English

Replicant support for FP2

Tags: #<Tag:0x00007f05daa1aea0> #<Tag:0x00007f05daa1ac48>

It’s a bit crazy, but I think that a similar proposal made to, for example, the openmoko community could have a interesting impact. But it may not be enough anyway…

It won’t work. If I recall correctly, too much of the FP1/U hardware was based on the choice of the “wrong” chipset. But maybe these guys could answer your question:

If the FP1/U hardware would work with a GTA04 board, there would be a chance. But I don’t think so. Else somebody would have already wrote that here. But hey, why not ask?

Is openmoko still existent?
Their official website was last edited in August 2013…

But to quote more from the same source as the original post:

Which platform (instead of Qualcom) they should have been choose for authorise the freedom solutions?

The could have chosen a modern platform that doesn’t check the bootloader’s signature, such as the i.MX6, OMAP4/5 or Allwinner.

1 Like

I agree although I don’t know if these chips will be any better. And this is why I won’t buy the FP2, alternatives were never really discussed. They settled with a new partner pretty quickly or at least this is how I feel about it.

The FP1 was a prototype to check if it would work at all. But the FP2 feels rushed. But maybe this is the only way to get money into the project? I don’t really know, I never saw discussion about this and I have to admit that I stopped following the whole project for a while, so I’m not sure.

But I would buy a voucher for a FP3 if they would tell me on what hardware they will focus the next time, just to support the project.

1 Like

I don’t think that Fairphone made a rushed decision and surely looked into the subject deeply. On the other hand they need to have in mind their most important driving thought: To prove to the big players that a fair smartphone is possible. I have never heard of these open chip sets before, but I am sure that they would not be a consideration for e.g. Samsung. But a Qualcomm chip in a fair phone, hey that’s something they might consider and imitate Fairphone.

2 Likes

[quote=“fp1_wo_sw_updates, post:15, topic:9574, full:true”]It won’t work. If I recall correctly, too much of the FP1/U hardware was based on the choice of the “wrong” chipset.

If the FP1/U hardware would work with a GTA04 board, there would be a chance. But I don’t think so. Else somebody would have already wrote that here. But hey, why not ask?
[/quote]

I refer to the FP2, and only in the case that Fairphone decide to open up the hardware specification, so third party could develop alternative modules and/or mainboard. I think that the FP1/U is too close for that, sadly.

I don’t know if Qualcomm was such a bad choice, either. There seems to be quite a lot of Qualcomm-related open source development recently.

I couldn’t find anything about free Snapdragon modem drivers, but it’s listed on the mainlining wiki page, so maybe we’ll see work on this at a later time.
There is source code available for Qualcomm Gobi 3G modems, however and the Snapdragon 801 product brief mentions Qualcomm Gobi 4G LTE Advanced for connectivity. Maybe it’s possible to reuse some of this code then.

2 Likes

I didn’t know about this stuff, seems very interesting… I hope someone will use it to develop something interesting regarding Fairphone 2! Sadly I am not an informatic

EDIT:

I have found this: http://redmine.replicant.us/boards/27/topics/10365 it seems that Replicant (and pruvacy) be and remain a dream for us fairphone users

I think it would be a shame for a smartphone project talking about open source stuff to not get Replicant running.

Well to be fair about 99% of open-source phone projects can’t run Replicant.

3 Likes

It’s a difficult topic, especially if discussed in the “app store” dominated world of phones (see: @Jerry). It’s either Android or iOS for the most. Other phones are not for the vast majority. But the FP project was mostly about changing the way of the manufacturers who create these kind of phones … and this will always crash with the “free as in freedom” concept that drives software like Replicant.

For me it would be enough to see what chipsets and OEMs were taken into consideration and how they there rated. For me it feels like they just took the next bigger player and hoped for the best (A’Hong ->i Hi-P, Mediatek -> Qualcomm SD 801 MSM8974AB-AB) but they don’t have any “hard” tech know-how at fairphone.

It’s just a brand (FP) buying standard phones from OEMs asking them to care better for their workers and use better solder if they want the job of producing a batch of 15k phones … with an added nice blue interface on top. Still android, still standard, still the same that you can get from all the others.

But will Hi-P changes their ways long term, after the 15k phones? Who knows. But this was the main goal for the FP project. Currently the are failing both goals (more freedom more fairness) from my perspective, but I think it’s better to try. I just don’t like how they “try” the FP2 – no published paper on how the decided to go with HiP or what other OEMs/chipsets were tested. That’s all. Right now I feel cheated.

It’s not their goal, actually.
FairPhone themselves are very aware that they will not be able to change the world by themselves. And with a projected 140,000 phones per year, that’s an unfortunate truth. With that amount of phones, their direct impact is very low.

The most important thing for them is to educate people and show the world that it is possible to create a phone without conflict minerals that was produced in a safe environment by people that are paid well. Hopefully (and perhaps naively so) bigger players in the market will pick up on the fact that there is a demand for phones that were built under fairer circumstances.

It’s also one of the reasons why I think it’s very important for FairPhone to stick to finding ways to make the production of the phone fairer and put openness of software and hardware in second place. I think that for the average consumer, the tale of a more fairly produced phone, without child labor or extorted workers is a much more powerful (and much more important) message than whether or not the software they run on their phones is open source. Everyone here has to admit: the average consumer couldn’t give a dime about open software (see Apple’s success) but everyone is distressed when hearing the stories of what happens in Congo or in factories in China.

4 Likes

I completely agree with you, @Jerry, however I would like to go back to the initial goal of this topic:

I’d like to start a constructive discussion about how to make Replicant on the FP2 possible. If there are people, who badly want Replicant support, I think they should do everything to make it work (by contacting either parties about it, or by talking to independent developers and asking them to look into the matter).

2 Likes

Make the process how the chipset and OEM are picked transparent and rate “replicant” support very high on the list of important things. Done. Right now, I can’t even find this information for the last current chip sets/OEMs decision on the webpage. I don’t care so much about “replicant”. I care more about the whole process. How is it done. What counts?

It doesn’t have to be “replicant”. “Replicant” is just an idea that is not intended neither by Google nor Apple. They want a passive consumer buying apps and paying twice; with their data and money. I’m assume @Jerry might disagree here. But I want a good consumer phone for consumers that like to think twice and that don’t need the latest tech. For latest Tech one needs real R&D something FP really does not have.

1 Like

I deleted three posts because two were flagged and hidden for being inappropriate and not contributing to the discussion, but it doesn’t make a lot of sense if everybody can still view the hidden content and reply to it, carrying on the off-topic discussion.

1 Like

Well then you should start a topic about the decision for the Qualcomm chip set, or - even better - request the information from the Fairphone Team through support.

Thank you for your suggestion, but if you don’t have anything else to add to the discussion I fear talk like this will bring us off topic soon. I requested the information (via zenddesk). I did not receive it (yet?). But let us stop here. If you will open up such a topic I will post my questions there again.

The topic here is replicant support. And replicant will not really work on the Qualcomm chip set due to its non-free nature. But if you read up a bit on the coreboot project you will understand that sometimes even those non-free chip sets can be supported, but often … not. It depends what kind of people you are hiring and if the chipset is worth all the extra work. All that is something the FP project is currently not looking into a lot … it’s more a brand and designer project right now.

So the on topic discussion for me is still about choosing the right chipset and if the FP project will hire the right people to make it more than a brand. If not, the FP project will always depend on the OEM sources for the software of “their” phone.

Don’t get too focused on words like “Android”, “Qualcomm” and “Replicant” they are all just concepts. This will show you how strict the “Replicant” coders are/have to be:
http://redmine.replicant.us/projects/replicant/wiki/TargetsEvaluation

Blobs (stuff without source code) “normally”:
https://android.googlesource.com/device/lge/mako/+/77f8e6b51a8f26ea9288a9186179511a92c07df0/proprietary-blobs.txt

Smallest Qualcomm Blob (Android One)/Qualcomm Snapdragon 800:
http://redmine.replicant.us/boards/27/topics/5175 but add these as well:
https://github.com/CyanogenMod/android_device_oppo_msm8974-common/blob/cm-11.0/proprietary-files-qc.txt

It would be great if someone could put the full real blob list for chip used in the FP2 here. Remember the chip will still not fulfill the “replicant” “whishlist”.

1 Like

@Reent is also on the forum and maybe he can give us an answer publicly. I’d also be interested in hearing more about this topic.

Although the below comment is more of a continuation of my concerns posted in this topic, the latest relevant thoughts and discussion has been here, so therefore I decided to place my question here as well. Please @paulakreuzer, move my post when you feel it necessary.

Ok, here it goes:

Continuing the discussion from Poll: If you could install any mobile operating system:

Because of the size of the FP company and the complexity of the coding of operating systems I am not expecting FP to sell me a phone with Sailfish, CyanogenMod, Replicant or something else. It would suffice for me to know that hacker communities could thoeretically develop a succesful port of an alternative ROM to the FP2.

Although I don’t understand, I still acknowledge the fact that open and free OS-es are not of great interest to most people. For me on the other hand it is a condition. Spending over 500 euros and then being dissapointed in the possibilities of porting alternative OS-es - as would have been the case if I would have bought FP1 - is just too big a risk for me. I would own with a fairly produced product, have ‘joined a movement’ and supported a good cause. Great things in their own rights. But I would also be owning a product that I don’t feel comfortable using.

@fp1_wo_sw_updates & @anon40444829 and others that might have more technical knowledge than me; if FP does publish the source code this month as was promised, would it finally be possible to assess the probability that hacker communities will be able to port other OS-es in the future? Or on a more practical note: would this be the missing information I can base my decision on to back the crowdfunding initiative?

2 Likes

I guess a mod needs to clean up this thread … it is getting hijacked. I will try to be brief: FP cannot publish the “whole” source code because they do not own the whole source code. Please read carefully what I posted above and try to understand what you want if you talk about “porting” “alternative OS-es” to the fairphone.

Currently FP is a brand like many others that buys a SoC from a company called “Qualcomm”. Qualcomm again also buys software and hardware “patterns” (cores) from different companies also (ARM, GPU) and etches them onto their chip or loads them into their chips by software.

An another company again is soldering this Qualcomm chip to a board for the fairphone project. The fairphone project pays another company to change the standard Android interface a so that it looks a bit different, but it is still Android. This code could be published by the fairphone project.

For most of the software on this SoC there will be no source code, see the so called “blobs” above. But often that is not important if you just want to port another OS. Some parts are others are not. Android can be ported with blobs, just as the available CM images or the FP software show.

So it is important to find out if the 801 is a good chip set to make a port to an “alternative OS-es” or not. Most people think it is not a good chip, but mostly for a lot of very complicated reasons (the modem for example would be a no-go for the replicant guys, a show stopper.). For normal users this could still be okay, but what is missing and needs to be re-developed again for porting an other os to the FP is the homework that the fairphone project so far has not published.

I really want to know why the Qualcomm chipset was chosen, there are lots of others out there.

I assume they just don’t have the right people/knowledge, but I could be wrong. Maybe there was a good reason, I think often in this business the OEM is more important than the chipset and that’s why …

1 Like