But I'm very glad I was wrong in my scepticism. This is freaking amazing on so many levels.
Edit: It’s been a while since I’ve been so excited about a project. Now I can’t work just thinking of taking a month off to play with this thing :)
We have made great strides, but in the U.S. social mobility is lower now, life expectancy is down, so if you have to be born to a poor family, being born earlier might be better.
How about "the percentage of millionaires who were born in a poor family"? I bet this is where US shines. Rich people in most other countries are mostly from old money.
Regarding social mobility -- well, you can live a good life in Denmark, but living great life is different. Janteloven is still very much a thing.
Sociology, economics and politics are difficult, valuable and all too often completely overlooked by people who have the luxury of being able to solve for x in their day job.
In many ways we've already tried that, for example in urban planning, and it failed miserably.
For a counter point, as applied to urban planning, see: https://en.wikipedia.org/wiki/The_Death_and_Life_of_Great_Am...
Social problems are best solved from a grassroots perspective, and they require negotiation, compromise, and HN hates this one, wait for it... politics (both official politics, and everyday interpersonal politics.)
Some observations are quite surprising, for example using coal and then oil for energy replaced huge amounts of manual labor and might have eliminated slavery, deforestation and hunting whales for whale oil.
What great strides do you think America/the developed world have left to make?
It's an amazing read as well, much like Pinker's other works.
Poverty: It is now known that there is a rising gulf between subjective and objective well being. A majority of millionaires do not think they are rich. I think this psychology will make it impossible to solve poverty.
Cancer: Entropy/cosmic rays scrambling DNA... I think thermodynamics will make it impossible to solve cancer (as in, all cancer) for everyone (very rich with DNA repair therapy in the medium/near future).
War: Actually living now in the most peaceful time in human history as a % of humans in conflict. I think the human desire for power/wealth (again, back to poverty) will make it impossible to solve the zero war scenario. At least nukes scared everyone from the big wars so far :)
This may be philosophically true, and probably even technically true in practice, but I suspect the vast majority of homeless would prefer better shelter than they currently enjoy. For whatever reason, these people have failed to attain better shelter within our societal system. Some people blame the system for letting these people down, other people blame the individual for not having what it takes to "make it", but I don't think you can call this a choice unless you call failure a choice.
Given that, there probably are things we can do to "solve" the vast majority of homelessness, so don't give up just yet. The few people that then legitimately choose to be homeless will probably be so scant as to be barely noticed by the rest of us.
I'd argue (and I'm not a millionaire) that being worth less than 10M isn't really that rich. A long-term pet peeve of me is that most people don't realise how poor they are, if you treat everything as being relative.
For most civilians, access to wealth and being able to access debt (and leverage) is tied to house prices, and a lot of house prices in high-paid areas are beyond the reach of most people. Thus, by extension, access to wealth is not easy for most people.
Trying to have access to wealth AND having at least one stay at home full-time parent? Even harder to achieve in this day and age. People born between the 50s and 80s basically were able to both get on the wealth ladder AND afford for one parent to dedicate themselves full-time to parenting their children.
War, meanwhile is a scalar concept of the platonic idea of conflict in general, between two individuals, but scaled up and beyond the scope of individuals to transcend to rival collectives.
So, cancer. Unsolvable for absolutely everyone, although it can be solved for some, but then dancing around the issue of wealth again.
Here’s where I’d like to stop for a moment, and dispense with the conceptual curveballs being tossed around, because now with wealth as a requirement for diamond sure shots against cancer being accessible only to the wealthy, we’re right back onto the homelessness/poverty concept.
I’m going to cut to the chase: the words “illegal to force someone to take care of themselves” is an extremely misguided framing of what homelessness is. Almost to the point of willful deception. That ain’t what homelessness is.
Okay, sure. You’ve got psychiatric, fractured people walking around in circles, and lead a horse to water, do no harm, you’re idea of compelled assistance is my idea of a cage, and what about all the sociopaths who might game the system like blood sucking parasites. Healthy people need not be beset by the lampreys of willful destruction and self destruction.
The U.S. Marshalls and/or the town sheriff’s department will trot on up to my door with cuffs in hand, and guns drawn, within 45 days of my bank telling someone that I haven’t scored enough paper points to pay tribute to the monarch of my roof. So, under threat of violence, not only is the degree of luxury stripped from my permission to simply exist comfortably, but perhaps I land in a cage. A real cage, not a conceptual economic cage of taxes and regulation. A cage shared with some of the fractured psych cases that might fracture me to match their proclivities and better suit their surroundings.
So, poverty. Back to wealth, yes? Points scored on paper, retained by banks that publish scores to highly available, networked computer databases for live transactions and batch processing.
We’re rapidly approaching a situation where electronic systems, computer software, and related hardware applications will obviate human effort in a wide array of scenarios. Transportation, anything involving sorting, organizing and distribution, rote fabrication and assembly; much of that can be automated, with or without magic decision making buzzwords for the edge cases that are currently mechanical turked with brute force data entry by humans in the loop.
This reorganization, centralized around advanced electronic systems of record and authority will drastically improve efficiency, to simultaneously create surplus and idle humans in one stroke, or many, many concurrent strokes, struck within a very fast, short span of time.
Now, broken humans, stupid humans, evil humans, and even just plain old humans, mediocre, unfuckable, and aged out; such a chore to be around and listen to. Why do anything for anyone that can’t charm your pants off? And then, there’s the biosphere to think about. Do we use automation to unleash a surplus of pollution, garbage, waste and toxicity upon our already fragile planet, but for the want of maximizing a bottomless pit of human activity? Indeed, when we play god, and cure cancer, feed the poor and house the homeless, do we even fix the ugly, mean freaks that no one finds adorable or even mildly interesting?
I think we can draw a bounding box around an extrapolation of all living humans in their current state of affairs, and what it would take to raise the standard of living, and provide a comfortable pasture for all the broken shambles of misfortune that creates poverty and all the mental illness that creates homelessness. We’ll probably even have enough surplus to snip out all those pesky tumors, and mend the festering sores that sprout more. We’ll probably be able to figure this stuff out, granted that biology is merely piles of chemistry, which of course is merely piles of physics, and that highly efficient resource sharing will produce idle humans bathed in surplus, many of whom will be pretty smart, healthy and motivated. My, that’s optimistic, isn’t it?
So, I dunno. I don’t see things your way, but then again, I have to spend eight or more exhausting hours a day, on this treadmill of income chasing to pay rent, eat food, run fool’s errands, commute, idle in front of a cathode ray tube, psychically recovering from the trauma of the stupidity inflicted upon me by this shitty mess of a groundhog’s day rat race, and then sleep for eight hours, so I’m actively prevented from helping you solve the problems you attempt to frame as so bloody insurmountable.
C’est la vie...
I was being a bit flippant in my comment because I take issue with the phrasing of these problems in the context of 'solutionism.'
We can, and have, eased poverty. Families aren’t having to eat their children as happened in 1920's Russia. We must strive to ease it our entire lives to add meaning, even if we know we will never reach the asymptote for the reasons I mentioned,
We must strive to ease war. Does that mean having a strong police-like presence in the Middle East, or just getting the fuck out? Our own society arose like a Phoenix from extreme violence, at we harming less developed countries by constant intervention in their sectarian violence?
We must strive to ease cancer, because we have to solve this before the heat death of the sun if we wish for humanity to survive afterwards.
Peace be with you
Oh and I don't know, about $250K worth of professional experience and skills but yeah.. we have.
That is the level of technology that we are at today. We are living in the future. And it is a future that Asimov would like.
The 3D printer OP uses is a Monoprice; it looks like the delta mini (https://www.monoprice.com/product?p_id=21666). It costs $160.
The circuit boards and the solder paste mask are made by another company and delivered by mail. It's been a while since I looked up costs, but it's quite cheap these days, and the watch's circuit board is small, so I can comfortably say it cost under $100 to get the mask and a few copies of the board.
The soldering could be done with a $200 chinese reflow oven, like the T962 (https://www.ebay.com/sch/i.html?_nkw=t962), or with a $20 toaster oven, an oven thermometer, and some careful timing. (Some people do their surface-mount soldering with just a frying pan and their kitchen stove!)
The rest is just basic hand tools -- screwdriver, file, sandpaper, tweezers, maybe a magnifier. Maybe a $5 Arduino clone to help with debugging and programming, and a $35 T12 soldering tip+controller for doing soldering test stuff and "I messed up" wires.
If the first try doesn't work, and it's not obvious why, you'll to spend more money.
If fixing the problem requires changing the board board then you need to order new ones, and the mask to solder them.
DC voltage levels, current consumption, and electrical continuity can be checked with a $20 multimeter. The presence, frequency, and integrity of signals can be checked with a good used analog oscilloscope costing $100-$200. Some things are easier to check with a new, basic digital oscilloscope costing $300-$400.
If the electronics are good but there's something wrong with the software, you'll need something to read the signals the chips are exchanging with each other. Basic, low-speed (less than 1 MHz or so) signals can be read and generated with the Arduino board, or a $30 Bus Pirate. If your project uses higher-speed signals, you'll need a logic analyzer. A good USB model is the Saleae Logic 8, which costs $400.
(There is overlap in the capabilities of digital multimeters and logic analyzers, so you can sometimes just buy one and use it for both purposes.)
Soldering was done in a reflow oven, commercial units start at around $200 or you can use a modified toaster oven. This board doesn't look too terrible to do with an iron or hot air gun either. You can get a decent combo hot air/iron station for about $60.
You'll want some test equipment, cheap multimeters start at free from harbor freight but you'll probably at least want to step up to a ~$15 model. Logic analyzers are handy and until fairly recently were quite expensive but now you can get them for $10. An oscilloscope might be handy for debugging the power circuitry here, you can get a cheap one for $20 or a nice one for a few hundred or an old nice one for about a hundred.
An adjustable bench power supply isn't really necessary but nice to have, those start around $50.
A professional EE's workbench will probably have many thousands of dollars worth of equipment on it but most hobbyists have much more modest budgets.
The 3D printer might have been the biggest offender, to be fair, but many libraries have some (as do many maker spaces that are hourly). Additionally, you can order a number of 3D printed parts online as well, for relatively low cost.
This technology can very well, and was to a large extent made to, be used for something else. I don't think researchers or EEs are stepping over homeless people more than anyone else. When it comes to doing good things with electronics I would be more worried about patents, or just finding the time.
It's more than just that though. Centuries of incremental knowledge passing from one generation to the next. A global economy that allows chips manufactured in China to be shipped to American for peanuts. All the collaborative effort that went into writing an OS that's available for anyone to use free of charge. That someone is able to be so specialized, they can do this in their leisure time.
It is an incredible amount of human cooperation and achievement over thousands of years. I suppose you could say that about many things today, but for whatever reason, this watch put me in a bit of a philosophical mood.
I guess: humans collaborating achieve great things; we should do more of that.
(You know what else reliably gets me marveling about humanity? Attending a symphony orchestra performance.)
Assuming perfect reliability, reasonably low sound levels, super high efficiency engines, and flying pigs.
And those 1000 KG bullets would be, on average, piloted by someone with the skill of the average driver today.
No thank you.
I'll give two: reduce traffic jams, and less need to build/maintain roads. The average commute time* is about an hour a day. Giving people back that time to spend with their family, to work, to play would be an amazing boon to society. The real issue is that the technology to make practical (cheap, fast, quiet, safe) flying cars doesn't exist. A practical flying car would also solve the housing price crisis in cities as people could fly in from the exurbs.
Doesn't this just reinvent the problem? You're currently spending a long time commuting due to congestion; you'll now spend a long time commuting due to sheer distance.
All the traffic collisions that currently happen in 2 dimensions (changing lanes without looking, failure to yield, rear-end collisions) would happen with flying cars but now in three dimensions. And when a collision happens that completely disables the car, it won't just crash in two dimensions, it will plummet to the ground.
To prevent that, we'd need dedicated air lanes that cars would have to merge in and out of, and on-ramps where they're only allowed to move up or down at dedicated spots to help with merging. Which we have on the ground already.
Yes planes fly fast. They're also piloted by professionals who dedicate their lives to piloting them, controlled by a central ATC tower, and heavily regulated by a governing body. And we still have air disasters.
There is zero reason to move cars into a third dimension of travel. If you want "cars but faster", we have a solution for that: cars. But faster.
Ships, also driven by professionals, collide more frequently than airplanes for the same reason.
Another important reason to go 3D is that you can go much faster with less fuel at high altitude.
I have an acquaintance who was evicted (due to property damage caused by her ex boyfriend). She was given a 3 day notice to vacate, and now she has an eviction on her record. The eviction basically means that she cannot rent a property in her own name for the next 7 years, even though she's gainfully employed and can pay rent. After over a month living in a hotel, she finally found an off-lease sublet. But that's only temporary.
Agreed, I can't believe how thin and professional it is! This has to be the most mind blowing thing I've seen for a long time. :O
I wonder how the creator of this watch managed to publish the source code, including the Dialog SDK; if there is a backdoor I am missing.
Flip side of the argument: "Someone wrote a GPL library I want to use but I don't want to release my source. Is anyone really going to notice, care, or bother to bring the hammer down on my company?"
Also, the OS is FreeRTOS (Open sourced by AWS/Amazon in '17) if anyone is interested
Amazon forked it, merged their fork with the original project and took over it in 2017. You can read about it here .
EDIT: actually, I don't think they fully merged their fork as there's a variant called a:freeRTOS
> As a special exception, the copyright holder of FreeRTOS gives you permission to link FreeRTOS with independent modules that communicate with FreeRTOS solely through the FreeRTOS API interface, regardless of the license terms of these independent modules, and to copy and distribute the resulting combined work under terms of your choice
It was GPL with an exception that allowed you link your app code to the OS, with your app becoming GPL.
The performance metrics part was weird though.
Curious to know if anything knows if that's actually enforceable. The GPL gives you permission to use an distribute a program and code. I can see where if you got your copy straight from RTOS, they can enforce that clause. But, to me, the GPL seems to give everyone the right to distribute the source without that clause. And it seems to explicitly forbid the addition of additional licensing terms:
You may not impose any further
restrictions on the recipients' exercise of the rights granted herein
I've seen this happen many times in open source. For example, to this day LGPL's LibreOffice, which started as a fork of OpenOffice, reviews all OpenOffice patches and cherry picks the good ones. On the other hand, Apache OpenOffice can't benefit from any of LibreOffice's patches.
Viral licenses like GPL have an edge and will win out if there is a community that actually cares.
The virality of GPL actually results in most companies explicitly banning the use of GPL tools and code to prevent the GPL from "infecting" their own code.
Or like $1,000,000+ if you are hiring some company to design and build it for you
Anything software was shit (ie poorly done/buggy), and samsung used a lot of dark patterns (you have to install an app to get the watch paired, which is fair, but it keeps nagging you to install a second app).
My friend and I were one of the first batch of people to get it (preorder a few months before release), and had some issues initially that they quickly resolved (by sending us new watches)
all they seem to focus on is a finished LOOKING end product. i rejected a job at samsung dev based only on that knowledge - who wants to work for a company with an almost institutional disregard for software?
Withings were empathetic but took a long time to respond to my tickets and eventually I just gave up. I've still got it because I can't bring myself to get rid of it. I loved that it looked like an analogue watch, got lots of compliments on it and everyone was surprised when they found out it counted steps (etc) as well.
Are the recent ones better? I'd consider getting the new one that looks like the old Withings Activite if it holds up to the claims on the box.
> I had one of the original Withings watches (before Nokia bought them out) and it lasted about a year until it stopped reliably keeping time. It got to the point where I was doing hard resets and changing the battery every few weeks and I gave up in the end.
Worse is better. Changing batteries? Failure to keep time? Over 200 Euros for a watch that doesn't last a year? Largely unheard of in the realm of automatic watches.
People usually don't believe that it's a smart watch and more surprised when i say that it's Nokia. I usually get the reaction "They are still relevant"
Any watch apps you can recommend?
Personally the biggest boon for me is workout tracking and health data. It's transformed me over the last six months or so (girlfriend said it changed my life—though that is a bit extreme).
There's even a VO2 max estimate if you track a run or walk outdoors for more than 20 minutes.
Actually I just use it for many of the things I use my phone for except for writing long messages, email, or web browsing. Which is kind of nice. I put my phone in my bag and forget about it most of the time when I'm on the move. It's been good that way.
So I'll end there because I'm sounding like a sales pitch. I wouldn't push anyone hard to get one, but for me it's been fantastic.
Edit: For anyone else reading this comment, my biggest complaint would be battery life. I won't get more than 2 days out of it without charging because I use the battery-intensive heart rate and music functions regularly. Resolved of course by charging it nightly beside my phone.
- AutoSleep: https://itunes.apple.com/app/id1164801111 or Sleep++ - https://itunes.apple.com/us/app/sleep/id1038440371
- HeartWatch: https://itunes.apple.com/us/app/heartwatch.-heart-activity/i...
I much prefer this complication to Apple's. It displays the most recently recorded heart rate measurement
- Dark Sky: https://itunes.apple.com/us/app/dark-sky-hyperlocal-weather/...
I also prefer this weather complication and watch app
- Overcast: https://itunes.apple.com/us/app/overcast/id888422857
Great watch app for controlling podcast playback, and includes the option to go phone-free (with AirPods).
- WorkOutDoors: https://itunes.apple.com/us/app/workoutdoors/id1241909999
If you go hiking outdoors, even infrequently, this app is great. It allows you to store trail maps on your watch, and optionally record workout of your hiking (with GPS data, etc).
- Baby Monitor - https://itunes.apple.com/us/app/best-baby-monitor/id43279139...
Drop an iPhone or iPad near your sleeping baby, check in on them from your watch (with notifications and optionally video). Not sure if you have kids, but this app is a bit of a marvel, especially when out and about or traveling.
- Fantastical - https://itunes.apple.com/us/app/fantastical-2-for-iphone/id7...
I prefer this app to the built in Calendar.app, both for iPhone and Apple Watch. The complication is great, especially in the larger sizes.
Of Apple's built-in apps, I frequently use: Messages, Now Playing, Walkie Talkie, Workouts, Maps, Activity, Wallet, Phone, and Music.
and a weird checkers program are the ones I use most. But I don't have the fancy new Apple watch. I have the original one, which is much more limited than current models.
In fact, even with the apps it's sorta gimmicky. Unless you use it for health purposes (e.g. heart monitoring), a smartwatch is not of the "my life change" category. At best the "I could do the same shit with my mobile phone, now I also have another device for them that I can do them in a worse screen/input method".
Actually, adding pockets to women's apparel might be a decent side-gig business....
I suspect it's something really subtle along these lines: Beautiful women can often get by being useless, so women in general acrew certain trappings of uselessness so as to associate themselves with beauty.
It's sort of like men and commuter pickup trucks (trucks are associated with hard-working men, so men buy trucks and drive them around town, even though they have no hard work to do).
And aesthetically, a pocket stuffed with things, especially in more form fitting clothing, doesn't look good.
Is there anyone that does similar type of posts? Or even a follow along type of deal to get feet wet with actually building devices (maybe it includes a list of items to buy first)? Don't get me wrong I love writing software but showing someone something you physically made as opposed to sending a link to a website seems so much more satisfying.
Love the little gifs to explain how it's done, especially the stenciling of the PCB.
I'm curious, how/why did you choose the DA14683 chip?
"I picked the MCU for a couple of reasons:
-Very low power transmit/receive for the bluetooth (<4mA). I get around a week of battery life
-High speed on the SPI bus. The display writes are limited by the SPI bus. It requires a 9-bit word (additional bit is used for data/command). So being able to run the SPI bus at 48MHz is clutch.
-Has PMIC integrated into chipset. I don't need to include additional circuitry for a buck-boost regulator or a lipo charger since the dialog part has it all built in!
The other problem is making the tradeoff between detecting this fast enough (to turn on all the systems that follow so that the screen can light up fast enough) and waiting long enough (so you can be extra sure that the user intended to look at the screen)
The garden path is easy. Making it work mostly unobtrusively is very hard.
But man! I was so wrong, and I'm totally impressed.
The first photo looks a lot like Android Wear, so first I assumed this would be using a prepackaged OS, but it isn't!
The source code is in on Github alongside schematics, PCB files and everything else:
Maybe I'm wrong and the pictures do do the process justice?
Either way definitely amazing.
The person who made it probably made a lot of mistakes that you're not seeing, and is used to working at this level of quality.
You could imagine this whole thing started on a breadboard and IDE at some point, but the final assembly is "print the things that I drew on the computer, glue and screw together".
I've been thinking about making a custom keyboard for a while now but, while I have no problems with the electronic and software side of things, I'm a bit overwhelmed with the 3D modelling required.
I tried learning Blender but it really feels ill-suited for designing physical objects. The author mentions considering openSCAD but I've heard pretty bad things about it so I haven't really given it a chance yet.
The person making the dactyl keyboard had a very interesting talk where he explain how he basically ended up writing his own DSL in clojure on top of openSCAD, maybe I should try that: https://www.youtube.com/watch?v=uk3A41U0iO4 (if you enjoyed TFA you'll probably like this video as well, it's a step-by-step explanation of how he designed his dream keyboard).
Otherwise I suppose I'll have to betray rms and try this freeware Fusion 360 thing.
One can drop down to Python to do code-based parametric there also, with cadquery. But even as a software developer, I have only done that for 2 out of over 100 designs.
I have also maybe 2 times used the OpenSCAD importer to include some existing designs as parts into my project.
Bip and Stratos are the watchier lines, they also have more minimal Fitbit strap-style product that I forget the name of.
EDIT: Here's another new entry https://arstechnica.com/gadgets/2019/04/garmin-revamps-entir...
> Like most Garmin smartwatches, the Forerunner 45 and 45S track all-day activity and sleep, deliver smartphone alerts to your wrist, and are easy on the eyes thanks to an always-on, sunlight-friendly display. Even though the Forerunner 45 watches are the most affordable of the bunch, they still have a heart-rate monitor and built-in GPS, so users can make outdoor runs without the help of a smartphone.
> It can also connect to external sensors like a running dynamics pod and heart-rate chest straps to capture more running data while you train. The Forerunner 245 watches have the same one-week battery life as the Forerunner 45 watches do when in smartwatch mode, but they'll last up to 24 hours in GPS mode or six hours when using GPS and music playback simultaneously.
That looks really promising.
Its display is always on too, though the backlight isn't so if you need night-time tracking it won't meet your requirements but otherwise it will.
It isn't top-of-the-range either, or even a brand spanking new model at all (early 2016 release IIRC).
Things have stagnated a bit I'm sure, but mainly because we are hitting to plateaus: one caused by the limits of battery tech & what size/weight we are willing to carry, the other caused by there not being many features left to add, at least not that most users will be willing to pay extra for (or lose battery life in order to support).
If people started selling the pre-soldered board as a smart-watch starter kit on Aliexpress , pretty sure someone would buy it.
This is the sort of thing that would make DIY popular, I can see why.
Sidenote, I'm a bit disappointed that people use imgur to showcase their projects, as I don't think it's very good for that. I built https://www.makerfol.io/ for this exact purpose but people seem to still prefer imgur for some reason, and I'm not sure why :/
To be brutally honest: because it's more likely to still be there in x years, for any value of x.
Of course if you _really_ care, you host it yourself, and then it exists for as long as you care. Imgur's an easy happy middleground though, so people use it.
* Replacing "The user's avatar" at the top with the user's name.
* On the project list, just setting the alt-text to "".
Otherwise: nice site.
Alt-text is to describe non-decorative images that add meaning. The parent commenter is correct on how you would implement accessible alt-text here.
In the top, your avatar is meant to to represent a link to your profile. That needs an alt attribute. The alt attribute should be the current user's name, "Your Profile", or something similar. When signed out, something like "Sign In" might work better.
In the featured projects, the avatar that's there is decorative. The litmus test for decorative images is that if the image wasn't there, would you be able to easily figure out what to do? The image is purely decorative, as the creator's name is adjacent to it. There you would add an empty alt tag, like `alt=""` (not the same as no alt text), as screen-readers know to ignore that image then.
I highly recommend this page by WebAIM on it: https://webaim.org/techniques/alttext/
In fact, I've used imgur for years and I don't have an account.
I have been looking at this wiki to identify a few good replacements to pebble time: https://www.reddit.com/r/WearOS/wiki/watches
The TicWatch seems to manage 40 Hours, something which my recently Pebble Time with a half dead battery achieved.
The main reason I'm severely disappointed in the market is the lack of long battery lifetime devices. And those that do run for a week on a single charge usually lack in other parts of the product.
And after all that, it's all so expensive. A Pebble Time costs around 60€ to 80€ used on Ebay, and that is after they went out of production. Most competitors don't go that low!
...only part that had me confused was using like 25% of screen real estate for showing time and putting flowers on the other 75%. Artistic I guess.
I think in the Reddit thread, he mentioned having to draw each character (lower-case and upper-case) pixel-by-pixel for the display.
it's a nice project, but in my vocabulary 'from scratch' doesn't mean "after an initial engineering and design phase i'll order a host of parts from a company', it means 'i'd better start wrapping memory cores, let me get the copper wire out.'