From e63c605d0709151f63fa258ae5669be99c8ea0b9 Mon Sep 17 00:00:00 2001 From: dcr-eyethink <151747088+dcr-eyethink@users.noreply.github.com> Date: Wed, 24 Apr 2024 18:24:54 +0100 Subject: [PATCH] plotly --- .DS_Store | Bin 6148 -> 6148 bytes analysis.Rmd | 11 +- analysis.md | 147 --- analysis.nb.html | 2006 --------------------------------- analysis/r theme-year.pdf | Bin 21937 -> 22009 bytes analysis/r ucl_all-year.pdf | Bin 6077 -> 6077 bytes analysis/rzall theme-year.pdf | Bin 7197 -> 7197 bytes docs/index.html | 1936 ++++++++++++++++++++++++++++++- 8 files changed, 1942 insertions(+), 2158 deletions(-) delete mode 100644 analysis.md delete mode 100644 analysis.nb.html diff --git a/.DS_Store b/.DS_Store index a54dd70654d1120b4b17684480df42cf948f05b3..0031aa0e8cd976166a9f4961b977015859599c94 100644 GIT binary patch delta 123 zcmZoMXffDO&!o!2kiwA9kjzk=lWrKCoS$33fB2016 & year<2023], qkey[,.(QuestionNumber=qnum,theme=theme2017,q=item2017, question=set_full2017)],link="QuestionNumber"), - pid_merge(full_data[year==2023], qkey[,.(QuestionNumber=qnum,theme=theme2023,q=item2023, question=set_full2023)],link="QuestionNumber")) -``` - -Now we want one number that summarises responses to the question. Right -now we have the distribution of responses for each. All the items are -framed positively, ie if they agree with the statement the student is -saying something good about the university. So we’re going to code into -a response variable between -1 and 1 that is positive if they are -agreeing, and negative if they are disagreeing. This is different to how -the NSS processes this. They have a positivity score, which I assume -just calculates the % of responses that are agreeing in anyway. That -seems to ignore some of the graded information we have in this dataset -tbough. - -For pre 2023, each question has 5 columns giving % responses for the 5 -options from strongly agree to strongly disagree. Convert these first to -numbers, and then sum to a -1 to 1 scale in a variable r. - -``` r -d[year<2023,ans_sd:=ifelse(A1=="",0,as.numeric(gsub(A1,replacement = "",pattern="%")))] -d[year<2023,ans_d:=ifelse(A2=="",0,as.numeric(gsub(A2,replacement = "",pattern="%")))] -d[year<2023,ans_n:=ifelse(A3=="",0,as.numeric(gsub(A3,replacement = "",pattern="%")))] -d[year<2023,ans_a:=ifelse(A4=="",0,as.numeric(gsub(A4,replacement = "",pattern="%")))] -d[year<2023,ans_sa:=ifelse(A5=="",0,as.numeric(gsub(A5,replacement = "",pattern="%")))] -d[year<2023,r:= (ans_sa + ans_a*.5 + ans_d *-.5 + ans_sd*-1)/100] -``` - -For 2023 onwards, We have total number of responses for 4 options from -strongly agree to strongly disagree, ie there is no ‘neither agree nor -disagree’ option. Convert these to a -1 to 1 scale as above. - -``` r -d[year>=2023,ans_sa:=ifelse(A1=="",0,as.numeric(A1))] -d[year>=2023,ans_a:=ifelse(A2=="",0,as.numeric(A2))] -d[year>=2023,ans_d:=ifelse(A3=="",0,as.numeric(A3))] -d[year>=2023,ans_sd:=ifelse(A4=="",0,as.numeric(A4))] -d[year>=2023,r:= (ans_sa + ans_a*.5 + ans_d *-.5 + ans_sd*-1)/(ans_sa+ans_a+ans_d+ans_sd)] -``` - -I’m going to exclude the questions that aren’t in the main list, and -that are about the student union. Then let’s check the distribution of -those values over all questions for each year. - -``` r -d <- d[!theme=="union" & !is.na(theme)] -ggplot(d,aes(x=r,colour=year,group=year))+geom_density() -``` - -![](analysis_files/figure-gfm/unnamed-chunk-6-1.png) - -So it broadly looks as though agreement in these statement overall peaks -around the positive, “agree” response, but that over the years, -responses have been slipping down for everyone. - -Now let’s plot UCL’s last 10 years for each theme There are vertical -gray bars here to denote when questionnaire changed, making comparisons -difficult. I am going to exclude the themes for mental_health, personal, -overall satisfaction and freedom as they only have single question each -that were only asked in a handful of years. - -``` r -d[,tm:=ifelse(theme %in% c("mental_health", "personal", "freedom","overall"),FALSE,TRUE)] -# this is a plot element with the new questionnaires marked. We can reuse it -yp <- geom_vline(data=data.table(year=c(2016.5,2022.5)),alpha=0.1,linewidth=3,aes(xintercept=year)) -pirateye(d[Institution=="University College London" & tm],x_condition = "year", - colour_condition = "theme",line = T,dv="r",violin = F,error_bars = F)+yp -``` - -![](analysis_files/figure-gfm/unnamed-chunk-7-1.png) - -So people love our resources! The rankings seem pretty stable over time -here. Rankings seem higher pre 2017 and there is a drop off in 2023, but -as the grey lines show, these changes are confounded by a different set -of questions (and responses). What does seem clear here is that our weak -point is our assessments. These are ranked low and if anything have been -getting worse. - -How do we match up with the average psych dept? Here’s all rankings over -the years, comparing UCL against all other psych depts. - -``` r -d[,ucl:=ifelse(Institution=="University College London",TRUE,FALSE)] -d[,ucl_all:=ifelse(Institution=="University College London","UCL","All other")] -pirateye(d[ (tm) ],x_condition = "year",colour="ucl_all",dodgewidth = 0, - line = T,dv="r",violin = F,error_bars = T,dots=F)+yp -``` - -![](analysis_files/figure-gfm/unnamed-chunk-8-1.png) - -So in general terms: we had a relatively great pandemic! At least up -until 2023 (the cohort that were in their first year during pandemic). -But again the picture is complicated by the change in survey. We can -split those responses by theme and compare against other depts. It’s a -bit messy comparing all individual themes for both, so let’s try and -simplify this by using a baseline. We can z score the r values for each -question, each year. - -``` r -d[, rzall:=scale(r),by=.(year,question)] -pirateye( d[ucl & tm],x_condition = "year",colour_condition = "theme",dodgewidth = 0, - line = T,dv="rzall",violin = F,error_bars = F,dots=F)+yp+geom_hline(aes(yintercept=0)) -``` - -![](analysis_files/figure-gfm/unnamed-chunk-9-1.png) - -So now the psychology sector average for each year is 0, shown by heavy -black line. Benchmarked like this, it looks like we had a good period of -growth from 2016 onwards, and in the pandemic years we were well above -the mean in almost everything. But again, our assessments are ranked -below the mean and perhaps trending down diff --git a/analysis.nb.html b/analysis.nb.html deleted file mode 100644 index df3e1c8..0000000 --- a/analysis.nb.html +++ /dev/null @@ -1,2006 +0,0 @@ - - - - - - - - - - - - - -NSS results - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
- - - - - - - - -

This is a notebook to analyse NNS data from 2014-2023, baswed on data -from NSS wesbite.

-

If you don’t have eyethinkdata tools, install from github

- - - -
devtools::install_github("dcr-eyethink/eyethinkdata")
- - - -

Load in the package and the raw data. Note that this has been -filtered to just psychology courses, and students on first degree -only.

- - - -
library(eyethinkdata)
- - -
Loading required package: ggplot2
-Loading required package: data.table
-Registered S3 method overwritten by 'data.table':
-  method           from
-  print.data.table     
-data.table 1.14.10 using 1 threads (see ?getDTthreads).  Latest news: r-datatable.com
-**********
-This installation of data.table has not detected OpenMP support. It should still work but in single-threaded mode.
-This is a Mac. Please read https://mac.r-project.org/openmp/. Please engage with Apple and ask them for support. Check r-datatable.com for updates, and our Mac instructions here: https://github.com/Rdatatable/data.table/wiki/Installation. After several years of many reports of installation problems on Mac, it's time to gingerly point out that there have been no similar problems on Windows or Linux.
-**********
- - -
full_data <- fread("NSS_2014-23.csv")
-qkey <-  data.table(read.csv("qkey3.csv"))
- - - -

First we need to translate the question numbers into the items and -labels. There are 3 slightly difference question sets. Annoying. So we -have to merge info from the question keys to three different sections. -For each, we will have columns for the theme, the full question, and a -one word question key q.

- - - -
d <- rbind( pid_merge(full_data[year<=2016], qkey[,.(QuestionNumber=qnum,theme=theme2014,q=item2014,question=set_full2014)],link="QuestionNumber"),
- pid_merge(full_data[year>2016 & year<2023], qkey[,.(QuestionNumber=qnum,theme=theme2017,q=item2017, question=set_full2017)],link="QuestionNumber"),
- pid_merge(full_data[year==2023], qkey[,.(QuestionNumber=qnum,theme=theme2023,q=item2023, question=set_full2023)],link="QuestionNumber"))
-d
- - -
- -
- - -
NA
- - - -

Now we want one number that summarises responses to the question. -Right now we have the distribution of responses for each. All the items -are framed positively, ie if they agree with the statement the student -is saying something good about the university. So we’re going to code -into a response variable between -1 and 1 that is positive if they are -agreeing, and negative if they are disagreeing. This is different to how -the NSS processes this. They have a positivity score, which I assume -just calculates the % of responses that are agreeing in anyway. That -seems to ignore some of the graded information we have in this dataset -tbough.

-

For pre 2023, each question has 5 columns giving % responses for the -5 options from strongly agree to strongly disagree. Convert these first -to numbers, and then sum to a -1 to 1 scale in a variable r.

- - - -
d[year<2023,ans_sd:=ifelse(A1=="",0,as.numeric(gsub(A1,replacement = "",pattern="%")))]
-d[year<2023,ans_d:=ifelse(A2=="",0,as.numeric(gsub(A2,replacement = "",pattern="%")))]
-d[year<2023,ans_n:=ifelse(A3=="",0,as.numeric(gsub(A3,replacement = "",pattern="%")))]
-d[year<2023,ans_a:=ifelse(A4=="",0,as.numeric(gsub(A4,replacement = "",pattern="%")))]
-d[year<2023,ans_sa:=ifelse(A5=="",0,as.numeric(gsub(A5,replacement = "",pattern="%")))]
-d[year<2023,r:= (ans_sa + ans_a*.5  + ans_d *-.5 + ans_sd*-1)/100]
- - - -

For 2023 onwards, We have total number of responses for 4 options -from strongly agree to strongly disagree, ie there is no ‘neither agree -nor disagree’ option. Convert these to a -1 to 1 scale as above.

- - - -
d[year>=2023,ans_sa:=ifelse(A1=="",0,as.numeric(A1))]
-d[year>=2023,ans_a:=ifelse(A2=="",0,as.numeric(A2))]
-d[year>=2023,ans_d:=ifelse(A3=="",0,as.numeric(A3))]
-d[year>=2023,ans_sd:=ifelse(A4=="",0,as.numeric(A4))]
-d[year>=2023,r:= (ans_sa + ans_a*.5  + ans_d *-.5 + ans_sd*-1)/(ans_sa+ans_a+ans_d+ans_sd)]
- - - -

I’m going to exclude the questions that aren’t in the main list, and -that are about the student union. Then let’s check the distribution of -those values over all questions for each year.

- - - -
d <- d[!theme=="union" & !is.na(theme)]
-ggplot(d,aes(x=r,colour=year,group=year))+geom_density()
- - -

- - - -

So it broadly looks as though agreement in these statement overall -peaks around the positive, “agree” response, but that over the years, -responses have been slipping down for everyone.

-

Now let’s plot UCL’s last 10 years for each theme There are vertical -gray bars here to denote when questionnaire changed, making comparisons -difficult. I am going to exclude the themes for mental_health, personal, -overall satisfaction and freedom as they only have single question each -that were only asked in a handful of years.

- - - -
d[,tm:=ifelse(theme %in% c("mental_health", "personal", "freedom","overall"),FALSE,TRUE)]
-# this is a plot element with the new questionnaires marked. We can reuse it
-yp <-  geom_vline(data=data.table(year=c(2016.5,2022.5)),alpha=0.1,linewidth=3,aes(xintercept=year))
-pirateye(d[Institution=="University College London" & tm],x_condition = "year",
-         colour_condition = "theme",line = T,dv="r",violin = F,error_bars = F)+yp
- - -

- - -
NA
-NA
- - - -

So people love our resources! The rankings seem pretty stable over -time here. Rankings seem higher pre 2017 and there is a drop off in -2023, but as the grey lines show, these changes are confounded by a -different set of questions (and responses). What does seem clear here is -that our weak point is our assessments. These are ranked low and if -anything have been getting worse.

-

How do we match up with the average psych dept? Here’s all rankings -over the years, comparing UCL against all other psych depts.

- - - -
d[,ucl:=ifelse(Institution=="University College London",TRUE,FALSE)]
-d[,ucl_all:=ifelse(Institution=="University College London","UCL","All other")]
-pirateye(d[ (tm) ],x_condition = "year",colour="ucl_all",dodgewidth = 0,
-         line = T,dv="r",violin = F,error_bars = T,dots=F)+yp
- - -

- - - -

So in general terms: we had a relatively great pandemic! At least up -until 2023 (the cohort that were in their first year during pandemic). -But again the picture is complicated by the change in survey. We can -split those responses by theme and compare against other depts. It’s a -bit messy comparing all individual themes for both, so let’s try and -simplify this by using a baseline. We can z score the r values for each -question, each year.

- - - -
d[, rzall:=scale(r),by=.(year,question)]
-pirateye( d[ucl & tm],x_condition = "year",colour_condition =  "theme",dodgewidth = 0,
-         line = T,dv="rzall",violin = F,error_bars = F,dots=F)+yp+geom_hline(aes(yintercept=0))
- - -

- - - -

So now the psychology sector average for each year is 0, shown by -heavy black line. Benchmarked like this, it looks like we had a good -period of growth from 2016 onwards, and in the pandemic years we were -well above the mean in almost everything. But again, our assessments are -ranked below the mean and perhaps trending down

- - -
---
title: "NSS results"
output: html_notebook
---

This is a notebook to analyse NNS data from 2014-2023, baswed on data from NSS wesbite. 

If you don't have eyethinkdata tools, install from github

```{r, eval=FALSE}
devtools::install_github("dcr-eyethink/eyethinkdata")
```

Load in the package and the raw data. Note that this has been filtered to just psychology courses, and students on first degree only. 

```{r}
library(eyethinkdata)
full_data <- fread("NSS_2014-23.csv")
qkey <-  data.table(read.csv("qkey3.csv"))
```
First we need to translate the question numbers into the items and labels. There are 3 slightly difference question sets. Annoying. So we have to merge info from the question keys to three different sections. For each, we will have columns for the theme, the full question, and a one word question key q.

```{r}
d <- rbind( pid_merge(full_data[year<=2016], qkey[,.(QuestionNumber=qnum,theme=theme2014,q=item2014,question=set_full2014)],link="QuestionNumber"),
 pid_merge(full_data[year>2016 & year<2023], qkey[,.(QuestionNumber=qnum,theme=theme2017,q=item2017, question=set_full2017)],link="QuestionNumber"),
 pid_merge(full_data[year==2023], qkey[,.(QuestionNumber=qnum,theme=theme2023,q=item2023, question=set_full2023)],link="QuestionNumber"))
d

```
Now we want one number that summarises responses to the question. Right now we have the distribution of responses for each. All the items are framed positively, ie if they agree with the statement the student is saying something good about the university. So we're going to code into a response variable between -1 and 1 that is positive if they are agreeing, and negative if they are disagreeing. This is different to how the NSS processes this. They have a positivity score, which I assume just calculates the % of responses that are agreeing in anyway. That seems to ignore some of the graded information we have in this dataset tbough.

For pre 2023, each question has 5 columns giving % responses for the 5 options from strongly agree to strongly disagree.  Convert these first to numbers, and then sum to a -1 to 1 scale in a variable r. 

```{r}
d[year<2023,ans_sd:=ifelse(A1=="",0,as.numeric(gsub(A1,replacement = "",pattern="%")))]
d[year<2023,ans_d:=ifelse(A2=="",0,as.numeric(gsub(A2,replacement = "",pattern="%")))]
d[year<2023,ans_n:=ifelse(A3=="",0,as.numeric(gsub(A3,replacement = "",pattern="%")))]
d[year<2023,ans_a:=ifelse(A4=="",0,as.numeric(gsub(A4,replacement = "",pattern="%")))]
d[year<2023,ans_sa:=ifelse(A5=="",0,as.numeric(gsub(A5,replacement = "",pattern="%")))]
d[year<2023,r:= (ans_sa + ans_a*.5  + ans_d *-.5 + ans_sd*-1)/100]
```

For 2023 onwards, We have total number of responses for 4 options from strongly agree to strongly disagree, ie there is no 'neither agree nor disagree' option. Convert these to a -1 to 1 scale as above. 

```{r}
d[year>=2023,ans_sa:=ifelse(A1=="",0,as.numeric(A1))]
d[year>=2023,ans_a:=ifelse(A2=="",0,as.numeric(A2))]
d[year>=2023,ans_d:=ifelse(A3=="",0,as.numeric(A3))]
d[year>=2023,ans_sd:=ifelse(A4=="",0,as.numeric(A4))]
d[year>=2023,r:= (ans_sa + ans_a*.5  + ans_d *-.5 + ans_sd*-1)/(ans_sa+ans_a+ans_d+ans_sd)]
```

I'm going to exclude the questions that aren't in the main list, and that are about the student union. Then let's check the distribution of those values over all questions for each year. 

```{r}
d <- d[!theme=="union" & !is.na(theme)]
ggplot(d,aes(x=r,colour=year,group=year))+geom_density()
```
So it broadly looks as though agreement in these statement overall peaks around the positive, "agree" response, but that over the years, responses have been slipping down for everyone.

Now let's plot UCL's last 10 years for each theme There are vertical gray bars here to denote when questionnaire changed, making comparisons difficult. I am going to exclude the themes for mental_health, personal, overall satisfaction and freedom as they only have single question each that were only asked in a handful of years.

```{r}
d[,tm:=ifelse(theme %in% c("mental_health", "personal", "freedom","overall"),FALSE,TRUE)]
# this is a plot element with the new questionnaires marked. We can reuse it
yp <-  geom_vline(data=data.table(year=c(2016.5,2022.5)),alpha=0.1,linewidth=3,aes(xintercept=year))
pirateye(d[Institution=="University College London" & tm],x_condition = "year",
         colour_condition = "theme",line = T,dv="r",violin = F,error_bars = F)+yp
 

```

So people love our resources! The rankings seem pretty stable over time here. Rankings seem higher pre 2017 and there is a drop off in 2023, but as the grey lines show, these changes are confounded by a different set of questions (and responses). What does seem clear here is that our weak point is our assessments. These are ranked low and if anything have been getting worse.

How do we match up with the average psych dept? Here's all rankings over the years, comparing UCL against all other psych depts.

```{r}
d[,ucl:=ifelse(Institution=="University College London",TRUE,FALSE)]
d[,ucl_all:=ifelse(Institution=="University College London","UCL","All other")]
pirateye(d[ (tm) ],x_condition = "year",colour="ucl_all",dodgewidth = 0,
         line = T,dv="r",violin = F,error_bars = T,dots=F)+yp
```

So in general terms: we had a relatively great pandemic! At least up until 2023 (the cohort that were in their first year during pandemic). But again the picture is complicated by the change in survey. We can split those responses by theme and compare against other depts. It's a bit messy comparing all individual themes for both, so let's try and simplify this by using a baseline. We can z score the r values for each question, each year. 

```{r}
d[, rzall:=scale(r),by=.(year,question)]
pirateye( d[ucl & tm],x_condition = "year",colour_condition =  "theme",dodgewidth = 0,
         line = T,dv="rzall",violin = F,error_bars = F,dots=F)+yp+geom_hline(aes(yintercept=0))
```
So now the psychology sector average for each year is 0, shown by heavy black line. Benchmarked like this, it looks like we had a good period of growth from 2016 onwards, and in the pandemic years we were well above the mean in almost everything. But again, our assessments are ranked below the mean and perhaps trending down







- - - -
- - - - - - - - - - - - - - - - diff --git a/analysis/r theme-year.pdf b/analysis/r theme-year.pdf index a72033b7381ad7a48ca3c8f7fb5e8dc26c22da1f..0eda0fca646ef1deb7c0860b941fa0020af8f249 100644 GIT binary patch delta 18043 zcmZX+bzD^6^FRJlUW$T%fOM%y=YqtNDgq)Uy>tmE-AG=Q?(PKyq;u&;T56H*hNZjv zcTs%(zK_q_e~#R{_s*GfX68KSIkSt%xpk6rD}?`F4o(PHs4(%LQGE)Hn+qB)aEk^g zpsBP%)wzkXrg;`{hM!Dy#6TYZK>PGehwv-WtID>hU*biByTq4A;X8s~-yaH=n~xiL zs48l{e(l!qhLV=5LC2ER4Jy^N!0u@c?n_ zELD9#6I;DT!I{CU2H?nZtbEaoscF8`n$lv8dH0O1KJ1ekvYbPua{2U!#`f)#&&#JT zVuQ*zGI4x@(9=HxWGb~Y)a@4>mJMioy!8~4LOMl`GZhit2e(2Gz7(-pbOI4LF}XO* zB$8)tew~oz2zKL94o5ZNK=mcncm0RJEnL&=4)pMl=i(ILcct<0JsXywq=Dg zecdi*jRVeb1KpryvrdsfPT=|W0nX0Mqu>iz$K4v)(#?#VTR2UKn!Gi3DwX*l?Qb0U zq*~tkuIe44UA_Ck8)g*JE`j1l|EN0Eu&B?ye!)aZqc1GtY00k^T@v;qm03YvjatKG zZJ4e8Td9gjO!;mZ3pZ5&&D4Rso$JD${RK4jn`H96*!l#9+PIcPfEV`{<6c9>l&$2% zBdWyGuOTP&;qs~?8YU)jfR>p-jZ<3pbBW_@zJ;k;Zlp(7Fb7Q}S957u$2p5}PQ-P@EB8r$DOXnk)!?zc29;V#Hd@oq3-v0= zd392+oXIW~lN!@ROs%Pa9LXdvGyg9JTEDK$&c+2|?{Su6+J4^;U~K~x@)K_vnGd!c zCesAINB&D`5F6G(I;W<7GH!M(x8xWYeALUHKNk0ux1hipn3|5_d>bef_4-jsYPYWy zu1fVoWb>*}V*pJZF#p!&$B^2Dc!ktk11!sgJdUeVD4P=gX^D7K%#^+2G1n%tpS*jX zaoI`CBYWSNLQjLcMMVOB|3lFvXAHHn|+^nbP`r>ovA1QBEs32B?L zy46a20QkB2edL;LRgib$3TCi+`fV_fjyRQ3z;fqS;LzJwfpi@aibD^F$Kp`h)C0B= zAL@g>5H;QWW4f7YsW`sMRk0x_X`tHNrf{FIBwu2i8c(bKagEZimgBu%DIdZiW@Qo9suLWQRt~2 zVE_wCfXK>@j9nOo`OMT!;-l}!nCn$Q$sU_^#8yB)4Pm@Ov3Zk}-K{_lIrozxK8zMn z?w?*|C)BM>rkOhNi#Iyf5z>5e#6L5oNAX zi%X|>@Srb%=8{-tF1VD?A%yjhcjR;>eb$O5=Ap#W54?13E6^BXu}pU0N@ffk3$YbV zyHBULv;AJow^e^)u)ZgYpjl;zs^k0boji5xA7_=EQs6*}b!jx&lf1ZTC&`CZTDU<>Bp>tE$RqE2<=x6J zKA75fJI%HuPUq0mbGFOCjFvzZt|jD0?(sH6SwCqyv57nN%fLkaief|L%P=#nqWb47 z>z=H0g|M2e(zax%mMs=3jtrS!hV*lNOoVP|@>)p)NW9{ek3#RJxQ1EIL=zwtey7Z5O?8`5%!7s1W1FkM+%+Wg9*o(ZyLr-67anwgpvWyp@F#ia42ys zzIuxD+sT$r1hcyJsHmw?x!Yi<@Kxyuy+R#;$*ZoYzU+V3$`T4mdFiuT{aQc5MNfoa zXkfM7)tZpGB!C4X|9x}r#5Ibga{Xyp@%WZ?bIr3M*Rx~?Tq9p>p0}_@9*iq$--o98rv zO+>`jEXEs9;lutJW(2@3v96?(RrS&}l7=3%4p3zW*XWfGD$(r?9FVCi4oRk;^cIX8 zw>j4<>yjdDSml|dz&Rb95DQJt)yRBc38A}Y4b3^F66h75AR1Tcon zdL4UUIX-h5R#?*-9pxiJGP#-EGLZ|p8qW8sXC`x<5%U1F^N+$C(1S<#X`7nO?GCZ! z#wNa61&)-M4>Rb?rI-uXOGCdzR6n5n+4i%h@Jf+IMeP@0>`6(MLg7YIA&9h zprjO$FJnE#WS-W0{3bVZG+w{`OU`oyHR|6XK_`IuM%G2x;IWWi+hLs92A~!gU#x#` zO3r!DTIUls(V!v5kQY&oBGfIT*>9|dB@|B3RF1&BYr1x~B}3_Eb1$;_RkrMin4u#` z3`ViXo3^5kOGD-!o_4cLg>Ze)B&^xMdjoHo53A)J%)Jdp!+x*OTD6qP+K4Q6YY)Ay zl!e>C!t=5^;Gz({7^7WfeUOBSaq_JFOVSjqYCEsm=~xqtR-h$qp0XOW?aha6$yX%( zvu23&P0>xM;fU{0=2SkxnD#_`&m7;njij86M-bgv%Z!hhrs*ith{6CZ2vpKfC8qNc zvO>J`rTcxx8rw9F9zoj9UOzBK5QtcIK>npu1|BIMf3$2J@!_!-(#ILDk^t>M?vG^i zi|lAGShBWFoScoZjf|;~7XDStzgiKw{_RkPsB%3&lM}D|nTV8c@v{EKy?namAIh?v zZJvu>$BD6X)F9{s#GB)v^Kzy{;hz5R`~4L;Bd}TzZ#OW`1yyzsNj(W9xC2aHYW|3= z{efnWAq%YDU7Wsi9bi4oHoyS8$ImU>COA&a%W3aB96ROnbyS0yRz;zVbsgc7HZ1~^ z`?aBsOa3pCq&n%CJ1u@dr8I>8kJmbw>B}XL6H=|DF0OC48(WdY&*mO2%Sen?Z7aj- z8VYnl^S38MGiuociV>Lh@I0~zz$OBmH(puk8BNZP$%r}!=}fQn3?G)meKmoJ?Af;s zM4`HO`?+4QNk%4}$9WJ$hTd)eGCl|s!-qLwaa-Ov!9k7|T9r$RXfc}&cWwb&mw^7p zqq-8w>Y*QprJi@3BdN#*Jp^m&%BULQcM9H{HZTFIJedXAgO2%sqRsT3&O_La3=KeJ zNt;_h#Jr$CNOm!%6E} zuiXN7e00Ojs7-)B1r6mwWNCQVsBvN8S<8Og{q<;7%)Dfs?%5DbK_Ne&i)8jY!J#~< zjM@rqi-Yd9;(GkmYt|(;eU^f^9Pe3AwN(qg~3QvJi=m+6_<6 zYy4__@d@>dNIdt#c3|>!2WH=4kX4vBSpW;{QD~Ub(F~+BGK!eTv$9t+>%ni!t|CpM z3utCoB6<%-NARA3OIji3JXspQ%Dn1L(74RTeL@1kqo!R7j&g73|N312eRH!AFvk-0Va$8NT4Nv^fA=sZU;dO_D-qgMr>7x}ZmmJL^ z32E6SzVrS>EI6VtMXX1y+xK{j20>jOm=+(g#?Vm)wQEr>bRT`zzj3AX{?GNVj`ATq*x5M z*+ze=oLXValUj3aA=VLFAk>0CC?QT26;?xf-2W6&PW|CuJ07UVYdaz?3Z#4|#FQ3v ziRob8J>MBp-6gaJu0CnTwU0b)ph=Hk-~qj*rhXLD1_6O(;+?ULC&ZW)&?qb$)4rBD zRyPfKAlGr9kU{*PS9+o^d6H-|;M@9`WNp020ck(kV$;r#(TKdsOC>rP3!}tIW8OP; z<&sqC0pI}p9GJeV8N{d!r65p#WV=*tszv*Zu+w7zt79u=g>fOAZth1Jh*-o1x|Mz= z6XZh4^5RDX?3?vH4*MT55Lj%;!A9gg9Y8>l`lOPhu=>gnAqH$gW9wDz7i`x z^NOU`&;T0@!b|F)gO+r3I6qk<{EgNA*sNg@xi>mKlX2M+6pCNA5

fucb?$R=|ifFnWL^X+8Eh+_tU%4(vv@N%%+K|7MND4Hd}OTmNbB*>;K2 z^uGynh)<~6_|ZdsUisl2w$lNgDf@ywSff=i8(0vM6OppZ910b9^q#o~Q`JFGX5q#e z>Kohe@}51~I*y}a@z#~0P7Twqju78L>8<7rKUL_U6BEER!KkX$`p4)iX3q$&ddLu# zA6N2U+^QmoSW^;=YyYCckvU)3#_?8ocqo-*EgqqG0!UFkL!|em#*%-c!N&|XSd|Gj zjFPV0_zkp+cy^?MJ3^D5t)_P!CVO}pn7T6SHiT(KXFB9UeQ5@dVPR;(zJg4cN?T%<96b6!OLL1>j%ca*|W#L}(5IJRe!yCiIE>zPr<%7;UK$Um)d#fSVO4@R=d zM<3!bF`iM_j5rW?y0ui9fS&#*a2(aIeosvMA!0w_Hb^D80UakM2oe7p=3XqgZMO*h z=Gds11toYc8lj@Tn@y_=gEfF+CUvT(FWX;5%z^)k9Kj|g>lUt3&9-o41}ED-L~HzKiHJBdvaFedu~J4KSm1De(YX_ftm6 zcmIJyChsNEkCEi88c%nhQD>zHz>j<0YtBGY2PK@)kQzMGsdZvWl}0V7zhp z4Tmq7$s5qF+2MM^&HIH$K4Y+|#Q7qfxEz{gCR|4?C3x4SAEIa zHk2qXiEND1g4irL7Smt+>^CaImNO2cijg8?V(dw!K{x8a<42^Y@%N+}jD`dEN5I4H zh{>fI;`7ASi-R5Df^%tLLl#84G@m_EvoaT04SSTUNocql0B8Jdkgu<R=iJMQPN&*)lqG3szTG3)H~MJYy` z;)`uYU#=HfeR(zc{rGuj)P((p);GoK1J>bci+$?NRcIIRrT+!pag$Aj6{m5$RAqWo zzYg@P6F$@G&(L4wyyO8aWYO9K(j69*2I{|Ip_``#kK0@!YPAO^kF`s;uyJst(LRAW zC~vgoQEuP@k4+@iHWx9IvCV>xweor{aW(tIyDB1a?7yL7@)f2g)JSnAb(VcxR)G7$iPLD0b4y$vmhTy8G8owD=GEC+6B{}H>Y?0mNjxi zHF<&kVOf{Jq zcBW7KhJy~t<)u3x;zSDG!>xt;^;K!n#|sYpVbP3i_lP)^6%Vd&eMjuW)Eo{zRGD^| z5rM=8+s+2-{;H;AjKtnMD^%~3$GsFkaW`cM5QKkHNcfb^KW&Y{TEV%$<{o`fzh`jirhyvT*`zEWIoSp4fUcHH) zld*p9sw8$av7Os&yYBp%u*1g$au_@g$*#x3D*uf7tE-YNw zdiBbo0eE*z%JITGBJ|lf!0|86et5_jXghvgj?0Yf&lK{eVFmS<8%(QK?aVQ4|7(beU17i0ui}~nCu~eOL|esF za?#^1ruSxqau?_v{5=Vjj(`@RmomR_Y&vq&@Jhu%U8Xv~Y-6aviR0r7!Qp6fnKYSn z;+e|YLIb!LN<-TeBgk=+Ez^dFJ4;|V8s7bKTt9JKa?&xMt#gXJM9{qT(0ohr2eq9GRd8{))w<=;XH0}mtXhiS1eu}^9aVxJ6w(H8H7!}gIW zFwbX-e&XyJ9ogCGPL|%Qkw0Z<X)#y9Gze-q>R#Je#NsCy=Xq&Wu8G@S5#>_AG+T|Dmk5J zr!t-YU+x4P{8;I+6w5cN!{p^v#tmOg;5?69Y~ZR)V`H{X1BDeq3zGD_-kU1@^Lsd*`H^8za)JC^Jq=p?*h_zgDDU2m-JeGQ?a;uS z79;ehLOd*;plB01Japr)4z5K1UM!9-FLc?G(!VYoX|aqtG~Lfl1vQnk+zPGWvDP1>6mLA@7A{{5h;7nV)VO9@Py7PY zQoE}kHV_vhcs@UIE_@42zaCILhWnL3z)EEWTP%ocf8C z3WtGBFfcWhdJvQvl_WrOaJp_+Hx?L8nuv{8-W<Q&6H{wmRCajsg4u%GFBr{Fn^5dq}+u;h*!I2Cq zcQN&(2Yo1Qh$HoL!$>3@0++2uBgU_@e!&hZdi$0YAMwuFGW+kh=)W~KIX?a3Ltigx zixJhc{K$igi?41BPJ_=%)g@+2nFZ2Q25bx<(7Gmb7dY~T2mzaAI_RfN8KH>GtiBVN z3^bEL*zubzOkuE15BYvAdQ~5>xvFnQAB(7>b!Lg~FI=%mL;Mmyev<)Rt z`TuN9>)?4#t1rQ|)Yb)slkO}qny4x`2nwUFT?7P^&S}r9W$k~f2lYR*we9u~+%|8k zLsuy?<4_}|+wk@{4>s#b)W-n8Q$|pUu zcTkq>IRZI9ynWzO*|K6p7@wy=RXJZ$8}2xom7t>#!sFriHKBK8`~Ewbj7|< zNPdWE8he=BzDaOoJFAy5hgp|?mx;01nZ{_JLVIs@U2{mY*trsPlIt_YEtXav(m!LgqMu#uN-GVTO1U$K5U)1Y#Et+BS0~;AQwC2#aderMSAn2~kRu0YsZg`dL z4MIviGt|BI243wb-#v=1TOExWXm&yGq8d+v7BG!?8~r+LFlNRroA5OBaFOg1fe(lte50VrJ;8MB{iRC8DaF#57*J@Sn#jS9MG=Fh?rT!r1zy5sM2dldjRfBML*#km-xE99q(5%!HxoFJ;aR2%a_T{!Jjk z(o8e1Y}{ ztrwpW)J<~OG^K0l#*%BkZDJwUQXnA%+Rr$im&01cCLn1wmL?J~1;@}dXiaDztcowU zddgtsCaGFe34C@dLHj9mn?+}y{{nC1`T9BD2(Li4i}@~`CeqD!;YWE0?3hB{;gqK6 z28+2xQ!EK(moq@OiSHO9KPY|>R=2=seshKY3Tvo2?yZG_^$Y}f%UmOo#a}CZ0~oK( zh}yvFS+~jQJDU+4nVweC{*mrk$m>`<5Hq!Y2`PLI?^@VCJ+btM;o)l4V}hP${zP(YwB zK_rwU%)JJ~m!+mty$by3jnQ3O4sim~oQf6-t=cf6I0?CvO-2&S*Jy3$Z=<=cl3is3Z;XF6-=;M(yxlE``s#P!zl$mkyffg}Vx4amz{OR%#bLz! z4H0lqyh)0=e1qA~G?iC7F?GV-<2BGYmuL!-DZ=+^(8@kY7(%1pxB7LM=!A1ZMpR%L z2XR$!(|v(oFU9<{r&|pG?*%d<#YR81 z8p#Ow*Vd2g0nciN1jqH0ZeT!Jc^11Vzdka|y#lhVPNCqni{eHxugA7BF5P$07d7Zo zwchrmA%qxYfgk#to*!rq)d~{Cki*w8w}!$pE>i^3EaP(Q#$k&!XB{v(M@Dw&2;UoK znTxbdK!70TMwFYc#nHP8=PBVta`HQ@vp>e~PC~REjjI$qX7jr$b~y+&))8G~S!iz{ z4uO#;gSZ#h@y2M>>sKM1N)P4v1l6#yZN=|%;+JXl`&+X|3o@vecPSyej$6%TbiaS9 zTF|)yDZ~efRc+*1e;6hJER|yJHFImduvu>Bk5N+As;hw6Y}S0%om5?{E-(5_pj7*L z?q@h%ZjB3=72HxPTP*3kuCHQlX?{cvpe(gqwxbBUV}m2f&h2De*AIz0Pm7Ul(u2@a zD4d9_6IyIH3@yEZyBFZGER+(+!=;NoZMbybu$CbeupEmU!F+1PvZVSEq$}gty)E>a zrw6J`r2is-QzdGPPh~0Mow`i^D`#;mmPw!V<;{}Mx(cz|jV?LDr_W7Y+C{7fIy{sx zBR11LvwtIpgh{7piY%-6z283Xwrt`T;@uV3Rt+y1DQqvD{D)DmKVXx z2Cv!^u=?xlGYy)S6j)dqZH`m-_%aD0Tz4K*d558#CmIsI`yR7JwvyQq9cl|u#+R7L)Jh+A}^1hAK%wE)c2Wzy@u&o&z9Zuq{RHZHRD? z@GBQcd4sogUUNh_NT@p0nsgYb*Fgm#Al3o>nKAD*X9lC1+h_V-2{WKOgyzPI1ut zn|@Ha%fF?&L3~d>$jrOhuYd9(>7VZ$V z8#w}srY@So!yNH*7&^fI%lQzhaQyl^Lj#F9Ha3n4-w zMO(`GQ5$2chiN1aSbM6@iDBf-n`bW_I)8T;iwG-h`z>nq;MrtokT@@{4Ysx3%qe8v@+_t_ zXr*30Hmr|ymQ-C(Q$5DB+I%doDNsd*u$G}(*KNXl%-nQ2YA>8!S#h$2%bTpA(O5yb zDqdPW5-__d`fg63d2cj&3yd{9LwZdw@e)gx=%PJEK18PLZ1|?SMzxiaI_pX22^mL| z=Z@p6Z*!_Tw151aM}w79vy86RoC4CeBwp*OjeB3q^tWDQnynr4$A0qOt?*0X!;)5N zOM+TY-Dhe`wB-pBj{I9bxeb)dRNpeg<_X22CF;G)N%4B)?tX**VH9VeFW6afw*)%z znmdZDYe%3|McOsy$`!d!HLu~OO?7vBL>n~ut@54l+ne%}l< zD6Fs>*0GKTR|l*_RM&MnooW8eva5a$25T3tdPx$_`ulm>t6NXAPc*x9Co$HXRIm{! z2Pj8jRv$=|%H95~u}p*R&vm{V`w%ZLr3ACKP$w??+?3E)V(}Fl6qZXvh*=2? zBw*GutM5wL{BuoiO;mP1Kk36bNN#^ATp`7|7m^&TkuF92O~|%VS+q@cF{ilyHduA| zweT&OmeBS6DpCk!m0he4)pUNhz1&ku;@Ou1yNg?*UPe}$+nL5!+$cz?9-C z?%M0_)!BEeg}a-e@co&V6&$qW{-T2XEj;lco(X~N4^TR%*-V!7xd5$CiuaC|Jq<>R zrIe7LEkPVAW1I*VB$RD6uUMO`k~q;!`J1Z|T^3RE!NH~w1#vS#^HdC;=Ro$vV6AbK z`J}nJ3C!7@;~Am6W=SIuL-Q=>{VvwO$0vrvCdY2k5@|M5P@`FNs0p*z94mE^;q&CG z&YzJ9iyHeRnA!ISrLh~$JeBD4ZF!Qb82SIS++AhYxMig4zQLAf_?LhzL#p^;S{uD) zKl^Jyvf)BBZ0~b1h)A(StY`eCnfsW(Z%Q?V+@O-7)0qW(9oOiEpNY6Z56Af$N+;?t zd8pQ^TbDoPW}-BkC7h9|V~msC)tT#&IWI;_tvyG^F0N;4B}gGBzbeuV)g#uV&99>t zv2pRI3$hNMJNLt8Yd7KpH7#`ps`;QJv~QT_UMn1}ZmeE!&2{V&KiP=-#Rt7dfGH*Y zfnv;RR@E!k^A1o7zgCBk!?)8H;2LzmMV~Ee<#;q)Ws;fS zS@yfdOmlj1y$`rUht2?qOoy&~O%)Jb&v<tW<^;h0rJ~ren%r%U{wGu6$N-NXJCp zaDMfT3Dw(km8(cDG&IXDS^H`m#4qE?O3Z?a;zz220o0+J8l3E=WZqU|YW)do91Op^ zv7K~;KiMgA7W${&+SnIWpr+H=>%;Quf9+5JEN}{}G{$ASCqu~}1TN})nC#ZcEaHN7 zuws1&+bt}YSoen?TP4!w&%t?{3CtgTw`jI1NMpyrVW?8Y=$Cn0$G-8s@OviMg5;h9 z4-mAB47Up}!bGRDqxntyob4?r-R0s*hjv+1Zof5Otn(;l;xs4{crW$VaW?W%V;GQK z2?=*Uk80&OO0yfDhym$B^|zn_OnEwpKQB+BL!XtA(`we;w%;qf;}Ep8u_V`DktMbp z=`eN=1U14u-aIj6o#Wkk)l+VpH>rAkb3!#mos`s(1ugn)x@7LL%sPGR+`1Z8emn!L zdM{nM(=j8bFg=Rgyz6BqV62A=kE>cDH`ll!CoZzdh`OBRK`VQH*dGKKst!Az&kMC_C- zqzsb3jupF4J#C;rsf&$(k}~DSM>84?Jq8g*fSq*+nh?T=LweGNui$T=#A&tEYB}m$ zlNR*`U>RWFH9TJVY`_0CJoRbgSI2DYDvCfF(Mh+x9ouPx^&lN!jdtSFe<+cbCA%@D6W z0`_y_Y{dAn8TDe>7*Q5Xz}`&QX|@yD5o3~GrG9P}wb2Yp$D0d5R@R^E+_}O1mlCXB zz2|!VhKdE9%KDw0Q`gfG*4EJob-CBfYFaUYZ((w=Jgm^-@q2~g<2bgm$i2S+d#IYZ zX4D}zWe;qVgjQu5G_;iCZdA*QEV$7w#sw`KbOIzx%(+?ZJXSR5Zmjy$S^W`UO%3eo z3j59PUa*cx#-ru}#@bagZ}D67*QhnQ#aKI5=z8wj?Q1d@VJm(#AtmP{3rN!+tIC=G z-5d?}?wL=~q#MY{e!Ocxn7nc7{<(&px$GgQ7Mp2hRAX4a?K(c4o`c?7&W9F8d#SbFEdNDsu?oznH$e=z&ZwG{T=o3qXJ8 z7Y3ZBns}a?zHfxe^JEni{hMgluHtyaKq@!0PpYb=q35Hi)LJlDyqvxT2Gr|bwv8t ze*ajXjiIc_NTJMJAjd^-)m~3}W5tOb9pd8K;^wxRABqMdW#AerA4%5ql3BQoa17d5 zhvYksej;l2Qbcxgg};=NO?T2N8+15$f#hcd`**SuyH$I~Nvoem7TXoXu8!W6tPWol zJmzp~R9yiN^?w~AKBGOUxG(s2S{xKDoK2oJgU6J8T;X6H5=w)>*KiyXgmQ4)xGn%z zv>oTW$paUAtudYgPZd2Fc%SYX=@YuiUuq)rI(YbYJqWmlEo>ykrxg)zYQc1~UFE;^ zl9AGa;nv&se$c~4QDL%G$2f|ajK*fiJ|bF}xP>4W@@e+VwnCrFjj1NC)+^JSGzsK? zHyYK?s(+2v#Jad}W&!70=EzrFewrIUH`HKLri{}dk&0INC6GWVTV0eKP9x2n|Buw9 zqo~e}(D5w}eLlXy1Z5k`73Fu3$OPVt@z~F%+5JuQ7P;19exyvC33Z@yWGq&l^nK$NN1f>EgAbW)tO_WWKT2jt=HA^8f##ND5iDWD;@?=IX#0C$M zZ!Y@3>{hmsB9llKIlnNn(Nrw+hBsh71W;dhPbOOq8QrEV^?qG)2qQ8QVmAbK+j8jU zmtKExB*D3Y(}{e8Bvf>fk@$4YO#0WQ!`n9b-?N=A(3Byt9tsx}&}(k5c%P7k)fZH7 zvUg?hL}@pPS{TdF2!$PWZDl-04=!s9H7aSnw``=f1@-fP%%btwueeY5vV8!X7pwsG zGTC41MF?#>E&2UC)t_AxuZ8E}SZnP!#1DKDSmSU&og~6E1K0tBv1#{${MZuKxR788yVPdErUb`t<)s!`5uU5gmSgE;oyp0!F%O@joV*i-?DEIZ9$l;)u1;WO+V$#w>3T3;SV z*Za@n2}i4npfgE-KcpAc&D*nlws-7Fgvye#?8b6iRfLE3aEO8IP>7N%zeHvGzr()K@-3-SZ2|B-&e1^x+(|)P@pewY0 zZ~OK!D5UYT$m8o5!e6Yzn$FgBWZM`czZhH_)rS%NSuQ5Ni}8M7r@`!jyVKjFo;|eY zc3KwH^&S-B*ZD4iYb5-_E4S~-c117ZeZt4K>*)$bHDdlGJbCF1ZWQyC@b7(L6$q|n zKarMgL%07C^!u?N6h_4v=_I9T%9j3aJI#( zUjU1hn6!PHxr2B5L#F}gr{@Rv9i54(Is;xPR$5Z>oSqWTJ#0L|yzoY225=NquN^9RZ=u0oB-C7xf@%U~ zvXBI>#`!<7ITMbfpk)=}-{%7U7@OmC2q*V}?XjQX$WBC^U~sTOYaz?g1h^~c<&D3U zvenW?){M0rRrG=@)^)HKv6)m4e3qU2)`>3j8ubZ*kD#u!X=? z8@jjt#`UJ3|HDJ>`py01`c?cN+rR*nNcC7;i!Te@V0deV-vQLsx$f#8AZe2$I(TB{ zL*5|PRBvyej8@hKsG5=F>xXU$({}rwm8}QItBHN(OFNjsoc%9kG;|JZO9Yxj0Wbg* z)38Elhdd?4#U=SG7Po&tH{oi=Sm~k6^ek$?pxZ%cViuJOj;y1z zDC-vLT5r1w$9JlG2~D;Y(AVI2Wm9p&a~5~yQEgpONv%qgaAZ~CNVM~2ONtx>F*O!j zWaQ>`!pag$t7PW+0^nFlW}aAi=_)+?fy_MUd5zQ-51F|eq|RdLE}6MGD>x1Uq~=;o z0)6~hUh()CTjOVNuXx$}!TYS=XHid`>fa_hyLe4yxc8(V%|76v!V<1S(r`{_ux1ud zHE~P`(C3~YbvY);pMv8RWa6AGb7(37GBH5_{>=}WsKEipQT7S=eQw{+IkQxU&jX=uEJ71QW4R~h9|b3(ke|M5GvJg*vQw@Q)is&vcE)WDA)3q* zI^wpX(wfXheT_F%Vw=nee8o>i=PlJA9dcc5a83Oxn1k(2W}8*yb!@+>zei^7&pe=7 zpH6CSLObMA{+iTW@-aBhk(e`)Dw@=PCNal+B+yqN?ltb@4scDguz8K|e&L$VBlH?K z{+4$lE4}GGzL22Q*FESxu0hV$FI($9J`e}~{oQ+9DpsjaI>39pnP_Xg$HseHfG~K! ze`uDQB~bj*?RfaAuU`ASN>%>i0OxAY0)V&#rn@RN{kjY;9R%uIuV#m@oO!3;U0mY4 zf4`|6vaxy!tN<6pZuUy8J>J?^mz}^F7skoI+<5hby3t8qq8;-RBO4jI5CuU`!GMda zbiMgT-+RM>vdpG6v;j#4aL<=tSdRNk@o#;L;pdr|FV@j`C}=n1c+SilmLR%df|)3M z)+CT~EGK$=l|Pq}#p7=KdP5TpcTS6q(`~Q#>iEO}7}ze}ay!dNx@y{9)1K-+ktLXn zI8q5Du>vUA$m-oLGTntQRMb$IOa`Q}qgO@iGek#Az~$!7WO;5*8(Q3)FrQs(+a*dx zmT;sZ_Bqi=|90ov^j5*3M$EB;_TA^M?^9zFd>ElL>28ptq?jvwCbZnf*PCbg+m7%@%CTWoo+XZFqKqgCQb%-uIsy!eQdx_eS)3{@MVoB*HaN0+!Vra5bp zSpQ_muDSaOeZ(G_*Q7ol;0t~^x-cQj`Y9ttdth*kF!;L*Y(9D4M}+C{95+86MNX#&r-@7v{t(rErkH$~}w(s=>p3T2YeK`2MA|mxAjqa3v|!2Xmy=D1fAO1UUsu3l#jElbnU!$wOj5u?+5=?` zt`X5WkSjs-j&jn8T(n=M-jdi(SwDGU_D4FQCG^OYAxc&}q844t&d?Kp6_HZpEg2N{ z^u9kOvz-x6^i8^ODdz56peJN*Y4!P29)&PGEcC@zu^=eZ5tKQ9muK%b%1992atSrc zM|U&EIg+c3X+1v@fF+q)O5;S6@5`U|4!UQ_Jr{(NN_X^r1504e6${^@-!eND4kQR_ z{7s<%GOW#7zY|6GPD&Z9&3;e-#PzpQo2od`dKYkGOW{)Qv*myjepLBog9Ud*PTGLu zQ0ih~G#^g-raq26NB5hJ*~SEj0S@t~5A~gg2nmDV;(ZNp3@1yy!S7sbb)OfMpM_)P z=fzk&udqRR$O~%t!YBKeoGXG~0Y`PZBRr%NQ2y%EYC(AOCA6uui1MnPa`R$-OPHhH z!QZW@p;QnKjqjHVE=AW*{ousQMQSzh%|rTHA!dPgI%a3~wR}N1$irKpfZE9la4AO~yO5}eE(a-AYUpi0CfImx7(OB&mN689ZPqTlYX@yh9 zN%yM1iK5CvcZbD7yd^Fhl)XM1Ac_-JRqWk)&8CD~eCE%PJ-jxCHmv+@4F^#F)Ec19 z`;7ki8kv{t#Ec$w9=WVDuYa*Of1%kp2RHBkT;t^C`k!mhA^&#*9FXVS j|5q3XgbTv)KhJV;U&{?QP delta 17970 zcmZ|$by!s0_dgEb>aC~<)Q|>hfsi8qh>CPdf2I(&8 z7&@dIeg|)SKiBpBT#x^pVehlgUVHU=?d4JSwZiOcUj+Z<;Na#B6@B#IQ*-hL4?kbH z&^0I$?4QNCIBvNBj>@O-fSq>sxiMJxtCLOkiv>Vy97%RQ z*Pef&_s9o0<-)H!v{5-||Lk$Gfm>>cHP`fRiL#-xTxH$%VXupDWL_NOQmb4#{!#t) z`q4-5$wg^7ra*n4k|2I;Ok26mC-aOe|G7{%)_h9QqDZ75XJ^cP;RgNc4eX<-U?zZ? zL!CN$5>r$ARK;n0Rz8<&c_AUbf2@p$;ETm%Yi$i%niE56oI?@fCZDAWH|sA>Am`7JM+b>@1uWx{kq$W}x#W))*|_EFUR z%uR^D*#KZRT>X`Cypf)b#A`I{a0c*q_7aC3JtM7az9IR$%8PWGDAu>D6V+v1Yp8YZg~P#p4;QfDK)S+ zGL^pCGBNZ6vE;<_%>&gO*dFM_1%9Eyl`~Jn%r9;>yFm4RPZk$bBu`Um8PImRXFld* zSY72Cx)5MpE9>&l(0jO=ApF(lUP8cO8?_0wk~2n-Q|{lAtaRPAUL!AbZD=N(#V_9# zP)JWL@~#+CjRBr-7D5vU1agma*oP!{(WEFP0D_b86z`)9=Iwf!FLb*ZjeN0MeLx_xJt@Vf$D>vHZHn9=h`ku z$GTO*cOFx3rsz%g;&SL@8NILGg~Z)?XjdH>rWnYM@q<3XNKJmA>N$LzBZXv?io(8DXrQ!5eDA3V?=YAgr_ic)-+#L7H|xjTu3*L zujeVqYDoS(WmHnspaqw^ok&wgG0#1|AN3Zw_prZmI#PRcnPaBBF707LD7bLYQuNUh zrx#izboB&?QhmTB+7_Wa6nM_`ICpAmMp&7x*PR+KrMlB`MlClZyD|EAU!F$57un%B zk~x!!Vaq4+e$7dKWIJQrbQaEW*{b(VP8k0qeD$nRG{4*)FP!|mMrKL$6s=5s{!mB3 zwYt5nEr7Gx<=5LKEu-D*jLQFZR5MqlJoQY0h! zA;>B>_jnyLG$@y!Wc#b{RJX$sIb5>;xHb=i)V(QE$Yuq`%ODSh`C z53rw;d7f={-N0G1s|k_R7@_f^DFs4wPxA(--8IWWaD}y~+0`l2duwr%h_>`{c1{9~ zg(D5luiHkfmWw=JA`a7|vb%-yLY9wB)QAqF%a(-MCQ=??{64A*F=KY^BQ8IR(s(hg z5Gk3N^E9tCvTmYTfM3hLB^^hzXsXyG<ok&}%^3eU^U(7oh@0ZsFB<|(e z8g=|z+NG|46Bx>~?ZL1iQjm`rXog`%Ry8yfYFvF9!!YJ6ovK+r+f`ZRklFd<;D9Xg zYJX;%%wB86x${9fpA1_Pc!>ALE3?0}NMxTmX|5Qm`VhJE!5Y6K;WOf3=`bhmVWR42 z*qEZeAo}Q)N4Y*qGhHJpQfVUmn4|-lzlDPZWYdcv5|Q1#eL*}oC)A^+J2=JY>>I>- zs$Li(P!VGgaHT0ie_l}DIW?>}(4d3n)f7)M``p9+pNyNaQvo&zR81RaCZ#HIc$`SzH?*A`5W+ z6wnA)Kxf=*9o2skF~TE~G=S$XpKaD=P6{r~hY!mQ3t$`MS6B2Pu-^x=!-`C?$Ah!^ z%rZpPqRG1JpRX}uZzjbQCJ&t^bTBz8VIa|yng_F3IEt-u?Lk?oW}S&63qC6^$ywHd z8T>*Ew}g`f$3-UpRbxrCjnj@YfVGv83gRhswn?JsuqQ${y$P)%?cOZD+P@UC=z*Ma ze~0){52rrP+jJ7(Wbob2dm$FFYGa`iy!8{V6hO(;8neao*ai8-b`Y-(`C{}xuBk-b z3CCvUaP9yl@oeeG3X?oQVjC5tr+`j`S<%8^n1T%(-NOK=KD09?$CYy;&+te5_Js2> z)-gox%<8H7w!b`P~=p5U}QNK*WJlLoM$S4-(;s@#pTP!f{Vh{ zraFZjiFv`N8geo$Bqbpz=v--#ZD5+7Z1wb3EETbmnYbzcW2M^K1cPRO4+~MeE?+?fG9Ib+HYt$O# z@Td-$$;;TJ{}yK;xnZ|*t2slZY#u0*9vRwD%Sp=kfY-bp=ta>jkk{N)u@E?~AuA)z zXHbcHxC6%;+uiUJUKYrNlGyUEJ{f{OSg|l>WfjC0DUS?U9DFmC;fk`xE~qt*au#SE zbDaFzV$#BP`Y;V+O-`9aVJ8njCWG1;F_J1}cIV*z$}McwSF={bsDj)HAP@JkWpR}Z zKFUv_Axv5-!o3W)qhRB>gEeVpgKJ1Rk2Bn}Ik`|vLc;-JRAtuwXHh=fNPgJPMN{Ed zR&Nu9Gk>FS?boJ}pVjo|h2$^&k(!MW5thB;&r}DDGNf!&^MpTp^$j4_9J`9>q7V&b z_AYV={h9tjaf;F6WMD#Bb5GC^vA^_1$-U_vV_|Y^l`pN!=g8c?i2-p#V}#x%m3i&k z_GTk%<8p)I@pMC5pZs91hwQ5FMU9l)`wPF9ySB_u!4-=6TEaXN`kI2HiQorIXN9c) z3FbgrbX*CtMoCSaR}LDN&d)fCW*s=bhlU3z?-{68-isDtfB>t`Z?b_#~!q* zrz}dF`O!IFN$65Q7H3(4yb7_1f_Ys*77n_D`j;+;LMyM~@PbI@sX{g5obThf()hfk z4+75e1`@{f)XV1oEE~QF9mNzTQ3NJ^Qdbq+{3eVza(7sTW8w49RF$g^!f%KIHHCIF znhtxv(O10PbEx$8cXVexLdTEGf7v^^)3IkOvD{+OHk6Ro=HSYS2}`!PfJ|%A&LoZm zD0aT&W4eGSl>O0!ji?-q+OgSEl~YMPnRHPkRpQN!m1qH<7Qcr%2+pN+SFd%9zsE}v zHhisDm@Wnqt3FwOK)xOQ?60Tzo!7k`+aIWuD{aN0;8Am7R;=}SxwGz zx2hOYKl(nQAo^X}MrZHC%VV0h$L6Yl%KzId-*o6E#lP>oYMs0Z3!N*e0+!jBh7Cuv zU4uBnnm}_Kz8OE-^+1uNo`n zN#El^rLe!rVO+9D{CJ@pS9s!QsZ=Fd`RBNoLK!yvyvpNo=M?Q*Jl8Fi-(f#3U!B?i zuwAp@SxNQR6EGIop%@J zp)A+Xbh~$Pqh5KUZ`xRW;E%T755IVn7Je&MKDrlDSCIZSPClTxVcpQ|Q+{iewD}F@ zIGZtXGvo4;gmAQ!61TTRdrU8`4xJgUeRDEtE!CKWQkjJ)kVQe>=SR$gZTlHdxM8K2 ze|=aO89v(&8qc4_t2VlOy=&!uDZ<6lUpt@MIT#e5EKH?s{P04~gg#JJE}k;NKj`>8 z=pJjwcVf=0X`?A+W_(Gnni?gje}(r#P)a>&h=Eqe#^^|i+i=e2s`ahTRM(Jvmcrz( zWAcC}lypDNIk>-2f3J_Qi^z}4nj>-`13L)FH1O00MivbUU= z3#PU`tQ-x*u_i(=xB>%7_WA%ge&t*EfI;6cfr#EhvAV7IFqZ2Q>@qoAEqI)1M6~zc z6bjv9f@|wr$HNtxYm{v`+~n5oIg}#1CndYDj){gu183rn`$CzJSs4iXX%{J!0!_3K znhc+m%1@amKkdT!KNdFSVVJFx)^EAt8EY55?Gn4soPofUfp+sGp4Zbn-1GbBSN+_U zkH31{EzGOUM?4EOS~(A-bH3lJ6wzl84!e)6(a~QTmXOJ6GBBD=BG5%0al_j0 zh?W&t2hPb%q#kl#9XwazeD*^df0sI`Y)In)$uaNS8JB=<5bh~}NN<4d(yd73XqJ|JkT!`L8bMn1D(@wJu5Rb3nXqb&AVSuQCBMe4e}=xm)u6D! zUt=|nB;yHH<47}49KDjOh9SH5th{M+nSD0IC!UjMww~fOTf9v-NAlD70twQsCkd!_ z^JiuPp+ZSqK7ax)hAX+g%CyO|J$>AsaBTGK0d>3e8HKdVZAKV>CzO&u@~( z>}vt9z45a;lj5kR*tV4Ct@SZf22SHGRjVJKM44Bv(UcXB)iFmY@4c0_goOCsx`H_@$`p$=LZTTlNV7~A>fD@$?jvUVKG?f06DY<)v(7k3bbN&(j`#lU7 zH6heR5%-u-@qq$2nkiVDyq2%s!8HKmAG_`(*VU9+T0omI)FmVDCQ=PfPZ{)vorQX_KD!hlcIPT1yQvLf(ipm=0qA!~!0fqgR zq9>thA4s6;ymA!ft2S04JZ`MBKjl3f!h6Ty?;?j<9&GMM1KF+%g$DYPJR+$1AM!Lg zu8e&Xhj8UGY+1xe*MKe^$dA$(iT)w;cW-`38h3Ft1HJSOn@NtQawqSml0aXjP{SQy)Q!= z>&{RH+_Fl3yB(fE=|voQ77SYx3^c6MpD3NO9!2#-3^dUvI!MLtq93p9$K3}-BY)Z6 z$}qEi1}f$lkw_$6_q9$U(FSMc_lfvT#&Gv-A&-LOZQZvvRt*e=M8Yb4!YcBptLSG= z(Aj4A0^`kqvo%rdG5qL+68pire7Fjrkcfm@b-@+L$P+2ti!j^ar5k~G|33#w{wgkG z1vW;pMnU_wnyqZ*y$B(Ns?4YWj-rHBgD*0zAU%CIKE@NlW9QNMIyIm3h!Qc|uiN^(;JyTlybL5@XE~sVWs7`-<3~_+cQ)A0jElr0MLx#Yt8{ z`5duW?M$F~K!QTv?5GI8~LcKWo@n2?-S1nt7&ej1rpsO-|3U zIAutHwG|3mFmGd3d8NZIFihA##Z@f*54LK0F{1h_8mc-FDqIaa?fA?_jvyb9Z9~Q^ zGXKfp(GXl$q2U}|HSjR_lsAcIbiKkI#6!O5JCTq7i?Z~!;sBS@euD{83^YK9HL8eVSmn4{thkCZE zlj$;=>A{oiNG0%%M(G;QG5+T3L-~)p`_+r8AoKC#GIK+v;|;}S9bln#JGG2L^6mp^ z|Ei-BejJk71}ecSw`Ka~4E6sqWH4p?relE{)trJ|>ljdocQ*!p#EH#<#LTa&>+ef?AkP&6$yie5nDjVzL4~53Fc`I}!(c_;jzW3kBChVmM zlZei(XraLnuUI8J1^4PvOY&F~W70PD}Wg|X2n{@53sZIi{>L1gr? zqon)y6!CfdhzC1rn|km0!@KjoqZ8Z;uPP+auJdH32OQTrzYV*otS~swt3jyAL(a0A z6vCj_2jL2X2A23W8<;+)_Y;E#n^yunQ4{|tLnIm%AM}aLASarbn)Gox*Ytfu4%pDW2(zw9J5hp1~YdSS2X7$zazKJ0HY;7 zMLy@NQRBA^h;Ytux@M}2rb-*26*sKzv8l<9q~H9H4{;+6PDyrA0_@6v-L}Kn{U?p- zJMUpPsj{S}WXgNLn|%Qbd&5yh_fY}0oXlnhqcG@J=WW5QT*-s^hpAV_y%BY2ANe`z z#%>C>18y(zEBWS>bjgTsQld=#FFgF#h^WJy2P`IS3B-Wy>k_!0L4}SYACpKYn5JD8 z=qH*Jg?jzYT}_gk!|}C6w*wlQGM&3-k<#k zG7_m>ZsL#|pzpD-L26Y&r4t&{V!C&GD;8bqI%1pdI;W)wS3H0k569W&b)-L8)}4?94!`Bn5*hwgi;88svF+s_Q zP--O>!&cVL>$-)*>Eyp`2n|OW8ryj7ag+h&V@TK%Yd&G(0*BO&*btSt)$3&?jLWYo zr!1R5wFde~!o*f4{I^NabSQH8wtQ6QZTYe0+udCj!(J>)WQ6u1-bBXSPL3vvTtQB&)$RNbZfIbkN7RK) z%Rcb?KM)vRsHM{n+^d`CFCzMVB}?XU`@lFOCAA4IpzLAw+^IT(S*L7$n&?> z^ZkKZIAq-90o@-~PrnUcw}Ys3y^%1G?7#yQFq6FtuDX+xk2xQ515x zxEC_^hhhh?;;}!$S+F9|ZBgz#xuGyQvQDiP!gGQmBRAz6cprd zScCCWH-zNluYxg%9L^QTTSXuA+Bm*FWPaKa>&VSNo|v6!CT<(q-wJLjoU-1bPU%~~ z`JI-6eW=E$1W33_8`i5>M=VWU@T~`IpXed?Df}68g{>6@ z3Rg&(%kI1bETcx=$XlgaD7wqfMO<>%*dZ4eRW;b6kAL=zuvOcQF{A693Uxb7a9n&S zs#srKg*pJX8~#X|+8?P#9dH?9Wb@QZFL@i@{$=4bnqT0Cy^UV-j-C+U*_TKy$fuKV z0~XRmCY=QnHNx-ycqe;DuWurrS_uSQ6hY9XMlm+py72XGbmRAUTl-tu6F`AELabXo zzTUw|x0_cj=lkPc1N;=5K*OeBvXZ zohMZyJ<0DP3+*h|nUAb%vom9fw`@y3(69mqXE|uh^{T&k|L^tJjpIx7zeaB`7#sb7 zZB7Kdx<`Bdb)8+Lk?I3ud2!p|xmRbtUiqR-d}#fK7ruY^r{AOcJg-zonW7LWH?Y|g zrz%_3*WclTNNG$hhurm;8Wt&?#c!<}H&?rurzn!IuPxv?bQXp;O7DF(+=l?Gd80L2 zV0Lr~Lu*9Ab8|-Sx^u_lX*N4*dfgF;7U16JeNze95{K7vx_K0&eg{3#JOlgzTHj#w zFBS)a&>L!p_H&BG#bZ8__?V+KjvbN3f23nLax-l|r%mI?z;^P;SeO@Yx7PC~!7T_i z9J&aL%*>y*LG%GZ5K-Gc!j#8R92?_WxD;%bbqIB6x}i%T*i*wmVih;^LQlCVYxuUt zfvRcV|9}&_Tpy)QB%yIL8;gJp@<}SQl#ZNI8Dv2X1&t~!O!gO!Z&kOF2q=VZ7#41f zGT;|q;oVv74MZHsizl%08;rDP_TvCGa&cKBV@EpIBRc1YSsG(@_XMWH(@8kyBU`zR*g zd|+kIF4=I7O8W~x=pfUfX~%S7#hmqpUA*<1_sb|WSEGCZ^CsxogTf%kR2`bkUFl~y zwO&ohWAYQrQ9m!w3g&>rRTN6A=2G}n+DmIue0MfW15^BDO1CBCRw9itu*D)JR}rv;Is48Bzlu zCQ1pw;Ii4}K?XY9i>#FX07e&t&3!JSGw)d84bdZG~VUfVc(`M~<#GDclm! zi%gKKid#c63t|gc3Uep2+MC24b_R-5bPCX?r}qTDGm0}YVukkW!@DBL8vMcE3syr< zbBvz#O>}_O?fjqWcELe?4A3HLRe!_HI2$)5ZQ78md7 zuxGK6WF0WNf_5r7R{<^L5`#i}L82o=d=8Go8XL=n0I|n7@6^)tMP(`>s~T9r5TB%4 zgIw03^{S7Jbae%^6W98{s)xxS@(NxgCM4xycxCfN?P;6l>qV{Gny3zk*~F|ImAq_O zu!M$@g?Zu)G|?=+Tjf3{f_h9W1R7$V;2YYRr!2FFWs~pXkNiATzGqd*- z@nB%!?E80G36IbCvo-{{s?1s!EtC_`3I~EgXwxWCQJ8O^)}yoB+j7sI#s9i3|1VRb zN-ggZMOfSI?s?s|7&w)1`9|RHj{g-apIuxVUmQy`jxx|g#kR-3%AtJBy_ht6lRH^j zOPzRNJo;}366*A=@PTuSrq@~ZS^q(85NU#L@2rXZ-|OByCqtdDI(xNj%=hb(z1D3- z2`0vzR!Z+br)FByTh#4()!Zv&XSog6dp5$a!-{OT;bn_I__6JKjO#T~p8F78r*WA4<`2rIin`cw^@?i`C zvtEc{bbQ6oV|gX*7+KB89U4vpamub6VH8@&=4PibB|fI^U`%G)pIfX%^cJQSaPDJ- z;oQg|gEfYWeRHGsSx8lY-{%xZL0`AP6`a6Hezu5_S*Ixx{7%UfR~ zWWI>oy~}GxkLc3{WnCMq0o=qYp*>Q|mH08GPYRR#sz;dkga-dMrBAK(t+h?t$8T4= zls^4cK`@>XOgbrSQp@A#LE!j~vB;J~8xEJ}K#hkh%oUFs`BK^M2f&`e$8~B zGX>dc{-Q9eYQm2ushE@pH=~Q2+PI{mW8Uyx|3zRy5?X|8*TxEl>)&QEu@^$v6kLSD zQ#qt241$&as>Eu8R(=p^4~6qK{(@g*)TS=gC6+FAb_$G|BIq4AP5_O^mARh zI}-sWijTG#YA(K%O%&VvPrlUYa6Zc#R3a|BzQZ47U6bLG1L{iWzK~a7~If8Q~moEMC z0@UJLMMq5~8w~cwtKzr2!-RMBZ;yk8vU#uawhULpJ{3K2ta)o|J!farYz1btm6vD1(kFrhC=KEJqclarXa1*-`zkB-^OJ4j;oc^Vj4!N(mAV} zb!Su*pH+ud-g++SbSKh^W`Sb9=?C$3Opw)E$<^p{mPmk!DVH#>aIl}J^7~49E^OR! z!jVcmu82GV(V4!UzRMp54GjiO{?8P#YFts@gxRmGzetev;`ZI^jz8CRAK*Ccq4*=L zJpA!|!>x-_b=la3-W+u>GV9CcikL)Ebkq_Bu)1ctw0_9D0=&#KI(qpRsX+hRg(hl@ zqHtMUYRuDaN~}h8+h{w{=47rMLVdaf4VFiWH!xcnMc6|@fsC167;#;A`yxrKAwFMd zNG@KdfEAe&`WmGm9uGH*kt%ZMm-56%UZ1$17#q*gs?}R1?I>6CSH%KootZASogpoW z!}t6`zBX*aJzHB}q++avfMQsaB25V^yCieFubc{8Veh`Qjn(~TIF(}IhO=)oh-UK) zKijw8QDcM3j}Gc?&p)B({|@zResxk&T0gH>ubdUXcn!?U>gQWu#OD^?BPri_EVFS1 z!={Ep&WJx`~Ya<)vK`%S7}VWoU!?`6IT?OWURZF3n0}lPO5*42BqV@ta_* z(E-bNrJ+W}r%<=|Ha!HWOpcm=j(>9Hc1fj+l*o`PB}mpnQ?d(^iwI=#kL#O~Lu4jK zeSp7OzL(9FeO~>|*ct2=x;5WJ+A74GBY|ss0d`o;j>lnLI?{<)O!mZhV7|kWtjS`E z2eZ=Ef4yTKE&~a_9(lEVD1XI(!%5uD#;Vgbq9CZ&LquvSc;DX{Uaopnk6s3|LA3z}oZ$`z61AR; zjq2pT(I2s@=%}<>WQ_O$>%|3hj%NijDa*f0I9bQXfcKrlmFn)lv-tGHDdkKOK@AS%i zu-oOts^2R5CcIaQ%zRdUDSEHi+ndVeYN_$yy|Urp(TF(n|3CtEH#wL-_AbEB6KXZ? zi1L+X?&o-h)1TF(PZL=M#e0RSvB5q=tL9~CfGu=(@^6$#t=}g5!G+P2BTl*7QtU>wEs|*X54!LWL>T znA6!_t2{nfr`;XSnz#4v;N~|2G={^U!%)w=PRAPjlPyHpx`Lm+7zywfPqPb`au|sH z(@#<^Dlaf)Vj*MeXg822K*bW&SDxD$L^d?Daj%s$@e>J3xict+fWq^XR9Nx-bmtSN5u$*H?WC;6fdFx2c~V>>c}Ea-ePIY; zDf%B&Du~vZ?Fw4Qqh}pHi)*canrD;SYrnyWnPqil_hEvQ$1#~k+6sICf!4+#n6%?< zBEd$(cN4bwe1G#^_d)5Y4x(OxnJ@7~l+*UxGz)}N=Ux4Isc$E{To&0DKd9@;V(<9O z18YmSye9Z|zkmkLqn23ukV~9b%yLp$6LIki=q*BRBTvm!6qT2`=TLJcc zVB-xH->v?!9G{$Ki%r&wvd-S_gn;+M=!TjzJ^XTxKdd<{+Wwhq;t9G^lv(cLuVQup4|h?9U3x%ldO%+Shq67lVlv*p90} z^?@SNu_WFjO|zcz4lwhYz~Om40rHQAQcYOQa+hk5Psl?fTH z561NQ%Ui4s4bt(%BK{P6o4Y(e82CC_q$TEK7>92yb*q=t>aK(Xj`CCBz zjaouIhK_~YxB)fwWYeA}e?XcOA6A)O&P5RvY&Vp6jjs9rBatGH8hY?TQ9uQ@p9x-6 zbh)wGa>U8AHdurGU#ow0UuRZ5Bs+k?wM*cuo?JXMiQ{4SX)32-c6G8OF%h9gYC z!Dr&q0m*V0j(wjCY=~p z9NZX|A>^^s(iHove^0jyW1#|HTg=#a4Ex)%R@X#cQy;`@8M6WwX~PuCtlu5!)6HZC zs|~Xji#W$eZEn3LXJJ+njw_~7iz#GLglQhcW+ozlGHgjZ*vrVi`^AbQ*fB}_V?eR~ zO3l*)hqsHqeb!IjAsaFp;O!9x1VV+$B-96TT@$OP^@I7JKmXTV(s1J=!$us~ z_jxl^lsp7C5qwyC4VtxC9zV)<#ZG=NNM1PQ?Tk4^Q+68h-5XwTMm4 z1+*bhn^SvqDnAar*Oz;0r1vs|*wTIx%bIG25R1|1O8p1puaTuGPSNq}L4dmEN*BxZ zzg1}|#QL{+mSFC}G{BqCLt25mYez|NIIlV4smOgQyW+8d- zuesy8i6k*`KD`KGuRHhAqd4Dc=Rcio_pa18Bc)k=+k{lEJ0sAU9hq!mGwKUV@x-&N+YCe<;mT$f{yzZq`$gdZp# zHariOUAes`W^HrQJVqfAUAJX2PIKhM1swj=q22x_bMLF+)P0zk<3*FI?s@R3$xTgS zJ&`h8{pFH_prf-H9#s)sXe15$^S}Uu%-kE2!(=N#7|KcnDt!x*1?;#+qFI@B!H za5OS3+%H=^ej%!%1?d(v_$ zc?QP~hsZCTJ1wzR4@Xg+&b@B=SU*qHy$p^B6+Gzqx?R!RYq}HmGh=uCsE6!k4va(C z!fyApT5^%e&4o(9VGaM?gp0uIkU`1Sq@J^#Z%{e2B-nEi3%h$?eNq;ZdVsT?p33b! zZr=HppkM`qcVb7-<9*kqz@MOlS^(fABIHSD>ICVJ zFrI@nDpdhg0^O}-4glZoDJ3Pfmfn}G%Q0%{`1#h%LGU#kpUm#djYXe#(DMkpj4vJj zkV~usqz<}wZiqo%$Zk>b38V`_w@XFREW2MZgman39#-^j)#r{7ILTeU#5!`#hAKEQ zm>}?|D5?K!=kyl9x}fu8~yW&UQTO_*zpaJ*j##}PFN+#llM;KTEv$! zDHDrN>bpK;osyE4Q0=tKG$GoJ`iY*c{OJhpDo&d{Vtc%QS~Do*cMG1#ZlK>*&I7bZ zCkFfRvRk^mo=R$A*DBcUQ(OYqNB(xtT)ZTsvu*k?#%Ie}K;z~x`=31?vtHYAM<)i0 zH~RAX{^4z4I5Q0DKE4F!+d4hB{X^NyTQ(m+e)LN5PA!~&-bnK4Lhh@TG@Yz1FXi)2 z2_1q%z|aFrO#rtY2X|zY%((F5`cG3#fuP_z`rQS`Eh9q$?Da$IlzeX&w!JTM=VWd! ziA27h8EQ!Tf((PmDqX!x28MVGbFF`szu32PwG!w&ir7e6#HrXyg^}!pkW*5R+)?WV zsY#@s_N(f7`n#gH@Lj@kp7ZOe#Fmz#-@%PEqVMH_7MvE9(?4%L>Gtg7+yZ6zX@+Ko z0MvQ+kAfa9HC1cNaQ6@q*l!V#1nTK7)6(fDPT*M^Y#H68rV@dR8>iujuh>dlqeVPX z`+B>10-bql$Cr~-n3e~{HG`0NP>TuhMFK9rxHMjX8}3x%B5=y;E8-|OmADAt{9Kdn zQW@=^$?j#7>t4E)^Q)ItDWG3GKl#me zIPxYgFb)!ef8cycT8f_hz3^1%?&T($G2+-ug(I-S4mNt(gCfx#$vXLXDU$CW=M+** zBz(e^<@yL&-UmUTg` zI{a~ee(&1lm%kho_eq{r^v-Z6QV{Nf@~FM|CuxI=ow~A~PiI7&&rw1`y3uL;lG9~k zmD|E~t4p|z@8Y^a>pAp7T10Ogf@epS+vM$mnqn+Hfbt*(Ae5AB=9~N3=YNLAms?Bt zN>kWn-iaOig8HSJLMt||0b0+9^V_hk43PD98@KQF>^IH5mS}%vwF|PfT4NT6{E@9p zl+BHlsp?0^>wbENST{la3Uj{cP47us}8btoa=lW-QY*@+|ItK+9yo|taz?4%4ieVD)3bCPznASk+PwF3ng<_e9E zV;oZt9uV-cb7-4x?owLFA!?7hd<4f`OI&vnr0pvzwYJ`D>QP!yDTCvB0^=0P;P{1Y z#Sz~L-^CpVtd*_AcBI{8%gh3z381ZkT35hu6{aTn_{UuVSGt63%ZkeFLn(po39Frs zy}h%~DdrWhzPB-xmb~Mq`Z<=1x5zClAVcLlCU|`L#8k@(%8i|OypH!pGG-Q(8fa^9 zKU7ruP6Bk@!XmFW`|FMSPprJ-LcFPU7Q@IaAf(mBlZ{;CLX03yjqfS|nAqWh;6;no zj^%Xe{zl-;Lf&-TW;Ib`=gAMgDHyqT9Q2cA?GEKVzNsHBmggyH=RNETTN)?LCsvyA zbD{2RBI859Tz4k=oGaK8w%*h!QdzL*3H7nN`iwQ8ENeAe@XL|RQ?BK#losuB)kE%) zuyNH;@Uod3zg%4JVxbPG!H^%TP=IfkkXvMPfh!xl$7pajb99vO%6W*uak0l1(obrS zESjO?s@#sbXV;A?*xvqrQ%6N+v5E(dJ8_Pq2tW!<^eS?1{o=dGY(2AztKNyUv)!BF zC6zEd&OES+(*Xa5F3<2PEwC)pV;aq81CIB1#%Fj#fM(jW5@dy}^jv5i_4!`i?vz8F z2}&6_o4Yt2+w{)wEy*9|D7nx#-3Kt~vCm}Q?^ zY|M8*z}B7J^>V#Vz7-x&fX&bg-FAFnfARtTE?v}1IPmOZu{p1z#Ry~k=(RQ!BR|aH z!E4|QOnG#7dxU;lTl=wdI(o2^bDrY=yAE-dk;&(2_is0XnjnlBVRU+Id~vkX{(Z)H z>S7^8^ulLGt*#<&$C!8`;y^8kEHq;4Ikk=K-W=uWR$bq7?L$R