Open Source Firmware

I just found a bug in the answer script of this forum.
I'm talking about that grey box in which you type your message. It turns completely grey when I shift - leftclick the border of the textarea ..

just wanted to mention that

Maybe they'll reconsider their answer if we all send a different mail asking them to open the source or release a SDK.
I mean if we're a lot to write and we show ourselves to be very supportive, they'll understand that the community is quite big and that we all believe in their player's huge potential.

("I have a dream..." )

It's a possibility... but might not right now, or we may make them upset instead. But somehow, we should show that we are a respectable community that isn't going to use it for bad reasons, and show that releasing it would help them and their sales instead of hurt it... Should somehow show them examples of the many awesome things out there that only are because they were open source.

I mean I'm quite sure that a firmware will be recoded (e.g. rockbox), if the player gets popular. In my opinion the best thing in their position is to make their firmware open-source.

puffel i dont think they should really make it open firmware. people could steal it and use it for there knock off players... and then people may but it because it looks like the meizu but If its for rockbox maybe. but it depends on how you see it

Exact specifications should do the trick, if we knew every little bit about the hardware, there should be no problems making our own firmware.
But I am no programmer, so there is a lot of people who knows more about this than I do.
But there must be some weird conversion of the Linux kernel somewhere that can run on this thing.
But don't know how you could install it. DFU tools might be able to, but not sure exact how it works.

I guess the reason why we won't see the firmware opened is all the closed things in it, like MS drm (there from next version, not that I want it) Mp3 and maybe parts of the firmware it self.

I have read the previous posts of this thread, as i will be very interested and ready to code if Meizu will accept to deliver a SDK and a source code.

But i have some doubts that it will be the case, because Meizu is certainly not the only owner of the code.
Let me explain:
- Meizu is a product developper: based on marketing request, they develop products. For each product, they buy ICs from other companies.
- In the case of the M6, the main IC is a Rockhips. Rockhips is the IC vendor.
- IC vendors sell ICs, of couse, but they provide also to their customers reference designs (ie an example board, showing how to use the IC) and the associated SW and FW, including the SDK. The FW consists usually in an operating system with basic drivers, decoder libraries and one example application, used mainly to make demonstrations.
- So, basically, the product developper takes the FW from the IC vendor and customize it, mainly by adding :
- new drivers to deal with the HW of the product (LCD, interface, tuners, ..., ...)
- top level application and graphics

So, as you see, to enter in an open-source model, you need:
- to have the agreement of Rockchips to give their SDK, the detailed specification of the IC and the example source code.
- to have the agreement of Meizu to have the detailed specification of the HW used and the application.

IC vendors are usually very reluctant to give their source code, because competitors can detect eventual patent violations and may re-use some ideas by studying the code.

As a conclusion, even if i agree that it will be marvellous to have open-source FW, i do not think that it will become a reality.

Are you sure the Meizu M6 uses Rockchips hardware? I always thought it was from Phillips.

According to my understanding of that:
http://en.meizu.com/userforum/forum_...p?TID=268&PN=1
Philips provide the DAC, but the main core is obvioulsy from Samsung. I have still some doubts: is samsung the foundry factory (ie subcontracted by Rockchips) or is it a full design done by Samsung itself?

I'm glad they're not going to release the source. For all the great things that could be done, there are plenty of retards out there, like the people who get off making viruses. It would be too easy for someone to write some very destructive firmware and try and pass it off as legite. That would hurt a lot of players. Meizu would rather be safe than sorry.

Originally Posted by spdsktr

Meizu would rather be safe than sorry.

Im affraid I would disaggree with your reasoning.

[joking] In my house we do not have a kitchen because kitchens contain too many knives & hot surfaces, people could get hurt. [/joking]

Seriously though, I have little sympathy for people who actively install publicly recognised malware. In the case of a malicious meizu firmware I am positive we would have a lot of negative feedback for these versions on the forum. If theres a lot of negative feedback nobody with any common sense would use it. If a user actively installs bad software on a piece of their own hardware, it is entirely the fault of that user.

It is after all marginly possible that the more recent 2.003.2 modifications (with no difference in firmware number) could be from an illigitimate source with malicious code within them. Has everyone bothered to thoroughly check this before installing the firmware on their player? I doubt it. But there seems to be a lot of positive feedback that it does what it is supposed to - so we trust it. This "trust" we feel is similar to eigentrust (see http://en.wikipedia.org/wiki/Eigentrust ).

Essentially, a user trusts an item of content based on the sum of all the product review ratings individually multiplied by a rating of our faith in the reviewer. Or:

T(x) = Σ(i = 1...n) R(i,x)*F(i)

where:

• T(x) is our Trust in product x
• n is the number of people we know who have reviewed product x
• i identifies each person
• R(i,x) is person "i"s rating of the quality of x
• F(i) is our rating of how much we trust person i

Since we trust community members at meizume will help other members, we trust the positive reviews of the firmware and thereby the firmware itself. If the firmware is bad - people will report this, and trust will decrease accordingly.

Some casualties may potentially occur, this can not be helped and is in some senses "necessary". Similarly, we do actually have a kitchen in our house. Sometimes people get hurt by the sharp objects and the firey oven within this kitchen, but we keep the kitchen all the same since we like to cook our food.

Note that there is NO EVIDENCE for believing there are ANY malicious meizu firmwares at present, this was suggested purely to demonstrate the idea of trust in collaborative communities.

p.s. its past 6am & I havent slept, so I hope the above explanation / analogy make sense.

Imagine the menu beeing some sort of cirle that you spin around when scrolling to chose Music/settings or whatever.. It would be nice!

Ive found an interesting source of some uCOS-II information, http://www.ovro.caltech.edu/~dwh/ucos/

Theres a full project report, using assembly & C. But even with this we would need a lot more data, perticularly drivers for the screen, usb, file store, decompression etc.

The firmware, if released as is, would only run on the specific hardware. But you fail to realize since one would have the firmware, it would be easy to change hardware components and build support into that firmware. For instance, one might add a mini-SD slot, and modify the firmware source with drivers to accept the addition.

I too am for open sourcing of all firmware, but it IS easy to steal, but there are also laws to protect them against it. The hardware would still be patented, and hence be illegal to copy for 20 years (possibly).

Most all chip makers supply source code examples for their processor chips simply by asking for it. You may have to sign a NDA however.

Once hackers figure out the bootloader via some JTAG connection, the M6 will be hackable most likely. It just needs to become more popular. Everything is hackable!
It will take some brave soul to rip one apart and start exploring.