Mediaflux Desktop is a web-based application that you access with a web browser. It interacts with the Mediaflux server via the secure HTTPS protocol. This application complements the Mediaflux Explorer (HTTPS Protocol) (a Java application).  While Mediaflux Desktop allows you to upload or download data, at the University we primarily recommend its use to curate data, control project structure, apply metadata and search files and metadata.  For more information, see the Mediaflux Desktop - How-to videos.

Accessing Mediaflux Desktop

The URL is:

Logging in

Multi-Factor Authentication (MFA) for Mediaflux is now available. To learn more and enrol, visit: Mediaflux MFA

Mediaflux requires the triple of (domain, username and password) to log in. All the Mediaflux login screens show these fields.

University of Melbourne Account

You can login to Mediaflux directly with your University of Melbourne institutional credential

• In the Domain field, enter unimelb (staff accounts) or student (student accounts)

• In the User field, enter your university username

• In the Password field, enter your university password

• Click Login

Open

Local Account

If your account is a local account (i.e. not using your institutional account), you will have been provided with the details for domain, user and password. Just enter them and click Login.

Using Mediaflux Desktop

To download data using Mediaflux Desktop

• double click on the Asset Finder tool

• right-click on a namespace or asset and select the Download as Archive option.

• the data will be downloaded into an archive that you can then unpack on the local machine.

This form of download method is not good for big data because it creates an archive.  For larger downloads, consider Mediaflux Explorer (HTTPS Protocol) or the Mediaflux Unimelb Command-Line Clients.

For more information on using Mediaflux Desktop, see the Mediaflux Desktop - How-to videos.

Open

How to "Copy" Existing Data Between Your Mediaflux Projects

It is possible to "copy" data between two Mediaflux projects (that you have the adequate access permission) via the Mediaflux Desktop web client. We are working to incorporate this feature into the future version of Mediaflux Explorer.

Please note the following:

• Currently you can copy at the folder-level and not individual files. You can copy the whole folder over and delete the files that you don't need at the destination project.

• All existing ACL on the data will be removed after the copy.

Please follow the steps below:

• Go to https://mediaflux.researchsoftware.unimelb.edu.au/desktop/?dti=no

• Log in using the following credentials:

• After log-in, double-click on "Asset Finder" from the left menu bar

• Click the "+" sign next to the "projects" on the folder panel to navigate to the source project and select the folder that contains the data you want to copy from

• Right-click on that selected source folder and click on "Copy collection" option from the pop-up menu

• Put your "destination" in the "to:" field, e.g.,: /projects/proj-migration_shared_test-1128.4.544

• If you want to copy the selected source folder to a sub-folder at your destination project, you can do so by adding the full path to the desired sub-folder, e.g., /projects/proj-migration_shared_test-1128.4.544/test/test_1 (this will copy the selected folder into the test_1 sub-folder of proj-migration_shared_test-1128.4.544 project)

Mediaflux Desktop - How-to videos

This page contains video guides for the Mediaflux Desktop which is the web based interface to Mediaflux primarily used to curate data, control project structure, apply metadata and search files and metadata.

• How-to videos from Research Computing Services

  • Logging in

  • Navigating the Desktop

  • Controlling Project Structure

  • Defining your metadata

  • Enumerations and Dictionaries

  • Create and assign Metadata

  • Modifying Metadata

  • Metadata in namespaces (directories)

  • Searching

  • Moving search results

  • Downloading your metadata

  • Applying Metadata to multiple assets

• Training Videos from Arcitecta

  • DTI (desktop integration) no longer available

  • List of topics in the Arcitecta training videos

    • Introduction to Arcitecta Desktop

    • Arcitecta Desktop Overview

    • Introduction to Asset Finder

    • Ingesting data with Asset Finder

    • Ingesting data from spreadsheets

    • Viewing Data in Asset Finder

    • Introduction to Asset Map

    • Metadata Library

    • Metadata templates for collections

    • Searching using filters

    • More filter topics

    • Opening and saving views

    • Access Controls (ACLs)

    • Administering Users

How-to videos from Research Computing Services

Logging in

See Desktop (Browser, HTTPS Protocol).

Open

N.B. Video shows old log in method, the link above the video shows the updated method, briefly described below.

To login with your UoM credential

• You must enter the protocol (https) and server/port details first (detailed above the video)

• Enter unimelb (staff accounts) or student (student accounts) into the Domain entry field

• Enter your usual institutional username (for the user field) and password (for the password field and click Sign in).

Navigating the Desktop

Accessing the components of the Mediaflux web based Desktop

Open

Controlling Project Structure

Finding your project(s) and controlling their structure.

Open

Defining your metadata

Creating metadata definitions.

Open

Some useful metadata element types:

Enumerations and Dictionaries

Creating your own Enumeration list and Dictionary.

Note: Dictionaries will always be presented as a sorted list of options, if you require control over the ordering of the list you should use an Enumeration which will preserver the order you specify.

Open

Create and assign Metadata

Open

Modifying Metadata

Open

Metadata in namespaces (directories)

Open

Searching

Open

Moving search results

Open

Downloading your metadata

One way (of many) to download your metadata from the Desktop

Open

Applying Metadata to multiple assets

Open

Training Videos from Arcitecta

You can download a zip file containing Mediaflux Desktop Training Videos from the Vendor of Mediaflux, Arcitecta.

DTI (desktop integration) no longer available

Note that since these training videos were produced, browser manufacturers have locked down the features that the DTI (desktop integration) uses.  Therefore the DTI is no longer usable; you will not be able to see your local files in the Files entry on the left hand side of the Asset Finder.  You can still upload data using Mediaflux Desktop by dragging and dropping into the Desktop window from your Operating System's file manager, however you will need to keep the browser window open for the duration of the upload.

When you start Desktop you will see a message stating that the DTI is inactive.  This is normal.

Open

Dragging and dropping data into the Asset finder from the Linux file manager:

Open

Dragging and dropping data into the Asset Finder using the Windows file manager:

Open

List of topics in the Arcitecta training videos

• Introduction to Arcitecta Desktop

  • Logging in

  • Arcitecta Desktop Integration Agent (DTI) (Note: the DTI is no longer supported for data ingestion, see detail above)

  • Basic features

• Arcitecta Desktop Overview

  • Introduction

  • Data ingestion (Note: the DTI is no longer supported for data ingestion, see detail above)

  • Metadata

  • Data discovery

  • Geospatial attributes and mapping

• Introduction to Asset Finder

  • Collections, Files and Metadata (Note: the DTI is no longer supported for data ingestion, see detail above)

  • Favourites

• Ingesting data with Asset Finder

  • Drag and drop files from local filesystem to a collection (Note: the DTI is no longer supported for data ingestion, see detail above)

  • Ingest options

    • Packaging

    • Naming

    • Geolocation

    • Metadata

  • Arcitecta Desktop Integration Agent (DTI) progress window (Note: the DTI is no longer supported for data ingestion, see detail above)

• Ingesting data from spreadsheets

  • Mapping rows/columns to assets with metadata

  • Applying default metadata

• Viewing Data in Asset Finder

  • Searching

  • Asset details

  • Asset history

  • Column view

• Introduction to Asset Map

  • Navigation

  • Background maps

  • Asset display

  • Layers

  • Saving/sharing maps

• Metadata Library

  • Metadata auto-extraction on ingest

  • Metadata namespaces

  • Metadata definition documents

    • Elements

    • Subelements

    • Attributes

  • Attaching/detaching metadata to/from assets

• Metadata templates for collections

  • Using a metadata definition document as a template

• Searching using filters

  • Filter selector

  • Metadata filters

  • Asset filters

• More filter topics

  • Filtering out assets without metadata

  • More on metadata filters

  • Date filters: relative and absolute

• Opening and saving views

  • Opening/Favouriting views

  • Saving a view as an asset

    • Filter definition

    • Column view

• Access Controls (ACLs)

  • Collection ACLs

  • Metadata ACLs

• Administering Users

  • Authentication Domains (local/LDAP)

  • User management

Plugin services available via the Desktop

Introduction 

The Mediaflux Desktop Asset Finder tool provides an extensible environment into which we, the UoM Mediaflux service operator, can plugin extra services that we have developed.  These are visible when you right click on a project namespace (folder) or sub-namespace that you have access to with the Asset Finder.   When you right click on a namespace, a new GUI pops up, and you can see various items in the menu. The ones towards the bottom above the destroy item and between the horizontal dividers are the ones that we have added (starting with Copy Namespace).   The function you are going to execute pertains to the namespace that you have selected. The functions available are currently:

• Copy namespace - copy a namespace from one location to another.

  • Tools like the Explorer and Desktop move the namespace if you just drag it somewhere else. Sometimes a copy is useful (but be mindful of unnecessarily duplicating data). See here for more information.

• Download from an sFTP server (e.g. the Australian Synchrotron).  

  • You will need to fill in the details of the sFTP server that you want to download from, including your username and password.

  • There is no browse feature here, you just have to specify the source path to download. 

  • The data will be downloaded to the namespace you have right-clicked on.

  • Use a tool like the Explorer or the Desktop to see the data arriving

  • This function

    • can handle source symbolic links - they will generates special Mediaflux symbolic link assets

    • can be multi-threaded, but for large files, little improvement will be gained beyond 1 thread

    • can notify the user when completed (so you can run in the background for long downloads)

    • can skip already uploaded files (checking by path and file size only (a checksum check is not possible with sFTP)

  • Example of main arguments for Australian Synchrotron

    • host : sftp.synchrotron.org.au

    • port : 22

    • path : /data/mx/15402d/frames 

    • username : nkilleen@unimelb.edu.au

    • password : <synchrotron password>

• Recursively compute checksums  in a namespace and email a CSV.

  • Mediaflux natively uses CRC32 checksums. It's pretty common to need others such as MD5 which is one of the choices available in this service.

• Recursively count files and sum the storage space used by a namespace(directory)
  
  

  • The resulting sum is in bytes and human-readable format.

  • See this page for more details

• Recursively validate asset and namespace names in the context of cross-operating system consistency.

Example and Details

Please note

• The progress meter you will see when you execute a service does not function correctly at present. This will be fixed in a future update - just ignore it for now.

Open

Hover over an item of interest to see a description of what it does.  Then, simply select the item that you want (left click) and then a new GUI will further pop up for that function. Fill in the required details and click the button in the bottom right of that GUI (it's label reflects the function you have selected) to execute.

Here is an example (selecting 'Recursively compute checksums in namespace') where you can email a CSV of the MD5 checksums of all the assets in and below the namespace of interest.

Open

Once you have started the execution, the GUI will be replaced with another to present you with any output and includes:

• Progress meter - the progress meters don't work usefully yet so just ignore them!

• A new button, Abort. If you click this the process will be aborted.

• A new button, Background. If you click this, the GUI will disappear and the task will continue to execute in the background. However, you will lose any output (including errors) that it might have presented to you in that GUI (so it's mainly useful for long-lived tasks that are going to send you output by email or there is no output)

Open

Sometimes, after the service completes, you may see an error GUI like this. If the error type is "Error: com.google.gwt ..."  just ignore that and dismiss the error GUI (a ticket has been raised for this) this is not indicative of a data transfer issue.

Open

Recursively count files and calculate the storage space used by a directory(namespace) in Mediaflux

A new plugin service has been deployed in Asset Finder within Mediaflux Desktop. This enables users to count the files and calculate the total storage space used by a directory.

Follow the steps below:

1. Log into https://mediaflux.researchsoftware.unimelb.edu.au/desktop/

   • Open

     

2. Open Asset Finder

   • Open

     

      

3. Select the directory (namespace), open the context menu by mouse right-click, then select “Recursively count and sum namespace“

   • Open

     

      

      

4. It will pop up a dialog to confirm, simply click the button: ”Recursively count and sum namespace”

   • Open

     

      

      

5. It will then start summing up the total storage usage and return the result in a pop-up window.

   • Open

     

Create Upload Share to Collect Data