Ok, Iâll check to fix that then 
Well spotted! Iâll take a look.
@Chris75 Small note: when you quote a message, try to use Discourseâs native quoting function rather than a screenshot. 
As the forum is international, messages are automatically translated in both directions, but thatâs not possible with an image. Plus, screenshots are often harder to read on mobile.
ok, got it!
For Windows users, I suggest this modification, a small arrow that appears on the right:
And which becomes an arrow on the left once you reach the end:
In addition, now when hovering over the filter area, the vertical scroll wheel scrolls horizontally (without Shift).
Finally, a thin bar appears when hovering or focusing.
The fix will be included in the next version of Gladys
Thanks for your feedback @GBoulvin, I can reproduce it and Iâve fixed the bug
This will be included in the next release of Gladys!
I think thatâs perfect!
Thank you!
Hello everyone,
the presentation is very good, I just have a small remark: when you create a device and add a feature to it, all features are deployed, I would have preferred that they were closed, we would then see a kind of list and when you select the desired feature, it expands. Because currently, there is a lot of information and I get lost. 
Hi @Psoy, thanks for your feedback
I actually think that having all features open by default allows you to immediately see what the dashboard will look like.
If we closed everything by default, youâd need an extra click to show each feature. That would weigh down the experience and make it much harder to quickly spot what youâre looking for.
I donât have the same approach as you. For me, it was simpler to hover over the title of the feature Iâm looking for and then open it, rather than scrolling through all the expanded features. Especially since the titles get lost in the definitions when you want to scroll quickly. Also, as I noticed thereâs a small arrow on the left of the title that lets you close or expand the feature, the idea would be to remember the state of the featureâs expansion. That way, everyoneâs happy . For my part, I would close all the features and, when I create a new MQTT device, the features would always be closed and I would open them as needed. I donât know if thatâs complicated to do? 
I admit Iâm not convinced, sorry 
and the idea of memorizing the deployment state?
Thanks for your answers. And I know youâre doing everything to make Gladys as pleasant as possible, so no worries
Thanks Psoy
On the idea of memorizing the state: technically itâs feasible, but Iâll be honest about why I prefer not to do it. If everything is open by default, itâs a deliberate choice: you get an overview of the device at a glance, without wondering what you donât see. As soon as you memorize a folded state, you lose this guarantee, and you add a layer of complexity (persistence, special cases, new features coming alongâŠ) for quite limited gain.
The design philosophy of the project is rather to assume simple and consistent choices, the same for everyone, even if they donât suit 100% of each person, rather than multiplying behaviors. So I prefer to stick with the current behavior.
But honestly, thanks for the quality of your feedback, itâs exactly the kind of exchange that makes you think 
Local startup in 700ms 
I found a small bug in the MQTT integration.
When you set a custom topic and then delete it, the device takes all the values from all MQTT topics. I suppose Gladys still assigns it a custom topic but at the rootâŠ
It would be better if, when the field is reset, Gladys no longer took it into accountâŠ
Issue created:
