Branching could be useful in different situations, some more relevant than others, but on big factories with a team with several developers, it could be very useful.A situation that occurs frequently is when we receive a change request which requires changes on core entities, this will break the environment , and will affect other developers that are working on a different feature, but are prevented of testing/developing without pain since the environment is broken.So it would be great of having an way of creating a branching per pta of a solution, were the developer could make the change that will break the environment isolated, finishes the development, could test it isolated again (remeber that i've mentioned a branch of the solution), if everything is ok, perform the merge of the branches, create a new solution and schedule the publish.Im mentioning PTA was a way of doing it, it could be by branch, if it was possible to work on different espaces on a context of a branchIf i was a smoker i would quit smoking with this feature :)

Problem: Personal Environments cost OutSystems unnecessary AWS compute time, but deleting them frustrates developers who lose their apps unless they manually export them. Idea: Make Personal Environments ephemeral but with automatic daily app snapshots, so they can be safely deleted sooner without hurting the developer experience. How it works 1. Daily App Snapshots Automatically snapshot all apps/modules once per day. Store snapshots in cheap object storage. Only snapshot OML + metadata (default: no DB data). Developer never loses their apps again. 2. Faster Environment Recycling Only do this if needed to offbalance the data usage in point 1 Reduce always-on uptime from 7 days to 4 days. After 4 days of inactivity: Kill the AWS compute Keep the daily snapshots 3. Auto-Restore on Spin-Up When a user requests a new Personal Environment: OutSystems spins up new infra Automatically reinstalls the most recent snapshot Developer resumes instantly with their apps already there Benefits For OutSystems AWS savings (compute reduced from 7 to 4 days). Storage cost minimal vs. compute cost. Personal Environments become fully elastic. For Developers Apps are never lost. Personal Environment deletion becomes painless. Better trust, better onboarding, better learning experience.

The default behaviour of datetime variables in Reactive apps is to convert from UTC to local time. This is a great in a lot of use cases, but hard to work with if a user want to see data in a different timezone.In this forum question , some 'hacks' are proposed:Use datetimes as textCompensate at server side for the difference in client timezoneThe prefered solution is to let developers set the config for this automatic conversion. The default behaviour to convert is good, but make an option to overwrite the default behaviour.

A ranking system for the Mentor Program where mentors earn points based on sessions they conduct. As their points increase, their ranking improves, helping us easily identify and connect with the most active mentors.

To add a link or a way to quickly open the Event Handler Action from Properties / Events of UI Widget. This can be extended to On After Fetch of Fetch Data From Server or UI Aggregrate. This will improve the developer experience to quickly open the Client/ Screen Action that assigned as Event Handler or After Fetch Event rather than scroll up and down Elements Tree For Button and link, we can quickly double click the Element in the canvas to quickly access the OnClick Event Handler already, however, this should be good if there's a similar way for other Widget like Dropdown, Switch, Custom WebBlock with Events, Data Fetch Aggregate...

I suggest excluding the static entities from the counting of Application Objects , as saving it leads to spaghetti code, and using it does not indicate the complexity or magnitude of the project.

Because sometimes you already know your timer execution has gone bad and you just want to stop it.

When building RAG pipelines in ODC against structured documents, the four native chunking methods treat all input as plain text. This works well for simple text stored in entity attributes but creates problems for documents that have structure: fenced code blocks, Markdown tables, and nested headings. A Markdown-aware chunking option would let developers preserve that structure during ingestion without needing a custom External Logic workaround. Three behaviours would cover most cases: heading path tracking so chunks carry their section context, code block atomicity so fenced blocks are never split across chunks, and table preservation so tables stay whole. The upstream Forge ecosystem already produces Markdown output via components like OmniDoc2MD . A native Markdown chunking method would complete that pipeline naturally. This is not a replacement for the existing methods. It is an additional option for teams whose documents have structure worth preserving.

It would be nice if we can make AI mentor target a specific application or have a list where we can enable/disable certain applications to be synchronized. In our environment we often work on 1-2 apps and want to see if we created additional technical debt. However we have to wait a very long time because of the extent of our landscape.

When moving from DEV to QA it is asked to provide site properties or use the default ones. It would be nice to do this for REST and SOAP urls as well. We had a situation where the WSDL (and we didn't know) was pointing to PRD. We changed it on DEV to so that the intergraded SOAP was pointing to the development version. When deploying to QA we considered that this would point to DEV again. But it took the URL from the original Url WSDL. So it was pointing to PRD. When deploying from DEV to QA the siteproperties are shown and can be defined. It would prevent errors to do this for consumed integration REST and SOAP URLs as well. Especially for SOAP. For rest it is possible to define the base URL and choose DEV so that when deploying to QA it's not suddenly PRD. For SOAP it seems not to be possible to provide a base URL Example