About "Better Model Exists" And Appeals

46 views
Skip to first unread message

SketchUp Guide Allison

unread,
Dec 11, 2009, 3:41:32 PM12/11/09
to Map Your World - 3D Modeling for Google Earth & Maps
Hi everybody,

We've had some confusion recently with people appealing "Better Model
Exists" reviews, particularly with Building Maker models. (Note: If
you haven't seen it yet, in Building Maker you can avoid modeling a
building that's already been modeled by using the Earth View in the
location picker. It will show you which buildings have already been
modeled.) With Building Maker, two models of the same building will
tend to look very much alike; so people have been attempting to
differentiate their model through increasingly small geometric
improvements. This leads to a kind of model "arms race" --

Modeler A: (creates a model of a building)
Modeler B: (models the same building)
Reviewer 1: (accepts A's model randomly because the two models are
indistinguishable)
Modeler B appeals: Mine's better because I modeled the air
conditioning box on the roof!
Reviewer 2: okay. (accepts B's model, A gets the "better model
exists" judgment)
Modeler A: Oh, yeah? (adds some detail)
Reviewer 3: (accepts B's model randomly, because they're still
indistinguishable at a glance)
Modeler A appeals: My model is now better because I modeled the
awning over the entrance!
Reviewer 4: okay. (accepts A's model, B gets the "better model
exists" judgment)
Modeler B: Argh! (gives up)
(time passes, and something triggers a re-review)
Reviewer 5: Hmm, these models look awfully similar. (accepts B's
model randomly)
Modeler B: Yay!
Modeler A: Argh!

If two models are being compared, and it is not immediately obvious
that one is very much better than the other, the result of a
comparison between the two is pretty much random. The reviewer does
not have access to the appeals history; she doesn't have any idea that
there's been contention. Duplicate check gets triggered when either
of the two modelers edits the model, and may also get triggered due to
events not related to modeler edits (other models appearing in the
same area, for instance).

What this means for you, the modeler, is that the appeals process for
"better model exists" may not have the effect you want unless you have
built a model that's so much better than the other candidates that
nobody would ever make the mistake in the first place. An example
would be replacing a Building Maker model with a phototextured,
geometrically accurate SketchUp model. Using SketchUp doesn't
guarantee that you won't ever have problems with "better model
exists"; especially with popular buildings, we'll often get many
models of a building and choosing between them can be difficult. But
it's a good way to separate yourself from a pack of mostly-equivalent
auto-generated and Building Maker models.

So: before appealing "Better Model Exists", we suggest considering
whether your model would have a clear advantage if viewed from, say,
about a kilometer away in Google Earth. If not, you can still appeal
the judgment, but be aware that the response you get from that
reviewer isn't necessarily the final word on the model.

Happy modeling,

Allison

Peter / G3D

unread,
Dec 11, 2009, 4:34:47 PM12/11/09
to 3d...@googlegroups.com
Hello

For the first time I send an email with a problem that I have emerged this
afternoon.

In my current models of layer 0, had next to the description metatags which
included an image of the building in addition to the standard description.
This afternoon have broken images on the balloons and also now comes into
view url to retrieve the image, leaving the poor presence balloons.


This morning we registered for the contest Google Model Your Town
Competition

Do you have anything to do with what happened.?

How do I change the description of a model layer 0, in a fast, in order to
fix my balloons?

Or is it just temporary failure of implementation?

Taking advantage of the mail, I threw the last question:

Serve the models created by me prior to 1 December?
are valid for the competition?


Peter
G3D

http://sketchup.google.com/3dwarehouse/details?mid=57bf18a288d17510859807d4c
639f399


Hola

Es la primera vez que os envío un email con un problema que me ha surgido
esta tarde.

En mis modelos actuales de la capa 0, tenia metatags junto a la descripción
que incluía una imagen del edificio además de la descripción típica. Hoy por
la tarde han dejado de funcionar las imágenes en los globos y además ahora
sale a la vista la url para recuperar la imagen, quedando los globos con
mala presencia.


Esta mañana nos hemos inscrito para el concurso de modelar tu ciudad,

¿tiene algo que ver esto con lo sucedido.?

¿ Cómo puedo volver a cambiar la descripción de un modelos de la capa 0, de
una forma rápida, para poder arreglar mis globos?

¿ O simplemente es un fallo temporal de ejecución?

Aprovechando el mail, te lanzo la última pregunta:

Sirven los modelos creados por mi anteriores al 1 de diciembre?
Entran en concurso?


Peter
G3d

http://sketchup.google.com/3dwarehouse/details?mid=57bf18a288d17510859807d4c
639f399











J.Stevenson

unread,
Dec 11, 2009, 7:16:07 PM12/11/09
to Map Your World - 3D Modeling for Google Earth & Maps
Hi Allison,

Excellent article! And it has encouraged me to "not give up" on a
recent "better model exists" appeal which didn't work out!

I recently lost my encumbancy as the author of the Leaning Tower of
Pisa. The "better model" is certainly more complicated then mine -
but I challenge its accuracy! Following the lead of your own Effel
Tower example, I feel that a superior texture over a 100% accurate
geometry (dimensions especially) should trump un-necessary (and one
could argue) frivolous additions of details to a model. Not every
body has a 3D-gaming super PC as a computer afterall, right?

The replacement model has attempted to model each individual archway
and floor separation. If you examine a photo of the real structure
(link below) we're talking 24 arches per floor (x6 floor) - just shy
of 150 arches! The replacement model has about 1/2 as many - so even
though it's detailed it's not accurate. And the "top hat" portion at
the top is too short (by about 1/2).

http://blogs.nlb.gov.sg/ask/wp-content/uploads/2008/01/pisa.jpg

So, here we have an arms race between a simple, yet accurate model
versus a complex inaccurate one. Isn't there an actual GE guideline
against "too complicated"? What should I do now? Thanks alot for
putting up with my rant!

-Jon



On Dec 11, 12:41 pm, SketchUp Guide Allison <allisonfloyd

SketchUp Guide Allison

unread,
Dec 21, 2009, 11:01:08 AM12/21/09
to Map Your World - 3D Modeling for Google Earth & Maps
Hi Jon,

"Too Complex" simply means that something in our processing pipeline
fails on the model. It's not a judgment in the sense of a human
looking at the model and making a decision about it; it's just
broken. If you disagree with a "better model exists" judgment, you
can use the "Tell Us Why We're Wrong" link to appeal the review.
(Incidentally, if you're commenting in here and would like one of us
to take a look, please include links to the models' 3D Warehouse
pages.)

-- Allison

SketchUp Guide Allison

unread,
Dec 21, 2009, 11:06:05 AM12/21/09
to Map Your World - 3D Modeling for Google Earth & Maps
Hi Peter,

Your questions with respect to the balloons are answered here:
http://groups.google.com/group/3dwh/browse_thread/thread/4699c46b4475c21d
Models submitted prior to last Monday should have gotten into the most
recent layer release (with a few exceptions, mostly involving models
that resulted in complicated duplicate review). As mentioned here:
http://sketchup.google.com/competitions/modelyourtown/index.html
the model submission deadline for the Model Your Town competition is
on March 1, so you're in good shape there.

-- Allison

On Dec 11, 2:34 pm, "Peter / G3D" <250...@gmail.com> wrote:
> Hello
>
> For the first time I send an email with a problem that I have emerged this
> afternoon.
>
> In my current models of layer 0, had next to the description metatags which
> included an image of the building in addition to the standard description.
> This afternoon have broken images on the balloons and also now comes into
> view url to retrieve the image, leaving the poor presence balloons.
>
> This morning we registered for the contest Google Model Your Town
> Competition
>
> Do you have anything to do with what happened.?
>
> How do I change the description of a model layer 0, in a fast, in order to
> fix my balloons?
>
> Or is it just temporary failure of implementation?
>
> Taking advantage of the mail, I threw the last question:
>
> Serve the models created by me prior to 1 December?
> are valid for the competition?
>
> Peter
> G3D
>

> http://sketchup.google.com/3dwarehouse/details?mid=57bf18a288d1751085...


> 639f399
>
> Hola
>
> Es la primera vez que os envío un email con un problema que me ha surgido
> esta tarde.
>
> En mis modelos actuales de la capa 0, tenia metatags junto a la descripción
> que incluía una imagen del edificio además de la descripción típica. Hoy por
> la tarde han dejado de funcionar las imágenes en los globos y además ahora
> sale a  la vista la url para recuperar la imagen, quedando los globos con
> mala presencia.
>
> Esta mañana nos hemos inscrito para el concurso de modelar tu ciudad,
>
> ¿tiene algo que ver esto con lo sucedido.?
>
> ¿ Cómo puedo volver a cambiar la descripción de un modelos de la capa 0, de
> una forma rápida, para poder arreglar mis globos?
>
> ¿ O simplemente es un fallo temporal de ejecución?
>
> Aprovechando el mail, te lanzo la última pregunta:
>
> Sirven los modelos creados por mi anteriores al 1 de diciembre?
> Entran en concurso?
>
> Peter
> G3d
>

> http://sketchup.google.com/3dwarehouse/details?mid=57bf18a288d1751085...
> 639f399

N. W. Perry

unread,
Jan 14, 2010, 11:42:50 PM1/14/10
to Map Your World - 3D Modeling for Google Earth & Maps
I've encountered this problem too, but under slightly different
circumstances. In this case, my model was the live one, and I went in
to make some minor edits. It went up for re-review, and some kind of
confusion occurred...it was rejected once or twice for obviously non-
applicable reasons. I appealed the judgement and was told that there
had been some chaos due to duplicate models, and it seemed yet to be
sorted out.

While this was going on, my live model was replaced in the New Year's
data update with another user's, and the revision I made was rejected
under the "better model exists" rationale. Because my model is more
complete and accurate in a number of ways (in all humility), I
appealed and was told (almost instantly) that edits must be
"substantive" to displace an existing model, so I was still rejected.

Here's my model:
http://sketchup.google.com/3dwarehouse/details?mid=ffa1e1d84cd61d043ef2ef433d070898

and the live one:
http://sketchup.google.com/3dwarehouse/details?mid=6e334434d73c8d74f852447597ea1ed9

I've read the argument that the difference between models must be
visible from a distance, and perhaps that's not so in this case. Also,
when I initially modeled this location there was no existing model
there. Really, it's just a frustration to me that I had a live model,
made a couple of small tweaks to make it more accurate and better fit
with the surroundings, but the result was that it was rejected and
replaced with one that achieves neither of these goals. I do believe
it was due to a glitch in the reviewing process, so is there any
recourse for me?

Reply all
Reply to author
Forward
0 new messages