Paul Krush

You get an easy CAD like 3D engine in Flash and people will build mass customization tools with it. It’s a given. There will be a million micro-factories that will rise up in Asia making every custom product you can think of using these 3D tools to directly communicate to the world wide consumer. I see this growth of mass custom as inevitable and unstoppable.

So I’m glad to see things are moving along on the 3D Flash front. Adobe has released a beta version of Flash 10 with some native 3D support, including hardware rendering. Speed wise it’s getting comparable to a regular windows application.

A company out of Russia called Alternativa Game, Ltd. has released a 3D Flash engine based on the Flash 10 called Alternativa Platform. Neat stuff. It’s private competition to open source Papervision3D

Microsoft has to be cooking up support for 3D in Silverlight, but I have yet to see any news on this. I don’t see how they can get to critical mass with Silverlight, but you can’t take them off the table as a player.

I don’t plan on working with 3D for Flash for a long while, but it’s nice to see it’s evolving into something useful for 3D consumer jewelry design in the browser.

I have said this a 100 times, but I just have to repeat it. Flash is on more machines than Windows and growing. These tools are going to give you the ability to release 3D design tools on any screen. Consumers will be able to design products on their iPhones, smartphones, micro-laptops, whatever, with the same interface as their home computer. Yes, a lot of the tools will be impractical due to screen size, but a least they have the connection/portal to the mass custom tools, databases, and community.

I had this idea for years and realized it’s not on the website. Just as you can write software to render frames of a computer animated movie, you can script out Rhino 3D to render people’s names into a name ring or name pendant for a post card. Say a jewelry store has 5000 people in their database. You could build a system that renders 5000 pictures for 5000 post cards each using the persons first name. You could render all sorts of address information into a post card. I think this would be eye catching as heck. Every year when Mother’s day gets near I flirt with this one.

Just to review each post card would be different. It would have a 3D computer rendering of jewelry designs that contain the persons name.

The scripted name pendants you see can also be rendered in a mass customized way like this. This is more of a highly specialized 2D font creation. You can script in a name and create all sorts of 2D rules on how swoops and hearts are applied. You can add 2D symbols till the cows come how in an automated fashion. After the 2D curve is extruded into the 3D model. You can also add 2 ½D art work into this. 2 ½D art work is neat because it can be represented by a grayscale image.

I had JewelryPostCards.com registered for a while, but I never wanted to jump into the jewelry marketing business, so I  dumped the name. Dell color laser printers are high quality and cheap enough to do stuff like this. You can buy software that helps you automated the address information, bulk rate info, and barcodes.

I think the money is just not in jewelry to do this sort of stuff. My prediction is these ideas will be fleshed out in Asia. They can market and ship directly to the western world. This has been going on with colored stones from Thailand and cell phone accessories out of Hong Kong for a long time now.

To fantasize, can you imagine having full 3D scanning glasses? Where you could pick up the 3D info on objects in your sight? Not just mesh info, not Nurbs, but the automatic generation of parametric models? You could look at an old ring in your hands, and have a model with handles (to change it) pop up in your vision. As you play with the forms and change the model a price could constantly update on what it would cost to build this new ring you are designing. An efficient, low cost, mass customization, manufacturing center somewhere in the world will make it in days and ship it to you. When will this happen? 15-20 years from now??? I have no doubt that it wil happen.

Here’s an interesting site:
http://genometri.com

http://genometri.com/pdf/case_risis.pdf

On the surface it just looks like they are just messing around with parametrics in CAD software. Their intentions are truly mass customization for the consumer. Genometri’s founder Sivam Krish is talking about releasing products in Second Life. That’s fun, but for me I just want more than Second life has to offer. I want the real Metaverse now!

Lately I’m not working on this stuff at all. Things don’t move that fast, so I am not worried. At this point I feel another 2-3 years will bring some of the core tech up to speed. Someone will write a Flash 3D CAD engine, and have it working smoothly. Better 3D worlds will come out. Computers and Internet speeds will get faster. Heck given enough time, servers will be fast enough to do the 3D modeling on the server. You can do that now, but it would be so much nicer to have the client’s machine do that. It’s not about tech; there is just no critical mass in 3D standards for the browser.

Here’s screen shot from their site:

This has all been talked about before. I am just framing the concepts differently.

As first glance most people would say this is oil and water, but I don’t see it that way. Steps can be taken to bring model sharing closer to software and software closer to sharing. With the release of uber-builders and scripting in Matrix Version 6 I think Gemvision is taking a small step in this direction.

It makes for an interesting exercise if you ask these two questions:
1.) How can Internet Model Sharing be more like Jewelry Design Software?
2.) How can Internet Jewelry Design Software be more like Jewelry Design Software?

These can be small steps toward each other, but end I see a product that is a combination of both these services. Yes, I’m calling jewelry design software a service.

Steps to bring jewelry design software closer to model sharing:
1.) Making the final model more parametric. It’s finished, but it has handles all over the place to change it.
2.) Some 3D Art can be standardized, and directly injected into a builder. Such as:
Standard panels for class rings
Standard tops for signet rings, pendants, etc.
Collar sections of rings.
Really any element can be standardized, and then offered as an option.
2.5) You can build a model and the builder can cycle through the standard art available.
3.) Offering the design software in an Internet browser. Yes this is a jump, but standards and tech to make it happen are forming right now.
4.) Put a live human into the jewelry design software and offer this service live. No Joke. Yes, even I am dreading the day jewelry design software has an Indian accent. Cool, but scary! It’s called Artificial Artificial Intelligence.
5.) Offer community and collaboration tools directly inside of the software. A share my screen/model button. This would be a lot easier if the software was web based.

We do some of these steps already manually, but sometimes it can be done with a person that does not know CAD. Like a store employee or even a customer standing at a kiosk.

Step to bring model sharing closer to jewelry design software:
1.) Offering models that have parametrics already built into them. An example would be a model with the Matrix WIP 6 handles already in the model.
2.) Offer models that have the parametrics automated. Like you can play with the controls before you downloaded the model?
3.) Offer models that the standard 3D art can be changed on the fly, and seen in real time with the sharing interface.
4.) Offer manual changes to the models being shared. (We will resize this ring model for $xx.xx)

Reasons why you would want to marry the two concepts:
1.) Profits. This creates a new product category for the wholesale or retail customer. They can claim that they designed their jewelry.
2.) CAD designers can start with a model that is closer to the final design.
3.) This brings more people to the table. Most jewelry sales people do not work with CAD.
4.) It’s the Burger King business model. Have it your way. This gives the customer far more control over the model.
5.) These concepts create more ways to get the designer’s creative juices flowing.

This is all about building intelligence into the design of the model. The model has to be simple to allow this, but it can be done in many different ways.

I use Gemvision terminology here, but these concepts are already in use in other industries.

By the way artificial artificial intelligence is a really cool concept. I have not thought much about it for jewelry design. It’s all about humans interacting in complex ways with each other with complex software.

The convergence is coming!

What do you think?

I discovered a genre of web sites and books about small software venders. The tech boom of the late 90s was all about this knock it out of the park thinking. These guys are talking about practicality and sustainability. I have started with the book Eric Sink on the Business of Software. It’s a collection of blog posts. They guy seems like he knows me, and all my weaknesses! Great work Eric!

I think I found another key to making consumer jewelry design software happen. The key is there needs to be competition. Or should I say if there was competition the market for consumer jewelry design software would be safer to enter. As a typical developer I want to go to a cave and appear a year latter with a full blown working system. I want to hit the ball out of the park, create something that has never been done, and blow everyone away. I want real time 3D tools in the browser, but 2D and 2 1/2D consumer jewelry design are not out yet. There is no consumer windows application. Silly me. It’s hard to be humble, but I’m beginning to think this project is better done in baby steps.

Boot strapping and baby steps….. I thought I was doing this with the crowdsourcing service bureau project, but then it took on a complex life of it’s own. Maybe I need to do a very simple working example system. Like a 2D band designer in Flash(in the browser), and provide links to 3D files(created on the server side). However simple and small this would start a flow of people using the software. With my experience from this project I could then refine my ideas about this world of consumer jewelry design.

Backing up… If you look on RingHunt.com I have done some very complex stuff. I entered a cave and I came out with what I thought were some very cool tools. My plan was that they would be so cool, that people would discover them, and motivate me to build better versions. I got some response, but I got scared with no money flowing in. Next time I need to think things through more, and have experience people review my plans. Which is one reason why I am blogging.

Did you know you can buy home, deck, and garden design software at Target?

Do people play with home design software before going to architect? If so, how has that changed custom home design? Do people rough out the house design, and let the architect detail it? This was one of the directions I was heading when I was working on that 3 stone trellis ring software in the end of 2003. I have not thought about it much lately until several people brought it up here in the forums.

One could write software that would let people rough out custom rings though a series of controls and parameters.

Having consumers rough out the ring first fits some rules I have for consumer design software to work in an independent retail store.

Those rules are; Will the consumer jewelry design software (or kiosks too):
1.) Keep the relationship?
2.) Build the local community?
3.) Be sustainable?

Would free consumer jewelry design software like this hurt an independent retail store? Sure I can think of many ways. One becomes defeatist when you go down this path. Pretty much you can argue that Gemvision cannot make its products better, because in doing so it will smother its current customer base. Every copy of Matrix they sell makes the rest of the copies less valuable though this thinking.

Does a $20 home design package cheapen the work of a professional architect? Of course not, this is mearly a better (maybe) way for the consumer to communicate to the architect. Some people might need to be educated on this.

Why do I keep coming back to consumer jewelry design software? This will be one of the ways jewelry will be designed in the future, and I want to be part of it. Good or bad it’s a compulsion I have.

These are some of the rings I did with a program I wrote 3 1/2 years ago. You enter in a bunch of parameters and it modeled the ring for you. I can do a much much better job at writing this software these days. I was just goofing off with some of these designs. It’s pretty fun using it.

Before writing software, I have to test my prospective user base for the crowdsourcing service bureau. I don’t want to put the cart before the horse like I have done in the past! There are so many great things I can do in software, but I have to make sure people are going to use them. It’s a little cocky of me saying I am going to start a new hobby, so I need to do some crawling before I code in the dark for months.

I have been selling on eBay for eight years. I am afraid I suffer from tunnel vision as I keep thinking eBay is the best method to get the word out. There are other ways to promote my site and find designers interested in using my methods (Duh!). I would like to have a local group that gets together to talk about their designs but for now the Internet is easier to find people.

I am going to offer free rings in the forums for people that want to test out making some very simple rings. I will let people email ether grayscale pictures or 2D Rhino files. I also am going to let people snail mail rings designed on paper. Alone I think automated paper to finished silver ring is pretty darn cool. It’s very simple to do.

At some point I will offer a free ring a day to the best design submitted to the database.

So I will make a submission rules web page and post to these forums:

Art Jewelry Magazine Forums
Orchid Forums on Ganoksin.com
Rhino3D CAD Software Forums

I plan on writing design tools that people can use within a web page to design rings. Now that real time 3D can be done in Flash, I am interested in starting with Flash with just images. I want to get my foot in before real time 3D flash takes off.

I think simple rings are a great starter product. These rings have no holes and a minimum thickness. This makes them very easy to manufacture; as easy as printing.

This type of design is called 2 1/2D because you can only change the flat profile of the design. The design is wrapped around a ring, but not designed in 3D. The design is represented by an image with the grayscale color being the height. So you will hear me say ways to manipulate an image, but really it’s changing the 2 1/2D information.

This is a rough list:
Centering of text on a ring.
Arraying artwork around the ring.
1D & 2D Arraying.
1D scale of the layout – stretch to array.
2D scale.
3D scale, the ring size, or thickness must change.
Pattern builder for the ring.
Booleans for the art work.
Art database.
Ring database.
Ring LxW look up calculator. Like a size 6 4mm layout fits on a size 8 4.5?mm layout.
Softening/blurring tools. Putting a chamfer on hard edges.
A cropping tool.
Import pictures tool.
Rail selection tool.
Height raise/lower tools, or darken and lighten the image.
Negative image tool.
Ring size calculator.
Font builder and font spacing tool (kerning maybe).
Vector curves – like adding a grove.
Basic setting tools for bezels and prongs.

I am thinking about these tools for simple rings, but they can be used for:
Bracelets, Links, solid or hinged.
Napkin rings
Huggie earrings.
Rondelles
Coin/medalian bezels
Flat metal strips for decoration.

Really these tools can be used for any 2 1/2D design work.

I think I see the Holy Grail for products designed by the consumer in an Internet browser forming. Adobe will be releasing ActionScript 3.0 for Flash 9 sometime in 2007. It’s currently in alpha. This release is orders of magnitude faster than previous versions. This release combined with faster computers is going to put us over the top for real time 3D in the browser. These 3D tools will work in almost all computer browsers without having to install them. The tools can just pop up, just like flash ads do at this point.

ActionScript 3.0 is not about 3D, but a powerful virtual terminal that runs in the browser. This lets people write 3D engines that use ActionScript. See Papervison3d.org and Kirupa.com.

Right now there are a lot of tricks you can do in 2D, but real time 3D in a browser that works on almost all desktops without installing anything is very hard. I see a small ad in a page expanding much bigger with a full-blown tool set popping up. In the past I thought Java would be the answer, but this was never able to achieve critical mass. Microsoft wanted to make darn sure that Java in the browser was not a way to replace desktop applications, and did what it could to kill Java. A server based CAD engine could work, but this lets the CPU power of the client go to waste. I have thought about running Rhino on a server outputting models and renders.

Flash is not just for computers; it works or soon will on kiosks, palmtops, handhelds, and large cell phones. The same tools that let consumers design products over the Internet can be used at a local level. Just like a person at Walgreens can print photos from a kiosk, a person can use a kiosk to make a custom gold pendant in a jewelry store. Sure this can be done with desktop software now, but kiosks would be easier to manage using these Internet tools. Also people could use their own devices in the store to run the store’s machines. This would truly be user manufacturing like Frank Piller is talking about.

I’m not sure where I am going to take this. I think it would be wise to let these 3D engines get developed for another 6-12 months or so to see what happens. I need to explore the world of 2D and Flash first. At least I now have more motivation to learn Flash. I am most interested in designs that are flat with 2 1/2D profiles. This limits the options and makes it easier for people to design there own pieces. With 2 1/2D you can use images to represent profile information. This is why I think a coin is perfect starter mass customization product. It’s just 2D printing with one more piece of information, height.

So can something like ActionScript 3.0 be the grease that really gets mass customization rolling? For simple custom jewelry I say yes.