RailsConf 2014 - Saving the World (Literally) with Ruby on Rails by Sean Marcia

SEAN MARCIA:Hi. My name is Sean Marcia.

I am gonna be talking about saving the world

with Ruby on Rails. And that's me on the Twitters.

At seanmarcia. In case you're wondering,

I'm an early adopter. That's how I was

able to get my own name.

I thought was funny.

So I, I work at George Mason University as

a software developer, and I'd like to thank
them

for giving me this opportunity to be here
and

do this kind of cool stuff. Thank all of

you and thanks to my Ruby group for letting

me practice this talk on them last week.

So today I'm gonna be talking about the, the

history of the project. Then I'm gonna talk
about

the technology we use. And then I'm gonna
hopefully

tell you how you guys can all get involved.

So, but really, I'm just gonna tell you a

story about bees. And my personal story with
bees

began when I was about six years old, and

for reference, that's me at six. Yeah. Looks
like

an orange on a toothpick.

But, but when I was six, my, my good

friend and I, we were, we were out, we

were wandering around, and we found a wild
beehive.

And, if you know six year old boys, you

can probably guess what we did next. We started

throwing rocks at it.

And, you know, my good friend, he just took

off running, and I didn't know why, and it

wasn't fun to throw rocks at the beehive without

him. And it turns out that the bees stung

him twenty, thirty times. And they didn't
sting me,

so. I kind of feel I have this karmic

debt to the bees for leaving me alone.

And so, you know let's jump forward to today.

And like I said, I'm a software developer
at

George Mason University. I was wandering across
campus one

day and I saw this guy was giving a

talk. This is Herman Pria, and he was giving

a talk about Amazonian stingerless bees, and
the honey

they collect.

And, you know, as a developer, I'm really
in

it for the swag, and since he was giving

honey samples of these bees, that's why I
went.

And so, I was listening to his talk, and

actually he told me what kind of animal that

is and I don't remember right now. But I'll

find out, because it's something interesting.

But, so I was listening to his talk, and

after the talk I went to, to talk to

Herman and, turns out he's doing all this
really

interesting stuff about bees. And you know,
bees are

fascinating. If you don't know anything about
them, like,

you know, there's 4,000 different varieties
of bees in

North America. If you guys didn't know that.

They're responsible for 90% of the wild, wild
plants,

and they're crazy hard workers. You know,
one bee,

in its lifetime, is gonna gather 1/12th of
a

teaspoon of honey. And that's all it's gonna
do.

But it's gonna visit like, 50,000 flowers
to do

that.

60-70% of all our food is because of bees.

Like, from the pollination they do, or they
pollinate

the food that our food eats.

But, the really crappy thing is, the bees
are

dying. And they're disappearing. And we don't
know why.

And this is actually starting to become big
news

and people are starting to cover it. Just
last

fall, Time Magazine ran an article about,
you know,

a world without bees and what the consequences
are

if we don't start doing something, and, which
directly

leads to what Herman is doing. He's researching
this

thing called Colony Collapse Disorder.

And if you're not familiar with Colony Collapse
Disorder,

basically what it is, is like, beehives and
bee

colonies will suddenly just collapse and disappear.
And for

no rhyme or reason. And it's, it's serious.
Like,

for managed beehives, like that's beehives
where there's like

a beekeeper watching them, about 35% of them
have

just died out and vanished.

For wild bees, it's much, much worse. In some

areas of the country, 90% of the wild bees

are just gone. Like, Virginia, where I'm from,
it's

about two-thirds. And it, like it's not just
our

problem either. Like, it's happening in Europe.
Like, they're

predicting in England, by 2018, all the bees
are

gonna be gone. And that's, that's really serious.

Asia, same, same situation. India, India,
it's particularly scary,

because in India, 90% of the pollination of
honey

comes from a wild bee that they can't domesticate,

and as, like, as we know, it hits these

wild bees harder. So if those bees get hit,

India's gonna be devastated.

And so, you know, some of the, some of

the theories about what's causing Colony Collapse
Disorder, maybe

it's pesticides. Maybe it's these two variety
of mites

that the bees for some reason can't, can't
clean

from themselves. Could be disease. Genetic
factors. Like, we

just don't know. And it could be a combination

of any of these.

I have one of my own theories up there

that, unfortunately, isn't gaining any traction
in the beekeeping

community. But, I'll let you guys try and
figure

out which one it is.

And so, hearing all this from Herman, it's
like,

hey, what can I do to help? And, and,

you know, Herman didn't, doesn't have any
technical skills,

and, and so, but he told me, hey, I

already have some beehives at George Mason,
and it

turns out they're up on top of a parking

garage, like.

And so up, five stories up, you can see

the top of another building nearby, and Herman
said,

you know, he would love insight into these
things.

Like, the hive temperature, because bees keep
the temperature

in the hive constant, year round. Doesn't
matter if

it's the middle of summer or middle of winter.

And, actually just in the last, last couple
of

years, they found out that, you know, when
they

pull these combs out of the beehive, and there's

the little holes in them, they always assume
that

the bees hadn't gotten around to filling them
with

honey.

But what they've learned is there's a special
drone

bee that breaks its wings off and crawls into

those holes and just vibrates all day long
to

generate heat. I guess that's a life.

But, so, and he also wanted hive weight, just

sort of to see the weight of the hive

changing over time. And then he wanted the
outside

humidity and temperature just kind of as control,
cause,

like, perhaps there'd be a week of really
bad

weather, and so, like, that's why the weight
wouldn't

go up. They'd be eating their stores.

And so, I was like, OK, so it, again,

looking at, at the situation, well, you know,
the,

immediately, you know, we see there's some
issues with

this. You know, like, it's open to the elements.

We don't have any power. Our budget was really

small. And we needed something that was really
easy

to repair and maintain because most of the
people

aren't very technical.

And so the first thing we handled was the

no power, like this was the low-hanging fruit,
and

we did that simply by, by some solar panels,

a deep cell battery, and we had some off-grid

power. And we found that our solar panels
and

battery could power our system for about seven
to

nine days, when there wasn't any sun. And
so

as long as we had sun at least once

a week, we were, we were good to go.

So the next thing was we had a limited

budget. And I, I think we started with less

than $200 or about $200, and this is how

things broke down, and, and why, you know,
we

used Raspberry Pis cause we got the first
three

free. Got a bunch of Raspberry Pi devices.
You

know, some SD cards, cables, like, who doesn't
have

a bunch of cables in their closet.

And you see the solar setup there was about

$140. That was about the bulk of our initial

costs. And you know, what I brought to the

table was this MacGiver-like ability to figure
out solutions

to problems. And, you know, like in the, an

example of that is, when we built out, when

we put in our first temperature probe, we
needed

to protect that probe somehow. And so, which
led

me to, you know, to quickly come up with

a solution.

I ran to Student Health services, asked them
if

I could get some condoms from them to cover

the probe, and they didn't even flinch. They
handed

over a bunch of condoms.

But it really made me realize that these,
these

people in Campus Health Services have seen
everything, because

when I walked in there, I was kind of

dirty, cause I was outside working, and I
was

carrying duct tape and rope. And, and they
didn't

even flinch. They just handed over a handful
of

condoms, so.

So, like, this wasn't perfect. But we had
a

start. And I plugged in Twitter so we could

get, you know, social media going, and some
Tweets

from the beehive. And so our, our initial
results

were, we had a Tweeting beehive. In the picture

you can't really see, and it's Tweeting hey,
it's

78 degrees in the hive. And the little picture

of the bee yard.

And, and we're going. And one of the interesting

things is once you're out on social media,
people,

I guess, they assume you know a lot about

bees and they start contacting you. And one
of

the people to contact us, send us a direct

message, was just like, hey, how do I get

my dog to stop eating bees?

You'd think eating the bees would probably
be enough

of a lesson. But, but you know, I, I,

I Googled, and it turns out lots of dogs

like to eat bees, and you know, little dogs,

medium-sized dogs. And even big dogs.

So, so before I go any further, who's, who's

familiar with this acronym?

OK. Some of the people here. This is something

started by Bryan Liles. Maybe five, six years
ago.

It means test all the fudging time. I always

get stuck up on the f. But you know,

no, no offense to Bryan, but he's wrong.

Really, what it should mean is try awesome
things,

forget testing. Because, if you, because if
you forget

testing, you can get results like this. It's
currently

501 degrees in the hive.

And you think, oh, why is that valuable? Well,

it turns out that people, people find stuff
like

this amusing. And suddenly, and, and you can't
read

this, but suddenly we're getting reTweeted.
City of Fairfax

is Tweeting us. And people are asking us,
are

we raising fire bees? It, it, is the hive

on fire? Are things OK?

And, and, it's kind of amazing. Like, all
these

Tweeting and reTweets and people getting in
on the

joke suddenly, suddenly was our on Gundam
style, just

making, make everyone aware what we're doing.
And it

really made things happen fast. And it brought
us

into the attention of the SweetVirginia foundation.

And what they are is they're a non-profit
that

teaches honey, honey bee education to, to
students and

adults. And, and they wanted to help. They
didn't

know we were raising bees at George Mason.
And

so they, they said, well, what can we do?

We can offer you space. And Herman wanted
to,

to, to have a class of researchers, and we

didn't have the equipment which was, you know,
one

of the biggest issues.

And so, so what we did, is we started

a crowd funding campaign, because it's gonna
be about

a thousand dollars per student. And we, we
raised

twelve thousand dollars, which was amazing,
so we could

have a class of twelve. And the, actually
there

was a waiting list of over a hundred students

trying to take part in this research.

But, so we had twelve, and then this, this

kickstarter campaign, sorry, Indiegogo campaign,
also was like, another

big social media raising awareness for us,
and so

suddenly, suddenly again, you know, people
are becoming aware,

and bees just become crazy popular at, at
George

Mason University.

And so the, the, one of the Fraternities is

starting a Bee Global campaign. The magazine,
the alumni

magazine is all about bees. And this is the

president of George Mason with Herman, there,
and, the

really amazing thing about the, you know,
the president

getting involved is suddenly we have a budget.

And so now, now, which leads us into the

technology and the interesting stuff.

So, for anyone working with Raspberry Pi or
is

unfamiliar with the Raspberry Pi, that's what
this is.

It's a credit-card sized computer. Like, this
is an

entire computer. And so I invite you to come

up after and look at it and play with

it. And if you want to come later, one

of the evenings, I'm probably gonna be hacking
on

it and playing with it, and so I invite

you to seek me out later.

But if you are gonna work with Raspberry Pis,

some advice I'd give you is to, is to

backup your card often. Like, take an image
of

it and don't, don't go the traditional dev-ops
route,

where you know you're gonna install everything
on there.

Cause if you've ever had to wait for, like,

Nokogiri to compile on your computer, Nokogiri
compiling on

a Raspberry Pi is about sixty times as long.

So it's better just to have an image and

then, you know, copy that image. And, and
buy

a couple, because they're cheap. They're $30
computers that

are full computers.

So Raspberry Pis have these GPIO pins on them,

which is general purpose input output, and
it's the

same kind of thing you have, if you've ever

opened up a desktop computer and look, looked
at

how your hard drive plugs into the motherboard.

It's the same kind of thing. And there's these

pins, and to reference, this is with one of

the sensors plugged into one of my Pis. You

plug it in and you may see some tutorials

online if you're gonna do some of the stuff

that says, that says solder the wires right
to

your Pi. Don't do that. Because if your, if

your Pi breaks or you don't know, it's easy

just to swap it out, take the Pi, throw

it away and plug it into the new one.

Or, I guess maybe not throw it away, cause

it might not be broke.

And so if you're gonna be working with Pis,

these are the three gems I'd recommend checking
out.

There's the gpio gem, you know. pi_piper and
wiringpi-ruby.

You know, I'm not gonna suggest which is best

of the three, cause they're all amazing and
it's,

I guess it's my own, Sophie's choice.

But, so, as to saving the world with Ruby

and Rails, I lied. I'm actually using Ruby
and

Sinatra. So I guess if you want to get

up and leave now, feel free.

So I, I went with the dashing gem. Originally

I was using, for creating a dashboard, I was

using the dashing Rails, but I, I realized,
I

don't need all the, the complexity that Rails
brings

to the table. Especially since we wanted the
solution

that was simple for other beekeepers elsewhere
to look

at and to use. And, you know, we don't

want to have to explain controllers and spitting
out

JSON and all these kind of, all this kind

of stuff that they just, they don't need.

And so, and so using, working with dashing
is

as simple as gem install dashing, and then
dashing

new. Much like using Rails. And if you are

interesting in more about dashing, I suggest
going to

RubyNation where Carl, Derante and Chris Mar?
are giving

a talk on it, cause their talk is phenomenal.

But after doing this, you have a dashboard
that

you can't see too well. But you can see

there's, we're using the green and the gold
cause

that's George Mason's colors. But we have
our internal

temperature, external temperature, internal
humidity and outside humidity.

And if the numbers seem low for the temperature,

that's cause we're using Celsius. We're, I
guess, cause,

we're not all on the imperial system. And
that's

the weight in kilograms. And, and what this
is,

is it's a, it's, it's being wirelessly served.
So

you walk up to the beehive, open up your

laptop, connect to the beehive, and then navigate
to

beehive dot local and then this pops up and

you can see the current statistics in the
hive.

And how did we do this? And we did

this simply with a series of chron jobs. And,

if you're unfamiliar with how chron works,
you, you,

the first one is running at zero and twelve

hours, so midnight, noon. Second one's running
at midnight,

six AM, noon, six PM, and, and so forth.

And, and that's it. Like, three, three simple
scripts

and it's gathering all our information.

And this is an example of one of our

scripts. So making a little, little, an instance
of

our sensor, defining a couple of directories
and, if

you see, directories are in the public, the
public

folder, and the reason, again is just to sort

of keep it simple. If one of the researchers

wants to, to get access to the CSE file

when they're logged onto beehive dot local,
they just

type in inhumid dot csv and it downloads to

their computer or their tablet or whatever
they're using.

and it's as simple as just writing the humidity

and the temperature into two different CSV
files and

that's it. And same thing for the outside.
And

so you're probably thinking well, it's hard
to get

it up on the dashboard. And, but, it's not.

Dashing makes this simple. Every thousand
seconds I'm, I'm

opening up the, the CSV file and then sending

it to the dashboard.

And it's as simple as that. It's, you know.

Nothing too complicated.

And this is the, the Python code, but I'm

not gonna explain. It was, came with the scale,

so I didn't bother redoing it in Ruby. But

now that our scale's having issues I'm going
to

work on a new solution with a RubyGem.

And the other thing we used is we used

Passenger. Just as for our server. We used
a

couple LInux packagages for our, basically
creating our wireless

access point. And the, a wireless, and DHCP
server.

And that was it.

This is simple. And I think it's so simple

that anyone here can do it. Anyone who's been

doing Ruby on Rails for at least, least a

week, maybe two, like there's nothing to it.
And

I want you all here to go out and

do stuff like this.

Because I think we're all amazing, and we,
we

sometimes, we sometimes get this sense in
our head

that, you know, we, we have all these problems

and we're living in this kind of ivory tower

that isn't really representative of, you know,
the rest

of the, rest of the country. Like, we're kind

of the silk on valet mindset. It's like oh,

I gotta get my queuing speed down from point

six milliseconds to point five eight milliseconds,
or I

have to, I have to get my tests to

run faster. Make DHH happy.

Or, you know, we have all these issues, but

you know the vast majority of people, like
I

work in the, in a university, and I see

every day people emailing spreadsheets back
and forth. They

email text files back and forth. And like
all

these solutions that are just horrible. And
really educators

they need our help. And they need us to

get involved in projects like this.

And it's super easy to get involved in these

kind of things. Like, I, I'm involved in three

projects right now and all of them I got

involved with just by going to see a Professor

talk and then after his talk or her talk

I, you know, I just asked them questions.

Cause, you know, professors, professors have
bigger egos than

we do, as, as developers. They love to talk

about themselves. They do. And they love to
talk

about their research. Like even more than
we like

to talk about our code.

And so I know, I know you're thinking Sean

you dirty Canadian socialist, you know, like.
Why, why

should I do this, you know? I don't want

to give my time for free. And I guess

like the simplest reason is, is I run a

Ruby meetup group, and one of the first questions

I'm, I'm generally asked my new people is,
you

know, how can I get involved in opensource?
Like,

like, cause, it, opensource seems like this,
this goal

that you want. But you just don't know how

to get to, if you're new.

And And, you know, doing projects that like,
projects

like this, it really is a, is an easy

gateway into opensource, and to, and to, you
know,

getting opensource credentials on your, on
your resume and

on your github. Because, you know, like it
or

not, when we do apply for jobs, they do

look at what we do in the opensource community.

And can't really see this, but this is one

of the, the extra benefits of, of doing a

project like this, is, is all the puns. And

it's like, you know, you can't really see
these

but I gave this talk last week and every

one of these are bee puns, like, hey, buzz

off pal! You know. Sean has a tendency to

wax on, and, what's the buzz about, and ooh

it's gonna be sweet. It's the bee's knees!

And so, so if this is the kind of

thing that interests you and, and you might
want

to go out and get involved in, I'd really

like you to check out Ruby for Good, and

it's a conference we're gonna be putting on
in

the Washington, DC area. The first to third
of

August. We're gonna be staying in the Dorms
at

the University. We're gonna hack on opensource
social good

projects.

It's gonna be 72 people. We're aiming for
$200

or less, and that's all inclusive of your
lodging,

your food, everything. You just have to get
their.

You're welcome to come a day early for no

extra cost and we're gonna probably do something
fun

the day before. Maybe go do a nighttime tour

of the DC monuments or something. Not quite
sure.

And, and also maybe to make it more appealing

for your work, we're gonna hold, we're gonna
have

some training sessions on the second day.
We have

one of the guys from the RSpec core team

coming out, gonna give a workshop on RSpec.
We

have a guy giving a workshop on Angular. And

another one of the JSON API committers doing
a

workshop on creating APIs.

And that's my talk in a nutshell. Thank you

so much for your time and are there any

questions?