We are happy to announce our freakin fast new search.
In the past weeks we have been working hard to get a faster and more scalable search into Memex. And we did it.
Before: Downloading 20k websites took around 25h and a single text heavy website sometimes took minutes to index.
After: On a medium powered computer with a proper internet connection, importing 20k websites now takes less than 90 minutes.
And even with this many documents, a 10k words document takes 500ms to index.
Yes. Swoosh so fast!
What you need to do get the new search:
Nothing. Just let the computer be idle for 20s for the migration to start.
It is interrupted every time you start using the computer again, so the migration does not interfere with your work.
Something didn't work. HELP!
No worries, your old data will not be deleted until we are sure that everyone runs with the new search engine.
In case you experience issues with the migration, please let us know here in the forums, we will help fixing it with priority 1.
Thanks to the people building the new search engine:
A big shout out goes to our team members Samir Sekkat for conceptualising this technology, and Jon Poltak for implementing it.
And also to the rest of the team for cheering, supporting, testing and improving the workflows.