Request - Make it easier to retrieve media

[ccicero]

Describe what you're trying to do

This issue is extracted from a broader one with an overall goal of making it easier to retrieve media and metadata from Arctos. Two specific requests based on discussions last month in the Media Committee:

1. Eliminate the maximum # records that a Media Search will display (i.e., remove limit of 500 records)
2. Improve how media metadata are displayed from the Catalog Records search download. Ideally, parse metadata into different columns similar to the media bulkloader. Perhaps this needs to be a separate download from the catalog record data?

Let's plan to discuss this at the next meeting when @dustymc can attend.

[mkoo]

Before we get into the technical details (ie Dusty) please articulate or give examples of what you are trying to do? Why do you need to do this? what are you ultimately trying to accomplish? (I'm just saving Dusty the effort of asking the question too!) Does "easier to retrieve media" mean download the media files too? And in what context? I will guarantee once we make it easy for anyone to download all your audio files then we will also get complaints about piracy.

both your requests are interconnected (probably) and both cant be addressed without the context. This might be better to be a discussion before any technical considerations. so let me know how you want to proceed. Maybe the media committee should meet to discuss

[ccicero]

Yes, the media committee needs to discuss but with @dustymc as we've already met and discussed some of this and need technical discussion of what's possible or not.

1. Eliminate the maximum # records that a Media Search will display (i.e., remove limit of 500 records). If you do a Media Search and there are over 500 records, you only get 500. Seems straightforward? Why limit the records returned to 500?

Example: Search 'Media Label Value' contains 'MVZ' and 'Created by agent' contains 'groth' - there are 2,277 media files.

2. Improve how media metadata are displayed from the Catalog Records search download. Ideally, parse metadata into different columns similar to the media bulkloader. Perhaps this needs to be a separate download from the catalog record data?

The media metadata that get downloaded are all mushed into one column and impossible to interpret/parse. We need the data downloaded in separate columns, comparable to the media bulkloader.

Example: Here's what a couple of records look like in the download (search Catalog Records for 'MVZ:Bird' and media = 'audio' and agent contains 'groth':

One one media file for a record:
[{"MC": "audio", "MI": 10293608, "MT": "audio/mpeg3", "MU": "https://web.corral.tacc.utexas.edu/MVZ/audio/mp3/427_2581.mp3", "PU": null, "TN": "/images/audioNoThumb.png"}]

More than one media file for a record:
[{"MC": "audio", "MI": 10293609, "MT": "audio/mpeg3", "MU": "https://web.corral.tacc.utexas.edu/MVZ/audio/mp3/427_2582.mp3", "PU": null, "TN": "/images/audioNoThumb.png"}, {"MC": "audio", "MI": 10293610, "MT": "audio/mpeg3", "MU": "https://web.corral.tacc.utexas.edu/MVZ/audio/mp3/427_2583.mp3", "PU": null, "TN": "/images/audioNoThumb.png"}, {"MC": "audio", "MI": 10293612, "MT": "audio/mpeg3", "MU": "https://web.corral.tacc.utexas.edu/MVZ/audio/mp3/427_2585.mp3", "PU": null, "TN": "/images/audioNoThumb.png"}]

Here's another example where there are multiple images for a record:
[{"MC": "image", "MI": 10001377, "MT": "image/jpeg", "MU": "https://web.corral.tacc.utexas.edu/MVZ/images/MVZ_img/cards/jpg/img_card_1360.jpg", "PU": "https://web.corral.tacc.utexas.edu/MVZ/images/MVZ_img/images/tb/tn_img_1360.jpg", "TN": "https://web.corral.tacc.utexas.edu/MVZ/images/MVZ_img/images/tb/tn_img_1360.jpg"}, {"MC": "image", "MI": 10007866, "MT": "image/jpeg", "MU": "https://web.corral.tacc.utexas.edu/MVZ/images/MVZ_img/cards/jpg/img_card_1361.jpg", "PU": "https://web.corral.tacc.utexas.edu/MVZ/images/MVZ_img/images/tb/tn_img_1361.jpg", "TN": "https://web.corral.tacc.utexas.edu/MVZ/images/MVZ_img/images/tb/tn_img_1361.jpg"}]

@ebraker has some good examples too.

That is very difficult to interpret or parse, even for Arctos operators who know the media metadata fields, and impossible for a public user interested in our media. And it gets worse the more media files you have linked to a record. We need a better way of downloading and displaying the metadata.

Downloading the actual media is a separate issue, let's first figure out a way to better download our media metadata.

[mkoo]

ok so really this is about parsing the JSON into friendly tables. (again wasnt clear from the request!)

Have you already discussed what the desired table should look like?

[ccicero]

It's about that plus not having the 500 record limit. We discussed having it look like the media bulkloader, but are open to suggestions.

[ebraker]

Yes, to echo Carla, the request is two-parted:

1. Make friendlier media download tables instead of the current JSON blob
2. Be able to page through and view all of your media. There is no current way to do this - the media gallery caps out at 500, and if you attach your media to a project, you can view 10 records at once but risk your IP being blocked when trying to open too many (hundreds) pages.