Coreboot + Linux-Libre could bring TONS more users

[sebt3]

Spending time and money? It's probably just a matter of removing some kernel module. Something like

Then why nobody cared to do so for pandora ? I mean there's seems to have many request for it.if that's so easy I would have expected at least would have done a downloadable with an install guide ?

 

[comradekingu]

There are several on the wiki, doesnt change the hardware selection though, which is the issue.

Running with no 3D drivers doesnt really fix anything. An uninstall guide would not make much sense.

 

[notaz]

Spending time and money? It's probably just a matter of removing some kernel module. Something like

sudo rm /lib/modules/*/kernel/drivers/gpu/pvr/*
will probably do the trick if you want a "bugged down but fully 'free' OS". I'm not sure about the 3G/GPS/wifi drivers, perhaps those are binary blobs as well.

Doing that you remove open GPL parts, not the blobs..

 

[Grench]

Is it a possibility, given the current hardware purchased?

32 pages summarized: it is within the realm of possibility, the only thing that isn't open source is the GPU driver. The reverse engineering effort has been ongoing for so long with arguably little success that it probably won't be available anytime soon. We would be stuck with 2D framebuffer access.There was a suggestion of offering two versions, one "free" and the other with the GPU binary blob, but RYF certification requires that the two versions be completely distinct with no chance of confusion, ED would literally need to have two different case moulds (at least) to make them unique enough for one to qualify.

Further discussion pointed out a lot of the restrictions on advertising that were a logistic nightmare for EvilDragon to try and follow. He could, but it would require more conscious thought than it's probably worth.

Finally there was a huge discussion about whether it would actually bring in any more devs: the Pandora (and by extension the Pyra) is a niche market, selling a few thousand but filling it's niche fairly perfectly. If "100% free" is required for someone to purchase it then the feature set is probably not what they were looking for in the first place and they probably would have been happier with another device.

That's how I remember the discussion anyway, does anyone remember things that I've forgotton?

I just had an epiphany. The Pandora/Pyra is a niche product and is probably the closes handheld anything to fully open source available: not just the operating system, but ED is even planning to release the hardware source files. Because of the niche and the willingness to pursue "freedom", it is in the FSFs interest to come here and try to get ED on board with the RYF, not the other way around. If they want a pocket computer/handheld game console that is 100% free out there, attempting to compete with all the Android devices out there, then they should step forward, pledge some resources to reverse engineering the powerVR, and make "us" (the collective us that is Pandora users) want to be part of the FSF and RYF.

You forgot the part about the FSF not being the freedom-loving-free-for-all-love-fest that many try to make it out to be. It is a licensing standards and sales organization. They sell/license their 'freedom' trademarks. They do not give them away.

So, is it worth it to -add cost- to the Pyra to get purchase rights to add a sticker with the licensed trademark from a self-proclaimed standards licensing organization who's clout -might- influence a hand full of Communist leaning individuals to -purchase- the hardware item from a -for profit store- using money they made -working-... See how this becomes a less and less likely positive outcome?

FSF/RYF certifications are a waste of both time and money. Move on. Produce the Pyra. Make millions. Roll in it if you want to. Make the successor in 3-4 years or so. Make more millions. THAT would be cool.

 

[_wb_]

Spending time and money? It's probably just a matter of removing some kernel module. Something like

sudo rm /lib/modules/*/kernel/drivers/gpu/pvr/*
will probably do the trick if you want a "bugged down but fully 'free' OS". I'm not sure about the 3G/GPS/wifi drivers, perhaps those are binary blobs as well.

Doing that you remove open GPL parts, not the blobs..

Oops! Where are the blobs? Anyway, you get my point...

 

[Binky]

Is there a significant number of people who refuse to by tech unless it is entirely free, open, FSF-certified and unencumbered by binary blobs?

My guess is not.

If I'm right, then the number of customers gained by making the necessary changed would be negligible.

The Pyra will be best in class anyway when it comes to freedom. Debian out of the box, Root access  Open schematics,  Developers with a real on-line presence, etc

If you care about software freedom, the Pyra is already lined up to be the obvious choice of pocket computer. Why should making it freer, at the expensive of a GPU, various other drivers and general user convenience, not to mention time and headaches meeting certification requirements, help?

There is a compromise to be made between usefulness and freedom. Most people here, including me (in general), err towards the freedom. I trust  the writers of proprietary drivers enough that I'm happy to risk security holes for the convenience of having whatever the driver drives. If there were a free driver, I'd use that. But in some cases, there isn't

There's also a distinction to be made between the commercial argument for freedom, and the ethical/security arguments for it.

 

[comradekingu]

What if you were wrong, if you havent looked at the evidence, then that could for the case of argument be the case. Is there something wrong with selling additional units, no matter the amount? To me it is a question of effort for reward. Because the reward is evident, and the effort may be small, but if there isnt any downside, then that is a free benefit.

The FSF has the most popular software license(s), and have recently gotten into looking at hardware. BSD can be looked at the same, and is even more starved for hardware support.

None of your points matter if it offers no real security. That is a logically valid stance to take. Sure you can sell units any other way too, but that is paramount, and its the same kind of concern that stems from being 'more better' in other areas. Your points are very fine.

There is no compromise to be made between usefulness and freedom. There is a short-term compromise between freedom and convenience.

I don't see the distinction, are you talking about freedom possibly only for the first person the software is conveyed to? Thats copyleft and non-copyleft.

As for saleability, I see it this way, trusting proprietary drivers that are in places that matter is just plain wrong. Since you care less, any is fine with you, but it becomes an exclusive argument as soon as you want to include also me.

And me buying the same thing as you, benefits you. You can take that to any extreme you want. As long as you don't have to forego anything, including adding effort, that can be as extreme as it wants to.

Some people _really_ get upset over systemd, gtk+, or even fonts and icons, design, colours etc etc.

Usually there is something to garnish from such sentiments. If they care about something what is only a problem to be solved, and not additional problems to be created, then welcome aboard. Then again you could end up in non-commercial plan9 / minix-land too.

If you look at the the solution towards the goal and who presents it, its a pragmatic approach. Of course you can let the matter be decided by giving it away to some loudmouth freedom-propeller dogmatic stereotype person. The FSF understandably cant be lenient, since that lessens their point, but they can differentiate without losing credibility in my view.

There is a great deal of difference between avoiding being evil, to being evil to the limits evil permits. And if the FSF isnt onboard with that, then so be it. We can be our own spokespeople. Its a very imortant point to communicate, and also it brings peoples shoulders down.

When discussing matters its easy to think everyone who has a goal of freedom is only concerned with all or nothing on the path there, which isn't the case. Curiously some people who aren't siding with freedom seem to want to paint that picture. I think discussing their agenda is more interesting than drawing it to issues that may not be feasible yet to overcome with respect to future freedom.

In all fairness the freedom we do have, we all take for granted. Which is an important thing to remember.

 

[ekianjo]

To me it is a question of effort for reward. Because the reward is evident, and the effort may be small, but if there isnt any downside,

Oh yes there are downsides. It means you need to have different product positionning and marketing depending on which version you advertise. This will make it a very messy business depending on whoever you are trying to communicate to. The most successful devices are not the ones which try to be sold to every consumer on Earth.  

 

[comradekingu]

Very valid point. But that is as far as RYF and the FSF is concerned. They arent the end-all of marketing good ideas, nor are they particularly good at it.

In my mind catering to more different types of people with different things that are important to them is a lot better than trying to make a generic device.

I imagine the points other than gaming is photography and audio-listening.

Championing the best solutions is the ones that bring the most freedom to people, this time around that meant powerVR drivers came to be part of the deal, but its modular, and possibly better drivers can be made.

Where the crux lies, is with things like the modem, a cellular network that controls the modem, which in turn shares memory with or controls the cpu is terribad. Today very few devices are good in this respect, since most of them are bad.

Its much more important to get things like that right than 3d drivers.

ED said jorjin  WiFi is currently considered, but that the WiFi wasnt final yet. The last time we tried looking into alternatives it was with the intent of having both eMMC and uSD internal simultaneously. Since then that has been solidified and fixed, in that its one or the other on the same line, not each with its own communication, which leaves more bus-lines available. So if we can make an effort there, it could really streamline the thought that no ugliness is in the places it can do security-damage in.

I don't know if its SDIO, or USB thats routed to where the chip is now, but a drop-in replacement could potentially be a big win.

(here is a post from last effort)

 

[ekianjo]

In my mind catering to more different types of people with different things that are important to them is a lot better than trying to make a generic device.   I imagine the points other than gaming is photography and audio-listening.

Yes, but The Pandora and the Pyra ARE Generic devices, no matter how you are trying to put it. They are not going to be the best at whatever they do - while gaming may be close to be really good at it (unfortunately we probably won't have high-value development titles on the Pyra, so while the hardware is going to be very capable, contents may be an issue unless people are fine to limit themselves with emulators).

For a photographer, there are tons of specialized devices out there that are friendly and easy to use - The Pyra could probably do the same thing but with some limitations and bad/clunky UI unless someones puts a lot of effort in making specialized software. Same for audio listening, the competition is very active already and while I think the Pandora/Pyra have very good audio capabilities, it's hard to sell them as "good audio players" when you see the UIs you have out there - plus the lack of external control when the Pandora/Pyra is in your bag.

And as I said before, you can't talk the same way to the photographer and to the audiophile - you'll need to have different arguments, different product positionings, different strategies to find them and reach them, and the price point may or may not make sense for them depending on the product positioning and the competition on the market. 

The same goes for putting "Freedom" forward to please FSF people, they may be out there but whatever argument you use to convince them won't be effective to talk to gamers, for example. 

The real question is:

- before trying to sell it to tons of categories of people. what is going to be the most "obvious" use for the Pyra that will gather a great amount of people at launch ? 

That's the question I would ask whoever is in charge of the product sale strategy, because obviously there's no time to prepare 20 different product propositions for the launch timing. And that kind of question is way more important than thinking about which color and layout to use for the Pyra website - in my sense people are working completely backwards. We should define the target first (or the first few targets first), then develop the appropriate communication tools following that. 

That's marketing 101, really. 

 

[comradekingu]

Sent pm about the bulk of the OT stuff. I think relating to freedom in the FSF way is why freedom seems to be a FSF-centric prospect. It isnt.  Which is why its a lot better to roll that marketing inhouse.

Getting gamers is one thing, which is obviously best garnered through focusing on games, but having freedom is an additional benefit.  The benefits of free software and copyleft is the type of gaming the pandora is superb at for example.

To understand that is nice, gamers dont just consume passively. Some care a lot about games, and the big gaming companies are doing their share to prove our point. Yes they have the most flashy looking games, but that is also the end of the good things.  So the pyra will be not good at directx heavy graphics games, thats a given. Lets forget about that area as a market.

But portable gaming isnt that. Its very casual since phones have no controls. Heartstone and minecraft tops.

I am an avid quality audio listener and i already have great use for the pandora, it falls short on editing for photography, but i see the potential with the extra and better bits on the pyra. The software for both these is already there.

Personally i couldn't sell video, or phone-capability, because im not that good/into those things. But that's a possibility.

 

[ekianjo]

To understand that is nice, gamers dont just consume passively.

I'd argue there are many types of gamers and I wouldn't call them out as a single "entity" out there.

But portable gaming isnt that. Its very casual since phones have no controls. Heartstone and minecraft tops.

Yeah, but mostly this kind of software is not made for Linux, it's made for Android or iOS. There are good games for Linux but they are for x86 currently when closed source, and the good games which are open-sourced are relatively few - you don't have hundreds of "good" free software games out there. 

Addressing gaming is complex, because there are so many things you can talk about. Getting developers on board (the ones who are already convinced that supporting Linux is the right thing to do) would be a great first step to fill the void in terms of new, original games. 

 

[_wb_]

The real question is:

- before trying to sell it to tons of categories of people. what is going to be the most "obvious" use for the Pyra that will gather a great amount of people at launch ? 

That's the question I would ask whoever is in charge of the product sale strategy, because obviously there's no time to prepare 20 different product propositions for the launch timing. And that kind of question is way more important than thinking about which color and layout to use for the Pyra website - in my sense people are working completely backwards. We should define the target first (or the first few targets first), then develop the appropriate communication tools following that. 

That's marketing 101, really. 

... if our marketing would be based on paper ads and TV ads, sure, then we would have to pick one demographic and focus on it. But that's not how the marketing will happen. We can have multiple target audiences, because our marketing mechanism is mostly anarchistic electronic mouth-to-mouth advertisement, not coordinated and focused ad campaigns.

Anyway, if I would have to pick one unique selling point of the Pyra, it would not be gaming, audio, photography, or any other specific use case (for which dedicated devices exist that are probably better at it, at least in a bang-for-buck sense), but versatile hackable portable general-purpose computing.

If you're selling a Swiss army knife, you don't sell it as "the ultimate screwdriver set" or as "the pocketable can opener, bottle opener and corkscrew combo" or even as "three knives in one". You also don't sell it specifically to soldiers, boy scouts, hikers, electronics hobbiests or people with tiny kitchens who don't have room for dedicated kitchen tools.

There is no "main" target audience or "main" functionality -- the beauty of the thing is that people can use it for lots of things and you can't even predict what they will use it for. Sure, perhaps retro gamers is the expected biggest subgroup (I'm not even sure about that, since the Pyra will not give them that much more than what they already have on the Pandora or even GCW0), but it would be wrong to focus exclusively on them.

 

[ekianjo]

There is no "main" target audience or "main" functionality -- the beauty of the thing is that people can use it for lots of things and you can't even predict what they will use it for. Sure, perhaps retro gamers is the expected biggest subgroup (I'm not even sure about that, since the Pyra will not give them that much more than what they already have on the Pandora or even GCW0), but it would be wrong to focus exclusively on them.

I'm not saying anything exclusive. I'm saying you have to start somewhere and you cannot start communicating in all directions from the start.  You need to have a message, and "it can do everything" is not really a message. Mobile phones can do everything too, but they are good at certain things and bad at others. You need to focus on your strengths. 

 

[_wb_]

There is no "main" target audience or "main" functionality -- the beauty of the thing is that people can use it for lots of things and you can't even predict what they will use it for. Sure, perhaps retro gamers is the expected biggest subgroup (I'm not even sure about that, since the Pyra will not give them that much more than what they already have on the Pandora or even GCW0), but it would be wrong to focus exclusively on them.

I'm not saying anything exclusive. I'm saying you have to start somewhere and you cannot start communicating in all directions from the start.  You need to have a message, and "it can do everything" is not really a message. Mobile phones can do everything too, but they are good at certain things and bad at others. You need to focus on your strengths. 

I disagree, I think "it can do everything" is the main message we should bring (together with its portability). Sure, any desktop or laptop computer can also do everything, but they are not nearly as portable (a.o. because you need an external game controller and probably extra USB dongles for 3G/GPS etc, not to mention a spare battery to match the battery life). And phones are portable but they cannot do everything: they typically don't have any real input methods, connectors, extendable storage, and their software typically attempts to cripple any general-purpose use. If anyone says "phones can do everything", I say: try coding on it!

 

[comradekingu]

it cant do everything, new directx graphic intensive games it will be really bad at.  But if the pandora worked, the pyra will have an easier time. Better supported OS, more mature linux, faster, wine ARM, GLshim, and so on werent there when the pandora launched.

Yes it can be a laptop, and thats probably the biggest market, since many people need small laptops, and will adapt and fit it to their needs.

 

[Binky]

If you're selling a Swiss army knife, you don't sell it as "the ultimate screwdriver set" or as "the pocketable can opener, bottle opener and corkscrew combo" or even as "three knives in one". You also don't sell it specifically to soldiers, boy scouts, hikers, electronics hobbiests or people with tiny kitchens who don't have room for dedicated kitchen tools.

This.

There are many things the Pyra will be good at (multimedia, communication, gaming, office-work, being-a-development-platform, not-spying-on-you, etc) , and focussing on any one of those, either in design or marketing, is a bad idea.

 

[ekianjo]

I disagree, I think "it can do everything" is the main message we should bring (together with its portability). Sure, any desktop or laptop computer can also do everything, but they are not nearly as portable (a.o. because you need an external game controller and probably extra USB dongles for 3G/GPS etc, not to mention a spare battery to match the battery life). And phones are portable but they cannot do everything: they typically don't have any real input methods, connectors, extendable storage, and their software typically attempts to cripple any general-purpose use. If anyone says "phones can do everything", I say: try coding on it!

Guys, Guys, Guys.

I was waiting for the "Swiss army knife" argument. If you used a Swiss Army Knife before, you should know that it's OK as a knife, decent as a can opener, and it sucks bad for everything else - try cutting a sheet of paper with swiss army knife scissors and you'd wish you'd be already dead. 

"It does everything" is not a message. It's just over promising and you'll get very disappointed customers when they realize your "Pyra audio player" piece does not have external controls to switch tracks when in your bag (or if you are talking about the shoulder buttons, good luck preventing them from touching anything else). If you try to tell them it's ok to code with it, yeah well sure it's possible but it certainly won't replace a PC with a large screen space to work with tons of different windows at the same time. 

My point is, yeah of course it can do everything, and you are preaching to a convert (hey, I started the "My day with the Pandora" series of articles on Pandoralive to showcase the "exotic" uses of the Pandora) - but it does not mean it's a good idea to advertise "everything". 

You can definitely say that the Pyra is going to be a very solid device for the following:

- gaming in general - because CPU/GPU + RAM + hardware controls, and improved hardware controls vs the Pandora.

- typing and everything that goes with it - because of the keyboard

- connecting itself to other peripherals, because of the USB ports

- always on device because of the long lasting battery

You could talk about other things too, but there are caveats everywhere: most of the software you want to use is not adapted to the screen size, many open source programs have shitty UI (no matter how good they are, capability wise - I use LibreOffice but i really dont like how things are organized in it, and that's just one of the examples), there are stuff that don't work out of the box... 

i.e. over-promising is a very dangerous marketing strategy. Because what people say after they get their device is going to impact the reputation and the subsequent sales. At the beginning the focus should be to ensure the first customers are VERY happy with what they have, before marketing it to the whole world. 

 

[comradekingu]

How did victorinox market the swiss army knife before people knew what it was, thats the question imo.

Tell in a honest way what it can do, and then people can decide if thats for them.

Laptop describes many-use cases, but quality audio and photography (other than gaming of course) are the focal points that i see where it shines. And those are also distinctive usergroups.  All within the serious segment, which is home ground. Then it can ascend into more and more usability for everyone from there. As seen with pandora.

 

[_wb_]

Of course it cannot literally do everything. It does not toast bread, for example. But it can essentially do anything a normal laptop or a normal gaming handheld can do, reasonably well. You will be some years behind in terms of raw power compared to big laptops, x86-only closed source stuff will not work, and obviously the screen and keyboard are small, but other than those obvious limitations, it's not really over-promising to say that it can do everything.

I agree that it's not a good idea to create false expectations. So we should make it clear that it will not run Microsoft Windows, it does not replace a gaming PC, small screens are small, and so on.

But we shouldn't undersell it either. If you sell it as if it were just a small tablet with built-in keyboard, people will ask why they should spend 500 EUR for that. If you sell it as just a fake Nintendo, why would anyone spend more on it than the original ever costed? Or as an expensive underpowered laptop with a crappy keyboard and almost no internal storage, but with a great battery life. Or as an outdated smartphone that can't really make phone calls and doesn't even have a camera.

The point is: it's a unique device with lots of strengths and weaknesses, but the versatility, openness and universality is the main advantage it has. Nearly everything it can do, can be done better by some dedicated device, but if you want to carry around all those dedicated devices, you're going to need two backpacks instead of just a cargo pants pocket.

And even for some of the things it is supposedly not good at, it can be surprisingly decent after all. For example image manipulation in The Gimp. On the Pandora, I first thought: such a small screen, no mouse, and a relatively weak cpu; this will suck. But there are also advantages: you can get images straight from an SD card or a USB hard disk (try that on a phone or tablet, and even most desktop and laptop PCs don't have SD card slots), and the resistive touch screen is quite suitable for doing image manipulation. Most desktop PCs and laptops don't have a touch screen at all, and phones and tablets usually have very inaccurate capacitive touchscreens. So all in all, I'm hesitant to say the Pyra will be less good for something like The Gimp than a typical laptop.

It's like a Swiss army knife: sure, the screwdriver and corkscrew may not be as convenient as their dedicated counterparts. But then you discover that using the swiss knife holder as a handle, you can actually get a better grip and get more torque than with a normal screwdriver or corkscrew. It then becomes clear that utility is no total order.