Hello,
After reading some news on the internet I found a company that apparently advertises open 3G modules. But I don’t have enough expertise on my own to be sure if this information is true and trustworthy. Have anyone heard about a 3G module called “adafruit fona 3g”? Is it better in terms of opennes than the 3G modules currently included with Librem 5?
It won’t fit as it’s not in an M.2 formfactor. Even if it was I see little reason to use it over the ones the Librem 5 comes with (fewer bands, only 3G etc.)
Not sure from where you got that it’s “open”, seems like as open as any other cellular modem?
Well, their website says:
If any of our products include software or firmware to operate, we also provide all of the source code for your hacking pleasure.
(here: https://www.adafruit.com/faq), but I’m not sure how true this is, because I could not find the actual firmware anywhere. But maybe I didn’t search carefully enough.
If it is already flashed on the modem there’s no need for it to work as it already works out of the factory, so they are not required to share the firmware.
It’s also illegal to share firmware for radio modems as of a few years back in both the US and Europe. This law was passed to lock down modems to their approved power and frequencies.
Wait… Do you have any more details about that?
The point is, regardless of this company and its honesty, I’ve seen several projects “in active development” trying to write an open implementation of the 3G protocols. Do you say that all of them are illegal?
The section which is interesting is this part
The TP-Link settlement was announced in the midst of a controversy spurred by those new FCC rules. The new rules for the 5GHz band require router makers to prevent third-party firmware from changing radio frequency parameters in ways that could cause interference with other devices, such as FAA Doppler weather radar systems.
Router makers can comply with the new FCC rules by placing limits on what third-party firmware is allowed to do. Alternatively, hardware makers can comply by preventing the loading of open source firmware entirely—which is what TP-Link chose to do.
Well, the article is not very clear (starting from the title that says “forces … to support”, while the article itself rather says “to ban”), and I didn’t understand one thing: was the issue with transmission power only or also with frequency? In the former case, I think it is technically possible to limit the power without limiting the firmware, by the means of electric circuits.
See also the topic in this forum with title “EU Commission delegated regulation on Reconfigurable Radio Systems”. (not linking to it as it is registered-users only)
While possible, this is never done in practice because different countries have different regulations on frequencies and power limits. Making different chips for each regulatory region is much more expensive. This is why most (all?) companies have opted for making the radio firmware closed source.
Fair point but … legally-mandated (or legally-encouraged) closed source is a terrible idea. It is everything that we are trying to get away from.
Are the limits in GSM/3G more strict or vary more from country to country than in wifi? In wifi, we have at least ath9k-based devices, and they are legally sold with actual open firmware available.
They are strict because you cannot simply replace something on Linux with open source of yours, meaning that there is no way to replace original firmware (regular updates inclusive) that was previously uploaded from Quectel, Telit or any other manufacturer (and FCC, RED, etc. Certified as such) without their strict approval. As I understand this, even if you posses proper kind of knowledge you are legally not allowed to make such modem work even if it is just for use on the Moon.
It is as simple as that the current Gemalto L30960-N3410-A400 have FW REVISION 04.004 A-REVISION 01.000.11 flashed onto PLS8 modules (Factory introduction: CW06/2019). IMO, no one can get such firmware to test it on an older version of PLS8 (either E or A) other than the B2B marketplace customers that previously bought at latest 400 pcs., and, of course, got those directly from the manufacturer.
Another way, sort of off-topic here, would be in case of/if (just example here) the NXP MIMX8MQ7CVAHZDB HDCP would have eventual capability (with particular chip that is built within L5) to be programmable from the customer side (for those interested to have this feature enabled), this refers to letter D, letter C means already HDCP NXP programmed (HDCP enable parts). Recently was promised that someone will “do better about keeping people informed” so let’s see, as for me it is still an open question, if I’ll be able to watch every YouTube video on Librem5 because nowadays Digital Content DIRECTIVE (EU) 2019/770 stays in between of any sender and every L5 phone recipient. I guess that @amosbatto might have appropriate thought on this subject (i.MX 8M HDCP) for myself and I appreciate almost every word he writes here, because they are important in content and in understanding of the broader picture, but they do not serve (always) as direct answers. This one I’m not accepting, as reality, either:
and still waiting if someone from Purism have anything to say about hosting of HDCP on L5.
I’m not “someone from Purism” but here are my thoughts.
My understanding is that the intention is not to do, or support, HDMI, so the question for HDMI doesn’t really apply. My understanding is that the underlying hardware can do HDMI, with a blob, but therefore it will not be used and there will be no connector for it anyway.
Video out will be via DisplayPort over USB-C (in one of USB-C’s alternative / alternate / alt modes). The question then arises as to whether that combination, including the monitor, will support HDCP (or DPCP) with DisplayPort.
On top of that … since many people have a monitor that does not support DisplayPort over USB-C anyway, they would be trying to use an adapter that converts DisplayPort over USB-C to … DisplayPort on a conventional DisplayPort connector / HDMI / DVI / VGA (I assume VGA requires an active converter). So the adapter becomes a part of the puzzle / point of failure / variable.
I’ve purchased (for another purpose) a display that does support DisplayPort over USB-C but when I get my L5 I will at least give it a try.
I won’t be testing HDCP though. It is of zero interest to me as long as it doesn’t get in the way. With the L5 I don’t think it is going to be possible to test HDCP.
If you want someone from Purism to respond, best to ask Purism directly via email - and to add detail to your question.
I live in Europe where Copyright Directive (EU) 2019/790 is in effect (348 votes in favor, 274 against), after a lot of unsuccessful protests in all big cities of Europe, and for some, to me “abstract”, reason (not being an expert) I bind HDCP with necessity that surrounds us in 2020 and beyond. Of course I can choose to live in darkness but I don’t think this is must for anyone. I don’t like mentioned Copyright Directive and I don’t need to buy L5 phone from Purism if they don’t want me walking around with it as my daily driver. Therefore as long as possibility to install and run HDCP doesn’t get out of the way for us from Europe, it is fine with me! Maybe this helps:
https://lists.freedesktop.org/archives/wayland-devel/2020-January/041147.html
As said: I don’t like, I didn’t say I’m lost or that I’m not in favor of your standing point against HDCP, but this doesn’t help me either or provide any ground to hate YouTube, just as example here … maybe someone will find solution. If I close my eyes, your answer directs toward MIMX8MQ5CVAHZAB (8M Quad Lite) and now we “know” few things more; so it might be there is no VPU either. Anyway, I don’t think that bringing MIMX8MQ5CVAHZAB to Europe will ensure CE smartphone-proof-certificate. We just discuss something that I’m unfamiliar with and selling nothing here. Is this an question, I don’t know either.
EDIT: Possible options, just in general:
• No HDCP
• HDCP 1.4 only for HDMI
• HDCP 1.3 only for DisplayPort
• HDCP 2.2 only
• HDCP 1.4 and HDCP 2.2 for HDMI
• HDCP 1.3 and HDCP 2.2 for DisplayPort
First off that article13 site is just a pure lobbying site. The citizens of Europe have nothing to gain from this European directive, only the producers which are the minority, so it makes sense for the majority to be against it. The site for example states the following
WHO COULD IMAGINE THAT THE OUTCOME OF NEGOTIATION WOULD BE, FOR INSTANCE, YOUTUBE BLOCKING 35 MILLION CHANNELS IN EUROPE as YouTube had claimed? It is nonsense. Why would YouTube inflict such harm on its own platform?
While true, article 13 doesn’t solve any of those issues. The only thing article 13 does is adds even more restrictions on what can be claimed as usage of copyrighted material, which means either taking all of the ad revenue from uploaders or blocking even more videos. And as is widely known, the YouTube claims system is completely broken and you often need a lawyer to fight false copyright claims. So the copyright holders today on YouTube simply claim as much as they can in the hope that no one fights the copyright claim (as it will be expensive and risks you possibly getting banned from YouTube). So what article 13 does is that it allows people to make even more false copyright claims and take money from YouTube creators. It would be good if article 13 gave any rights to equally fight back these false copyright claims, but it does no such thing. I don’t want the EU to have as shitty copyright laws as the US and this takes us another step in that direction. Even people who work with international copyright law have a hard time fighting false copyright claims.
https://www.law.nyu.edu/centers/engelberg/news/2020-03-04-youtube-takedown
HDCP requires non-free software so it would make sense for the Librem 5 to not support it.
Thank you for your comments! Yes, my link was intentionally pointing at this, nothing else than just a pure lobbying site. Involved, together with particular EU commission, actually didn’t ask masses about their opinion. And in the meantime this particular Article is carrying number 17.
I wrote few of my thoughts as I just want to stream-down something that authors self provide to the public over Internet and at the same time don’t want to use Firefox to watch YouTube videos because of EME specification (as uploads are not of my concern anyway). Instead, I’d like to have some open source standalone .deb app, like devices that are using OpenEmbedded-based Linux. Also, I feel like the first rule of somethings is not to talk about those, and therefore I’ll leave someone, that insist on an easy way, to watch (and listen) any “free” video[1] through its favorite browser. Or implement even something easier like YouTube.apk from GooglePlay (might be it doesn’t need HDCP at all).
[1] “Somewhere Over the Rainbow” - youtu.be/1Eyy9dhq2lw
at this point in time it’s rather U.S.E (united-states-of-europe). long live the welfare-super-police-state !
about snoople and it’s copyright claims … it seems they have had enough of our data that they can now afford to randomly start shitting on people. surprise-surprise !
On the original topic: It looks like the Adafruit FONA 3G module is just a breakout board for a SIMCom SIM5320 module.
I don’t think there’s anything special about this module. It would be special if it came with open firmware, but it surely does not. Without open firmware, it’s just a lower-spec module than what’s already going to be shipped with the phone, and it’s one that doesn’t come in the right form-factor. You could adapt it, but it’s not worth the effort.
I suspect that when Adafruit talk about providing firmware, they’re being a bit imprecise and they actually just mean firmware that does not come with any off-the-shelf parts: the firmware you would need if you were going to buy the component parts and build their module in the same way they built it. I think they’re happy to classify firmware as “hardware” if it comes in a component they’ve bought in.