diff --git a/docs/graph/postgresql.rst b/docs/graph/postgresql.rst
index 0cb0c38..5d8c17e 100644
--- a/docs/graph/postgresql.rst
+++ b/docs/graph/postgresql.rst
@@ -1,98 +1,98 @@
 Setup on a PostgreSQL instance
 ==============================
 
 This tutorial will guide you through the steps required to setup the Software
 Heritage Graph Dataset in a PostgreSQL database.
 
 .. highlight:: bash
 
 PostgreSQL local setup
 ----------------------
 
 You need to have access to a running PostgreSQL instance to load the dataset.
 This section contains information on how to setup PostgreSQL for the first
 time.
 
 *If you already have a PostgreSQL server running on your machine, you can skip
 to the next section.*
 
 - For **Ubuntu** and **Debian**::
 
     sudo apt install postgresql
 
 - For **Archlinux**::
 
     sudo pacman -S --needed postgresql
     sudo -u postgres initdb -D '/var/lib/postgres/data'
     sudo systemctl enable --now postgresql
 
 Once PostgreSQL is running, you also need an user that will be able to create
 databases and run queries. The easiest way to achieve that is simply to create
 an account that has the same name as your username and that can create
 databases::
 
     sudo -u postgres createuser --createdb $USER
 
 
 Retrieving the dataset
 ----------------------
 
 You need to download the dataset in SQL format. Use the following command on
 your machine, after making sure that it has enough available space for the
 dataset you chose:
 
 .. tabs::
 
   .. group-tab:: full
 
     ::
 
       mkdir swhgd && cd swhgd
-      wget -c -q --show-progress -A gz,sql -nd -r -np -nH https://annex.softwareheritage.org/public/dataset/graph/2019-01-28/sql/
+      wget -c -q --show-progress -A gz,sql -nd -r -np -nH https://annex.softwareheritage.org/public/dataset/graph/latest/sql/
 
   .. group-tab:: teaser: popular-4k
 
     ::
 
       mkdir popular-4k && cd popular-4k
-      wget -c -q --show-progress -A gz,sql -nd -r -np -nH https://annex.softwareheritage.org/public/dataset/graph/2019-01-28/popular-4k/sql/
+      wget -c -q --show-progress -A gz,sql -nd -r -np -nH https://annex.softwareheritage.org/public/dataset/graph/latest/popular-4k/sql/
 
   .. group-tab:: teaser: popular-3k-python
 
     ::
 
       mkdir popular-3k-python && cd popular-3k-python
-      wget -c -q --show-progress -A gz,sql -nd -r -np -nH https://annex.softwareheritage.org/public/dataset/graph/2019-01-28/popular-3k-python/sql/
+      wget -c -q --show-progress -A gz,sql -nd -r -np -nH https://annex.softwareheritage.org/public/dataset/graph/latest/popular-3k-python/sql/
 
 Loading the dataset
 -------------------
 
 Once you have retrieved the dataset of your choice, create a database that will
 contain it, and load the database:
 
 .. tabs::
 
   .. group-tab:: full
 
     ::
 
       createdb swhgd
       psql swhgd < load.sql
 
   .. group-tab:: teaser: popular-4k
 
     ::
 
       createdb swhgd-popular-4k
       psql swhgd-popular-4k < load.sql
 
   .. group-tab:: teaser: popular-3k-python
 
     ::
 
       createdb swhgd-popular-3k-python
       psql swhgd-popular-3k-python < load.sql
 
 
 You can now run SQL queries on your database. Run ``psql <database_name>`` to
 start an interactive PostgreSQL console.