Page 12 of 19 « First 21011 12 1314 Last »
Reply
Thread Tools
lma is offline lma
2010-12-10 , 01:35
Posts: 2,802 | Thanked: 4,490 times | Joined on Nov 2007
#111
Originally Posted by javispedro View Post
I find it funny that you mention RedHat, considering they do practically the same Nokia does. (They have large closed source software comercial offerings
Such as? I'm not entirely up to date, but their typical M.O. in regards to closed components is to acquire and open them. Even the "crown jewel" management stuff that used to be closed (satellite) is open now (spacewalk). It's possible they still have the odd closed product (I don't know) but their core business is open source. They make their money from selling support, but you're free to grab their update SRPMs and build them yourself, or use any of a number of distributions (both community and commercial) that do exactly that.

but it is usually presented as a opensource friendly company because they use and, more importantly, contribute to open source projects. Like Nokia).
Far more than merely friendly. If RedHat went away tomorrow there would be huge gaping holes across the entire F/OSS spectrum. I think it's fair to say that Nokia is nowhere near that important so far (though yeah, KDE would be inconvenienced). Consider also that RH don't have physical products (if all you're selling is bits, traditional thinking says that there's a lot of value in keeping them closed), and the relative sizes of the two companies.

This is not meant to be disparaging to Nokia, far from it. They do contribute a lot, but RH are in a different league altogether.
 

The Following 2 Users Say Thank You to lma For This Useful Post:
pycage's Avatar
pycage is offline pycage
2010-12-10 , 05:55
Posts: 3,404 | Thanked: 4,474 times | Joined on Oct 2005 @ Germany
#112
Originally Posted by Venemo View Post
Not true.
The main problem is that they don't obey their own rules. Download a while/light theme and see the calendar app. It remains black... (there are other examples as well)
Yes, but that is not the policy framework you're talking about.
The policy framework is the bullying entity inside the OS that dictates what your app may or may not do at times, like e.g. may not get CPU cycles while the user is pulling the lock slider.
__________________
Tidings - RSS and Podcast aggregator for Jolla - https://github.com/pycage/tidings
Cargo Dock - file/cloud manager for Jolla - https://github.com/pycage/cargodock
Last edited by pycage; 2010-12-10 at 05:57.
 

The Following User Says Thank You to pycage For This Useful Post:
lardman is offline lardman
2010-12-10 , 09:03
Posts: 2,102 | Thanked: 1,309 times | Joined on Sep 2006
#113
Originally Posted by danramos View Post
So the question, then, is... why isn't Nokia writing open-source applications from the start (or writing new, or getting open-sourced apps and using them) on these devices, so as to avoid the whole problem altogether? Why does Nokia feel the need to obligate users to use closed-source software, even the ones written by Nokia themselves, on a platform being sold on the benefits of being "open-source"?
There are 3 reasons I can see for having some closed source packages:

1. Expediency - if you write and then release some piece of code, it is faster to not opensource it as opensourcing requires the legal dept to get involved to check there's no secrets being released, no inappropriate comments, etc., etc.

2. Secrecy - some part of the code may deal with chips which are important to the company and as such they don't want to broadcast information about how they work (e.g. Retu and Tahvo in the N8x0 series as these were also in use in other phones). I'm not sure this particular reasoning is very logical, as it's quite simple to reverse engineer if you have the time and resources.

Secrecy can also apply to chips or sw for which Nokia have an NDA with the originating company (e.g. the GPS chipset and the PowerVR)

Lastly secrecy can apply to things that Nokia don't think users should fiddle with - e.g. BME to avoid us blowing batteries up, CAL to avoid us overriding lockcodes (I'm guessing here)

3. UI differentiation - the thin veneer applied to OS applications serves to make the Nokia apps and overall UI look different. There's one reason why you'd want to keep this closed, and that is to avoid people copying it (whether you think they would want to is a moot point here). This would apply equally to competitors (assuming they might want to copy it) and to Chinese knock-off sellers.


Just my opinion
 

The Following 4 Users Say Thank You to lardman For This Useful Post:
javispedro's Avatar
javispedro is offline javispedro
2010-12-10 , 10:35
Posts: 2,351 | Thanked: 5,243 times | Joined on Jan 2009 @ Barcelona
#114
Originally Posted by lma View Post
This is not meant to be disparaging to Nokia, far from it. They do contribute a lot, but RH are in a different league altogether.
No arguments to any point; there is a large scale difference.
Last edited by javispedro; 2010-12-10 at 10:37.
 

The Following User Says Thank You to javispedro For This Useful Post:
jstokes is offline jstokes
2010-12-10 , 11:02
Posts: 235 | Thanked: 339 times | Joined on Nov 2010
#115
Originally Posted by lardman View Post
CAL to avoid us overriding lockcodes (I'm guessing here)
Nokia released the libcal headers in the maemo-binaries repository for Fremantle, which is all that's needed to fiddle with the lock code
 
Texrat's Avatar
Texrat is offline Texrat
2010-12-10 , 17:12
Posts: 11,700 | Thanked: 10,045 times | Joined on Jun 2006 @ North Texas, USA
#116
Originally Posted by fpp View Post
Anyone willing to bet N9/Harmattan will be any different ?
Kill it. Kill it now.
__________________
Nokia Developer Champion
Different <> Wrong | Listen - Judgment = Progress | People + Trust = Success
My personal site: http://texrat.net
 

The Following User Says Thank You to Texrat For This Useful Post:
lma is offline lma
2010-12-11 , 11:17
Posts: 2,802 | Thanked: 4,490 times | Joined on Nov 2007
#117
Originally Posted by jstokes View Post
Nokia released the libcal headers in the maemo-binaries repository for Fremantle, which is all that's needed to fiddle with the lock code
These headers consist entirely of /usr/include/cal.h, which just declares some function prototypes. That's completely useless if you want to, for example, retrieve the factory-installed WLAN MAC address of an N810.
 
jstokes is offline jstokes
2010-12-11 , 12:26
Posts: 235 | Thanked: 339 times | Joined on Nov 2010
#118
Originally Posted by lma View Post
These headers consist entirely of /usr/include/cal.h, which just declares some function prototypes. That's completely useless if you want to, for example, retrieve the factory-installed WLAN MAC address of an N810.
So, erm, how do you think the MAC address is retrieved in the first place?

http://git.slonopotamus.org/?p=opend...3f;hb=HEAD#l70 built using the header from Fremantle's libcal-dev works with the libcal already present on an N800.
Last edited by jstokes; 2010-12-11 at 12:29.
 

The Following User Says Thank You to jstokes For This Useful Post:
lma is offline lma
2010-12-11 , 12:35
Posts: 2,802 | Thanked: 4,490 times | Joined on Nov 2007
#119
Originally Posted by jstokes View Post
http://git.slonopotamus.org/?p=opend...3f;hb=HEAD#l70 built using the header from Fremantle's libcal-dev works with the libcal already present on an N800.
That's an independent implementation (many thanks to the author btw), and doesn't #include Nokia's cal.h anywhere. Also, there's no libcal present on an N800, the functionality was hardwired into dsme before Fremantle.
 

The Following User Says Thank You to lma For This Useful Post:
jstokes is offline jstokes
2010-12-11 , 12:38
Posts: 235 | Thanked: 339 times | Joined on Nov 2010
#120
Originally Posted by lma View Post
That's an independent implementation (many thanks to the author btw), and doesn't #include Nokia's cal.h anywhere.
True. But it's not hard to make it use Nokia's libcal.

Code:
#include <assert.h>
#include <unistd.h>
#include <stdint.h>
#include <cal.h>

int main (void)
{
	struct cal *cal_s;

	cal_init (&cal_s);
	const size_t mac_len = 6;
	char mac[mac_len];
	uint32_t *data;
	unsigned long len;
	if (cal_read_block(cal_s, "wlan-mac", (void **)&data, &len, 0) == 0) {
		assert(len == (mac_len + 1) * sizeof(uint32_t));
		size_t i;
		for (i = 0; i < mac_len; ++i) {
			mac[i] = (char)data[i + 1];
		}
		printf(" [%02x:%02x:%02x:%02x:%02x:%02x] ",
			mac[0], mac[1], mac[2], mac[3], mac[4], mac[5]);
	}
        free (data);
        cal_finish (cal_s);

	return 0;
}
Originally Posted by lma View Post
Also, there's no libcal present on an N800, the functionality was hardwired into dsme before Fremantle.
Nokia-N800-43-7:~# ls -l /usr/lib/libcal*
-rw-r--r-- 1 root root 7760 Feb 11 2008 /usr/lib/libcal.so.0.3.0
Nokia-N800-43-7:~# ls -l /mnt/initfs/usr/lib/libcal.so*
lrwxrwxrwx 1 root root 15 Oct 21 2008 /mnt/initfs/usr/lib/libcal.so -> libcal.so.0.3.0
-rwxr-xr-x 1 root root 6496 Oct 21 2008 /mnt/initfs/usr/lib/libcal.so.0.3.0
Last edited by jstokes; 2010-12-11 at 20:12.
 

The Following 2 Users Say Thank You to jstokes For This Useful Post:
Reply
Page 12 of 19 « First 21011 12 1314 Last »

« Previous Thread | Next Thread »
Thread Tools

 
Forum Jump


All times are GMT. The time now is 00:22.

Contact Us - Home - Archive - Top