Justus Perlwitz

Designing a Better Web Analytics Platform

Not many things a business owner has to go through are as scrutinous and unpleasant as a data privacy audit. In the case of Germany, this will include a state commissioner for data protection putting your companies data processing and storage practices under the microscope. Luckily, as it turns out, most of the time you’ll already be doing the right thing. Salting user passwords, allowing customers to delete their account, and allowing users to request all personal data that is being stored about them should be business as usual for most of us.

But there is one thing that will always be a problem in a data privacy audit: The common reliance on US-American service providers for cloud services, data storage, and web analytics. This includes Google’s famous product Google Analytics.

What does Google Analytics do? It allows you to pervasively track your website and app users, understand what they like, how they convert, and what their business value to you is.

How is this problematic? Using Google Analytics in Germany and other EU countries is legally questionable. I was personally told by a city official last year that in order to comply with German privacy laws, you need to close a legal agreement with Google Ireland that will guarantee you processing and storage of user data in conformity with the law. There exist countless templates for this and you can usually just mail this over to Google Ireland and get it counter-signed. But that’s not the end of the story.

At the end of the day, when you really think about it, you are handing over your user’s precious trust to a faceless multinational corporation that has a well-documented track record of not caring too much about their user’s privacy or third party information that they have been entrusted with. And if you ever want to migrate away from their platform, there is no easy way to retrieve all your customer data and move it someplace else. So clearly, choosing Google Analytics is a convenient thing to do, but it will really only get you to your goal half-way.

Another big limitation is of course the quality of results when deploying Google Analytics on high-traffic pages. Sampling will start at around 500,000 records, meaning that when you are investigating small user segments, you will not always get the full picture. More about their approach to sampling is described in this article by Moz.

I have spent the last few weeks trying to understand all the alternative solutions to web analytics out there on the market, both proprietary as well as open source. One of the most prominent alternatives is of course good old Matomo (formerly Piwik) that you can self-host. It is very unfortunate though that Matomo is hard to deploy outside of LAMP stack environments, meaning it runs on PHP, assumes you have a normal HTTP server like Apache running, and really does not like to be dockerized.

It is also completely tied to a MySQL database, which surprises me. Given the size of the project I would have expected Matomo to have a flexible storage backend that can use any available SQL database. Since I am a self-proclaimed PostgreSQL fanboy, I would of course like for that database to be PostgreSQL. Furthermore, the fact that while attempts at dockerizing Matomo exist out there, none of them really strike me as easy to deploy solutions, as they grow outdated quite quickly. Matomo itself does not officially maintain a Docker contain ready to be deployed.

While 10 years ago a reliance on LAMP would have been totally acceptable, I feel like in 2018 I just can’t deal with it anymore. We live in an age of containers, reproducible deployments and many more luxuries that many of us have gotten accustomed to.

So I was thinking what every programmer at some point has thought: What if I implemented it myself?

A question of course that can either lead you to the conclusion to never think about this question again, since creating something like Matomo took a decade, and how will you ever replicate that effort as a sole developer? Isn’t this just classic Not Invented Here Syndrome?

Or, another take on this could be the following: Sure, Matomo or Google Analytics can do a lot of things, but really what I need is only feature X, and perhaps feature Y. And as long as I choose my solution to be modular and flexible enough, I can easily add the rest when I need it. So, I will reinvent the wheel, but only to the point where it starts spinning.

  1. Anatomy of an Analytics Platform

So what exactly is it, that Google Analytics does?

I would claim that there are 3 major architecture components, that almost every analytics platform is offering:

If we really reduce our project to a simplified combination of those 3 components, we can get the job done quite quickly. For the sake of example, we will implement it in the micro-framework Flask as it allows us to create the beacon and storage component without using a lot of code.

  1. The Beacon

For a beacon, we have to provide an HTTP endpoint that can accept requests coming from a user’s browser that contain detailed information on what site the user visited just now and other data about this interaction like browser language, screen dimensions, and so on.

We can either choose to add a small piece of JavaScript that contacts this HTTP endpoint, or we can just add a small image tag to the end of a page. For the image that we serve we can either serve a real picture, like a 1x1 pixel GIF, or just serve up empty content with a 204 No Content HTTP Status code.

So we get out the good old Flask and start hacking away. All we’re really looking to do is to create a beacon endpoint that can

It’s important to keep in mind that we have to use the HTTP Referrer URL here, since the HTTP path used to retrieve the beacon will not tell us what page the user visited, but in the case of a Referrer URL, we can see on which page the tracking beacon was requested. This page is exactly the page that the user’s browser is displaying. To illustrate this, I will show a simple curl request to the handy site https://www.whatismyreferer.com, once with referrer, and once without a referrer. The referrer will be http://www.google.com/.

Fun fact: the HTTP protocol specifies the referrer Header tag to be spelled ‘Referer’.

%%bash
curl 'https://www.whatismyreferer.com/' \
    -H 'Referer: http://www.google.com/' 2> /dev/null \
    | grep -e 'Your HTTP' -A 5

Output:

        Your HTTP referer:
 </h2>
 <p class="lead" itemprop="description">
  <div class="alert alert-info" role="alert">
  <strong>
  http://www.google.com/      </strong>
%%bash
curl 'https://www.whatismyreferer.com/' \
    -H 'Referer:' 2> /dev/null \
    | grep -e 'Your HTTP' -A 5

Output:

        Your HTTP referer:
 </h2>
 <p class="lead" itemprop="description">
  <div class="alert alert-info" role="alert">
  <strong>
  No referer / hidden     </strong>

We can see that by adding the referrer header, What is my Referer knows what our referrer is.

  1. The Storage

We will also have to create the database used to track page views. The only table we need will be called beacon_hit and it stores individual hits of our beacon, which in turn are page views. It contains the following columns:

Of course in the real world we would track a lot of additional information, but for the sake of example we will keep it brief.

import os
import sqlite3

db_path = "/tmp/analytics-beacon.db"
# Ensure we start with a fresh database
os.remove(db_path)

conn = sqlite3.connect(db_path)

conn.execute("""
CREATE TABLE beacon_hit (
    id INTEGER PRIMARY KEY,
    timestamp TEXT,
    url TEXT
)
""")
conn.commit()

When the beacon is then hit, the following information will be added as a new row to our database in our Flask application.

id timestamp url
1 2018-03-09T18:28:10.357074 http://localhost:8000/

  1. The Beacon, Continued

We can now create the tracking endpoint that stores page views to the database and serves an empty response. As described above, it will need to respond to every request with 204 No Content. We can easily do that in Flask by just serving a response tuple containing

("", 204)

and we can call it a day. So let’s combine those two requirements in the following track() function. First, we need to import Flask and create an application object. Then, we can define the API endpoint.

from datetime import datetime  # for timestamp creation
from flask import (
    Flask,
    request,
)

app = Flask(__name__)

@app.route("/")
def track():
    conn.execute(
        "INSERT INTO beacon_hit (timestamp, url) VALUES (?, ?)",
        (
            datetime.now().isoformat(),
            request.referrer,
        ),
    )
    return "", 204

We can use Flask’s test client to then call the beacon and trigger a row to be created in our database.

app.test_client().get(
    "/",
    headers={"Referer": "http://localhost:8000/"},
)

Output:

<Response streamed [204 NO CONTENT]>

As expected, the result is a happy 204 No Content. We can then print out what the database has stored:

for row in conn.execute("SELECT * FROM beacon_hit"):
    print(row)

Output:

(1, '2018-03-09T20:16:17.045070', 'http://localhost:8000/')

Since we want to show some dashboard data later on, we will fire off another 200 test requests. We visit two pages, page_a and page_b, 100 times each.

for _ in range(100):
    app.test_client().get(
        "/",
        headers={
            "Referer": "http://localhost:8000/page_a",
        },
    )
    app.test_client().get(
        "/",
        headers={
            "Referer": "http://localhost:8000/page_b",
        },
    )

  1. The Dashboard

Here, we will try to get things done with minimum necessary effort. We will use Pandas to read in the data and create some plots. Our goal is to

We import Pandas and get started:

import pandas as pd

To read in our data with Pandas, we will read in the whole beacon_hit table as a Pandas DataFrame. To achieve that we use pandas.read_sql.

df = pd.read_sql(
    "select * from beacon_hit",
    conn,
    index_col='id',  # Automatically use the table's id column as our index
)
# Cast the timestamp to a proper datetime column in Pandas
df.timestamp = pd.to_datetime(df.timestamp)

Let’s take a quick look at the data contained in the table and now read in as a DataFrame:

df.head(5)

Output:

id timestamp url
1 2018-03-09 20:16:17.045070 http://localhost:8000/
2 2018-03-09 20:16:17.089820 http://localhost:8000/page_a
3 2018-03-09 20:16:17.091435 http://localhost:8000/page_b
4 2018-03-09 20:16:17.092620 http://localhost:8000/page_a
5 2018-03-09 20:16:17.093510 http://localhost:8000/page_b

That looks excellent! Let us now plot page popularity using a simple Pandas value count on the url column.

%matplotlib inline
df.url.value_counts().plot(
    kind="bar",
    rot=0,
    figsize=(10, 5),
    grid=True,
);

Output:

<img alt=“<matplotlib.figure.Figure at 0x1125e0ef0>” src=“data:image/png;base64,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”/>

We can clearly see that /page_a and /page_b are equally popular with our Flask test client.

In order to show request distribution over time, we have to be a bit clever since our test data has timestamps that are very close. We therefore use a Pandas Grouper to group the rows by milliseconds using pd.Grouper(freq=‘ms’).

counts = df.groupby(
    pd.Grouper(key='timestamp', freq='ms'),
).count()
counts.tail()

Output:

timestamp url
2018-03-09 20:16:17.283 0
2018-03-09 20:16:17.284 1
2018-03-09 20:16:17.285 1
2018-03-09 20:16:17.286 1
2018-03-09 20:16:17.287 1

That’s good enough for our purposes and we can simply plot the data:

counts.plot(figsize=(20, 10));

Output:

<img alt=“<matplotlib.figure.Figure at 0x10f3ebc18>” src=“data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAABJkAAAIvCAYAAADXiQp5AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4xLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvAOZPmwAAIABJREFUeJzsvW/sNUl233Xq3t+dfRazhGR3MWHHYfdFJNsReIlGGxAWsQ0y68iRhUCyVyiJEGYlFCQiISRAyDbJC5BYW1EIxCywsiwSw4vEwVh2Ykuwa4zlkFnHxOs/2Nbajmcx2fGM99/s/Ok/xYu+1V1dXafqnLq3b9/b9f1Io3me3+/27X76dFedOnXO9xhrLQEAAAAAAAAAAAAAcAmHrS8AAAAAAAAAAAAAADw+CDIBAAAAAAAAAAAAgItBkAkAAAAAAAAAAAAAXAyCTAAAAAAAAAAAAADgYhBkAgAAAAAAAAAAAAAXgyATAAAAAAAAAAAAALgYBJkAAAAAAAAAAAAAwMUgyAQAAAAAAAAAAAAALgZBJgAAAAAAAAAAAABwMU9bX8A1ede73mXf+973bn0ZAAAAAAAAAAAAALvhk5/85O9aa9+d+9yugkzvfe976cUXX9z6MgAAAAAAAAAAAAB2gzHmtySfQ7kcAAAAAAAAAAAAALgYBJkAAAAAAAAAAAAAwMUgyAQAAAAAAAAAAAAALmZXmkwAAAAAAAAAAAAAW9A0Db300kv0xhtvbH0pxTx79oyef/55Op1ORccjyAQAAAAAAAAAAABwIS+99BK94x3voPe+971kjNn6ctRYa+mVV16hl156id73vvcVfQfK5QAAAAAAAAAAAAAu5I033qB3vvOdDxlgIiIyxtA73/nOizKxEGQCAAAAAAAAAAAAuAKPGmByXHr9CDIBAAAAAAAAAAAAgItBkAkAAAAAAAAAAACgIr7ne76HPvKRj1z9exFkAgAAAAAAAAAAAKiEtm1X+250lwMAAAAAAAAAAADYAb/5m79J3/qt30qf+tSniIjoIx/5CH3pS1+ij3/84/T+97+ffvqnf5o+9KEPrXZ+BJkAAAAAAAAAAAAArsh/9r/+Iv3S//uFq37n1/5T/xh995/8I8XHv/XWW/Tiiy8S0VAutwYolwMAAAAAAAAAAADYOd/+7d+++jmQyQQAAAAAAAAAAABwRS7JOLqEp6cn6vt+/Psbb7wx/vkrvuIrVj8/MpkAAAAAAAAAAAAAdsBXfuVX0mc/+1l65ZVX6M0336Qf/dEfven5kckEAAAAAAAAAAAAsANOpxN913d9F33gAx+g97znPfTVX/3VNz2/sdbe9IRr8sILL1gnYgUAAAAAAAAAAABwK375l3+ZvuZrvmbry7iY2L/DGPNJa+0LuWNXK5czxnyVMeZ/N8b8kjHmF40x/37kM8YY85eMMb9ujPn7xpg/6v3uzxhjfu38359Z6zoBAAAAAAAAAAAAwOWsWS7XEtF/YK39OWPMO4jok8aYn7TW/pL3mW8hoj98/u+PEdFfIaI/Zoz5A0T03UT0AhHZ87E/Yq39vRWvFwAAAAAAAAAAAAAUslomk7X2d6y1P3f+8xeJ6JeJ6D3Bx76NiH7QDvwsEf3jxpg/SET/KhH9pLX21XNg6SeJ6INrXSsAoD5+43dfo1uXC7/8xTfpC280qmM+/fKXVNf5hTcaevmLb2a/E6yH9v5KbAbWBe/EhHZsfO3Nlv7hF97If9DjM597nd5oOu2lASLqe0u/8buvqY555Utv0ue+/JbqGO3cE9J0Pf32q19WHfPZL7xBX1TOkXulxGZboLWZtfbi8faNpqPPfO511TH/3+ffoC+/1Yo/X/KeATmS+6u12b3x6JJEl17/TbrLGWPeS0T/HBH9neBX7yGi3/b+/tL5Z9zPY9/9YWPMi8aYF19++eVrXTIAYMf81iuv0Td+5OP0s59+9abn/c4ffJH+ix//FfHnf+0ffpG+6Xs/QX/vtz8nPuY//7FfoX/nB3ltuk995vP0Td/7CfrUZz4v/k4gp+T+5mwG1gXvxMTvfP51+qbv/Th94lfl/tRf+t9+jT703/2s+PPWWvrgX/wp+qt/5x+UXGL1fOLXXqZ/+Xs/Tr/zefki+8/9zz9P/+nf/JT48yVzT8gP/73P0L/yfZ+g19+SBxP/9Mf+L/q+n/zV4nPuCa3NtkJrs5/7B5+jb/reT9Cvf/aLxef8H3/2t+iDf/GnVIvgf/2v/Az9t5/4tPjzJe8ZkCO5v1qb3RPPnj2jV1555WEDTdZaeuWVV+jZs2fF37F6dzljzD9KRH+diP6ctfYL1/5+a+1HieijRIPw97W/HwCwP37vy8Ou26uv3XaX8NXX3qRXvyQ/p7s+3TFvJv9d7ne/9wA7pI9Iyf3N2QysC96Jic99uSFrdWPjq196S/X5prP0xTdaevU1ZO+V8OqX3qLeEn3+9Yb+4O97u+yY196irpe7yCVzT+w73mx7+vJbLb39uaP4GIyFA1qbbYXWZuOz9Vp5xtqrr71FX3yjpba3dDoa0TGvKOfZkvcMyJHcX63N7onnn3+eXnrpJXrkBJhnz57R888/X3z8qkEmY8yJhgDTX7XW/o3IRz5DRF/l/f35888+Q0TfEPz84+tcJQCgNtquH/7f9zc+r1Wdsz07mKpjOjv+++Lf2Y+fA9en5P7mbAbWBe/EhLsHque3t8rP435fQvkYo7Opf64Spnl2vWdpz2htthXq9989FxfMeePz2Vk6yeKXBf4Xxqk1kdxfrc3uidPpRO973/u2voxNWbO7nCGi/4GIftla+33Mx36EiP70ucvcP09En7fW/g4R/W0i+mZjzO83xvx+Ivrm888AAOBinIPS3Nh5aHurOmdT4KQ3vaUm8Xl3fs13Ajnu/jYKBzpnM7AuJTbbK83ZoW8Ujn3T9brnvdtm/N0LZWNMr7LpNebIkutslc/SntHabCu0NnNz3SVzXqPcKLTWFvhfmBfWZBpj4ve3xGbgvlgzk+lfJKI/RUS/YIz5+fPP/hMi+kNERNba7yeiHyOiP0FEv05EXyaif+v8u1eNMX+BiP7u+bg/b629rXgKAGC3uJ2T7uaZTL0q/b3zduvkx6TPMX0nHKc1cPdXZ2fdcwGuS4nN9krZ82uLxrVbj7974RY2cvPDJe9EyXW2yuvcM1qbbYXWZu69v+T91/pGpe+M9hggZ/LD4/cX9//xWS3IZK39aSJKFsraQQ3rzzK/+xgRfWyFSwMAVM64W3/rTKbOFu74645Jfd79Dpkz61Byf3M2A+uCd2JivBfKHf+2t2StpSGJPU2L+30RJTYa5p7bZnGUzLNth6xOh9ZmW6G12TUyGcfvEAaqclkz8XNs4yfWQu7+ltgM3Bc36S4HAAD3xKQ7ctvJq+l7pT5FgaZF12dr3N3nwPUpub85m4F1wTsxUfT8Kseppsf9voQSvaSm61e1afQ7uoLr7HXXuWe0NtsKrc1KdN+W36HTS2qUnye6ji4Z4Mnd3xKbgfsCQSYAQHV0V3Cgy86rE3ienHSlcGrCKbrG4gHwFInyZmwG1gUCrxNlgW3dolG7QARzSu7fMMaUlGpfQfhbeJ19b8laPBcOrc22oMRm0xhzufC3tJSqJOCJcWpdciWPJTYD9wWCTACA6thCeNZaey6LKhD+VpfLDaUr3O+H74TjtAbaNH53TMpmYF1KbLZXykp0dWLh0/3G815CqY1uLc7eKMtdSkTn94zWZltQ1ijgGuVy/ez/2c8XlG5C+HtdcjbcStYCXA8EmQAA1TFlLtzOeegKUq/bzE5P9Jjzv4lbv01tpeE4rUGJWHvOZmBdSmy2Vy4RyBVnMl1B+LdmpvunyzYrEf6+KNtEKR5+jTKqPaG12RbkxJtjdFfwvyZ/Sif8rfO/Lhe/Bzy5TH0Ifz8+CDIBAKpji530SwJGZa2n48dcozU14NHusBJB4HJrSmy2V4qEv5WlVe0VMhlqpi2Yv9q+L7LpJTbS2rlF5sgMrc22oORdvob/pc0yKrnOkvcMyJkylRhfFePBw4MgEwCgOqbdt9tNXmPpm1JfiahU74Cpc4fo7qpotSKGY6CTtSUlNtsrpcL1RPLnF2PQZZTcP63umzYLiTvn8H9dSRPew4FH0OprCrKSRv/rEr0vpY5eU/A8Y5xal1wWXInPDO4LBJkAANXRKR2U65yzZPGmSwmfHcPuDmHiXpOS+4vuZtuCd2LipsLfuN9FaAWJuwJx5qsIfyvnWW0J1J4psdkWlNjsup0L1wtsQ/h7XbLC3wjyPTwIMgEAqmML4e+SFPGSXcLcvw3C3+tSJsqL8qEtgcDrRNHzqxT/xf2+jLGUTXy/9eWgW5Q0oWx14lHuxUXP1hWEv6W+0SVlwBCiX4dcw4gSm4H7AkEmAEB1XKOFbvE5185kyvzbtvi310SR8DfKRDYFwt8TtxX+xv0uQS2oXVLCW9DZlPsOCH/reZQS3l0Lf1+hZBTw5N73kmcL3BcIMgEAqmOLzJESB7ot2CWUTtzYHVqHkpbOELjcFrROnyjZPdY+vxiDLkMvqD2VJ1kr1LC5hvD3mAmiCz5iA6TMZlvgbHVz4W9ls4ySDG6MU+uSFf5GJtnDgyATAKA6ttDAyaUGR48pyLDIiSU2G/zba6IkmAiBy21BBsWEVqyZSP/8Tp/HGFSCtpufv0gWa9hcwUb6kiYs6h0lNtuC0Z9QdcC9XpacVgeuzP/COLUGuUx9aGI9PggyAQCqY0zXvqHzVlRGdVFHuly5HCbuNSgpM4HA5bZcozRoL5Q49trnt6RrJpjQd/Ob7KIW4b4kk0kZvL2G2PheKLHZFpT5NXoZgEu/o6hTLzZ/ViX3vsMvenwQZAIAVMe4Q3VD56HxnCJp+rt2krXWjhM3txtckqkA5GifLYnNwLpsMR7cK2XC9dqgB8pDL0F7//yFtVac/aJAQKFAOd7DMpttwSXC35cEzxplaWXJs4Vxal3yWfcu+wzjwaOCIBMAoDq2yFzwnSG1EKpy8Raeb/6dSEFeE3Uav8BmYF3wTkyUZDpqM1aQIXAZF40xSnH2i0qa+rLrxHNRZrMtKLHZNbS39GOOPisG88K65LLLtpC1ANcFQSYAQHVsof1QpIuREUZcfF5wDohZrot2915iM7AuEBidKFk0doXPPBZvZWht5C/S5Jkfzqa3K2ly81x352LXt6DEZltQYrMpQ6jcxp0yy6ioXA5Bz1XJ6ZTi/j8+CDIBAKqjpJ3ttc5JpO+IIs18agQ6Dm7R0N2x4/rIaBfQEpuBdUHQY6Jk97hRZqxoS13AnFJB7eEYqT7S5ZlM2uucz5F1v4slNtsC32ZqUfkrPFvirPCCDqLa5xfoyOl5aTMhwf2BIBMAoDq2mLxmO5PajijKlPDwfNHvxO7QKmhLASQ2A+uC1ukT2hKRIYNhfmz+HAjqXYK+jLpk7rncRvqMK5QOOx5F+LstCIZdRfh7zJ5c73m+xnUCntyGxnT/6x4LHhkEmQAA1VEibnvxOf1dWmnJgjLjyp+sOeerKchUAHK0pZgSm4F1Qev0iWZcvOkEdf1jpcdAULeMRlkq5D/X2rnnGuVy8uv0nqXK38USm23B/P1XPls3zJLzRaalZX3XuE7A04zZZZyvOm2y1l4++6ggyAQAqI4thGcvyWSSl6EIMpmQgrwq2lIAic3AumzRCOBeuUxUWqmPgqBqEWobXTT3XCMQsN6ztFdKbLYFRaLyVxDUVgt/X1TWd7/3/5HJZjJ5dpKWRYL7AkEmAEB1bNGy3N+ZlE6YWnHLzt/9ZByj7go71IBHu4D2bYZF9zYg6DFxi+e3w/2+CO39m+vm6LKfLrGR9jr9ea72RWWJzbbAt5k44/oK77+62UBBwALj1Lo0mc3eufg9bPCIIMgEAKiOboOSsVsLf3PHTIKZ9+u4PjIQ/n48oBE0oRX+Lnl+Iah7GdPirCRbUrrA3lj4u/JFZYnNtqAruM5W+fzGcONOV5DNpy3fvOcg3yOTF/7GBtyjgyATAKA6NhH+7vW7Mm2mZn3xeYFw6qSTgUl7DS4S/oYzuwkQ/p6Y7sX1xhzumN4S9Vg8qNG2Y283yja5SPi78gBkic22oEj4W/n8hvjNBop8oxtdJ0gzjQ9pX3X48/2+A4AHQSYAQHWMQr83dN5mQp7KTCatuGV4vtlnet13Ah1aEWmI3W4PhL8ntE0RSp5ff9y9Z1Hje0UvqJ0vo14eo+tsGkN9nT3GQkeJzbag5F2+tPHKrESv5NlSBsMh/L0OuTEG48HjgyATAKA6NslkuoHwt0Q4dQvR85pQZzJB7HZzkMk0cRPh74KsAjCR0zIJmWXRatvMX1Iup5xnkdU5UWKzLSjKELrQBykT8dZfp/Y9Azqywt8YDx4eBJkAANVxDQdafc4CIc9cOvHi84J24tCfWRd115uCFvDguuCdmLhkzNF2bRr+jHuuxd1naYaFf4+1QsmlC+x5SVNJwL3u56LEZlvg20z8PF6YTX3Jhp3qGOV7BnTkxhjME48PgkwAgOrYoptUmWOk1LTwO6iwbWGRtbEmah0Sgc3AuqC73MQlY05ZO3E881rcXCJtCDEfY3TZT6X28Y8TXye6SY2U2GwLfJvJn8drZjIVBDCVAfTaOx2uRa6iAMLfjw+CTACA6tiiZKwoxfuicoP0xI2doXWYyo1KhL9hky3Q2mzPXDbm4Jm/BTcR/r4wu+8SoeXhmLrfxYcR/i4K3lzv2ZJq9ZT5X/CV1mTa0OA2RDEePDoIMgEAquNS4clLzqk57xrCqeN33rHj+siohb8hbrk5EP6eUAt/Fzy/82MwDmnRjuF+Ge6thL/bgvkOTRAmSmy2BUXC/xf6IEXC351+zIGvtC5Z4W+MBw8PgkwAgOp4FOHvRplWLmn5OopZYtJeBbXwN9r0bg5KSCe0pSwlzy+Evy9DL6hdkMmk1OYK8RfmRc9S5e/iw2QyFWWsXeaD3Ez4ewM/sSZyYwzGg8cHQSYAQHVc6kBfck7NebUC5SLhb+jPrAqEvx8PCH9PuHswCDfn70eR8Hf/GAvoe2QmqF0i/K3uLne7TCYIf0+U2GwLZsLfK/k1y+PXH3NK3jMgx1o7al3xmkzIZHp0EGQCAFTHVIYmW0hd5ZwlZSVKEd5GUMM+ZTLBcVoDbUca32YdFtybgC5CE9osgZLuUk2nOweY8O+xVJDYX/xrxZlL34mS62wLSpr2SonNtmAm/K3MZCrdVCnR6tEG7UqeXyBHMm/4doINHhMEmQAA1VHSEeXycxZkMl1QFsH9u5C1sS5aUXm06d2eLRoB3Cva53GWIbBiO3EwcHmpkDLb5CqBAGnwAcFHx6OUUd8iQ5s7nqhQ+Fs0ruFZXBNJ84euIJgI7gsEmQAA1dFs4EDMdu+FjpHbGRSnoQvEQt3EDTHLdXDPVkkLeKSEb4PWZntm1qBAMEaUZCWVZD+Bga7T37uSMcZ9t7RsMsTPyhQLlBeUQe2VR5kXtOLM1tqLS/abgnLb+bMlKJcreM+AnEawOTF7tiofDx4VBJkAANXRbRBk6nq9A+0mYmuJeknpiiBbyn0n0o/XYcyKKRFBRuBvE7Q22zPazKSiDE2UyxUjWZyFlIhIXzpHFm2qIHNh5FGEv+fPSf46r6G75R8nL8XUnbfkPQNyOsG88SjZfIAHQSYAQHU0XU/GDH++1eTV9FZ9zrabjpHsBrvJ2pj4rqITs3S/v5UeVS30vaX+fH97aWAwYzOwLiU22zP+mCMZp/znV9NdahzXsHhQ4e6xMZqya28eEXfW8p8D/TvhX6dGq+uSc+6JEpttwSzzURGU1jy/y+/Qa3f5Y44oGHaF6wQ8jWDemM8T9/sOAB4EmQAA1dF2lp49HYnodpNX2/Wqc7q0cneMaPft/Jm3n47RBaJzyNx3Ipvpurhd2refznZWlBu9/XSE8PcGlNhsz8zGHEX25NtPR1Uwwd1vjEE63D1+++koz4jtevX9ns1XBe+EO2YY1+TBx0vOuSdKbLYFbd/Ts9OwlJRcZ+s9F6UbXe65f3Y6qPS+xjFeIfytec+AHGe3Z0/8vNF0/lxU93jwqCDIBACoDt8xutXk1XZWdU7fkXLH58+RduwX3wnn6ar4DjSR3ma1795vQYnN9kzb9boxp9cvxtq+x/0uxA+KysvlLL3tydk0P/e47D7NcxA753idipKmS865J7Q224rBZu5dlmcIXRJAc2POs9NRrNXTeoFtjf+FeXkdxiDT6ZAU/tYEMMH9gSATAKAqrLWznfRbORCNcictdMZk5XJu4o47X35AY7iO+3VeH5Gmm+4/kXSRnrYZWJcSm+2Z2TilyMR7pshk8hemGIN0uDFcdb/7np6OBzodjWiMaRaBV72NSq7Tz96p/bnQ2mwrGmWGduMFiIgu0/t69hTP2GavU+V/6Z9fIMcfY1LC3xqfGdwfCDIBAKrC+TTPnrutM9v1/XhOjfCkO0aykzOWw50OTLmcVX8nkDMG8Z7TLNLTNgPrUmKzPdN2unHKv3/SAF3T9eP9RjaljrG89jlNuZyl08HQ00E2xrh5YZwjSwIBvXedCuHvZ3guiEhvs63oeuu9y4pMpgv8L/875KWY3nVqpAcU7xmQ448x3Jzb+uPBHb8DgAdBJgBAVTRBNs+tAi1NZ+ltoxaSvIWuZme36y0dDNFzT0dW+Hv+nXCerklYjih5tnI2A+tSYrO94sqkNGOOf/+k984vg0CWgI4uuN8STZuut/R0PNDT0Qj1aIKSpoJxqeu94Lm0zXwPrS6H1mZb0fRWmSF0+Xg7K5cTZ0/62lGKQKviPQNyfD+c657c+plMlY8HjwqCTACAqvDLk4huK/x9Ohp1yYKmjKfpzo7pwUQd+6YLvrPyrI1rEwqrSxzgnM3AupTYbK8syqRE2ZOegKs0mNDpGhqAiXFe0JQodT09HY14jGmDeaJI+Hsm7KsQG0e5HBHpbbYVQyBArqNV8vwuvmMMgvJ6PrFjisr6btwgphamYCM/xsxlLe73HQA8CDIBAKoi1CW6mfB3b+moSH8PNZmki4Ong6Gno4k6fGEnLSzwrsuYKaYsN3I2q333fgtKbLZXuiAAL9vx15fLtf1ULodnXkdYbiTq6NXZ8xgjyza7xjxRVNLUWTodDR0Sbc1rQWuzrWg7S889HcgYXYaQpsRu+R1TEFRTinlJWd892+ARkTTcGBr04P4/MggyAQCqYrtMJkungzz9PdzpEbUT7wfH9HSIlyiEO9TQGrguCxFpoVi7s1ntu/dbUGKzvTLdi8Ps75Jj3pZoRR0yE/6u+H6XUJJl1PY9PR0OdDoI557FHFmgm1NS0tT3dDoe6Ol4qP650NpsK6brPMgytEMf5IJMpmeno6IUs1z42x0PrsfCv2U2RTVzEbg/EGQCAFRFqAlwqzTcoVuModNRplExlcspUtE756THM5nCRSRSkK9LuDsnLWVJ2QysS4nN9soiCC0KSPR0PBh67skoSld6ZFMW0hRkGTXnDKGnozSLNph7LugA9vaTXDh5CrhjLNTabCva3tLT0dDxYIoytMuEv/OdyWLnVb0zyPpelVALMRbE8zcj7vkdADwIMgEAqmKpS3Q74e+n4+HsjOkzmUQivN3g8A07wfH04/l3wnG6JosSR6GdUzYD61Jis70SlkmJsmRcWY+iC1brCTwje0/HotxbJM7eTyLSGp2tCxZ4/nVKhZPnc2Tdz4XWZlvRdJaeNBnaBbpvy++Ynk+N8Lcr69PokkETaB0kGW0zjbY7fgcAD4JMAICqWOhN3GjyGtLKz7u0otI3vTM27AQfzjvBkZ2hxb8djtM1aRYZANJyOd5mYF1KbLZXxo4/ylbfYyaeImPlkiyZmllmAMg1mU5SPcCrBgLk5S5On27I9q37udDabCumhibCDO1FNvUFWXLPKYS/+yEzbJAS0JSMyt8zIEeiOeps9gTf6GFBkAkAUBXtYiF1o3I5T8hTlVb+nKad+ODwcdlSy++E43RNRuHk5zSBwclmELe8PSU22yulwt/HcTEsLV3px/uNZ17HmOHq7p9QY+l0zhCSCoUTXSaGP5bcaQTKeztm79ScUUikt9lWjDYTXudyvL1A+PtJJ/x9HJuiyP0vzXsG5IyB7MQYMwSddRsY4L5YNchkjPmYMeazxphPMb//D40xP3/+71PGmM4Y8wfOv/tNY8wvnH/34prXCQCoh82Ev/tpx1+0+xyUtom7CB0P7K5iqLlyz87rI+K37R7+Li1xHGyG0qHbU2KzvTLdC4Xwt9ttPsparfe9pd7ifpfSegtsImlJo9MDLBT+LggEjN+huc6+p9O59LJ2oWWtzbZizNA+HsQahET++38j4W8XsBCL3+vfMyAnHGOi/uqsI/P9vgOAZ+1Mph8gog9yv7TW/pfW2vdba99PRP8xEX3CWvuq95FvPP/+hZWvEwBQCZI03XXOe3YahenvTXidQgfuye3WRQJISzFLOE7XRLI7F5KzGViXEpvtlXFhpWr1rVu8ucXac08HtKovoCnIMmr8LNoiPZpLSpr0+nQnZDKpbbYVU4a2Uvj7gkxyF9h6Ohrq7RC4lpx3FFJXlPVhXliHpQB8PPNes4EB7o9Vg0zW2p8iolezHxz4EBH90IqXAwAARV3brnJeTyBTJ/ztsgqE+j5Hfucn7BoEnYHrUhLAzNkMrMtWQed7ROL4x45xwvWaRaZbQCNDQEcYAJKWUasCgVfoQloinOyEv+89sHILtDbbCl+sXZOhfZHeVzdkuJyOfGey2HldWZ/M/9K/Z0DOQgsx5q+ONpNlyYH74y40mYwx/wgNGU9/3fuxJaKfMMZ80hjz4cSxHzbGvGiMefHll19e+1IBAA9OSde2q5zX7b4JnbEpGCZ3xlxK+JBiHxFSvELXIMATOk7S8oGUzcC6lNhsrzQFGSyNtgzYBZmOB7SqL6DpgwCQuIz6LM6sEP6+pLOTO+ZtTwqBcn+OrPy50NpsK5ymoDZDW7N5FvuO0zlgRJQfp6y1wzGasr6C9wzIyZXkzm0G4e9H5S6CTET0J4no/wxK5b7eWvtHiehbiOjPGmP+pdiB1tqPWmtfsNa+8O53v/sW1woAeGCusZNWQteX7firusud04vQvQ2LAAAgAElEQVSPTAe7rf7ttbDQGRBmgqRsBtalxGZ7Zew+qSyXm4S/BYvM83dODQqweNAQ6upJ77kTkdZ01rqkrNp1ins6yrOhulm2b93PhdZmW9F1vqC2XmuySFS+H6QHji7IlLk/TnvyOIpIy0tGsSG3DssxZm5DZ7OxEcAdvwOA516CTN9BQamctfYz5/9/loh+mIg+sMF1AQB2RpjJdKvJy999k5yzWzhjkk5P9pwVExeRDrtH1bygXoOF4yTqzjXZDI7s7Smx2V5x48HbnhRB0t6q9FHGxYN75rF4UNEtnlfBXHLWzZFqHV1jjnSbKqejLBBA5AIr59Lhyp8Lrc22wgXDnoTvcsnm2eI7vDlz+M70mN2OAQuj8L8uv07AswjiBfOub7MTxoOHZfMgkzHm9xHRHyei/8X72VcYY97h/kxE30xE0Q51AACgoV1oMt1Q+PsgT39f6mLInHTnSMU6xy2+s+IF9RqMOg4aUV7PZnCkbk+JzfbKYmwU7vifjt6CL/MMu+D301Ge/QAmioS/xzbzOuHvS3QLh1KX4ZzS74Dw94TWZlvhhL9PB2GGUEG25PKc/SgI7X9n7pwnRfkhhL/XJdQcDe/vGGQ6IOP1kXla88uNMT9ERN9ARO8yxrxERN9NRCciImvt958/9q8R0U9Ya1/zDv1KIvphY4y7xr9mrf1ba14rAKAOSrq2XeW8vaXjUd5FLCxtk7WqttOuYkL4WyPsC+QsMwAutxlYlxKb7ZUi4e9RuN4t+Hp6LrF/6c7hMlYg/K0jzACQzQvTolwalCK6LLvPlTS5QEBOe8dae25Zfv+BlVugtdkWOJuNwt8KQe2LyuWc8PdBpuvkzqlrvKJ/z4CcJvBvF5lMM5tB+PtRWTXIZK39kOAzP0BEPxD87NNE9HXrXBUAoGbCFrq3ch7arh93djUdfnStqs/6KEcT/Xc14S4idoeuSug4SYW/UzYD61Jis73SlGTincc1p4+Su39uMXE8i93f6wL6XhkFiTUljV43LmlQisibIwszmZyNiRTZJufSy9ebup8Lrc22wLfZ6XigL7/VZo8JN/mKhL+9ZgNE+Xeg8QLbx8NB2HhF/54BOQs/PLi/vs1OQh0tcH9sXi4HAAC35BrCk1r63lJvySsFkAt/a3Qx2t6OGQLW0qJkblkHD8fpmixTwGXlAymbgXUpsdlecePBc08HMkaeiee6YA1/Tx8zLh7OotBYPOiYBLXN+Pcco26OtH37QrvvgpKmg+y58IV+T3gu1DbbglFQuyBD+5Kube4dkAp/u3NOHS1lgVbtewbk+HMNUaxcbrLZPb8DIA2CTACAqhgXlU+3y+aZNAE0HX7CduJSx+jAligsRDfhOF2VxeJMJIQ6t1nti6tbU2KzvTLpYBzEYqtDiZNCH6WbgglYPOhZlicKtY4OikDAtcSZFdfp5qqp62Ddz4XWZlsw2syVON4qgHku0Zt04HLlcudgmKoLnv49A3IWQehFudykyfTENLIB9w+CTACAqph2SAYHQpI6fbVzuvR3hfDk2xQ7fk03d4zCrBiXdq/5TiCnRPMqtFnti6tbA52yiUkHQy622p4zLqT6KI0//t7xAvpeceWJT8IFtq+bMwQCZDYlmuaJopKm8DpzZZTeovJ0NFUvKktstgVTwFhe0lSi+7b8DhegkM2ZofC3pPxQ+54BHVOHwHiZddgREBnejwmCTACAqmhmOyS3mbwafydNOGF2QVmfxOnuept07LsrOHiAJ+yc0wkc05zNwLqU2Gyv+IvGp6NUyNfOykpyY5v7/aRPV+/9LqELMixyNprut0L4221wHI9kTFkJ73idbhGZea/c74/nwErNi8oSm21B45U0SbPPXMDsbSc33pZlyTm9KiKl8LfY/9K9Z0BHM5YjDjYM592F8HfF48EjgyATAKAqZtoPN1rkjE7jUd4pw33mueOBDkJH33X0OTGO/ShmecKCeg3cs+RKMWXdudI2A+tSYrO94p69k0IXp+n6cVwb/i4ri0Kr+jKGzEevVEiYxeGOkYwvnZdtdhI2qohe52HKcMtdZxhYqfk9LLHZFoy+lEKgvBk1CMvnu7DZQM438kWkNf6X5j0DOlwQ78QE8WbC38KsWnB/IMgEAKgKf5Fzq13CNlxYKdLKtYGp03kn2P8O/zqeDtidW4u2s3QwRIfzPZYt0tM2A+tSYrO9MtPBUIhE+4sFqT6K0+Oo+X6XMHQp9TWwckLrk9aR1KZjtu84R5bo5pxLmoTCyb5W16ny56LEZltQNF6cNQgvCd64ZgOT9qRQ+HvsaCnTt9S8Z0CHC+JNGdxBJpNns3vO5gNpEGQCAFRF65dr3KiLTROUiMgWbz0ZM5TYaTqiHGdBpHDiHpwzY7CgXgOnM0BEqvIB32awyW0psdleCTMopCLRR7/Tk3DBd3SLB5RBqGh7O3Tzkt7voFS77S1ZKzvmdC4vKusAZoNxTS78XfuissRmWzDZzPlSwqD0wdD5sbiN8LfXBe+oaWgArcTVWAbx4plMTy77DH7RQ4IgEwCgKpxT44I3t8jmmZ1TKGradHYsNZB2pGsCHYzlxN3PvxOO01Vpu3689ydp9lloM9jkppTYbK+0i4W+UPjby0zICn8XZD+ACa0gsa+bM+mf5AOB5pzdJ21UEbvOWaaCNBBwwTn3QonNtsAXZ5ZnCE0bXadjWeOVYczRB1pPh8PZ54Pw99a4IN7UMGKZdU90Lt2G8PfDgiATAKAq/FrvY2EpQOk5n1x7ZmFre5cdIHW628AxWoopDhP79J2YuK+J26UlonOWht5mcGZvS4nN9oq/0D8KO2+GpSu5sW1ZBlzv/S5hvN/CkudpgS230WKDo6SkKdTekWp1nbV2as5wK7HZFvg202Rou7mutHOeewfUwt+KLDntewZ0hLpa4XPQzeYi2bMF7g8EmQAAVdGdO5MYcxY1vYHzFgp/d5KShXOGCxGJuuD1vaXeOrFWri1sP5YGDd+JBd41cSLIRHTunHW5zcC6aG22Z8byl7Ngs2QBOAp/H4QLPl9UGuVyatwYLhb+DrSOiPI26vogu69gntBq70xzJMrlSmy2Bb7w99PRyIS/OzvOdaUC78352RILf8+eLZlEgvY9AzpG4e+x4UmQde/ZTJr9D+4PBJkAAFXR9P0sc6G7gfMw7fj5Arn5kgXN4rcNAllEMeFvO07qTzcKsNWE2/0kGpyjXBBPYjOwLlqb7ZmZCLp0x/+cCebGlWwp1kzvpyyToWZchtCwUSIpQ5uXQBLlx5ihM5x8gyN1nXKBcgh/O0pstgUz3RyN8Pd5rjudN9y0tJkAReycRGfhb2kmufI9Azpc90ljhnu8zLoPhL/hqz4kCDIBAKqiPXfzIqKbdTcqCSa0nqMv6Ug3dePwuq6EE3c/X1BjgXddmiBTTCqCnLIZWBetzfZM45eyHGVB6HbRJUj4zB/Rqr4Ed7+JhkCddPPhySvJzY0x8w2OwnK58yKS01xZft5l0Q3PRW+HTM8aKbHZFvi6OU74O5eh3Xg+SGnzEfdsTR1ZpQHMcyaTuFxO/p4BHa77JFF8jJkHMFFW/aggyAQAqIq51tE2wt9Eeaex8YJhEl0M9++YdfRZTNwQ/l6TbuZA5xfpMZtB4PK2aG22ZzqvlOUU2V2O0Z4zQ7mOliGh8Deedx2NN3/FMgBin3eflY4xrjMckVwPcHHevh+6eY0Zbjrhb/cdNVJisy2YgmFThnbuOjs/S66w8YoLUEizvDo/aKcQ/ta8Z0DHcoxZSju4350q3/x5ZBBkAgBUhUuDJiJxff41zjmcT9ERJdDFyGYy+buKbocvIvw9/05M3NekDRzo3A5rzGbYsbstWpvtmba3s4WVRlNMLvw975yFMhQd8/krb6NJaF0z93jZvoWlKu05YCkWKB/nyEP1beNLbLYFM5uNjSsEfs1hypa8hvC35JzD+Yz4eda+Z0DHTAA+kqkfCn/fY5AV5EGQCQBQFS7Vmqh8J63knERhuVwuABF0vRI76VO21EKTyXfwClPVAc9MRFrQvU9iM7AuWpvtGf9enAQLwKbXB0lnz/yNxt894QS1iYSbD/0U1NN045pl9xUFAvrZOfPl4b4OUd36dCU224JJN0eXyTgL3hQED0Lhb6nel9OBEzVeUb5nQMdCH5QplzsdhkymWrMaHx0EmQAAVdF4GUJPN9IlchPkzBnLllL5ASFJ6dXUGWraVeyDz/gd6+peUK/BQvNKWMqSshlYF63N9kzbzTta5rOSvHIZhag00fTM15w5VsJska4oo3ZaR0T5zI+mD7P7CgIBSuHkqYzy4M2RdT4bJTbbgrnWkSZDe9I6Kspkcs0GlHpfM7FwwXujec+AjmGM8XXfGOHvsyaTtfdZMgrSIMgEAKiKeXnMbdJwW9+BPjtYua52c62YfDCs8zIEuBKFbrF4qNOJX4t5YDCvNyOxGVgXrc32zEwEXSIq7XUEG8e17DMfCH9XfL9L6LzFWUzLJPZ5IleGpshk8sTwS96J2XWKBMrnz4V/7bVRYrMt8HVzpALlV+lcODYbcIGtXMblXPibSDJO6d4zoGMYY/jnYG4zWZYcuD8QZAIAVMVsJ+3Gwt/ztrt5x2hWsy7csZ47X0vh73kdPBynazIL4gkEc2M2q3VhtRVam+2ZztMhOR3zYrd+d8STMPsEwt+X4WfiaoS//UC2RPjb1+4ryShqOt11zrLi7jiwcgtKbLYFc5sJhb/7y7OpF8LfWbFxN07Js+S07xnQsRxjAmkHL8v7JLQzuD8QZAIAVEWYyXRL4e/TOfXXXUfymK6fOj0ptDcGsVVG+NtfRB5krXyBnLnjJNHRWtqs1oXVVmhttmf8jj+S7pNtJENAVC5zMGSMEeujgIm2m4ty5zLBRj3Aw8HLCMjbddyMKCwVWlyn8Fk6JTZJaqHEZlswdZc7yN9/X+uoQBdybDZwOMh9qZn2ofQ6de8Z0DET/o6Wy52Fvz2fOZf9D+4PBJkAAFWx6Jxzi0ym3ttJE6b+zvRRDnJ9lGNCHyUsg6hZf2YNmn4STj4KdLQkNgProrXZngmFv/NZSdNu8/EgH9em1tX3u4C+V1qvtbpE08aN8cMYI2w60U8bHKXZfW3fz1qUS7ujzgKWlc5PJTbbgrjWkVz4W9LBcnG8V1ap1fvyxylJ51fNewZ0tL0NxpiwSU0k0FrpePDIIMgEAKgKP42/tBRAf85IF7Gs+Kqu69W44Dumhb+nzi7QGbg2fpbcSaB5JbEZWBetzfaML4IuyWBpS8a1WYYAnnktTT/vziUW/vZKryQ2movh6+YJa+1gZ4V4eONlxYyll5XOTyU22wJfN+covE6/w62kg+XieK9Ej0io9xWVK8g/j5r3DOhwulpE8YYbofC3OwY8FggyAQCqYl4ud6NMplnnHOFu8uI6da3BiZZOuu/gPRWkqoM0C80rTRp/5QurrdDabM/MRNAFQWi/1bp4XPO7e+KZV9OGNhIs6t1nR3FmjfB3QRaHL1xNJBNOHheVB3np5V4psdkW+Lo5clF5XQfL2PHDsf6YLRT+9qUEVOL3dWe4rkHjC39H/PCZzSD8/bAgyAQAqAq/a4hEK+I651wKeUoyk46ek5MT1BzFQmf6CMs69yPa8q5G2weZYlJR3oTNwLpobbZn2m4u/C0vQ9G1BvczGYbz1nvPtcw0xETZkp44s7AkN+xsqp0j/eC5+3/uWYp22qz0XSyx2Rb4NpNe5yxDW5CFFDueaCq1lUkJeDpwUl0y5XsGdMzHmGVJrm8zCH8/LggyAQCqopntpB9u0jXE1wTQtNA9eU5OzuF233dKpK77dfCDTgYm7WvS9XYMDJ4EnbMkNgProrXZnpkF4A+HrNCqu1fH82LgKHrm+5kmi/89II+vKXgULLCnMeYgHmP8QGBJm3lfT0V6nVOJGDptlthsC6ZgmF/unXkeA002rf81Bra8DThJANPXB/O/h0P7ngEd86zZ5RgzL912NkOg79FAkAkAUBVh15CbCn97bXdzQaNZhx9J6ZWXRs7t/LSL0iBM2tek6eaCudewGVgXrc32zCwAf8wHtn1NMaJz1ougVb2v9zOct957rsV15yOSadrM9WhkGlitFwiULOJT5yQSCpT75XKVd9ossdkWzK5TKKg91zrSj7dN8GyJ9L58n0/aoED5ngEdywzi0Ff1NzxQVv2oIMgEAKiKppt3vbml8PfxYMbd3bxjNC1+j4K08ibipIeOke/glXR2AWnCBXS+007eZmBdtDbbM1q9urAs6iRYNDa9HcdAPPM6nKD2bKNAmiF0OEwi0oLg99RZq6ADWFQ3Ryb8jU6bZTbbAl83Ryz87WWwSDK0l8cHwt9C7Th/jMpdZ8l7BnQMY8z0HCykHWYbHnVrtD0yCDIBAKpitoNyY+FvvxRAJPyt0EeZiUgzugNzB0+/Qw3SDI6RPMVeYjOwLlqb7Zm5CPpQTmstfz98TTGi8/1TtAaX6qOAga6fL7AlmjZujD96WTHZ4HfQ2VTdAayfgufuO7IC5TPdnPvN3rkFJTbbgpjNNMLfxwL/yxdFH/4v047zxyh37Rwl7xnQ0cyyJSPC30HJrjsGPBYIMgEAqmIu/J1fSF3nnD0ZM9+llXQFGh0pgT6Kv+Bz+ijLtrDzzi61al6sxbxt91LMcvn5vM3Aumhttmf8APxJkJkwBc+9TDCJ3o/XdWw4R733XMMyc0wn/D1ucAh0c568QKC2nHHZZj5/naHYuH/ttVFisy2Yax0VCH8LspAWx3vNBoiEmzndPKDhf0/08wXvGZDT95asnetqhQEk32bS7H9wfyDIBACoimG3Xr6Quso5C0pE/MXvSaCPEi74YruEvoMnSTMHOmYi0godrZnN7nAxsWe0NtszYQDe/Sz1eaIpO0DS7t7X+5myCuq95xpGDSwvKyMrYOxl0YqFvz09QImeEnudigzBYb4Luw7W+VyU2GwL5lpHQuHvIDClDTBPHe3kz2dM+Dt13pL3DMhpxmw0PwidEP4WZv+D+wNBJgBAVbSeM3u8kfZD6wW2pAuroaxEIfwdpJHHtDR8rY0ntOW9Ok2nE8yN2azWhdVWaG22Z5quJ1+Tyf0s9XkiT/hbGFh9CrMfsIATMTUKkGsKtn4WrXCDw88iOB4M9XbIPhBfZ1SrSyC0HM6Rlb6LJTbbgrnWUd5mTuvoNAtKXyj8LdT7CoW/U8eUvGdAznJzbTk+zLrP3XGgFaRBkAkAUBVt34/BpdGBW9mBaLrYTpqkha43CWfK+qa20dOC2W/5aq0dvtOlmReUQYA0/u7bSaHJNO9uBmf2lmhttmfCclr3M/bzsWCCsgx4OAeeeQltkMUhFWc/BdomKuHvgkDgVC6nEyj3Mwol17lXSmy2BXPdnHz22SILqSCTaZoz/VIrmd7XcM68/1XyngE57l7OMogDewxZtYHwNwJ9DweCTACAqvADLbdy4Dp/J02w4zf8fr74dd/Dfr5bOnC+8xVLM0cK+PWw1i7Kjay9zGZgXUpstmdmASAn5CsoK/F3nLPjWue3rq5be0eLu7d+UDor/N0tO2vls5/8+Uq/wBuvc9RYyl9n188zCrXn3BMlNtuCmc0EmY9hUFqShRTivt/vfthl7k3YKY4oHdguec+AnMX4EAkyzWxWuUbbI4MgEwCgKtqZw3EbB67tvZ20g+vCwk+YscXv8D35Tk+cPkrMwetuIHpeC80iBVxebqTRtAHXo8Rme2amA6coK/HvX24h0Mw0sOoOJmiZMgDkmjZ+Fq2kVIhoPl+VLPCa2HVquqlWvqgssdkWzMrvBRlvS60jfeOVkjGnjQXDBJs/l2hHAZ7RF/V0SsM517dZ7RptjwyCTACAqogKf688ec0EMgU7aeHid2pvn98l5NrChg7e1BoZE/c1iJUCEMnS8lM7emA9Smy2Z7p+GYCXCH/73eJEej/hArrS+61loYEl6Draes0eJCLSTjdn0QFQEXhtw+s85rNmZ80xCs65J7Q224q51lF+I6wLNrpKGq+UjDmz0itBiW7JewbkhLpasYYbvnafNPsf3B8IMgEAqqKN1XrfQPhbs+MXCkJLxMLbSOmKv1sXilnWLq56bcaOKQpnNmYz7NbdjhKb7ZlZAF4Q2F7cP0GQdKb7hNbUKmLZqJKuo84+xphsw4dxEa+Yr9jr9IKJ+evUzZF7RmuzrZhrHck3zy7xv6YSXfmYMxORLtCak7xnQE4beQ5CP3SYiyZpB/848DggyAQAqIawDG1cSK0t/D3TV3IZRHpnLK2PEmbFzIW/myBwhTr36xI6TtMiPVU+sLQZgn63o8Rme8YPwEsyKBb3L1L2ENLMdJ/yC1MwEdXAypUnelm0RPlF+ULMXTBfsdepEE6eZU8JSsr3jNZmWzHXOsoHAhbdVAv8ryn71y+1kge2JRpXJe8ZkMN1QvbLJv3ssyMymR4WBJkAANUQlie5hdTaQr9dN5UCHA6GjBGWoQT6KLljjgdDxni7x9180p59p6AcBshpw4WVsNwoZTOwLiU22ysLHThRMDy8fyZ77zqv8cK0yKzvfpfQ9bGgdFrTpvPazBPlA4Fh966SOTJ2nRrh7+mcdS4qtTbbiqjwd8JmiwwWZ+eCTCZfLykr/N0vRaQl/pfmPQNywq66TqPPN4lvs9o12h4ZBJkAANUQdvOSLKSuct6o0ygofVPsEjaeWCvRUqtgmaKMLIJr0oSOqUT4O2MzsC4lNtsrnA6cpKxEE0yYNV6ovDxRSxPMXxJNm8bLCBiOTWdlTHPPfJ4oEf72M4ZF3VSRUUhEepttRdvrBMoXWkcFjVdiwt9ZIXtfB06gRVnyngE5Cz88UpodsxnmiccDQSYAQDWEGiKShdRVzustrIic05gKPhQIf2dS7EMHbyxJgON0FUJ9palzVrrc6BHKIvZKic32SljCIGkjv7h/B0m3M79V/W3G370wCmorNG3arg/GmHRJ7mKBXdABcNGi/HAQlMtNZZTGmCF4UGkmk9ZmW9F0/SLjLalBGJZiFrz/y5K7g0yXbHHO/Limec+AnJjwN9F83p1p91Wu0fbIIMgEAKiGmOAg0fq13m2QsfJ0yOhiBIs3mT7KJJzqzjHbGQocPIlTCOSU3N+YzeDI3g68ExOhWLOk82Z4jCirYJb9UG9QrwR3nxbZI5kSJc0Y0wYbMZIMldg5Z9cpyWTq7HKOrHQs1NpsK/xAgDHmnLEm13A7Fgg6h6VWknsz+F8KX6rgPQNylo0BlplKvnZf7RptjwyCTACAaogJDhKtP3mFDvTpmNZYmHaTw0k4XRbhdoTcMf7nF2KW2B26KpMDHYiaam0GR/ZmlNhsrywXgDLh74MZdOaI8uMaEdPpCc+8iJgGFlF+Xlhk0WrE3MfyIr3w9yTOLAs++td5EpRe7hWtzbYitNkwf0kCmBcIfy9KrdLnnK5TLmRf8p4BOZJ51/eZa9doe2QQZAIAVENYz3+rNOi2j5RFCXaTx9I2wWKs9RZv4zmC9GP/u05IAb8qYQr4lKWhs1mtC6stKLHZXuF04LKaYsrF8FAiGi748MxL4OavtNjypG1CpBD+DhfYiiDHsgV8Xji57efXKcl+2itam21FLEP72ptny++YC38PAUxJltzUeOVgciWm+vcMyFmU00YqCqJdoDFPPBwIMgEAqiHcQXkq2EkrO+9c+PvpkNYR4Mr60mLhdnaOU6DjsNBcqVjkeA2WIsh5xyhnM7AuJTbbK5wOXG4xNgsMCLR3Wi8wheYDOhaZIJKSxmCMUQt/FzTHiGl1DdevuE5Ba/q9orXZVixKHIXPVhjAVD1bkTFbovc1yxjO6DiVvGdAzrIBz/L++jarXaPtkUGQCQBQDaPwd9BNStNCt+i8EWcs1UJ3sRMsFNWcZUsFJQpsd7k7TMN/RDgR6ZydUzYD61Jis70S3gtxJt6sxEnWReyEMaiIWDc/opxuViAinSnJbZh5QvNOlFxnE2Z1HvKt6feK1mZb0SgFyhe6byXPVmTMzgp/98FmzsEkfb6S5xfImaQb+HkgtBl8o8cEQSYAQDUsMpluVB7jC08O5xXuJh+CnZ6M3kFK90kitgjKKQsMLm2GBfftKLHZXmE7P2WDpPJMhq63ZK2viYfFm4YwA0BURt2HeoAZTaaemXsKhL/V1/kA2Tu3QGuzrYgFb0SNAo7l/tciACTR+1qUpWe0owqeXyBnOe9GhL8Dm9Ws0fbIIMgEAKiGNshkulWt98IZ0wp/C1qrN91chPMYdLBjxSzv0Hl9RJalmE7HQWmzCgMcW1Fis70y6ZDM70WuFGveBSs3rs3H38PBkDFYvEkJMwAk4uzxMUaeRVsk/N0z15kr6wsXlRW+h0R6m21FaLN88CaewaLxv9yYY4xG72sZtBONU4r3DMiZNjzTwt/hBkatmY2PzKpBJmPMx4wxnzXGfIr5/TcYYz5vjPn583/f5f3ug8aY/8cY8+vGmP9ozesEANQBK/y9uiZTRPhbsJu8EP5OZcV0gQjnMV4uB+HvdeBKMbU2w27d7Six2V6JCer6P48eE8vQlJQBL0SN8cxL4DIAkkLeizEmI/y9CDYWCH9zmoIZHZx5p816A+5am23FQvg7G7y5gvB3RGycKFOWfmGgtUSXDPBM2lx8w42ut3QM5pZag86PzNqZTD9ARB/MfOb/sNa+//zfnyciMsYciei/JqJvIaKvJaIPGWO+dtUrBQDsHq4UYPVMptBpzHX4CZ10QRemZeo6I/wd/tuxO3QVxsVZKCp/gc3AupTYbK9wIui5spJQUFdTBuzOU2swQctSj0aQIRS0mc+Wy4XZJhcIf7uAZZlAeb3BR63NtmIp1q7rmlvSeCUmNj58dyKbbxHAzDVe0b9nQM5C+Dsy78Zshnni8Vg1yGSt/SkierXg0A8Q0a9baz9trX2LiP4nIvq2q14cAKA6QlHTW2XzNItyucM+9D8AACAASURBVLQz1nRxZywt2NqPi2V3zLxbR6hlgEymaxJ2ZToJsuRyNgPrUmKzvcIKf+c0xYLFcGrxFo5Bw3nucwF9jyzGcPEYI1+sNYtyOf0CuzlrdfklTbnvWGqw1NtNSmuzrVgIf2c2ScLNsxJNtkWzgUwQdNSBC5+t5LO4TcZ7LYTltJMfPvw8ZjP4Ro/JPWgy/QvGmP/bGPPjxpg/cv7Ze4jot73PvHT+GQAAFNP18R2UtZ2Hrg+0C3LOGCNInE8JTwl/xzWZUOd+HcIW8BLB3JzNwLqU2GyvhKUsEr26RdfMw4Gs5cepcAwazoNnXkp4/06CTNzlGJMuye3CjNeCBXYotDw+S4nv6PrwWap3Uam12VaENstdJ7t5ptL7Wmr1uGtJnTPMkkv7X/r3DMgZ/fCgNNv5vTGb1azR9shsHWT6OSL6p621X0dE/xUR/U3tFxhjPmyMedEY8+LLL7989QsEAOyHcPK61aKy6fpl5xyFdskkvioviwiFvxe7iBWXBq2BW5wdxxT7vFh7zGaads7gMkpstle6MeA2Dy6knscu0hGMiA9IhGMQEZ55DeG8cJRsPijHmKVuToEmU/BcSK6zCa4ztxGzZx5lXmh6S8cgeJMbL4i8MsoCva8uFBvPBIDCgIb7s0iTSfGeATnTGBPXfQtLt4nq1mh7ZDYNMllrv2Ct/dL5zz9GRCdjzLuI6DNE9FXeR58//yz2HR+11r5grX3h3e9+9+rXDAB4XJYaLBsJf+ecnLB0RSL83Qcp9qHwN9c2tlJH/tqEpZilNsNu3e0osdleacaAWyD8nSnFimWscGNbGwSy3J8R6JbBdQBM6vstxpicHk1wjoKNmOV8JxMoDzNUqs1kUtpsK9qw3PuQEf6+gg/S9OGGXfo7Qv0fouHZzmnNuc/5xyLj8jqwwt/nn0+abvymKXgMNg0yGWP+SXMu2jbGfOB8Pa8Q0d8loj9sjHmfMeY5IvoOIvqR7a4UALAHWm4htbbwd9/PW30LO/w4Z+o4LvgUZRGh8HcQuDre6N9eC8vFmeuopLQZHNmbUWKzvbIIwAvLPUNx1tQxoQaW+zMC3TLaridjIpkg2XI5X8NGJs48zpEFC+yw61juOvveUm9pOUdW+B4S6W22Bc5mqgxtF1y4oNHCQvg7M0654FwYwMwFPLXvGZDTBhlt4XMQy2Q6HeEbPSJPa365MeaHiOgbiOhdxpiXiOi7iehERGSt/X4i+jeI6N81xrRE9DoRfYe11hJRa4z594jobxPRkYg+Zq39xTWvFQCwf0JBx1sIf1trqensbGcy1y1mcoycJkBehDfMKng6Gurt4AweDmaTf3tNTDoOYSlmuc3AupTYbK+EAaCj4F4MwYT580vEZz+FmWPuzxiDZDQL3SKZ8PdTELxJj0nzYGNJSdNCqyuTbRLOd8P56y2P0dpsC2I2GwTKW/aYcPOspPnIstlAOgg6lejKM8lL3jMgpwmCjWHDjajN7rTDIkizapDJWvuhzO//MhH9ZeZ3P0ZEP7bGdQEA6iQU/j4eDBmzrvPg5sWF3oQkkynY6clrb8x3fogGZ/BthyOEv1cmXJw9SbIMMjYD61Jis70SiqAbY87d4tL6Pc9OikymQFR6OB+Ev6WwgtqqMSZt024ReM1vcESvMwgYpa4zqptTdbmczmZbELNZVvg7CEyVNF5ZBjALhL+zXfD07xmQ0/WWDobGjbQwGy0q/H3APPGIbC38DQAAN2OcvA7h5LWe8xDvbiLrwjJ2N5EIfzOCmNPEHQh/H/LfCeSEizM/YMTB2QwCo7ehxGZ7Jewq5v6cE/KdZTKMi7FMJlMwFuJ5l9F0odB6WpzdWhvtbKoR/i5ZYDcLQfj0dYZCwO7PNb6HJTbbAmezY5DxkxwvruCDLMYcYQBzWS6Xfgc0zy/Q0YTloEE2WlT4u+Kg8yODIBMAoBpiwrPDImc9Z3Y8p78zJli8EcVaq2cEXiMdfdzEvBSz1O8iAp7Q6R4DRhn9k5jNEPi7DSU22ytcACgp5NvNteZOmayCsIMd0SDuem9ZGvcKv8BmSoUii7VjRpw51M0xxqi7m3Wdnc13x9x1BkLAw/nrDD6W2GwLJpvNfalkowBGi0fjfy069WZK2SaNsXlQQxU8zzy/QMcgGL/MRhu7y3WRDY+MWDu4TxBkAgBUw9S1InTg1pu8oufMiN2GwohFwt9Bnbtz5NxlIKBxXcIg3kU2gzN7E0pstldKtEvanhH+Zjs9xTJJ69Xe0RIKaoeLs8Xno521ZO3bj2GQQxEIaIMOYNKug/N5+VBl5kKJzbYg7FZL5MSZ08LfTwdD535PxcLfs452me+YSqLnOnC5joya9wzoaHu78MHdz/3/L7T7sCH6cCDIBACohlD8evhzuj5/lXNmAluhMKIx5rzg04lIE00TdnNeEDoHD8Lf16Xt5x1pnKbNJTYD61Jis70SiqAPf76udsnUwjwQdMUYJKLpwgyLdFC66ZdBvZxNQ52y4c86G4XlMLngYzjfETl9n/rewxKbbUHMZseMj7LQmioU/p5nT6bfgXjQLp0Vo33PgI6m6wOR//n9jdkMDSIeEwSZAADVEAp/E60/eUXPmekW03ZzYcThmHT5QBt0RJn0DtzEPV8QOtFzCH9fhybYYSU6O90X2AysS4nN9goXXNBolxwzi8ZoJlOl2jsltEx5ImejsIEE0WTToZHzklCnbPizrrtZmAmSu84ukhVTazepEpttQRfJPss3Cuhn48uhoPHKcszJCH9HnudsMEz5ngEd3UKzLdAPjdgMDSIeEwSZAADVENslHCavGwh/BwGjpDMWtOklyguUt+HucThxB87Z+J0VOvJr0AUlIkT5jiiczWrUItmCEpvtFS64kNUuiZWusBkrkWBCpdo7JWgFtcegXrDBQTR1PV2cIxbkUHY3W3YAKxD+rvQ9LLHZFsTf5fx4ER9vy4W/xwBQTgcuGKfSXfAg/L0mTdDwJGy40UY2PGoNOj86CDIBAKohtkv4lNFHuvicMe2CQz6TydcQIMrrODWc8LebuPt5irL7DFLAr0PTzXUGiPKlQJzNsGN6G0pstldKhL/bvp9p9+TKX9xYtBD+rvB+l9AGmSCjIDEX1GNEpIlSYuH9Oct1HjxQZTIF5TB5gfLY5k+l72GBzbYg9i7nGwXYmQC3O0bzbDVdoJeUKxmNaGKeDvkyYM17BnSEmm1hJ+Qp0BqOQfWNB48OgkwAgGqIiXCvXS4X75RhqLdEfWIHOsxkekosxrrekrUU3x3yuss9RRbUWOBdh1gQL6WlkbLZvelv7BWtzfZMifB3EwTDp+zJjD7KIphQ3/0uIWwUEC7Olp9fzj2jCHdCLDw2T+h0c5hymJxA+Z3rEN2CEpttQfxdzgt/n4JMpidleXIbdn4TloyegucxfZ269wzoCMeYqRPy8OyP2n0Q/n54EGQCAFSD64a02KW9QSbTKea0cE53v3T0T0fD6idFRXsDrYL4d6bT24Gc2OIsJZibshmc2dugtdmeCUXQifJjYxgMzwYwxrEQralLGII3c109okRQL6Z1lMn8CBfxwzHKQEBQDiM5JxEtsp9qfC5KbLYFMZsdM4GAMPjojtf4X0u9pEzJ6DjPyoW/te8Z0BFu7oyNbc426SI2qzXD+NFBkAkAUA1DoCWWrr1mJlMkQyCTsRLupA3HJAIWTNc8okD4O7p4gON0DcKONETpBbTEZmBdtDbbMzER9JNWuyTTRSxWBnHKtBMHE23fzzLHXDdETo9mGmPkHQCHeSKce5SBgKAMWHLO4TzzZ8na+nRwSmy2BbGs8JxAeSj8TaT3vzjhb75zIZMVkykD1rxnQAfr37pOyEyWHOaJxwNBJgBANTSMA72m8xDrlDFpVPAdUcJgWEqkOx7ImmcVhAtCovVFz2si1BkgSmtUSGwG1kVrsz1TElxYaJdkFsNTGQR2qEuINW9I6SXFm06kA4FNbCMm02UwZNnQIHPOxFhY27tYYrMtiGu45YTo48+v6tnq+6AD5lwWIHZOomX2ZLoLnu49AzpiY4zfcCNqs0ozGx8dBJkAANXAl8esWC4XcaBH7Z1EACLULkiJdMcDWctMpmi7djhOVyFWjpjStEnZDIvu26C12Z7R3ou+t9Rb3fPbMgtoLB5ktN1y8yHVdSleepXXcYqVVavEmftQnDlXRhlrWV5nwL3EZlvACX8Pv0vJAMSeX/mzFXaoy+rAMZt86cYruvcM6IiNMf5mQ9RmGb0vcJ8gyAQAqIYhc+HW5XKxCTPvdC+vk9/xS2bF+MLfsawNOE5XIeqYJjRt0plMCPzdAq3N9kxMBD2lXTIuhmPC37lnPlzwVXi/S4hp2qQCgVzTCfdd8WPSpSyi6ww2c/KBrVhJ0/3pEN2CEpttQVT4eyyXTfk1lwX1wxLnMeNNJSqfabyifM+AjvgYM827zs+dZayt3AUarAOCTACAamiDbkhE55KxGwt/n4Iso5AmlnGVEP5uYoGsQPi7CcQs3XV0cJyuQsxxSmnapGwGZ/Y2aG22Z9rOzvRViNKZjlFB3UzpShPLfkBrajFNF8sE4YOi0ZKmAuFvv5RFfJ0FAuX3rkN0C0pstgWc8DdROkM7liWnebYWwt/ne9NpximnfZh4bzTvGdAR3dDwSnKdn3sMtLdSgUFwnyDIBACohrBrCNH6O1Sjk5PIMloc00Vaqydq0qOBrIjwdxhg06aqA55YEC/lmEpsBtZFa7M9ExNBTwdJE5lMWR2yeVkUnncZsRbw/uJs8flIUC8rwt2nS1lE1xkIf+cFyuPlMf6/oRZKbLYFUeHvnCZbrPmIMksu1JZ0gQjuO6asGF3JqOY9AzqipdleSW7MZrnAILhPEGQCAFRDE6kFPx7W1QRxE6bvjE0dUfigUZhV8JTYTe5Sgazejv9fficcp2sRDeIlApgSm4F10dpsz8RLWVLlnmWaLETzsRBaJ3I6LiiayRCKlqElsjiiGW0KG3WxTJCUQHnkOo+ZQMBeKbHZFjSRTZJs19x+qTX5pPS/Qm3JvPD3MpNpuk5ubta9Z0DHMMbwm6hRm1U6Hjw6CDIBAKqBK4+5hfD3KZKundIuWThjid3kcecnJ/yNrI3ViOk4pDRtUjbDovs2aG22Z+K7y3xwwZXhzsSac6VY5wwBY+bPfNfzbc/BRNNz2aiazlo5TSZOm0sn/B2dv3KZTIdl9k5t72KJzbYgqh2VyxCKdRVT+F/RZgNZ4e/IOJU5RvueAR2xTLG58HekZHRslgMbPBIIMgEAqiHe3WTdRSXX3YQo5YzFupskRHhjItLBOSBmuS5sR5ps6VDMZgj83QKtzfYMG1xg9ZVSwt+K7lKZrE4wEW8BLxD+1nQAvMI8ob7OaEe1+9MhugUlNtuCePAmI/wd6yqm0GSL6UDlOxemMsMU4vfwla5GNFPskBf+JkK53KOBIBMAoBpCrQgiV2t/g0ymwzKTKSn8HblOtsQhIpQYdnqKiVnWmrWxBtziLCeCHLUZnNmboLXZnlELf0czGTLC39HW1fe3gL5X4mUmCeHvVCAwEfw+ReYJ6RxprT2XZiv06aLZvvenQ3QLSmy2BemOtgnh72iHW9m/a9K3nOt9HRO+USrjKuV/ad4zoCOWKebr/8X0vtyfO/irDwWCTACAami7WK39jYS/owEgPpNpWdqW2rHmA1l+nXs8RRmO0zWIdqRJatokgo9wZm+C1mZ7Jtp9MiHWHO2CVbDIxA61nOgmSUKcPVWGxopwx7S5FKVCY7ZJuIgUCJTfuw7RLSix2RakMrRTGyuXNF6J6VvmviNalp5rvKJ8z4COeLBxmndLOjKD+wRBJgBANTRB+1ui9XWJol2YMhNmrNVvSvg7nmIfK5dbOnhwnK5DTmcg9nn3mfHzELe8KVqb7Zm4CHpKVDoi/H2QBDCYcrkK77mWVtlavY0FAjNlaDHhb1VJ03hOuXDyNEcus2Jqm59KbLYFsZKmXGBwyJJb+l/SwEGs2YD7eyqAeTwEOnACTUzNewZ0cMLfYyZT1GaYJx4RBJkAANUQ626ydmvaWKeMU2bCbKIi3fxucqrTi3PgGm4RCcfpKsSCeKfjgc3QkNgMrIvWZnuGK2XJZU+GpStPiRK7hllkDt9X3z3X0sSCooKS3JiIdGpRvlzEy0vKp3PGyqLkAfdTpc9Fic22ILpJknu2umWQ+VSQJRfPtJNv2J1yGVfK9wzoiGeKTUHo6CZrJjAI7hMEmQAA1RATnj2euxuteU6iuWOUqy/vet1ucjeWG/DdeaLfiY4pV6ONdaQ5GOoKbAbdgdugtdmeiQcXEs0GIiK8RMOYkhrXYotM//sAT8d0Q0zdb6J5INDZizsmtsA7HnibLs4ZCT4SDRkv3HtVMkfulRKbbUFK+Jt9tiKbfE8K/2sqxYyVOKfOucxKSl2n9j0DOrpIppg/xjTd0ma5wCC4TxBkAgBUQ6w8RrNLW3TOiGNUJPyd2K2L6aMcxzIUT/hb0T0K6IgJJz8dDypNm9FmcGZvgtZme2YIwEe0TjKCuuH9Ox340pWmW+r9HMdSoPruuQYnqB1bnGXL0Lxxfxpj+Jbv8XIkXSZTONekxZlT3aTqei5KbLYFJd1RW2ajS1suF+/8xo9TsXMSxf2vkvcM6GgymldxXbj7C7SCPAgyAQCqIb6oXDebJ7YYEwl/R52cXOmK56QHbXpjKcpwnK5HLIiXdn55m6Fc7jZobbZnUjoZ1i7HnViQlGjolqgqXYHwt4hYK3YiqfB3TECXPyYqrCzNNuEymVLX2fdkDNEhJvxd2btYYrMtcDaL+TX8xkofabRwBeHvRFlvE82e4v2vkvcM6Ij64Ye58Hdsw4MIvtGjgSATAKAa4t1NhjTo2ELqKueMdTeRCH9H0rXZz0c6vRwOhg7GE/6OpCinukcBHbEgXlLTRmAzsC5am+2ZuAg6v3scE/4mcg0KEsLfkVKX4fz13XMNrKB2qoNlpKQp1wEwlkX7pMh4TQp/JzJxFyVQd6hDdAtKbLYFUZtlAoNtFwn4KAS1uTHnlMg+baOBLd7/KnnPgJwxUyzRPTlqM5RVPyQIMgEAqqHrl8Kzp5WdWaeF5HfKKBL+TuhixDq9ELnSH0/4+wI9BJAmHsRLCH8LbAbWRWuzPcM5/u53i88nMlaSwt+RoN7wu/ruuYaUoHZqHiEKOoBlsk2GOTK2GXGh8HdCOLmLlMeMwt+VvYslNtuCmM0kovKLkn2FoDY35gy+kWbDjve/St4zIMf5m0s/fJp3o00okOX9kCDIBACohjZWHrOyM9v0vbpTRjRdOLNbN3xm2UVl3B2KiFlq9BBAmmhWTLLEkbdZjcLTW6C12Z6JiaCHJbfzz/PC36kyYE74G8HuNHxQLyF63Fk6BGVozqZdIis2tsCzVmajWBmwu87UJglXHlPbu1hisy1oIiVPIuHviF8jzdDixpxjIlCVEv6O+V8l7xmQM2Xq8WWTTVQXrs7x4NFBkAkAUA1ca1qi9dJw20SnjFRHlGir34SgJlFEH+WsL2OtHTqmRBx/TNrXgRORbplSzKTN4MzeBK3N9kx09zhRppMS/k4FPZady+rMWNEyldfKBYlj88gxU3YS6+w0ZbTlbRQrA85fZ7yzYeo690qJzbYgZrPJl1raeZAkoKXum0ITk92YSQSq2j6uMTZ8H18GrHnPgJwxUy9SNjkKf3f6jVlwnyDIBACohmh3k5V3TGNCqsfMOWOT7PFgqLdEfbR0ZZliTzRpFXBilqmOP0AOpzPggolRTZuUzeDMrk6JzfYMJ/ztfrf4/JhVECsrUXQuQ2tqEWMHtlhXpkSGK5edphX+9q+h+DoTjSv4ObKusbDEZlsQF/Hnr7NhM3flGUINkyWXClTFsuRSAcyS9wzIcfc3Nsa4ksdYllytmY2PDoJMAIBqiHY3cQ7HSs5stLvJ6Iwtz9n3lnob360bvi+SVcBkxTwdh9IrTszS6SHUlrVxbcY0fnb3TZ7JBO2H21Bisz0TF0FPlZW4LlgK4e+Ob02NZz4NV4b2dEhlcSR0tjjdrGhZtdxGYyZI7DpTwt/Mdda2sC+x2RbEbJYKDLaJoLS08QqXJXc6JJqixPQtEzpwJe8ZkNMwmWK+39PFsuTuMJsP5EGQCQBQDV0k4HNa2ZntmK5uRPEJcxKejGcVxDIs2B2+wyCmyItZDp/HvH0ZfEca3pnN2QysS4nN9kxMBD1ZVtIlgqSJYMKyDMKNv3Xdby0Ns8DOCX+zrcCVwt/+NSSvkxNnTlxnVPg706lsr5TYbAuSwt+JzF1ujJFkabHC30e+KUpUizIxrpW8Z0COSPg7ajMIfz8iCDIBAKohKvy9sjPbMEKq3Dl5R4pPRee1CgbHKOWcDd+JifsSWjaIl1qk8zarrVRrC0pstmdimUzJshJO+Dsldt8vM0knUeO67reWlKB2UgMrsKkxhu3GNWr3sfOVIhAQ0VhiS5qijS4qfQ+VNtuKJiI9kBIo5zOZ5Fo7RcLf3TJoKhP+lj+/QE5qc2cU/u56NtBd23jw6CDIBACohnint3U1QeLaBfw5cxkWscCU2zmMBabavvcCGpxQJybuS8gGBiPObNJmcKRWp8Rme0Yt/D2WRcUEXOXtxDWi0jXDado8JXT1mkhQz31HNIvjCpsRbCZISvg7UsaeKg/fM1qbbUUbCQSkSprymaOSAGZC+Jvt1BvLkktozRW8Z0DOeH8TDTfi8wSEvx8RBJkAAFXguptwzuxak1esbbcxhhXd5kvbUg7cEDwzZukYNZ3lRXorLUm4Ns5mR66LIJPJxNsM9libEpvtFSeCrhL+Znf8EyK8iY5U9yRqfI+wmSAJXb2hs6lZ/JzTzeI6a6XKu2PnJFo2NHhKCpTrSpr2jNZmWxGzmUT4e5H9pNDEnDZm5NmTURHpxDlL3jMgh9Oi9Dsux7PksCH6iCDIBACoAk40cv3ucstMJnfe2ITZMY7UqFHBHBP+u4gmrYIusSDkvhPImRZWcWc2Vv6WshnssT4lNtsrrAj6gQ/Ac1kFT4luUV1Ukwk71BK6MXMsXvrDjzFMVkyqgQTzHEjKtTp2ntUJlB8rXVRqbbYVSZsphb+JZONty2TBJEtGlSLSJe8ZkMMKq3v3Nyb8nevIDO4TBJkAAFXAtaZdu2tFrG03Eb/7xqcTJ8rlIroDw3cMpSu8mCU6O10DXockLfydshlYlxKb7RW2lCVV1svtSB/5sqgm1jkLmUwixswxxfwVE5EmSmQyjXNkuY2axDyb0uoKA5xOh6i2LFutzbYiZrOUQDkn/O3mQImoebLZQKpEl9GOir8z+vcMyGG1EL2GGzGbpToyg/sFQSYAQBWkWtMOv19J+DvStpvIae8khCeZgBBX5hA7x1L4G5201oAtcUw4szmbgXUpsdle4e5FqpTFjRmLsqgDryk2tBPnzlHP/S6BCwCNi3QmeyS2wcEFAtnsNIWNWiYTxO8eFdJESq/cddT0HhLpbbYVsTK0VGAwFXwkkpbLxTvvpoS/my6RyZTwvzTvGZDTZDY02s5GbYas+8cEQSYAQBWkWtMS3Vb4m2iYNKM7fowuxiTSHXfgQj2V4ZhBEJMVs8TEfRW4EsdjIoiXshlS8tenxGZ7ZSqnje8uc6VYB0N0CI45Hg0bTGij+ijuHPXc7xJGDTH2/sUzmcLPu2Oin2ebEbgMFUEg4PwsLa7zkGgzHxH+JhoW9jW9h0R6m21FGwkEEPHXmR9v8/+2jnk+Twde+DveLTGhNVfwngE5XMmj33CjTXa4rGs8eHQQZAIAVEGqNS3RysLfXAAotePHLPg4EemoWOhx2OHjxSwxcV8DXtS00GZYcK9Oic32SsMFtkfHP15WEg8MJER4IfxdDCuonbh/SRHpgpImUSZTouSOFShP6dNV9lxobbYVnM1OTFnf1GiByWRUBDBjWTAa4e9U45WS9wzI4XTf/IYbbWRuQSfkxwRBJgBAFWwm/K10oHmtmFTpFXOOsyPFLh6QyXQVeL2vhHByymZwZFenxGZ7ZQrAc2NjvLRKUwbszgPh7zKypWzRMYbb4GDK5RidLU1JEzvPJoSTY4tKd0xtz4XWZlvB2ezIicpngjeyAGah8Dd7PxOBVswLqzCNMfENDVcuF9r4cDB0MHVt/uwBBJkAAFXALSqntrsrZTJ1nMZCfGcyV9bH6aNEHanj4TxpM4sHxQ414GFbfydbwPM2q61EZAtKbLZXSoW/NWXA1tpoRypkCMho2GxUfpHOax3phL9HMXzBZkTDBRMSGxpDB7BYVgyvtbNXtDbbiiEYphGVj/s1Gv+LC4IeEwG4lrmfJ+Z+lrxnQI6z09IPn0py+c67vK4buE8QZAIAVEE71trHnd+1au1ZjYWDieqQuOsInfSUdgE3KZ/Owqnjd4aCmQqtDcDDljhm24vzNgPrUmKzvZIV/o7qwDGaLIym2DSuQfi7BK61emqR3qVEpBmbEsX0lNw7kZ8nOmaTJClQHilpGr6jPn06rc22gg/eMMLfXPBG4X+xzQaYwLY7hteOir0z+vcMyOE126Z5l7NZqhQb3CcIMgEAqiCndbSW3sHQLWbpjHEdUXitmNSCjyk3OOs+cWKWGq0NwNOxKeBp4W/OZrUtrLagxGZ7hRPlzQl/c2XAXKez4ffL4LkxEP7OkWutzgl/x7NN4uLMnHafRpy56DojWl3uOmt6D4n0NtsK1mas8Hc8ePM0Bh9lwt+xZgMnRt/SHcNmXDFac+7fMfs8hL+vQjtuaPANN7gs2VQpNrhPEGQCAFQBu6hcWfx6cHLi5XLpFrpMurai3GAU/s6WMGDivgSue18qiJe2GeyxNiU22yvsvTjyC0BW+PtcosudI14WdV+ixvcIL6jN26hlsmK4ktxcFodK+Hsx1/DX2TEdWKsU/lbabCs4m3HvMhe8OSl8kKGUMB7Y6i1Rz3RMjB3Ds54jLQAAIABJREFUZVyVvGdADtsh0Gu40TJ25kqxwf2CIBMAoAq4hdSTN7mtdV5W+JsJGA2/lzv6bLlBKPzNOXhwnC6CE04+HngHOmkz2GN1Smy2V7L3IpqZFB/Xcl2bYs98SlMFDLRMNupTcozhS7U5DSeixBwpeCfavicTyTZJXScXCHhKtKbfK1qbbQVrs0zwJgxgTmOMTPg7zEQnyjRFYTLDjpzwd8F7BuQ0XBDPy9RvGL2v48FQd0fvAMiDIBMAoArGUgCu1n4l56HtbFTgeWj1Gy+jIop0esoIf8cdvqH0ihezRMeUa9CyGQBp4W/OZrDH+pTYbK9wge2xdJDtCMaLNYet6hvmHO48Nd3vEnKC2my2GSugqxD+VmRxNMx8lxJObjtG+LvC50Jrs63ggzeM8Den1aUU/uaC1P45HH1vqbfLc7rzpjKuNO8ZkMNueJ5t+Gbbk2VsNpQ4wjd6JFYNMhljPmaM+awx5lPM7/9NY8zfN8b8gjHmZ4wxX+f97jfPP/95Y8yLa14nAGD/8KKm62bztH0/CmzPzsuUAvDp2okFH6P75BZ8U+tfuZg4kDPtzsVFpOPtxdM2A+tSYrO9wunVnRKlxLygblw4vWXOQeRakNdzv0vIdefiss3iGxycODMzRyY2OGLXyS3qifTC37U9F1qbbUVS+JvRjSTSdR1cfgev3eWfY/w8o//jjolnXOnfMyDH+a/LMWa4v282vM1OFZbPPjprZzL9ABF9MPH73yCiP26t/WeI6C8Q0UeD33+jtfb91toXVro+AEAlsFpHiVTrazDs7HIOtHzHL6Ud1TK6T04ocWpRjs5Oa8BmgqR27zM2A+tSYrO9Mo0PzAKQe36Z4Ln7vQ+niUd0f6VA9wjXvj2l1ZcSZ09ucFwQCOBFe/nvSAl/1/ZcaG22FVqbccGbKZtapveVCmCG8yZXBjxcB+d/6d8zIMfZiBtjXm+64e8IOu+CVYNM1tqfIqJXE7//GWvt753/+rNE9Pya1wMAqBdukXNS7NKWnjcqYpjRLtE4+ly5gWvtO3UNgvD3GnDObE4rgrMZHNn1KbHZXsmVsnCLsajwN7NoTAp/Mws+MJET1I4GbzpeRDo25nfcc6BojtGywQd+Q4PrAFbjc6G12VawXduY6+SCNxr/i9+YiXd+4zpauvMmhb8V7xmQ03HPwfl+v/7WOcjEdpvE/X8k7kmT6d8moh/3/m6J6CeMMZ80xnyYO8gY82FjzIvGmBdffvnl1S8SAPCYjMLfrNDvbYW/s8KTiqyjlun05FLCOTHLGjtprQG7+5lwoJM2gyO7OiU22yvswiol/M21Wmfu33i/I888hL/ztH0fbd9eZCNmjOFKSI+aTCZOU4jRzbHWsl3DanwutDbbgtFm0cBgJpOJC95IhL+5jqyMlEDL+HzuZ1wmufY9A3I44W/nm77Rduffx4POYSAR3DdPW18AEZEx5htpCDJ9vffjr7fWfsYY808Q0U8aY37lnBk1w1r7UTqX2b3wwgt4+gAAUdidtFHUdCXh78QubUr4OyyxS3W9ajjdp8OBekv0VhtPUT4ik+kqsA50QvMqZbOuH4STjVn+HlyHEpvtFS4IfWQWb0T5YMJCH4XRxCOaMi4BT8MFpVPC371lx5h0Z63LhL8118m1NCeqVPhbabMtSJW+Hg8Havpu8fNcGZpE0JnvyMqUyzHndMdw/pf2PQNyWmYecEEll8nECbyv5aeDddg8k8kY888S0X9PRN9mrX3F/dxa+5nz/z9LRD9MRB/Y5goBAHugYRaVh4Ohg1knm8dae04rZ1r9MmVUw+/ljn7L6D65ifsNV+fOdPSA43QZRcLfGZvBJusC4e8J7l4YY1iBXFaThbl/nCYe0TC2IUMgTU5QOz7GMCLSx/hiraSz6eKcfeY6lYGAmt5DIr3NtiBlM1ZUPpsteYHwNzNnjiW6nP/FZFxp3zMgp2GCym7ecL4q9w7cS6AVyNg0yGSM+UNE9DeI6E9Za3/V+/lXGGPe4f5MRN9MRNEOdQAAIMHtvsV1cNZpjdqmzsnuJqfLeDpmMRFvDe4mbkZ0k+kEBXTkRKTjgcG0zeDMrkuJzfYKdy+Gn3HBcEaThVk0Zs9R0f0ugRXUZu53qn07Z9NpjpR3Nl1cZ+65YHRzNIGAvVJisy0osRkr/J3oOhg7byoAxHW0jN/PA9upV/OeAR1dP5SDhlna7v5yvurwmfqCzo/OquVyxpgfIqJvIKJ3GWNeIqLvJqITEZG19vuJ6LuI6J1E9N+cH7j23EnuK4noh88/eyKiv2at/VtrXisAYN9wHZSI3O7b9Z2H5Dkzu8mc8DdXFpHSKnid2R2avhMT9yW0zO5cSkcrZ7N7WVDslRKb7ZVUF6ZTIhgeH9fii0auDJjIiQXXc79L4LI4OHH2dPv2eOYYFwhIZbRJr5MTTubO6a59jc2fe6XEZluQsllW+Jspk5J3LozrGBItx5yk8DebcaV7z4COXIfAsbscsxnxeoP7/0isGmSy1n4o8/vvJKLvjPz800T0dWtdFwCgPkYHjlnkrJHN487JCeTGzpnv8MN15+Edo9ebLipmyXWCAjqmUqClmKUx8eyznM1qCnJsQYnN9grXfZLIZVAwpVhMxgXRMqsgpePCddoEE6nW9kTLMrSc1lFS+JvJUJHMkdx1cl3EmsRGzNPBUFfROFhisy3gyiqJ3Lsc1/uKZ7DIs6lzY84ySy7l83H+l+49AzqazrLjC5G/IRoPRGGeeCw212QCAIBb4BaVMUHBtfQOUhkCKeHJ2DGH8+I3Kvyd6GBHNNS5c917huvExH0J7v5FzEynQzwtP2cz2GRdSmy2Vzgx1uFnh3i5Z28XAtFE06KR00fhnvmayhNL4ASJOXH2VPDmmBP+ZnSzRMLfTPB8vE6ujJLZ/KnpPSyx2RakS1+ZZ4sR7T4psqn5jqyM3lfC53tKNF7RvGdAR8s0PHGBJ6fJxAl/38s7AGQgyAQAqIJpkcO1Sl6jXC7hjCV2/I6RHT8i3tFvO8ukeHtBJohMr4YrfYvZbNCoYIS/EzaDM7suJTbbK5zwt/sZW1aSGFNY4W8me6+m+10C1779xC6wXekVs6kS3azgN0WOTEbb4jqZbBNOODm9EVPXe1hisy3I2YwL3sTmO2OM2P9qGFH5rPA3mz0Z97807xnQwQbxgiY1nG90L9l8QAaCTACAKkgKfx82EP5O7PjFnLfhGBMX/j4HphafH3eH+uh3OgcPwt+X0XZ91HEi4rM0cjaDM7suJTbbK8XC34lMPC5jJZ5VgMVDDrZ9+1jGExfU5sYYa/mSRk4TSPJO8G3m09fJlmpW9R7qbbYF+fJaplwuMl4QncdbUQAz7huxwt+JLnhsNp/yPQM6OiaIJxb+hl/0UCDIBACogkl48natUacdP0b4m9lNjjlvw/csF7/W2nOdO+8Yvf5Wl/7OO9khfVS4rjdEcR0Bic2w6F4Xrc32DCfKS3QW/ua6MKUEcjnhb0WnTTChFv7OtG8fjllmPxnDl5SLhL973XVC+HuixGZbkBT+5kTlmTJKInnjFW7MmbofcllymsYrEP5ek6zw91tn4W/mHahp82cPIMgEAKiCsVyO7W50fedtFP5mnLHYzmRqxy92nUlBXU/4m/1O1LlfDFf6RhS/vxKbwSbrorXZnuFE0IkS5Z59vFyOb1XPZ0udhKVYNcMFRacMIWaBnRThDjvSpbNoRR3AmGwTTjg5VaJX63uosdkWJIW/2czHeAYLkcvqLi/FnLJ/GeFvpsSOD57L3zOgI6XZZown/M2ULGKeeCwQZAIAVEHb2WiHNaJ4htC1zjl8P78zGWv1ze34xZzudEq4r8nEfCf0UC6GK30jOu/GF9hsDSF6MKG12Z5JZbBwZVJDWQnfqj42rg3fB0HXEhqmvJMTJE4F9Y7cojxRQnpiGlVErzPRZIK9TuZZantL1tbxbJTYbAuyNlOUoRG5rG5hJlMqy0sj/M02XtG9Z0DHMMYwz8HhMGoycaXb91AuCuQgyAQAqAIu1ZpoWFSuMXm1iUwmt1Oz1MXgd/xOx2XpyqiPwKSEEw27Q/x3Qg/lUrjSN6K4jpbEZrDJumhttmfyGSxcWYlmXOOzpWoriyqhOwvVhxhjogLZ0xiTaPgQGZf4LFrZAo+7TlagPBFwd9dey8KyxGZbkLIZV34/NFrgNs8O1EmEv5kABTdnJoW/medZ+54BHSk//OloxkwmXlS+jrFgLyDIBACogpbphkTkar1XEP7O6JD4n/GPSTn6XBchriSP6JzJlHDw7mF39JFpmd17Iud0y3VIIPx9G7Q22zOpDBauZIkvK3GZTJpnHplMObj27UTx0p+p9CrV8n15TCoQIBL+zmabxK8zXqpZlw5Oic22IGkzTgYgkcHyJOycx/lG471hNnPiHelSjVfk7xnQkSybPBgvk4mzGfyiRwJBJgBAFSR3UFZa5KTTyuM7k1zN+vA9y8Vvk3JMx7awlzt4gKfJiUhzpUMJm2HHbl20NtszTTKwvSwrsdZSx4xTU1YBI/zNlNjgeU/DtW8nipeyNamgHlOq3fbpeUKihzKUy/HCvktx5oRWYmWlwyU22wJns1QXtqgMQCKTXCb8nSnFDDXGOt7/OrEZV7r3DOhok1mzh6m7HCf8jSDfQ4EgEwCgCrhUa6JzNs8awt9JUdOULkbiOiMLA+4cEP6+DV1no4tnorjejMRmtZSIbIXWZnumS5SyxMpKphbmvKYY98wfmRK7msoTS+AEtYnipWzJjoFMSWOTOoe0Axgn/O1K3yJi4+7fsDjmjsSub0GJzbagSbz/JTIAx4MsgNn28RJnrllGlxynuIwr3XsGdHCaV0TDc58V/kaQ76FAkAkAUAW5MrRVhb+TGSvyHb+hXI4pi0g49m+1iYm9snbta5DrCLgsl8vbDNll66K12Z5JZrAcEs9vojuiRvgbJbt5OEFtongmWCqLgy0vyp5DEghQCn+nrpPJftorJTbbgqTN2E5v6QCmqBST8Y3GYOTieS7xv3TvGdCR8sNPxwO91fLi909HQ70l6iuamx8dBJkAAFXAaUUQOUHtFTSZBMLfsVbfyVa/CnFL/9+bqoPHAu8yUhkAp4hYqMRmsMm6aG22Z/IZLGFJb6LEiRvXkto7KNnN0TKCxERxG6X1AOPlRSnhb2mDCE5QPydQnhL+rmUsLLHZFiRtxskApDoXCsuT2VJMZs5MjVO8/6V7z4COnPD3+OfEBhx0sR4HBJkAAFWQDN6spcmU0iHhuu3kFr8qcUt/0k4F2DBpXwK3e0/EZJ+JSkTgzK6J1mZ7Jp3Bos2e5LqI9WQM1068rvtdQlKcPSb83acyAhLC32wpiywQkH2WVMHHugLuJTbbgpTNjqUNTQQ+CJcNlc1k0oxTyvcM6Gj7RAOeQybINDYCgG/0KCDIBACograz0QUO0XqBllGHRLWTxjvpx0ha+eRI8SK8RPGAxvSdmLQvIZkVE8mSk9ispnKtLdDabM+kRNCH0kGuo2VK+DuSIZAs2bVkLZ55jmSWUUSQOLnAZrJN2r5PzJGKQIDmOkVzZB3vYonNtiBlM7YMLZHBchKUy6aaDWSFv5XjFIS/1yPnhzviHRbvJ5sPyECQCQBQBU3X8+2Zj+ukQTfJ9PdUG2l51yvn8KXO4Y6NfyeyCC4lJZz8dOCzDFI2gxDyumhttmdSIuing074m8+SSXVtuh9R43slmWWUFGePddaKNxdoEnOPVDcrHUxcXmey6+CoQ1THc1Fisy1I2Swp/H1B58LUmFMi/F2UzQfh74vJ+eGOtJ3hGz0KCDIBAKqgS+xQxTKErnVOoviuzJHZmWx6O6ach6Sc9JS4JVF815FocOThOF1GagEdF+WVCJLCJmuitdmeyYmg8+Vyid3mhSZLunX1cB313HMtSRsdEsLfiZLcMCujS2SbHAW6WWO2ieK96nr+OqcMlToWlSU224KUzXhRed1zEZIS8T4eDBmz3JiZuvWlsmKWm3aa9wzoSPvhXiZTtBEANJkeDQSZAABV0KSEvw/rlMdM3WLkYpVtl6pZXzo5beIcIuFviO5ejFY4WWIzZJeti9ZmeybV0fJ0XJbTNslFZlysOamBVVkwoYR0kC6i1SfoYLkQSu74LoMngW5WqoOg+/myCUJeRLqWsbDEZlsgsVks6JlqXZ/zv1Ii3kTnUjamKYqmQYH2PQM6hvvLZ806kj4zgkwPA4JMAIAqaLtU17aVhL8T6e+cwHO6xWtC+DvhSA2/T3V2waR9CUML+LTezPzzeZvBmV0Xrc32zHAv+CxPTqw5pY+iaWjwVFkwoYSkoHYkACQSkY7MJXwpS174O3XO4eexZ+lcOhzNXKhL6LfEZlswbZIohL9TGW6CAGYqADecN95h8WCIDlrhb8V7BnSkuycPP+dsxs0t4H5BkAkAUAU5wcE1dtHdZJgSNV3svmWEv2O7z0SMuKWffgzh79VIBgYj91diM6Tlr4vWZnsmndW1HBtTZT3GmCH7Kar3wwcwiO5D1PheyYqzMxlCsaxYTkC37dLC37lmBLlMprjw93DMMZYVU5nQb4nNtkDSHXXZLCCRwSLwv1LNBtx5F9lTCV+KzbhSvmdAh0T4O2Vjovt4B4AMBJkAAFXQ9AnBwYi47VXOKRD+DnUEut6y5XKnSKentLilJ6TIOnh1iRyvQfLZiuhoSWyG3bp10dpsz7QJMdZYY4CUCC9RXOetS2idnMYd6nruuZYuIagdG8NTeoCcUHJO+Dv3TnSJzN3hWmKbJAnh7+P9iF3fghKbbUFS+JvJZGoymeQ5GzeZMYfr4pqSHiCKCZTr3jOgIyn8fbZVyv8luo9sPiADQSYAQBUM7W8TO1QrOA8pp5HblWkT+ihPB0MdJyKdEEoM/zz/Tgh/X0qXETVmhb8hbrkZWpvtmZxYcyxgRMSX4D5FdZzywt8Yh+JkBbUPy/JOiYh0uFjjWsS7Y+TZJvLrTIpI35HY9S0osdkWSGwWfbYuENTuEnPm8PP4Zk42K8a7zpL3DOjI+eH+/xe/x2bEw4EgEwCgClLitmsJ/abEKjmB55T4aiwY5py5aMvXWbeOhPB3JU78WjQJZ/YUETWV2AyO1LpobbZncmKssU5xRHwwIabzlsuWGr63nnuuYcqITWWbxcvQVNkmyc5aAuHvxDnH6+SepTsXu74FJTbbgpTNOLH2lPB37LlYnDMR2Bq+I95hMZX5FF5nyXsGdKTm3TGTKWMzzBOPA4JMAIAqaJPBG0O9JeqvvEuV7BbD7PjlBTLjXYRide6Hc2vf4XwQs1yL1LN1PPAt4FM2qynIsQVam+2ZtPD3gaydZxmlhL+HY+LdzjgtjlHQFVkCUdy9TN2/WFCPO+bIzT2Z7L7cmJQ6J3udfX9uQZ/YiKlkLCyx2RakbHZkbNYmMlhiz8XinJIxJ+IbZccc75iS9wzoSPvhg21zNkPG6+OAIBMAoAraVObCSsKzSeFvRuA52eo3qmnhsmK4TASXgly+eABpch0BOeHklM1qKtfaAq3N9kzqXsRKlrJZBYeYCG9qBxs71ClygtpPEa2+lB4gJ6id6qzlbGotPy7lMtxiwsmpwBYnzrxXSmy2BclgJPtspf2v3LufKiV0P482UUlkTw2fWWYyad4zIKfvLfU2PW8Qpcuwie7jHQAyEGQCAFRB2/cJEch10vKHltDMLi0j8NwmxFdPkdbqqU4v/s/ZEoZIi3KgIyciXWIzpOWvi9ZmeyZ1L8byl0gmEzemxNrdtx0//nIlNmCgzQallRlCY6nQMtuEF2fO62ZNZcBy4eThnOmug7VsgpTYbAtSNmNF5VOt6wU+SK7ZwOlwWOhVpnypSa7Ay2QqeM+AnMmG6SBStsSxkvFgDyDIBACogvRu/Tp6B21GSJWIohpLydK2RYlDRoTX7Q6lMpngOF1EUszyLKzuZwBIbFZTkGMLtDbbM0kRdNeFyRsjUsK/7ueLcU0y/uKZj5IPSi/F2dOlb/HyRMl8lbJRqjx8+Hlkk6RL60AR1ZO5UGKzLUjaLFLW1/eWrE1lqOR9EEmzgbjYuDwrpuQ9A3La0YYQ/q4FBJkAAFWQEv4+RRyj65wz1bbbLd4iu8mZrlezgIWgta///+V31lUadG2stcJnS2cz7NatR4nN9kxSBD3ShSlV1kPEZBWksqUi+ihgIiUITRQv70xmmxziQT3JfJUMMmUyFWLCyU0isHUSZE/tif+fvXePuicr6zufXZe+QHPpBlqwu2lQkIsiKi2omBkWo0i8ERNNjFneEpeZMSZxJlmTmBkvozOTTJKJa2bpSuIYjMmYqDFxSRIiwVGXjBGnURC5iI0EpJtLNzTQNDTd51TV/FGnqnbtem67zjlv7Tr1fNZiwe9933Oh9v3Z3+f7zGmzJWDbDDEol027FcbfqlRMfREVrG/NGWeGnqEN+XVAajNrg/VgQSbDMDYBZ25bnGkDF3szWXU3fsJNDmrCS7wmFxZuU80cR9cW0u2cv+nWtJnd1p2POW12yeyZKkxo/+0NcvWqAsn4t/sbY8pe8KPJkTmcU5tgpsftvxXrFXPAU31PzAfqiM+8JOa02RKo2qz209BkhZtUeKVP0Y1ReQtFVKbfM36cGXoGBTe/1kjpdFsJOl8CFmQyDGMT7Jky3YMs/8TG32zq29T4u/t8qroGFpjaCRLkbsEm33NjqUGnRpTYd33LV4Io2mwrKSJLMKfNLhlNFaaR8bfGIHeiKuBSV+yGmkNSABTZ1DiZVwgR5sxCqjb2GvR7MmlR4Zhi0yg3li43p82WILbNpEuVAlFLhuxqYW+EFRtgiqgMqhjE+DtinBl6tOsu18YAU4sJI10syGQYxibYc8aTZzKe3Vdc6tv0ZlJjbun/XfcZANyBrzNTtNSgc9BXihMCmKiSiWkzS5c7H3Pa7JKZbfwd6R0nVRHbyvOORfIyKZGqo9zzds4dVBlxqdr+d2G/J1M9ampQbsbfHXPabAlUbYZUo+SMv9vXKJRMTNAI8yWjPhNThs0ZZ4YeKR1RMv7GzNqNtLEgk2EYm0BnPHtqJRNXtnt6KyMaQmOBKUn9pDD+bt/HNk9zqKTbOWTTrWkzC/qdjzltdslIJujt39Te3wvG34iqgJ8L0/GbSRE5KE0YfxPt073XnFRtTSAg6nuyaezpqHeugtg2Wwq2zQgbAABe4QbA9y1NsYFwvq4iC6/M6b+GHikdsWsref9rbbAWLMhkGMbF0zQNWzmnPNNmljNSdc61m0YkDUXyRxlJvA9lp7Gyx+17HfLcpVTBBG5I14hoZolsZjVttpUAxxLMabNLRWuCjqWVcKqCicFzVZMqmXMVXrgUeuUdY5gbzt87Rm3SvdfInFk4AGpMd8XviRgn831vWxcgsW22FDsmvRZTW8t9S5EuJ+2NsHS5WmFkP9p/xY8zQ4923RXnIGuD1WBBJsMwLp4+F/yKjX45I9X2c8c3k0O6gXCbHJhVUsGz7jP8/ybfM4HN6xrp2oJMM0E2s5o2s/Y4H3Pa7FKp+hQG6Vlg8xSX7qlX1piakkdTWr1pwoIQtNqkey8sVTtGoRL9PbMMNyhn/MCc207wMbbNlqItFBCjtpYCRPL4H1LZaDXkNEWXS+tD0thnjDNDj5SOKBp/2151dViQyTCMi2dIjxE2HOdIl2OCCWU2vpmUq7DgGyNOYt9VgJJMd7eykT81XVvEmLVr2mwLKpqlmNNml0r3/zGnvEsQlcFOen7ZVHGhSh1O4ACdIsO6wFcQ3QVBI+p5t+/lRnNMJZl2qwIB8vcM21iaC0skMHWpxLbZUmiqtu2QoDRXjdL/OwxxzkEC27uqphVX/b4H23/px5mhR7u/5dYVgG2sy5eCBZkMw7h4Bhm0XuJ9CjizcYCpwfNg/C2kDwQbI1Zif1iwxVQVux2ahZwigihBFG2Wwo31pTKnzS4VyQS9xJRMgvF3iRjX72s6XW5LQb059Mo7Id1oHyiZ2LUnG6c0SubMmpTGvbTOogblDdn3ANJR71wFsW22FFybYQblQ3rt/D2IPOfggW3ZbHxq/B0zzgw92v2tXKBn+TFg6LAgk2EYF490g3KuG6o94zfRfu641LdoCI1WpJMl9ux7Km6oDRptWd5dZJtZe5yPOW12qQzlxePmHOe4eQo58DFz4ZbSE+cgloDHUnIZM/f2vRwaOKQ+I0cUKtHfM0dSmhjfnO5zN2P8HdlmSyG1WZHF9a1BoTK/cmHob9l+Lp2W3vdnTMlk89RZkIy/uzYRjb8tyLcaLMhkGMbF093S5lKZ7jMYf1MHse5zo4y/kY3+TvB9KqU8d4XppkEjSeyxjZGmzSx98XzMabNLZQi4CXNjkO7Jqk8wI2rGb8YUAjy9cXLE89sJvm/tHDM1ZxbTizjjb6kEPFIdjTORHr7nNubC2DZbCl2bRXi4RQQwyQIm+TSVkEvrwwuvxI8zQ4+U8igaf2dy8QEjLSzIZBjGxTNsUATj2TN4MsnpcshmTPLFCI2/BR+H7rPY90zghnSNSBJ70vhbaDPbyJ6POW12qahN0Cdzjl7J0L7ejL/nojEkBpimNIpthKZqC+lFnPG3ogT8xKC8pk2k+++5kX4R22ZLIbYZYSpPB4gijL85U/mJkkmRihkEz9nPsHnqKIZ1l9/fUm2WZQ4yZ89/TViQyTCMi0c0/lbcpM1BY/y9Q5RMYuqKfwOtNf4+wnTToNEYkvp/B6Brs62kiCzBnDa7VCQTdCxNSlIyFPm4/zZNAxUzF/bzr81BKIPCQm9IvBNMpMM0tEq64MhlFcFQqj7CoFyh6tzCOASIb7OlEKvmBoUrJJWcRk2t2RtNKhfWNVnQAADbf8WPM0OPtoIl12ZFImo+Q4cFmQzDuHhE4+/Dxu7UpWlFI8/cjT6zLyceYUhcCbd13eeLZpYJbF7XyGB2y7fZqJ0VbbYFFc1SzGmzS2U6K3BbAAAgAElEQVSWcb2gZChz1wct2r9XqmRsDkKphOeHpfFUjNF6+5rA+FtI1e5+zo2JOd+zVfvyapNqI8HH2DZbCrHNgjS0vdS3Mnm+lVKcMePvXdWwz5Pef1m63DnQmLcD0KpaANx7y0gXCzIZhnHxSKWVB+XCGYy/We8S3PhbKvE6Mf4Wbn78/8a+Q/c+Rjzi7Ryi0tC0mW1kz8ecNrtU5BSRqdJRo2SYYypthwecWcbfUupVmCok+ZT1fihcSpPuPaapw/yhMvTauVRi22wp9oLX5MRUvuYVbpr9l1RsIKxo131PNi09D/dfZvx9TiTj78HawawELgULMhmGcfGIpqZn8iXaSXLtoNT3rhY2Ob2sPDCR5lQFfcWO+aabBo1snDztW5o2s6Df+ZjTZpeKqBBA5kZJyVDmY+NvyRDaOXeYCy//ec9hJ/hmkcbf7BwzDgQOxTGkz1Cky0V9T4WJ9EYO9bFtthQ7IRg2Mf4WvLo0amqp2EB5SCVsmvE8xX7P0Ph7xjgz9OyEy52yvxCV0me3MR9cAhZkMgzj4pHk2rliAz33cznpbz4p9SspmbBS1XTVJoBhwTYzy/OwF1IxMZWGps2sPc7HnDa7VEQTdFTJJJedR5VMkvrJDm8okm8WVryh9QOUVDFISpPgR8MGAqoGMtca9OKfSaReMsGDsC9dMrFtthRSm9Gm8lTfktXUouINSXGWAlM5oeaLGWeGnmF/y88xolm7Pf/VYEEmwzAuHlG5cCY1TyUYeRb5CYy/pdSV7nbIjL/Pwk5SglDqswjjZOO0zGmzS2WYc6RUlsC4Xui/+3pQFQxpEvzhwW6ocSRDbcw4WUzVDlOFhOpdXYCANf6udZVO/feoBLVJkWebGIcA8W22FKo2Gyl3dWlS3P83ec0c7436YgNzjL8jxpmhR5uazafPZvb8V4QFmQzDuHh65YJ4w3raxWsnGH+HKSJSiVfUhLeSS0B3n4W+p93OHYWmtDLA1IdENE61jdTZmNNml8os9WRVC1Uzxwe+Xa9kkAxdbQ7CkI3TsZRcoehEqDYR1L7YBcfkeyqMlsP32FVC6uWGjH5j22wppDajjL+lNCk+XU5OJWz/rum/I/eZAJh3VPw4M/Rojb+lNttCQY5LwYJMhmFcPN3mQSqhe+p0jX0tlPrOsqjNWJ+yEEjRpZsfAIXpbgKb1zUitVnft8JUFqHNbCN7Pua02aWiLSsdppVoUle65yelSXSvsTkIR3p+Q3+NWRdwc+ac+gxF4FU0WiYCljEm0pdMbJstharNkMszSSHEp8vxSqYwxVn6TIAuRXfcF7ufs99zI0HPUzPMMUcafycwBgwdZw0yOede6Zy71zn3FuL3zjn3fzrn3umce7Nz7gu8332rc+6uw3++9Zzf0zCMy0aSQYe3YKf8XPbGf7IZE256kAo/ksS+u5WTzCxTkOGvEXUJ+PCQLrSZpQ6djzltdqlo58ZJupxg/A0wHCqkNIn2c5zNQQSSwgJL7xTVZlSq9hGmx60hPK/c9T+r+86yifQ25sLYNlsKsc0maWid39d8NbWoeAv6lkbJRO+/9OPM0DP0g/lKplTGgKHj3EqmfwoAL2d+/8cB4JmH/3wnAPxDAADn3E0A8AMA8CIAeCEA/IBz7sazflPDMC6WSpJBn+mGqhI23dNS30Iaz+Hn1UjJJEjsc/52CDPMNPRUwgG6QNVncptZe5yPOW12qWjnRn/OqeqaT4sKTKKlNIn2czLr8wT7QwU25/g0nipQuMbMMZVwwaEpECF6CmXTtaYSzJm3VLI8ts2WQmwzom8do2Sq6jjjb6k/d68JCxTEjjNDTyWlqXcXooJ331bmg0vgrEGmpml+HQDuZ/7kFQDwz5qW1wPA451zTwGArwCA1zZNc3/TNB8BgNcCH6wyDMMgkYx+NcaTc9gLpeqLSalv4XsiN8HiZ/TpcpLppm2c5iC3GaE+E2X843LMxumY02aXypy5USph3j+/w9wmlbZvX2PqPYqdsoJo7Bwzy5yZOWDvxEDA+D005sypmF1fBbFttgRz2kzsWwo1tVhsINjHSBd23WtGiqsZ48zQIxa2Efaq3d/YOrEeioU//xYAeK/377sPP6N+fvG8+0OfgO/7xbfA93zZM+EFt9+09NcxjJPzmrd+AD7yiUfgG1/4VPT3H/jYp+D7fvEt8KlddbLPvO/jDwMAfVvvnIM8c/Bv3ng3/M4ffeRkn/tIJd/4v++jD8E3/5PfAgCA9330of7nGN3t4U/+xn+G17z1A+1rPvYQPP+2x7Ofwb7nYVP747/+LnjV774PANrn8R1f+nT4Lz7rSehr7nz3/fCjv/JOqL0gyGOvL+Hvf/3z4fprcvQ1/8cv3wVveM/4zuFrnv/p8KfvuA39+7s/8kn4wVe9DR7en64fnAOpzbqf//Tr3wO/9o5729co2+xbXvn/9T+74doC/vaffB48/lHXoK/Bnu9/9eyb4dte/HT077Fxdm2Rwfd99XPh9ic8Gn3Nz73hvfBvD32k4/m3Ph7++lc8C/37hx6p4K///O/CAw/t+p9lzsFffukz4I6n4evbr//BffAT/+9/PmuAbU6bAQB8+4ufBi999qehr3nLPR+D//0/vmN0yzpnTMS22bF0cyO1scfmxre//wF4/q1y//3uf/FGuLbI4OOf2rOf0b3m9e/6cD8XXhVPe8Kj4Yde8dmoeqGqG/gb//rN8MEHPnWl3ynkXfd9Qjhgt3P4j7z2D+D/fv17AKBde6RD+X0PPtw/73sfOKyRQrDxZ+98L/zmH34Y/Zvf/8DH4dFEX/e/5//0qrfCY68v+59LJtLv+fAnovrFNXkG3/uVz4Fn3HwD+vt/9+b3wc/e+d7Rz55582Pg+7/muejf76oa/vuffzN86MGH2c99+ec8Gf7ci25Hf6dZz2LbTEOROfhrL3sWfM4tj0N//0tv+QD89G+9R/1+HVKbvfPeB/vvefdHpPl2uq8Jefv7H4AnPPpa8fv8dz/3JriuzOHhvS6w/ea7P9Z/zznjLM8cfM+XfRZ8HrGeY+vZk264Fv63r/9c9Bk2TQPf/4tvhXd/+BOjn/+ZL7wNvvpzPx39jN//wAPwd3/pHckHX/7o/k8CgDzHSCmOnJrvPR/+BPzwv3tb3/4AANcWOfzg1z4Xbr3xUehrfvq33gO/9JZxv7vj9pvgr37ZM9G///indvA3/vWb+3VNy5NuuBb+wZ/5vKjXrJ2lg0xH45z7TmhT7eCpT8UPrWvik49U8Lq7PgR/7kXr//9iGBg/d+d74Z6PPkQGmd74Rx+B177tg/DsJz+GPJzFcv01Obz02TfDbTfhiwwAwDd+4W3wtvc/AA8+HLdwcNxx+43wpc/EAzUAAC/77CfDe+7/ZP+Zj72+hK/+3KfA47wNuM/jDr+/56MP9a95zlMeC1/+XPzgCwDwkmfdDPc9+DBcX+LP8ok3XAMv/+wnwwc//qn+Pd9yz8fglsdfRwaZXvu2D8Kv33Vfv7H62EM7eN1dH4L/5r/8THIz+1O/+W7InIPbbroeAADe+cEH4eFdTQaZ3vDuj8Avv/2D8NynPBauLdOtUSG12aOuyeHrPv8WePeHP6Fusxc/84nw63fd1//9g5/aw133Pgjf/EW3w5c844noa8Ln+677PgH3ffxhMmARjrNH9jW89X0PwMue+2QyyPSv3vBe+P33fxye8Wnt4e3ujzwEb/yjj5JBpnfe+yD8+ze/Hz7jSY/un8+b3vtRePaTH0MGmV79e++H3/zDD5H96BTMabO3vu8BeNIN15JBpl97x73wq+9ox4Rz88ZEbJudgm5ufGrE3PiMm2+Ar/rcp5B/f8fTboQXPf0meKSqYVfV4BzAi5/xBPjsT6fb9E98/i3w2rd98KTzr8S9DzwMr7vrQ/C3vvI56PO87+MPw8//9t1w203XwxNvoA+45+bmx17Lzhe3P+FR8NJn3wwf+eQj/fO74/Yb4Y8xa8+XPffT4K57P97//aOuzeHLn/tp8OTHXYf+fZ45+IYX3ArvvO9Bso1uvfF6eMln3Ux+5nOe8hj4Y898Ijz48L5/jy982o3wxZ/5BPI1L/+cJ8MHHviUul/sqwZ+756PwUue9SQyyPQLv3MPvOHdH4FnP+UxANAGb19314fgf/yq50CGHHDf99GH4BfeeA/c/oRHwU2PxoP8f3jvg/DAQzsyyKRZz2LbTKJp2vn2jqfdRM5Br/rde+DOd98Pz3nKY1XvCSC32R//nCeP+uLjH1XC133+LeQe5PHIvibktpsexY6Bz7vt8fBFn3ETPLyv+/d40dNvghfcTjutfNXzPh0++UjV//2ccfbGP/oofO4tjyODTOF6dv8nHoHX3fUh+J4v+yx46hOmc+5Duwr++evfA7c8/nq4+bHtnPP77/84XFtkZJDpdX/wIfiV378Xnn/r49D+mwo3PfoaeOHTbiK/4y03Xg9/8gtugS/6DLpv5YLx9+vf9WH45bffC5/96Y+Fa4oMHt7V8Lb3PwBf8/ynkEGmn73zvfDuD30CPvMwX7z3/ofgbe97gAwyveMDH4dX/94H4Bk33wCPuU4fRqH6/yWzdJDpHgDwTxi3Hn52DwC8JPj5r2Fv0DTNjwPAjwMA3HHHHavXMJoJr3Hp7OqGvXHpUll+9Ju+gNwknoP/5eued2Wf1fGVz3sKfOXz6MNaSJY5+NFv+gL5Dz2ed+vj4Hm30oe7Is/gH33zC0Y/e/Hf+RVeul7VcMO1BfzCd70YAAB+9R33wrf/5J18u1Y1fP0LboUf+JrPBgCAb/4nv8Vulrv3+sff/AI2OJg6zjn4kcjbqy946o3wr/7rL+n//dvv+Qj8qX/4n1jD0fD5/qV/8Tvw++9/gP77YJzd+8Cn4IX/6//Dp8NUDXz+7TfCP/vzLwQAgL/9H94O//Q33s18Rvte3//Vz4WXPKs9fD7vB14j9K0Gbn7MdX3fWgKszV76939NeP7t737hu74EnHOzxkRsm10VsXPjM25+DPzsX/ziqNd810ueAd/1kmdEveZYfuJ174L/+d+/HXZ1DdfD9ADQtd1feekz4RuIYHgKXFfm8Mpv+8Ko13z5cz+NPVCHOOfg733D82O/2ogn3HAt/PO/8KKo17zi826BV3yePpHhY5/cwfN/6D+KqVfPevJj+jnmx371nfD3XvMO2NU1XJth/aB9r7/2smfB1z4fP+h/x0+9oVdIop95gvUsts3quoHP+FuvFuagBp7+xBtOOt9+wx23RY2XOfuakKc98dHwM98ZN+d804ueCt8UcaGPjbPP+h/+g7gu+OvZL77pHvirP/Mmcp3t+tqf/9Knw1/40vay4RU/9htCf27f62f/4hfDdSsOZJR5Bv/gT/N7pTLP4ME9t29sn9NPfvsXws2PuQ7ee/8n4Y/93V8V9xwv+ownwP/1LXcAAMAPvuqt8G9+527xM374FZ/DBluN8xt/S7wKAL7lUGXuiwDgY03TvB8AXgMAL3PO3Xgw/H7Z4WcXT2+uaMZyxoWyr2q+Qo3Cv8M4L+1tkb6c8GCULJnCjl/DmsgqKlJthcFIOa5NYsbZUAKaa5M68jOmfj957vhS1YKvy1KIY6KuR6axs8eEzY1XRmhQHmJz0PrIFSbS+2o6j7U/p/qBIvUqk+a1q+9LWeYgc5JZO5+iZ/Bo1gW/zQuhkh42x4t9S1FJ71IQ943982ufc67aO0XuawQDc2PgrEom59y/hFaR9ETn3N3QVowrAQCapvlHAPBqAPhKAHgnAHwSAL798Lv7nXM/DAB3Ht7qh5qm4QzELwZpsTOMtbOvGrFCDYBN4EvSGpILwQbPT6Bb0Lkb08lrhFK04WZhy2gMR7E2iRlng9l1TCCrrQjWNA3qaYOZr0rmteFnpIJkQIw9G4DIMRHZZsZxDBXT8DayOWh9aIp47APjeqlynuYQX+Rxh9+rosgzVp26F0y8DZ62YEHEutCv5cScgwQjW4NyuW/lCa6bp0bcN9bhvqYrNsBfaIZ7U9U6vIHnfSxnDTI1TfNnhd83APCXiN+9EgBeeY7vlTJDKU8LMhmXya6u2VuZbkPEGUsa56XMMlE14xuad2m+1MLcNA3skNdw81y3qeJKOm+FbixQhpf089WPs1Kx9rQltodx6ad3X1MgQaZ62oZlzt/87qo6ybEvPs8KfzbxY8LmxqtC6vM2B62PUggYAbTj6IZyOP4M/YBKYZLHXZlnSa5npaj8aEgjZkOmbXf9eta1P7WW930t81+TwScfYVLE6gbK3KEXPZeGZPzdj7PD8+v+u5L2HME6LAVm27+zcSNhTygxBimlpcsZl0lVK5VMdkuwGOKtLHLzA0BvnKr+dmms2uA2C4Mk2ZapIfUKXxe6xzhuE/lw4b+3Zu3ZVzWuACC+15BmMv5eXLtL5bGXQpLpV2FahPBsTtFmxnGIqStI/zXSpk1ZbccjBaU6pIJEmlS3QpE21b7H1SuZYtK5jDjkdSFQzWX8ulBhSibNmrmROSrP+GBuFaSyaYQbE2VjlkHT0PtZTKFt4GyjV64IM/42Lp1dxRt/75GAhHG1FOKtLO5pESUBz52uH9iB2ku94m8/p2lp+nGmUTLtqmYkyZd8h7rv67+mzDLeKDXRDXMh3VgH31tKfT9FmxnHMaRS6OctI32kOWZySSKtX4i3XIicYr7MeqZRDG8hzepciAo2Yl2g13L8Qk4qxLKVOarM+H3jjrg8460GwktTQdloe1M1tlNJDDP+Ni4drfG3pSgsR6kwmgwDRgDyTXAZbLY0qg2TJHvPVwrizTCR7saZc04sD7yv64mMv30v3svEb0PZVD5NI9pS4blSjCT3SnXEEW1mHIdswmtz0BrRzDEltn4JijZu3BWKFPP2Pa5YyST5vNVppievhUJI/56YSktrJmIyrzH+TnHNPAeyyj4owCHsnbrXxNg/mFefHntCiaExLTSMNbOvG9H3BcAm8CURzSwnt3O8yTFqAC0oQ/ZVDc5tw8xSQgziIZseSY2GjbMi470IJtJ/MfiFt3uMKW8qFJI6ggi8imPiyDYz5iPdWFs1v3USvX5JAWFEXYJ9phQgXmI9KyR/mQ0FKM5Bu2bGrwv0nINcPigCK1tRt8r7xpmVj5FCNlZ19Hi20StXhMa00DDWzK6qReNfANvYL4l4K1uPb4Klm59QwgzQyZ4lmbn1AYBBAUa1CWYqqzHYBkBuTFlZ/tjcur8UEQ5nY/WTbG6d4s265Lmyq/FnI42JY9vMmI84b9lhYpVozJjDtQhANv7mxl0pVKRaaj0TFcPBvGXEIe2VwvWsFAIYmMl8KVQIDAtIXDLivjHYozjnRCXYxP5BSqM2hasae0KJ0fVzzrTQMNZMVTdQNwA1czDNHEC2kUUzReRb2Tjjb8z0tMgF4++qNsXGAUnJRBmrx44z0SR24sUllIDHlEyZxsQ0vbEv3SZXNZEWIRq8HtdmxnwkE15Li1gn0hwzVZccr1yQU5qWWc80imELos5HXhcaNIARZfydOahEc/FtzFG5UDAmLMABoFGCEebsESmNBs42euWKcM4dyieaksm4TLpbCOqWYLch6W+qaIwmc+QmWLr5CTdbkoGjbX5b1MbfEbdx2DiT1p5d3UCOpoTpFWxtupzU7umN/0Jxm1yMbqylZ3OaNjPmozfhtXloTZRSSm4wjiTj750i2FgcFBZNQ/elJfpRoVB+WEr6fNp1gX++eF+T1szxhRz/GdsJFJbSvhEpHMLtZ5ummVTnk1LdsTYycOwJJYgkvzSMNdPdAnBmq1uR/qaKlLbTSuz1N8GoBFwoRWuGpANa4++YdC1snMnmtYTxt6QEGUn/ZQPtFI2tS9GoPjQTllRep2kzYz5aE16bh9aFZMYcphfJnnfT1NbpZ8pq3iX6USmpU4NiDkYcZbTJvLBmIibz4n6sajbThhrvs3CcFkyKPpW2DqDwwExwn5Ia2+iVK0MyLTSMNdNt2LjUH7utXxYxlS24+cmFm2BMAp4fUhqom99U06aWQDKv7DY9eYZtZvXjjNvA1Yc011DG336+Ps0kV5QXzxPcMGvGRB51cD1NmxnzUaeu2Dy0KnLBjDkcR1J6DJbaGqJJaV7Ek0k4lIfFHIw4xKBHsJ5pLhIAxuuCpurrVtRoxSFdjto37hFlXsFcaFJp6+3vhGq+Nm5EbLeSIJJpoWGsmS4QQZsYp6lk2BKlUGUsvC2Sbn4weXH3Gk42bgqCFsm8kr2NixhnnHktprwphaBI/xq/3SUD7apOUq0jyvRDlVfGP5tTtZkxH9mE1wxe10gpmTEH6UWlEGzs1kJuXtKM90WUTJlgSG7r7FFISrFwPZPWzNnG3xtZF8R9I2JkzynBdoi/kvQZWEETA8eeUIJI1SAMY810NwecNH0rtzKpIt6cBW2kNoCOMEbeV9u5ndPAtcmguJiqy2LGWc4Esrqfhzes7Xvx6XKhwkfqWyneELbPRv+9xWfTj4nj2syYz/C89f3XSB+V0S86j+nVJSGauXCJfsTN6QC2zh7LqdcFyvhbUjJtReEqpqUGxUkA+L0T5heaC5cP+7oGZwU4VGyjV64MybTQMNZK0zT9gkya6tVWVWxpCmEOCg2I1UomLNWKeo3J+Edw60J/G+ffzvdm4fpxxpnE4ibeOhPT0JOCPfQkmhLWmofqv7d4G4ocJua0mTEfza04gKVFrI12jtEb/XbKB9H4m5mXNH1pEeNvwX7D1tnjkIqkxK4LO+TCqOvPXIrYVlJ6e+N0JpVtWtCETtHHfCPFAhwb8sA6FntKCdJ6lVi6nHF5+Bs/+iZiO9LfVCkZo0SAtu18CbhzTrgtwiXg7e/o19hCPsCZVw7GtNONUsw446T/3fvg6XL0a/LMgXN+IEXysEg3XU66TUbHxJnbzJiP5lYcwNIi1kYZmVosjrt6+poQTV9aoh9J9hvtWm79ey5ykZQ6MJnXFcsYpVFncv/cSsqj7GmFGH8zZ+odMralNOqq3k41v2PZRq9cGYVQRtow1oo/adM3/NuR/qZKIfg47JEy8znj49SrNqJSreyG1Yczr+RMpGPGGWdiilVUkTZ8rfImlK7LfStV428+7QQzHJUDr8e2mTEfqXR9n0ph89CqyBlFZp+mihj9imoHZl7S9KVFlEzMHNSruqx/z4ZTzQFM1zO5iAdi/K0oIrGVlEdN1dbJOsyowLGxLVXz3W1IOXYstltJEMm00DDWih+E4G4abQJfFpXJcaiC4Q7USAWPXjbO3M7ZgXqAN6/EbuMEM1tknJVMSlj/GRFpJnvEVHZO30oBSYHVfu+pTJ9LB23/5rg2M+YjmvD25cRtHloTnCITu/CQDfdlJZPcl5ZZz9o0a+n/l/XvuXCqOYDpeiYpuLE5R0yj3pDCVdo3hgU4utdIYxsrBEDvTadrvYFjTylBJNNCw1grfr/mTIxtAl+WIndR5drb1zAbe0QCrjELTzFtaik06YioiXTEOOPanTLx7t6Les20n8gln1O8WS/yDJqGT1kIv3fBbG5P1WbGfEQTXsSHzEgfXpE5Daxoxl2Y9hsiGzovs56xczpSzMGIQ2Uy760L3aPmFNwAuPchZ0S9Fa8+cd+IKAY5c3bMBqD7DCrFzgpw6NlGr1wZBWNSZhhrxr/x4Uw2UzxkbokuNYsymtxV001NyWzsB9NoRJLMvMb6wQBrXomZSCvMbKdBEcZcHDVv59NM2nLCYQoZHYzETHlTYTA5Zza3EyP187eZMR+tCa8dKNYFl+6Npf324470bZmm/YaIfWmxdDnFnG79ezbcmomtZ865Vs0rrgtj42//d9hrtrIuSPtG1PibU2j3lz1xBU3sskeHPaUEadNOLF3OuDz8jR9nbmubnmWRfAMqMngg+WDoZeN7JJC1ZTjzyrkm0uE4KxjpP37jJytBYoKRGoPdpRC9NGrccFS6sT62zYz5aMthcwoWIz24whVD9a6pt9wx425QP6S1nsXO6UYc3POl1rMiy+RiAxFp1Fha+qUyKJmYdRhRT0um/tjlmRl/H882euXKsHQ541JRG39bcGFRCjEANL0tagsWREjAJaPUDd3OaWDNK2vkNq73cdCPM84klrvxY42/EbUU16+6v0mNOTJ9bi0/VZsZ89GVw7Y5aG3wcwxm9CuMu0o23JfKni/Vl9g5CFF1GXHwqjl8PSsYX8LBMwxRMrEXcttoQ5WiOEKhzRp/M4HqrTzvY7HdSoK0qSqmZDIuD7Xxt216FqUUNsxY+oAmQIGpNth0OTtQ93CqGD4dUT/OOJNYNr2LCS5i/YTrV93fpIacDoOlkNLpcqdqM2M+YnXEhcrOG8fBzjFo2q9g/I0od6efKRg6L7SescbfiKrLiIMrZEGtZ9GXOcKF3G5DwfBS3DcSl2cR80EpXfaYyl6NPaUEkYxRDWOtaIy/rXLD8nCHr6puoGlgupAzJd55STJtjJxi2tRS8Ga20xtp6QCNjTPVZ2DlmDnlTfgZGW2gjSneUmHwxWCUCuhhQjD+PrLNjPlIZs1LlZ03joM7xHfjN+bCQ3OolMqeL7WeFYzpMTanG3FoTOYn6wK3V6oayBxAhimGE0vFXAK5YAyiZJo7H9i6cDTb6JUrgzPqM4w1sx8pmcx3JFUGefZ0kcX8lQAOt8fkQk6rNqjXVAuVfE4VzrySUxnFjDPeVwsJFApG1ehnMGqdoW+l1+5iAAhNIc3Ykubt3xzXZsZ8OhNezmcrxb5o8PDpMdODfymokDSHSklhgRkSXwVs6iAyBxlxdP5KWJGU4SIhNKLm90rYxUz7fmb8LQfcMB9Ieh3mqvlx5uy2Luiwp5QgJVP22DDWzFjJRB+Y7bZ+WbgN82BYPE21ijKzFEx3dwuVfE4V1ryye75Y6lXEOOMMczGj6q49OVNN7DP898M/I712FwNAiPE3t5ZXJ2oz4zi6SpoY+6qxOWiFlDlTJAExu84yB5njVUNbOH4AACAASURBVIrSuOsDAeR7LLOexc7pRhzsXoky/hb2SmE/KZmLmfY121kXVMbfqDeifj7o1uSKfI3tTbVso1euDO720zDWjL9IkrcEVqZ7cbiFfLj5iTCzRAwwRdNdkySP4I2/6XTEmHHGmcRi/hLDwYrxR0Aq67Sfr+9bKcAFgOq6gRpLIeVSVU7UZsZxSCa8dmO9PngzZkKJKxRWEKvLCcrcpdazIsugbto5KgTz/zHi4Ey5Z+2VMCWTZEyv6J+Xgmiwjxl/M9lBaNq6xi/U1mEVtnomSCulNCWTcXnsFEqmXdX0XhnGMnAL+Y5Jl5sVoGCMUvMEgw1LoTKvRP2S9OOs4FLymM+Yky4X07dSYPj/ihzWyDHBmO6eqM2M45BMeLdyeLskuEqn3bibzn18YQVp3KW6nnHzraXLHQ93WTZrr4TMOdy60DTNwVR+G20opbZGG38j5vdildcNeWAdiz2lBOFukw1jzaiNv20CX5RZSiamKiaX907LmM3424c1r+Ru4yLGmcok1vuMPs2EvWHV+0skrWRi0mFog9fzt5lxHJIJrx3A10es0W//Gi6FSUqXk1KaFlrPOC85M/4+Hv75ztsrYUEpAFxZU/WBwm20oSalfNY6HJm2buuCjm30ypWRMx4BhrFm/Nse1rjPJvBFGaT/EcbfOWNmeXgf/za4LwkfYeC4ZTjzSuw2TmUijaSMSCaxmL8EKV3HKr0wfWtHpLKkAJcOg6WDdv+m+vep2sw4jlIISNgctD46ywncjBkPCJdCdVTR+FtjHr6IkokJ6puS6Wi4IinUesYVG9hhFzNMgY2ttaFq3zjxgVQYfyNVXm1vejz2lBKEM+ozjDVT+UomUs5uZbqX5tTG37u6gTJ34NxUkkwZtO4WKvmcKpx5ZXXwZBg/X/5mHRtnJZNm0gcXw8otrMKKNv7G+kpFBLJSoGQOE4M6IhgTTArOqdrMOA62AmDVJNkXDZ7OlBebY7qD+kTJxPnRVbKCUFIyLbWeDYdyWoFpc8p8uKAHtZ5xlzlVPZ1zuDWzv/zZSNAjF/aNqPE3lwqLKBudc+xrTGWvZxu9cmVwpoWGsWZGxt9cOdaN5JenypDSRB+oQ3+JgvGSq+rpzY9klFqZJHlEzpU9JswuAeLGWZFl0DTU4YzwMskzxsukRn2f2s9nVHIJbpgL5uDa3yYjfZyuInSaNjOOg5u39nW9mTSUS6JgFIEVofzgzYHltYjzbOu+y1LG3wC4kmnXr+U2p8ylvyxDlWL4epZzcw6ixOHWBcrI/lLhxhlVgCPPHatK8t+3/xxGbbY3b0Q1tnomSMlUHjCMNaMx/raKPsvDVdfo2jAs4cotyrFVxpqmORyorR90lIyPww6Rb+t8BYjXRNx6c9L/PXIryxpoExu+FOhTFiKqNLUVq87bZsZx8H4dFtRbI1xBgh0TEObT5SQlE30x0zQNetFyFajmdFtnZ8MWsiBTM4W9ElEsA2vDlNfMc1Ay6Z+U0XqZ0RdhQ7phqNCmxR672hTFWuwpJQh3WDOMNeMvkrzJ5jYWzFThFnLKaLLgFnIkbYrd/CacNrUU7GG4RoJ4QhlebJz1aZJou9MV1DjpP+UvwRqlJriBK5kAEObrAHBIlyMDcKdpM+M4CsYDsw2SptcXDR7OQH/fp8shKTXMWA0vVcjP5FKaFkyXi5nTDT3cXon06hOCHqTxN6eW2sg8pQmaYkG6ummVTtPXEPMBE3Su7PJBzTZ65coostbbBDMtNIw1M1Yy0XJXk6IuS/f88du5Gcbf9TRtqmQl4F1qli1RHazxN6KK4YxoqXGWc6kVxK13nnHlwulASkzJ5xTIGQUWtdHPudvQE7WZcRycB+YeSfc00mdQHdIH0cl6xKT9atJj+rHL+BgusZ71czqjgrHA9XxyVjU3w/i7mhp/d4Es1Ph7a0omxb4RM/4G4APA2F6I2s/uTFGsxp5SgnCmu4axZnx/EnTTQ5QXNq4W7rBLGU1K6VxYel37GVw/2MbGSQNnRFkhRpRZ5iBzceOMNYnlNsyc8TcRSEGNvxNO3xg2qsxhDUmLoL03TtNmxnFwJrxm/L1OSkb5QY0j0fhbTJdLcz0rNcHxBOfbtaAqkjLxr4wz/u76Fmv8vZF1YXgW+gIcXEGNvhDAJLDHGH+boljNNnrlyuBukw1jzfiTNnsTsZFbmVRhN8wVbhbKmxxPDXTZ9ALrBxMk80pM/UOZ2ZKycoWSCTX+Ziq95JSSiTGVT1HJxKUskCmkzI01ZWwf22bGcbAmvEg5cSN9unGIHsqJcZRnGT2/Kky7Oa+5Jccul55MzemGHq5ICpWOmHNVXDGvvk69w6R7pbhmngNeUYz7rbGvqRrIXHvB42PG36fBVs8E4W6TDWPNSMbfKXuybAmuLO9gAD1NN4iprMOrpbblM6CBM6/c1VPFEEB3O68fZ2wAiDD+Lpiqd5iCjfMISTl9g5PckymkV9BmxnGwxvWKqmJGepRsIJsy+uWUC7Jpt3MO8gxPvRwCAcsZf3Opg1tRwZwDVZEUzKuPMZXG1MIA1KUfHli5VFjjb2IdHl6DqwyxcVlmTNEOM/5WY08pQbgbU8NYM90GLCdMNi1NKg342zl8k54Lm/TwQJ1lDhyRGkQFsraMZF6J3axRZrbUOOOk/1Vdg3NxXiYV4l3AeYQM80N6WxNO3TfcoOq9N07VZsZxcMbfdphYJ924wlWynNEvly4njzt67C5o/M1c5pjx9/Gwxt/EelYI/l9TlZ3C+HsjyppcsW+kqvNRykbM1J9Ln20LmmzjeR+LrZ4JwpVCNYw10926XF/m6CJREXJX42rhb+eocu1MOhdyOwfQ3RZx6QXWDzp4JQ1uRFkSqWzUOOv+jfkdsMobzsSU8EfA2j3lcsxcysIwJrDDBGP8fYI2M46D896wKkLrpB+rbEB4ag7MB8vlflAS471acD3jqnFRxRwMPf16xjzfmMuHHWIt0Bt/c5+xoUAhuW8kFPAls3ZjF2EA9OVD0zTk2m1MsaeUIFyqimGsme6gdF2Z83JX29gvCufNM6QbTNPfyNQg4rBGpQalXGVsKYbbedz7AbslzwlZPhkoZL0LGOUNlyaJKJ/a9+NSMdPbmgwKY73yrmAq7+1P1GbGcVDPG6CrImTPe23wKbn4OKLU1d1rNAEiSs275HrGZUb0Shvr47Nh90rEesb1NdT4m2vDXpmX3pp5Lsh9I1GAI2fW7nZsE3tT9rLHxoyG7fTKFcGlqhjGmtlXbcrNtYWgYLFNz6Lwt5+UaoNJ5yLy3qn0AlMyTWEDQIR8m/LJosaZZBJLGVVj/aRpmoOsnDJ81xulpkDBHFxJ4+8sg6YhKumdqM2M4yjyjDHhNePvNVIwqk/S6JcpWU7NryEloeZdcj0bnoVeaWPo4dKoqfWsYPx+9siayaV/bjHoQQXpuAIcALTxN76vwS8fqEtWA8dWzwThUlUMY810KTd55tgSpCZFXZZeXswcqDHjbwAmnYs4ULOBLFvIe1jzSkRiDwCHin/6ccaZxLY3rPhncJVepqancrWjFNM3+JQFOoXU/334mlO0mXEcZcZVxTQl0xopOeUCZfRLjLv2fXTpMUWW9alxo89ccD3jLieGtdzmlLlwmSfUelYyfj9Yirlzjkzr3RGBlUsmdt/Y9W9snt8RFUSLHA8E9vuaBPcoKWJPKUE4+aVhrJku5Yby76HkrsbVwpocE1WuuHSufYUb6FKpKlSAYstI5pWo5DuLG2dSxb+YQCHZTxglU8rpG3zaCZFCylwYnarNjONgfbPM+HuV8MbflNGv1A8Uxt9Eafol1zM+PRkv5mDo4TJPqPWsyDM+sI20B6Xe2W8wjVpUGZE2APg+BRuXlG/W3i5Ao7DVM0G420/DWDNdyg3l32NlutOANf7uqlxF+PlQxqlUalAvM7fboh7OvHJHmFdSZrbUOOOCi7uKUDJRHkJE1ZuS6Scpp29wB1cyhbQzUkefz2nazDiOMnd06ooZf6+Sboxg7UoZ/ZaEf1pVN9A0urWIMv7eE/PDVcCmvhPFHAw9fJEUwvhb8OrD+yeurNmk8Texb6TWYU5lT1s54OvwztbhKOwpJQh3+2kYa2Z/uBmmyoPuiRx242phFS2Hn2Gl7Nvf4wszLknGzRW3uHGS4Mwr91WN3s7TSjFKVs4pAHDj71bJxPmQUL5PnIF2elsTLmWBMv4eUinO12bGcVAKgaZpWn+UBPuiwSMZf8cY7seMu7YvMRXtllAysWs5PqcbeuYUssgZr74dqbTj+2eKa+a5oPaN1DrM750oKwdCyURcnhk42+mVK8KMv41LpTO7bc1WmTQpu11blCxzkDnB+JtSwRALM7axJ2XPCQcbloIzr6SUYrHjrF97iFtv/DOowEt8P6nqtNM3qLLGtMcVfwA5RZsZx0GZ8PZ+NYn2RYOGm8f25IUHb7ivSpfLCH+6BdczvoiHeY4dS59GTaR/Y+sZl61CKe0K4jJni8bfVMGYWcbflNdkRqkSt/e8j8F2KwnCDQjDWDNdyk1BGH93i26qh8wtQVXbIY2/M0aSXDWMCoYu+Wz9YGCQfOtNpMvIcdb9G233Ck+tyKmD1Yx+knr6RnuI5MxAqeeJH0BO0WbGcbSGz3T/TdEfzODhFJm7mlJkEob7vXJXly5HmQu375Ge8bdd5BxHF7AkAxhIv2H7J2L8DdCuC1z61pbakU5lI1L0OeNvQs2Xi/5q23nex2BPKUG4w4RhrJne+Fu88bON/dKQCpX+5izCz4cw0CUl4AlXGVsKyXiaktjHjDNW+l9Tm1/B+Duin+yJDXYqSP01DBqx6YcnajPjOCjD5+7AYnPQ+pAUhLiqlugHfTqS0vibS6ddxJOJN6Y2RcZxzFnPOK8+KjBFKZm2aERNpw7iAaA+0ErsU1Djb1Jlv73nfQy2eibIIL80JZNxWXQpN6S5rZXpTgZKkkzdFnEqGLpiCpEaZAv5BNa8smrQm/bYcSanVugDhb3xd1hZRzD+TlmpUwilkydpEVzloRO1mXEclAnvEDhMtz8aOGw1KUpBSCkXiGA5Rkmm2Cy3nknG1KbIOA7p+VKqOYBp0KNpGqhq/DWUirbbP6W8bp4aKqWcWofFyzMqqMcZf9vlg4qzPiXn3Mudc+9wzr3TOfc3kd//iHPuTYf//IFz7qPe7yrvd6865/dMjWFjakEm47JozW4Pxt8R1TiMq6et4IFvujPX+jaFfw8QF6AgJeALlnxOFc68kjWzjap6w0n/8Vtvrp+0nzFud+ccY5Cbdsn4MjY1kEl9P1WbGcdBmfDuLKi3WqSDP1kinqtYdYTx95LrGTunm/H30QzPV7+e5cQZb0h9izD+3qDikto3zjH+pnzJ2qCzGX8fS3GuN3bO5QDwYwDw5QBwNwDc6Zx7VdM0b+v+pmma/9b7+78MAJ/vvcVDTdN83rm+X8pw8kvDWDOdQbEZPqcPJ0nGS77S6VwVZfxNfEbVL+TWDzpmGX8L5pUTWTkr/ccPZ5TijavKxI3/lDdvBeG5QpUo79MiiHS5U7SZcRy+ei/P8v7nQ4pTuv3RwOEUhJQPUZeO1DQNODe0ORVARj83d/CpvX5+uArYOZ2Ygww9XZEUfF0g1kzC+Jsyrgboik6Y4hJArrQXl7ZOWDkQvk9LVopcI+ec8V4IAO9smuZdTdM8AgA/AwCvYP7+zwLAvzzj91kNg5TSlEzGZdEpWsgF08p0JwNZSYsovS6lc2EbbNL4e4MbJwnJvBJ/vrRiCABJZZM8hCKq3nAeQiX1msSNaEnPFeIg2h8m0MPBadrMOA6q3D134DPSZq6xMsB0fo258CgJf68lD6ac/caeKOZgxEFWJiTWs65/kupJIsWOWjPb32+nHenzA34xIFebJC5A2X3Ndp73MZzzKd0CAO/1/n334WcTnHO3A8DTAeBXvB9f55x7g3Pu9c65P3G+r5ke3C2MYayZfd0GKMzwOX3owy7tzQPAGX9HqGBsIZ/AmlfWhJktdeghxlnJrD2UiWlJGsTThzPSVD5x42/ac6X10fAVEACSF8Rp2sw4DsqeoDf+Trg/Gjii8TfhwQIwDcZwKUyT96CqTy64njl3UI4TgeuU59u1QKdv8cbfYZtw/YReF7aXviUFgCivybhAoD3vU3C2dLlIvhEAfr5pmsr72e1N09zjnPsMAPgV59zvNU3zh+ELnXPfCQDfCQDw1Kc+9Wq+7ZkpiJs1w1g7nRFimWfEYflg3Gcbn8XhPCooT4v29/QhfPoa6gbQyrWHcF59lIl0QZkaE+MsZxUAtJdJ3QDUdTPy6eI8hNrvhZmFp238TXuuEGWQGS+IU7WZcRyU2qwbAyn3RwOHD8gTRr+eWfh15TRtEhurk/cg1SbLrmf0Wp52evJaIH30GBNvgOleqVsnqLXEjL9bCmLfKBfg0PuSSf5qFpzVcc6w+j0AcJv371sPP8P4RghS5Zqmuefw3+8CgF+DsV+T/3c/3jTNHU3T3PGkJz3p2O+cBF1UFVsgDWPN7Ks2/5nyZOlvDTe0YKYKZ+iMK1rw27mmaUj1E6WC2ZlqY8JspVjEOJONv/Ebv+47jP6euFVsvxfhd3AoDJAqZU4Yf1cNkUJKp76fqs2M45CUTOYLtz44f0DK6Je63KWqZOKfy1ekWmpuK8l0LrzSnhEHvVfC1zNKaccZxJPpcodLv1BFe8nQ6mk+bZ0uSoOnz+6q1qNt9Bm2LkRxzqd0JwA80zn3dOfcNdAGkiZV4pxzzwaAGwHgN72f3eicu/bwv58IAC8GgLeFr71Uuqgq5r1hGGumM5qk5a5W0ScV6FtZ3F+J2thTefLtZ9C+T913MFqogEVVN9A0VFqa9HwjjL9Jo2qq3RnjbyZdNuU2LwgPsTYtAg+mdb/3OWWbGcdBHUA4TzEjbVgFIWX0S6TLxaSpyvuaZfpSQaW+m5LpJMSuZ6R6kglgcEbUKa+Z50BcI0kbAML4mwkEhudwM/6O42zpck3T7J1z3w0ArwGAHABe2TTNW51zPwQAb2iapgs4fSMA/EwzDhc+BwD+sXOuhjYQ9nf8qnSXDnebbBhrpjOAtsoN6dNKkqlNOq1kCl/DKVpKysPC+sGEoVJZmNZDH2CkG7+J8XeGB7IADmod7nA2kf7ThzPO+DvlIEqR4UqmHemvhKdFcGXRY9vMOA7K+JuqVGSkj3OOLkFOBFa6S5BpsFEfICI92xYeu1R60b5u4HovNdCYR5Hh9g87SsFNVB3l+gmtZNqeeXtJeozFK5mkQGC7Jxn/ffsZ23rmczmrJ1PTNK8GgFcHP/v+4N8/iLzuPwHA88753VKGu002jDXTBSiom5+lZeXGQMneztHG39ONE22gW+SOLP0LYP3AhwpYcIqLgjj0UOOsN4mlTEzZlDB98Iszok05HazMM3hoV01+XjHVEwGwilWnazPjOCgTXktPXDdFluFrC1MlEyAuhWn6HoSPS2civ1i6HDOnX5eKNe96KYl9TEWsZ8NaHmH8TQYwt2feTnmMUQU4OOPvXYUrGwexR+DRxlQANKbYbiVBuDLShrFmukWAMv6uIvwPjPNCbdIp429KBdMrmSKNvzMHIyPprUN5x3DGtEWOm0hz44wK/O3IdDk8kMIZf+eUIogwiE8FqioeZfxdkOqI07aZMR/KhLfqVQW2TV4jdKU3quIXfvDvx50mXY6Y16q6WXQ9I+d0S5c7CVTlN2o9o1LMKeNqgLbgA6qWIgpIXDIF4TFGr8N0ulxV42OAujQd9jXbeuZzsaeUIH4E1TAuiV09GH9Tmx4AuyVIAfqwywcbwkN49x45ZfxNmTHaIj6CknzvOKXYjHFGpVZUghfX9HDGpEnmjPF3wod66hDZphLSY2KSLnfiNjPmQ3lvcAc+I304tTR6EBUOlZp+UOTUxcyy61mbzqWf0404uIAm7/+FV7RE7QhI9Q5uX3DJUPtGqgBHq26aWg0AHCoAskHn8PLBLntisNklQbLMQebwqKthrJnOD6E7rE0qNxxuIrZUKSNVyCpXVLCB8jZh0k5ICThx27xlKK++PXOz1t34xYwzLuUjqipTRcvKWVP5hNu9yAjjb6JKE5X6fuo2M+Yjp67Y814jpEk/ZfQrzK+a4HeRx6mnrgoqBXq3wVSrcxC7nlEBDNb4m7qYSXzNPAf8vhEfpyURaBUrABJp1Ft75nOxIFOiUO75hrFmOmNfsnKDybeToeDK8nLG3xGl7EtqU05ImLfMcNMeYfxNVCrlxhlpEisZf5O3srjfAdbuS9/4S3DqCE5yH25uWePvGW1mzKevPER4yaXcHw2aklEdUsbKANg8Fmn8TVZkXTDIRKQXUXO6EQdVJIVazyivPsn4GwuS7IggySVD7RupAhwAeKp7XTdQN7RvJAB2abqsv9rasKeUKNTNg2GsmS5AQfmO7WzTkwxcNRPOZ2CSbjDD+JsyY9wyOXn7SSsu8hnjjK5uhvsdlMKtLJWaQhqlJnxDSHvJ4c+zJFJIOePvOW1mzGcYV4Txd8L90aChU9fwtYWaX2MM9/NDqmuoQlx6PaPTi/A53YiDKpJCrWfUnCMZf+Nr5vaUTN04C6EKcHSvodPW6UDgpI0On4ul2BlTbMeSKG31HVMyGZdFb/xN+I5tsVJGqrSSZFxizylaSANdIkVhj2zKt7hxkuieOVlqHbsxnTHOMOl/0zRtkAMNGBFVmbiqd8zNespGy0XmoCIq1HCphFTFxVO1mTGfXsFCjauE+6NBQ6WuUUa/w/waHvwjlExMBdDl0+Xw1HcLoh4PWcmQWM+oOWfXrwv4WrJG9e856DJ9JsFcZo0skfRZrigNdxEOYEomLfaUEqUkUlUMY810AYqcUr1scMFMlXZTE1GNo5cXh4oAvpQ9AL6QWz8Y05lXkh4BkWa21PPFUtm6l3O+Q6HChzP+bi9RCI+QhG/Wi5zydeAPrpQ64lRtZsyHNv42o/U1w3oKEnMSAKbE1XuwkH2prpc1/ibOE2b8fRqoIinUekammHNefZTx9wYvH0pi38jbAEzVT5yVQ0Fc9pjxdxw2uyQK5f1gGGumWwSoG7/U02W2REnczu2I2znq5odL56JUMHvrByiY5we7USLVZfTzxVLZeA8h+mAFQFWwoyrlpO07xKUScpX3SOPvE7WZMR+pOqKlKK4T0qSf8BTsx10wj1URHix9X0IC7kuO3ZJJS7fD8vG01VIjLh/IIAlzIUcZ2Se+Zp4Dbt9IGn8j6ul9ny5He7TZ5cNx2OqZKG3JUVMyGZdD0zQT42/sJsc8AtIgZ3wc8BK7dJoPAECOGmDim3LrBziYeeWOuVkbNrP655sjxt+9iTfShnP8JSilLhXATIWcUkcQ6TD9s6ECcCdqM2M+fZoUceCzZ75OSkSJ2xn9ot5yVCXICA8WMqV54bGbk5W1zOftFLR+P/r1jEwxZ7z6OiN7zO8r5TXzHFD7RqoABwC+d+rHNvIa0gPzoE6zKq86ttUzVwRlJGcYa6W/Gc4cWblhRxjoGlcPWZ2nbtCAUZY5yBzjzcOm2KVllJoqmHllpQgAxYwzzCSWO3CTnjaH12B7vksz/t4TPmXOObSIRx+AO1GbGfMx4+/LBFN+cEa/pzL+BsBTxhc1/iYKCVHFHIw4yCIpkcbffSoWpoo99J/wY5ZWyS1Bn9qK7IWocZZn0+p8O0alSHu02WVPDLZjSRSqMoZhrJWhPGvG3h6bDDUNihw3/t7XNRow6l8zI51rWjbajL8xMPNKNpVtxjjDUtn2zA0r14Zljt/4FYhaqv1eaXuEzEnzw4zUubLoNjdeLVRahBl/rxt0HmNSXciy8jOMv6fjfWHjb+LSel/jxRyMOKgiKVS1M6qf9KlYTBGJSTB8g75aeZ/aOt0LkcbfWTYJSu25502lNNplTxT2pBKlIOSXhrFW+sNw5pgbv7TTZbZEa2ZJVUzhAhT6dC5OybS1jZOG2INT75MVMc4wk1jeIJOSldObX9b4O+HgIlXVhjtEYkbqp24zYz5SdUQL7K0TNLirmMem6id9P+jmu9TWswJRYFZ1A01jQdRTQBVJodazoZ/EGX8D2OUDgJdSjlVxZC97cCsH3htxqupOeY+SGja7JAq2QBrGmvFvDchSv4mny2yJkridYwMU2XTe6m6PMEkydcDbogRcQ4kpxbiNUu+TpR9nmPTfDxBj3wmAOFgx5uKrNP4mKlDtmUMklhp46jYz5tO1KeYLB2DG32sFU5fwRr/EuKv0HixUpc3W+2hB42/kcoJTwBpxUEVSqPWMVslwF3J0YGprbcgaf5Pr8FQ9zVk5UDYAW1SOHYM9qUQpkNtPw1gzO+9g1S+YyGYs5UPmlihyB3XTmqX6dMaHGJihM58aRG/KbSGfkmNKMU4Vw5hI0+ly2OFMlvHHfAbWT/zCAKlSEKlslPF39xpqc3uqNjPmQ3lg7esanDPj77WCGX/38xjrhTY/dZtSMnGH36sAS33fRyi0DB5MJcOtZ7J/JVMkZRL0MOPvDnYdRtTTvLKROqNsTzl2DNvqmSvCjL+NS2Nk/M0omba2YKYK7Q3DpEEh8xZnoEumFySeNrUUWAojV2p9zjjDlEy9Lwm2+eUOVhGKt31N3yqmwrDRnwZSKWNgzHSXU8nY3Hi1kGkRdaMyezbSpMji/OsGFdJ0vdP2A8o3Z2lPptg53YijRC4SpPUMT2Hk+id9MZvymnkOuGAupTwtkKIpmgtQdP9re1M1NrskCrZAGsaa8W8NukVimnZit/WpkFOKCiZtp5238IUcq0g3RwWzZbAURo3xd8w4yyONv3OmDWMM4quavlVMBUr1UjGHyHym8bfNjVcDafxdWeWtNcN6MmHpMd2eBAkg58pDJd2Xlh27ORfUtwPz0WAKY2k9w1MYFUbU6NqT7pp5Dsi9qaCenqat0/MBrXA14+8Y7EklCmUkZxhrxfd1Ic1tN7hgpgppd2l1sAAAIABJREFU6Mzc5JSIbJxLDSLz3itTbWCg/j6c5xVlsM+MM+yGlTMkLal+wnxGq9TFg2UpH3p61QvSx6mNJ2f8fao2M+YzpEXoUyCN9CnzDFU6AFBGv3TFL63ah0qxWXo9w+w3uFQhIw5MYSytZ1gFcW5doJTlu2p7Xn3UvpELABXMngOvAEgpXC1dLgabXRIFi7oaxprxfV2GnOrp7fHWpL+pMhg6D4ts7zNAbLpRFQyTGpRTee9m/I2C+fv0SjFGYh8zztAbVo2MH9nAkeqebOr31X1myuoRuqoNrXrBUkhP3WbGfIZgOqbYtC3yWuH863CjX9r4W7sW0amuCxt/o2nsdDEHIw5MYSytZ61n2HTOca71bArhPMNSXjPPAZ06yKzDGbJ34qwcuKCzrQtq7EklSp5NJyDDWDP+rYFVykgfbJGVzEKjjb9JtdSyRqmpgplXcgcnShXDjTPUJJYrrcx5FzD9BGCsCNoxKoNUIMvdM+lyuDfMadvMmA9lwrt0ipNxHCWiLuHUO2c1/l54PWvTk/XFHIw4sCIp0nqGXcjtGP8v0vh7g3slat8YbfytqPI6XYdNyRTDtnrmisCM+gxjzQwGxY6+JTBz22ToNjt+0EjyGeB8MNDUILIUrR2oMTDzSq5N5owz9HDGBgoJBQCjeMP8JfZM4CUVBjPQ6caTOxyQprsnajPjODCPMC4F0kgfvEomVyKeUi7QY3vyHsT8sHTqZXlYN5pm+P/Wm0xvLEBxDrBUNmk9KzLc+Ju7rGg/A/E+3Nheid430mMV25tqrBxwG4BtPe9jsNklUbAF0jDWzM67RaRv/LZXKSNVBt8s73ZO8hnAvB/YVKsubRIzjbblKaREVEYa4++YcYYezjSycjTdiFZLtX/jH3rS9wjB/HvquoG6oRUBBeoNQ6sI5rSZcRxoBUCrcLlq8DRV+sLDOUeWOdf2A67S5pLrWTen+hYcXNVXI44hQDn0HcnzCk9hpNWT1OWDGX8PcAU4UI82tsorV81vW8/7GOxJJYoZfxuXRreJL33jb3SRsGkpBbBAAFedB6DdOIVeclzFlKGiz3RTbge8KajxN3Nwwja/APw4Q9VozK03pQDYM94FJRKYWofx99S/Z6jSxJVOxlNVTtVmxnFgSqZWiZduXzR4SlSdRgfku5+jqtojlUzc4fcqQFPfmRRoI47eR89X5tYzjL8ZHzjq8mFXb8+rjy4Ywxh/ExcJAHHKRtubxmGzS6KUiI+DYayZbrLOM0em2Ows3zkZsMNupzjKI3wGuM0sdSPF3ehtGSxdrts4YYaXxYxxVjCllbHPcM4dSmRPFVaUCWeObOD8+SFV0DS/mn7+AFQK6WnbzDgONBC4Qa+TSwKbk6RLEtQcOOJQSVVkXXo9ww7Mw1puc8qxYD560nqGzzm0iTfmGVbVDTTNUEBlK9Cpg3wBDtKjjfVGnI7llPcoqbGtnrkisBsVw1gzO8+HhE6xsXznVJjjm1PmU5+BrmIKtjCTee8Ll3xOFdZEOiqVTTD+Jg5n/C2hvqIS5vfFlRNOBUxCz/k6tD/HUhxP22bGcaAKFjNaXzWYDxFn9Nv+nEiXUx4quapXS65nmIkx55VoxIEpxaT1jPIIopVMnbUA8hkbm6dIE3SpAAepZIopBGBVR2OwJ5UomNTXMNaM7+tClwe12+NUGEo6Rxh/R1ZMoTbl3GZhy+DpHO3tnXNMwCJinPGHMya4GGP8jcjdOd+nVMCq2vRpwJyR+pnbzDiONniLHVjsea8VzoeIDQhj407ZD0j1w8LrGXYol+Z0Qw92aSKtZ5R/pWj8jaZqb6sNuWDuHONv7NLUOUdWqDRFsR5bQRMlR/JHDWPN9Ga3WYYumO2/t1cpI1W42znOZyDcOPEVU/Abqao2c0UMXClGb3rKGeMMO5xJt964AkBj/O1vmHmVQQqgz0YoBV5kiPcGs1Gd02bGcZRU/7XDxGrB1q+hwi1nxjydX7XjDqsyBsAffq8CdL5dQVB/LXR9bbwu8OsZZvzNreUFFihcgfr3HGDBXKkAB278LSgbUfWTVR2NwZ5UomD5o4axZnwjROq2fm9lupOhwG7nvEAhBmX8TRuFTw/tTdMsbpSaKkXmpibpnNnljHGGmsQKt96ol0lFK5m6w3tMyecUYI2/mQAcFhg8ZZsZx4FWALR0uVXTB2sj0ovwtEm9ghBLdZUOv1cB5yW3tQDFOUCNv4X1DL98kI2/sQuOrQUK8T2KIqUfWYf998M+B7dy2NbzPgabXRKlRCKohrFmukWXMv5umuaQWmUTeApgFTy6TTptTolVaaI3Tlje++BXY8tTCG5eyZhdzhhnuF8Sn2aCKQB2dU2ayubooSd942+sRDln4g1A36Cess2M48BMeHfmvbFqsLVFVGRm2WR+3UUoCDFl7q6/XFuuL2Gp71wxByOO4dIkwvgbu3zQGH8jarTNGn9XUyU0b7Q+tgHYCfNBifhTmvF3HNvqmSuiyB3UTXsLYhiXwJBqhRt/d119awtmqnQLKZo2RRo6x1VM4bwibCGfgppIM+kcc8YZ3u68rDzPpgq2PRfIQg896afLYRv9PgDHBNTO3WbGcWB+HVVth4k1w64tTPp2OFZj+gGWllYlEDzn13KbU44FuzSR1jPUv5LxgcMuZs34O8JvDVGCVcJeE93X1Mumvq4Ne1KJ0m/CgyiqYawVX5qKltTd6IKZKlggYGhDxpsHM/4m/z6+WteWiU3nmDPO8MMZH0hpvaKmh/SolLwVeIRwB1fW+BtRn52yzYzjwA2f6XRPI324tYVMYUIK7uwi0lTxsbv8embG3+cF3StJxt+IfyXnA4ddPmw3XQ7zGOPXYdxjVHNpiu1rtvW8j8FW0EShqlQYxlrx/Xycc+0NPyIv3tqCmSp4IOCwkLOb9JiKKbSZpd2wTsGqjnKl7Z1zEy8CaZzhptxSmgmi1uGMv1G/r/Q9QjgDXbZiFaLuO2WbGceBGj4z/ddIH9SHSFCX4P1An6baG38ntp716UWY8XfC8+1aYH0MI/wrubW89wPEFMYba0N03yhdhCEebVyV1/a9EOU4c2lqTLEnlSjDZtaCTMZlMGy22gk9lAtvdcFMFc74m0x/Q25+2Epa2KZc2CxsmTAwC8Abf/eviRhn3OGM9pdAjL+ZNJNuDlibiSnquaK4QcXSIk7ZZsZxhM8boDvw2fNeK7wSl0ltxVSHyn7QTXeY2mRR4+8ZqYOGHmyvJKlkUP9Kzvj78BmVKZn6dRjdPwg2AKFHG6cwjPXANKbYCpooBTIgDGPN9Df+h0W0DA6m0qJsXC2c8TedBoVUTBGMv50bcuPbv7cbVoouiOebV+7rmt0oxY4zziSWfs00+LXjUsLQG38+FTMFsAo/lXSDiqjPqhO3mXEcbbonosSzw8RqwcvKC0a/2TSFifNPC3HOTdJj+yD0guvZcJmjTx009GCXJsO6QFdYnaTL1XTQg0sx39rlAx4wircBkFRJWAGunaXLRbGtnrkiqDLGhrFWdr18+KBkyl0QXEjf+HdLcCamdBoUbvzNLcpF5nAJuC3kEzDzSqnaSew4w9aezvyWlJUjSpCKM7fuFUGI8XfChx7MF0Oq0oSVTj51mxnHURDG9TYHrRcuFZtLScLKyscc4sPS9EPwfEElE+IVJQVBDD3YpYm0nlH+lVRxB6zq6G6jajQ8mCusw8jeibNyaF9D7Gs2FtQ7BntSiYKVkTaMNRN6l0yDC8sbZBoDJbJJ3wk3Z1iVpl3NG+iGnjW7FShaloLyyeJu48Ky3BoPIYAgACQob3AT00jj7wTSSiSwCj+S8XeXSjhSn524zYzjwNM99WlSRnpgPkSaClRounfEnFTk44uWFALEBapOtcucU4FdmkjrGepfyfkYYsbfG/bVCveN4jqMXmwJe9OgoEnTNGb8Hcn2euZKGCYUUzIZl0Ho5zNdJNI/ZG4JtOxxv5DTAYqqnh6oxbx3rOqNHagn4OaV/KYnNLMVN79EuWs+KDJVAOxqplIO5ve1At+hvA+OxRh/42k7p2wz4zhQw2cmdcVIH8qHiFNklpjnXWQ/KPMM9xhcsC/N8bAx9HBFUkjjb8z/i6vI2gVJTPUNAMi+UViH+5TRkZpP2JtmY0XxkLZuY0aLPalEwVJVDGPNdLc03QYvlKKu4ZC5JbDbz70gSUZLJQs3wWVwo2eqDRrKvJI7KEQbf6NtyJtdhjd+baARSOn/1oy/Acab29bvx4y/UwE34eVvuY20wXyIJKNf3AA+TtE2qZqbwHpGzekA9Fpu6Jlt/I30Nepipqs6ivpXJrxmnosy8P8cnkWk8bdk5YCZ5duYUWMraKJ0E415MhmXwj5ImyqDg6mZ26ZFSRyO29/xhs4jzyDG+BtgqoKRPmPLoEE8wVcgdpyhJrGC+W2ZjZUgUirG2o2//e8tGn8jBxBJcm9z49WCmfDumNQVI30w9Y5o9Buk/TZN074mRskUHExTWM9K5NLa5pTTgRVJEY2/8xn+leHFbL3dy4dpMHeG8bdQ5bVVJerTbY0p2+uZKwHbhBvGmtkFaVP55FZmuwtmivSeFsHhGEBOtdrV+tvj6W2RpQZRYApX0UQ6cpyhJrGid4FDN9ii8bf//2MFaZJdNcTR8xfSYbDU91O3mXEcqOGzeW+sml69E2n0G1MhDH8PXGGRhPF3MEdzqYOGHnRdFtYz1L+SMf5u3wuvOrpFZc00mCsV4ECMvxVWDphZvl2A6rEnlSiYSZlhrJnwliYMLgyGz9tbMFOEuvkBUAQogqARt0kvgtuirZbl1YA/X0EpluMbU0llNDH+Zg9n2SSw2H7fiH6yklvCMkitktMPMR+t07aZcRzh8+4NXm0OWi3YJUmopsZec+yFR2j8vUsg1bWk5vTE59q1gGWeaApsTPwra1492QY9EOPvDQY9JvtGsQAHcrHFFCcBQOaDDXtgzWV7PXMlYD4OhrFmQm+DUIq6T0BWbgw452hJslAmNmxXaeMU3jYDmIwfAzcYlZViMeMMk/5Ln1EGXib9LW5EIGstHiFFaMotBICGA0icus/mxqujCOY5SyVaP5QPEbsWTfrBYdxFBIiKiZ/a8n1pzrph6MHTv/n1DFMMSz5w0yDodoMek32jsmru9IJOX4BDUi0bU2zHkiiYkZxhrJl94G0QSlFTqMJijKE2zGLee1hJi9k4lWGVwRV48ywFdTvPq2LixhlW+lsyv41Vo6F+Xwffp9TTNyYKTPEGdbq5bf1+TtdmxnEUWVgRbPmy88ZxYD5E0Z43MxSERZZN5s72PZbrS+i6UcUZmhs0Q0Bz6mNIrWdoYCo26NEFMDeouAz3jeoCHBGFAIo8TrVsTLEnlSjYbbJhrJkwbaqcbOwtJSQ1igw/7JILObKZDb24Jq8JNvY7QS21ZfDqfYLxd+Q4m62WQtRotHfX1B9hLelJRWQ1RCz1XfL7sbnxainDNBQL6q0eLCV3J114hIfKGf2gDAPECaQBo6oZoZiDoQdXJdVCXyP8K6M8w7arZMozyj+NvwAN9ym8coxYFzb4vOeS/o5uo2BGcoaxZjDjbzwVy6alVJgqVGQzSwAkQCH4+aAbe1vIJ2Aqo53GlDtinFG+WlHKGyFFBAu87IRgWSpM1H2COTAdtDtdmxnHQSpYLMi0WgYlU4TRb5imOkOF1K6Zaa1nWPEBaQ4y9JD+X0JqZvuatk2aplGpvrGqo2tYN09NW50vxi90hrKRTH21caPFnlSiYBt9w1gzE+PvcJHY8IKZKmWQ9979b6mCRygB580VcR8M2wBPQQMWgoFrrIk0ahIrBIBarwh9NbQsc5C5ON+nVChzwgxUGBPh8zllmxnHURzM3DsT3sHg1eagtZL3e+gIo988i/J5Qd8jS9j4O6h+ZvPJacAVxnIqFsCwVmpUcznh1bfFvdJUUay7AA0DU+J8YCr7o9hez1wJ2C2MYayZUJpahouE4G1iXD3FxC+pPRzTPgOYQmXsxTX9DDfZnAGY6S4GGsTT3MZFjLM56XLlJOVR3jBPKtIJwchUmKiMtDL9M7aZcRxhuksXaLA5aL0M1aQiqndloedN/LiLVXVeBVQ1Tzssnwa0SIricg1gaBONF2UY9EhBJbcU1L5RStEPzdm5vSmZRr3B5z0X27EkSncLY8bfxqUQegCQptK28UmGSRqJ5CWDBCgq4bYorDJYCelHWwY1r5RS2SLHGS79F7wLKJ8irt3DinRCMDIVQv8pqQJVgVwYicbfNjdeKaEJ75YVApdCicxjre8b73lTNwD1YazOOVROLmYSMP7GVF1rCeqvhem6IFyuhXOOoqpuGPTYtPF3sG+UC3B0e9OYdLlsEpTqfm7oOOuTcs693Dn3DufcO51zfxP5/bc55+5zzr3p8J/v8H73rc65uw7/+dZzfs8UMeNv49LYB1L10GRzzq2hcV5KJH2A29CQxt9SRZ9RSl53aLcDdQhavU/wGYkdZ6j0XzSqbm8V+3QjhSFpkWdT4+8VjP1wc7uvGshcmwKI/j16wJOUYTY3XiVlcACx9MT1g/kQ7YSKamEGwZxUt9D4ewgELNeXnHOT1HdTMp2WMvTiEvva2Phbc5FQTMyut2v8PTVB559fiSiZdkKgtfV9MiXTMRTnemPnXA4APwYAXw4AdwPAnc65VzVN87bgT3+2aZrvDl57EwD8AADcAQANAPz24bUfOdf3TQ0z/jYuDdT4G5nAKb8f4+qZtpEcMGr/LghQsBuncHNm/YACDeKJvgJx46y/9Q5u/LgAR+5Vi/ODhlxAMtzArcX4O0eMv7nnPzxP/Wti28w4jknqiuApZqQPbvzdwDWFIoWpruEayFQH/+l74OqHpccutt+yoPXpwJ6vpJIBGOacnULx1qbLTZVMWwwWhulyw8VApJJJ8sBCU19t3Gg555N6IQC8s2madzVN8wgA/AwAvEL52q8AgNc2TXP/IbD0WgB4+Zm+Z5J0tx47UzIZF0K46IbG33ZLkB6YBDxXKZmGiilSaXrS+NsW8glowEJQMuWhd4HoIYSnd3GHpDC4uFMcrCbBmpXcrE/NgWUTb4DA+PvEbWYcRx4YI0v+Hkb6oMbfyoDwrg82xveDPHeB11wa6d9lNlVHLh34uiRilWLhZY5WyRQWVckZj8xLpi3WEGH8jZypxb3pQW3dKbS7cW3jRs85Z71bAOC93r/vPvws5E85597snPt559xtka8F59x3Oufe4Jx7w3333XeK750EQ76uKZmMy6A9WHnpcsQiscX88lSZpgYJxqmBBLxPFZDS5ZADtZnuTukDQIfnVdcN1A1/CIodZ6hJbOCnNv1e4yCTxuy2yBDj7xWM/RK7sRbMQwHO22bGcZRBn7c5aP30h8ogIMwb/Y733f24iwgQhV5zqfSlyTpb8XO6EQfmxcWtZ+G6oDb+tpRHAKD3jaTxN3Km3gn72TIj9jW2DqtZ+kn9WwB4WtM0nwutWumnYt+gaZofb5rmjqZp7njSk5508i+4FFjaiWGsmV2wqQlzqu32OD1Cv6QuHYoizHvXmJ6WYZUxM1ckCSXfktll95qYcYYpAGQ12ngDpzFvxyrSreHQE25uZRPvcer7OdrMOI6un1b9YcLmoLXjnJtUixONfg+/6/uBoqz89D2yYOym0ZfC4gzSnG7EMV0X+PUstEQZ1gU+6LEP0uW2mroV7hvlAhxU2rrenH3LHlhzOWfvvAcAbvP+fevhZz1N03y4aZqHD//8CQB4gfa1l05/WLMgk3EhhJuaSU61lelOjjILPABqwfg7qH6m2aRPJeBp3PymSPfsd/2mR36+ZVD2WBpnmEms5JcUmoXvBOk6AFKRbiXVjkqkkl5MxUVNWkRsmxnHMfRfM/6+JMJLEmmO6efX0Pg7Rsk08ZpLYz0rkf2W9e/TERZrkPvaeM0c1nLJqy/8jG22IVaBlS/AgdsAaNRmMcp8Y8w5dyx3AsAznXNPd85dAwDfCACv8v/AOfcU759fCwBvP/zv1wDAy5xzNzrnbgSAlx1+thmw22TDWDPhpqZAqjQBbNPEMFWmEntlsGGSNiWZWY43C1v1GZAYbtrDw7DgeRU5zibSf6mqYHArOwQKpe+1Pun/xKdMSIuYlKo+U5sZ8ymCA8jegnoXQXhJsq+lsvLBJcmMQ2WRTYMNKaxn2H7L+vfpmHpLCgVPAvXkENCMuZjZrhptGkCWi2kAADIG+AvQ7u/az0hDlbgmzlZdrmmavXPuu6ENDuUA8Mqmad7qnPshAHhD0zSvAoC/4pz7WgDYA8D9APBth9fe75z7YWgDVQAAP9Q0zf3n+q4pgpWRNow1Ex4ii4lRopnqpUaRZ/CJR6r+35Lx91BlbOxpwVdZcf3fA3SHdusDGKF5pWR2CdAZ0caNs6kpfw15xGZMY/yNbRKvK9PfvIVBUSmFdGizLgB3njYz5pOHbWRBvYsgR9JO44y/4w+VrWm/voLVVRHut8z4+7Rg6wK3ns2Zc8ILDqmAxCUzvQgTCnAgxt+S2iz0cepea+NGz9mCTAAATdO8GgBeHfzs+73//b0A8L3Ea18JAK885/dLGefcoXyiKZmMy6BVwQTG34iB7tI3fsZAGADa14Lxd7CQ7zTpcogh6VY3ThLhpkdT/WjOOGvbPfCXiKgupzb+DjeJ1551S3ISUJ+XmGdzpjYz5jMx4VWkrhjp084xYXqMwvi71s+v0/eYpn+nsJ4VmYMqopiDEcd0XeDXs+mcI6dmtoGVMGi6zTbEjL+5cYadqaXX9JdnEcp8Y4w9qYQJ0wkMY83s6qnxd9201ZYAZBM+4+oJ56BKrKQVpgbJN8GY8fcavHmWIAxYSGaX3WtixxlqEqsxt96E8fe0FLjK+Lu/DT1PmxnzGdI9g3nLnvmqaeeY4FAZkx4zQ9GGGX+nsJ6FJd/N+Pu0hMpc0fg7LAjRVy7j7QgmHpkJ9K0lwIy/pWfhK8GquoGmkTywzPj7WLbZO1dCaP5pGGsGM/4GGB+YrTRoWrRz0HghV+W9TzyD9MbfYTDSGCCNv1mfrPEBWjPOJiaxscbfnaeN5PsUGn+vYPyXYTqMVHExeDaaSnFz2syYT1+qugsEWvGBi2B68JeMfqfjzv+5hjJz48NvIutZmcfN6UYcU/8vpan0ZC0XjL8tXQ4ACE9HoT+X+aAE03hgkcbftharsSeVMG0ZY0uXMy6DcFMzMQvesPQ3VaabGkHJFARBNAa6nTKkaQYVwRqCDUvQeQHEGH/3RSQixtnEJFaUlU/Hsv/Z+Gdgxp3pj/9pqWql8XefgiOnRcxpM2M++SQtwgxeL4HJwV9Ql4QFd+aky+VZBk0zBA1SWc8mptGWln5S0LR/RV+LMv7G0qg3qmSa7BsVyrzcS2nUVDEeVMjj/SznT2mM2WbvXAlFNjbqM4w1E1YzmZgYr6SE+ZYIPQB2ygoeVRBs4ANT483WWoINSzBRxagUQ/HjzDeJbZpGTK0ogu+lKf3t3yoCrKfaUZguVwlKhYmvQ7e5PXGbGfPpnu3EhNfmoVVTBMbfYtrvRHUYH2wc3iOtAHGRTYs52JxyOsJ1YS+kspXEnMOpZEpEyZSCSm4Jism+UX4Wpbev0YztcCzvFPtZY4zNMAkTRq0NY810pXw7ukV2uMnhzYWNqyf0AKjqmm2jaXUe2SixuxXyzRXXEGxYgtC8UqMYmjPOfONpzY3f4MU19mSSVAOh8fcaqraUoXlo1aief+jzcuo2M+ZT5uMDy06hwDTSJw+Nv4X1Cxt37c/jjL9H75GIb06JqFNtTjkd03WBX8+mBSHkCqJ5aPxd82vPJYMVG5GeRZFlvfm9ZmyXwWdUiv2sMcaeVMKE8kvDWCtN00wOShNpeiIGmcYAmi6nCjboDXSHFLshnchuimh8LwKNZ0i/GYsYZ/7aM6g6OHNrooJaxK1sKt4lEkUeqvt4w9E8c+AcYvx94jYz5lOEab52Y30RtHPMEPhuGt08totIbZ2+x3QNTKEf4Wu5zSmnYpJGLSpcw4IQuqDHxOw6gVTMJQj3jSrjb+/5aav5AYzXYef4QKAxZpu9cyW07vkWZDLWT9eN/Ql9MNXzc6pt8k6JUE0pGX/3B+rAM0h1o+fdMNkiTlN6snyNZ0i/GYsYZ34Vlm5TxnoyEVXvJB+nScnnFWyYi4kCS/c8d2duM2M+w4114MWzgv5o0GAB+SjVYR0fbAzNglNRm0wKbJjx90kpJv5fOuPvUMkkBT3Gfl9ppGIuQbhv1Ngs+Cn6Kt/IsDhR3Ww2qDcXe1oJU2Tj8quGsVYwU8PpLcF2F8xU8YMNALrDbumlKGiNvwG8TbniRmrLjMwrK/kwPFUMKo2/I3y1sLGcOYBMCEyNPEJWMv7RilVi6eQMQiXTqdvMmM/wvPUKTCN9Cq86qibtFxt3AHFBphxRMqWwnk2reVrg+pTEqr6nc45inQ09gjZu/A3gV36TL6lGVgMa429EmZ9CwHhNbLN3roR2gTQlk7F+BrNbz/gbMTG2m+O0CFODWtWGJtUq3vg75kZqy4zMK/sgnmZjqh9nvknscOCOMP5WmMqWWTYy5ZWMUlOhOHxvv6qNlObnB9R6g9cTt5kxn4kJ74yqYkZ6+Cm5GqPfibpaoX4KmRh/J7Ke+elcmmIORhzTIil8cLEML9cUfj+h35fkkXnJhMbfUgGO7jUxpv79WPaU+SmM5TVhM0zClLnrqzQZxprBNmuoua1N4EkRegBoKnjkfqqVwn8GK+VrkmQaXxUz20RaJSufYfzd+Z8ojKrzPDBKrddxSxhW+Kvqplcv0K8ZVARdO5y6zYz5FPl0DgLgKz0Z6ZNnmbcWKYK7WTfuBqVImTtw7gjj70TWs9Z+Qz+nG3FMi6TwaZIFoZ6UjL/Hr0kjFXMJps9PfhZlnkWZ+pfB865WchGWEva0EsY/rBnGmsEm9KESmWf4bBN4UszxAMAO1Bqz8K4fVImyIycVAAAgAElEQVTc/KaKfyOtMpGeMc6KkQJATsmbbPgUMv7Su1XECgOkSjEJAMm3yUXmpgGME7eZMZ/e+Nvr81K6p5E+pZdaPKxFekXmHJ+4PnXYSzdOYT3zPZk0xRyMOIp8mOM161mYmhlT7cxPEdtq0CPcN0rKMYBOURwYfyuUTHt/HbY1IYpt9s6VUAZGcoaxVjBTw9D4cKcw0DWulokHgGLTPTJbVRjoTk2jt+szoMH3ydKk9aDPV1Hq1099kz5jsuFTVFQqPLWU5gCYCpMKVMrA6646b5sZ8ynCA58F9S4CLCCv85bz16K4cTcEiIc1MIW+5NtvaIo5GHH4xt+a9cw5F6zlMdXOzFogXCO1BTj8se2/D0aYPmvrcDzLz3wGSVgS0zDWCubNg5UH3eqtTKpMK6AojL/9A3Wv2pA39v4N0xoULUsxMuXWBPFmjLMS9dXSBwp15uJxFddSYVqBSqkMizD+trnxaplUKlqJqs7gKdAUMc085q1FkeMOMwtOoS+VXnryHENzg2fk/6VczwrsNRHG3xqPzEsl3DeqC3B4BWYAJEXxdB1OIWC8JuxpJUy7QFqQyVg/2AYvPJhKOezG1dN5AFQRaVAFspnl/Wfic+u3DGZeqTGRjhlnuZfetVcomXrp/ygYqTf+xgoDpAqW5iD6T2VulIoFcPo2M+YzeIrZ874kylGKmOx5g3mwxPaDMFCVSl/KM+et47Jqxogj94y/tZ5ufrbKsC7ojb81HpmXytQEXVeAw5/jASQPLGRfs9HnPRebYRLGzyc3jDWDbfAwafpWF8xUCT0AWkWFQpLcyYtrxW1RcMDbss+AhpEpt8JbI1TeaI2/d3W4+dUcznwFgLzh66q0ralkfCihr1T+U3GpgXPazJjPJF3O5qCLoIg0+g0DRHN84vq5MLH1LHZON+Lwi6RovCi73/tBUOf4oEeqfl9LMHi6RlxsedWSu/VbVdBkpHBdfiyvCXtaCeMbyRnGmsE2eGF5UCvTnR5+ILCuG6gbPs2ne8001UpTZSWtks+p4t/Gabw1Jh5CinE2MomNMsj00+XkzwBo54adop+kQiih1/lP+T5l52kzYz79gSUi3dNIn1b1GW/0OzL+jk6XC0370+hLqPG3zSknwy+Sol3P2tR3z7tL4SkE4Pl9zTCmvxSGAFDEOuzPB5q0dSR9NoWxvCa22TtXQmvUZ0omY/1gG7xQmm4b+/TwN8waA+ju974xZfszRWWykYGjLU0U+GFBb2ar80vyTGIVn9Ebf9cxgZfhVnZN6RsT/ynFIdI/TJyrzYz59Ca8vvG3zUGrB/W8UfgD+mqHucbfo7GbQF/q5qCmadRruaHHD0ho1zN/ztFUMpz4fSkCU5cKum+MSJdTGX/3vk9m/D2X5Wc+g8RPOzGMNYNt8MLyoJoSpMbV4ht/a81CfS85nWcQYvxtm1+SkXllRFnumHGGmsRqAoURVW/8W9k1GdH6Evqmadp0OdF7Y3ievfH3idvMOI5RQMLSEy8Cv+KXxvMmzxw456cj1dHpMX7wHKBLl1u+L5UZtpbbnHIqfG9J/V7JRV5WhPYFaVQuXAJs36gy/g6UjapCABGfYYyxp5UwhWckZxhrBtvgTSqX2S1BcvipQZqABsDYS05T/Qwz/rYbVpoijzP+njPO/LUnyvjbTxFR+CN0r+mNUlewgRuCY7XK5wUgaDOVF4TNjVdNGfR5O0ysnzKfGn+L1VGzLEodETIn4H4V+CoYzZxuxOGr4LTrWWj8rfn77m8BusuHbbZhuG/U+KeFFwkAgqLY1uGjsVU0YcrcWbqccRFgG7yucplvsmmbnrQYq03kgAZA4D+j2NiHZpZb9hnQUOaIkonzFZgxzorcTRUAzGd06Ua+gk3TTwBaDwutUWoK+BtPfVqE71Om8IKwufHK8atiWlrEZVB4c+VOOce069fwmth+EFYqTGU9833ezPj79PgqI01qZvf7UWBbESTp/rauG2gUHpmXSlj5TV+AI9ybyuuw76+2houwlLCnlTB+1NUw1gy2wQtNDCvzwUiObtGu6qafi6RyzKN0rqqBzAFkMbdFG/YZ0JBncUG8sAqLZpwVmRuVSfbfh3yNV6hiX8llu4dDz3qNv4cgX1xahPSaOW1mHEeeZd7ztqDeJeBbTmiCuwDt2BuNu8hD5TB2h0pjKcxrfVDfUzLlNqecDF/JpE2Xyz3VnCaw7avktu6rVSJpqeLz9i4SNBdbXfps5c0h0r7GGGMzTML4+eSGsWawDR5WHnSrC2aq+LdzvQRcNKccVz/TVhmLqUy2ZcpA8i0F8eaMs5FJrPLW25f+azyExsEa3QEwBfwKVGrvDS+Aca42M46j9BUsiahPjONAK35J85ivfpqhaMPTv5fvS72HTe2l+dqB+WT4RVK069nY+1CxV/KLZazoYuYczElLLb0AslbZWGbjIiim/otj+ZnPICm9m2HDWDPYBm9IlzFz21TxPQC0KU3hgVrayPaVyUYBClvIKfzUK1UQb8Y4801ih3ZXBBd9JYhS+r+r9AfAFChHG31dutwogHGmNjOOI1Ti2Ry0frCDvxz8diPlQuy489W/AOmsZ6V3KNfO6YYe30dPu575lWI1lQzRYhkbbUN/36gtwFHk2WhsAygvTUeqxG0+77nY00qYIjfjb+My6Cb2kfF3UB7UTPXSw6+usVNuaspwUVYeqPd1s3mfAQ2+iXSlCeLNGGe+SawmJa97zdhDSGn87QVr1hBI8Usn7/p5TX42owDGGdrMOI4y86piWnriReAf/PvAiuLSYxehjpi83vM+qhJaz9A5PYHg16XQtXEVsZ4V+aD+bddybbGMIV0uhQDmEhRIUE9XgCOsmqsIBI6U+dt83nNZfuYzSFojVUuXM9bPIB/2jL89E8OmaQ4bOpuSUqJbtHeeybFcwcP3GVAYQHtpU1v3GdBQZmPjb30QTz/OhnYffIc06ie/6p3c7sjN+goCKf5GX+vzUnpruS4twubGq2Zk+GzpiRfBaG1BKtyirwn7QWSAKNVgzmhOVxRzMOLon281+FfKAU3fq08OYPjrwrBmbrMNh+IYemWeXzlyp1Q2lnlcBUBjjD2thPHzyQ1jzQzGvsOUM5L+dgfZFRwyt8TIN6fSGn/7PgMKCbOnDLGqNzK5l9ajMrucMc78AFB34JKNv70KanUjmsqObiLXWF2umltFSFYlzWkz4zh84++9pSdeBKNKkMqATx5ZJRN7PUCXopfOejYyplYWczD0jJ+vbl04yvhbqTC+VIZn4V1OKtbV6uA1qfZT9NduM/6OxlbRhPHzyQ1jzWAbvLEi4BDASGAzZgz4h92dVgI+8hmQF+Xh0O71g43ezmkYK4YUZpczxlmeYwoAeQO88xVsCjPs7jO0iqAU8Kvi7ZHgOcYogHGmNjOOY2zCK1dHNNLHP/h3c5PUrmUWBstnGn/XfoB++XnND35p53RDTz5SiukVrkNfm2n8vdE2xPaNGuUYQOdppbw8y7I+iLWrzasvluVnPoPEX6wMY81gSoWhPKhnypnAZswY6A67VR1h/J0Hxt/KA/WurjfvM6DBVwztFN4xc8bZXOPvKqIE/FqNv8cpODp1hB/AOFebGccxMny2w8RFMDL+jjD6HfrBDOPvbFgzU1Iydd+hipjTDT1+kRS18benktEomUrv0q+3oNhoG/r7Rm0BDt+Uv1MUOye3UTdezPg7HntaCeNvZg1jzVTEBq9TP2z9ViZVfA8ALOURY6QI0NzOZdiNlC1NFEU+KIYq5WG4MzVWm10iBqOylP8I42+lP0IKjG5Qtcbf2VDEQ1u5LLbNjOMoRt4bZvx9CYxVRfMuSWLTkcpRYCud9cxXwWw91eocjNOoI4y/66FIiubvu8/o1WgbbcNR6qB6HR57TWrW1JEyvzLj71iWn/kMElMyGZcClaPelWPfbfxWJlX8Kld7pcrIP1BrbufybNicpWSUmiplNq5UphkznZmtdpyhJrGK14w8baJ8n9Zn/B2VFuFXtVEEXgHi28w4jjJUFdgctHqwohKaAgYxwfIQ5xzkh4NpSuvZWAWjm9MNPaMiKdr0LV89qQhgFGjQdJtt6O8b9evw2JRfow424+/jsKeVMP5tsmGsGaqyS1eOfb/xW5lUGd3OKY2//QO15nbOOde+xpPxp5BekCq+eaXG+Lt7Tcw4G5mYHtpS+pi2qqDvaaNVMq3r0NM976qOS4sYzNp16ojYNjOOIz8oxwC6IGn6fdHg8T1YKuXBvxt3AF26XPy4yw/VJFNKl/PPE1o/GkPPkCY5XJZJ61k+uZCTq6MBdGtmOgHMJXDOHYrMxBitj/ezKiVTUAHQxkwctoomjG/UZxhrhtrUlHmrytj6rUyqFP7tnDIQkGdB9TPFQt69pt842QGPJCyVrHm+seNs5GVySMkTvQsiD2fYTeQaNnDds9tVnreJWEGxrSLUNI3KrwrA5sarplUVeFUxN3p4uyRyzx9JbfydB0rRGWtRmbX+dCkVNBjW8rSCX5eC7zFYqftaoHCNUH33gZUVrJnnogsAadNSR+mzWkVx5imZ6sYueyJZfuYzSCxdzrgUdsSC2JW7N8PnNCm9dANtCehW0TKUiVVJkg8LuVW9kQm9NTQHmNhxNt6M6Q5apWdIrqugNgQwscIAqdI/m1Eqm/55nqvNjOMYm/DGGz4b6TEKrFS1LlieDxW/dtU8JVO3Bqa0npUjdaoFrk/NKBVLXSTFU8nErJl+KmYCAcylGPaNyr1pYM6uCRh1adStetzGTCz2tBLGjL+NS6E63NKEG7wiSAlZg5JhS+T9gdrbOEVUJtPKi7sD3tAPbGmi6M0rD89LZSJ9MLPVjrPc9zKJUKP1wUVFYKpAAphrSFHK/f7dp7LJaREAwwHvHG1mHMfI+HtG6XojPUaBFWWbtsqFocrgnH7QKVQ6ZW4K69nI+1BZzMHQM6wLEZUMPZXMTlHJsPf7GnklbrcN+32j1vg7CDrn6n2N+YXOZfmZzyDxo9aGsWaoW5quHLulSaXJyANAWZFmVJlMnfeejVQbtpDT+J5A2pK6rRGtfpyVQSBFcxgpgw2f3E8Q4+8VtPtghq9X942M1Gudkim2zYzj6Pw9AECdhmqkje8pqDX67RSEQ2pr/LjrjL+1XjFXAVppz4JMJ2NUJEXr1edZC1TKdbbo+5YVhOj2jdq01CIIOmuNv0e+TzZmothu71wB/YCoTclkrJsdkTbVlWO3W5k0wYy/Yw/UmrSTcmJybEsTxaiMca1TGcWOs5FJbIR3gX/DKlewW6fxd5Y5yNy80slxgVebG68SvyrmXhmQMNImrPilNvr11REzDpVFlo0rcybQl/w06zV54K2FUUBTWcmwvWQ9pOgqA5p90MMChV7ATZ+eCNBdECnng+4zLD1xFva0EsYvV2kYa2Zf49LUMstag0y7lUmSPjUr0vgbAA7Gp8pKWofqW1plyJaZHJw0KqPIcTaWleu8C4bDmc5DCEsvWMuhp8jbQ2RnFC2lw/iBwVZ9dvo2M46jzFsFS103UDc2B10Cg/F3rQ6WTw7xM8ZdmeB6NkqzVhZzMPT4RVK6SoYq4+++2IC+UuzI+DuBvrUUc42/uyIzKm/EQxq1meXPw3YtCeMbyRnGmqHKsw75zlamO0X6MrGjtB2dCqarYqNTMtlCrsX3Mmr9kjQbpbhxFnqZaA9n/u291E/89AKqMECqdNWjdtr/r6Hx9xnazDiOLkiqVSEY6TPxr1OnIx1nuN8pVLRKx6ugzIM53RQZJ6Vbz6qqVq9nRZZB00BfRVRXdbRVuJrxt2f8rbRZ8LODdmpvRHd00HnL2NNKGN+ozzDWTFfZJaQ1yLQJPGW626Kdso16P5+DxDguRcE2ThK+wrW7kZZfk0WNs9DLJEZWvotUS3XG31hhgFQpArWDmBYRekGcoc2M4+hMeC0N5XKYBFZUwd1slKY6px+0BU30XjFXwbBu6Od0Q89oXVauZ6M2UVoLtGm9ZvwNMC0YIxqtez7HUTYAfgU7WxeiWH7mM0j8tAjDWDMVYaQaGj6vJV1mSwyHr3nG39qKPr5qw/oBjR+w0Bp/d2a22nE2SpPUGpKGAWOt8XfdHQDX0+b9IVJ7gxpUtTlHmxnH0Rl/WzW/y2ESWIlQMvUpvDOCu+HFTAp9KVR12WH5tBSeMle7nvlefdo06smF3IYvHwYfyEglU9Te1PVtqvkMY8x2e+cK8NMiDGPN7IhKDoOpnt3KpEpohBpt/K2q4BEYf2944yThm1fGqozijb9rdWpFt+Graq3izVMZEIUBUqVPrdLeoAab23O0mXEcnTrN0uUuh7Dil1pVW3upbjONv/e170+3fF+azOkJfKdLYlYlw0kl3vhqZ1tOo+5T2bQFOLygM5Vdgb1m7/kvpqBKXBP2tBLGl/YZxprZVzV6a9AtmDaBp0u/YZ5xoNZv7A83UkplyJYJS1HHlOHVjrP+hvVwUNJuxlqfIp2XSV+lrTflXU+bD9WjIpVMVUzg1ebGq6TMXW/MDmBz0CUwSslVGv2WBz+lzrx5rvF3an1pNKdX9aaDE+dgov5V+v0AADy8r9p/a5V2EVVcL5k8i7NZKOeowPMMKk+VaJc9cWy3d66A7vajW6gMY61QtzRdpQybwNOl2zDv6xoy1wYHOPyqYbtK5zPQbRa0ypAt4ytcqaqNk9fkceNs5GUSccPamb3735P9XlnrO9RK19fT5mWg7tNUEQKAQ9CoOUubGcdRZBnUDcAj+0OQdEX90cDxAytao98wTXWW8XeWJbeezfGnMvSMi6To1rNujXxo1waZNOtCfkjf6i5zUkjFXIqu8ps2LXWcPqs0/s5CGwAbNzHY00qYwjusGcaaodQQfZqU5Zcni+8vobsJjvcZ6GTPlSmZRCYHpyhfgUhT7kOgUOtdsI/c/HaBFK10PRW6281Y4++q29yeoc2M4+j6/KcOBz6bg9aPnyJWaY2/T3CoLA6l6bVKx6sgzxw4F+dPZcRR5O6wLijVv4c26IJMaoWrVeIFgGnlN3kdDoy/lUomv9qkJhBoDNiuJWG6AWPpcsbaoYINoeGzbXzSoysTq68U55WJVW/ss5FqI4VNeaoUI5VRRIWUiHE28jKJqIY23vBpVQPrNv52Tn+DujtjmxnHER74LKi3fvziAmrj70O7dylMc+alodJmWutZmQ0VK1P5TpdE/3y1xt99YFsfjCz6wIqlUfeV9tQFOIa9aYxHm19tMgVV4pqwp5UwfdqJGX8bK2dHHKyKyJsI4+rxjb9VqqRsfKBWSZJ7429LVZEYVSqL2MxG3fiNvEyUCrYsVIIoU+wijFJTofNL2kV4YgFAf/A8R5sZx9H114f+//buPWi2rKzv+O/Z3X3myswBGTSAA4NiFZcgFBPIxRRJBTKEpBhNsCAiCpJQQqAqlUAVxIgIlZCImrISjIwFokmFqyY1Ggx/BEhK7oNguMjoADIyI3eYCzPnnPd9+8kfvXf37ve6nrXf3b129/dTNTXnvG9f9tl7r8te61nPupCeHwVlOzhYnhK5UA82NvdBziDTfMC9rPZsHjmaWKcjpkkSHU383dxrSW1mtbxEbNsjmZrl9lJKvtBFlP3ONC2VwyQ4kIVlvdYyZvZUM7vZzG4xs5cf8vt/YWafMbP/Z2b/28we0vrdnpl9ov7vxj6Ps1SLhwkimTBsu0esf26SbDYV+DavLy/VqFok5U7rpM9ec2F3qqknLpuaf0e9tp6G/EhLu7IdsWvjgfcEy9motSQvdaCwuWbn5jltEpOF13mKhtR5GzWRTInRfc35bJbQ9HHN0M1k3/3L+R6+pcHyxGW/Tdlc3Af5ib+bAeJS2rPRPHJ0WMuTh2IUjBSbT8xEEn+32szZEsjtvY6jqjkXaQNA88meevImpT4YVbacq49yEzLu64PNbCTp9ZKeIulLkj5qZje6+2daL/u4pGvd/R4ze6GkX5D0zPp397r7Y/s6viFYPEwQyYRh25m6Lj2ks0bi7/ItoozSOk6jfREtSYm/W3mfpHJmfkvU3qms6Wie+J5gOVss1a7DypMimeKzspuQ+Dvy4LoTGnilblyl8b77l8ix4ctK/L0/kiljsLHZkXWePLyQuq2JHE3dzAExk1Ew8Xd9Dc7V91rqoMc9F2b31rYPhI/rzUbCib/3Aom/9y9pLKQsD0WfZ+sJkm5x98+7+wVJb5V0ffsF7v5ed7+n/uuHJD24x+MZnEWBIJIJw3bUlrmTer3zHsltizWuZ4uiib+bmeDUJXZNjiGJkOTjjA8M4qXnFUgtZ02S2L36QSl1N7T2cSUn/p5HyQ3nmo/D0X37orx6uGboZn+5og4aPjOro3em6Yl+M5b9HvYZO3tT7RWWY3DRljNA0Yf5EufE9uxA4u/EJeZNTqFtX9K76DfGE3/vBRJ/S7F+DRb67LU8SNKft/7+pfpnR3m+pN9v/f1iM7vJzD5kZj/cxwGWrh3aBwzZUVEw46qaD2DM/k4FXppmm9i9aWLi72p5di4pomW+gwf3wUkWnZ70wYecctYkMd2bJu6Gtq8zlpz4u4mSG9A1H7ej+xJ3T5SCZYK6caUO7C7HjPVGWKpjAhtXdBlsbJY0ldaezSKZ6jq9kIGvTbLYJCWtXchK/N2akNv2iYem35i+AUezy2ugPsiIzMdCb8vlIszsxyVdK+lJrR8/xN1vM7OHSXqPmX3S3T93yHtfIOkFknT11Vev5HhXZZGkjOVyGLbd6dGJv3dauQuowMszGZnO7dQ7xQUeqO8NDDa0I9rGW55n4CQHdsFK7MxGy9k8SWzi0ooDib8T8w41ib+H1GGejBYDQEnnslmK1fM1Q76cQVKUrxlYSU78vf+hMmOwcdx8515Z7dk8MfXUdRmDqKduPrgY2EFUarcLgTaTgcJ5vzF1A455JNN0mp74+xQGnbdZn7XMbZK+t/X3B9c/W2JmT5b0M5Ke7u7nm5+7+231/z8v6X2SHnfYl7j7De5+rbtfe9VVV53e0RdgkSyUSCYM2+4R0RDzWZnEmQis3nyb2L1Y4u/IYMPiO4aVAHod9nd6kkPsg+VsXDV5shJD/5udcnZiS8KaJKZD6jA3y2F2U6P7RvHZ0JxrhnyT+eAtyxM3yXxJbj3gc5LJvnospz2azJfoldWetaNghlTfDkWTY3A3sT2bHGgXAm1mYvqCTdZE+6ZG2bd3eY1Gm837NVt+zqP6PFsflfRwM7vGzM5IepakpV3izOxxkt6g2QDTV1s/v6+ZXVT/+f6S/oakdsLwrWBm8+SfwJDtHpHYd1yvL0+dicDqtRN/pw5OSK3ZucRk4bt76TNS22x/YvXUZKHRcjZPEpua3HpfBFvacVWzmcjERKmlmMy3KE89N8sz1n1dM+QbZUQIonyzAeFZHZMycLj/PsgZjBlVVd2epe0kuSrtxN8MWp++SWtAM6U9y23LZ/czebWaXV53Eu/n5iVNcva0CO19bTeDsyG9LZdz910ze7Gkd0saSXqTu3/azF4t6SZ3v1HS6yRdLukddTjpre7+dEmPkPQGM5tqNhD27/btSrc1mkRywJAdFQ0xqWweVr7tDWap5g+7iY3yeBR/WJu0EmbSiB8vZ/Yzp5wtBv7SEmTm7Co4qWwRJTeg8t/sirk7TXuIzHlwpW5cLZZFbKZZtEMkkmn/g398kGjSWupaUns2aqJTj0hfgG6aNvOoyP39sqKSq1bi74LurXVY7PKaFmVvZrP0D4FNaQ70a2iLQ3rNyeTu75L0rn0/e2Xrz08+4n0fkPSX+zy2oWgSyQFDdlTYeNPROb+bFu6K1RvXy3b2pmkDAc2AxPlIYupWRNu2h4CfZLwvKiayQ0qknE3q7YFTl7Lt3wI+dRlZJClvKZYSfycNGO0rEz1dM+Tbv3U99dBmiNYxB+qxjIfKZklT6o52q7LYxIOdyfowbkWKRZZRR3P1NQMr2z5Q2PQbIxuHjKsq3EeR2v2a7T7nUZytwjUVCjBku0dEQ7RzlZDYtkyTeW6exAfqfY1y0oxeK2pj22fnTjJfjnghsCNNRjmbPyilJv7O2J2rGcga2sx6M5scju7L6NxSN67GfFdMEn9vlFkd48mJfnPypx34jKqat5kl3UeTKlanI6aJyE5tzxYDmukTcovNMoa1I2sfmn5j6gYc0qx8R5L6E+HaDbVM4cZ17gdgyI5KhDjfyWU3LZQdqzffJjaY+HuRk4mojdM07/TsxkLsZ+9JL2dNktid1Ai20eI7ZseVmFy8jjIY0sz6Ijoi9dzsXy7XzzVDvvkg6S6JvzdJk9c0Ndphsq8ey0r8Xb+ntPasqbdSN3NATHsDk7Tlck2dE0j8XU/6pfbHNtmi37gXitCOlO32M4pE4u8ozlbhmpwVwJAdNXPWDk3f9gazVM02sakJoHMSf7cfwmnEj9dcgnOB5RzznECBcjYZVTq/O5V7Wl6SSes72t95nGYSZS8xIqgU7cTfkeVykWUROdcM+Ub77t8hDXriaLMlTJ6c6PfgfRAve6N2v6ag9qyJHE1tyxEzzy25lxbJNM/3E1ii27SZpe1cuA6LfEnpEzGjykJL3+abdgT6NVgop/bDoUYk/sYG2Jke3gi0l9hse4NZqvbsXFIn6ECixNh7uA+Ot0hemT6Il1PO2mHlaQOF8SSmTZTBTmJS3lLMoyMSc65UlamyWMg9deNqHVwWQfd4E4wr0/ndwNLt/ct+O0QylVZ2F5GjZeWK2hTjelfB3Wk0ai7QzgYj8zbZImJwL/l+nlQWSuJ9cPnsdp/zKGqZwpH4G0O3N3W5Hz5L034w3fYGs1TtRJMpDWwz07NI/B1LFk7n92RLySsDSaQj5WxcVaGO1f7OWFoS02HOyo7rPC+R4x6PqqzE39SNq3Hg/h3Q/YijjUcW24Si2j/YmJH4uz1pUlB7Nk9MnbiZA2LGI5tHikWWYkWSzI9HpqlLF996lxcAACAASURBVBgoXFoJEWmHIxMJ7fpgVJnMKDcR232HDgCJvzF0u9OjBxvaW3tve4NZqnFVd5wSk4U2kTaRpUHzZOGFzfyWajwy3RsaxIuXs3FlsR3s9i2TTAkrnyXubCKZhlP+FxFY6bshLp3Pnq4Z8o0zljSifJOqCu/eJcV27zz4GYt7qaT2bJHPZ1jLk4dinscwsT1b3CdNHsNYIuqS7q11WO43pif+zq0PaBPiqGUKN653sQGGqhkkPWzmjMGF8s0Sf/uRSx4PfU+7Yx9MFk5DfrJxZTofWZaWUc6WOmORrb/ra5gy49feXnxIM+vz2eTdveTjbg8y9XXNkO9ALjkG9jZCl3rMLC8Hy/IAcTllt9nEI9KWI91ik5S09qy5BuczJx+2PVdfu9+YmkOv22QPZSZqu+/QAZiMZls0AkPVDDIdulyuCUW9kD4TgdWatGc/UxvykcVCwFth49wHJ1sK+U46v/FyNhlVoYSk8zwkwdD1oSb+luqEo4HzGSsT1I2rNG7dvxK5NzbFuF3uAoO75zok7Z4vdS2s7E4q04V6MwcGUU/fYpOUtHbh4BLzwE68gWTXm2oxSJe+i+O4ivZryizLQ8EZK9x4VGmPxN8YsJ1jlsu1t41mN58yLbaJTX+gHreTKwY7WzzcnWwSXMqWU87GlcW2+m1t/Z36cDZbvjG8mfXlnCsZidR7umbI1966ntwbm2OpLUpawru4D3IjCJfKbkHt2XgUq9MR02ySspM4IZez6yibpCy0+43Jib9bZSCUa7KwsjwUDDIVblznrACGKimSiQazWDkDQMvJFQNRG+SfSTIeVVlLQGK7ywUTf2d0fsdVNciZ9a6J1Pu6ZsiXM3CI8rUHmaLRJbn3QantWbROR8x4NIsUk9Las2bX0VBbvpQjqJx7ax2WluivIPH3tp/vHJyxwo1Hs6UqwFA1g6SHJv5uNZjbvr68VO3Ztsg2sfP8M4GojXuJZEoyi4qJz8ZFytly8vb0sPJIEs6l7xjQdc9JONr+t/Z1zZBvKb8H53tjTFoD8qsqd+NC27NoBCxiJsHJn9nrqlZbntDOZgysbKpJxoDbUm5ENqXpHbVM4ZrwS2ComuWehyb+rhsG97wEm+hfc13cYx2nZmx8lNKxr++DKfdBkvYMe9rOb/FyNqoquR/8vqNf37pPAkvImu8o6WHsJM2/derpxz2qbF4m+rpmyDdu3b+c783RJOmX0q7raZS7Uu+lpTp9QPXtULTr+MiGEIf9+cjXjxb3Vm7OsE3RLqvpUfaLPkdK2czp12Bhu+/QASDxN4au2R3x0OVyo9jDMlZvMop1gqTl65oyW7R0HxBFcKJ2WYmE2O9/73Ha1y3lmizdJ4HlctHjKsEk47jbs9R9XTPkW6qzeADfGNE6Zvk+6Jb4e/+f121CndKrSXDyZ/a6xXuSBkGD7cgmy+k3Lved0iO0I9+BBc5Y4cZVNX9IB4aoGSQ9NJKJjn3xcjrMkw4NOUmOTzYJPgjllLNxcNBovHQN05eQRY+rBDkDEtGH16Gem6HKGThE+aLlaBIcwD/8M2KTLKsSrdMRs9QGJp7fpi2YjNI2G1i6t7Z80KNdT6eWs2h9EI00w7LtvkMHYDyyeeJkYIia+3d0TOJviVmCUo0zOsyjKtaZHQWjZrZd+xxFloDsf2/qd6QMGo2D11xarhMOqx9KFT3/0vI16OuaIV+ThFfiAXyTRNuW5QjCzOVyhUY/LD2Uc4+fuva9k9qeNe9JX8Yei3zaZMvnu5/o6eWBw3LK8lBwxgo3GVUk/sag7UyPSfxd6IwfFtoNa0p+pdl72teVqI3TtlxuTj6/OeVsEuzMLt0nGZFMQ3qwjy4llOJlgrpx9ZqHDh4mNkc0SnZ8CpEiS2W3oHptkjEIgnQ5UXDzOqenZdebbLn/kHb+RtHIxoyBLCxQyxRuXBmJvzFo88TfhzSiNJjlGwcfjmfvCeafYalKSHRwJqecRUP/R5WpifbPSXpa0sPYScYZyzujebSoG1evOc8si9gc4+CAsJnNHyazl8sVmvsop95Cui7LqNPb5Xh/bFPlROZFJ4iI/utmu+/QARiPTDtEMmHAdvaOjmQi4XP5og/Hs/fElhyQKyKmU+Lv5OVyGdE69XGlJ4gf5uBi9/PZzzVDN+P54ALne1PkJEqe3weZdVJOm7kK0QE3xOS0Z9E6p9R7ax1yNsdYvkbRnEyUmSjOWOHGVUUkEwZt97jE31Wswsfq5XRMmyiMWXRLMGqD++BE4eWIGeUsJ+HlKNhhHuoyyZwIrOXcZv1cM3TT1EOc782Rsztq1/sg5ztXISdvHtLltAvzxN9Ziau3+xF+Eoy23v+6lDJArr5utvsOHQASf2Pomt0RD8sBMMqYicBqLYdnBwcbshKFcx+cpCkrZrNO0Imvz5nxWwoTj0XrpHb4Bpv4O5jEW1rc131eM3QTXbqC8o0yIhG63gfFJv7O2AEU6dr9nXDi7+Q2kxxBjZxzkZOaoSk32z6ol4MzVrjJqJonTgaGaKceJD1swGGSMfOD1VpO6BybLUptlNsdXnJFnKzpkCbnyMqaYY134BYRbBue+DtnKeEKrhm6iSbhRflyoh2aspf7UJmzI+sqkMS4Xzl59EbBOoeo74XlROuJib9z+kL1eygzcbSkhRtXNk+cDAzRYrncweomJ18CVitnt53mYS0nmWVJM7+lajo9OUm8+0r8LbUezjJmFYf0YN8l8Xef1wzdEMm0eXKWe8/LauZDZantGUmM+5WTlDvelrNcrpETZZ8zQbSIZKLMRG33HToA41GlnT2XOwNNGKZmudyJib8H9JC5TfIGKGKJU0n8HdNck/QE2znL5fJzceUNLg7numc9uAZ3LqNuXD0Sf2+eaKLf2Xu63QelJmcudfBrU2RtkhLtKxV6b61D3gYc8fpgErxGWOCMFa4ZnSWaCUM1j2Q6pIKeMLNWvJztmCfBmZ/JQCNa1qXTcsRgQtLZ+/t5OBtq4u+shKPNsogerxm6iSbhRflyEiVH269jv7Og9oylVv3Kac/CfaVC7611yOrXrCAyHwvbfYcOQJN7Y5dBJgzUPPH3IRU0CZ/Lt5woMZacMjkvwUAjWtZlFMwRkFPO2tchPalmLFpnsIm/c87NqP9rhm6i9RbKt1yOgu1XZp2U852rwO5y/WrX06ntWZe2fNtzBI0y2uHmupjF224GZuPouRSuGand2SP5N4ZpZx7JdLCCLnWrXyzkLGmKhhezXXvMOBoVk1HOuuUyieWjiBxXCZbOTXD5YZ/XDN0slqHSNd4UOW1LdLD8wHcWGjHEEtx+jTPas0WUTEbi7y0fKMxJ/J0zYMQy6nycscI1BaJZcgQMze5ek5PpYHVjZuHt7rFanRJ/5+SfoSE/0XwQL7GTmVPOcpLENseT/vr2dwznumclRQ/mq6JuXL1oEl6UL6dtieaWO/CdhUYhkvi7Xzl9pUm0zcz4jk3VJfF3ZJB1zDLqbNt9hw5Ac3PvTIlkwjA1Sz2P6rAxS1C2nFnZaG6enBnAbRZN/N1+bc5yufQE7tuR+HuS8RDJDGr5okl4Ub51JP4eFdqeMZnTry67uEbb2Nl7yrm31qFTtHWgv0E7nI8zVjgSf2PomkGmo5IULpaRbHeDWaouM8Gp19TMwsuJttkk41zFE4xm7MISzGUy1ITvS5FMyecmtsSx/VrqxtXgfG+enES/87Ka+RBvZuHNGVYhZzMHpMtJyj0J9n9zEtlvquV+Y+z85bTDQ5oIK8V236ED0HRmWS6HoVoslzsikok8GEXLSfydk9NiMXtMQ36SUcZsXDSZbXPtKpOqnhJkDjWSKW/r5Pj93TUBMWJI/L15mvIZSfQ7OoXIhRLvpXa9vO1Jo/uw1FfqKcl8zndssmj06SiYymH5OzjfUfRcCteMupL4G0PVJP4+qoKO5pfBak0yBgJywoujSaO32WKnsshsXKycLWbvcmb8NnuZZHuWOvVhrTk3fV4zdEMdtHnmkQ6hHCzdy92kwHtpOWl0Oce1KbI22GjahYycTAwULspXdKe4UCRTgWV5KDhjhWsKzi7L5TBQu9OpRpXJ7KicTFTgJWs33smJv3NyBjFblGwe8h3K7xMrZ6Mq/h3z92Qk/h7SQ89ymYjNWPd5zdDNhDpo4ywG5NOv6SSj/Trqe0u6l3LqLaRbysUVTfydeJ9Ulal56bYvl5Pag0ax5XKR+iDar8ECd2jhms4lkUwYqt2pH9vRYplU2VaR+Hv22ax7T5WVvDJYzjpdw+CSvMhxlWB5xjr2b+3zmqGbcUb0HsrWKdFvh8Hd6CYIq5BTbyHdcqRYamRSvM7JmcTbVNHz163vRJmJ4owVrmkUyMmEodrd82NnXEj8XbblJU39JU4tMVFqqXJCvuMJMuMd2dwledKwEn+bWXh2c5FUur9rhm4435snp12JRkcc+hkFbmSRk18R6drtXmp7lnOv5Wz8samikWA5y2dPI7JxW3GHFm6e+HtKJBOGaXdveuxDJ4m/yzbKiDbJicAocXlBqRYRQ5HzmxdlFLuGseMaauJvKZ53LOf+pm5crdOIYEFZspZuZ0SXHPze8tqz5pgimzkgXU5S7pw6p8QouXWJRhnlnLucjWwwQ0tauGZ0lkgmDNXOScvlCtyFBQvt7ZhXkvib2bkT5ST+jpazScYAxyQ68DLQxN9SfHazmTnt85qhGx7eNk9ORNFpDBBNCmzPcjZzQLpOib9Dg6AMejRyE3/3uaEJFjhjhVtEMjHIhGHa3Zse+6Cas4wEqzWeL3+LzRbFkhyzVCXVJCfEPljOcsL4o+9pd7CP2higVLk5rvq8Zugmp1yhbDlRtfPl3qcQyVTSvbRYKlTOMW2SnPZsktHvWQyC0i6E0wBkpHIosSwPBXdo4Zqbm8TfGKrdPT9huRyzMqULP1DnRDKxNChZXrLQWDnrlpA0uCRvgGU/nOQ8YzaUunG12M1v8+QtLe4eQVjivURi+37ltGdZ7QKbpMyF+xw5qRwKLMtDwRkrXFNZsVwOQ7U7TUv8TQVerug1WkXS6G2WMyMdvYZZHebge4YcqROdQV3FNUM37Oa3eSYZUR+nMfhd4r005EH9IchampnRLrBJykI0+jQrBxbL1rNxhxauKQgk/sZQ7U6nJ+RkYlamdNFcJXmzRXSAU+Ulr1zFNYx9x2xZwTDLfjjhaE4iderGlWJQb/Nk1ZWnEFVbYntW4sDXJhnl1PEZOYJyktlvqvgGHDn1Ae1CLs5Y4ZrR2R0imTBQO3t+bAPAeufyTYId5pzw4vGAo1pWLa9jGsxdkJP4OyepZlUNsvMWTvzdYbkcdeNqkPh780Q3I5i99hSWyxXYnk1Y9tOrZpOUvneKy0lHsKnCO9pmTCRMaBeycYcWbpH4m0gmDNMs8ffRlXPOwyxWazyqNA4ks8xJ4k1IcrpJRp6RaDnLuobNoEjguEaVDXJGNhqpkJNfibpxtXKS8KJsTXRJaDnSKST+LnH5d1WZKuP+7tO4qmJ1fFYep/Ki5NYld9fcnL4p5SaOnkvhyMmEodudHp/4OyfEGKs1Hllsi91OIeA0SycZZczGRctZ3tbKGe8Z2SDLfvT85ET3UTeu1ohlERsnJ1JklPHgf/AzymzPxqMqVD8jJtqe5bTlLKNeiPYbF21qvG86KqwsDwFnrHCTeSQTg0wYpt09n88MHoYkhuWbVFXo+sxnizJm9JgtOlnOVuvRcpZTLvPeE7u3SjEZzZZGJG9VvYJrhm4WESzUQZtiVfXYgc8otD2bVEZ90qNoe5YTWTO/Pxn0CJezxbnLOd9lleUh4A4tXDNSvbvHcjkM0+50euyMS84Ww1it6OxcTs6AnOinbdUp8Xc48iYjYW7wXhli2Z8dd9+5N6gbV4kEr5tnVfXYwc8osz0bjyoiYHoUbc+yor6rSpXNlj9uu/AGHJ02TSmrLA8BZ6xwTWVF4m8MVWribzr25cp9oI4tm+KBOlXWNrzBcpY1UJiZVHOInbdxFXtYW8U1QzfzeouH8I3RlLvIUpfFe7on/i6tPYu25YgZVxYeMGrel/yeUew7Nlm0nE1y6oMC86sNBXdp4Uj8jaHbnZ6Q+Jv15cUbj6reEyWWuOVzqaI7m0nxcpaTJDbnug828fcoc8a6x2uGbuZLL3gI3xg59dhiCVP35XKl1W3jepkv+tFskpIqJ0H8UKN/+xDfgKND4m/ahbBez5iZPdXMbjazW8zs5Yf8/iIze1v9+w+b2UNbv3tF/fObzey6Po+zZEQyYeh29/zYBmDELEHxxlVm4u+MZJYkJT1ZTvLKnHI2SxLbdwTbcBN/Z81Y93zNkI9E65spWlZPM/F3ae3ZuCLxd59Wkvg7OJC1yaL9xtzzHfkOLPQ2yGRmI0mvl/T3JD1S0j82s0fue9nzJX3L3b9f0n+Q9O/r9z5S0rMkPUrSUyX9av15W2ee+JtBJgzUzt702BlBZo/Lt6pklpFEytssK4l0RjmbVJaXIDNyr1TVIMt+/rnp95ohX069hfKtoh477DNKbM9mx0V90pdoe5bfV+IaSqvegKOssjwE4x4/+wmSbnH3z0uSmb1V0vWSPtN6zfWSXlX/+Z2S/pPN7pTrJb3V3c9L+oKZ3VJ/3gd7PN4ijSqTmfTlO8/ps1++c92HA4Tde2Hv+MTfI5IYli68NCgz8Te5ItLk5PvIKWfRJLE5WysPN5IpmHtjRdcM+ciBtZlWUY8d/M4y2zOiYPqVvUlKMOp7iG1mH6LlLKuPQuLvbH0OMj1I0p+3/v4lSU886jXuvmtmd0j6rvrnH9r33gf1d6hlu/yisd7ykVv1lo/cuu5DAbJcdtHRVc3ZSya68pLJCo8GUVdeMtHd53aTX3+fiyeqTKHrevaSM9wHiS4/M9a4Ml15aeT8xsvZlcH3NK+9z8WB47p0oisCry9F9H5d1TVDvrOXnJEkXcE53yhnL42Vo7OXxuuxA59RaHsWrdMRE23Pzl46q3NC7ULwft5k0XI2GZkuOzMK92sqmz2LI8bc+1mGZWbPkPRUd/8n9d+fI+mJ7v7i1ms+Vb/mS/XfP6fZQNSrJH3I3f9r/fM3Svp9d3/nId/zAkkvkKSrr7768V/84hd7+fes06dvv0O3fuOedR8GkO2JD/su3e+yM4f+7u7zu/rqnef0sKsuX/FRIdXX7z6vC7tTPfDsJcnv+cztd+oHvvvy5NmfO8/t6Jt3X9BD739Z7mFulc9++U497P6X68w47fzmlLMvfuM7Ontpeidub+r647+4U49+0JXJ3/HlO86pqqQH3Ofi5PeU4JvfuaB7Luzqwfe9NPk9q7hmyOfu+vTtsfsX5bv1G/foysCDeU49tl+p7dnt375XZ8aV7n/5Res+lI2U05596rY79KgHXpG85Ovb91zQXed29b33S297NlVOObvlq3frgWcv1qVn0gaNdvem+pOv3K1HPvCK3MPcOGb2MXe/9sTX9TjI9Nckvcrdr6v//gpJcvfXtl7z7vo1HzSzsaQvS7pK0svbr22/7rjvvPbaa/2mm27q458DAAAAAACwlVIHmfpcYPhRSQ83s2vM7Ixmibxv3PeaGyX9ZP3nZ0h6j89GvW6U9Kx697lrJD1c0kd6PFYAAAAAAAB00NsCwzrH0oslvVvSSNKb3P3TZvZqSTe5+42S3ijpv9SJvb+p2UCU6te9XbMk4buS/pm77/V1rAAAAAAAAOimt+Vy68ByOQAAAAAAgNNVwnI5AAAAAAAAbAkGmQAAAAAAANAZg0wAAAAAAADojEEmAAAAAAAAdMYgEwAAAAAAADpjkAkAAAAAAACdMcgEAAAAAACAzhhkAgAAAAAAQGcMMgEAAAAAAKAzBpkAAAAAAADQGYNMAAAAAAAA6IxBJgAAAAAAAHTGIBMAAAAAAAA6Y5AJAAAAAAAAnTHIBAAAAAAAgM4YZAIAAAAAAEBnDDIBAAAAAACgM3P3dR/DqTGzuyTdvO7jAABstftL+vq6DwIAsNVoiwCctoe4+1UnvWi8iiNZoZvd/dp1HwQAYHuZ2U20RQCAdaItArAuLJcDAAAAAABAZwwyAQAAAAAAoLNNG2S6Yd0HAADYerRFAIB1oy0CsBYblfgbAAAAAAAA67FpkUwAAAAAAABYAwaZAAAAAAAA0NngBpnM7H71/23dxwIA2E60RQCAdaMtAlCiwQwymdnlZvaLkn7HzP6Sk0wKALBitEUAgHWjLQJQsvG6DyCFmb1E0rMl3SHp65LOrfeIAADbhrYIALButEUASld8JJOZPU3SdZKe7e7XSXqspEfUvyM0FADQO9oiAMC60RYBGAIrMbrSzK5w9zuP+N0vSfq6u792xYcFANgitEUAgHWjLQIwNEVFMpnZ2Mx+RdJ7zOzxzc/q/zfH+h1JO/t+BgDAqaAtAgCsG20RgKEqrTJ6nqTHSXq3pJ+WJHffrX/XhFx9QdKP1b+brvoAAQAbj7YIALButEUABqmIQabWGuLfkvRcSW+WdNbM/lH9+6q1a8KNkr5iZo9a9XECADYXbREAYN1oiwAM3doGmczsovZfJcndz7v75yX9maTfk/RjZnaJu09bFe4DJF2sxQg+AABZaIsAAOtGWwRgk6xlkMnMXiPp1WZ2lXQwvNPddyS9V9K3VIeHalHh/rGkN7j7Z1Z3xACATUNbBABYN9oiAJtmpbvL1ZXnDZIuSHqFpFvdfdfMrpT0Kkm/5e4fr197RtLjJb1U0h/W77nB3e9Y2QEDADYObREAYN1oiwBsqvGKv+8+kuTuz5QkM7tc0t3ufoeZvc7db69/bu5+wczuK+lvSXq4pBdSkQIATgFtEQBg3WiLAGykXgeZzOxidz/X+tE1kv7MzCaSfl3SpWb2WUn/0d1vrytRd3c3swdI+teSfsbdf63P4wQAbC7aIgDAutEWAdgWveVkMrOflfRhM7u+9ePPS/oRzUJCP6tZZXm1pN+QpLoS/VEz+x53/6qkH6IiBQDkoi0CAKwbbRGAbdLLIJOZPVXSP5D03yU9u15HLHf/Qv2zF0l6h7v/iaSfkvT9ZnZNHSZ6qaSdevR+evg3AABwPNoiAMC60RYB2Da9DDK5+/+S9OOS3iDpLkkvbP365ZK+I+kR9d8fJ+nDkr7k7ne7+2+6+zd8lRnJAQAbh7YIALButEUAts2pDjKZ2fzz3P1P3f0vJP22pKeY2dX1z89L+peS/raZ/bakN0r6g3p7TgAAOqEtAgCsG20RgG1lXQbGzewnJH1R0ieO2uGg3p7zZfV3vWzf7/6upPe7+3eyDwIAsNVoiwAA60ZbBAAz4UEmMzNJ3yPpv0maSrpF0uWS/rm7f83Mfl7SR9z9f9avH0n6AUk/J+lTks5Ier27f+XU/hUAgK1CWwQAWDfaIgA4KLRczsxG9Zrg+0i6zd3/jmbJ6r4l6T/XL7uhVZGau+9JukjSEyU9X9InqUgBALloiwAA60ZbBACHS4pkqkfdXyNpJOldkq6Q9Ax3/8nW72+T9Ex3/z91pbtX/+6MpN+V9AF3//l+/hkAgE1HWwQAWDfaIgA43omRTGb2JEkfk3RfzUJAXyNpR7MEdU+QpLrifFX9n9x9z8yuM7PHuPsFSU+nIgUA5KItAgCsG20RAJwsZbncVNIvufsL3f3XNVs/fI2kV6oOBa13T/gfkr5mZg8xs4s1q3zvluY7JwAAkIu2CACwbrRFAHCClEGmj0l6ex36KUnvl3S1u79Z0sjMXuLuU0kPlrTn7l9093Pu/lZ3/3w/hw0A2DK0RQCAdaMtAoATnDjI5O73uPv5Zi2xpKdI+lr95+dJeoSZ/Z6kt0j6Q2m+0wIAAKeCtggAsG60RQBwsnHqC+sRe5f03ZJurH98l6R/JenRkr7g7rdJkqdkEwcAIIi2CACwbrRFAHC0lOVyjamkiaSvS3pMPUr/s5Km7v4HTUUKAECPaIsAAOtGWwQAR7DI4LqZ/VVJH6j/+w13f2NfBwYAwGFoiwAA60ZbBACHiw4yPVjScyT9MjsjAADWgbYIALButEUAcLjQIBMAAAAAAABwmEhOJgAAAAAAAOBQDDIBAAAAAACgMwaZAAAAAAAA0BmDTAAAAAAAAOiMQSYAAICamZ01sxfVf36gmb2zx+96rJk9ra/PBwAAWDUGmQAAABbOSnqRJLn77e7+jB6/67GSGGQCAAAbw9x93ccAAABQBDN7q6TrJd0s6U8lPcLdH21mz5X0w5Iuk/RwSb8o6Yyk50g6L+lp7v5NM/s+Sa+XdJWkeyT9U3f/rJn9qKSfk7Qn6Q5JT5Z0i6RLJN0m6bWSviDpVyRdLOleSc9z95sD3/0+SX8k6UmSxpJ+yt0/0s+ZAgAAOIhIJgAAgIWXS/qcuz9W0sv2/e7Rkv6hpL8i6d9IusfdHyfpg5J+on7NDZJe4u6Pl/RSSb9a//yVkq5z9x+U9HR3v1D/7G3u/lh3f5ukz0r6m/VnvlLSvw1+tyRdWh/7iyS9qdupAAAAiBmv+wAAAAAG4r3ufpeku8zsDkm/W//8k5IeY2aXS/rrkt5hZs17Lqr//35Jbzazt0v6nSM+/0pJv2lmD5fkkiap39163Vskyd3/r5ldYWZn3f3bmf9eAACAEAaZAAAA0pxv/Xna+vtUsz5VJenbdSTREnf/aTN7oqS/L+ljZvb4Qz7/NZoNJv2ImT1U0vsC3z3/qv1ffcy/BwAA4FSxXA4AAGDhLkn3yXmju98p6Qt1/iXZzA/Wf/4+d/+wu79S0tckfe8h33WlZvmZJOm5eYevZ9bf90OS7nD3OzI/BwAAIIxBJgAAgJq7f0PS+83sU5Jel/ERz5b0fDP7I0mf1iyJuCS9zsw+WX/uBzRL0P1eSY80s0+Y2TMl/YKk15rZx5UfbX6ufv+vSXp+5mcAAABkYXc5AACADVDvLvdSd79p3ccCAAC2E5FMAAAAAAAA6IxIJgAAAAAAVb06oAAAAERJREFUAHRGJBMAAAAAAAA6Y5AJAAAAAAAAnTHIBAAAAAAAgM4YZAIAAAAAAEBnDDIBAAAAAACgMwaZAAAAAAAA0Nn/Bzzga7x6YBYaAAAAAElFTkSuQmCC”/>

  1. Wrap Up

What was our initial goal? To find out what the minimum amount of moving parts required to host your own analytics solution is. We were able to condense this to three parts. Two of those I have made Open Source and available as Francis Beacon. I am very happy that it is possible to implement your own analytics solution. Especially in Data Science, the more power you have over the data source, the better your results will be.

Date created:
09 Mar 2018

You are more than welcome to share your thoughts via email