Skip to main content

Sprint 37–39: The many things before, during and after our launch

This week we finally went live with Colloq. Here’s what happened during the weeks leading up to our launch and the few days after it.

Preparing for the big day

On the infrastructure side, we finished the server backups and now have continuous, automated backups for all our servers in place. We also were able to setup Syncthing, a decentralized distributed file sync application, after a pull request from Tobias was approved and included into their latest release. This allows us to add or remove servers individually, by having an automated setup process, which also works well with our service discovery.

To catch server errors that our users might get, we implemented an automated email notification system, which sends us a grouped summary of all exceptions for the last hour. This also aligns well with our philosophy to only use 3rd party services if necessary.

Eventually, we enabled page caching for all pages without cookies to decrease the server load of our application servers and improve the page speed for everyone even further.

On the backend, we fixed an issue with our cities table not having a name but only a localized name in the database. This solves a problem in our city autocomplete field that was not showing a city name, but only county and country information. Now users are able to see the city name of every city that is in the database.

We also added an Atom feed for events and implemented sending of newsletters from the application to our users.

Front-end-wise, we worked on some visual improvements for the Event Cards that now show the Event Series if available, improved the Coverage link cards, as well as the Speaker and Session card layout.

The cookie notice banner had an issue with setting the cookie expiry date correctly and we made it more bulletproof that it remembers your choice for longer now.

Doing a quick review with the amazing aXe Developer Tools, we identified a couple of accessibility issues that we were not aware of and fixed them shortly before the launch.

The launch day and the days after

The launch itself couldn't have been smoother for us. As planned we published our blog post and sent our first newsletter to our users. The servers did do their job as planned and we did not have any serious issues with our infrastructure.

Since then we received a ton of feedback from our users and already shipped quite a couple of bugfixes and improvements, while there are some more in the pipeline. For example, we added a clearer explanation for the allowed Markdown syntax, fixed a double escaping bug on our blog, fixed a hover issue on the organizer option menus, and improved various copy text on the site. The fix for our contact form and the deletion of event articles that unfortunately didn’t work correctly was shipped shortly after it had been reported.

Thank you

Last but not least we want to thank every one again for your great support! After all, Colloq is a platform for people and we couldn't provide this service without you.

Now we will continue to work on new features, but don't hesitate to send us your Feedback. It is highly appreciated.


Want to read more articles like this? Subscribe to the category What’s new or the feed for all our articles.