How many submissions will there be?

[Dan U.]

I was just curious if anyone has a guess at the number of submissions
expected. Being that there's teams all over the world competing, I
would guess quite a few, maybe in the thousands.

I think it'd be neat if there could be a submission counter put up
somewhere like they have on the netflix prize page:
http://www.netflixprize.com/leaderboard

[Haitham Kaddoura]

Great idea :)
but I don't agree with you regarding ("all over the world"), you have to read the terms carefully :)

Regards,
HK

[Dan U.]

yes, maybe I should have said all over the world, except a few places.

On Jan 4, 11:46 pm, "Haitham Kaddoura" <hmkaddo...@gmail.com> wrote:
> Great idea :)
> but I don't agree with you regarding ("all over the world"), you have to
> read the terms carefully :)
>
> Regards,
> HK
>

[Peli]

I think we can get a good estimate by counting the participants in the
various Android groups and forums. I guess, any serious participant
will run into some problems with the (early preview) SDK at some point
and would ask some question here or in one of the other discussion
groups:

Number of members:
Android Beginners. 2072
Android Developers: 6803
Android Internals: 966
Android Challenge: 1308
Android Discuss: 855

Then there are independent Android sites, the biggest one being
anddev.org: 585

I have no idea about the number in foreign forums (e.g. a Chinese
forum or other countries). At least, German speaking ones are
practially non-existent, with all developers seeking help in the
English language forums (where most help can be expected). For Chinese
forums, this may be different, but they will suffer from the
restriction that the submission should support the English language.
(if you need to look for help in a Chinese forum because your English
is not good enough, what are the chances of writing a good English
GUI?) If someone has numbers, this could be helpful.

I think, one can safely set an approximate upper limit of about 7,000
- 10,000 eligible Android developers worldwide, but not many more.

The number of open-source projects for android are:
hosted at Google code: 106
( http://code.google.com/hosting/search?q=android )
hosted at sourceforge: 14
( http://sourceforge.net/search/?words=android )
Of these I'd say by skimming through quickly that currently a half is
only a stub project, some others are projects by Google employees who
are not eligible to participate.

Many of the discussion group participants may be interested in the
competition, but may not have the time / ability to implement a useful
challenge entry or take the time to set up a working APK including a
readme file. A third to a forth of the developers interested seem to
be beginners who have little or no prior Java experience, but would
like to learn it.

Then of course there are groups of developers, some of them formed in
the forums publicly, others with friends / colleagues, their number
being between individual developer and open source.

Some business entities will try to recruit people here in the forums,
but others will try to implement things by themselves. Nevertheless,
they will need help from one of the forums, so they are included in
the numbers given above. The lack of visibility of the business
entities in the forums lets me conclude that most serious competitors
either are somehow affiliated to the OHA, or they have business models
where they can make more money by investing their time into customer
projects instead of into a competition with uncertain outcome. Some
others may simply port their existing Java applications to Anroid with
little effort.

Given these numbers, I'd guess there will be:
* around 700 individual developer submissions worldwide, of which
about 150 may be regarded as "not completely trivial".
* maybe 100 indivudual group submissions of which maybe 30 are
serious.
* around 40 open source submissions worldwide (this is an
astonishingly low number! But where are the open source projects?)
* around 60 business entities submissions worldwide, of which maybe 30
deal specifically with the Android platform.

Making a total of around 250 serious entries.

Out of these, 50 entries can receive an award. So, for any serious
participant of the challenge, the chances of winning are actually not
impossibly low! (the chances are around 1:5). (paradoxically, this
very statement may already increase the number of participants again
and lower the chances somewhat...)


Assuming proportional splitting, there will be around:
* 30 individual developers
* 10 groups of individuals and open source projects
* 10 business entities
receiving an award in the first round.

Since the composition of the teams can not change after the first
round, the winners of the second round will be unproportionally high
represented by developer groups or open source projects and business
entities (who can invest more man-hours to polish up the application),
that could take up a large fraction of the 20 final awards, leaving
little room for individual developers winning the second round.

Let me know if anybody comes to different conclusions...

Peli
PS: Anybody interested in joining a growing open-source project as
developer may have a look here: http://code.google.com/p/openintents/
;-)

[Dan U.]

Wow, thanks for the detailed response. I see what you mean about the
final awards though. I'm a single-person team, so it might be tough if
I get that far. On the other hand, I do have some time I can devote to
polishing.

[Bhagaban Behera]

Great analysis..........
thank you........
Shows how serious you are........

[ian]

Thank you for the estimation. And I came to a similar number by a
different formula. I read an article saying that there were about 500
pieces of shareware/freeware available after a new mobile platform was
released in 5 month (sorry, I forgot the name of the platform). So
there could be 350 serious entries for Android since it allows only
3.5 months time preparing. Considering learning curve at the begining
stage, the number could be dropped to 250-300. That's my guess.

I don't expect many creative ideas will pop up at this contest. If
there would, it should have already existed in today's smart phones of
all kinds. What I am seeing here is, Google would love to see many
helpful apps are ready before they launch their real phones. The
Android is not significantly better than others (perhaps worse in some
aspects except it is free), wonderful apps is the ticket for google to
enter into the banquet of wireless market.

On Jan 5, 5:26 am, Peli <peli0...@googlemail.com> wrote:

> (http://code.google.com/hosting/search?q=android)
> hosted at sourceforge: 14
> (http://sourceforge.net/search/?words=android)

[Peli]

Hi,

> Thank you for the estimation. And I came to a similar number by a
> different formula.

That's always reassuring, though you may have been biased by my first
analysis ;-)

> I don't expect many creative ideas will pop up at this contest.

Never underestimate the creativity of those who usually don't get a
chance to show it.
All three, location based services, accelerometer, and always-on
networking, are three new key elements that are mostly absent so far,
but can in combination lead to completely new applications that today
may seem superfluous, but tomorrow you will not want to miss them.

We will add the following features to our (yes: *FREE*) shopping list:
* shopping list will show up automatically when you are in one of your
favorite shops (location based service)
* scroll through long lists by tilting the phone, remove marked items
by shaking the phone (accelerometer)
* share your shopping list instantly with family or friends (XMPP
services)

If you have any of these combined in today's smart-phones, show me
where.

Now, this was just 3 out of 7 (= 2^3-1) possibilities to combine the
above features, so expect a phone that you have never seen before...

Here is the free shopping list, by the way (including source code):
http://code.google.com/p/openintents/

Experienced developers and good designers are still welcome to join
our open source team.

Peli

[Kornelius Tuggerson]

Why don't we try and list all the serious contenders in this thread?
So far we have:

Free shopping list:
http://code.google.com/p/openintents/

and Geocast - The incredible location enabled social networking/file
sharing/viral marketing software :)
http://code.google.com/p/android-geocast/

[Dan U.]

Are these really what would be considered serious contenders? If so,
it gives a good benchmark for what to shoot for.

On Jan 20, 5:35 pm, Kornelius Tuggerson <victor.seme...@gmail.com>
wrote:

> Why don't we try and list all the serious contenders in this thread?
> So far we have:
>
> Free shopping list:http://code.google.com/p/openintents/
>
> and Geocast - The incredible location enabled social networking/file

> sharing/viral marketing software :)http://code.google.com/p/android-geocast/

[Kornelius Tuggerson]

Well I definitely consider my project the front runner, and its only
about 30% done right now. There is still plenty of time for
improvement.

[Dan U.]

Is your project one of the two posted? I looked at the links and
couldn't find your name.

On Jan 20, 7:20 pm, Kornelius Tuggerson <victor.seme...@gmail.com>

[Kornelius Tuggerson]

Yes, http://code.google.com/p/android-geocast/ is my project

[Peli]

> Are these really what would be considered serious contenders? If so,
> it gives a good benchmark for what to shoot for.

I'm part of the other project mentioned above, and I can just speak
for our project, but I have the impression that most of us have not
professionally worked on mobile apps before, and we can merely devote
our evenings and weekends for programming.

There will definitely be companies contributing who can afford paying
GUI designers and 3D animators for applications that will take your
breath away, so don't blame us if you don't win anything, even if your
application is far better than ours.. :-) You should definitely shoot
as high as you can...

Peli
http://code.google.com/p/openintents/

[Dan U.]

I also don't have experience writing mobile apps, so I really have no
idea how good or bad I will do, plus I'm the only person on my team.
I'll be crossing my fingers.

[whitemice]

>>Since the composition of the teams can not change after the first round, the winners of the second round will be....<<

I'm sorry, are you saying that in order to enter "Android Developer
Challenge II" you have to have already submitted to "Android Developer
Challenge I"? I don't see this is the FAQ + legitimate question.

[Peli]

> I'm sorry, are you saying that in order to enter "Android Developer
> Challenge II" you have to have already submitted to "Android Developer
> Challenge I"? I don't see this is the FAQ + legitimate question.

I've been talking about the two rounds within Challenge I:

There will be Challenge I ($ 5m)
- round 1 (for everyone)
- round 2 (for those 50 who won round 1)

and Challenge II ($ 5m)
- exact form has not been specified yet.

see terms&conditions, section 5 and 8a:
http://code.google.com/android/adc-submit/tandc.html

Peli

[Kornelius Tuggerson]

I have a some j2me and windows compact experience and I know for a
fact that as long as your app handles being put in the background well
and is responsive and good looking its got a decent chance even if its
only marginally useful. A lot of the submissions will have poorly
implemented threading that won't play well with being minimized, but
not gc'd and then maximized again. They'll probably test that kind of
stuff first to weed out the amateurs. I can see there being a 200-300
entries that get that stuff right, but not many more.

[ajd]

So amateurs (like me) have no chance in this contest?
I missed that in the judging criteria.

On Jan 21, 8:38 pm, Kornelius Tuggerson <victor.seme...@gmail.com>
wrote:

> > Peli- Hide quoted text -
>
> - Show quoted text -

[Dan U.]

I wouldn't say you have no chance. While I don't have any background
in mobile development, it doesn't seem too terribly difficult to
handle things like your application getting put in the background. I
think its more of a matter of how much testing you do.

[Kornelius Tuggerson]

ajd don't let my previous post discourage you. I did not mean amateurs
like "People new to the android platform", in that sense everyone is
an amateur. I meant it as "People unaware of the unique requirements
and challenges that mobile device programming poses".

New mobile programmers often don't realize just how vital the "going
in background-> coming back from background" part is, how vital it is
to preserve some minimal ui state info in between sessions, how
crucial it is to keep long running operations out of the ui thread
etc. If you have no previous mobile experience one of these things is
bound to get you. If you want to vastly improve your chance of winning
read http://www.amazon.ca/Wireless-Java-Developing-Jonathan-Knudsen/dp/1590590775
it does a great job of covering the basics that are true for j2me and
android. I am not trying to discourage anyone from participating in
the contest, this platforms openness to competition is one of the
things that'll make it a great success, so keep at it.