Development

musdb

• Added a tool for the AI-aided detection of displayed subjects in images for tagging
  • Has to be explicitly turned on on a per-collection basis, as it makes sense only for certain types of objects (like paintings, drawings, photographs)
  • Usable within the tagging overlay on object editing pages
  • See also: blog post about the feature (German)
• The maximum length of contents in the data field “edition” of literature entries has been extended to 50 characters
• Uploaded PDFs may now be up to 40 MB large
• New command line option to reset all permissions of a user account to the default provided by their user role
• New event type: “Changed”

nodac

• Added AI-generated suggestions for tag definitions and translations of tag names in a sidebar
  • This is also re-used to identify duplicate tags

Dissemination

• Poster presented at CoRDI 2025 (Archived) on August 27, 2025, in Aachen: “To Educate or to Enforce – The Case for Underhanded Methods to Improve Research Data”
  • PDF
  • Abstract / Zenodo

This content is licensed under a Creative Commons Attribution 4.0 International license.

Image credits: “Devanagari_letters.svg” is licensed under Creative Commons Attribution-Share Alike 4.0 International, by बडा काजी, and retrieved via Wikipedia Commons.

This content is licensed under a Creative Commons Attribution-ShareAlike 4.0 International license.

As always, many thanks to everyone who’s made that possible!

Image credits: “Telugu consonants.svg” by Psiĥedelisto, licensed under CC BY-SA 4.0, via Wikimedia Commons

This content is licensed under a Creative Commons Attribution-ShareAlike 4.0 International license.

As always, users with browsers set to using Tamil as their main language will now automatically see the Tamil translation of the site navigation etc. Others can select to navigate the site in Tamil using the language selector at the top right.

Image credits: “Tamil Community – Worldwide.PNG” by Coppercholride, via Wikimedia Commons, licensed under Creative Commons Attribution-ShareAlike 3.0 Unported.

This content is licensed under a Creative Commons Attribution-ShareAlike 4.0 International license.

Say, a museum’s previous database used actor roles over an event structure to express who created an object. As such, the museum entered that the object has a linked actor X that is linked to the object as a “main creator” and a linked time Y marked as the “creation time”. During the import, these roles (“main creator” and “creation time”) are then translated to museuem-digital’s event types to form an event: The object was created by actor X at the time Y. This works, because the terms “main creator” and “creation time” have been matched to the creation event type.

If a term is not yet matched to a corresponding value of a controlled list in museum-digital, the importer will simply abort the import. On the one hand this is a way to uselessly require resources for an import that cannot be completed anyway. On the other, it is tedious. One recognizes yet unmatched entries only one by one.

A Small New Tool

A small new tool, available at concordance.museum-digital.org, makes the process a bit less tedious. Users can upload all the import data from a given field (e.g. the actor roles) – one a line – and check whether they are already matched using the concordance lists or not.

For entries that are not yet matched, the tool will offer selection boxes to perform the matching using the graphical user interface. Once all entries have been matched, one can then generate the relevant lines of code to enter the missing entries to the concordance list upon the click of a button.

While simply checking and extending the relevant open source lists should be trivial even to most non-technical users, this way is certainly more convenient. Importantly, it also removes the need to run the import multiple times until one does not encounter errors caused by unmatched entries anymore. And, well, it certainly is also more convenient to match to regular human language values than to the internal IDs of the target values.

The concordance checker’s MIT-licensed code can be found here.

This content is licensed under a Creative Commons Attribution 4.0 International license.

By 10:34 the error was fixed by the technicians in the data center and all systems at museum-digital are operational again.

We are sorry for the downtime and looking into ways to better mitigate similar issues in the future.

This blog post will first deal with the issues of the old design. A summary of the aims of the new design follows as they were often formulated as a direct response to the shortcomings of the then present. Finally, a tour of screenshots through the new design concludes the post.

The new page design will be published on the night from Tuesday to Wednesday.

For those seeking only a quick overview there also is a small playlist on Youtube.

Problems of the Old Design

Inconsistency

A natural problem of the old design was its growing inconsitency. Quite a number of features and even some sections of the frontend were developed after the last re-design (early to mid 2017). As new features were added, they often did not reuse existing styles and instead used slightly different reimplementations of the same.

Hidden Features or “To View or to Act?”

The likely most consequential issue of the old design lies at its very foundation. In 2017 we had done the redesign aiming to create a design that was very much focused on presenting museum objects in a clear and attractive way.

To underscore the issue of clarity, as well as for their then increasing popularity and the posibility to implement one simply even for users of the Internet Explorer (see below), we chose to go with a stream design. Some sidebars existed for presenting additional information, but the aim was to always clearly focus only one thing. While the design gave quite some space to gaps between different sections to improve focus, it ironically reduced our freedom to label buttons (as can be most evidently seen in the settings bar for searching objects).

In summary: Many features were – at times intentionally – hidden. And thus users did not find them or even learn that they existed.

Internet Explorer

This is not necessarily an issue by itself, but a self-limitation that has now become unnecessary. Was our aim to keep the page at least somewhat usable using Internet Explorer 10. Microsoft’s Internet Explorer did, until its demise, not support many of the newer web standards and trying to keep the page simple to maintain while usable in the browser meant that we could only use older styling options (e.g. CSS grids were not supported).

Lines

Especially during the re-design we noticed that the old design had overused lines. Lines to differenciate different sections, lines to form tiles, lines around buttons. While it had not necessarily occured to anybody I personally have talked to (or they did not say so), a comparison with the new design reveals that the number dividers and separating lines effectively lead to cluttered user interface.

Basic Principles of the New Design

Users Want to See, Read – and Then Do Something About It!

As stated above, the frontend has had quite a number of hard to find features – especially interactive ones like a the option to compare different objects or a watch list (the latter even being hidden behind an option in the now obsolete settings page) and export options. Often, they were hidden by being styled as normal entries in a list. This list in turn could either be found at the very bottom of the page (objects) or in a sidebar. This stems from the idea that users of the site primarily want to see and read things and should not be burdened with anything else unless they are explicitly looking for it.

The most radical change of the new design is our leaving that assumption behind. The new design is instead based on the assumption that people want to see and read things – and then do something about them. Be it to save them, to compare and further investigate them, or to share them.

Buttons (styled as such) offering options like the export of a data sheet, subscribing to a feed, comparison, and the watch list are hence now available immediately below the headline of any page offering such options. Whenever the screen size permits, these buttons also feature text explaining their functionality instead of being made up entirely of icons.

Do What Feels Natural

The old design featured some idiosyncrasies. To take the most blatant example: Right next to the page title there was a search bar. The position of the search bar at the top of every page suggests it being a general search bar for all sections of the page. In reality, it was a search bar only for objects. The new design now features a general search bar in its place.

In a similar vein we tried to identify unintuitively positioned or superfluous links. Some features (e.g. the maps for displaying a museum’s location or the locations of all museums represented in a given instance of museum-digital) were integrated into the pages linking to them (the institution’s page or the institution page respectively. They are now accessible as tabs, allowing users to quickly navigate back after viewing the map.

Don’t Repeat Yourself

The old design featured quite a number repeated content. Where possible we tried to identify such duplicates and offer only one, clearer way to access a certain information or visualization.

Hovering over a tag e.g. made links to the timeline and map visualizing objects linked to that tag accessible to users immediately from a given object’s page. These links were removed in favor of letting users click on the tag, see the objects linked to the tag, and then navigate to the timeline or map from there. On the one hand this more clearly communicates what is actually visualized (the objects linked to the tag rather than the tag itself), and on the other it allowed us to remove a flickering and likely unintuitive interaction.

Shapes and Positions Over Lines

This speaks for itself: While the old design used straight lines as dividers a lot, we tried to communicate the separation of different elements primarily by their positioning, size and shape in the new design.

A Tour Through the Re-designed Frontend

Start page

Object Overview: Display options

Extended search

Object overview: Facetted search

Object Overview: List Results by Museum

General search

Collection overview

Museum overview

Institution pages

Object Overview: Interactions Visualization

Object Overview: Tag Clouds

Object Page

All these objects can be viewed together on museum-digital:global.

museum-digital:musdb has supported EODEM since February. But the usefulness of a standard is determined by how widely it is adopted. Thankfully, there is now a second collection management system to implement EODEM: Zetcom’s museumPlus.

You can learn more about EODEM on the project’s website. See also the very insightful notes from the project’s co-coordinator Rupert Shepherd.

Over the weekend, we have added a new feature to the project page, www.museum-digital.org: a calendar for trainings, meetings, and the like. The events for this calendar are pulled from various shared calendars via the respective iCalendar files and compiled into lists of the upcoming events.

While the primary source for events thus far is the calendar compiled by the German association for museum-digital, the museum-digital e.V., it also includes events in English. We can thus schedule the meetings ahead of time with meeting URLs set ahead of time without having to write a blog post every time. On the other hand, (publicly) scheduled events will be kept for sure.

And thus, publishing the calendar also means that this time around, we will surely do better and continue the series of monthly user meet-ups on every Tuesday of a month, 5 to 7 p.m. consistently for longer than we did last year.

This content is licensed under a Creative Commons Attribution 4.0 International license.

To allow administrators and users to get acquainted with the updated design and new features ahead of time, a preview follows below. The update will be distributed on January 11th.

A slight redesign

Sometime in late 2020 or early 2021 – when the reworked dashboard was released – we introduced a new, different design to musdb. While the old design had sidebars with a margin to the window border, the dashboard’s sidebar goes all the way to the windows left end. Where the old design positioned all page contents (minus sidebars and navigation) directly on the background, the dashboard features clearly defined boxes for each section of a page.

Bit by bit, pages that had undergone major updates (e.g. the institution-wide settings page and the image editing page) or were newly added altogether (task management; calendar) also received the new page layout. That way, we could slowly phase in the new design and hopefully managed to warm up users to what musdb would look like in the future.

But keeping two different page layouts side by side also comes at the cost of a harder maintenance and (obviously) a less consistent user experience. With the update, the “new” page layout will hence be extended to all pages.

A few additional improvements beyond even what the dashboard suggested have been made however. Sidebars are now used for displaying additional, directly usable information gained from a given entity’s data much more frequently (e.g. a copy-pastable address block of a contact is now displayed on the contact / address book page). And sidebars of editing pages now (almost) always come with an indicator displaying where the user currently is (“collection”) and the ID of the given entity. This hopefully allows users to better reference the entities down the line – especially when integrating musdb with other applications like Nextcloud (see below).

The clearly distinguishable boxes for each section of a page are now also used on pages for adding or updating entities. Again, to allow users a quicker grasp of where they actually are, these pages also come with much more visible headlines.

In other news …

In November 2022 we introduced maps on which the location of a museum or an object’s closer, unnamed location (e.g. a finding spot of archeological objects) could be determined by a simple click on the map. The colleagues from Baden-Württemberg requested to be able to enter geo-coordinates directly into an input field that interacts with the map, as they already know the coordinates they were to select. The maps thus now come with a button on the top right that allows the user to open a dialogue, in which the location can be entered using pastable geo-coordinates.

User-Defined Reports in musdb

Thus far reports in musdb were exclusively pre-written and provided together with the rest of the software. But in the end, museums are often bound by local regulations or already have forms or reports that may be generally used for a given purpose. Users can now define templates for report formats themselves and generate reports based on a search result, an exhibition (and its objects) or a loan (and its objects) independently. This may e.g. be used for automatically generating loan contracts.

To define a report format, one needs to hold the user role “museum director” and navigate to the institution-wide settings page. At the bottom of the page, one can upload a report template with placeholders marking the spots where object information is to be filled in by the system.

To simplify the implementation and improve security on the server side, only plain-text reports may be uploaded. HTML may be the most useful format for textual information with formatting; CSV for tabular information.

Timed Generation of Reports and Exports

One feature that was often requested, but thus far hard to implement, is to enable museums to generate exports automatically and without user input. This is now possible, both for XML reports and the new custom report formats.

Timed reports and exports are configurable on the institution-wide settings page. Each timed report requires the setting of:

• a start date (when should the first report be sent?)
• an interval (weekly, monthly, annual)
• a selector; usually a search query, written in the query language for searching objects
• mail address of a recipient

Note: As musdb is used by many museums together, we had to set some limitation on this feature. If an export file size exceeds 10 MB (which is also a size that many mail servers simply would deny for attachments), the configuration for the automatic report is automatically removed and a warning mail is sent to the recipient.

Literature entries

Probably the aspect of musdb with the most requests for improvements is the handling of literature entries. Almost any request for making literature entries interoperable with other software (e.g. Zotero for bibliography management or library catalogues) essentially requires a field for defining the type of a given literature entry. The same goes for most common citation standards: Within a citation style, the way of citing for books differs from that for articles, for webpages, or for archival material.

We have now added that rather essential field and – as that is possible using the new field – display a BibTeX representation of the literature entry in the new sidebar of literature pages.

User-specific defaults for adding new Objects

Museums have specializations, and so do people. It’s not rare for people to almost always enter objects of e.g. a given object type, especially if they are working within the context of a project focusing on a given collection (“Digitize all paintings of the museum”). Similarly, all users in the museum likely use the same units for values and measurements of objects.

To speed up the data entry in the case of such fields with unchanging contents, users will now be able set default values for “direct” text fields of an object. Unfortunately, setting defaults for links [e.g. to collections or spaces] and repeatable fields is much harder to implement and not yet covered by this update. Default values for the form for adding new objects can be set in the personal settings.

Take note that defaults can only be set for fields that are displayed on the object addition page. The most basic and generally required fields aside, one can determine which fields are available on the object addition page in the institution-wide settings.

Loans

To be able to better represent the process of a loan in a museum – from the request to the discussion with insurers to the final sending of the objects – we have now added a concise but hopefully reasonably complete checklist of the steps a loan takes within a museum in the sidebar of the loans page. The checklist covers the most common steps within a loan lifecycle and allows simply marking a progress in working on the loan. The last user to update a given entry in the checklist is displayed if a field has been updated at one point to allow following the progress later on.

We were also notified of a very obvious, but thus far overlooked case: Loan requests that are denied. We added the missing field to cover this status of a loan.

Finally, it is now possible to links loans to exhibitions. All loans of an exhibition can be listed together on a tab of the respective exhibition page.

To return to the checklist for a moment: One of the more noteworthy selectable points in the loan checklist is “metadata exchanged”. Adding this point may be opinionated, but hints at the next steps. In November 2022 we added the option to import loan object information following the upcoming EODEM standard. We hope to be able to implement an EODEM export before the update is pushed to the production instances, so that at least museums using musdb can handle loans with minimal duplicate data entry.

Exhibitions

As mentioned above, loans can now be linked to exhibitions. A new tab on the exhibition page allows listing all loans that happened in the context of the given exhibition.

The list of objects of an exhibition is now also redesigned. When linking an object with an exhibition, it is now possible to enter the exhibition room in which the object will be displayed. If this information has been entered for the objects of an exhibition, the list of objects of that exhibition will be grouped by the objects’ locations.

Integration with Nextcloud

Keeping with the theme of allowing for a deeper integration of musdb into the actual everyday work of museums, we have added the option to integrate musdb with a museum’s Nextcloud instance. If the Nextcloud integration is activated, a new widget will be accessible in the sidebar of most editing pages (e.g. for loans).

This widget displays a reference ID of the entity (e.g. LOA-000000005 for the loan with the ID 5). If this ID is present in a folder or filename on Nextcloud (say, there is a folder for everything concerning the loan, which will then be named something like “2022 Loan Brisbane [LOA-000000005]”), musdb can identify the folder or file as belonging to loan and list it in the widget. If the ID is present in a folder name, the folder contents will be listed in musdb.

For this integration to work, musdb connects to Nextcloud using WebDAV (unfortunately we needed to use some properties exclusive to Nextcloud’s and likely OwnCloud’s WebDAV interfaces, which makes our integration incompatible to other storage solutions that also use WebDAV like Google Drive). And to connect via WebDAV, it needs the information to get an authorized access.

To configure the Nextcloud integration, one hence first has to set the base URL to an institution’s Nextcloud instance on the institution-wide settings page. This only has to be done once per institution and simply gives musdb the information necessary to locate the Nextcloud instance.
Once a base URL for the Nextcloud instance has been entered, the username and password for Nextcloud (ideally an app token [can be generated in Settings > Security in Nextcloud]) can be entered on the personal settings page in musdb. Once those are entered, the Nextcloud integration is activated.

Institution and contacts pages

The layout aside, institution and contacts pages have only been minimally changed. A simple but maybe useful small widget has however been added in the sidebars of these pages: An address block to quickly copy-paste the address to e.g. a letter head.

Objects

Finally: Objects. Objects have seen the addition of a lot of new fields, mainly for administrative purposes.

On the “administration” tab of object pages, one can now reserve an object. If an object is currently reserved or will be so in the next week, an indicator will appear in the sidebar.

For logging an object’s history within the museum, we added a number of logs for:

• Damages to an object (Tab: “Restoration”)
• Conservation and restoration treatments for an object (Tab: “Restoration”)
• Scheduled Checks (Tab: Administration). These checks cover e.g. condition checks, but also audits of whether the object information in musdb is complete. This section comes with a notification that can be sent if a check is upcoming.

Deaccessions can similarly now be covered in musdb.

Again reaching for the most practical applications, we have finally implemented linking an object’s actual / permanent location as a “space” rather than simply identifying it using a text field. This way, it is now possible to search for objects that are not currently in the location their expected permanent location.

We also added some simple text fields that are often present in imports and suggested by the Canadian Heritage Information Network’s software requirements checklist. Namely: sex of the object (for biological specimen), the color of the object, and form of the object.

Post image: Courtesy NASA/JPL-Caltech.

musdb

New fields

Other page

Batch updating object information

Instituion-wide settings and adding new objects

Notifications

In other news

• The calendar feature (accessible under the puzzle symbol in the main navigation) can now display tasks or make them subscribable via WebCal (thus implementing a “reminder” as had often been requested)
• PDF of all linked information is now in A4 and uses a two-column layout
• Ukrainian translation
• “Simple A5 PDF” now covers inventorization fields on rear side

md:term

• If two actors have been joined and one has an old links to the page of the actor now deleted, one is now referred onwards to the new, single actor entry. The same works with transferrals between vocabularies (an actor that was transformed into a tag, etc.).
• Ukrainian Translation

Frontend

• Ukrainian Translation
• JSON-based settings for specific institution pages have been removed
• Bulk download of object images
  • An overlay with a message from the museum can be displayed during batch downloads (see above)

CSVXML

• Almost completely rewritten
• The served page now is completely static and all checks and conversions run directly in the browser. This way, no uploads actually happen and the application is completely uncritical to the server’s security. On the other hand, this allows for installing CSVXML as a progressive web app and using it offline.
• We also added some explanatory texts did small updates to the design of the page. A footer now links to the source code and offers to refresh all cached contents of the page (this may be useful when visiting the page after a long time, as the whole application is cached in the browser for offline use).
• A new check also checks for the file encoding. A warning is provided if the data does not appear to be UTF-8-encoded.
• CSVXML is now released open source under the AGPL.

Importer

While the individual parsers for the different export formats are updated very often, the core scripts of the importer are very stable. November 2022 however came with a large update to these core sections, as more categories of data that had not before been covered by the importer (many of them new) can now be imported:

• Contact information (e.g. for object owners; loan partner institution)
• Object’s movement log
• Minimum and maximum temperature, humidity, and lux of an object
• Loans
• Events / Appointments

In terms of the parsers, we extended the LIDO parser to cover the new fields suggested by the upcoming EODEM standard for exchanging loan object information.

The meetings will take place next week on Tuesday, and following that on every first Tuesday of a month, at 5 p.m. and are open to anybody interested in museum-digital. Generally, we will have a very rough agenda, where we will first discuss the updates and new features of the last month in museum-digital’s tools and then open up the meeting for general discussion and questions and answers. A short summary of the most recent features will be published here on the blog.

The first meeting will take place on Tuesday, August 9th, 2022, at https://meet.jit.si/museum-digital-meetup-202208.

This content is licensed under a Creative Commons Attribution 4.0 International license.

For the start we have compiled primarily presentations there along with those logos currently in use at museum-digital. Going over the archives also presented a welcome chance to explicitly license most of the files under a Creative Commons Attribution 4.0 license.

One especially nice aspect of it is, that OpenSearch can be implemented in just a few lines of code. It was thus easy to bring to both the frontend of museum-digital and musdb, to make searching objects right from the browser’s search bar possible, just as you can quickly search Wikipedia or Amazon on many browsers by default.

Since musdb does not feature a simple, unqualified search function (one always needs to specify, whether one searches for a tag, place, etc.), OpenSearch offers using the full text search on musdb.

This content is licensed under a Creative Commons Attribution 4.0 International license.

Image credits: Eva Gonzalès: „Girl Awakening“, Kunsthalle Bremen – Der Kunstverein in Bremen (CC BY-NC-SA)

This content is licensed under a Creative Commons Attribution 4.0 International license.