Rebirth

Inspiration

The history of the internet is very important but very difficult to access, and it is also fun and nostalgic to revisit old versions of websites that you can no longer normally access. Although tools such as the Wayback Machine exist, they aren't easy to use. We want to create a modern, simple and accessible tool for accessing old archived versions of websites.

What it does

Our app allows the user to input a URL to any website, and our app, using the Wayback Machine API, will search for any archived versions of the website, display these to the user and allow them to select a desired dated archive to return to. Once the user has selected their archive date, the app will display the archived site in its former glory. For example, you can go back to YouTube in 2015, or old Reddit. The app can also summarise the selected website using AI. It will provide details such as notable design features, historical context and impact. There is also a "Search History" tab, where your previous searches of archived sites are stored in a database for you to return to at any time - or delete if you no longer need that URL.

How we built it

We first created a shared project using Microsoft Visual Studio Live Share.

Using python and HTML, we created a web app using flask. We started by simply creating a form allowing the user to submit a URL. We then accessed the Wayback Machine API, allowing us to search and retrieve all existing archives of a given site. These are then displayed in a dropdown form for the user to select one date they would to like to see their chosen website from.

We later added a database, using flask-sqlalchemy, to store previous searches which can be returned to, or deleted. The website was cleaned up and given a modern feel with CSS-stylesheets. Finally, we incorporated AI into our app, giving helpful summaries of websites, as well as recommended websites to the user, populating our homepage with more information for new users.

Challenges we ran into

Getting the correct URL for a given archived site on a given date was challenging - we encountered many bugs, and issues with the Wayback Machine API not getting the correct link and not displaying a web page. Actually saving and storing the date as part of the logs in our database was also challenging, as we had to get the date from the URL.

Accomplishments that we're proud of

Very proud of how seamless and easy our website is to use, as well as how clean and modern it looks thanks to fantastic use of CSS from Miguel. Proud of the convenience of having a search history that allows you to return to websites you previously visited between sessions. Happy with our use of LLMs and AI to produce summaries and more information