Is there a risk of losing my work by developing my data model only in the browser memory?

Answer

A thin-client (browser) Git client affords the developer the flexibility to commit work to the code repo while not requiring a desktop installation. While browser failure resulting in work loss is rare, a good practice to avoid any work loss is to commit / push code often to personal or development branches.

This approach is consistent with the other browser-based Git clients such as Microsoft’s Visual Studio Code for the Web:

When working on a remote repository, your work is saved in the browser's local storage until you commit it. If you open a repo or pull request using GitHub Repositories, you can push your changes in the Source Control view to persist any new work.

The alternative would be to require a thick client or terminal plug-ins. This approach carries its own set of complexities related to security and configuration while not eliminating the potential of work loss should that thick client suffer a failure as well. If a customer decides they want to use a thick client, they can. VSCode connected to AtScale repositories is supported, with only the AtScale UI serving as the AtScale developers' environment with syntax and dependency checking.

Was this article helpful?

0 out of 0 found this helpful