nemik

While OpenWRT is really good and often the best one for really constrained Linux HW, adding features like secure-boot and secure storage to the builds can be much more difficult than on Yocto. Especially when vendors like NXP provide that functionality in their Yocto BSPs. This can be important for commercial projects because a lot of customers are now asking for such features to secure end-devices.

For more training data, I wonder if you could make Lego parts in Sketchup or some 3D program, then render them in a 'scene' similar to your camera setup using a renderer like Maxwell or V-Ray or whatever. Then you could maybe be able to generate unlimited numbers of sample images to train on.

I'm doing a similar experiment now to train a model to parse out an image of a blood pressure monitor that's a 7-segment LCD display. To do it I separated out each segment of the display as masks with Gimp/Photoshop and then I can create my own images by just overlaying them on top of an image of a blank LCD display. That gets me basically unlimited training photos. If you could render the 3D parts from various angles, colours, etc then something similar might be possible.

Also, you said you're doing modified VGG and into 20k classes. That works, but another thing to maybe try is use binary_crossentropy as the loss function and a sigmoid (instead of softmax) on the final activation layer, to be able to do multiclass classification. Then your labels could be a vector of shape possibilities, colour possibilities, or whatever you could divide your 20k classes into.

Thanks for writing this up and adding all these detailed notes! Super awesome work on this, it's very cool.

The air puffs seem like the most unreliable parts at the moment?

Maybe if you could have some manual labour, one thing to do would be to mount a project above the setup to shine a graphic you generate onto it.

If you have bins labeled, you could project the label next to or on top of the part as it's moving on the conveyor. That way if you have human pickers, you're not constrained by the number of bins with each needing its own air-puff solenoid.

His way is a lot safer. It's never a "$1" chip, you also need a power supply for that chip, breadboard (or dev kit) which puts price into cheapo webcam range.

On top of that, breaking out to LCD pins or their drivers is not easy. They're often very fine-pitch pins, if they have pins at all, some can be QFNP packages making it even harder to get to the pins. There's a big risk of just straight-up breaking the thermometer doing this, so if you attempt it, it's best to buy a backup.

Personally, rather than thinner/lighter I'd pay for it to be a bit bigger but capable of letting me open it up and replace RAM and the SSD as was possible up to the 2013 ones.

:( I think you just called yourself stupid.

Check out DirtyPCBs too. Based in Shenzhen but offers next-day DHL. Usually a 4 to 6 day turn-around time for me; great prices and good quality too.

If you're in a bigger city, also check out some smaller shops near you. There are a few nearby in the suburbs of Chicago for me and sometimes I go to them for quick turn-around time and a price that's higher than DirtyPCBs/Seeed's, but much closer than Advanced Circuits'.

https://www.amazon.com/ALITOVE-WS2812B-Individually-Addressa... are perfectly cromulent LEDs and less expensive than Neopixels. Not as cheap as the RGB strips, but if you pick off more points for colour analysis, you could adjust them along the edges, if you really wanted to. Nice work on the project!

What keeps it from crushing you while you sleep?

Vim for everything (C, Python, JS/HTML/CSS, etc etc) except Android development, I use Android Studio for that.

This here is right on.

I will add that if you can afford the time and effort to do so, it would be good to design your system in the beginning to work on multiple providers without many issues. That means trying as hard as you can to use as little provider-specific things as you can (RDS, DynamoDB, SQS, BigTable, etc). In most cases, pjlegato's 1) will still apply.

But you get a massive side-benefit (main benefit, I think) in cost. There are huge bidding wars between providers and if you're a startup and know how to play them off each other, you could even get away with not having to pay hosting costs for years. GC, AWS, Azure, Rackspace, Aliyun, etc, etc are all fighting for your business. If you've done the work to be provider-agnostic, you could switch between them with much less effort and reap the savings.

As the hiring party, how would you feel about a hypothetical deadline from the hiree on how long you have to evaluate them and present an offer? Would limiting your time for that evaluation be unreasonable too?

This is very cool, thank you for making it!

I just have one concern: Does the AGPLv3 license mean I need to open-source whatever other service I make that might be making use of this?

It's what fuses and diodes are used for. You know how when you write a program you never trust user-input? This should be the same thing, but physically. A well-designed USB port wouldn't fry the computer like it did.

I also advocate for Java, but mostly because I invested in companies that manufacture RAM.

No big deal, you can make it work. axTLS is a pretty lightweight TLS implementation that you could strap to your TCP framework and make it work fine. It's also BSD-licensed so you can put it in proprietary stuff.

You can program the ESP8266 yourself. For its TCP stack it uses lwip which you can also compile yourself. The 802.11 stack is more of a black box though, but not necessarily the TCP stack.

Hmm, look's like Vishay's VCNL3020 in there? Or the good ol' Sharp GP2Y0A21YK0F.

Left wondering why you need a base controller and the dual-PIR's though, especially if the PIR's are connected to DC voltage anyway and not battery-powered like some other commentor said.

I get you would wanna use wifi to talk up to your cloud and maybe that's what the base is for. But seems like a waste if you got something expensive like a Raspberry Pi or other ARM A-series in there. There are much cheaper wifi chipsets out there now these days that could do the same thing without needing all that cost and space of an ARM Cortex A-core, plus wifi, Linux, etc. You could probably get away with putting a small wifi chipset in each of the dual-PIR housings and still come out ahead in cost. The ESP8266 sounds perfect for it actually.

Either way, a pretty cool product.

So what do you think of https://en.wikipedia.org/wiki/London_Agreement_on_German_Ext... ?

Perhaps everyone in Germany would be better off still re-paying that 50% that was just removed.

Hmmm I was under the impression that cash was part of their operating expenses. They have retail stores, their campus, most of their employees and all their most expensive employees are all in the US too. But you're right, it's probably still enough to eat 3 months worth of royalty payments.

Also likely, this isn't going to be the only or last fuck-you Apple is going to be giving artists or content producers. The ones who agree to be screwed for the 3-month opening trial will be push-overs who can be leaned on again when the time comes.

From http://www.aboveavalon.com/notes/2015/6/13/apples-cash-dilem... it seems like Apple may not actually have the cash to do this, not in the USA at least and it would need to pay taxes on the cash it is holding outside the US to pay the majority US labels/producers/artists.

Not that this excuses what they're trying to pull here or anything. By trying to weasel out of paying any US taxes, they can't spend their own money. So in this case they really might be too poor to make this deal work otherwise.

Good for Taylor Swift to stand up to this kind of crap.

I'm unsure of how to reconcile the conflicting things Cook says, then does: http://www.businessinsider.com/apple-china-security-checks-o... China will ask for keys and Apple will give them up. I guess because China's a much bigger market than the US.

I find this policy incredibly sexist and am very surprised it came from a woman. If the cause of this was women's supposed inferiority at negotiating, does Reddit prevent women from having sales roles where good negotiation on behalf of the company is important? How about as attorneys expected to negotiate for the company, are women excluded from those jobs?

Agreed, but touchscreens and SD or USB means connectors, which are often a significant part of BOM cost and take up space. If you can do it all from the one-chip/module you have in there it's a big benefit when it comes to making something small and cheap.

The CC3XXX "SmartConfig" was complete garbage the last time I was evaluating it on a CC3200. It didn't work on 802.11n networks, only b/g. And that was when it worked. I got it only once trying on both an iPhone and Nexus 5, I would hate to have to answer any customers' support questions about using something so flaky.

The audio thing does require an app though to do provisioning, which I suppose isn't a problem for the kinds of people ordering this thing.

My preferred provisioning method is to let the little Dash-like device broadcast some setup or config access point. You connect to it, get an HTML UI to input target AP credentials into, and submit the form. It works from just about any device.

Just because it's 32-bit doesn't mean its code can't store 64-bit ints. There's just a few more instructions that are needed to manipulate it. But it's not a big deal.

That's how computers are able to store, retrieve, and manipulate any variables (like strings) that might be bigger than their address space.

"There's a widespread belief that an Android device without Google services (really, this means Maps and the app store) is unsaleable outside China (I'm not entirely sure about this, as I wrote here)"

Like you said in that link, Amazon has tried this and I'm not seeing it really working out all that well for them. I think the only reason phones without Google Services work in China is because the entire country blocks them and gives their domestics an artificial advantage. Amazon might be there too but only because they're too small to even worry about. Not so with Google and its Android services.

Very cool! But ATMega's can be pretty pricey as well as connectors like USB. Bluetooth LE would be a pretty good alternative especially since it's wireless.

Check out platforms like the mbed (http://developer.mbed.org/) which can work with Nordic's really nice nRF51822 chipset. This is the same one in the Estimote iBeacons and it's a very nice platform.

They even come in module forms that are pre-certified with the FCC to make integration with end-devices more friendly. Modules such as the PTR5518 can be pretty cheap (http://www.aliexpress.com/store/product/PTR5518-Coin-size-nR...) and small, and they just could get soldered to your existing circuit. Best of all, the mbed source code you write for the firmware works directly in those modules too.

Good luck with the project!

I'd recommend starting out with Eagle, just use their free version in the beginning, it's more than enough. The best tutorial that helped me most when I was in your same spot is: http://aaroneiche.com/2010/06/24/a-beginners-guide-to-making... It's a little older but applies just the same to the newer versions of Eagle (7 is the latest).

I'd encourage you to start with a simpler circuit for the first time. Then later it won't be too hard to expand on it and adjust trace widths for higher/lower current, and ground planes (or lack of them under RF). Good luck!

And unless you really want to learn how to etch your own boards with chemicals, just use OSHPark and have them send you really nice ones for cheap.

The choice of GPLv3 for a license doesn't make much sense to me. The bottom of http://we-io.net/whyWeio.html seems to allude to embedding it into end-user devices in production. With GPLv3, wouldn't all your custom source code also then have to be open sourced?

As for battery life, the AR9331 SoC is used in the TP-Link MR3040 uses a lithium battery similar to old Nokia ones, and it lasts for about 5 hours. So this product could certainly work well with larger batteries, but really no chance with AAA or AA cells.