Android & iOS tools

On 27/10/14 02:08, edward.ming.lee@gmail.com wrote:
>
>> Putting aside the fact that my impression is that Android phones and iCrap have both had Bluetooth for just about ever, this is the sort of thing that I'd only do for a customer if it was understood -- in writing, in the contract -- that I'm doing exploration, and that I get paid by hour of effort, not by line item of success.
>
> Bluetooth has been there for audio for a long time, but not for socket API.  For Android, it started with 4.3.  So, not many people have worked on it and i am not even sure how usable it is.  Part of the reason i am taking this job is also my desire to find out if it will really work.

Curiosity is admirable, but not if money, reputation
and lawyers are involved.

To try determine whether this is a person you want to
have as a client in the long-term, could you spin it as...

... thank you for trusting my technical expertise
and judgement ...

... the biggest risk to your business plan is that
technology X is new and unproven ...

... to minimise your expenditure, I suggest we de-risk
technology X, and then make a reasoned go/nogo decision ...

... cheaply and speedily achieved by giving me a contract
for 1 week to investigate. If OK, then it is money
well-spent on a good foundation for the remaining work.
If not OK then you haven't lost much and can move onto your
next idea/plan ...

On 10/26/2014 7:08 PM, edward.ming.lee@gmail.com wrote:
>
>> Putting aside the fact that my impression is that Android phones and iCrap
>> have both had Bluetooth for just about ever, this is the sort of thing
>> that I'd only do for a customer if it was understood -- in writing, in the
>> contract -- that I'm doing exploration, and that I get paid by hour of
>> effort, not by line item of success.
>
> Bluetooth has been there for audio for a long time, but not for socket API.
> For Android, it started with 4.3.  So, not many people have worked on it and
> i am not even sure how usable it is.  Part of the reason i am taking this
> job is also my desire to find out if it will really work.

So, your assumptions are that the job exceeds 2 "weeks" of effort (however
you define that).  And, the upper bound must be "considerably more than you
are willing to INVEST in *yourself* (by undertaking a similar task to learn
about the functionality in question in a SAFER environment).

If you are only willing to learn on someone else's dime, then ask yourself
what that learning experience is WORTH to you (in currency units) and then
imagine what it may end up COSTING you if you decide to learn on *this*
client's dime.

Personally, this just has the smell of A Really Bad Experience for all parties
involved (possibly even irreparably damaging your reputation with this client
and any other clients whose ears he may have!)

Buy a demo kit from <someone> and tinker with it in your off hours.  *NEXT*
time, you'll be better qualified to prepare a more specific estimate of the
time/effort required with far less legal/professional "exposure".

On 10/27/2014 2:01 AM, Tom Gardner wrote:
> On 27/10/14 02:08, edward.ming.lee@gmail.com wrote:
>>
>>> Putting aside the fact that my impression is that Android phones and iCrap
>>> have both had Bluetooth for just about ever, this is the sort of thing that
>>> I'd only do for a customer if it was understood -- in writing, in the
>>> contract -- that I'm doing exploration, and that I get paid by hour of
>>> effort, not by line item of success.
>>
>> Bluetooth has been there for audio for a long time, but not for socket API.
>> For Android, it started with 4.3.  So, not many people have worked on it and
>> i am not even sure how usable it is.  Part of the reason i am taking this job
>> is also my desire to find out if it will really work.
>
> Curiosity is admirable, but not if money, reputation
> and lawyers are involved.

+1

The hardest thing for "contractors" (ditto "potential employees") to do is
walk away from a sale.  Especially if it looks like you've got a LOCK on it!
(the same thing applies to designs!)

Several years ago, I "no bid" a contract that would EASILY have turned
into a long term "consulting gig" (i.e., many years of guaranteed income).
Mainly, because I knew the customer was being unrealistic in his expectations,
didn't understand what he wanted (*SO* many red flags, there!) and would,
inevitably, have been put in the hot seat by board members who knew even LESS
about what he wanted.

"Shit rolls downhill"

No amount of money/security is worth that sort of grief!

[In the time that has passed, the project NEVER got finished and that VP was
given his walking papers -- despite (or, perhaps, BECAUSE of) a boatload of
money being thrown at it!  Amusingly, I am in exactly the same situation
with another potential job at the present time.  Same temptations.  But,
far less inclination to "nibble"!  :>  ]

> > Bluetooth has been there for audio for a long time, but not for socket API. For Android, it started with 4.3.  So, not many people have worked on it and i am not even sure how usable it is.  Part of the reason i am taking this job is also my desire to find out if it will really work.
> 
> So, your assumptions are that the job exceeds 2 "weeks" of effort (however you define that).  And, the upper bound must be "considerably more than you are willing to INVEST in *yourself* (by undertaking a similar task to learn about the functionality in question in a SAFER environment).

Considering BlueTooth socket API by itself could take two weeks, and multiple  graphical activities/views with screen switchings.  I would say at least two months, if not longer.

> If you are only willing to learn on someone else's dime, then ask yourself what that learning experience is WORTH to you (in currency units) and then imagine what it may end up COSTING you if you decide to learn on *this* client's dime.

I am not a total newbie in Android, so I am charging them fractional cost of development.  We are bounded to have some new things to learn in any job.  This should not be a negative factor, but the two weeks time limit is.  I think they have some live presentation of the software that have not been written.  That's their problem, not mine.

> Personally, this just has the smell of A Really Bad Experience for all parties involved (possibly even irreparably damaging your reputation with this client and any other clients whose ears he may have!)

Any reasonable client would understand that you can't build an app in two weeks.  They understand it too, but they are just hoping for miracle.

> Buy a demo kit from <someone> and tinker with it in your off hours.  *NEXT*
> time, you'll be better qualified to prepare a more specific estimate of the
> time/effort required with far less legal/professional "exposure".

I think i have the hardware needed.  But Android Lollipop (5.0) now wants 64bits Linux KVM, and Eclipse is to be depreciated soon.  It's hard to keep up with the Googles.

On 27/10/14 20:13, edward.ming.lee@gmail.com wrote:
> Any reasonable client would understand that you can't build an app in two weeks.  They understand it too, but they are just hoping for miracle.

In which case they are unreasonable clients. Such clients sometimes
use lawyers to "prove" (to themselves or others) that "I was on the
right track, but the contractor fouled up and let me/us down".

I'd politely decline the work. If you think they are actually
potentially reasonable clients, you could indicate your answer
might be different in other circumstances.

edward.ming.lee@gmail.com writes:
> I keep telling them i cannot do iOS in two weeks, but they are
> tempting me with big checks.  Perhaps enough to buy iMac, iPhone and
> iPad all together. 

As others explained, you really do have to turn it down.  And that check
doesn't sound anywhere near big enough for 2 weeks of that level of
intense development on a consultant basis.  It might be more interesting
if iCar was added to the list, if there was such a thing.

>A customer wants both apps in 2 weeks.  Gee, i can barely do Android in
two=
> weeks.  iOS is completely new to me.  I can run ADK on Linux.  Are there
i=
>OS tools available on Linux?  How about Android Java to iOS obj C
translato=
>r?  Any good tools i can pickup in less than 2 weeks?
>
>I keep telling them i cannot do iOS in two weeks, but they are tempting me
=
>with big checks.  Perhaps enough to buy iMac, iPhone and iPad all
together.=
>  But i don't want to learn iAnything.  I would like to stick to Linux if
p=
>ossible.
>
>
Although I agree with the consensus that you would be nuts to try to do
this, I thought I should mention Cordova (http://cordova.apache.org/) which
allows you to program in JavaScript and then generate the code for
iOS/Android/Windows etc. I have only heard of this- I cannot vouch for it.	
  
					
---------------------------------------		
Posted through http://www.EmbeddedRelated.com

On 29/10/2014 18:57, antedeluvian wrote:
>> A customer wants both apps in 2 weeks.  Gee, i can barely do Android in
> two=
>> weeks.  iOS is completely new to me.  I can run ADK on Linux.  Are there
> i=
>> OS tools available on Linux?  How about Android Java to iOS obj C
> translato=
>> r?  Any good tools i can pickup in less than 2 weeks?
>>
>> I keep telling them i cannot do iOS in two weeks, but they are tempting me
> =
>> with big checks.  Perhaps enough to buy iMac, iPhone and iPad all
> together.=
>>   But i don't want to learn iAnything.  I would like to stick to Linux if
> p=
>> ossible.
>>
>>
> Although I agree with the consensus that you would be nuts to try to do
> this, I thought I should mention Cordova (http://cordova.apache.org/) which
> allows you to program in JavaScript and then generate the code for
> iOS/Android/Windows etc. I have only heard of this- I cannot vouch for it.	
>
> 					
> ---------------------------------------		
> Posted through http://www.EmbeddedRelated.com
>

I managed to write two Cordova-based apps for Android from scratch, 
knowing nothing of Java, Android etc. Basically, the application is a 
web page and a chunk of Javascript does the work.

If you want to use the Bluetooth LE stack, start with the Evothings 
environment.  This is basically a framework for building your app 
interactively from a PC with a prebuilt runtime environment (the 
Evothing client) running on the Android tablet.  Once your get that 
working you can turn your Evothings project into a Cordova one and build 
an installable package.  Then your app can be built for Android and IOS 
without any changes (well that's the theory!).  Get the Evothings stuff 
first - source code for lots of example Bluetooth apps.

good luck!

Mike