Today we shipped a tiny improvement (we hope) for the editor.
The editor now uses the browser's spell-checker and highlights words that may not be correct. It is now also possible to right-click within the editor and access the browser's default context menu.
Today's update brings a few improvements to the "Articles" list, as well as a choice for the default view mode (Dark / Light / System) and two bug fixes.
Articles list
The "Articles" list provides a quick glimpse into all articles that exist within a workspace. It is sorted alphabetically by default. Due to quite a few requests, we added a new view that will allow you to group those articles by category. The view preference is stored in the current session and will eventually reset to the default sorting.
Default view mode
Dark mode, light mode, system preference... everyone has different preferences and some change during the day. The default mode for the frontend view (where you and visitors view your workspaces) has always been light mode. Visitors were able to switch to dark mode if they preferred.
That switch still exists and won't go anywhere. But we added a new option to the "Appearance" settings that allow you to set the default mode. Visitors will still be able to use the toggle to switch the mode however they prefer. This is stored in the browser's local storage, and if a preference was already set by a recurring visitor, changing this value in the settings won't overwrite it.
Bug fixes
We also fixed a bug that would throw a server error when a draft article had a shadow category assigned. Draft articles don't appear in the frontend, but we missed a little where clause in our database query that would allow them to sneak in when shadow categories were loaded. This is now fixed.
Another bug we fixed was causing paginations to miss track of which page you were on. So when going back and forth, at some point, an error 404 would be thrown. Really weird, but yeah, that is now also fixed.
Today's update brings a lot of smaller improvements to the backend, as well as a new feature and a bunch of bug fixes.
Filelist
First off, we added a filelist to get an easy overview of images that were added via the editor. You can see how often files are referenced (used within articles) and also delete them if they are not used anymore. You may also copy the link to an image to re-use it in other articles without uploading it again.
With this update, we also separated the export. The workspace export in the workspace settings does not include files anymore. Files can now be exported directly from within the filelist. The export will also include attachments that were added to articles.
In a future update, we will also include article attachments in the filelist for easy reference.
Business Lifetime plan
We also added a "Business Lifetime" plan for people who don't want to subscribe yet to another service. This plan is currently reduced, sort of a "early-bird" special if you will. So grab it while you have a chance.
More about this new plan can be found in our dedicated blogpost.
Bug fixes
We also fixed 7 little bugs, including an issue that would appear in the team list after having more than 15 people on your team. The pagination should now respect a list of more than 15 members.
Also, when an article was indexed for search (= can be found via the search in the frontend) and then disabled again, it would still be searchable and an error 404 (not found) would be thrown when trying to open the article. Our indexer now removes articles that are not enabled anymore.
Many little UI changes
We also changed A LOT of tiny things across the whole backend area. From colors, to spacings, to font sizes - almost every part of the backend was slightly updated or modified.
One last thing...
It is now possible to directly chat with us from within the backend. No need to go to our dedicated help center article anymore. The chat can be enabled and disabled at any time.
Today we added two small features, new integrations and also fixed a few bugs.
Read-only role
One highly requested feature after we introduced frontend authentication was a read-only role. Essentially a role that was only allowed to view the frontend (after authenticating), but without any access to edit or even view the workspace backend in any way whatsoever.
This is now possible. Roles can be set to new (invited) users, as well as existing users.
After-save redirect
After saving an article - either as a draft or published, you'll be automatically redirected to the category overview. It is now possible to change this behavior. As an alternative, you can now choose to stay in the article edit view after saving. This is a global setting per user, so if you have multiple workspaces, this change will apply to all of them. To change this setting, go to your account settings at https://eniston.io/settings and find the section "Redirect mode".
New integrations
We also added 3 new integrations:
Brevo Conversations
ChatNode AI
Tawk.to
Bug fixes
And last but not least, we fixed a few bugs. Two of them were pretty annoying if you encountered them.
When changing the title of an article from "A" to "B" and later back to "A", the slug of that article would be changed from "domain.com/category/a" to "domain.com/category/a-1". If you manually linked to that article, a 404 would be thrown. This should be fixed.
We also fixed a bug when creating a new article within a category (not via the + symbol in the sidebar). The active category was not pre-selected and upon saving the article, it would be stored in a different category (the first one that was created within the workspace). This should now be fixed as well.
Finally, right? There are now 3 different methods to prevent external access to the frontend of a workspace:
Restrict via IP address
Restrict with global password
Restrict with email and password frontend authentication
1 and 2 were already around for a while. They do what they promise, but are not always the best solution. Restrict with email and password authentication brings a lot more security and control benefits.
Frontend login with email and password
One of those benefits targets teams, or any scenario were multiple people with different accounts worked on a workspace that was restricted with a global password. When one of the people left a team, they could still use the password to access the workspace. So a new password would be needed and everyone on the team had to be informed. That... sucked, obviously.
Now, when a team member is removed and frontend authentication is activated, this former team member is not able to access the workspace anymore.
In a future update we plan to integrate more finer control so it will be possible to restrict certain user groups from categories and articles, both in the backend and frontend. We also plan to implement third-party authentication so our systems can connect to external systems and use JSON Web Token Authentication (JWT).
To enable this new authentication method, log in to your account and go to the workspace settings at eniston.io/base/settings#access. As with the existing global password screen and category password screen, the background color can be changed in the "Appearance" settings (Header Background Color).
One of the things we had planned for a while was to rebuild the "Docs" theme. The initial design worked fine for most cases but code-wise, it was bulky and not a joy to maintain. Today, we finally shipped an updated for the "Docs" theme (and also minor refinements for the "Default" theme).
In general, both light and dark mode now feel more modern and snappier. At least that's our impression. :)
Pinned navigation
Optionally pin the navigation bar to the left side
There is now a new option to "pin" the navigation to the left side of the screen. This can be toggled by admins. By default, the navigation remains close to the content. With the pinned navigation, the appearance reflects many popular documentation examples and sticks to the left side of the browser window.
Toggle categories
We now also added the option for users to toggle categories. By default, the tree view shows all articles within their categories. Users can now optionally collapse categories in the tree view. This is probably a welcome addition for many users. The states are stored in the browsers local storage.
As of today, you can change the slug of a category.
What is a slug?
A slug is a set of words or numbers used in a URL for a specific webpage, such as example.com/this-is-the-slug. The slug is used to identify specific posts or pages, or in this case a category.
Slugs are usually auto-generated based on the title. And as with all web urls, they should (and usually are) be entirely lowercase, not contain spaces and be unique per workspace. Spaces are usually replaced by a hyphen or underscore. Punctuation marks are generally removed.
Slug preview
How to change slugs?
Slugs can be changed within the category settings after the category was created. Simply click the "gear" icon and find the section "Category slug" in the panel.
In a future update later this year, we will also make it possible to change article slugs.
In case you ever wanted articles to appear in multiple / different categories - without duplicating them - we got something for you:
Shadow categories
Show articles in different categories without duplicating them.
Article settings
How it works
When assigning shadow categories to an article (via the article settings), this article is connected to these categories. In the frontend (where you view articles), your article will appear in those shadow categories, in addition to the main category. To prevent duplicate content, the article detail view is still opened in the main category. Your article exists only once and will be found only once via search.
Releases integration
We also added an integration for "Releases", one of our own products. If you have an existing Releases account, simply copy the workspace identifier (found in the widget settings) and paste it when creating a new integration in eniston.
If you don't have a "Releases" account yet, you can save 40% off when using the code ENISTON during checkout. Show your product updates directly in your help center with Releases.
Today we released a small bugfix that mostly impacted newly created accounts. An error was thrown when attempting to create a new article without first creating a category. There has to be at least one category in a workspace to be able to create an article.
Apologies for the bad start to those this bug impacted.
A small feature many of you have asked for is now available: clickable tags.
When adding keywords / tags to articles, it is now possible to click on them on the frontend to get a list of other articles that share the same keyword.
In a future update we will improve the keyword selection in the backend to make it easier to re-use existing ones.
Today we shipped a few new features and lots of minor UI changes across the backend. Most of the UI changes are a reflection of our recent survey. More updates to improve the usability are coming soon.
Duplicate articles
It is now possible to duplicate a single article within a workspace. Set the new (or same) category, define a new title and duplicate. All article settings and attachments will be duplicated.
What about workspaces? We received quite a lot requests via email if it was possible to duplicate an entire workspace. We planned to ship this feature today as well, but it's not quite ready yet. We plan to release this mid October. Stay tuned.
Highlight important resources
Today's update also includes a new section called "Important resources". If you want to directly point your users into a certain direction of most relevant articles or pages, you can now create up to 6 sections that will be shown on the front page of your workspace.
After releasing our new search, we received a few dozen emails asking for the old search back. In general, this is something we won't consider as we spend a good amount of time building new stuff like that. New things are (mostly) different and maybe need some time to get used to. But we heard you and we agree that a stellar search is crucial for a knowledge base.
We knew how to get the results we expected and were before, but it was not clear to many of you. That is a clear indicator that we didn't do our job – building a new and better search – good enough.
The old search is back, and even a little better than before
We spent the last days implementing the new old search the way we should have done in the first place. But it was a "quick fix" done in a hurry when we initially did it. It is now more maintainable for us, and brings more relevant search results to you and your users.
We apologize for the trouble some of you had and we appreciate the feedback.
