diff --git a/README.md b/README.md
index eecdfdd..dd4b9c1 100644
--- a/README.md
+++ b/README.md
@@ -19,10 +19,12 @@ triggers the supported site badge, but other Forgejo instances should work too.
 - Medium: Title, author, and date published fields are autofilled from the handy [`NewsArticle` JSON-LD schema](https://schema.org/NewsArticle)
 included on every page.
 - Substack: Title, author, and date published fields are autofilled form the handy [`NewsArticle` JSON-LD schema](https://schema.org/NewsArticle)
-included on every page. The supported site badge may not show for custom domains (i.e. not a `*.substack.com` url), but autofill 
+included on every page. The supported site badge may not show for custom domains (i.e. not a `*.substack.com` url), but autofill
 should still work.
 - Youtube: Title, author, and date published fields are autofilled from either the `watch7-content` metadata or the [`VideoObject` JSON-LD schema](https://schema.org/VideoObject).
 
+***WARNING: All data will be cleared when you close the extension's popup. If you are really proud of a note and want to come back to it later... save it somewhere else I guess?***
+
 ## Wait, what is this even for?
 
 I have a page on my site called the [sharefeed](https://eleboog.com/sharefeed) where I share links to things I think are interesting.
@@ -36,4 +38,4 @@ author: string? // The author of the linked page.
 publishedDate: Date? // The date in which the linked page was **originally published**.
 accessedDate: Date // The date in which the linked page was **accessed by me** (i.e. the date of this entry itself).
 note: string // A message to be attached to this link, usually detailing why I decided to share it.
-```
\ No newline at end of file
+```
diff --git a/popup.js b/popup.js
index 4794443..50dfd1d 100644
--- a/popup.js
+++ b/popup.js
@@ -32,10 +32,10 @@ function parsePage() {
     const youtubeMetaElem = document.getElementById('watch7-content');
     if (youtubeMetaElem) {
         // url fix
-        entry.url = document.querySelector('link[itemprop="url"]').getAttribute('href');
-        entry.title = document.querySelector('meta[itemprop="name"]').getAttribute('content');
-        entry.author = document.querySelector('span[itemprop="author"] link[itemprop="name"]').getAttribute('content');
-        entry.publishedDate = document.querySelector('meta[itemprop="datePublished"]').getAttribute('content').slice(0, 19);
+        entry.url = youtubeMetaElem.querySelector('link[itemprop="url"]').getAttribute('href');
+        entry.title = youtubeMetaElem.querySelector('meta[itemprop="name"]').getAttribute('content');
+        entry.author = youtubeMetaElem.querySelector('span[itemprop="author"] link[itemprop="name"]').getAttribute('content');
+        entry.publishedDate = youtubeMetaElem.querySelector('meta[itemprop="datePublished"]').getAttribute('content').slice(0, 19);
     }
 
     // github scraping