Maemo is CSS contaminated (not entirely open)
 

I'm quite disappointed with what I've discovered. Maemo.org requires users to surrender their serial number and sign an EULA in order to update what was erroneously expressed to me as a totally open platform.

Exactly what components are not GPL'd?

Has there been any efforts yet to replace them? Could the black box components possibly be replaced by porting something from OpenMoko, Android, or the like?

I'm a developer who was highly motivated to contribute before I discovered this. Would it be feasible for a GNU purist to grab the GPL pieces and assemble an image that runs free of CSS contamination?

Also, from a legal standpoint, what rights do we give up if we sign the EULA? I wasn't even able to read the EULA w/out supplying my s/n.

Re: Maemo is CSS contaminated (not entirely open)
 

According to the manual (some of these have more restrictive licenses than GPL, some have less restrictive licenses):
This software is based in part on the work of the Independent
JPEG Group.
The Graphics Interchange Format© is the Copyright property of
CompuServe Incorporated. GIF SM is a Service Mark property of
CompuServe Incorporated.
Adobe® Flash® Player. Copyright © 1996-2007 Adobe Systems
Incorporated. All Rights Reserved. Protected by U.S. Patent
6,879,327; Patents Pending in the United States and other
countries. Adobe and Flash are either trademarks or registered
trademarks in the United States and/or other countries.
This product includes software under following licenses and
developed by following copyrightholders:
• Computing Services at Carnegie Mellon University
(http://www.cmu.edu/computing/)
• Tommi Komulainen (Tommi.Komulainen@iki.fi)
• The OpenSSL Project for use in the OpenSSL Toolkit
(http://www.openssl.org)
• Pedro Roque Marques
• Eric Rosenquist, Srata Software Limited
• The XFree86 Project, Inc (http://www.xfree86.org) and its
contributors
• Paul macKerras (paulus@samba.org)
• Purdue Research Foundation
• OpenEvidence Project for use in the OpenEvidence Toolkit
(http://www.openevidence.org/)
• Eric Young (eay@cryptsoft.com)
• Tim Hudson (tjh@cryptsoft.com)
Copyright © 1996 NVIDIA, Corp. NVIDIA design
pending in the U.S. and foreign countries.

Re: Maemo is CSS contaminated (not entirely open)
 

- i don't have any problem with entering my MAC address to get the os update...

- yes, i have to agree that the technical information and specifics regarding the platform are not outlined in a cohesive/comprehensive way... (the APIs are ref'd, but one needs to do some work to get up to speed - LOTS of reading!)

- from what i've been able to glean from postings and list searches is that the binary 'blobs' are for wifi and other chipset functions in the tablet (power management?)... others may want to chime in here...

- you can get the GPL'd portion of base distribution and other apps here:

http://repository.maemo.org/pool/

(choose your target os release, such as os2008)

- AFA development, you're still in control, as any software you develop and license under the GPL (V2? V3?) is still protected by that license... or why not just use the BSD license?

- you'll get to read a EULA when installing the SDK - specifically Nokia's binary bits... it should be embedded in the install .sh script... you can get the script here:

http://repository.maemo.org/stable/chinook/

- unfortunately "integration@maemo.org" did not see fit to allow one to peruse the embedded license in the Nokia binary install script unless one has a working Scratchbox install (really stupid, IMHO)... there *should* be a command-line option to view the license using the 4MB script...

<sigh>

- various license links are here:

http://maemo.org/intro/licenses/

(note that the way i read things, these licenses only apply if you choose to use maemo.org's resources... set up your own site, host your own software, set up your own repository, use your own license - JMHO, IANAL)

:-)

Re: Maemo is CSS contaminated (not entirely open)
 

Who expressed it as a "totally open platform"? This is purely a fiction of your imagination, as it's been made quite clear here and other places that the platform isn't "totally open".

Of note: typically "CSS" refers to a Cascading Style-Sheet.

Re: Maemo is CSS contaminated (not entirely open)
 

You might want to take a look at the Mamona project or the efforts of installing Debian.

Re: Maemo is CSS contaminated (not entirely open)
 

+1 Mamona looks *very* promising...

http://dev.openbossa.org/trac/mamona

Re: Maemo is CSS contaminated (not entirely open)
 

@jgombos: Your "serial number" is just your MAC address. If you're *really* paranoid about them seeing it, I'll be happy to just email or pm you the EULA, but really... (or are you saying you don't have a tablet?)
Anyways, lots of stuff is actually closed source. Chunks of the media player, parts of the desktop, some home applets, some statusbar plugins (brightness and volume at least), wifi, dsp (sound...) and bluetooth drivers, as was mentioned before, probably quite a few userland programs, including the browser UI. It *is* disappointing. :( Some of us (including me) are trying to get Debian working nicely, but for now closed source stuff is needed for sound, bluetooth and wifi. It's not a terribly useful machine without that :/ And you realize most of Android has no source at all yet, right? It's binary only until Google decides to release their stuff...

Re: Maemo is CSS contaminated (not entirely open)
 

Maemo is open. You have to enter your MAC address at tablets-dev.nokia.com to get a Nokia OS image, *based* upon Maemo.

Re: Maemo is CSS contaminated (not entirely open)
 

Specific to these devices, if some parts had not begun as closed, then the tablets could not and would not have been released. As has been noted numerous times on this forum as well, continuous efforts have been made to replace as many closed aspects as possible with open equivalents. However, as long as technologies like Flash are supported, they can't possibly be 100% open and I don't know how someone would get the idea they could.

Re: Maemo is CSS contaminated (not entirely open)
 

Thanks for all the responses folks. Y'all have addressed my concern quite comprehensively. Considering that Mamona exists and is in moving forward, I have no issue with developing Maemo apps on a temporary basis until Mamona is stable.

Regarding the MAC address, I'm very distrusting. I can only see how the collection of that information can be used against users and developers. Eg. A developer who becomes bound by the EULA (and the agreement acceptance is tracked via MAC address) produces an OS or component that somehow damages the profits of those standing in front of closed source components. From there it's just a matter of picking out whatever in the fine print in the EULA that can be used to threaten or sue the developer.
OpenMoko has found a way to do this w/out closed parts.
I believe I've started the only thread of this kind on this forum. But searching has its limits.
Frills like Flash could easily be a separate install. And if Flash and other non-essential technologies were the only parts preventing Mamona from happening, Maemo may not have been packaged with them in the first place (debatable) but certainly Mamona would already be complete, and it would have captured the interest of most developers; certainly all the GNU purists.

(BTW, Flash does have a GPL alternative)

Re: Maemo is CSS contaminated (not entirely open)
 

I understand you're distrusting, and I think that's wise and prudent. But, since they require you to accept the EULA to download it, what difference does it make if you provide the MAC? That you downloaded it is sufficient to show you clicked through the EULA, and the MAC of your tablet is in no way imprinted in the OS or components you might make, so I can see no way they could leverage the MAC entry. I understand it to be strictly a means of preventing people with no N8x0 (and no cleverness or motivation) from wasting their bandwidth. They already know what range of MACs they shipped, and if they can't tie it to anything in particular, it's no use, and I guess gets sent to the bit bucket as soon as it's range-checked for validity.

Re: Maemo is CSS contaminated (not entirely open)
 

The components that I feel could and should be replaced are those three mentioned in an earlier posting: wi-fi, bluetooth and possibly the dsp code (at least the former two are always open source in linux kernels for other devices). There shouldn't be any reason for having closed wi-fi and BT drivers.

Re: Maemo is CSS contaminated (not entirely open)
 

Jgombos, you are comparing apples and oranges when you invoke OpenMoko. Nokia hardware is closed. I don't see that changing. Plus, there were components involved from the outset that had no open source drivers or other means of access. Again, going 100% open would have meant that the Nokia tablets available now would not have been deployed.

Also, my statement about the subject being discussed numerous times is 100% correct.

As for flash, the GPL equivalent is a bit behind the commercial, is it not? Either way, that leads to a tangential argument that I won't indulge. It is what it is.

Bottom line, I understand your desire for 100% openness-- but again, there are very valid reasons why that is currently impractical with the Nokia tablets. I've barely touched on them. Others have already epxlained in greater detail in previous threads on this forum. I am sure that with the right approach the forum (or Google) Search would find them. The important thing to note is that as the platform has progressed, there HAS been continuous effort to open the tablets more.

EDIT: I find it interesting and ironic that you want increased openness from a commercial manufacturer like Nokia and then express fears of lawsuits over a simple, innocuous device registration. If you examine that dichotomy objectively, isn't the latter paranoia?

Re: Maemo is CSS contaminated (not entirely open)
 

There is no mac required if you update with the nokia update tool in windows.

Re: Maemo is CSS contaminated (not entirely open)
 

- someday Flash too will be free:

http://www.gnu.org/software/gnash/

(most swf7 features supported now, such as streaming video from YouTube or Lulu)

:-)

Re: Maemo is CSS contaminated (not entirely open)
 

I actually wish SVG would take over...

Re: Maemo is CSS contaminated (not entirely open)
 

At least you do not need to type it in ;)

Re: Maemo is CSS contaminated (not entirely open)
 

The Nokia tablet is a "form factor experiment" of Nokia, the target of it is not GNU / linux / <insert any other form license> opposite to for example openmoko or GP2X .

It's target are mainly *users* like me who like to have a device that simply works. 99% of this audience has no interest in getting a "GNU clean" system that has less functionality (because of driver issues with some parts or so) then a system where parts are closed source "binary blob" but work as advertised (or close... :) ).

If GNU purists succeed in making a system that is better then the provided SW stack I'll bet a lot of people will be more then happy to use it, but until then most "normal" users won't care to much - wich is nokia's target (in the long run) audience.
The fact that Nokia is trying to make a working business model based on (partial) GPL software is only to be applauded as it will only give more credibility to the opensource/gnu/.... "movement".

bootnote: I never heard someone not buying a TV because their TV firmware was not "opensource"...

Re: Maemo is CSS contaminated (not entirely open)
 

As long as this topic has been brought up again, I'd like to just ask a couple questions to anyone who can shed light on them. I can completely follow the logic for why wifi (and bluetooth and dsp) drivers ended up closed source: NDAs, patents and license agreements are a minefield in the hardware world, especially in embedded systems. Likewise, commercial flash support is an unfortunate necessity for the majority of the NIT's target demographic. This is nothing new, and from a business point of view is a pretty simple choice to make when push comes to shove (ie, do we want GPL drivers for wireless or the best/cheapest wireless chipset we can get?).

What I don't understand is all the ridiculous little bits of code here and there that are closed source. Things like the backlight and volume statusbar applets use publicly available APIs to control the hardware, however for new programmers they provide helpful examples for to write a proper statusbar applet. Ask rm_you how much time he spent trying to figure out how Nokia got them to look the way they did. Also, the browser UI, which seemed to be open source for a while then went back to being closed source: What's up with that? I'm willing to bet dollars to donuts that the closed source part has nothing to do with flash support. And random things like the "Web" menu on the lefthand panel and many of the home applets. And what's the business case behind making the image viewer closed source? It depends on open source libs for image loading and processing, but the part that would make a great, simple example of a hildonized app is closed.

I'm not blaming anyone and I'm not some open source zealot. I understand that it's Nokia's code and they have the right to do whatever they want with it. And really this is all somewhat rhetorical, as I think I know the reason the PHBs don't want to have this stuff open source: competitive advantage. If Intel is going to get their MID into the same market as the Internet Tablet, then Nokia wants them to at least have to spend the R&D cash rewriting all the apps they'll need.

Anyways, that's all, I'll shut up now. :)

-John

Re: Maemo is CSS contaminated (not entirely open)
 

I still don't think he was able to figure it out.Agreed with you regarding the annoyance of these being closed, but I suspect the reason isn't competitive advantage. I suspect they don't want to spend the time checking each app(let) to verify that nothing in it comes from an NDA, or is otherwise constrained from release. Not that there is anything like that, and they already know there isn't, but policies probably exist that it must be checked and certified, and that's too much hassle/their legal team is too busy/whatever.

Re: Maemo is CSS contaminated (not entirely open)
 

No, it's sort of a compromise right now. It's the GTK tooltip widget, same one as OMWeather uses. :\

Quim didn't really give us an answer when it came up during thoughtfix's interview (thanks again Quim and Dan!). As far as I could read, it basically came down to, "They're closed, it's not gonna change anytime soon." No real reason was given, though.

Re: Maemo is CSS contaminated (not entirely open)
 

Kinda difficult when there are undocumented hardware components. Where you have documented hardware, gnu purists already succeded in providing a better system (ubuntu, mandriva, and surely most other linux distro nowadays) than the commercial alternative (windows).

Re: Maemo is CSS contaminated (not entirely open)
 

Considering that it sucks power like there's no tomorrow and its performance is mediocre, I'd say they chose the cheapest wireless chipset ;)
(Hey, I don't blame them for that, I also bought some of the cheapest wireless adapters I could find, but they have open drivers and perform much better than the tablet's one)

Re: Maemo is CSS contaminated (not entirely open)
 

Bluetooth driver is not closed, the source is in linux kernel. What is closed is the bluetooth firmware blob but that's pretty normal (see http://www.bluez.org/download.html for various firmwares) .

Audio could be replaced relatively easily, it is possible to access audio from ARM side and linux drivers exist or is possible to adapt them. Then DSP will simply sit unused. See linux port for Palm Tungsten T/T2 (OMAP 1510) for example, I think it uses same audio codec as 770.

wi-fi is a problem, see this mail for introduction
https://garage.maemo.org/pipermail/c...er/000001.html
basically there are two closed parts with open glue (=cx3110x.ko) between. To make it kernel-independent one needs to replace umac.ko module. As I understand it the firmware blob and the umac blob fit together so using some random open wi-fi stack with the firmware may be hard (just guessing).

Re: Maemo is CSS contaminated (not entirely open)
 

Do you have a reference for that? I thought cx3110 was pretty good in that regard.

Re: Maemo is CSS contaminated (not entirely open)
 

Man, that was beautiful. Thanks.

Re: Maemo is CSS contaminated (not entirely open)
 

wgetting a big file uses 75% cpu (according to osso-statusbar-cpu), I suppose (but I may be wrong) that a sizeable part of it is due to the wifi transfer (on another arm system wget takes just 30%).
I also suppose (but again, I might be wrong) that cpu consumption translates to power consumption.
Btw, I don't really care about wget performance, I noticed the problem because it's very taxing to stream video to the tablet, in part due to missing video optimization (even with the great work done with the mplayer port) but in part due to the wifi (lack of) performance/efficiency.

Re: Maemo is CSS contaminated (not entirely open)
 

I retried with the output redirected to /dev/null (writing to the flash takes its toll on the cpu) and "Osso mcspi" takes 41% cpu, while cx3110x takes 12%. So the cx3110 isn't completely to blame (though 12% isn't that low), but overall wifi takes more than 50% cpu.

Re: Maemo is CSS contaminated (not entirely open)
 

I guess I'd be more concerned with power usage figures on the wifi itself? It uses CPU to drive it, of course, but I expect more of the power is directly consumed.

Re: Maemo is CSS contaminated (not entirely open)
 

Network transfers are CPU-expensive. This really says nothing at all about the chipset itself.

Re: Maemo is CSS contaminated (not entirely open)
 

wif-fi chip is connected over SPI bus so in fact the cpu taken by OMAP McSPI/0 belongs to cx3110 too. This is known issue and does affect video streaming. Looks like the driver is currently not optimized for transferring lot of data. See discussion here
http://garage.maemo.org/pipermail/cx...ry/000012.html

Re: Maemo is CSS contaminated (not entirely open)
 

Yes, that would be my speculation. The existence of bluetooth is largely justified by the fact that it consumes only a small fraction of energy that wifi does.

I've also noticed after using wifi on a Palm TX for a while, the device gets very warm (ie. wasted energy). The device does not heat up for non-wifi use (that is, BT use or hours of navigational use). This suggests the waste is in the wifi radios themselves, not necessarily by way of CPU workload.

Re: Maemo is CSS contaminated (not entirely open)
 

A lot of the waste related to wi-fi is in how the chipsets handle (or doesn't handle) power saving mode. On the NIT the chip drops to power saving mode after only a few hundred milliseconds, and as most of the data traffic is in bursts anyway it'll stay in power-saving mode most of the time in practice. If you turn it off, or change the timing (which may be necessary for some troublesome routers) your NIT will get warm too, and battery time dramatically reduced.

The SDIO wi-fi card I used to have for my T3 got hot too, and the T3 could only take about 20 minutes of wi-fi on before the battery almost kneeled.

Re: Maemo is CSS contaminated (not entirely open)
 

OpenMoko demonstrates that these devices can be released w/out closed components. You may be correct in the fact that closure of Nokia's hardware is a hindering factor to releasing an open OS -- but my "complaint" in that regard didn't come with any stipulation that the hardware be simultaneously closed. I would simply consider closed hardware another deficiency to circumvent.
That's indeed favorable if it's true. It would mean less work for Mamona. But it sounds like BT and wifi are closed drivers.
That may be true, but the connection isn't obvious to me. Certainly Adobe would not have been able to hold the NIT back. Which closed source vendor getting a piece of the action would have caused the show stopper?
In the absence of a link, I'm not compelled to accept that (since I already tried searching before I introduced the thread).
Actually I wasn't referring to an alternative that would access shockwave pages, but an alternate (open) technology altogether: DHTML. It's a red herring anyway, since Flash is not an essential component. I don't believe most linux distros come with flash.. it's generally a separate plugin. Certainly the inclusion on flash in the image is not essential to NIT success.
The latter would only be paranoia in the absence of commercial players. Since it's a mixed bag, of course the commercial players will look after their interests - we can count on that. Surrender of the mac address was not expressed as a means of "device registration", so no, I don't have a reason to trust their use of the information. Therefore, I will use the windows tool to download images.

Re: Maemo is CSS contaminated (not entirely open)
 

That's a good point. I think when someone told me it was open source linux, I mislead myself from there.
From the utilitarian standpoint, yes, this is a very good thing. But the deontologist need not applaud; Nokia is not tossing the GNU community a bone as a favor, they are looking to use the GNU community to add value to their product. And there's nothing wrong with that motivation, but call it what it is.
OTOH, if the TV was advertised as open source w/ customizable DVR, and the buyer discovers the DVR API is closed enough that they can't create what they want (a widget to cut out commercials) they might not be as impressed as they were initially.

Re: Maemo is CSS contaminated (not entirely open)
 

I did a quick search, and here's links to a few threads relating to the closed portions of the tablet firmware:

• http://www.internettablettalk.com/fo...ad.php?t=11031 - discussion of proprietary kernel modules on the tablets and the GPL compliance of same
• http://www.internettablettalk.com/fo...t=5613&page=15 - similar discussion with some additional details on which bits of the tablet software are closed
• http://www.internettablettalk.com/fo...ad.php?t=11066 - discussion of possibility that Nokia followed in NVIDIA's tracks on binary blobs
• http://www.internettablettalk.com/fo...t=10997&page=2 - discussion of binary blobs in relation to holding up getting Ubuntu Mobile on the tablets

Re: Maemo is CSS contaminated (not entirely open)
 

OpenMoko uses binary blob AGPS driver that outputs NMEA for gpsd.

OpenMoko still doesn't even have Wifi.

OpenMoko is still in development and may never see a production run.

So I don't think Nokia should be using OpenMoko as an example of how to do things right. Nokia have been bringing IT's to the market since 2005 with very little closed source material. I sure wouldn't mind if they opened up some of the nokia developed code like the connection manager, but at least the device works.


Update: I also found this on the hardware page.

Re: Maemo is CSS contaminated (not entirely open)
 

Even though I mentioned the closed wi-fi driver and other drivers in an earlier posting, I think the _real_ problem, one that actually affects us, is the closedness of important parts of the UI and the provided software.

Re: Maemo is CSS contaminated (not entirely open)
 

I strongly disagree with your suggestion that Nokia may be "abusing" the "gnu community" for own profit. The main point of GNU is that it's FREE (not as in free beer). So no-one can complain that anyone is using it "for their own gain" - that's inherent to the whole GNU thing itself.
From the other side the uptake of GNU/Linux by company's like Oracle, IBM , Nokia etc did boost the credibility of GNU/linux as a whole and provoked exposure and investments from big $$ company's, leading to more investments in GNU/opensource and provoking added value to the GNU community itself, which is what the whole GNU thing is about AFAIK.

The whole "stallman GNU ideology" vs the more pragmatic "let's simply get things done" (aka Linus) is nicely documented in the "revolution OS" documentary by the way.

I agree with your TV example, but for example any TV that simply say's "build on/using Linux" does NOT imply that *everything* is "open source" in that system.

Re: Maemo is CSS contaminated (not entirely open)
 

Hm, what jgombos wrote (included what's in the quote) didn't come over to me at all like you interpret it.. he said ".. and there's nothing wrong with that..", which IMO is quite far from "..your suggestion that Nokia may be 'abusing' the 'gnu community'". Certainly there's nothing else in that quote which supports that.