote:
>>
>> Hi, I'm the author of the dormant Sneer (Sovereign Computing) project. http://sneer.me/project.html
>>
>> I met Dr Paul in Barcelona in 2012.
>>
>> I wonder: are you guys happy with Serval adoption or would you like for it to be more widely adopted?
>>
>> Cheers, Klaus
>>
>>
>> --
>> You received this message because you are subscribed to the Google Groups "Serval Project Developers" group.
>> To unsubscribe from this group and stop receiving emails from it, send an email to serval-project-dev...@googlegroups.com.

>> To view this discussion on the web visit https://groups.google.com/d/msgid/serval-project-developers/CAMAooZE9%3DcKzLbsr3rC6VDzKoPpdyuRzu9yO%2Ba5b-ceQooUk1w%40mail.gmail.com.
>
> --
> You received this message because you are subscribed to the Google Groups "Serval Project Developers" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to serval-project-dev...@googlegroups.com.

> To view this discussion on the web visit https://groups.google.com/d/msgid/serval-project-developers/CA%2B_T8-Cb-YcUtTyLDCjGZ--L8khVMkHkYQeO7gx7n6%3Dq3wUAJQ%40mail.gmail.com.

--
You received this message because you are subscribed to the Google Groups "Serval Project Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to serval-project-dev...@googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "Serval Project Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to serval-project-dev...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/serval-project-developers/cee5a0c5-cef2-461c-abd6-46d60d22a02c%40googlegroups.com.

Paul Gardner-Stephen

unread,
Sep 8, 2019, 8:27:29 AM9/8/19
to Serval Project Developers
Hello,

This is exactly one of the use-cases for Serval that we work with the NZ Red Cross on.

We agree that maintaining and Android app is one of the harder parts.

 But also making a fully self-oganising mesh, so that you don't have to think when you put your various links in place, but just have it magically work, and work around when the network partitions -- these are also significant challenges, and ones where we have put a lot of thinking and effort over the years. This is what the Mesh Extenders do, making what we internally call "radio lego" where you can use all sorts of strange radio links, put LBARD over them, and then just use them with your android phones as though it was all compatible internet links.

Oh, we also have a mostly finished iOS app in the wings.

As for a roadmap, we haven't had the resources lately to do anything formal on that front. We'd welcome collaborators helping with this. We already work with the SEEMOO group at TU-Darmstadt, which I think is not too far from yourselves (compared with Australia, at least ;). Indeed, it seems that you guys are in the neighbouring Bundesländer.

That all said, my plan is at the end of the year I will be based in the Australian Outback to focus on knocking these bugs out and undertaking some serious field testing.  I'll also likely be back and forth to Germany to visit SEEMOO, qaul.net and various other folks. So if you would like to be part of the action and join with us to bring this technology to maturity, we would welcome that.  Obwohl mein Deutsch ist gar nicht Perfekt, ich bin eher bequem auf Deutsch zu arbeiten, wenn es notwending würde. Wir wollen mindestens, dass Serval bisprächig Englisch-Deutsch sein soll.

MfG
Paul.

--
You received this message because you are subscribed to the Google Groups "Serval Project Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to serval-project-dev...@googlegroups.com.

muebau

unread,
Sep 8, 2019, 8:34:26 AM9/8/19
to Serval Project Developers
I will send some logs as soon as I find some time to sort that out.


Paul Gardner-Stephen

unread,
Sep 8, 2019, 8:58:51 AM9/8/19
to Serval Project Developers
thanks!

Paul;.

On Sun, 8 Sep 2019 at 22:04, muebau <mue...@gmail.com> wrote:
I will send some logs as soon as I find some time to sort that out.


--
You received this message because you are subscribed to the Google Groups "Serval Project Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to serval-project-dev...@googlegroups.com.

serval-...@yanosz.net

unread,
Sep 8, 2019, 10:47:16 AM9/8/19
to serval-proje...@googlegroups.com
Heiho,

thanks for your answer :-).

Am 08/09/2019 um 14.27 schrieb Paul Gardner-Stephen:
> Hello,
>
> This is exactly one of the use-cases for Serval that we work with the NZ
> Red Cross on.
>
> We agree that maintaining and Android app is one of the harder parts.
>
> But also making a fully self-oganising mesh, so that you don't have to
> think when you put your various links in place, but just have it magically
> work, and work around when the network partitions -- these are also
> significant challenges, and ones where we have put a lot of thinking and
> effort over the years. This is what the Mesh Extenders do, making what we
> internally call "radio lego" where you can use all sorts of strange radio
> links, put LBARD over them, and then just use them with your android phones
> as though it was all compatible internet links.

Just sharing a few thoughts and experience we had with fieldtracks:

I guess, that most of the mesh-techniques are already available:

Some people from my university started a spin-off for a wireless
backhaul network, focusing on "just works". It provides internet for
rural areas (e.g. in Africa) including DVB-T. Researcher are trying to
integrate LoraWAN modules nowadays, that focus on 868 Mhz / Lora. [1] -
unfortunately, it's not OpenSource and we're not using it.

Some LBARD-style IoT networks have OpenSource components (e.g.
Lora-Server [2]), providing interfaces based on MQTT.

freifunk.net, The Things network (TTN) a some of animal protection
communities have been are tinkering with various outdoor mesh nodes for
years. E.g. designs for TV-white space shifters [3] and MPPT [4] exists
as Open Hardware.

Patching things together is not as easy as pie, but people have done
similar thinks before and you can find tutorials online.

But .. messaging is a different beast, I guess.

If the app is supposed work in the global internet as well as in a
isolated mesh network, it requires a middleware different from firebase
(among others), but still restricted to smartphone's platform.

Providing a decent UX across multiple channels has hardly been done
before. Briar is close to that (imho), but its focus on different users
/ use-cases makes it unusable in our context. Creating a fork won't help
much, I guess. There isn't much to build up on.

Other programs (firechat, qaul, conversations) do not provide a unified
ux over different channels (wifi direct, global internet, etc.) - most
do not support multiple channels at all.

In addition to that, there's to be an active open source community
adapting platform changes of the systems in use.

In fieldtracks, implementing the end-user application consumes about 10
times the effort the mesh network. I guess, that it'll be similar for
serval. Due to this, we ruled out to create our own messaging app in the
foreseeable future.

> Oh, we also have a mostly finished iOS app in the wings.

That's great!

> As for a roadmap, we haven't had the resources lately to do anything formal
> on that front. We'd welcome collaborators helping with this. We already
> work with the SEEMOO group at TU-Darmstadt, which I think is not too far
> from yourselves (compared with Australia, at least ;). Indeed, it seems
> that you guys are in the neighbouring Bundesländer.

Nice ;-).

> That all said, my plan is at the end of the year I will be based in the
> Australian Outback to focus on knocking these bugs out and undertaking some
> serious field testing. I'll also likely be back and forth to Germany to
> visit SEEMOO, qaul.net and various other folks. So if you would like to be
> part of the action and join with us to bring this technology to maturity,
> we would welcome that.

Sounds cool, you can drop by at the C4 [5] or the wireless community [6]
weekend. Battlemesh [7] is also an interesting event to visit. Visiting
the 36c3 [8] is also an option.


> Obwohl mein Deutsch ist gar nicht Perfekt, ich bin
> eher bequem auf Deutsch zu arbeiten, wenn es notwending würde. Wir wollen
> mindestens, dass Serval bisprächig Englisch-Deutsch sein soll.

I'm looking forward to it ;)

Greetz, yanosz

[1] https://www.wiback.org/
[2] https://www.loraserver.io/
[3] https://github.com/elektra42/Freifunk-Libremesh-TVWS-Converter
https://wiki.freifunk.net/images/5/54/Freifunk-in-TV-Whitespace2.pdf
[4] https://github.com/elektra42/freifunk-open-mppt
[5] https://koeln.ccc.de/
[6] https://wiki.freifunk.net/Wireless_Community_Weekend
[7] https://battlemesh.org/
[8] https://events.ccc.de/

john whelan

unread,
Sep 8, 2019, 12:01:41 PM9/8/19
to Serval Project Developers
One major thing to consider is reliability.  One a LAN or closed network Spam and malware should not be a problem.  Once you connect to the internet they become a problem and all the clients need a much greater level of protection and updates.  Old smartphones not running the latest version of Android really are a security problem on the Internet.

Email relayed should be fairly safe but watch the attachments.

Cheerio John

--
You received this message because you are subscribed to the Google Groups "Serval Project Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to serval-project-dev...@googlegroups.com.

Paul Gardner-Stephen

unread,
Sep 8, 2019, 3:28:55 PM9/8/19
to Serval Project Developers
Hello,

See inline.

On Mon, 9 Sep 2019 at 00:17, <serval-...@yanosz.net> wrote:
Heiho,

thanks for your answer :-).

My pleasure.


Am 08/09/2019 um 14.27 schrieb Paul Gardner-Stephen:
> Hello,
>
> This is exactly one of the use-cases for Serval that we work with the NZ
> Red Cross on.
>
> We agree that maintaining and Android app is one of the harder parts.
>
>  But also making a fully self-oganising mesh, so that you don't have to
> think when you put your various links in place, but just have it magically
> work, and work around when the network partitions -- these are also
> significant challenges, and ones where we have put a lot of thinking and
> effort over the years. This is what the Mesh Extenders do, making what we
> internally call "radio lego" where you can use all sorts of strange radio
> links, put LBARD over them, and then just use them with your android phones
> as though it was all compatible internet links.

Just sharing a few thoughts and experience we had with fieldtracks:

I guess, that most of the mesh-techniques are already available:

Some people from my university started a spin-off for a wireless
backhaul network, focusing on "just works". It provides internet for
rural areas (e.g. in Africa) including DVB-T.

Internet over a few hops is very different to a robust DTN. Of course, I love to see advances on both fronts :)
 
Researcher are trying to
integrate LoraWAN modules nowadays, that focus on 868 Mhz / Lora. [1] -
unfortunately, it's not OpenSource and we're not using it.

We have active work at the moment on including low-level LoRa waveform radios into the Mesh Extenders.  No LoRaWAN or other proprietary bits, just using it as a nice packet radio. Current student is using the MicroChip modules with weird closed firmware, but then one of the SEEMOO folks will come out early next year and work on the little 2€ SPI modules for a cheaper and more open solution, probably with an onboard Lattice FPGA that supports open tools for the SPI to UART interface.
 

Some LBARD-style IoT networks have OpenSource components (e.g.
Lora-Server [2]), providing interfaces based on MQTT.

freifunk.net, The Things network (TTN) a some of animal protection
communities have been are tinkering with various outdoor mesh nodes for
years. E.g. designs for TV-white space shifters [3] and MPPT [4] exists
as Open Hardware.

Again, I assume real-time internet-like services, going by what I know of Elektra's work on TVWS, rather than a DTN designed for a "normally partitioned" network? But again, I love what these folks are all doing.


Patching things together is not as easy as pie, but people have done
similar thinks before and you can find tutorials online.

But .. messaging is a different beast, I guess.

Yes, if you want it to be super robust for networks that are not only "normally partitioned", but "never connected"/"always partitioned". This is what Rhizome tackles as the underlying transport for our messaging services: MeshMS (like SMS) and MeshMB (like Twitter/Facebook).
 

If the app is supposed work in the global internet as well as in a
isolated mesh network, it requires a middleware different from firebase
(among others), but still restricted to smartphone's platform.

Yes, and we haven't tackled this piece of it yet. We want to get clusters of devices on the edge happily and reliably messaging first, then get back to that problem.


Providing a decent UX across multiple channels has hardly been done
before. Briar is close to that (imho), but its focus on different users
/ use-cases makes it unusable in our context. Creating a fork won't help
much, I guess. There isn't much to build up on.

Yes, Briar are doing some great work, but their use-cases are a litle different.
 

Other programs (firechat, qaul, conversations) do not provide a unified
ux over different channels (wifi direct, global internet, etc.) - most
do not support multiple channels at all.

Right.


In addition to that, there's to be an active open source community
adapting platform changes of the systems in use.

In fieldtracks, implementing the end-user application consumes about 10
times the effort the mesh network. I guess, that it'll be similar for
serval. Due to this, we ruled out to create our own messaging app in the
foreseeable future.

We'd love to join forces to get rid of the duplication in the app dev part.

> Oh, we also have a mostly finished iOS app in the wings.

That's great!

> As for a roadmap, we haven't had the resources lately to do anything formal
> on that front. We'd welcome collaborators helping with this. We already
> work with the SEEMOO group at TU-Darmstadt, which I think is not too far
> from yourselves (compared with Australia, at least ;). Indeed, it seems
> that you guys are in the neighbouring Bundesländer.

Nice ;-).

> That all said, my plan is at the end of the year I will be based in the
> Australian Outback to focus on knocking these bugs out and undertaking some
> serious field testing.  I'll also likely be back and forth to Germany to
> visit SEEMOO, qaul.net and various other folks. So if you would like to be
> part of the action and join with us to bring this technology to maturity,
> we would welcome that. 

Sounds cool, you can drop by at the C4 [5] or the wireless community [6]
weekend. Battlemesh [7] is also an interesting event to visit. Visiting
the 36c3 [8] is also an option.

Yes, there are plenty of opportunities, it all comes down to cost, and leaving my poor long-suffering family at home too much and too often. I'll have a better idea of my movements later in the year. Congress is always hard, because it falls in the Christmas timeframe, when it is especially hard for me to travel due to family committments.  I need a rich shiek who goes by the motto "a man and his family are inseparable" and who will pay for the whole family to travel with me each time ;)
 


> Obwohl mein Deutsch ist gar nicht Perfekt, ich bin
> eher bequem auf Deutsch zu arbeiten, wenn es notwending würde. Wir wollen
> mindestens, dass Serval bisprächig Englisch-Deutsch sein soll.

I'm looking forward to it ;)

Greetz, yanosz

[1] https://www.wiback.org/
[2] https://www.loraserver.io/
[3] https://github.com/elektra42/Freifunk-Libremesh-TVWS-Converter
https://wiki.freifunk.net/images/5/54/Freifunk-in-TV-Whitespace2.pdf
[4] https://github.com/elektra42/freifunk-open-mppt
[5] https://koeln.ccc.de/
[6] https://wiki.freifunk.net/Wireless_Community_Weekend
[7] https://battlemesh.org/
[8] https://events.ccc.de/
--
There's a ripped off cord
To my TV screen
With a note saying:
"Im not afraid to dream"
-- Donkey Boy, Crazy Something Normal

--
You received this message because you are subscribed to the Google Groups "Serval Project Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to serval-project-dev...@googlegroups.com.

Paul Gardner-Stephen

unread,
Sep 8, 2019, 3:29:36 PM9/8/19
to Serval Project Developers
See my above comments: EVERYTHING needs to be authcrypted if you want to have a sensible DTN.

Paul.

yanosz

unread,
Sep 9, 2019, 4:34:29 AM9/9/19
to serval-proje...@googlegroups.com
Heiho,


Am 08/09/2019 um 21.28 schrieb Paul Gardner-Stephen:
> On Mon, 9 Sep 2019 at 00:17, <serval-...@yanosz.net
> <mailto:serval-...@yanosz.net>> wrote:
[..]
> Some people from my university started a spin-off for a wireless
> backhaul network, focusing on "just works". It provides internet for
> rural areas (e.g. in Africa) including DVB-T.

Yepp - I guess, that's different from fieldtracks: Tracking measurements
have to fulfill realtime requirements. If position information arrives
too late to be displayed on a map, it can easily be discarded .. but
that's not such a big issue, if the incident is shown to the user by
marking a node offline.

>
> Internet over a few hops is very different to a robust DTN. Of course, I
> love to see advances on both fronts :)
>  
>
> Researcher are trying to
> integrate LoraWAN modules nowadays, that focus on 868 Mhz / Lora. [1] -
> unfortunately, it's not OpenSource and we're not using it.
>
>
> We have active work at the moment on including low-level LoRa waveform
> radios into the Mesh Extenders.  No LoRaWAN or other proprietary bits,
> just using it as a nice packet radio. Current student is using the
> MicroChip modules with weird closed firmware, but then one of the SEEMOO
> folks will come out early next year and work on the little 2€ SPI
> modules for a cheaper and more open solution, probably with an onboard
> Lattice FPGA that supports open tools for the SPI to UART interface.

Cool :-).
 
[...]
>
> Patching things together is not as easy as pie, but people have done
> similar thinks before and you can find tutorials online.
>
> But .. messaging is a different beast, I guess.
>
>
> Yes, if you want it to be super robust for networks that are not only
> "normally partitioned", but "never connected"/"always partitioned". This
> is what Rhizome tackles as the underlying transport for our messaging
> services: MeshMS (like SMS) and MeshMB (like Twitter/Facebook).
By intuition, I wouldn't expect that a DTN must have much impact on osi
layers 1-3. Given a suitable middleware (MQTT is a strange, but
interesting choice, I guess) you can do things within the application.

At my university, some military minded people from Fraunhofer FKIE even
use JMS (ActiveMQ, afair) and enterprise'ish stuff .. but I like to stay
away from 'em ;-).

Of course, you can do things "in the network", but that might turn out
to be a research project hardly used in practice. That's fine for
research - it just doesn't help me to build a real network with low effort.
 
>
> If the app is supposed work in the global internet as well as in a
> isolated mesh network, it requires a middleware different from firebase
> (among others), but still restricted to smartphone's platform.
>
>
> Yes, and we haven't tackled this piece of it yet. We want to get
> clusters of devices on the edge happily and reliably messaging first,
> then get back to that problem.

Ok, sounds cool - I'm very interested in the latter ;-).

[..]
>
>
> In addition to that, there's to be an active open source community
> adapting platform changes of the systems in use.
>
> In fieldtracks, implementing the end-user application consumes about 10
> times the effort the mesh network. I guess, that it'll be similar for
> serval. Due to this, we ruled out to create our own messaging app in the
> foreseeable future.
>
>
> We'd love to join forces to get rid of the duplication in the app dev part.

That's interesting.

Atm fieldtracks is written in Angular not doing messaging yet. It'll
evolve into a PWA in the near future - most probably, it won't have much
access to APIs such as wifi etc. We're going to have a first, but very
small field test by the end of this month (treatment units on a parking
lot).

I guess, we might play with chat-over-mqtt in the next months, but
that's not high priority.

Thus, I don't think that there'll be much duplication soon, but things
might be different in 2020 / 2021.

Greetz, yanosz
Reply all
Reply to author
Forward
0 new messages