Monthly Community Update -- September 2022

Hi all,

This is our September community update! Read last month’s update here.

August was a month of steady progress, both in terms of shipping product improvements, and in terms of the no-code movement growing. Two things in particular we’re excited about:

  • We’re partnering with Notion to offer perks to entrepreneurs. This was one of the most-shared announcements we’ve made to date, which we think reflects a general excitement about the no-code ecosystem maturing and popular companies working together to make the lives of no-code founders easier!

  • We’re very happy to be a sponsor of the upcoming Nocode Summit, Sept 29th and 30th!

We welcomed seven new people to our team this month:

  • Rosie, joining us as a recruiter
  • Aaron, Ryan, Spencer, and Nidhi, joining us as engineers
  • Jerrel, joining us as an engineering manager
  • Alicia, joining us as an account executive

We’re continuing to improve and streamline our hiring process. We made a number of changes this month to how we hire for Technical Product Support Specialists to make the experience faster and friendlier, which is a role that we often try to fill with experienced Bubblers from the community, and would love you to apply. Our full list of openings is here.

Changes we made this month

If you’ve visited your home page recently, you probably noticed we’ve overhauled the design: our goal was to make it more attractive and easier to use, especially for people who have a lot of apps. We also updated our support center and contact us pages.

We were very excited to release a major performance improvement to rendering pages on Bubble as an experimental feature, and plan to release it for all apps once we are fully confident in its stability.

A few other product improvements:

We also made some important behind-the-scenes improvements, including moving our internal data pipeline to dbt, which should make our internal product and business analytics faster and more robust.

This month in numbers

  • New conversations via bug reports or support@bubble.io: 9,988 (up 10.4%).

  • Average first response time to messages: 1h 32m during business hours (down 12.3%)

  • Average response time to messages: 1h 28m during business hours (down 14.3%)

  • Tickets closed by the engineering team in the past 30 days: 148 (up from 95)

  • Average days to closure for tickets closed by the engineering team: 21 (down from 48.9 days)

  • Average days to closure for high priority tickets: 7.3 (up from 6 days)

Things on our minds

We are still spending a fair amount of effort building internal tooling for our teams to prevent abuse of our platform, including people who are using us to build phishing websites. We believe that abusive actors led to the incident two days ago where Avast and other antivirus tools were briefly flagging apps hosted on bubbleapps.io as phishing websites. Following the prior incident last month, we put in effort to segregate the domain names we use to host assets, and as a result, this incident did not affect apps hosted on custom domain names. We also believe the relatively fast recovery was due to some of the changes we made, which made it easier to identify the harmful apps as being on separate subdomains from the rest of our applications hosted on bubbleapps.io. We are continuing to invest in our tooling to efficiently moderate the Bubble platform, and expect to make significant progress here over the next month.

What we’re currently working on

Following up on our latest post about pricing, we are continuing to do technical work on capacity and auto-scaling. Our focus right now is on making improvements to the way we track capacity usage to ensure that the metric is stable with respect to changes we make to our code: we have had incidents in the past where a bug we release or a change we make significantly impacts the capacity usage of applications, and want to reduce the degree to which capacity usage fluctuates over time prior to enabling auto-scaling. This stabilization work should be behind-the-scenes: the total capacity usage of applications should not significantly change, and instead should become more stable moving forward.

Other ongoing workstreams:

  • Our work on an improved version control interface continues: we have working code for many of the features in the initial release, and are hard at work on finishing it, as well doing the design and planning for follow-on releases

  • For the overhaul of our network architecture and infrastructure, we are starting with improving our credentials and configuration management, and aim to have our first production system using the new configuration management by the end of the month.

  • In order to increase our overall reliability, we are working on the way our alerting and observability systems work. We’ve launched a proof-of-concept of it and are playing with it internally.

  • We are working on our new components system to improve how easy and fast it is to build beautiful apps on Bubble. We are mostly done with the initial development, and aim to enter internal testing and QA in the next two weeks.

  • Our push to migrate code of CoffeeScript continues; we are now down to 18.3% CoffeeScript in our main codebase. Typescript is now up to 3.2%, with the remainder being Javascript

Wishing everyone a fantastic fall,

Josh and Emmanuel

42 Likes

Amazing update as always!

So looking forward to these one!

Thanks
Zubair

1 Like

Hey @josh,

Great updates! Love to see the design rehaul on the home page and pages of the main bubble app :slight_smile:

1 Like

Special thank you for both of these. :raised_hands:

1 Like

Really looking forward to this too!

1 Like

Thanks for the update. Unfortunately, some of the changes to Repeating Groups completely broke our paid features yesterday, and we’re now doing damage control trying to make sure we don’t get too many cancellations. From Bubble support:

"We recently rolled out some improvements to the way that we guess the number of cells in infinite scroll repeating groups which makes the scrolling process a bit smoother in most cases. This had the unintended consequence of causing nested repeating groups like this one with no minimum height set to load too many cells for both the parent RG and the interior RGs. "

We use nested Repeating Groups in some of our most-used paid features, and this is causing RGs to load all items at once (there can be 300+), which is crashing the app and massively spiking capacity usage. And there’s literally nothing we can do about it because it was a change Bubble decided to make without notifying us, without allowing us to test, and without us pushing a Bubble version update. Beyond upset right now. We’re just starting to gain traction, and now I’ve got to worry about app-breaking changes being pushed without the opportunity to test first.

6 Likes

Music to my ears. Finally! Exciting :heart_on_fire:

Thank you @josh and Bubble Team

3 Likes

Gosh - I love this!!

I’ve just joined Bubble.io and I’ve never read an update like this before.

You talked about Bubble being used to build phishing websites.

Such a level of transparency with any company is unheard of in my experience.

Keep it up. Things like that will keep me hanging around and encourage me to share my experience.

8 Likes

I completely agree with you on this. Bubble needs to run a stable production framework that doesn’t receive any changes unless the community is given prior notice and preparation.

If they release an update to a dev/staging cluster then users would have an opportunity to report the bug and Bubble could fix it. Rapid iteration is very good and can be very bad if not done well. I think this policy should be given further consideration.

Two seemingly good options:

  1. Lock production/live apps from receiving any changes without said notice
  2. Allow apps to run on the current version OR the previous version of the Bubble platform. Then if they push a change that breaks, individual apps could choose to rollback to the previous version of the platform, report the bug to Bubble and switch back to the current version when the bug was patched/resolved.
6 Likes

Always good to be in the loop. Kudos for that.
See you in No-Code Summit.

Thanks,
Ali Imran

2 Likes

Thanks for the update. Could you share what the followi g goal will allow?

“For the overhaul of our network architecture and infrastructure, we are starting with improving our credentials and configuration management, and aim to have our first production system using the new configuration management by the end of the month.”

Is it for stability or will it open new posibilities?

1 Like

Replied to @samnichols and @bubble.trouble here: What is Bubble's policy on pushing changes? - #4 by josh

1 Like

Both. Goals are:

  • Increased scalability and reliability for the main cluster: we’d like to split it into multiple shared clusters instead of a single one
  • Improved security
  • The ability to host apps in multiple geographies without requiring users to upgrade to Dedicated hosting
33 Likes

fantastic Bubble, always innovating and making it easy for us.

Wow! This would be FANTASTIC! Can’t wait…

13 Likes

Any ETA when this will be available?
Choosing an own AWS region would be helpful. World is changing pretty fast and most of my clients wants to store their data close by. It’s a bottleneck at the moment.

17 Likes

:+1: :+1: :+1: :+1: :+1:

10 Likes

That’s AMAZING!!!

Awesome, thanks for the update!

Much appreciated @josh!

Really like you are prioritizing the UI. Here’s a bit of feedback and improvement suggestions to the frontpage amongst other things. Would love to do a second round overhaul myself as a UX nerd but hope your team can use it :slight_smile:

3 Likes