The Linux Foundation
This is an old news already but we are now working with the Linux foundation.
I've noticed that companies are seting up different forums and clubs to standardize/promote/develop Linux/open source based technologies for mobile/consumer/embedded devices. I have my doubts... I do not fully understand how they plan to work. I'd go directly to places like the kernel.org or to GNOME to get things agreed, aligned, and -the most of all- developed. These are the communities that do the actual work and I'm not sure how these additional forums add value.
It may be that big companies are so used to do high level standardization and high level business deals that the simple community work doesn't seem real to them. They do not know how to deal with open source development, so they try to change it into something they are more familiar with: NDAs, business deals, and standardization forums.
I believe the Linux Foundation is slightly different. It is a forum dedicated to a specific open source project, Linux. And it promotes the community aspect of the work at hand.
I've noticed that companies are seting up different forums and clubs to standardize/promote/develop Linux/open source based technologies for mobile/consumer/embedded devices. I have my doubts... I do not fully understand how they plan to work. I'd go directly to places like the kernel.org or to GNOME to get things agreed, aligned, and -the most of all- developed. These are the communities that do the actual work and I'm not sure how these additional forums add value.
It may be that big companies are so used to do high level standardization and high level business deals that the simple community work doesn't seem real to them. They do not know how to deal with open source development, so they try to change it into something they are more familiar with: NDAs, business deals, and standardization forums.
I believe the Linux Foundation is slightly different. It is a forum dedicated to a specific open source project, Linux. And it promotes the community aspect of the work at hand.
posted by Ari Jaaksi at 3:06 PM
35 Comments:
Ari, you say you "do not fully understand how they plan to work". You know what? I do not fully understand how you expect Maemo to work.
Not long ago, your company made the decision to not further provide software updates for an almost new product, the 770. If this was just some closed source device, nobody would have noticed or cared. However with Linux bases systems, consumers have different expectations. Consumers expect to be able to upgrade almost any piece of hardware to the newest version of their favorite distribution, especially if the hardware is made by the same company as the distribution.
The OS2007 hacker edition shows there's really no technical reason not to officially upgrade the 770 to OS2007. It just works. (I'm not talking about Opera etc., which, as an application, is not part of any ITOS)
Even worse, your disputed decision about the death of the 770 line shows how much of your ITOS is, in fact, closed source. The whole thing is not open at all. Development cannot be done in the community now that Nokia dropped it. We simply don't have the sources. This is not what consumers expect when they read about a Linux based device.
This is why I, at the moment, do not understand how Nokia's Maemo strategy is supposed to work. I can only hope that some day we'll get some other distribution up and running on the great hardware you put together. (Ångström, anyone?)
Quote" Development cannot be done in the community now that Nokia dropped it. We simply don't have the sources. This is not what consumers expect when they read about a Linux based device." End of quote
Linux and openness was the main factor for me initially to choose 770. I had the impression that this would be the safe way, upgrades and hacks we sure come around. And all this could still be achieved with open source OS, by the Maemo community. And once again Nokia would be one to win in this. The 770 as hardware has a lot of potential that should not be let unused
I am wondering what Nokia wants to protect by keeping essential bits of the OS closed. Does it want to keep the competitors ie. possible cheaper chinese hw-copiers and look-a-likes out of the market ?
Call me a conspiracy theorist but the only possible reason for which Nokia would want to keep some sources closed would be so that customers, frustrated with its abandoning of an "old" device, upgrade to a newer one.
Well, I've got news Nokia. This is not the cellphone market. The people who love Maemo do not do so because it is 'the flavour of the week.' They do so because they love the platform. What Nokia did to us was despicable on many levels: they abandoned a whole bunch of users and they caused a fork in the development process, further dividing their user base.
Read the forums Dr Jaaksi, and you will realise how many customers you lost--the amount of bad publicity Nokia is (justly) getting is staggering.
In fact, with the more enthusiastic users you not only lost a happy 770 customer, but you also lost a N800 (and possible future upgrade) customer. Plus all those possible customers that speak to that disgruntled user. And the possible customers that speak to them...
Maybe I'm being melodramatic. Who knows? At this point releasing the sources would do much good indeed for the public face of Nokia. It would also reverse some of the damage done to the Maemo community.
And on a completely unrelated note: Nokia should get its act together as regards technical support and repairs for the 700/N800 both in Europe (UK support is terrible!) and in the US.
I hope you are reading this, Dr Jaaksi, because Nokia would do very well to listen to its customer base.
I could not agree more with you. Cheers, PP
http://ppires.wordpress.com/2007/04/11/yes-mr-ari-jaaksi-you-are-so-very-right/
> Call me a conspiracy theorist but the only
> possible reason for which Nokia would
> want to keep some sources closed would
> be so that customers, frustrated with its
> abandoning of an "old" device, upgrade
> to a newer one.
Conspiracy theorists, allow me to lay out an alternate scenario for you:
* The N800 has a Texas Instruments OMAP CPU with an integrated DSP and video accelerator. The 770 has neither (afaik).
* Nokia probably chose this chip for cost/performance reasons
* TI surely provided Nokia with drivers & libraries for interacting with the DSP and video accelerator
* I would bet money that TI's drivers are closed-source
* Nokia probably said, "We can waste two years developing our own open-source drivers from scratch, or we can ship now with a small amount of closed-source stuff"
* Guess what? They shipped the device asap.
As soon as open-source drivers for that DSP and video accelerator exist, you'll probably see a fully open-sourced image for this device.
(I have no affiliation with Nokia, but I have worked for one of TI's DSP competitors. Anyone who has ever worked with chipset vendors can vouch for the obviousness of this scenario.)
I will not purchase an N800 device because N770 is not being supported anymore. There are still serious issues in the kernel which are not addressed in the latest 2006 kernel. I don't want a hacked 2007 image either. I want a new 2007 image for the 770. It's only been what, 18 months since the thing was released, and already it's an abandoned project? Without access to all the source code, development cannot continue on the 770. Now we just have to hope someone can reverse engineer these and provide proper linux support. However, this puts a bad taste in my mouth concerning future Nokia hardware, and with news about WiMax support coming in 2008 it seems like even the N800 won't last more than 18 months. The PSP is 2 years old, the NDS is 3 years old, and both of those handheld units are still being developed for... home consoles have a 5 year lifespan. Is it too much to ask to have a 30 month lifespan for a handheld?
@matt
> * The N800 has a TexasInstruments
> OMAP CPU with an integrated DSP and
> video accelerator. The 770 has
> neither (afaik).
The OMAP1710 CPU in the 770 has the same DSP that is present in the OMPA2420 CPU found in the N800. While the OMAP1710 lacks the video accelerator present in the OMAP2420, Nokia have no intention of using the video accelerator. There is no kernel support (or planned support) so essentially the hardware in both devices is the same apart from speed/memory/camera. Abstracting the OS and supporting both devices with future OS releases is not that hard as has already been proven by the "OS 2007 Hacker Edition".
If Nokia care about their reputation and the future success of this platform (which is by no means assured given the current community pain) they should reconsider their position with regard to ongoing support for 770 - the easiest option would be to officially release OS 2007 for 770.
Until Nokia sort out this mess, people will continue to hijack Ari's blog posts with 770 related rants - whatever else he has to say people aren't interested in, as his previous promises have now been shown to be worthless. He needs to regain credibility within the community he is trying to build, but is successfully destroying.
I'm pretty disappointed with what I've been reading since buying an N800 on Monday. I now expect nokia's relationship with the open source community to deteriorate if the n770 is abandoned without allowing the community to support itself.
I'm tempted to get a refund for this n800 and get a stack of gumstix boards instead. It is my first nokia product and would no doubt be my last should I opt for a refund. I've a few days left to make up my mind. I hope to see more discussion of this here, as it's an important issue.
Get over it!
Nokia decided to create a product using open source software. They've always said that parts of the software will be closed. (http://www.maemo.org/presentations/ari_jaaksi/ME9_LinuxWorld_2006.pdf) But show me another major device manufaturer that comes even close!
They've made many contributions to GNOME and other projects. They are the main sponsor of debconf, Guadec etc.
The created a product (770) and provided several software upgrades to it. Then they launched another product. After the second product they have released only one bug fix release to the old hardware.
Ari has ben very open in his communication. He said immediately -- no OS2007 for 770. I'm not sure if that was a good business decision -- but that is what he said from the beginning.
So Ari has been open and telling us what is happening.
> But show me another major device manufaturer
> that comes even close!
FIC, OpenMoko? Do some research first.
> After the second product they have released only
> one bug fix release to the old hardware.
This bug fix was prepared in November 2006 and released after the N800 in late January 2007. It had one or two bug fixes that related to WiFi - that was all, nothing more! It's timing and paucity of fixes is highly suggestive of a tactical release - it is meant to persuade people like yourself that Nokia were still supporting the device when in fact they gave up months before.
> So Ari has been open and telling us what is
> happening
He indicated there would be ongoing support for the 770 - many of us took that to mean ongoing bugfixes and software updates (NOT new features, we don't expect that but we do expect Nokia to fix their buggy software) however apparently he only meant that you would still get your device repaired if it broke. Big deal, that's implicit in the warranty so it's reasonable to expect "support" to mean "more firmware updates". Let's be honest, Ari wasn't open or clear - he was mealy-mouthed. He'd make a great head of state, but he'd probably bankrupt the economoy in his first term.
Anonymous said "major manufacturer". I don't know about FIC so I would hesitate to call it major, OpenMoko is definitely interesting but not major. So Anonymous might very well have done his research.
I can understand the disappointment some 770 users may feel, but I think the proportions of this are getting out of hand. Compared to what was available before, 770 was progress. And progress has intermediate steps.
I'm not the best technical guru around but I have it on good faith that OS2007 hacker edition is, to a degree, proof that Nokia could easily support the 770 by merging the two OS lines together, thus making it easier for themselves and developers alike to develop software and fix bugs for both platforms. The cherry on the cake, of course, would be having a totally open source system, although I don't think that's about to happen. I have seen several instances where devices were rendered _much_ more popular by being totally open. Their names escape me right now but if I'm not mistaken there is a particular router (Linksys?) that has benefited greatly from open source. In the Linksys case I am not sure whether it was Linksys itself that opened up its router. In any case, Nokia would probably be better off it they gave the full sources to the community. Just my opinion.
"I can understand the disappointment some 770 users may feel, but I think the proportions of this are getting out of hand. Compared to what was available before, 770 was progress. And progress has intermediate steps."
Qt was progress, and still there were lots of people who disliked their QPL and disliked KDE, as a result (this point is moot now, since Qt is GPL). Shall we let N770 (and N800, I gather) be flower of one day (as the Spanish saying goes), to be dropped after just a year and a bit? I had heard of fast turnaround cycles in the mobile industry, but we aren't talking mobiles here. Some people may be able to shell $400 each year and be happy with it. Many others aren't.
But I digress. I would like to be sure of Nokia's commitment to F/L/OSS on both N770 and N800: I find both to be a good option between mobiles and laptops. For the time being I can't. In the end, if Nokia really isn't going to support N770, let them do what LEGO did: open the programming specs so people can substitute the hardware drivers, for example. That is the right thing to do: if you don't want to use some knowledge any more, you don't just hoard it, you let it go (I already had a problem like this with Agfa and an unsupported parallel scanner: in the end, I decided I wouldn't buy Agfa again, and not recommend it to anyone). Because this is what happens: discouraged customers go away and won't talk positively about you if you let them down like this.
You may consider this is getting out of hand. I consider this is the foreseeable answer to an out-of-hand decision. In the end, whether it is a major manufacturer or not isn't relevant: you can either be fair or get criticized. I know Nokia is big enough not to care about N700 and N800 users (and I'm not saying that they don't care, actually), but don't ask me to shut up and smile on top of it.
> I don't know about FIC so I would hesitate to
> call it major, OpenMoko is definitely
> interesting but not major. So Anonymous might
> very well have done his research.
Hang on, you admit you don't know who FIC are yet are prepared to give an opinion? Strewth, *do some reasearch*!
FIC (First International Computer, Inc.) is a large Taiwanese computer and components manufacturer - they also manufacture products for other companies worldwide so just because you haven't heard of FIC doesn't mean they're not major, you may even have owned one of their products!
Being a recent N800 purchaser having used Windows Mobile and Symbian as my mobile platforms, the Linux world is an interesting place. The ire that people direct towards Mr Jaaksi is astounding. The fact that Nokia are even in this venture is a sign of how far Linux has come in the consumer space. They are learning how to work with the community.
I think the largest divergence between the parties views might be in what is considered the correct life span for a device. Nokia as mobile handset creators are locked into a vision of the 12 to 18 month life cycle with 2 or 3 major updates, versus the PC/linux community which seems to be more on a 24-36 month cycle with constant software updates.
I certainly am hoping to use my N800 for the next 2 years as an all purpose device and then upgrading after that and keeping the N800 as a dedicated device for a specific use.
I'm sure Mr. Jaaksi has read the complaints and I'm sure the Nokia powers that be know what the more vocal members of the Maemo community want.
MRP
Quote"I can understand the disappointment some 770 users may feel, but I think the proportions of this are getting out of hand. Compared to what was available before, 770 was progress. And progress has intermediate steps."End of quote
Simply answer me. Is it fair towards customers that you buy a product as new, pay full price. Product´s software is buggy and no corrections are to be expected, but buyer would not have known about this, unless he spent time reading Blogs and forums?
Not to nitpick, but those of us 770 owners who worship the IT forums did not really expect Nokia to let us down this badly.
I still cannot understand why Nokia cannot see the sense in either (i) merging the 770 and N800 product lines by releasing a supported OS2007 that works on the 770 (the hacker edition showed us that their excuses are unfounded); or (ii) releasing the sources, or more details about the closed parts.
In both cases, sales would probably increase and their reputation will probably be salvaged. Just my opinion...
I can understand Nokia not caring about couple of hundred (maybe thousand) unsatisfşed customers. What I cannot understand is after defending so much for an open source product and getting a whole community behind it, how Dr. Jaaksi can leave this community of 770 developers in a situation where almost no further useful development is possible.
Very disappointing...
There are valid points, both pro and con, but there is way, way too much hyperbole from the severely disappointed sector. As a data analyst and quality specialist I understand well how emotion can taint objective data, and that natural tendency can cause a defensive reaction from the very people you wish to consider your points.
So I ask the disgruntled community this: reframe your comments. Your feelings are certainly relevant and worthwhile, but they can cloud the true message. Try to strip the anger from your responses and adopt a more persuasive tone.
Ask yourself: what is my goal? Is it just to vent? Or do I want change? If it's the former, ignore me... but otherwise,poisoning the well won't facilitate positive change.
On the other hand, Dr. Jaaksi, as a colleague (more or less) I implore you to overlook the ire and desperation infiltrating these pleas and drill down to the core disappointment. Nokia can and must improve in critical areas if the platform is to attain critical mass. I strongly believe it's at a make or break point as we speak-- at least within the consumer space.
Dr. Jaaksi, I'd love to chat with you more if you can find the time. If you didn't save my work email feel free to use my hotmaila ddress.
Any comments on involvement of Nokia in this development?
https://intel.wingateweb.com/publish...003_100eng.pdf
texrat, I think the message is quite clear, as far as I see it: dropping support after just over a year _and_ having a lot of the code being still propietary after the fact isn't "playing nice". Of course, it's their product, but you don't make new versions sell by leaving early adopters high and dry.
I have read Nokia provided some driver code and it didn't have much success. Since I haven't found (probably my mistake) references on that, I can't say whether that code was workable or not, nor whether that code was enough to have an open system or not. Certainly, if it is workable and enough to support an open system, then those affected by the support drop can put their money where their mouth is and build up a support community. Unfortunately, all I know about this issue points to the system not being open enough for such community. Feel free to prove me wrong: I'd rather be wrong for a minute than for a lifetime.
I thought the readers of this thread might be interested in how the first embedded device community of open source developers created new application software back in 1981. What these engineers did as a group back then is very similar to the challenges we face today in finding value and community:
Community Software Development for Embedded Devices
http://www.embeddedcomponents.com/blogs/2007/04/community-software-development-for-embedded-devices/
baka, I didn't argue with any of your points. ; )
It's an opensource product from a monolithic company; this is a difficult mix at best: management of large processes tend to require more discussion and socializing than typical OSS projects, and developers may have to trade-off their personal pet-bugs versus corporate incentives.
Ari may be fighting for exactly what we want, but be limited by the fact that he represents Nokia. Nokia may want us to have all they they have, but there's an implicit test, QA, and support burden for a product that is out the door. These wouldn't be bugfixes (a-la-Patch Tuesday) but re-issues of the OS (Tiger/Jaguar). For free.
I'm not defending it, it sucks, but I'm saying that be implications of a business limit what it can offer, and steer the effort a bit towards obvious business cases.
In about 9 to 12 months, I'll bet that Intel gets Open Source "right" with their new MID platform. Intel is a monolithic company, same as Nokia, but with one difference - they understand open source and how it can be used to sell hardware.
Nokia can continue to use their existing excuses for cocking up the delivery by using a mixture of closed source and open source, however nobody will be listening though as we'll all have moved on to Intel based devices. Nokia will be out of the game - a game they started.
Time is rapdily running out for Nokia before Intel eats their lunch. Nokia have not rewarded the community for their loyalty (ie. 770 fiasco) so Nokia should not expect any loyalty in return - when (and not if) a new kid arrives on the block offering better candy, he'll be the communities new friend, goodbye Nokia.
The comment above is so childish and naive. It shows that the commentor doesn't have a clue!
C'mon, this Nokia vs. Intel comment above is far from reality. It's not Nokia vs. Intel. It's Nokia and Intel now, together. Nokia can proudly say they pioneered in this field, but still they get together with their competitors FIC and Intel to push development further. I think this is great and will be good for Nokia and the Mameo based product line.
BTW: Intel said, IIRC, that they started to develop their Linux-Platform in 2005. Does this ring a bell? 2005? Maybe they were part of the Hildon/Maemo-development all the time? (Which could be a possible reasons why Nokia keeps some components closed even though they wouldn't have to from a technical point of view.)sw
Internet Tablets exist because they are based on mobile phone technology - the kind of technology that Nokia buys in spades, and gets dirt cheap which is why they took a punt on the devices in the first place.
If Intel enters this market, they could stand to dominate it within a short while. What choice for consumers then? An Intel x86 tablet or a Nokia ARM tablet? Tough choice eh? Assuming battery life is similar, who in their right mind would choose the Nokia tablet?
So what choice for Nokia - start building tablets using ultra low power Intel x86 chips? Not likely, unless Nokia start selling those chips in millions of mobile phones and get a stonkingly good bulk order deal from Intel. x86 simply isn't cost effective compared to ARM. Perhaps Nokia could compete at the uber low end, but I doubt it.
As for the comments about Intel/Nokia being childish - why? Believing that Intel will open source their platform in a way that Nokia has so far failed to do, why is that childish/naive? Working with the community rather than against it (which is sometimes how Nokia come across) will work in Intels favour in the long term.
Ari, could you please share your thoughts on GMAE with us? Nokia is part of it, but so are future competitors. What does it mean for Maemo users/developers?
The comments about Intel are childish in their presentation and naive in their content. To suppose that Intel will execute better than, or very differently from, Nokia is speculative at best. Ther eis no track record other than Nokia's on how monolithic corporations (especially those beholden to stockholders) will execute in this space. Nokia has made many missteps in this new frontier and if another major corporation does better it will be mainly due to their hesitance to be a market leader and rather a follower who capitalizes on the mistakes of predecessor(s). This is not in and of itself a bad thing, but if Nokia can correct course now they still have the advantage of first-to-market and can continue to parlay that into future success.
The black-and-white viewpoints expressed on this subject are disingenuous. This is not an Old West shootout. And those who think Nokia's gaffes in this new arena are without precedent are ignorant of history. This occurs every paradigm shift, folks. Consumers have a choice: wait until the market matures to jump in, or take the risk of being an EARLY ADOPTER. Sorry, but I have no sympathy for anyone indulging in the latter without conducting due diligence. New markets are not for the meek. It remains to be seen if Intel or anyone else can do better.
Ari, please explain how they were
able to run an external monitor
from the N800 to an LCD panel, this was done for the CES 2007 launch
To the above:
By running VNC Server on the N800 you can display the N800 desktop on a normal Windows or Linux PC that is running VNC Client - the latter obviously being connected to a large LCD panel.
"new markets are not for the meek."
well, then how about if nobody buys any of the nokia770, just because everybody knows what you know and does not want to risk?
What would be the reputation of Nokia in the field of handheld computing? What would all the blogs, magazines, etc be speaking about? That nobody wants it, coz nobody wants to risk? Coz everybody will buy it only if everybody already has it? What would then be the budget used to develop the next machines, if there would not be any income from the firsts one?
I am not a real developer, but from a user point of view I need to tell you, that having a device, that not only crashes evry 10 mins while writing a mail for it, but is also no more supported by (at least) the developers, coz they are already compiling their stuff for n800, and no more fore n770? This is a little bit too much. So, Nokia, try to be loyal to your customers who did invest their faith and money in your succed in this handheld field, and did not go for Palm, and other device! Ok?
Post a Comment
<< Home