Scribble: Wait, Kickstarter Is Vetting Projects Now?

PenFirst rule of reading anything: if a headline is an interrogative, the answer is a resounding ‘no’. This might be the one exception to that rule.

This Kickstarter is actually fairly interesting. Not because it’s an obvious scam, mind you, as there’s very, very little to actually call a scam. It’s noteworthy because it was on track to be a highly successful campaign but it was shut down by the creators just days after its launch.

Before getting to the unsavoriness of this Kickstarter campaign, a little bit of history is in order. Several years ago and before crowd funding was a thing, a designer came up with a rather clever if completely improbable idea: a color picking pen. Simply hold the end of a pen up to an object, press a button, and using technology and/or magic the pen now writes in that color. There are obvious shortcomings in the design like using red, green, and blue ink cartridges for color mixing – a classic case of confusing additive and subtractive color models. Still, this is just a design concept and over the years the idea of a color sensing pen that mixes ink has bounced around the Internet. With enough people willing to throw money at their screens in the hopes of actually getting a product as interesting as this, you just know it’s going to be on Kickstarter sooner or later.

Enter the Scribble Pen. Yes, it’s the same idea as the 5+ year-old color picking pen, with a few of the technical challenges already addressed. They’re using a CMYK (plus White) color model that can theoretically reproduce just about any color, and do so on any color paper. How are they doing this? I have no idea, but the whole campaign is super, super sketchy.


Above you see a screencap of the Scribble Kickstarter on launch day, with $112,000 already in the bank. It has all the markings of a very successful campaign; completely funded in five hours, a slick video, a great story, and wildly popular with the press. They were featured by Huff Po, VentureBeat, most of the Gawker empire, and… wait, Hackaday? How did we end up there?

Oh, this is why. A Hackaday links post where I point out how sketchy their project is. That really doesn’t help them in the, ‘try not to be sketchy’ department.

I’ll be honest. The Scribble pen has been on my radar for a while now, and I’ve been wanting to do the complete deconstruction of this already-failed Kickstarter. Why? Because the specs on this pen are insane.

On the now-cancelled Kickstarter, the specs in the pen are extremely ambiguous, listing the processor inside the pen as simply, ‘an ARM.’ There’s nothing wrong with this, as you can find ARM microcontrollers that will fit on the head of a pin or the body of a pen. The Wayback Machine archive of the Scribble website is a little more specific: they’re planning to use an ARM 9 processor. This is just a guess, but I think they mean ARM9, also known as the kind of processor you would put in a phone, or something that requires a lot of computing power. A color sensing pen, if you can get around the mechanics of dispensing ink in a very small form factor, would never need a processor that powerful. It’s like using a Lamborghini to go to the supermarket.

Actually, no. It’s the complete opposite of picking up some groceries in a Lambo. It’s highly improbable an ARM9 will fit in this pen. It’s like using a dump truck to go to a supermarket.


According to the cache, the diameter of the Scribble pen is 15mm. The smallest ARM9-based CPU I can find is the Atmel SAM9260, a 217-LFBGA package that is 15mm square. There are references of a 10mm square ARM9 BGA package. Add in the wall thickness of the pen, the board the processor is mounted to, and the Scribble design is really, really close to an impossibility. That’s notwithstanding the absurd choice in processor, or the challenges of dispensing five colors of ink in such a small device.

I held off on posting anything about the Scribble until now because I was waiting for the team to get back to me. I emailed them last Monday, asking them about the apparent change in chip choice between the cache and the now-defunct Kickstarter. They didn’t get back to me; somewhat surprising because we’re apparently we have coat tails cool enough to ride on. Since then, Kickstarter sent an ultimatum. This is directly from the people behind the Scribble pen:

Dear Backers,

With your amazing support, we reached and surpassed our initial goal within 5 hours.

In view of the many requests to better see and understand how the Scribble Pen and Scribble Stylus works, we are currently shooting another video in more depth which will be available sometime next week .

However, We received an email from kickstarter dated 08/12/14 requesting a new and more illustrative video within 24 hours, so we had to cancel the project with hopes of restarting it again  with a much more detailed video.

Yes, apparently Kickstarter does do some vetting of projects. Scribble will be back again once these guys can shoot some more video. It’s not like you can easily fake this sort of thing. Oh wait. You can.

At the time this Kickstarter was cancelled, the Scribble team was sitting on $366,000, with a few projections saying they would see about $5 Million by the time the campaign was over. I’m actually somewhat impressed the Scribble team didn’t decide to ‘fix it in post’, if you will, and continue the campaign. Don’t worry. They’ll be back, and next time they can legitimately use the Hackaday logo in their campaign.

102 thoughts on “Scribble: Wait, Kickstarter Is Vetting Projects Now?

    1. Tell me how you would fit all what is needed in the size of a pen?
      – Color sensor
      – MCU to read sensor, compute values of primaries colors and control color mixing.
      – Some injector system to control the volume of each ink in mixing
      – 4 or 5 primaries colors reservoirs.
      – battery

      It’s a lot to fit in such a small shell.

      Because micro-processor are now engraved at nanoscale doesn’t means that everything came be made at that scale.

      1. I disagree

        Color sensors come in 2x2mm packages and can be put on a flex. mcus come in 2×2 packages too (LPC1100). injectors come in small sizes too (your inkjet printer cartridge has a few *hundred* of them in the 10mm^2 area on the bottom of it. color reservoirs need not be large (they never made claims, as far as i can see, to the useful size of the reservoirs). We can say 1mL of each color. Lithium batteries come in all sizes and shapes, including [very convenient in this case] cylindrical.

        Software is trivial. The RGB -> CMYK conversion does not need an ARM, nor does driving printer nozzles.

        With a good industrial designer, this project is not only possible, it is in fact not even that difficult. It is slightly simpler than designing a modern bluetooth headset.

        1. Anyone see the potential problem here with Hackaday posting something that they purport as impossible, waiting for an onslaught of comments and ideas on how to resolve potential issues? Aside from the fact that it’s just plain good old marketing driving publicity, a site with the level of credentials such as Hackaday could be the most efficient way to troll some of the most technical / intellectual individuals from around the globe into resolving design issues of a potentially lucrative startup campaign.

        2. One detail, you would need to make your own print head, there is no such thing as a 5 color, 5 nozzle print head at this moment, all heads on the market are something like 50 nozzles but 1 color, and then one of those per ink cartridge.

          The project is not impossible, but it requires a very heavy startup capital, probably beyond a normal kickstarter campaign.

          Then there is the practicality, the ink cartridges would be laughably small in any pen you could hold when subtracting space for all the electronics and battery.

          The project is possible with today’s technology, but it is physically and economically impractical for a single purpose novelty device.

    2. I’d assume because of the mechanics of dispensing the ink. I’m no pen expert, but conventional pens have a single ball per ink, and can only draw in one ink at a time. Mixing the inks to dispense with a single ball would probably require pumps, or at the very least valves which would be difficult to fit in the form factor provided.

  1. Perhaps Kickstarter is beginning to realize (after the iFind debacle) that obvious scams and their attendent negative publicity can hurt Kickstarter’s credibility and thus their ability to attract viable (and therefore profitable) projects?

  2. I don’t think this is impossible to achieve (minus the weird specs like ARM9).
    The most difficult thing to achieve is the color mixing, but it should be quite possible with current things people do on a chip with microfluidics.

    But one thing that will probably not work as well is that you have to still consume some previously mixed color before you get to the new one.

    Yet except the fun part I doubt you could find this thing to be that practical. A stylus that I could use on the phone/tables, yes, but not an ink one.

    1. “But one thing that will probably not work as well is that you have to still consume some previously mixed color before you get to the new one.”

      Perhaps not, if, given your mention of microfluidics, the mixing chamber is small enough. Might still get a bit of left-over color form the previous run, but It could be kept quite low.

      1. The mixing chamber can be small enough, but what about the piping up to the ball?
        I have no idea how much that is, maybe it is 1-2-5cm of writing maybe it’s 1m, i have no clue.

        Now i am starting to wonder if there is any diy way to achieve such a small mixing chamber + flow control.

        1. You could theoretically fit the entire mixing chamber into the cavity just behind the nib / ball assembly of the pen. Not entirely sure about the remaining connections to the fluid reservoirs. We do some work in our research group here that involves a two step laser machining / chemical etching process to produce micro-channels in silica glass. You could imagine a system where the pen nib / ball assy is attached directly to the mixing chamber (think lab-on-a-chip style assembly). I don’t see why this process couldn’t be applied to making pens. One would need to know more about the actual physics of ink flow in a pen (ie. flow rate under normal writing conditions).

          There are various solutions for flow control and mixing as well, though this is not my area of expertise.

          As for DIY.. not sure about that one. How versed are you at building your own (ultrafast) laser system? Again, well within the realm of possible given the variety of parts available on ebay. ;)

          1. You are right, you could probably stuff everything behind the ball.

            Now i am starting to wonder if they are actually capable of doing this or made this campaign specially to get ideas from people about how it might be done.

    1. Because when some backer gets butt-hurt about having backed a fraudulent project, they’ll need something to point at, and say “Look, I was deceived by the snake-oil pitch, but THESE people knew it was bogus, therefore Kickstarter could’ve known it was bogus, therefore they’re culplable for my loss!”

      It’s shaky logic and IANAL, but posts like this — bringing attention to the crowdfrauding campaigns out there — are probably the best bet of eventually shaming the platforms into changing the way they do business.

      Social engineering is considered hacking by many. Attack and defense are two sides of the same coin. Consider this post to be part of a defense mechanism against social engineering.

    2. If I were to back some project on there, I’d want to look around the Web first to see if anyone had any warnings about it. It might save me from throwing away money and being scammed. So by questioning like this, we might be helping people.

      Why HAD in particular? Cos a lot of people here are likely to have knowledge of the subject in question.

      On the good side, if nobody here can come up with a criticism for some particular product, then it’s probably theoretically sound. Doesn’t stop all dishonesty, but it means anyone selling miracles had better have them up and working.

    3. Because we’re people who enjoy hardware specs and are perhaps one of the better resources to help debunk obvious nefariousness?

      I mean, is what you propose that this community not be interested and also let con men get away with crazy schemes?

  3. So, which is more important, the project authors’ freedom to lie about their project, or the backers’ freedom to throw money at something they’re too stupid to realize is a fantasy?

    OH WAIT, those aren’t at odds here! Both parties apparently want it to be this way, LET THEM. Just rename to, and if you feel left out of the action, start your own fraudulent campaign. Free money from people who lack critical thinking skills — it’s just like the lottery except it’s not the government raking in the bucks, it’s YOU.

    (At this point I’m honestly not sure if I’m trolling.)

    1. Hah! I love the name (“Caveat Emptor”). But, things like this are known as pyramid schemes, and are generally frowned upon by local authorities. (Now look who’s trolling). My guess is that kickstarter has to do some minimal level of due diligence, lest they get themselves put out of business by a state prosecutor, but then I’m an engineer, not a lawyer.

      1. well they do say it is battery powered and usb rechargable.
        imagine if this thing runs out of batteries in like 15 minutes and then you need 2 hours of recharging. it might even be cheaper to buy this limited use pen for your kids than normal pens that work 24/7.

  4. I don’t know how physically feasible this is in a pen sized object, but how about four single inkjet heads and a sensor similar to an optical mouse? The biggest problem I can see would be to ignore the pulses of ink during movement detection, both in flight and subsequently when they appear on the substrate.

    1. i think inkjet heads are way too complicated for this. they rely on vaporizing the ink and projecting the vapors on paper where it cools back, turns into liquid and gets absorbed by the paper. the other method is to throw droplets of ink to the paper using ultrasonic devices.
      normal pens work by gravity and the ball takes the ink out one bit at a time. for this pen to work you will just need to find a way to control how much of the pen that goes out is from each color.

    2. I suspect that a “ballpark” writing speed would be sufficient to make ink-jets in the tip work well. Tip-speed +50% -30% should do. An optical mouse sensor that could detect rotation and translation with a telecentric lens could do that. (the mouse sensor would be on a parallel axis so wouldn’t see the drops in flight)

      Might not even need that much. A switch near the tip and settable ink flow rate would give lots of artistic control on it’s own. It’d be Sort of a micro spray painting pen then.

  5. I keep pointing out WAY dodgier shit than this to the hack a day tip line, the “blueshield” crap that magically broadcasts a “magic” field that some how will protect your “loved ones” from WiFi, bluetooth and other “emanations in the aether”, imediatlt springs to mind.

    As hackers WE are the first line of defence against shysters “selling smoke” to the “civilians” that rely on us to vet this shyte!

    We need more articles like this

    1. Meh. Stuff that’s blatantly dodgy is the low-hanging fruit. You only really notice the very stupid people and the very smart people, but most people are smart enough to recognize pure alarmist snake oil. The Scribble pen is less obvious because it requires a certain amount of technical experience to recognize the major obstacles. Similar to the cordless 3D printing pen a while back, or the supercapacitor AA battery…unless you’re already familiar with the state of the art in power density and thermodynamics, you won’t see the inconsistencies. Hackaday is better suited to analyze those types of scams rather than the latest cell phone foil sticker gimmick.

  6. If I really want to be amused by sketchy kickstarter campaigns I’m pretty sure kickstarter has their own website where I can do that. I come to hackaday for hopefully other content though. I guess what I’m trying to say is I’m getting a bit sick and tired of the articles here that showcase crappy kickstarter projects. I know there are plenty of them so it is low hanging fruit to write about. Still, seeing this here is just annoying, and lazy on the part of hackaday’s staff. If hackaday has to just produce less articles a day though I could live with that. For me calling out kickstarter scams is getting to be a crappy feature. I know, don’t click on it. I still have to see it on the main page though.

    1. Perish the thought that HAD posts useful information that you’re not interested in. Maybe they should run their article choices past you before publishing?

      In other words, just because you’re not interested doesn’t mean the rest of us aren’t.

      1. Isn’t the fundamental point that it is not a hack. Even with the most lateral interpretation. There’s a bazillion other sites that cover scams, tech and ‘news’.

        1. The issue is however; “Does everything on have to be hack related?”

          I say no.
          the HackADay prize posts about who’s judging aren’t about hacks, neither are the hackerspace tours, meet so-and-so etc…

          it’s a website about hacking, engineering, andall the social circles surrounding said activities. this post most certainly qualifies as HackADay worthy

    2. You should realize that HAD staff knows the number of page views for each of their posts. By coming to this page to comment you’ve reinforced to them that this type of post is popular.

      I suggest an experiment is in order. All of you who don’t like this type of post just don’t click on the post to begin with. If the overwhelming majority never opens the page, HAD will focus their attention elsewhere (away from Kickstarter frauds). And if they don’t… well, at least I won’t have to worry about reading your bitching!

  7. The ARM9 size point is mooth as ARM9 as avaiable as “Synthesizable and hard macro”. even if there’s nothing on the market there (which I doubt) you could even licence their hard macro and create some bare die ARM9…

  8. The video is totally bogus… When he draws the second time, you can see its reflection on that cylinder next to the paper,… and the line is blue, so is the tip of the pen…

    1. …Are you referring to the video in this article made specifically to demonstrate how easily and quickly someone can throw together a fake video that demonstrates the concept described in the kickstarter?

      Reading comprehension is important.

  9. Brian these sort of articles aren’t very constructive. You’re basically taking the worst part of HaD where commenters rip a project to shreds and turning it into a feature. Yes, it’s a bad kickstarter. No, I don’t think 10mm is the smallest arm chip. But what is the relation to HaD for this sort of article. It’s not a hack. Your post isn’t about how to make one as a hack. So what is the relation here? Why is this relevant to a site that normally tries to bring cool hacks into the spotlight?

  10. This is the problem I have with the HaD editors: They don’t know what they don’t know.

    This Kickstarter may or may not be a scam; I have no real comment on that. But in the “proof” that this is a scam, we’re told that because Brian Benchoff knows how to look up “ARM9” on Digi-Key (or wherever) and not find a COTS chip smaller than 10mm, then by golly, there can’t possibly be any other packaging option available. Even though bare die and other options are often available from manufacturers. And I say that as someone who works for an electronics manufacturer that is regularly inundated with semiconductor company “sales engineers” who are constantly offering us all sorts of packaging options– some of which would easily apply here. You won’t find these with a casual search of Digi-Key (or others), but you will if you actually sit down with representatives of semiconductor manufactures.

    In other words, your ability to search for information with Google, Wikipedia, and MathWorld isn’t the last word in… well, anything.

    Oh, and yes, they made the grave sin of originally mentioning an “ARM 9” microcontroller instead of the ever-so-different “ARM9”. Why, that extra space is clearly indicative that this is a scam, because every electronics geek knows there is no space. (This is like the pointless arguments you hear from other geeks like “clearly your Star Trek fan fiction is of poor quality because you reference the HB27c WarpPulse drive when in fact it is a HB27d WarpPulse drive” kind of pointlessly pedantic nonsense.) But what about that dreaded (and oh-so-telling) extra space? Well, here’s how it works in the real world. The companies I have worked for have these things called marketing departments that often don’t have a freaking clue about anything more technical than where to plug things in. These people will generate marketing materials and other documents which feature violations of the laws of physics, horrendous descriptions of technology, and embarrassing use of terminology they don’t understand. Many times, these will pass by engineering first for review and correction, but sometimes in a fit of rushing to the presses, they burp out these screaming honkers of stupidity.

    And yes, every group that puts up projects on Kickstarter should pay loads of attention to the technical details and make sure everything is right. And yes, people who see these amateur efforts have every right to point out the errors, questionable bits, and call “scam” when warranted. But please, before you rush to pump out a story, pass it by someone other than fellow editors and pals who know only as much as you do. You might, for example, post a story *asking* if something is a scam instead of breathlessly rushing to declare it so.

    1. > But in the “proof” that this is a scam, we’re told that because Brian Benchoff knows how to look up “ARM9″ on Digi-Key (or wherever) and not find a COTS chip smaller than 10mm

      Actually it was going to, looking at all the ARM9 cores, going to manufactures web pages, searching for all the relevant cores, and looking at all the datasheets. You *do* know I had months to do this research, right? And that did include bare dies. I’ll admit the 10mm one slipped by me, but that’s a significant investment in time.

      And your entire comment doesn’t address how overpowered any of those chips would be for a color sensing pen. My technical assessment, while imperfect and not considering putting an ARM on custom silicon and other crazy stuff for a first-gen product, cannot be called incomplete given what we know about this pen. What you see above is extremely close to the best technical assessment possible given the limited information.

  11. Next project: hypodermic syringe where you hold the back to your forehead, it analyses the disease you have, then it DNA recombines a serum and you inject yourself with the cure on the other side. Easy as a pie.

    ‘If you don’t support this project you like hitler.’

    To quote HaD: “‘If you don’t support this project you like hitler.'” and “return to the top”

  12. Lets me think about it

    – the idea is brilliant
    – the idea is self-evident
    – there are companys that earn Money with selling printers
    – they have the ink, the printer head etc.
    – one can assume that they have a lot of developers, geniuses, outsourced research

    And still this product does not exist.

    1. i think in practice it is very hard to sell it to mass market. I see it as an invention we could have a for a long time but we don’t have it because it is too expensive to make for the gain it brings.
      i bet that the color on paper is still significantly different that what it captures. I mean, pro photographers using expensive cameras, expensive calibration devices, expensive printers and expensive paper have trouble getting a photo to look like in real life. and these guys can do it in a pen for $119? i doubt. you could sort of make it, but not significantly. maybe a 50 color crayon set is better than this thing at matching a real color.

  13. I think it *is* possible to get a pen-like device that can perform as they say. However, a pen-like device that is the size of a pen with the hardware they want…. that’s a different story. I think it boils down to a bunch of art students deciding to make something, and then realizing their mistake later. They probably thought they could get any component at any size, or that “oh, we can add an ARM9; that’d be easy, right?”

  14. I think you can make this device with an 8 bit micro. I mean, what do you need:
    – I2C rgb sensor that provides 3 values, 8 – 16 bits each
    -1GB of memory (this is really stupid to have) over SPI, eg micro sd – no problem.
    -simple bluetooth module via UART – what’s the big deal?
    -some PWM channels to drive the ink mixing

    if you think about the functionality you don’t need more than to capture the RBG sensor data and output it to the ink mixer. sure, some processing, but not much for 3 lousy bytes of input.

    so i think it is fair that Brian Benchoff got stirred up about the ARM[ ]9. there is no need for such a powerful device, whether there is space for it or not. Let’s not forget that arm licences the core, so if you want to make an 8 pin ARM9 running at 1GHz+ with 4 GPIO, then you can (***Let’s assume it is technically possible)

    Now if someone can explain to me what a person could do with a memory of 100.000 things and how they could manage that….

    1. He gets stired up over anything. Things don’t come in the 74LS or 4000 CMOS series anymore. Even the ATmega is to big these days. When you want to build something simple, the chip is so huge that you can’t fit it in your shoe. By going up in core and MHz it suddenly gets smaller. No need for the processing power but the package is way smaller?

  15. Can’t you even in your mind imagine how a pen like this would work?
    There is no way in hell they can mix ink to do this on demand.

    Say that you scan black and draw with that.
    Next you want white.

    How will it do an instant change from black to white?
    Just figure, no black ink left in the pipe and instant white?

    It cant be done.

    An inkjet printer spits the ink on to the paper.
    No mixing problems, all colors separated.

    This is through a tip.

    Not going to work.

    1. It wouldn’t need to be an instant change. It could, on each colour scan, require the user to do a bit of scribbling to get the old ink out.

      Still, how to get a ball-point, dirty with black ink, clean and writing pure yellow without washing it seems rather difficult. Unless the inks are something really special. Maybe they’re using invisible ink.

  16. This would be hard to pull off, but it’s decidedly not impossible. The sensor is easy, as stated above, they are pretty tiny these days. But in reality? You don’t need a very powerful ARM processor in the pen. Have something just powerful enough to transmit and receive data and output values to the ink controler…and a bluetooth module to your smartphone with an app.

    That way, you can not only sense colors, and duplicate them in real time, but store them for later use…or manually program specific colors of ink without scanning.

    As for the mixing issues, provided the right viscosity of ink, you wouldn’t even need a mixing system…you could do it on the roller as it goes. You have each of your ink reservoirs deposit ink incorrectly onto the back of a rollerball. A minute amount at a time, and the action of rolling the ball over the medium spreads out and mixes the inks over the surface of the ball. It would be a question of fine tuning viscosity to get the proper flow to mix properly on paper.

  17. It would be near impossible to replicate colors exactly. First off printers inkjet or laser do not mix CMYK colors to make other colors they layer them to make the eyes perceive them as a solid color. So the pen would have to spray to be able to layer the colors. Which means it would have to track how far the tip has moved. Second how would they account for the different brightness papers, white is not always white. And third how would they account for lighting. Look at a printed picture under different types of fluorescent lights, incandescent light, led lights and in sun light and the colors will look different. A printer that is calibrated in an office with no windows will print documents that look like crap outdoors. So I think this is total BS.

  18. “Several years ago and before crowd funding was a thing, a designer came up with a rather clever if completely improbable idea: a color picking pen.”

    These were popular for girls in grade school in the early 90’s (liberal brain washing makes that comment seem sexist in 2014).

    I was equally surprised when a friend tried to tell me about the ‘exciting invention’, a color picking pen, a few years back. Seems you can ‘invent’ things all day as long as no one remembers the original.

    1. To be specific, I am talking about the manual single color selecting units that are referenced as 5+ years old.

      About two decades ago the popular version was about 2″ in diameter with over a dozen color choices. These were more a novelty item, as school progressed I would actually see the smaller <1" units more and more, they used Red,Green, and Blue selectors, which are actually practical in school for work, grading, and notes.

  19. I don’t know why I care, maybe it’s because I’m amazed how out of touch people are with mixing colour but it is simply impossible to mix five inks to get every colour! The article mentioned it -its subtractive colour mixing not additive. This is a hoax. It’s amazing people bang on about sensors and processors and the pen tip ckeaning method when this simple physical impossibility is staring them in the face.

  20. in their video, they never showed an entire demo of the pen. everything was disconnected, so we don’t know if they have a working prototype or not. looking at their timeline, it seems that they “have” a demoable product, but there is nothing to prove that. disregarding all technical challenges, this along makes this project very sketchy.

  21. Benchoff needs to find a new hobby. He is obsessed with kickstarter scams. Honestly, you could probably create a blog solely dedicated to crowd funding scams.
    This isn’t a hack. This site is called Hack a day.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.