diff --git a/README.md b/README.md
index 86b82da..e1e95af 100644
--- a/README.md
+++ b/README.md
@@ -9,7 +9,8 @@ Feather is a RSS reader client. It can:
 * and do all that while using less resources than a web browser showing a blank page :D
 
 That's right, using an innovative technology known as "plain files", Feather allows you to read your feeds in any file manager that exist or will exist in the future.
-See [usage](#usage) to admire it working.
+
+![Reading an article](images/read.gif)
 
 Note however that Feather is meant to be used alongside a RSS reader server, meaning:
 
@@ -26,37 +27,31 @@ All demonstrations are done under Linux with GNOME Files, but a file manager is
 
 Feed categories are directories, and each file in these directories is an article.
 
-![]() TODO
+![Browsing articles](images/browse.gif)
 
 #### Sorting articles
 
-![]() TODO
+An article file's modification time is the article publication time; you can sort by publication date by sorting the files by modification time, or sort by feed by sorting by filename.
 
-search + sort
+![Sorting articles](images/sort.gif)
 
 #### Searching
 
-![]() TODO
+![Searching articles](images/search.gif)
 
 Tip: if you have nested categories, search "html" to list all the articles in the category and its sub-categories in the same view.
 
-#### Reading an article
-
-![]() TODO
-
 ### Marking articles as read
 
-![]() TODO
-
 Deleting an article will toggle their read status (will take effect on the next synchronization to the server).
 
+![Marking an article as read](images/markasread.gif)
+
 #### Handling read articles
 
-Surprisingly, the read articles can be found in the trash. If you restore them before the next synchronization, it's be as if nothing happened.
+Surprisingly, the read articles can be found in the trash. If you restore them before the next synchronization, it's be as if nothing happened. However, restoring the file after synchronization will not work if you want to mark the article as unread.
 
-![]() TODO
-
-Or, if you'd rather show read articles in the Feather reader directory directly, you could add to your configuration file:
+Instead, if you want Feather to also track read articles, you could add to your configuration file:
 
 ```toml
 # Grab both read and unread articles into the local directory
@@ -65,9 +60,9 @@ server.only_sync_unread_articles = false
 html.filename_template = "{% if unread %}☐{% else %}☑{% endif %} [{{ feed_title }}]\t{{ title }} ({{ published }}).html"
 ```
 
-Note that if you delete a read article from the reader directory, it will be marked as unread and the file will be recreated during the next synchronization.
+Now both read and unread articles will be stored in the Feather reader directory, and if you delete a read article file, the article will be marked as unread and the file will be recreated during the next synchronization.
 
-![]() TODO
+![Marking an article as unread](images/markunread.gif)
 
 ### Updating with the server
 
diff --git a/images/browse.gif b/images/browse.gif
new file mode 100644
index 0000000..dfd854c
Binary files /dev/null and b/images/browse.gif differ
diff --git a/images/markasread.gif b/images/markasread.gif
new file mode 100644
index 0000000..9c74ce5
Binary files /dev/null and b/images/markasread.gif differ
diff --git a/images/markunread.gif b/images/markunread.gif
new file mode 100644
index 0000000..41c6633
Binary files /dev/null and b/images/markunread.gif differ
diff --git a/images/read.gif b/images/read.gif
new file mode 100644
index 0000000..efc7c05
Binary files /dev/null and b/images/read.gif differ
diff --git a/images/search.gif b/images/search.gif
new file mode 100644
index 0000000..5128033
Binary files /dev/null and b/images/search.gif differ
diff --git a/images/sort.gif b/images/sort.gif
new file mode 100644
index 0000000..68a4b50
Binary files /dev/null and b/images/sort.gif differ