ACUS

["eberhard speer jr." <se...@ducis.net>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi,

while I have a number of 'topics' in mind for presentations/articles
on DDR, user-agents and DeviceMap (more on that below) I'm sorry to
say that even with Travel Assistance, for me it's financially
impossible. If it were held here in Athens, I might manage :-)
(jobs/business...help ! etc)

But if I can provide some one with 'material', no problem.

By the end of next week, I hope to release an updated version of the
API running via my site, based on the current DeviceMap code, together
with some 'articles' on how to implement such an API and some
'back-ground' material on device detection, user-agents and the
difficulties of developing meaningful tests (fully automated here now).

And that -- tests -- bring me to the release question.
I agree with Bertrand there is room for improvement in DeviceMap's
accuracy but based the test-results I obtain now, I would say
DeviceMap is pretty much as 'accurate' as the 'old' OpenDDR code.

Improvements now depend on the 'tokens' in BuilderDataSource.xml.
These are now chosen with regex in mind.

I've made a rather in-depth analysis of the test results and
identified the specific areas where things go off the rails with
regex-minded 'tokens' in the current resources ("Test Results -
Update", 02 Jul 2013)

So, basically the question is if and how to maintain this
'alternative' data-set.

Also, while the Java version of the OpenDDR code is probably still ok,
the .Net C# version is effectively dead. In that version
TwoStepDeviceBuilder only worked thanks to a bug on .Net's
implementation of OrderedDictionary
(https://connect.microsoft.com/VisualStudio/feedback/details/384059/ordereddictionary-enumerator-doesnt-throw-invalidoperationexception-after-collection-was-changed)
That bug has been fixed in versions after 3.5.

esjr
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.8 (MingW32)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/

iQEcBAEBAgAGBQJS2QaNAAoJEOxywXcFLKYcF/cH/0eXP+hUUFJEoqXEUGvtfeni
0Wxtuab30DjM4mtOeZf3JvdGC6uoQpTZUfkMPkDfAPLENsAKW165XXm4dsIYgtJG
yLdKV9vShdYT7iiexoHEIZfI442HhwXTcEVuYk0tOv1pd+aDmWRhAgjsGR2EXKYY
rc+9lOeAKF8W1ERGHg/loNST3MML/xBFUIdB7gMRBwQAW4tYYRPIgsgIXI/xqYnT
rAlN6aR6NpktBjzdBwJm0EWWFY83TVvGm7BGHiP2mR6MTNnjKcF+t5aaZ7ZdvLLP
VwuWSPwRznD5Ey/dpqm0y4svk68ZvmwxpWYj5+JMtdXyxvdWq7jCkEKkUTXTsM4=
=HIdH
-----END PGP SIGNATURE-----

Re: ACUS

[Werner Keil <we...@gmail.com>]

All,

I did not submit anyhing, partly due to attending Nordicana Festival in
London where I relied only on tablets and had not much Internet usage
otherwise.
I also thought, let's try despite our slightly smaller team (~4-5 active
committers, none of us seem to work Full Time on it though) to graduate
from Incubation first, then chances for either Apache or other events could
increase a bit[?]

I haven't heard from the 2 German based events so far (JavaLand is highly
unlikely, the other one won't be till Summer) but offered it on two
occasions, maybe I'd propose it for GeeCon in Poland, too, then we have to
get things going towards Graduation...

Cheers,
Werner

 Werner Keil | JCP Executive Committee Member | Eclipse UOMo Lead, Babel
Language Champion | Java Godfather

Twitter @wernerkeil | @JSR354 | #EclipseUOMo | #Java_Social | #DevOps
Skype werner.keil | Google+ gplus.to/wernerkeil

* Social Media Week Copenhagen: 17 Feb 2014, Copenhagen, Denmark. Werner
Keil, Eclipse JCP EC Member, JSR 354 EG Member, Agorava Cofounder will
present "Bitcoin, Payment Instrument or Object of Speculation?",
"Enterprise 2.0 with Agorava"

* Social Media Week Hamburg: 18 Feb 2014, Hamburg, Germany. Werner Keil,
Eclipse JCP EC Member, JSR 354 EG Member, Agorava Cofounder will
present "Bitcoin,
Payment Instrument or Object of Speculation?", "Enterprise 2.0 with Agorava"


On Fri, Jan 31, 2014 at 5:09 PM, Werner Keil <we...@gmail.com> wrote:

> Guys,
>
> Last day to think of something for ApacheCon. Do you think we should this
> year?
>
> The PhoneGap "army" as usual throws in a lot of money, design and
> manpower, see AeroGear artwork:
> http://burrsutter.blogspot.fr/2014/01/ios-apns-android-gcm-push-with-apache.html
>
> I also heard there is interest in OpenOffice, so I could "recycle" my
> OOoCon Talk about healthcare usage of OO there. That one is actually quite
> in a good shape, but for DM unless some of you provide suggestions and
> their input I'd be limited to what was done about OpenDDR in previous
> years. I have abstracts also the one for Developer Week, but I can't say,
> if it'll convince ApacheCon hosts or not[?]
>
> Cheers,
> Werner
>
>
> On Fri, Jan 17, 2014 at 12:43 PM, Bertrand Delacretaz <
> bdelacretaz@apache.org> wrote:
>
>> On Fri, Jan 17, 2014 at 12:35 PM, Werner Keil <we...@gmail.com>
>> wrote:
>> > ...Is there an easy way to demonstrate the C# part like the Java (mostly
>> > OpenDDR derived) portion?...
>>
>> I have no idea, never used it and not planning to do so.
>> -Bertrand
>>
>
>

Re: ACUS

[Werner Keil <we...@gmail.com>]

Guys,

Last day to think of something for ApacheCon. Do you think we should this
year?

The PhoneGap "army" as usual throws in a lot of money, design and manpower,
see AeroGear artwork:
http://burrsutter.blogspot.fr/2014/01/ios-apns-android-gcm-push-with-apache.html

I also heard there is interest in OpenOffice, so I could "recycle" my
OOoCon Talk about healthcare usage of OO there. That one is actually quite
in a good shape, but for DM unless some of you provide suggestions and
their input I'd be limited to what was done about OpenDDR in previous
years. I have abstracts also the one for Developer Week, but I can't say,
if it'll convince ApacheCon hosts or not[?]

Cheers,
Werner


On Fri, Jan 17, 2014 at 12:43 PM, Bertrand Delacretaz <
bdelacretaz@apache.org> wrote:

> On Fri, Jan 17, 2014 at 12:35 PM, Werner Keil <we...@gmail.com>
> wrote:
> > ...Is there an easy way to demonstrate the C# part like the Java (mostly
> > OpenDDR derived) portion?...
>
> I have no idea, never used it and not planning to do so.
> -Bertrand
>

Re: ACUS

[Bertrand Delacretaz <bd...@apache.org>]

On Fri, Jan 17, 2014 at 12:35 PM, Werner Keil <we...@gmail.com> wrote:
> ...Is there an easy way to demonstrate the C# part like the Java (mostly
> OpenDDR derived) portion?...

I have no idea, never used it and not planning to do so.
-Bertrand

Re: ACUS

[Werner Keil <we...@gmail.com>]

Bertrand/all,

Is there an easy way to demonstrate the C# part like the Java (mostly
OpenDDR derived) portion?
I don't know if they accept this talk, but at least the Developer Week in
Germany is a "mixed" event with Windows Phone or C#/VB/.NET sessions, too.
Unless we archive the C# part for now, it would be especially interesting
for them to see more than one language.

Werner


On Fri, Jan 17, 2014 at 11:49 AM, Bertrand Delacretaz <
bdelacretaz@apache.org> wrote:

> Hi,
>
> On Fri, Jan 17, 2014 at 11:31 AM, eberhard speer jr. <se...@ducis.net>
> wrote:
> >... I've made a rather in-depth analysis of the test results and
> > identified the specific areas where things go off the rails with
> > regex-minded 'tokens' in the current resources ("Test Results -
> > Update", 02 Jul 2013)
> >
> > So, basically the question is if and how to maintain this
> > 'alternative' data-set....
>
> Could that be loaded in addition to what's under
> data/device-data/src/main/resources/devicedata?
>
> If Werner keeps contributing data in there, having two distinct data
> sets that are merged might be the easiest.
>
> > ...while the Java version of the OpenDDR code is probably still ok,
> > the .Net C# version is effectively dead...
>
> I don't need C# myself, if someone wants to fix it fine but I'm also
> ok with dropping it for now. Maybe just move it under contrib.
>
> -Bertrand
>

Re: ACUS

[Bertrand Delacretaz <bd...@apache.org>]

Hi,

On Fri, Jan 17, 2014 at 11:31 AM, eberhard speer jr. <se...@ducis.net> wrote:
>... I've made a rather in-depth analysis of the test results and
> identified the specific areas where things go off the rails with
> regex-minded 'tokens' in the current resources ("Test Results -
> Update", 02 Jul 2013)
>
> So, basically the question is if and how to maintain this
> 'alternative' data-set....

Could that be loaded in addition to what's under
data/device-data/src/main/resources/devicedata?

If Werner keeps contributing data in there, having two distinct data
sets that are merged might be the easiest.

> ...while the Java version of the OpenDDR code is probably still ok,
> the .Net C# version is effectively dead...

I don't need C# myself, if someone wants to fix it fine but I'm also
ok with dropping it for now. Maybe just move it under contrib.

-Bertrand