Re: Closed Source Packages in Maemo
 

Because there is already an open market for PC. It wasn't built by opensource operating systems, rather Linux piggybacked on the commercial success of DOS\Windows to take advantage of aftermarket x86 components as their host.

To understand the futility of your argument, you have to see it from Nokia's view. (Don't be selfish)

As I said, if you can come up with a good and thriving example of a company that releases a brand new device, with 100% opensource codes that is not directly compatible with anything before it (Your PC argument failed this condition which I have stated in my previous post) then we can talk on even grounds.

Re: Closed Source Packages in Maemo
 

WHOA! Where is all this more open Chinese crap coming from? I WANT ONE! :)

Re: Closed Source Packages in Maemo
 

Indeed, I do not see the whole market crashing down due to these perfect chinese replicas. What I have seen of them, are people going "oh crap, got ripped off, need to get the real thing now as this sucks".

Re: Closed Source Packages in Maemo
 

PC's aren't the same as cell phones (and mostly not the same for other pocketables)

PC's are commodity at this point. There's very little research that goes into the actual design of a PC. It's mostly in the case, and in what add-ons to bundle with the PC. Most PC makers aren't making their own motherboards, power supplies, display sub-systems, etc. Nor are they doing extensive evaluations of which CPU architecture to use ... and they can mostly use an off-the-shelf firmware/BIOS, with minor customizations.

NONE of that is true for mobile device makers. For cell phones, it's almost all custom hardware. Which means a whole different level of overhead for research and manufacturing.

The value in PC marketing is in "packaging, assembling, burn-in, and support". The value in pocketable devices is all of that ... PLUS the R&D that goes into design and prototyping both the individual components AND the overall devices ... PLUS the manufacturing of all of that.

Not the same market space. Not comparable situations. Anyone can buy off shelf commodity PC components and build a fully usable, fully functional, PC That is just as good, and in many cases just as physically "attractive", as a brand name PC. I'm not aware of ANY such capability in the pocketable arena.

For example, the TuxPhone isn't as fully functional as a all-round pocketable device. It is no where near as aesthetically pleasing. And, last I checked, it's a monster (in looks, size, and usability).

When someone is manufacturing pocketable motherboards, pocketable system cases, pocketable displays, pocketable keyboards, pocketable ...etc. that are all able to fit together into a single usable specification/standard (like the way you can with a desktop system) so that ANYONE can build an entire clone of an E90, E71, N95, G1, Touch Pro2, riser phone, or flip/clamshell phone from off-the-shelf components, and with easy to purchase and install copies of the OS ("just put this in the microSD card slot, boot from it, and it'll install into your firmware") ... THEN it will be comparable. Not until then.

You can't make an easy/direct AND meaningful comparison of commodity product business models and custom product business models.

Re: Closed Source Packages in Maemo
 

Hi, please take this answer as a personal opinion and please don't limit it to Nokia since I believe it's mostly the same for any other company in the same position.

Sure but if you look at the trend, "materials" are decreasing the % of relevance while software increases it.

"fail to imitate" fails to reflect the reality at least in two angles. One: cheap copycats are only one part of the story when a considerable % of the production of "western companies" actually happens in "easter soil". Two: companies actually protect such "innovative features with hardware" through patents since "western companies" are so willing to imitate than anybody else - if they only could.

And you mention China. Have you noticed that they are moving fast from subcontracts and imitation to genuine know how and innovation, filing own patents at EU / US levels?

Sounds like Qt. Nokia WebRuntime and Ovi SDK point to that direction as well.

If you look at Maemo you have the % of closed source mainly in two layers: hardware adapation and applications/services.

For the hardware adaptation you actually need to start convincing the chipset vendors since most of the closed software is licensed by them. If you answer "choose open hardware!" then you probably will compromise the previous points of "material of incomparable quality" or "innovative features with hardware (...) fail to imitate". So you need to have an alternative business model to chipset vendors in addition to the alternative business model to device manufacturers.

At the applications/services level... it's really complicated to make sustainable and competing innovation with open source (as much as I would like to see it). Commoditization sure, but pure innovation... It's possible, but complicated. It's like playing against Deep Blue: every time you win your competitor can immediately assimilate the lesson and use it against you with less investment and effort.

Yes, you can also copy closed source but the difference is time to market. If you release your closed apps only when the device is launched and in its way to the shops it's really different than developing them openly all the time. Specially the competitors caring less about "incomparable quality" might be able to ship a product with your new software even before yourself, while your "incomparable quality" standards keep you bugfixing (providing the bugfixes for free to the customers of your competitors).

All this might be worth if there is a critical mass of users and oss contribution around certain application. Maemo made a bet with Modest, and the contributions were also modest (yes, you can blame our mixed-open development but still). We are making another bet with Mozilla and the equation results better since the Mozilla engine is used by millions, tested by thousands and heavily contributed by hundreds.

The browser is a good example of open source innovation, but note that is an area where all relevant players seem to be moving towards OSS models on top of the Mozilla or the Webkit engines. It is much easier to compete with open source when your competitors are also doing the same.

Nice sentence, but easy to rebate in a business plan for Nokia. RIM and Apple are doing good profits this year. They seem to be scoring well in quality and customer engagement looking at the levels of satisfaction of their users. Yet they achieve that not through freedom but quite explicit control. Software freedom doesn't seem to be a cry of the millions of customers of Series40 and S60, the platforms that are bringing the big profits to Nokia.

You are missing the first step "Get millions of fans through the traditional multinational labels business". That was the case also for Hole, Robbie Williams, Gilberto Gil, Radiohead and many other great artists I love and have an attitude pro-CreativeCommons, file sharing, etc. Or do you know a professional band that hit the charts creating open music since Day 1?

They might come in the future, but not today. And this is similar to what Nokia could say about Maemo. Maybe one day it will be 100% free, but not today.

For all these reasons I think you would have a hard time convincing the Nokia shareholders (and even most Nokia customers, current and potential) with your business plan.

Good that in Nokia we have a good bunch of people thinking in open source innovation together with beautiful products and profitable business, all of them contributing to actually quite innovative business models around free software. This is why Qt was relicensed, this is why Symbian is moving to open source, and this is why Maemo will keep being a very interesting platform for freedom lovers.

Re: Closed Source Packages in Maemo
 

That is an interesting problem and one of the more important one in the end for the long term viability of an open platform... If users can't get to the hardware it's game over. Hopefully hardware manufacturer could get pressured by such big players as Nokia to release source, at least a year or two after its bleeding edge chips are out ... ?

But not compromising on that choice makes total sense, of course!
It's Nokia's business to make kick-*** phones :)

Considering how competitive the market seems to become... I'm wondering if, Nokia being at the center of the real platform direction / innovation, shouldn't right now completely aim for a 100% FOSS platform...
It would completely satisfy the FOSS world and, even if you do get a bunch of clones, it's going to make the market share grow, as a platform! Growth which ESSENTIAL right now against Apple's lead...

As you said, the hard part would be to try and balance this with the fact that you're competing with others on the same game! Since Nokia now owns Qt and could start taking a bigger role in other leadership roles it should be adapting quicker to new needs... It's hard of course to exactly point at what needs to be done as this is the million dollar question but i think that the first company to set foot in this kind of model will have a tremendous advantage.

I'm just throwing ideas here but when you see softwares like MySQL being able to sustain themselves while being GPL or Qt for that matter, I think there is something to be learned about the history of the PC openess or the TCP/IP history and how it has beaten the other architectures or network protocols.
I may be hoping for history to repeat itself...

Maybe a temporary license where we can see the source but not share it might make sense so you can still sue rip-offs and after a while completely free it? I don't know what might work but it's worth thinking about ...

As the platform is moving toward a portable PC-like experience, I think that selling proprietary software when it is the "bleeding edge idea of the year" might make some sense but anything else has to be considered commodity and moved to a community/leadership model. The proprietary possibility has to be left as to be able to create an app market for those who still think their great idea is worth 3.99$ and creating a real diversified market of license type.

Modest is probably too new and came out directly from Nokia and thus it will take time to build a community for an e-mail client used on what is right now a fringe platform... I think Nokia did a great job with the community and really is going in the right direction, it might just have to learn a few more things about how to deal with a community.

Software like Thunderbird came out of desktop environment groups whose goal are to build a complete desktop.
It's going to be hard (though Qt will help) to standardize a new desktop environment to build a community like KDE or Gnome, but if successful it would be an incredible success and advantage over competitors for Nokia!

Hopefully, as the smart phone is really in its infancy, we will see this community grow very quickly in the coming years and I hope Nokia can get as big a part of the pie as possible... But now is not the time to try to cash in too early! Though, time is running out...

I'm wondering why is everyone moving toward a FOSS solutions... The browser has become a commodity and there is not much you can do to innovate except for speed and usability (GUI?) without going through standard bodies, and that's a great thing, otherwise we wouldn't have the web!

As you said real innovation in the FOSS vs FOSS is hard and require great ideas and i think having your hands deep in some key communities!

I'm wondering, is the new browser front-end (GUI) in Maemo not FOSS ? (I know the back-end now is, and that's great!)
But, If not Why not?
How is that different than the media player which is even more of a commodity to me?

I'm guessing this is a question of market segment... and marketing, which Apple has always been good at, and people sucker for it...!

But it's true, most people don't care about their freedom!?

But we're moving toward a more complex world of information and the bet is now on whether more powerful device are going to be useful or not... And i think that for the market which will require the best adaptable devices the FOSS model will be an important key to success.

But yes, very specific dumped down solutions for different markets will probably always succeed. So, meanwhile, of course selling simple phones over and over again is going to generate a bigger margin of profit than trying to push the envelop like the NXX0 serie is... And while people looking for simple solutions are easy to please, those with complex problems will require more than a glowing fruit or soap bar. But of course, marketing does miracles that are actually horrible technically...

I wonder if we will get to a point where the software platform for cell phone will be moving toward a common base and/or formats without the type of lock-in that we see right now from Apple or Microsoft and all...
These type of barriers and vendor lock-in these days are really a barrier to global full scale cooperation and huge social networks unthinkable today, that and the carriers...
What if the 2 billions cell phone on earth could exchange more than SMS as their most complex common language?
This won't happen in any interesting way without open standards and FOSS...
Though Nokia got that by trying to sell phones directly, bypassing the carriers responsibility on the lock-in situations. I'm just hoping it's not to take their place!

Just playing devil's advocate here, but maybe the whole model of trying to be #1 is becoming outdated in the context of music and it's going back to what it was 60 years ago when we had 100 time more choice in music and diversity of labels now that people can access any music from anywhere and not get forced into the big label pop crap...
Same goes for every phone on the market where the diversity of solution right now is still limited by technology and Nokia's tradition of trying to hit every single niche market is a good idea in that respect! Though in the near future we'll see devices capable of a lot more and thus an open platform is the only way forward to bigger social networks and thus markets.

The whole model is on the verge of a big change.

With all this said, I can understand how hard this is going to be and doing it progressively through Modest and other test is a good idea! Keep going :-)

Go Nokia Go !!! ;-)

Re: Closed Source Packages in Maemo
 

@ Yssss M#131:

You say that my example is not just because the computer is developed around the model Win-tel.
But it's wrong, my example is correct.
If computers had developed around a free OS, the computer now will not be less open. Instead it will be more open, freer (as free speach) and better quality.

The computer market is open today because it has been well established. And it was created thus beacaus it's a working model, companies make money and user have got choice.

So no, my example is not futile.

@ johnkzin M#134:

The N900 is not so different from a laptop.
You say that PCs are assemblies with parts of different manufacturers.
It's True, but who made the processor N900? Nokia? No, that Texas Instrument. It's same for the graphics accelerator for the audio amplifier, etc... They are not manufactured by Nokia.
PC manufacturers are producing components that can not produce themselves by other manufacturers, it's the same for Nokia and the N900. The N900 is built the same model that a PC: a processor, memory RAM, memory to store OS and user data, input devices and output devices, etc. ...
Finally, the N900 is a PC, for pocket, but a PC. It is more nigh of TouchBook (a Netbook ARM processor) than the 3310.
And Nokia too says the N900 is a Mobile PC.
So yes, it requires more work to create a N900 than a laptop, but why should that stop Nokia adopt the same policy of openness, for the software, than it is possible on the PC market?

Your arguments do not convince me, he instead makes me think more than I have right and that everyone would gain something if Maemo became 100% FOSS (exept some drivers): Nokia, the Maemo community and users.

Currently, the following model will be ideal (from me):

N900:
- Nokia allows, as a PC, install the OS than user choice.
- Like on a PC, an OS is pre-installed: Maemo 5.
- Basic, except for some drivers, Maemo is 100% Free.
- Proprietary supplements (Skype, Adobe Flash, codecs audio / video, etc ...) are automatically installed if the user seems to need it. (like Ubuntu). If the user does not need them, they are not installed.

Maemo, same model as for Ubuntu:
- A non-lucrative foundation is created: The Maemo Foundation.
- The Maemo name and logo have become the property of the Maemo Foundation.
- The Maemo Foundation aims is to manage community development around Maemo, organize the promotion of Maemo in event around Free Software and ensure that Maemo is Free.
- Nokia becomes the main sponsor of Maemo. Because Nokia needs Maemo to sell its N900, Nokia provides developers in full-time (as for Canonical Ubntu) and fiances.

@qgil M#135:

I agree that the business model I was talking about and that you quote may be impossible to present for the material. But, there are experimental projects underway in this regard.

But the business model that I present some lines above them is functional. Moreover, it solves the problem that you say in the software "competitors use the Nokia work before Nokia."

Indeed if Maemo is developed by a Maemo Foundation, everyone is equal, problem solved.

Because here, the problem you says is not resolved by the actual model:

In the hardware layer:
A competitor has the hardware drivers that he will sell, so anything that Nokia make don't change anything. Proprietary software is only acceptable in a pilot, if understanding of this driver permit copy the hardware.

In the Application layer:
With the exception of OVI MAP, any software application developed by Nokia could be replaced by a FOSS equivalent taken in on GNU/Linux Desktop. I explained this a few posts before this one.
And ultimately, the competition will have more FOSS software application with more functionality. So, in addition to taking customers to Nokia, the competitor will get a larger community.


The problem you are saying now is not resolved by the model chosen by Nokia now, but the model I propose in this answer resolve it.

So ultimately, I see no reason for the presence of proprietary software other than some drivers.

So, why?

Re: Closed Source Packages in Maemo
 

Personally, I feel there is nothing wrong with proprietary locked-down hardware as long as the drivers are up to scratch and not putting huge restrictions on what you can and cannot do.

Now on PC sadly this is not the case, mostly due to the huge variety of chipsets and software revisions that cause instabilities. However something like the N900 is like a games console, they are all identical so you can tweak everything for that specific hardware combination so these instabilities should largely not exist.

Its all about APIs these days and that is not a bad thing. Sure it means you can't push the hardware quite so hard, but for a mobile device especially the last thing you want is someone bypassing the OS and crashing everything, as the end-user will just blame Nokia. Just how often do people blame Windows for crashing when its just a bad combination of hardware and software.

So yes, a lot more care and attention is taken tweaking the hardware for cost, features and battery life - so again they have to ensure they can make back that investment and being 100% FOSS would not allow that. FOSS is never quite as polished as a commercial venture which is why a device like the N900 is so appealing in the first place. Because it has the flexibility of FOSS but the refinement of a commercial product.

Re: Closed Source Packages in Maemo
 

Alex Atkin UK, many of proprietary software are not as good as their equivalent free.

This is not what model is chosen than defines quality, but who working on it and the number of people working on it.

But Free Software has a big advantage for the number of people working on it and in fact more likely to have competent people working on it.

Re: Closed Source Packages in Maemo
 

While I use a number of Open Source programs, let's not go overboard. There are some Open Source projects which are massively supported with top level participants. There are also many more which are developed by a few average programmers who maintain the software sporadically and drop it unexpectedly. Much as I like Open Source, I have to admit that, in general, proprietary software is better tested and more stable when it is released and it is more likely to be maintained because people are being paid based on it being sold and upgraded.

By all means, find Open Source alternatives, use them, support them. But recognize that there are valid reasons to pay for proprietary programs as well.