analysis and plots for draft results section

analysis_data/scratch.R

@@ -0,0 +1,34 @@
+library(tidyverse)
+main_csv <-"~/analysis_data/120725_unified.csv"
+main_df <- read.csv(main_csv, header = TRUE) 
+
+author_closer <- main_df |>
+  filter(comment_type == "task_description") |>
+  filter(author_closer == "True") 
+table(author_closer$isAuthorWMF)
+
+
+new_authors_summary <- main_df |>
+  filter(comment_type == "task_description") |>
+  group_by(source, AuthorPHID) |>
+  summarise(
+    task_count = n(),
+    first_task = min(week_index)
+  ) |>
+  group_by(first_task, source) |>
+  summarise(
+    new_authors_count = n()
+  )
+ggplot(new_authors_summary, aes(x = first_task, y = new_authors_count)) +
+  facet_grid(source ~ ., 
+             scales = "free_y",
+             labeller = labeller(source = c("c1" = "VisualEditor", 
+                                            "c2" = "HTTPS-login", 
+                                            "c3" = "HTTP-deprecation"))) + 
+  geom_col() +
+  labs(
+    x = "Date of first task",
+    y = "Number of tasks created",
+    title = "Task count by Author's first task date"
+  ) +
+  theme_minimal()

dsl/final_bivariate.R

@@ -16,6 +16,38 @@ data_summary <- dsl_df %>%
   ungroup()
 
 library(ggdist)
+priority_plot_summary <- data_summary |> 
+  filter(priority == "Needs Triage" | 
+           priority == "Unbreak Now!" |
+           priority == "High")
+priority_plot <- priority_plot_summary |>
+  ggplot(aes(x = week_index, y = proportion, fill = priority)) +
+  facet_grid(source ~ ., 
+             scales = "free_y",
+             labeller = labeller(source = c("c1" = "VisualEditor", 
+                                            "c2" = "HTTPS-login", 
+                                            "c3" = "HTTP-deprecation"))) + 
+  geom_col(position = position_dodge(width = 0.9), width = 0.8) +
+  scale_color_viridis_d(option='turbo') + 
+  scale_fill_viridis_d(option='turbo') + 
+  geom_vline(xintercept = 0, linetype = "dashed", color = "black", linewidth = 0.5) +
+  geom_text(
+    data = subset(priority_plot_summary, source == "c1" & week_index == 6),
+    aes(x=week_index, y=0.6, label='Opt-out deployment'),
+    size = 2.5) + 
+  labs(title = "Triage priority proportions for new tasks by week created",
+       x = "Weeks from feature deployment",
+       y = "% of items tagged",
+       fill = "Priority Tag") +
+  theme_minimal() + theme(legend.position = "top")
+priority_plot
+ggsave(
+  filename = "120825_triage_priority.png",
+  plot = priority_plot,
+  width = 12,    # inches
+  height = 4,   # inches
+  dpi = 600     # high resolution
+)
 
 data_summary|>
   filter(priority == "Needs Triage" | 
@@ -27,33 +59,51 @@ data_summary|>
   scale_fill_viridis_d(option='turbo') + 
   facet_grid(source ~ ., scales = "free_y") + 
   geom_point() +
-  labs(title = "Triage priority proportions for new tasks by week created",
+  labs(title = "Proportions of Triage Priority by Week",
        x = "Weeks from feature deployment",
        y = "% of items tagged",
        color = "Priority Tag") +
   theme_minimal()
 
-dsl_df |>
+c1_ttr_plot <- dsl_df |>
   filter(priority == "Needs Triage" | 
            priority == "Unbreak Now!" |
            priority == "High") |>
   filter(week_index >= -26) |>
+  filter(source == 'c1') |>
   ggplot(
     aes(
       x=as.factor(week_index),
       y= TTR/168, 
-      color=priority,
       fill=priority
     )
   ) + 
-    facet_grid(source ~ .) + 
+  facet_grid(source ~ ., 
+             scales = "free_y",
+             labeller = labeller(source = c("c1" = "VisualEditor", 
+                                            "c2" = "HTTPS-login", 
+                                            "c3" = "HTTP-deprecation"))) + 
     geom_boxplot(outlier.shape = NA) +
     theme_minimal() + 
-    coord_cartesian(ylim = c(0, 112)) + 
+    coord_cartesian(ylim = c(0, 52)) + 
     geom_vline(xintercept =27, linetype = "dashed", color = "black", linewidth = 0.5) +
-    scale_color_viridis_d() +
-    labs(x = "Weeks from Release", y = "Time to Resolution (weeks)", title = "TTR by Task Creation Date and Triage Priority")
-
+    scale_fill_viridis_d(option='turbo') + 
+    geom_text(
+      aes(x= 25, y=45, label='Opt-out deployment'),
+      size = 4) + 
+    labs(x = "Weeks from Release", 
+         y = "Time to Resolution (weeks)", 
+         fill = "Priority Tag",
+         title = "VisualEditor Time to Resolution by Triage Priority") +
+  theme(legend.position = "top")
+c1_ttr_plot
+ggsave(
+  filename = "120825_c1_ttr.png",
+  plot = c1_ttr_plot,
+  width = 12,    # inches
+  height = 4,   # inches
+  dpi = 600     # high resolution
+)
 
 dsl_df_long <- dsl_df %>%
   pivot_longer(
@@ -101,7 +151,7 @@ weekly_summary <- dsl_df |>
     median_comments_before_resolution = median(n_comments_before)
   )
 
-ggplot(
+tasks_created <- ggplot(
   weekly_summary,
   aes(
     x=week_index,
@@ -109,7 +159,11 @@ ggplot(
     fill=isAuthorWMF
   )
 ) + 
-  facet_grid(source ~ ., scales = "free_y") +  
+  facet_grid(source ~ ., 
+             scales = "free_y",
+             labeller = labeller(source = c("c1" = "VisualEditor", 
+                                            "c2" = "HTTPS-login", 
+                                            "c3" = "HTTP-deprecation"))) +  
   geom_col(position = position_dodge(width = 0.9), width = 0.8) +
   geom_vline(data = weekly_summary |> filter(source == "c1"), 
              aes(xintercept = -29), 
@@ -133,9 +187,35 @@ ggplot(
              aes(xintercept = -3), 
              linetype = "3313", color = "black", linewidth = 0.5) +
   geom_vline(xintercept = 0, linetype = "dashed", color = "black", linewidth = 0.5) +
+  geom_text(
+    data = subset(weekly_summary, source == "c1" & week_index == 6),
+    aes(x=week_index, y=120, label='Opt-out deployment'),
+    size = 2.5) + 
+  geom_text(
+    data = subset(weekly_summary, source == "c1" & week_index == -33),
+    aes(x=week_index, y=120, label='Opt-in Testing'),
+    size = 2.5) + 
+  geom_text(
+    data = subset(weekly_summary, source == "c2" & week_index == -12),
+    aes(x=week_index, y=20, label='Deployment Announcement'),
+    size = 2.5) + 
   theme_minimal() + 
-  scale_fill_viridis_d() 
-
+  scale_fill_viridis_d() + 
+  labs(
+    x = "Weeks from Feature Deployment", 
+    y = "Count of Tasks Created", 
+    title = "Phabricator Tasks Created by Week and Author Affiliation",
+    fill = "Task Author Affiliated with WMF?"
+  ) + 
+  theme(legend.position = "top")
+tasks_created
+ggsave(
+  filename = "120825_tasks_created.png",
+  plot = tasks_created,
+  width = 12,    # inches
+  height = 4,   # inches
+  dpi = 600     # high resolution
+)
 
 outcome_summary <- dsl_df |>
   group_by(source, isAuthorWMF)|>

dsl/rq1_plots.R

@@ -79,19 +79,33 @@ declined_summary <- dsl_df %>%
   mutate(proportion = count / sum(count)) %>%
   ungroup()
 
-declined_summary|>
-  ggplot(aes(x = week_index, y = proportion, color = task_status, fill=task_status, group = task_status)) +
-  geom_smooth()+
-  scale_color_viridis_d(option='turbo') + 
-  scale_fill_viridis_d(option='turbo') + 
-  facet_grid(source ~ ., scales = "free_y") + 
-  geom_point() +
-  labs(title = "Task status for new tasks by week created",
+task_status_plot <- declined_summary|>
+  ggplot(aes(x = week_index, y = proportion, fill=task_status,)) +
+  facet_grid(source ~ ., 
+             scales = "free_y",
+             labeller = labeller(source = c("c1" = "VisualEditor", 
+                                            "c2" = "HTTPS-login", 
+                                            "c3" = "HTTP-deprecation"))) + 
+  geom_col(position = position_dodge(width = 0.9), width = 0.8) +
+  scale_fill_viridis_d(option='magma') + 
+  labs(title = "Task Status (as of February 28, 2025) by Week",
        x = "Weeks from feature deployment",
        y = "% of items in status",
-       color = "Task Status") +
-  theme_minimal()
-
+       fill = "Task Status") +
+  geom_vline(xintercept = 0, linetype = "dashed", color = "black", linewidth = 0.5) +
+  geom_text(
+    data = subset(declined_summary, source == "c1" & week_index == 6),
+    aes(x=week_index, y=0.9, label='Opt-out deployment'),
+    size = 2.5) +
+  theme_minimal() + 
+  theme(legend.position = "top")
+ggsave(
+  filename = "120825_tasks_status.png",
+  plot = task_status_plot,
+  width = 12,    # inches
+  height = 4,   # inches
+  dpi = 600     # high resolution
+)
 
 
 

mgaughan-rstudio-server_31713063.out

@@ -1,17 +0,0 @@
-1. SSH tunnel from your workstation using the following command:
-
-   ssh -N -L 8787:n3439:48979 mjilg@klone.hyak.uw.edu
-
-   and point your web browser to http://localhost:8787
-
-2. log in to RStudio Server using the following credentials:
-
-   user: mjilg
-   password: eTkReY42xI/86STHG+4f
-
-When done using RStudio Server, terminate the job by:
-
-1. Exit the RStudio Session ("power" button in the top right corner of the RStudio window)
-2. Issue the following command on the login node:
-
-      scancel -f 31713063