Android Mythbusters (Matt Porter)

47 views
Skip to first unread message

PeeDee

unread,
Nov 5, 2009, 12:38:00 PM11/5/09
to android...@googlegroups.com

Al Sutton

unread,
Nov 5, 2009, 2:31:58 PM11/5/09
to android...@googlegroups.com
I read it, to me it sounds like its' written by someone who doesn't
understand the complexities of mobile 'phone development.

Flexibility costs. It costs in terms of storage, processing
requirements, and battery life. I've yet to see a full Linux/GNU
installation on a phone sized device that can run for 24 hours without
a charge, and I doubt that it would be possible, but hey, I'm always
willing to be proven wrong if you can provide me with an example....

(and before anyone says N900, even Nokia don't try to call it a
smartphone, they call it a Mobile Computer).

Al.
--

* Looking for Android Apps? - Try http://andappstore.com/ *

======
Funky Android Limited is registered in England & Wales with the
company number 6741909.

The views expressed in this email are those of the author and not
necessarily those of Funky Android Limited, it's associates, or it's
subsidiaries.

On 5 Nov 2009, at 17:38, PeeDee wrote:

> If you haven't seen this:
>
> http://laforge.gnumonks.org/weblog/2009/11/04/#20091104-android_mythbusters
>
> --
>
> You received this message because you are subscribed to the Google
> Groups "Android Discuss" group.
> To post to this group, send email to android...@googlegroups.com.
> To unsubscribe from this group, send email to android-discu...@googlegroups.com
> .
> For more options, visit this group at http://groups.google.com/group/android-discuss?hl=en
> .
>
>

Disconnect

unread,
Nov 5, 2009, 3:19:46 PM11/5/09
to android...@googlegroups.com
FWIW my n800 can go just about that long. Certainly longer than my g1.
(Wifi on, cell radio nonexistant.) Turn off the cell radio and turning
on wifi, and I've got desktops that'll survive unplugged longer than
the g1..

Andreas Kostyrka

unread,
Nov 5, 2009, 3:36:33 PM11/5/09
to android...@googlegroups.com
On Thu, 2009-11-05 at 19:31 +0000, Al Sutton wrote:
> I read it, to me it sounds like its' written by someone who doesn't
> understand the complexities of mobile 'phone development.

As it happens this was mentioned by people who have done mobile
development in the past ;)

>
> Flexibility costs. It costs in terms of storage, processing
> requirements, and battery life. I've yet to see a full Linux/GNU
> installation on a phone sized device that can run for 24 hours without
> a charge, and I doubt that it would be possible, but hey, I'm always
> willing to be proven wrong if you can provide me with an example....

Well, it's news to me that my G1 can go for 24 hours without a charge.
Guess if I go to GSM, deactivate anything in the background, and don't
touch the phone it might manage that.

But then, all my cars have an USB charger. I've got a mobile USB charger
that works from batteries.

Hmm, interesting, for my former smartphone I had to buy the car-charger
when I needed it, because between uses it got just lost. With the G1 I
continuously keep in mind where the next power-up opportunity is.

>
> (and before anyone says N900, even Nokia don't try to call it a
> smartphone, they call it a Mobile Computer).

Well, actually the hardware seems quite similar to the N97 Symbian
device.

It's probably being called a "mobile computer", because it came from the
department that has been doing the internet tablet stuff. Actually
Symbian devices are the smartphone offering of Nokia.

And so what if it's a "mobile computer", if that subsumes the
functionality of a mobile phone, which it seems for the N900.

The size of the N900 seems quite like mobile phone.

Just for a comparison:

G1 => 117x55.7x17.1mm == 111.42 cm^3 volume and 158g weight
N900 => 110.9x59.8x18mm == 119.35 cm^3 volume and 181g weight

the battery life time that is officially claimed (up to 5 hours talk
with UMTS and 9 hours with GSM) sounds similar to my G1 too, and it's
quite possible that they are utter irrelevant projections as for the G1
too.

So what if Nokia claims it's a mobile computer. Actually my favorite
price comparison site lists it under UMTS mobile.

It has the size, the battery life, the weight and usability of a mobile.

Andreas

signature.asc

Gergely Kis

unread,
Nov 5, 2009, 5:12:43 PM11/5/09
to android...@googlegroups.com
Hi,

Just for info: Harald Welte was the first system architect of
OpenMoko, so he has built mobile phones in the past.

At OpenMoko they did try to build a GNU/Linux based mobile phone
software stack...we all know the results. I know that I am being
unfair here, there were many reasons why OpenMoko failed, but one of
the reasons was that they failed to create a good software stack.

Personally, I found in Android what I missed in GNU/Linux approaches:
a clean architecture.

Android was not designed to be Linux or GNU/Linux. It is a completely
different system, which just happens to be using the Linux kernel.

Also, if I am not mistaken, some of the Android architects came from
Palm, where they already built embedded devices for some time.

Regarding the lack of System V IPC: There is a clear comment in
bionic, why it was not included. (It creates leaks in the kernel)
Instead, you have Binder.
There is a similarly clear comment about why e.g. pthread_cancel() is
not implemented and why C++ exceptions are not supported.

So Android did not throw out anything. It was simply built from the
ground up by choosing already available components (Linux, sqlite ...
etc.) and implementing others to fulfill the requirements. This also
meant that they had to cut corners where necessary (e.g. time and
resource constraints).

Of course Android is far from perfect (both technically and as an
open-source project), it is enough to just look into the
android-platform list to see that. There are many ugly places in the
code...etc. But its system architecture is clean and I like it way
better to work with than the GNU/Linux alternative for mobile phones.

Best Regards,
Gergely

PS: I think the linked presentation in the blog entry provides a fair
overview: http://tree.celinuxforum.org/CelfPubWiki/ELCEurope2009Presentations?action=AttachFile&do=view&target=Mythbusters_Android.pdf

Mike Wolfson

unread,
Nov 5, 2009, 6:33:10 PM11/5/09
to Android Discuss
FWIW, I recently took my SIM out of my G1, and am amazed how long the
battery is lasting, without the radio being in the loop.

I won't be able to report more, as the new phone isn't worth keeping,
so I am going back to the G1, and the SIM will be reinstalled.

Bottom line, eliminate the radio, and even the G1 battery will last a
long time.
> >>http://laforge.gnumonks.org/weblog/2009/11/04/#20091104-android_mythb...
>
> >> --
>
> >> You received this message because you are subscribed to the Google
> >> Groups "Android Discuss" group.
> >> To post to this group, send email to android...@googlegroups.com.
> >> To unsubscribe from this group, send email to android-discu...@googlegroups.com
> >> .
> >> For more options, visit this group athttp://groups.google.com/group/android-discuss?hl=en

Serdar

unread,
Nov 11, 2009, 6:17:32 AM11/11/09
to Android Discuss
> > Flexibility costs. It costs in terms of storage, processing
> > requirements, and battery life. I've yet to see a full Linux/GNU
> > installation on a phone sized device that can run for 24 hours without
> > a charge, and I doubt that it would be possible, but hey, I'm always
> > willing to be proven wrong if you can provide me with an example..

That's the problem.
This is the reason why it makes hard to port android, because it's not
flexible.
You should design software for hardware not hardware for software.

Look at the android-on-freerunner Project.
We are stuck on cupcake because we want a clean stable android working
before switch to other versions, but we are still not finished.

Greetings Serdar
Reply all
Reply to author
Forward
0 new messages