My name's Conor. I work in the City Council as a land use person, but I also like to do this kind of mapping
and data science analysis in my free time. And I try to apply it to my work as well. But i'm trying to approach this as a hobbyist,
and and you know, bear with me as we move
through this. I'm my! You know my my biggest goal here is to encourage other folks To you know. Take that first
step in the data analysis, sign of things. If you don't feel super
comfortable with coding or with, you know, you have questions about the open data,
but you're not really sure where you started.
We're gonna work through an example That I've worked on in the
past.
That kind of edited it to the to this session that
looks at a portable housing. And so let's start. You know. I try to target folks who like,
I said, or maybe novice probably is interested in this kind of data,
and how It impact their lives,
or or the the people who live in the city. You know, conceptualizing your questions how
to answer them. I always start with. You know,
to something i'm curious about. See if I can get an answer out of it out of a little
bit of working with the data And get a basic understanding
of how the data and our as the programming language
work and how you can use them in your own hobbies and interest.
And then,
to most more like most importantly,
it's just gain a sense of confidence about Like what you're able to do. I know
that for me that was the biggest barrier which is thinking that I could do this kind
of analysis on my own, or answer these questions that I have. If you do want to code along,
I and one of the slides will give you like a short, fitly link to a Github
page where you can just see the code. You know. I'm not going to be
typing anything out as we go, because I don't trust myself enough,
but we're just going to be looking line by line and see.
You know exactly what everything that every step I take like,
how it helps get to our men product. So if you do have our as the language and arts video as the software downloaded on your computer. And that's great. If not, we're just following along. And hopefully you can use that link later to To you know practice when you get home.
And he doesn't even interest. And that's it very low. So a bit about our of our studio, so our in the programming language that
was created for specifically the Cisco analysis and
data find a lot of links. Love languages do. Other things are really good at this. It has a little bit of of visualize
visualization programs as well, which will be using it's an object or into language.
And the way I understand that is A lot of the code involves
assigning things to an object And then manipulate,
changing the object, and then eventually, maybe looking at it.
That
object would be. It can be a visualization,
you know. It can be most. It's mostly gonna be a data frame, you know,
just like X and y like you see on Excel. And It. It consists of base art,
which is sort of the fundamental package and library of the of the coding
language. It has basic arithmetic, basic syntax to actually be a
functioning language. But then all sorts of people have created their
own unique libraries and packages To build on a web base. Our started. And then our studio is the
integrated Development environment, which basically means it's just
how you interact with the language. It's really helpful. I found it to be
a really clean and easy software. Program to code in our with it
even tells you when you mess up. Which is great, which happens a lot. And you you can track all the
objects you can track your files, and You can even do the spreadsheets in in our studio,
which will be doing My very simple mythology,
which is just that it's. It's a very common methodology for data science we
have.
We start with that question. We pull the data we need How we clean it up so we can
work with it. Then we work with it, and then you can map it or
view it. and then analyze your results. And we can see, you know,
if we've answered our question if we need to go back to one
of these steps and try it again. So it's a very straightforward methodology,
but I think it's helpful reminder of what
this process really looks like. And to our question or my question. What is the affordable round for each
neighborhood population? And how does that compare to the rents of affordable housing
build are preserved in that neighborhood. or Basically how affordable is a city's
uportal housing to its local community. And again. I think I want a disclaimer,
You know this is just one way of looking
at the data. I'm not saying what's happening is good
or bad. I'm just saying that this is something that I
feel like it's often overlooked.
You know,
in my job you get a lot of people coming in who can't afford to live in the
neighborhood they grew up in. And they ask for help, and we it. It
can be really difficult, because even the affordable housing that's being built in
the neighborhood is not affordable to that. And so this just looks at that. How You know how that varies across the city in
the city. Some neighborhoods, with a mostly wealthier population, might have some
affordable housing that they could easily afford. But some lower income. Neighborhoods might have a bunch of a portal housing. But if it's not hitting the incomes of those of that population, then that's maybe something to think about. So this is the fifth meeting. If you want to just take a set, I'm like,
yeah, I think this is the easiest way to do it. It should pop up as a Github page with just all the code Github. It's a couple because it keeps the the formatting with the code as as opposed to like a Google Doc, so I just leave it on the Screen for a second, and then we'll look
at the data that the data that we'll be using.
And then we will dive into our studio. And and I think you know you might
have some time for questions at the end. but feel free to just interrupt me as well. If
you have any questions about anything. No, it is he sensitive,
I am. I'm not sure, actually. Okay. Thank you. And again, like. I am not a software developer like I
am not. I didn't study computer science.
So i'm not saying that my code is the
best code. It is messy, and but it works. And so I think that's also
part of the learning Curve is understanding how to make
your code work for what you need it to do,
and make sure that you can come back to it and not say like,
what is this with with what is going on? So with that We will go over to The data set so hopefully, some of us are here this page. It's the open data portal that we all love. And so what I will do oftentimes is, I'll just start looking through the data sets are available. Which is some keywords. So i'll put in affordable housing and just see what pops up.
You know there's tons of data that I that You can just come across on
this data set on this on open data. And so I encourage you to
look up whatever you find to be interesting. But this is
the data set that I've used For our project today. Just looking at it. It's started being tracked in 2,014 when we are. De
Blasio started his housing plan Because a lot of the income restricted housing that the city builds and records started in that time. and it's continued into the present day because the current mayor
has continued that policy. And again. What they. This is the kind of the data
they use to say. Oh, we built this many units of affordable housing, where we preserved
this many units of affordable housing.
Preserving,
meaning. It can be exist in housing that they continue affordability for,
or they implement new affordability measures on existing housing,
and then built the new construction of income,
restricted housing. And so, if we look,
we can actually just view the data or look at some stats. got a lot of views.
a lot of rows and 41 columns. And so if we actually want to
look at what those columns are.
We we have some, you know,
project identifiers some basic address information. And they actually get into. Because that's how we're gonna
find out where that building is, what neighborhood it's in. And
then they have this breakdown of The units. And these these terms, extremely low,
very low, correspond to income levels. And I know that you know extremely low
Income is 0 and 30% of area meeting account. And if you don't know what area needed income is, it's a term that the city uses to measure. It's a portable housing. It actually includes all 5 boroughs, as well as a few counties outside of the city. and they take that number, and then they take a percentage of that income, and then that'll be The income limit of a unit. So if you're building a unit, that's 30% of the air you need income. It's 30% of whatever that
meeting income is for the whole city and Westchester, and I think another county. And so that can be kind of
skewed right that number, I think, right now for a family of 4 or family 3. It's well over a $100,000. So what that means, you know,
for for a lot of people that's not a very accurate estimate of what
people are making realistically in a lot of areas.
So that's kind of the
the impetus of this whole project was to be able to map how that
policy kind of mismatches the need. Do you repeat that policy? So the policy being it's actually regulated by the Federal Government. We are allotted an area by which they
measure the income. and so the area that they chose is the 5 Brows plus Westchester County,
and maybe Suffolk County. Which are wealthier neighborhoods than the 6. And so what happens is you get this huge number Of what is affordable,
or what is the average New Yorker making? And so when that number is
around like a $130,000 a year.
That's what they base and anything
under that they basically consider affordable. right. And so if you're
making $90,000 a year as a family of 3, You can be eligible for a portal
apartment again. Everyone deserves housing everyone. It shouldn't,
you know, like there's a lot of Perspectives on this,
but it often leaves out people who are maybe making 15 to 30,000 a
year. There can be a portable of housing partners surrounding them,
but they can. Still, they still cannot afford. That meant,
because that's That's how like the system
sort of faces it's a portability. And so that's what I wanted to look
at is comparing people's real incomes To the income you have to make to
afford one of these affordable units. So we can just get right into
it. Are there any questions at this point about the data. not
where we're. We're headed. Yes. What should I? Oh, yeah, do you want to turn your microphone? Oh, yeah, do you want to turn your microphone? sorry.
The question that I wanted to Is the issue of the Which are really, I believe, glamorize 82, And Using this formula To validate what it is they're doing. And these are, You know, the people who are along with the politicians who have been allowing this. I don't know if you are going to touch on Apartment size and The price of these apartments,
or the whole philosophy behind that,
because people so desperate now in New York City It'll take almost anything and like,
I said to you. from what I can see,
our senior citizens are being pushed Into the very tiny. As I said to me,
prison cell size of apartments. Oh, sure, I think you raised a really
important point. I think there's a lot of Issues around the way that. Like what you talked about. It's
not just about the rent, but the quality of the housing for the for the
residents who are living there.
I think that Not just yet like for a square footage
of an apartment, but also a number of bedrooms that are available. The
the unit mix of a building right because you can have an affordable
building affordable housing building, but maybe, like the majority, are studio apartments,
so that already cuts families out of the picture. I'm: so mostly talking
about the income Mismatch. But that's a like that's an example
of like another research question to be able to answer with similar data,
right? Because we have right here The bedroom size of each unit. So you
could do the same thing. You could measure average family size of that neighborhood
and compare it to the average room, size Or the average unit size that they're giving that neighborhood right. There could be a huge mismatch. You could have a bunch of studios while, like the the maybe the average family is 3 people. And it's the same thing with seniors. Yeah. And so yeah,
there's a lot to touch on.
And I think that those are examples of great
research questions. I don't know. I'm going to be touching on it specifically today. Just another good question.
Were you able to look at whether Hpd is really part of this this
year in terms of apartment size. And These buildings that are are
being built along the city plan. But that's a whole separate yeah,
I mean, I think Well, this these buildings are all under HD.
Supervision. They usually like, receive some sort of subsidy,
and I in in in exchange for that. That's when they implement
affordability regulations that I mean, there's a host of policies
that results in the creation of these units. I can't exactly what leads to what,
but they're usually partnerships between
a developer and a city agency. Any questions? Yes. thank you. Yeah, just quickly. The rent. How is the rent calculated, based on? So That's a good question. I
believe the way that they do it, and the way that i'm there's kind
of a common metric that people use where it's.
30% of your
income goes for your rent, and if you're above that,
then you're considered to be burdened. And if you're not, then that's considered,
you can afford your rent. And so what I was looking at is they actually have in
here. They have the the income levels. And so you can either compare income to Intel,
or you can prepare rent to rent right like what an actual portable rent
is to the rent that they're providing, or you can compare an actual income to the
income. They're assuming that unit Holder is gonna have. But generally people
consider they measure it as 30% of your income Per per month.
Okay, so We'll switch into the phone. And you know, I think, yeah,
that, I think is raised a lot of imported pieces about like the context here,
right of like the policy and how the city decides to do
things. And I don't want to know. You know I don't want to, necessarily,
you know. Call anyone out In particular and reiterate that, you know this is Like why this sort of open
data access so useful right is that you can kind of just then use
this data to decide for yourselves What what what people are doing,
right, what people could approve on. And kind of help. Educate
yourself on the issue. That's that's how I've learned so
much about.
This is through Access to this kind of information. So to get started we have, like I kind of tried to break it up into the at the
elements of the methodology. This is our studio. You have 4 windows you have. Like your coding window. Your console, which is kind of shows all
the processes that you're doing some files that you're accessing, and then this is
where all the objects are going to pop up as we create these objects that we're going
to be working with.
And again, I thought it might be a little confusing. But when I
say an object, I mean, usually a spreadsheet Right like a spreadsheet file that you
would find on your computer. As we're creating them or manipulating them,
they'll populate in the environment. So the first thing you have
to do is install the packages that we're going to be using.
The first is called tidy verse. It's kind of a family of
libraries all about data science for our. It includes a ton
of functions that you can use to add columns,
to remove columns, to just manipulate the data.
How you need to, how you need to use it. Shiny and shining themes and Chinese dashboard are the mapping or sorry, not the mapping, but the The right, the visual, the it's a way to host your own little web page of the
visualizations that you've created. Leaflet is the mapping
package. So it helps you create these nice,
colorful maps that interpret the data The way you want them to. As that Stand for simple features, I believe,
are spatial features, and it's another spatial data package that we use a few
of those functions in in this coding session To help manipulate some of that
data we excel. We use one excel spreadsheet, which is the census data,
because that's separate right, the affordable housing data set
that we're using doesn't talk about income of that neighborhood,
so we need to pull that separately And that package.
Lets us do that.
And then our Socrata is the package that helps us take the open data that we
just looked at and import it into our studio. So those are just sort of the
general purposes of the packages. And For after you install them you have to basically set them up for your environment. You have to say, I want to use this from this one, this one, this one, this one, this one, and this one. And then we're ready to go.
We have all our packages ready. We haven't, created anything or change
anything. So the environment is still empty. But the first thing we're going to do is gather all the data that we're going to be using. So we're pulling the data. The first thing we need are called Ntas, which are neighborhood tabulation areas. They're the closest thing we have to a neighborhood geography
in the city. So that's gonna be In our final map Gonna show each neighborhood. And the
the next thing we're gonna have to do is aside the data that we're trying to look at to
those neighborhoods, so we'll do that in a bit.
So first we have to Import that,
and then you'll see a pop up in the environment. And today,
so there's 262 neighborhoods in this data set. They have pulled variables, the variables which are like columns just say
basic information like the name, if they have any kind of code identifier. But it doesn't say anything about what we want yet. We just we just need that spatial information We're going to only select a few Of the columns that we care about. We
don't need all 12 columns. So to do that we're gonna run this next function.
Select Nta 2,020 andta name and geometry. Those are the 3 columns that we care about,
which are the individual identifiers of those neighborhoods,
as well as the geometrical it the geometry,
information of those of those neighborhoods.
So like what shape are they? Where in the world,
are they? Next. we're going to pull
in our affordable projects. Which is the data set that we looked at.
I'm going to call it affordable projects data Just because that works for me like there's really. I believe,
like there's no way to you. Don't have to call anything anything else
I called Ntas's affordable project. Data sounded right to me. But when you're working through this you
can call whatever you want. And this reads the crowd of function
like I mentioned from our Socrata basically Goes to the website we were at. takes
the data and brings it into our environment. Now, you usually you need a token
and an email and password which is mine.
You can use mine, I you don't. I I guess I would recommend you don't use mine forever, but you can use it for that. It's very low base again. They don't really care. I don't really care. So you need to say who you are and who's pulling it So to do that we just Go down. And what i'm doing right now, Casino, is It's like a control? Enter which
just runs that element of the script. And anytime there's a new element. It'll stop it only does one at a time. And so I've tried to break it
up into sections here to kind of make it clear what we're
doing. But you just do one thing at a time,
and it's helpful. It's. It's in the particle that way.
Some languages just work
bridges everything all at once. But I like our because you kind
of just do 1 one action at a time. Are there any? Is this? Are
there any questions in this book? Okay, yes, really quick check for the one
I've for the dependencies to download for like 15 min. Is that normal? Yeah.
So the dependencies can take a long time. So especially tidy verse,
because it's okay. But what we can do is you have the code right,
so if it does Finish up you'll be able to catch up pretty quickly,
just because You it's all there for you.
But I guess, like we'll just keep moving. Just so I
can explain what everything is, and then you can run it on your computer, maybe at the end. Does that work. Okay, Great thanks. So we've downloaded the the neighborhood
shapes. We downloaded the affordable project Spreadsheet That we already worked at. And
now this download file function is going to download the census
data that we the income data. Which is the last remaining, because the data that we don't have from that one spread that we already looked at. So do not be download, and it actually just downloads the file on your computer. And then we upload it Into this into our studio. And again, whenever we have A piece of code and you press control, enter it runs that code. So now we'll assign it as Acs data Which again just works for me Sometimes. That's where it gets kind of
messy when you have to have a bunch of different names for things. But I think everyone
kind of figures out their own system that works.
So now we have all the raw data. And now we have to start Cleaning the data and making it
look like we want it to look before we actually start that same with it.
And a lot of that can be a little tedious. Especially since there's a few
different data types that you have to work with. The main ones are
character data and numeric data. All of this this big paragraph here
is just changing what our numbers and just telling the computer that
their numbers and they are not words Right, because it can only run
map on numbers That makes sense, but it's basically you'll see it
just says as numeric as numeric. All of these unit numbers are numbers,
and you're just making sure that the computer
knows that the numbers. But first we are going to create this,
a portable rent Column, which is what we're talking.
About. 30% of your income. and that's literally what that map is. Right here.
I'll play it right here. If you see this. I'm literally saying that this is the income column from the census data, and you're dividing it by 12 And multiplying it by point
3.
So you're just saying 30 of your monthly of your monthly
income is what you can for. So over there it is. And then we'll run this big chunk,
which is just cleaning the data which is making sure that
the data is formatted properly. And and this filter is in a
longitude. What that's saying is some of these, you know there's going
to be errors in any data that you work with, and some of them are
missing. They want to to latitude data which we need,
and if we try to map that it's going to send error. So we're just gonna remove the ones that don't have a longitude because we can't work with that. And we're going to turn it. Do that. We are going to attach the
Nta data which has the shape file information. I mean, the yeah, sorry the be it Geometry information for
the neighborhood that it's in.
Because we don't have
that right. This data set as it just comes as affordable
housing projects. Is it just has where that building is. But we need to know what neighborhood it's in,
and what that neighborhood looks like. The will run that Affordable projects by Nta.
This one is basically them saying How many units. And how and putting all that
information and converting it to the neighborhood. So after we run this. I'll run this, and then we can take a look at it. Yes. This one. So before We have this right, each row. So this is like a regular spreadsheet,
which is why I look Mark studio. Each row is a development. right
each row it it shows you its name. The project start date, and then this is like
some of the information that we saw before right. The income units a lot of it's 0,
because they only have these middle income units Which will come up in the now.
And it's just every problem
is a building. But what we're eventually going to need is how
every road be a neighborhood Because we're mapping the
neighborhood. They're not mapping the buildings we're seeing
how the neighborhoods compare To the So that's what this is. The most recent Function that we ran created this
data set. which is every neighborhood. And it's got that unit information in it. right? So it took the unit information for
each building if we figured out which neighborhood it was in,
and it summed all of those unit accounts up. And so we're kind of creating
new data sets that Don't exist on their own. But we need
them for this mapping project. And we have our geometry information
that's those are the geometry.
That's the geometry information for the
neighborhood. right? So it's like the shape of Dream Point. And where in the world that shape? Is Any questions at this point? After that we just have to
do a little bit more configuring. because, as you can see,
we're slowly getting to a data set that we can map right. And so we just
have to keep manipulating the data until we can have something
that has all the information we need Her neighborhood. And then
we can actually start looking at it. To do that, we're going to create a final
data set that i'm called Nta affordability. And what that is, is, it's the income data. the affordable rents data. the affordable projects data All by neighborhood right? Because
that was the whole point of this of all this maneuvering is having a a
spreadsheet that showed every neighborhood It's income level. It's a
what would be affordable to that income level,
and then the affordability of the units That were built there right? So We'll run these functions to.
Essentially,
I think I don't need to go. You know. I want to make sure
that we can actually start at the app. But all of these do is basically get there, and You'll find that sometimes you have to. It's
a very round about way of getting there. but Sometimes you have to remove things
a lot of trial error. It's a lot of Googling. But you'll eventually get as long as
you know that that end goal is that data.
Set that data frame that you
can. That has everything you need. All of these little steps are
just helping you get there Total units, just adding some
problems that we didn't have before. And then this this one's important. Just because this is my logic for saying whether or not you can afford an apartment. So i'll just I'll just look at this. I it's not super clean, because i'm. But what it's saying Is. You have these levels of these income levels,
right? Like, I said, 30% of the area meeting income 40,
and that's not really how people's income works. Your income can be anywhere in that range. But what I'm saying what
this is basically saying is. Yeah. The affordable rent to that income is greater than this number. Which is one of those AI levels. then you can count it toward what's the portable to that neighborhood? If it's greater than this number,
which is a lower. Am I level? Then you can count those units as well. And then so on and so forth,
until you get to the lowest.
Am I percentage level
that those units are built at. Does that make sense so like
this will be the affordable rent To some of those units. That's a high number,
I think, $30,000 a month, 3,600, But it's the affordable right in the nearby,
if it's a wealthy neighborhood where
the average in Ken. That's that's Less than 30% of their income. Then, in terms, then, that's an affordable unit for that for that. And so on and so forth,
you'll see that this number. which I would say is. I think, about 30% in my this is there's been. This is These units are going to be a lot more important to
a lot more neighborhoods.
And that's what's going to show up in the map. But just to explain like this is an important section, because this we're actually defining what's a portable 20 over 100 and what's not? Does that make sense? Any questions about that? Okay. This is just again Rounding some numbers. converting percentages to look
what we want them to look like And a little bit more cleaning.
Making sure that everything's formatted correctly. And finally, we can actually start mapping the data. Because now we can actually take a
look at this nta affordability spreadsheet and see if it has everything that we need to
have. Let me take a second. Okay, great. So First column. this is the neighborhood. This is like every neighborhood has this. Id number. The borough it's in. This is the income, app average meeting,
or Median household income of that neighborhood. and then, therefore,
the affordable rent, which is 30 of this per month. And then extremely low. Income units
very low income units, low income units. What is this this one's like? Oh,
middle! What this was middle income. This one's Something in between those 2,
and then the total number of units. We have an Nta name again,
just because sometimes you have duplicates,
and then this is the shape of the neighborhood. So that's everything we need.
We can compare This rent To to these units right? And so You see here. You just go back and it could miss a few calls. Okay. But Okay, let's just start with the data. And so the next. The next step is basically Using this shiny package that we that we
installed at the beginning. The sunny package Does like front end, back end the whole thing to basically host what you
mapped onto a website. We're not going to be putting it on a website today, but we're gonna look at what
it looks like to design that That structure. You have the user interface and the server. Those are just the 2 objects that you create as a part of That Yeah. Or a part of that website, the first one You're just creating this page blue
page it just when you say fluid page just means that it will expand and
shrink as you move the page around.
We're putting a map in there, and we're
giving it the height of 700, because that's what fits the window. And then this server is sort of the back end.
It's like the functionality of Yeah, that's right and to say. And so what we're doing is we're doing a color. We're doing like a caller. Coordinated map. Different shape of purple. That's
kind of how I decided to do it where the darker the purple,
the more affordable the units are to that area. And then this output map is basically
the functionality of the map that we made. These are all basically just
inputs of what I want the app to do. We're using this nt affordability. App.
We're adding a base map which is just It's like a it's like Google maps or something that you see on any kind of on my map. It just shows What the world looks like, and then
we'll add our purple color data on top of it.
Where where we look at you just shows
where we're looking, and then we're going to actually add the neighborhood to
the color of the affordability. And then So this just kind of sense. You know what those
neighborhoods look like. Is there a order? What's the color
gonna be like? How bright do you want it, like? All of this is just all of these settings. And then what I added is a little. Pop up,
menu That when you click on it
it tells you more information about that. Everyone. The information that's in the data set that we just created. The first thing is the name of the
name We in household income Number of units that the city says are
portable in that area, and the percentage Of those units that are actually
affordable to that neighborhood.
You're gonna. And a legend that's just the bottom that just tells you What you're looking at Just gonna give it a title. And if there's no portal housing,
what is it going to say? And then that's basically it.
We can run this. I just want to make sure. First,
I'm going to double-check that everything is Install properly so i'm just gonna clear. I'm just gonna run this again.
So I
might have something. But What we're gonna do is see if the
map turned out the way we wanted it to. What we're gonna do is see if the
map turned out the way we wanted it to. Okay, great. And this is this is in
product right like I'm. When I first made one of these I was like,
Wow! That looks pretty Cool, I think that, you know. And then you know A lot what's right about our and shiny,
and the the the packages that we use is a
lot of this. We didn't really. Do you know we told it. We
told it to put this base map on, but we didn't typed out Valley stream,
or we didn't like. You know. Tell you where
the water was. A lot of it's already built in,
so it makes it a lot easier for folks To create a map like this in
whatever it was like 90 lines of code as you can,
and then we can just start exploring it.
The gray obviously no affordable housing recorded. So these these
neighborhoods. Haven't even Build affordable housing,
let alone Whether or not it's Affordable to that neighborhood right These white or purple Our neighborhoods where they built
the portal housing, but compared to the what people are not even really can afford,
it's not really affordable Right. And this we're not even looking
at market rate or luxury housing. Right? We're we're just looking at units of
housing that the city calls affordable. But we do have some darker
areas where you know we can zoom in to like a wealthier
neighborhood. and if we click on it. This is the pop of I was talking
about financial district, better park, city Meeting household income, $183,000. They built 190 units of form of housing,
and All of them are affordable to this person And that makes sense. Right? That's
a high income. So that's great. They build a 190 units that are portable to
those residents.
But right next door. We have Chinatown and 2 bridges For the meeting. Household
income is $27,000 they built far more affordable housing units right. But the city says,
Look, we built 2,600 formal units in this neighborhood
for preserves 2,600 units in the same of my. But For this into order. 14%
are actually a portable right? And so like you can do it
that way you will. But like That, that means something,
and you know no, and but like the the sort of like Cherry on top
is that every one of these units Is a lottery unit, right? And for every
unit there is at least a 1,000 applicants. So It's like there's levels of scarcity here, right? And I think that this is just one perspective
on the reality we face with housing.
But like this isn't something that the
city is recording right? They're recording 2,600 units. But we can see just off of 90 lines of
code and data that any of us can look at That only a small fraction of those can people in that neighborhood move into right. So like. This is where I feel like
it. Kind of becomes real is When you know, in my in my job like this is our district. So like. We have people who live in that neighborhood,
and they lived there for generations. and you know they're getting
evicted something's happening. And they hear about these buildings
going up. and we say, you know The city is not gonna bar you
from paying 70% of your rent of your income for your rent.
But that's not sustainable,
so they could apply. They have a one in a 1,000 chance of
getting in. and then they're heavily red burdened. So it's like that's kind of
like what made me interested in this data And interested in, you know,
doing this talk and everything but That's kind of what we're able to show, with just a little bit of coding knowledge, in my opinion. And so you can just look throughout this map. You know. Let's look at
Jackson heights $62,000 a year. They built only 59 recorded units. and only,
I guess 6 of those units are affordable to Jackson. Right. You can look at. But then, you know, on my side. $127,000 a year, they built a lot of units and they're 100% affordable to that paper head. So, anyway, Pat. you know,
we can keep looking at this, but Sorry That there's still a lot of like the
grade. I mean? That's okay. But You know Richmond Hill $72,000, but they they haven't
built a portal housing there.
We're preserved in total housing in
there. So there's a lot of gaps. And you know it's it's interesting.
Can you have like areas That are not like necessarily low income for the average income burner.
but you still 0 of those units. Our affordable. It's just,
you know it. Just kind of you find things that you
wouldn't necessarily expect. And so, just like now that we
know what it looks like. I hope that You know this isn't too
daunting I of of of of a project to kind of take on yourself for
whatever data you find interesting. And hopefully,
some of this makes sense in terms of what each of these lines does,
and how it leads you to that final product. But You know, I think,
that once you get past that point, and I don't know when everyone's experience
level is in this room. But for me. Once you get past that point of saying,
you know I can do this. I can have these things on my own.
You
know it opens up a lot of opportunity to To kind of showcase some of that information. So that's going to be my presentation. Hopefully,
we can have a little bit of discussion if
people have like takeaways or Questions or thoughts about you know
what you might want to be able to visualize Anything else. So thank you. Any questions. Yeah. Just because it was really killing me. Moi
is honoring. Yes, yes, thank you. Moderate income. Yeah. Between low income and
middle income. They call it moderate income. Okay. Okay. okay. wait. Which question. Okay. We got a couple. We look here. So it's interesting. This is the thing.
You know some neighborhoods. You know the the meeting kind of says $65,000, A good number of recorded units And for that in order.
89% of those units
are portable. But again, that's the median income, right? So there's a host of income. So the people who need housing are you the most Are still missing this. But I I can't. You know. There's limitations to
the project right like I can't Say every I don't know exactly.
Everyone's income. The only, you know, reliable income number
that I can use is the median intent. And so it really just it. It's really kind of scratches the surface. But again the meeting is in the middle. So there's half of the population of that neighborhood makes less than that. So that's Harlem, South. Can we do the
rest. Of the Yeah, Exactly. So like you have 57% East Harlow. Maybe some of the most housing
units built or preserved on the neighborhood. And only 20% Are portable. I know there's been okay. So yeah, cell phones has seen a lot of development recently. They reported 25,000 years, and this is the plan That was initially.
You know,
created by De Blasio housing New York. It it it was a number right.
I believe it was 300,000 units. By 2026, And that was the policy With building or preserving 300,000 units,
526; and so that kind of creates an incentive To to report as many units as you can,
because you want to hit that number. So if your if your strategy is quantitative From the beginning. I feel like you're almost starting on the
wrong foot, and you're kind of manifesting this mismatch because you want to hit that
number. You You don't want to say you failed. Your like your policy failed,
so they'll record any number, any unit they can. And So they have they'll just say 25,000 units in in my habit. but like who's who can? Who can live there? And I know that, you know I I feel like I keep interesting because I don't want to like.
Just in case. But you know there it's so.
It's a complex issue. I don't want to like skew the way that some i'm trying to show that some data,
Steve. I don't want to do the same. I'm not saying that you
know lots of people can move into neighborhoods,
and I think that a big, you know There's a lot of policies that are about
addressing segregation and everything. But I think there is something to say about a neighborhood of forwarding its own housing, right? And so people are filling those units,
and they're they're meeting People's needs some, you know, some people. but it just.
you know. Another thing to think consider Is like. What about the people
down the street. especially in neighborhoods where housing
and security is a serious issue. And then you're building the Charging the department. you know. When they Crazy. And the Uber came
in in this mindset of raising Downtown, working and putting up a condominium All over the place, I mean.
I just don't know where they expected the New York
is that when they have to go I don't understand same thing in right? Well,
because I I mean, yeah, I think it's like These are some of the things that we can kind of Back up this kind of research,
and I think that that's what's so interesting about it is like you. Can. You can show people
this like you can show what you've done And and you can't.
And this is the date.
This is data coming from this city. So you know what is. What are you gonna say like our
data is wrong. No. So you know. And I think that You have another question. And you have Nina. Yeah,
i'm just gonna say like it's a secret thing, and I definitely have ideas of what I would want to do. That's like like. Why, why use the
neighborhood level, maybe to the immediate. I think it's important to note that If I'm. Understanding correctly.
it's not even fully show. The situation Is this is really just showing an
assessment of the cities attempt at forcing developers, but affordable housing.
It's not actually what the actual rents are In the thousands of buildings that
already is this. So I would love to like on top of this to do how the rents are
increasing in the non new development. Because I think this yeah,
this is a great example of how my the affordable housing,
and they force developers to do, and not But the rents are all for low enough, so That's something I would love.
And then I have one more question, and I can can't remember it. But Yeah, this is oh, yeah,
Are you aware of any that birds Is city, government, or State government,
any legislate in any talking at all? Any policy income level from city wide,
new, any smaller taxes. Is there any talk about that. Yeah, that's a good question. Well,
to your first part that's like amazing, because that was the whole point,
right? It's to like spark New ideas for how you can
build on this or answer newer your own questions. So
that's great to hear. And then. In terms of you know,
adjusting that how we measure area needed income. I think there is a lot of debate
around that. I think people Always kind of kick it because it's. It's a it's a Federal Government thing. It's it's a high Jurisdiction, almost so.
People always say,
like, oh, that's out of our control over there. There are,
like some, I think,
actual question new questions that raise with local AI. But I think what I kind of
come to this realization that You can,
whatever the benchmark is. That's fine, but it depends on it's it's what you do with it, right? If you say, okay, this geography. the the
area needed income is $130,000. Okay, that's fine. So we'll just only build
0 to 30 of that. right? Because, like You're doing a percentage of that number. So that number could be any number.
But your percentages have to Equal what people need. Right? So it's like you as long as you're not counting that as a portable. And you're just focusing on What people actually need. If
that's 12 of that number, then sure.
But just go with that number.
You know what I mean. So it's like almost not even about there. You need income,
but it's about what percentage of that you're
targeting in your policy. I think what percentage of that you're
targeting in your policy. I think What we we were asking for, and then my just for Harlem and the we saw we saw what the problem was. You cannot bring in
Westchester and all this other stuff Into Harlem and say that that's an app. I mean, we just out about in terms of what those folks are making. Most of those folks are making and what they can pay for, as far as correct was concerned.
And, as you said,
hundreds of involved in my thing is, we really need To start requesting it from here.
Because we have Congress people. It's just not the same That's the reality of For their runs for office and all that kind of stuff We really need to. I guess maybe you
go and start and reach out to other Mary's And deal with the Congress people who are reping us. And HUD created this this
policy that it is not working in terms of housing for for for
human beings in this country. Yeah, I think I think there's like A lot of we that this could
be addressed. It's a man made issue right like how we,
how we calculate affordability Can always change. And so
whether that's at the Federal level, or if at the city level, we just address, we just shift what we consider
a portable based on that number like those are
things that can always happen. Are there any other I was gonna ask about like what
you put on the microph10, I was going to ask,
how were you able to get the map into the code? So you were taking it from like the website,
and it just read it.
And yeah,
so that's a good question. Yeah. So the base map. What you see is that's anything
account what we need, what we made Are these shapes of purple, right and gray.
That's what we put on the base map. The base map, which is this detailed map of
everything else that was one line of code That just said, Put in this map. Because that's that's never changing right. I'll I'll show you how is this one? It's? It's It's a base map of, and it's called like carto,
is this company that does this work,
and it's it's within the leaflet package that we that we import it.
That allows
you to add a base map. I like this one, but there's a bunch of different
themed maps that you can use To just kind of Orient
yourself right? But what we did was just create those purple
tiles of of neighborhoods. Yes. So where would you find like
that string that you added to the provider tiles. This, yeah,
that's it. So they actually they have to positron is like the theme name of this one. Also,
in addition to carto, a couple of the companies
have have pace map tiles as well. But I think you can. You can just look it up on mine.
Usually like Carter, will provide a list or like for for specifically leave Flip for R.
polytron pedal. That'll work as a base map.
This to be in the like. We put our package
documentation. Yeah, it'll be in documentation like that, or Cardo will provide the list of
names of maps that they make, you know Sure. Do you need to have a carto
subscription to use the positron basement or yeah. So everything I did was totally free,
which is great. I wouldn't have done it otherwise.
Like all I had to do before,
this was obviously download, or just for you,
download our studio, which is also create. All these packages are free reed to Krata and the Api for the to call an Api. When you, when you just pull data directly in. you're not like downloading the data and then uploading it back. This was all free. I just had to create
a like a username and password itself. So maybe they track it. But
I thought it was worth it. And then, yeah, everything else is is totally clean,
open source. Yes. How much control do you have in terms of the Ui and Ux when you're creating a a map like that? Yeah. So so our tiny Does a lot, but it's far,
I would say it's far more limited than like Coding language that designed For web design. Right? Some
Some people actually who are really good at our You can
you? Can.
You can code HTML or something else in our but
it's it's the HTML language in terms of language that's
specific to our and what we did It's all because it's all so easy.
It's all pre-designed for you. So you want to mess with that some things
you can mess with some things you really can't. How how interacted can you make
them out? You can make it pretty interactive. I mean I've I physically
sort of almost like a simplified version of one that I've made
where it's different tabs you can do filters. You can search addresses.
You can do all that stuff. Okay, cool. Yes. alright. Yeah, it's cool. If you don't have an answer, I don't know how long you've been doing our but you that you said that you didn't know how to go it and stuff before the any recommendation for Learning our Yes.
So what what I did was,
I did a free trial code One week it was a week
that had a lot of free time, and And then from there it was
just some of the pot. There's this thing called stack overflow,
which is the best it's. It's literally the best. Yeah,
it's. It's. It's a. Blog. It's like a. Q. And a log for
code.
And so if you Google. If you Google. A really specific question about our like,
how do I make the Them purple from light to dark. Like someone has asked that
on. And so for me it was literally just Googling. You can read
documentation. That's it's all free like I never paid the code account. I
mean the the name so put Academy is a is a website. It's it's a
it's usually a subscription base Tutoring service.
but they also have free trials. So it gives. It gives you a ground
to like. Learn the syntax, and get just your head around it. Enough
that you know what to Google. Basically But that sort of was my. and and it's like, I, you know, like I was not a poker person,
and i'm still not, you know, like. but like you're still able to do stuff
like this. But just a little bit of Yes. I was going to. Okay. Makes all of its own documentation free as well,
like they have their own documentation on how to use our.
Sometimes it gets a little in the weeds, but It's all free, like. I think you should just
do as much, Googling as you can really Anything else. Thank you. We thank you guys for coming..