White Box Robotics Pre-order

How many people run Windows on robots? Some, but certainly not the majority. This is why I think the guy isn't going to make it, he hasn't thought it out.

What you run on the robot is important, not in what people are familiar with, but what works in general. If you look at the devices we rely on, wireless routes, access points, ethernet routers and bridges, many of these are running an embedded UNIX variant or RTOS, and many of them run Linux.

It is prefectly reasonable to make GUI apps that run on your PC run on Windows, but when making an embedded device, Windows is a poor choice. I can tell you a lot of stories about Windows on Cellphones.

I would treat the robot like any other embedded device, have a web server serve the application interface (web serivces) and let the user use java if they like. That way, they never know or care what the robot runs.

Reply to
mlw
Loading thread data ...

Just the opposite. For PC-based robotics most use Windows, because they simply connect up their existing Windows laptop. I've watched this trend for a long time, and you can guestimate it by looking at the various robotics gallery Web sites.

Robotics is not yet mainstream, and isn't likely to be for some time, science fiction stories notwithstanding. Linux may be preferable for a black box device, and some form of it may eventually become the de facto standard for robotics, but TODAY Windows allows a larger market share simply because it has the much larger share of desktops.

Again, these are business decisions, and plausible for a startup looking for momentum. You also forget that his design does not preclude, in any way, also supporting Linux. For all you know he may do both, but come out with the Windows version first. Or, someone else could do a Linux version, and create a cottage industry.

-- Gordon

Reply to
Gordon McComb

I think you're jealous that someone else is commercializing what you think is "your" idea. The fact is, this fellow announced his plans a year ago, and has been working with VIA for even longer.

Look, some people may want to build from scratch. Some people may prefer to buy a kit of parts. Some people may want to purchase something ready made, and use it as-is, or attach things to it. All these markets exist, and can co-exist.

You can't control the markets, but you can develop product that addresses those markets. That's called savvy business.

-- Gordon

Reply to
Gordon McComb

And good business that has good marketing also paints their products all nice and shiny, knowing that the appearance of a product is what sells it to most people.

Put an old car in the classifieds with horrible paint job. It won't sell. Put a coat of paint on it, and everyone says "wow, nice paint".

It doesn't matter what is on the inside. If the exterior looks good, people buy. Judge a book by its cover. Most people don't read them anyway.

Rich

Reply to
aiiadict

Not jealous at all, a PC based robot is hardly my idea. I simply don't think the whitebox design is targeted toward any real or potential market.

One of the problems with the "marketing" argument is the definition of the "market." Sure, there may be a thousand different segments to which you could market, but if each of the segments have a low population, you will never attain a market size big enough to make your project profitable. In order to make a profitable product, it has to have a wide enough appeal that, at least some, benefits of scale start paying off.

Everyone who has a PC has different needs for thier PC, but that doesn't mean that there are infinite markets. Unless you can specifically identify a vertical marketplace, you have to create your product with the broadest appeal.

Unless you define and develop the robot sufficiently to perform a well defined and marketable task, roomba for instance, your robot will be relegated to tinkerers, people who are interested in learning or building.

Now, it is not as though aesthetics are not important, they really are, but not at the expense of the product. Make no mistake, the whitebox robot is an experimental platform, nothing more, nothing less, and as such does not seem all that well suited toward that goal.

Reply to
mlw

I wouldn't dismiss the looks of a robot being important. Are looks important to operation? Nope. But it is important to how people react to it. For me, I like to build my robots to operate around and interact with humans. I enjoy when I write some code and get my robot to self navigate from one room to the other, but really, what's the end result? It got itself from one room to the next. Big whoop. Now, if I can get the robot to enter the room, recognize someone, say hello to them by name, tell them a joke cause they know that person likes jokes, ask them if they would like something to drink and then go get it, then that would be a big deal.

Do you need looks for the above? Nope. But a robot that looks like something people imagine rather than a bunch of parts from my shop really seems to intrigue non-robot people 100x more. When my non-robot friends see my robots they think they are cool and amazed I was able to build it, but they don't connect with the robot itself. I whip out my old omnibot and they want to play with it, cause it looks like a "robot". To me, it seems that when non-robot people see my home made robots, they are thinking, that's great and I'm amazed, but I can't wait to see it when its done.

If you just want a truck chassis, there are several robot platforms and systems already available. I think there may just be a market for this Whitebox bot, not just cause its looks, but because it gives the consumer a robot that looks like how many people think a robot looks like and is pretty much ready to go. I'm not talking about you and me, but people new to robotics look and say, that's what robot I want, not an erector set or some small boebot type robot. Oh it runs Windows? I know how to use Windows. They don't need any tools. They don't need to build anything. Just plug it in and start playing with it, as they get more and more into it, they start seeing how it works, start seeing what limitations it has and start thinking of how they could modify it.

I know people that when they started into robotics, have spent more than this robot's cost getting parts, tools, and spending tons of hours researching and building just to get a robot that looks like an upside down trash can that can move around. Many of them ended up losing interest and/or just didn't have the time to devote. I think this robot has a place in the market and gives people a ready to run robot that looks like what they imagine a robot to look like. The cost might seem high to many of us, but that's because we have the ability, the drive, the space, tools, the time, etc to build our own. I think the draw to robotics for most isn't the idea that they get to spend hours and hours researching a bunch of fields they know nothing about and then cut and drill material and fashion a robot, just so they can get to the point of making a robot than can be made to do things. I think there is a large market of people that just want to get a robot and then make it do things. This does that and on top of it gives them a design that they and people they show it to will relate with.

-C

PS, when looking at the cost of this robot, its easy to look and say man, $1200 is way too much. But is it? What does the cost of all the parts cost? The motors, the wheels, batteries/charger, VIA board/CPU, memory, hard drive, camera, metal, etc. Now, if you built your own, you would need to have at least those parts. So the price of this robot is the delta. That delta might be easily worth the cost to many to get a fully built and well designed robot that took 0 amount of their time to get to that point and they get to focus on the part they want, making it do whatever they want.

Reply to
Hoss

Yes, the White Box platforms seem to be targeted as an experimental robotic platform.

If I had had one before I started my book, I would have used it as the main example.

Once this company gets going and lowers the prices a bit I think that these will go well in the educational market and in the experimenters market.

-- D. Jay Newman

formatting link

Reply to
D. Jay Newman

I notice that some are using Roomba as a robot base.

[...]

So you agree with me? Your robot is limited to those computer science students interested mainly in software development?

Anyone who is not a student, people like yourself, already have the expertise, have nothing to learn from it, and can build their own version if they like.

A product with the broadest appeal will have an entry point for each interest group.

[...]
Reply to
JGCASEY

Most robots are only an electronic version of Jacques de Vaucanson's mechanical duck, so they may as well look good as they aren't going to do much else :)

John

Reply to
JGCASEY

That duck could even poop. I wonder what market there is for a pooping robot? That would give the poop-cleaning robot something to do!

-- Gordon

Reply to
Gordon McComb

Heh-heh -- you obviously didn't read "Mason and Dixon", or you'd be more impressed with Vaucanson's Duck.

Reply to
the Artist Formerly Known as K

Who woulda thunk there would be so much business in PC modding. Yet there is, and there are some folks who spend an enormous amount of money on this "hobby." There are entire Web sites that sell nothing but mod parts, like cases with built-in Lava Lamps, or fluorescent panels that pulse with the sound of music ("The 'bots are alive, with the sound of music...").

The White Box robots are really highly-modded PCs. The biggest difference is that they have wheels. There has never been a rule that a PC can't have wheels. With a wireless keyboard and mouse (easy) and an RF link to a monitor (fairly easy) your PC is now your robot. Or maybe your robot is now your PC. Either way, will it matter?

Tom Burick wants to blur the line between the PC and the robot. He's not the first one to do this. But he has taken the extra step of paying a few hundred thousand bucks for injection molding dies to make it look more appealing. No one buys a PC mod that's ugly.

Jay wants to build a PC-based robot with tracks so it will climb up stairs. I keep trying to get him to move to a single-story house. Much easier solution!

-- Gordon

Reply to
Gordon McComb

It is the kind of thing you see in "sharper image" catalog. They will be popular among people with a bunch of money, until one of us comes out with something affordable that everyone can buy.

Rich

Reply to
aiiadict

I don't see that. Robosapiaen is a toy the performs on demand, roomba is a carpet cleaner. What is whitebox going to do?

While price is very important, utility is very important.

Reply to
mlw

Would you really?

The question is whether or not they will be able to lower thier prices. They spent a lot of R&D money to get where they are. Unless they can find a viable market to support them, they will have to keep the price high.

This "pre-sale" tactic indicates to me that they may not be able to fund themselves.

Reply to
mlw

Probably as much as your robot. Your platform doesn't specify anything beyond what the Whitebox robots will have off the shelf -- a PC, an OS, some path-planning software, a camera. How did yours suddenly become so capable? (The Whitebox machines will use the ER vision software, which is actually pretty good.)

While we're on the subject, Roomba is a poor carpet cleaner. It picks up little specks of stuff, but that's about it. People buy these so they can say a robot is cleaning their house. Then, when no one's looking, they get out a real vacuum.

Don't get me wrong, I welcome the Roomba, but its utility as a vacuum cleaner is questionable. It inspires more than it cleans. I have nothing against products that inspire.

-- Gordon

Reply to
Gordon McComb

It isn't a bad mousebot.

Not really. My $500 linux robot project is aimed at engineers and hobbiests interested in doing a robotic project that may not be directly involved with the chassis.

Actually, I look at it 180 degrees differently. I am creating and documenting a system, hopefully, in such a way that people need not know the intricacies of the various components to accomplish their task.

I hope to provide a platform that does all the neat stuff, is easy to build for yourself, and lets you focus on what you want to do.

Exactly.

Reply to
mlw

Yes, but that isn't the issue is it? I haven't decided to sell my system yet. Regardless, $1200 and $1700 are substantial investments.

yup

I won't debate the usefullness of the roomba because I feel it is dubious at best, but they have a marketing focus.

Reply to
mlw

Irrelevent. You asked what the Whitebox is going to do. "Your" system, at $500 just for parts, is equally substantial if it has the same limited function. Price is relative to value. A lot of labor has to go into building on of "your" robots, and labor has a value.

You know, $1,200 isn't all that much when the machine is first and foremost a PC, especially a Windows PC, which has wide acceptance as a desktop appliance. It's not much more than a modded PC, and guaranteed to be different than what anyone else has in their dorm rooms.

Has it occurred to you that you might work with this guy to offer a Linux alternative? The hardware is quite similar, down to the VIA EPIA board. You'll have to support steppers, though. They'll be in your neck of the woods May 10-12 at Robobusiness, along with Evolution Robotics, and other leaders and shakers. You could tell them how wrong they are to use Windows.

-- Gordon

Reply to
Gordon McComb

What constitutes "easy to build" to you?

-C

Reply to
Hoss

PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.