Sept 12, 03. As of today VRStoremaker is just a movie and a start
on a tool
in vrhotwires, but I've decided to post a page about the project so I don't
to write things over and over to people... None of this is chipped in stone.
A note on malls
This is an application that makes
it easy for QTVR authors to make virtual stores. Stores where you're in
a panorama and you can click and drag on the items in front of you and drag
them into a shopping basket. Like the example above.
The stores should play back using qt if the viewer
has it, or java.
After shooting panos and stills the photographer
just drags them onto the vrstoremaker tool and after filling in info like
the price of each item, presses 'publish'.
The store is served by the VRStoremaker team and everything is handled
automatically. ?(Haven't figured all of this out yet...)
a) demo a store and gui in Washington DC. oct 1 iQTVRA gathering. Distribute
free CD's with the ability to create stores on them. Put a demo version
up on the web around this time with the storemaking capability enabled.
ie: get people making these stores now!
b) release final VRStoreMaker 1 .0 app feb 1st 2004?
APP IS $29 for author. Target author is a photographer who can make good
vr panos and objects.
Fees store owner gets charged when they sell an item:
3% Credit card company.
3% web transaction/ split payment?
1.5% VRStoreMaker team...
This is total speculation with regards to the credit card and transaction
That side of it is not worked out!
There is no risk to the store owner. The author/photographer risks his
the VRStoremaker team does in making and improving the app .
But the Store owner just sits back
and watches if sales increase.
If they already have a web store, this can
be just a second way to sell products over the web... a high bandwidth/multimedia
button on their page.
To use the app you make panos in your store.
You then shoot each item against a greenscreen.
In some cases you make object movies on an item.
You then open these in VRStoreMaker and go though the tabs
Most of the tabs are greyed out at first, but as you step through them
next one lights up.
HOTSPOTS: You choose a current store item's hotspot.
SubTabs on Hotspot Tab:
ITEM IMAGE: Import the item image. Link hotspot to its dragging
PRICE:Set the price of this store item
INFO: Set and create? a translucent flash text info pane on this item.
OBJECT: set an object movie view for this item.
Submit Item. Loop to tab 1 for next item.
When you're finished setting all your items up the next tab lights up.
STYLE: choose between a variety of shopping carts and info panels for your
PUBLISH: publish store. In the first version this will just be a movie
Someday, pressing publish should put your store up on the web in a way
people can shop in it right away...
OK I've only done a bit of work on the app but the first version
is sitting in VRHotwires, so that's not really the hard part.
Making a windows/mac app that just does these and plays back well
in java will
be a challenge, but
The key is in designing a business model that pleases everybody.
So anybody got any input on this, because you're the one's who would
The author/photographer would get 2.5% of the stores sales, but in some
cases, like an antique store,
they'd have to do a lot of updating. An ongoing service.
The option of charging a first time set-up
fee to the store would be up to the photographer.
I mean the rough idea is that the photographer should risk 2 or 3 days
with digital equipment. But if it's shot on film? Should the storekeeper
vrstoremaker would get 1.5% of sales,
and the credit card companies/transactions services might take up to 6%
or who knows? Currencies have to be very international because the qtvr
project is very international.
so when you walk into the store you're saying 'there's a 10% overhead'
to the store owner.
You have to be aware that in bulk transactions here in canada, say $100,000
an ecommerce sale might only cost 2.5%... that's the transaction and the
credit card cost.
So at 10% we're talking
a fair cost to the store...
(realistically they're probably paying at least 5% for ecommerce, if they're
doing it now...)
but hopefully to the store owner that isn't on the web, they'd be comparing
their brick and mortar place.
Let's say they're selling a scarf. They probably have to pay 50% of the
price they sell it for to buy it.
If they sell it in the brick and mortar store they have to pay rent,
and labour, to sell the scarf,
in many cases taking up almost all of the other 50%... usually at least
on the web it cost just 10%.
(over time I think this should be more like 5-6% somehow...
but now it's an exclusive distinguishing service... )
The app can't be free. I mean I'd give it away for free at this iqtvra
thing, and other shows,
but on the web it would cost something, just to make sure the surfer
isn't just clicking
for fun. Maybe $29 is a good price. I wouldn' t be totally opposed
this token fee going to
Actually where it should go is an important place: Mall and Street building.
The payment and transaction and credit card stuff is in limbo.
As of now we're trying WorldPay.com
I've been thinking that part of the VRStoreMaker 1.5% should be to
serve the stores. When you press 'publish'
it goes over the web to a VRStoreMaker server. I think the stores
should have a 15 meg size limit.
Also a percentage of the 1.5% should go to
Mall Making. Maybe .5%.
And .5% goes back into the tech. That's my secret motivation here. To
fund a general improvement in VR technologies...
Really, the integrative task, the linking of the stores is a key.
It's interesting that the same store can be in several malls.
example, I live in a little town in Holland.
The store I shoot is really the only interesting store in town, from an
international perspective...a fine pottery store.
It appears in the Mall for dutch stores, and also in the pottery store
shopping mall, and in the 'Gerry's favorite stores' mall...
Theoretically, if there was a search engine included (there has to be)
a street of stores that sold that item could be generated! (For example
by a server with Maya on it)
the iqtvra might want to play a part in mall and street building
The goal would try to link together interesting VR's in interesting
ways, providing interface to tons of panos.
Also crowd draws like amusment parks and games would have to be sponsored.
I can't help thinking of how many stores there are at disneyland.
Now that I write it down it suits the iQTVRA in that it makes them in
charge of meeting place, but
the best way to do it would be with MAYA or something, and that doesn't
really fit that well.
I mean the interface to 50 antique stores all over the world, could be
a nice street to walk through with
50 doors. On entering the door, you URL link to the store.
Jumping all over the world is key,
but shopping for just what you want, without having to use the keyboard.
(until the credit card number)
Perhaps it's better to just hire someone and make it happen, but the
malls really should come out of the
group of authors somehow. All authors should have input.
Then there's the real streets. If VRStoreMaking really took off then
another business model becomes to
shoot walkthroughs of a city's streets and charge the store owners a
fee to link on to your street...
That should be part of it somehow...
There are quite a few towns already shot in VR. Maybe this has already
I read this back and I see I haven't really justified the percentage to
the VRStoreMaker team completely...
They would be embarking on a development path that would take years to
Things like real-world inventory feedback would come first. In such a
VR store items that are not in stock wouldn't
show up on the shelf. This kind of imaging might have to be done as a
spriteorama in qt. I don't know how well it would work
in java. Exchanging movie data with a databass comes into.
then work would be done on sound...music as you shop would be in 1.0.
But sound label descriptions for some things... ? audio conversations
with a storekeeper.?
As time passes realtime video confrencing with the person behind the
counter could be added to some stores.
Also they would have time to work on cutting overhead. The cost of a
store could be whittled down to 6% if
the VRStoreMaker team took over more and more
The goal is to provide a shopping experience that is very much like
real world shopping.
No text has to be read, although info on each product is available in
The interface is no interface. Just VR. Full Screen and high res.
I walk into the sporting goods store, I look at the baseball bats. I
drag one into my shopping cart, I say 'purchase', and the store sends
it to me.
I walk back out into the mall and hang around.
Eventually I can meet people and talk with them...Making VR a social
experience is one of the technologies we can already do and want a budget
Why should you as a photographer, work on this project on spec?
VRStoreMaker's development strategy is to keep everything low key and down
to earth. A few guys put sweat equity into an idea and if it works they
If not, well the losses are minimal...
Once there are say 10 stores made, a shopping mall, or street can be created
and the doors on that should url link to the stores.
That's the first milestone. A small street or mall in qtvr.
We'll call it the 'Hyperstreet' and it should have stores from all
over the world
To do- scripting.
-Use dynamic 'create sprite' call to create a second shirt if asked.
- implement ability to put items back. (and decrement total)
-handle zoomed pano. (zoom sprites)
-make is so while dragging an item you don't pan.
If We Build it they will come.
A QTVR Shopping Mall described:
The first hyperstreet in QTVR is just
two nodes. Perhaps a multinode or two
single nodes, it is opulently 3d modelled, with raytracing. The kind
of image that
might take the computer(s) a night to render. Marble floors always look
nice when modelled.
Each node has 4 or
5 storefronts, surrounding it. Perhaps the mall is Shaped
like a figure 8.
Any kind of distinguishing features
like a waterfall or something. might be nice.
The stores all have
signs and windows to look into, but you can't see much,
and some windows are the magic kind that show different things from the
outside than the in. (depending on how it was shot, some might have a stock
Transitions too, when the
door is clicked might vary or be standardized. I think
it would be cool if they were photographic somehow. So you click on a door
and you zoom in on the actual store... (lower bandwidth versions might have
And then you're in
the store. In England or France or Germany or America or Canada or wherever.
it like this. Any day now, someone builds the mall in Maya. The stores are
left like empty stores in a mall look. It gets posted and improved.
Over time people
make stores. They take pictures of the storefront and send them to the 3d
modeller. The modeller integrates the imagery and store-signs into the 3d
model. A new mall is put up with a working store in it.
because it's a novelty. The hyperstreet is a sculpture in light. An art piece
peace. A happening. All over the world brought to my living room.
In the first mall, payment might have to be dealt with by the individual stores.
are probably not impossible but I haven't solved them yet.
Some fee to the server, and the mall builder might be asked for, but since
it's a free-for-all of people trying to get it set up in any way they can,
there's no way to
require that (I suppose a store could be 'evicted')
I'm thinking of three stores that are really part of the mall. The first would
be a funny 'travel' store that actually links to qtvr's from all around the
world. Since there are so many cool qtvr's already up on the web, this store
would be a links page in VR.
would be a movie theatre. Where linear qt's could be played as miams.
would be an arcade. Where you could play video games.
These three places
could leverage existing content and draw people in.
Then there's the stores
and whether anyone would actually buy anything.
I don't know.