Where I work we use NETMF for prototyping AND in production. Essentially we put as much as we can in managed code, if we need better performance, etc. we write native code and invoke it from managedland. Prior to 2012 everything was in native code. Why is NETMF only a prototyping technology?
It think the Gadgeteer module concept is the rapid prototyping stuff.
As soon as you go to higer numbers, you make a specialized PCB, then you can reuse the modules for next prototype.
The software might stay the same or just Needs to be slightly modified for the RTM Hardware.
I for example use a G120HDR for prototyping, but I only use plain NETMF Software (no gadgeteer).
But I have several Gadgeteer HW modules which I connect to my HDR (directly or via Breaout boards) to reflect the upcomming specialized PCB as good as possible.
NETMF is a âgiftâ to the community, the remnants of the abandoned SPOT initiative from years ago. Support from Microsoft is âlightâ, new versions are released occasionally. Support from GHI and other vendors is usually better, releasing new versions more often, but still based on old (or very old) Microsoft releases.
Recently there was a small indication that there might be more resources dedicated to NETMF on Microsoftâs end, but weâve seen no movement yet.
Commercial products based on NETMF might move GHI to improve/continue support, but I see no reason why they would move Microsoft to improve/continue support, as itâs open source anyway, and Microsoft has no âskinâ in the game, receiving no compensation for any resources dedicated to it. It shows up only as a âlossâ in the RoI calculations.
In summary, if GHIâs current products meet your needs (and likely they doâŚ), then go for it, you wonât be disappointed. If, on the other hand, youâre waiting for some big advancement of NETMF, then donât bank on it, thereâs no guarantees that Microsoft will ever take it anywhere.
If Microsoft drops support for NETMF anyone could take over the project.
From my point of view GHI would be someone who could do this.
It also could be maintained by the community.
I think if NETMF would be enhanced by the community and the results are monitored by someone (like GHI), then NETMF could be much more already.
But since itâs good as it is, I can live with the current situation.
I agree. Someone with a more vested interest in the technology should be in charge of the releases. GHI is all-in but Iâd rather it be a board comprised of companies & contributing community members. Gadgeteer should be much further along by now than it is.
It could be solidly argued that GHI is already âenhancingâ NETMF. Certainly the base PK distribution is of little (or at least, significantly less) value compared to GHIâs distribution(s). The problem here is that GHI has little to no motivation to enhance the PK distribution, or even any OSS distribution, because they make money off their âPremiumâ enhancements.
Thatâs a proposed enhancement (or, application, really), not a motivation?
Unless youâre proposing that NETMF be altered to require access to the (for-profit) Azure cloud service?
Or, it just occurred to me, that youâre most likely saying that if NETMF and IoT were more widely deployed, it would result in additional Azure utilization. That makes sense, but I doubt itâs a big enough market at this point to drive the needed resources.
GHI has a financial interest in keeping enhancements out of the PK that would compete with their Premium products. Forming a coalition that would result in the improvement of the PK would go against their best interests.
@ godefroi - I just wanted to say that GHI could take over if MS drops it.
This does not mean GHI has to integrate the Premium lib into the framework.
But they could integrate the OSH lib.
Also anyone else with a deeper insight could monitor a community driven development.
A fully open development without supervision might also work, but bears the risk of unstable enhancements and a inconsistence framework.
@ andre.m - yep, I could too, if my Family gives me a couple of month to dig into the framework and porting kit 8)
If we (at the company I work for)integrate NETMF devices more into our system, then we âwouldâ have a good reason to take it over too
A world where GHI âtook overâ the PK would look, essentially, exactly like the world we live in.
The PK gets little to no development love now, and wouldnât if GHI âtook it overâ. Thereâd be no reason to. Theyâd put their energy into the premium and (to a lesser extent) OSH libraries.
@ godefroi - My thinking was that GHI has an interest that NETMF does not die, because a good part of their products depend on it.
May be thatâs not the best for new features inside the framework, but still better than seeing it die.
Another approach would be to have an open branch where the community can add new features, bug fixes, âŚ
Then use some kind of voting which should make it into the next release.
Then someone would have to take the duty of creating such a release.
Well, but as long as MS officially supports NETMF this is just day dreaming (not sure if nightmares or not :think: ).