Print

Print


Hello,

As the firm that coded BWF MetaEdit in collaboration with, and funded by FADGI (www.digitizationguidelines.gov), it is extremely rewarding to see it being adopted in so many places, and to see expert users like Mike providing in-depth advice on its use. It has been downloaded over 4000 times and I have run into it being used in studios and workflows worldwide. It is great to see this conversation taking place here, and by all means it should. Additionally, I would like to encourage users of BWF MetaEdit to use the SourceForge Tracker and Forums (http://sourceforge.net/tracker/?group_id=314696) as much as possible so that we can track bugs and feature requests in a centralized place, and have a user forum.

Gregorio, regarding your original question, it looks like you are sending non-newline characters through the CLI assuming that <br/> or /n/r would be interpreted into the newline and carriage return ASCII codes. However there is currently no translator to do this substitution. As a workaround you could construct a csv or xml conformance point document in a file and use that file as the input to bwfmetaedit. Or as Mike suggests you could simply enter this via the GUI. If you think a translator would be a good functionality to add please add this request at the SourceForge page as a feature request.

Bill, as you can see from Mike's response you can use MetaEdit to embed XML data in the aXML, iXML and XMP chunks. However, before moving forward with any of these you should become familiar with the specification that revolve around each of these standards.  The report found at the following link will take you to a document that provides some high level information on each of them - http://www.digitizationguidelines.gov/audio-visual/documents/AVPS_Audio_Metadata_Overview_090612.pdf

It is important to note that MetaEdit does not currently validate the data  that you place in these XML-based chunks. It assumes that the data is being validated in accordance with the respective standard before and after embedding. In other words, if you are embedding iXML data in a file it is up to you to make sure that the data conforms to the iXML standard. In this regard, the aXML chunk is the least stringent. It only requires that the data be valid XML and that's it. However, if you had a defined schema for embedding in the aXML chunk you would need to validate to the schema before embedding.

I would also like to point those who are embedding metadata to an important study conducted by the ARSC-TC (http://www.arsc-audio.org/pdf/ARSC_TC_MD_Study.pdf) which studies and reports on the persistence and integrity of embedded metadata in WAVE files they are worked on within various applications. The study reports some unfortunate results regarding how applications handle embedded metadata currently. Hopefully enough people make enough noise that developers get wise to the need to handle embedded metadata in better ways. In the meantime the report provides a toolkit for running metadata integrity tests within your workflow to make sure that the metadata you are embedding is not being removed or altered.

Best!

Chris
______________________________________
 
Chris Lacinak
AudioVisual Preservation Solutions
350 7th Ave., Suite 1603
New York, NY 10001
 
office: 917-475-9630
mobile: 917-548-8632

http://www.avpreserve.com
Facebook.com/AVPreserve
twitter.com/AVPreserve

On Jan 26, 2012, at 5:01 AM, Mike Hirst wrote:

> Bill,
> 
> I delivered a talk at the British and Irish Sound Archives meeting in Glasgow last year in which I explored the possibilities for embedding XML. You may find it useful to read the notes I wrote following the talk.
> 
> http://mikehirst.netfirms.com/bwfmetaedit/
> 
> Best Wishes
> 
> Mike
> 
> On 25/01/2012 20:50, WIlliam McQuay wrote:
>> greetings all,
>> 
>> We are interested in the possibilities of including an xml as part of our embedded metadata. Is there an app that would allow us to do this? We currently use MetaEdit to embed and evaluate the MD5's.
>> 
>> Thanks,
>> 
>> Bill
>> 
>> 
>> 
>> 
>> On Jan 25, 2012, at 11:59 AM, Mike Hirst wrote:
>> 
>>> Gregorio,
>>> 
>>> I forget to say that parameter variables need to be uppercase. The example you supplied would be written thus:
>>> 
>>> <?xml version="1.0" encoding="UTF-8"?>
>>> <conformance_point_document>
>>> <File name="CUL-OR01-Gerhard-000101-pres.wav">
>>> <Core>
>>> <CodingHistory>A=ANALOGUE,M=mono,T=Studer810; Sn1234; 15 ips; open reel tape&#x0d;&#x0a;A=PCM,F=96000,W=24,M=mono,T=MytekADC96; Sn1234&#x0d;&#x0a;A=PCM,F=96000,W=24,M=mono,T=Wavelab; Sn1234&#x0d;&#x0a;</CodingHistory>
>>> </Core>
>>> </File>
>>> </conformance_point_document>
>>> 
>>> with command line:
>>> 
>>> bwfmetaedit --in-core=CUL-OR01-Gerhard-000101-pres.xml CUL-OR01-Gerhard-000101-pres.wav
>>> 
>>> 
>>> Best Wishes
>>> 
>>> Mike
>>> 
>>> On 25/01/2012 16:26, Mike Hirst wrote:
>>>> CodingHistory is over-written other tags are unchanged
>>>> 
>>>> On 25/01/2012 15:45, Gregorio Garcia Karman wrote:
>>>>> Thanks Michael. Will that merge CodingHistory with existing<bext>
>>>>> metadata?
>>>>> 
>>>>> On 25.01.2012, at 14:21, Mike Hirst wrote:
>>>>> 
>>>>>> It might be easier to put the data into XML before importing the data.
>>>>>> 
>>>>>> Data from the file test.xml:
>>>>>> 
>>>>>> <?xml version="1.0" encoding="UTF-8"?>
>>>>>> <conformance_point_document>
>>>>>> <File name="test.wav">
>>>>>> <Core>
>>>>>> <CodingHistory>
>>>>>> a=analog,m=mono,t=Studer810; Sn1234; 15 ips; open reel
>>>>>> tape&#x0d;&#x0a;a=PCM,f=96000,W=24,m=mono,t=MytekADC96;
>>>>>> Sn1234&#x0d;&#x0a;a=PCM,f=96000,W=24,m=mono,t=Wavelab;
>>>>>> Sn1234</CodingHistory>
>>>>>> </Core>
>>>>>> </File>
>>>>>> </conformance_point_document>
>>>>>> 
>>>>>> imported into the file test.wav using the command:
>>>>>> 
>>>>>> bwfmetaedit --in-core=test.xml test.wav
>>>>>> 
>>>>>> gives results in accoradnce with EBU R98 1999
>>>>>> (http://tech.ebu.ch/docs/r/r098.pdf)
>>>>>> 
>>>>>> The same data can be imported from a plain text file via GUI.
>>>>>> 
>>>>>> 
>>>>>> On 25/01/2012 12:15, Gregorio Garcia Karman wrote:
>>>>>>> ...like this I meant:
>>>>>>> 
>>>>>>> bwfmetaedit --History="a=analog,m=mono,t=Studer810; Sn1234; 15 ips;
>>>>>>> open reel tape<br />a=PCM,f=96000,W=24,m=mono,t=MytekADC96;
>>>>>>> Sn1234<br />a=PCM,f=96000,W=24,m=mono,t=Wavelab; Sn1234"
>>>>>>> CUL-OR01-Gerhard-000101-pres.wav
>>>>>>> 
>>>>>>> 
>>>>>>> On 25.01.2012, at 10:51, Gregorio Garcia Karman wrote:
>>>>>>> 
>>>>>>>> Dear list,
>>>>>>>> 
>>>>>>>> ...thanks for the useful and hilarious responses!
>>>>>>>> 
>>>>>>>> Here a more technical question regarding bwfmetaedit. I am trying
>>>>>>>> to automate the writing of metadata to a set preservation master
>>>>>>>> files using FADGI BWF metaedit's command line interface. This works
>>>>>>>> well, but I am having difficulties writing a multiline content to
>>>>>>>> the<bext>  field<CodingHistory>. I am trying to do something like this:
>>>>>>>> 
>>>>>>>> 
>>>>>>>> but the<br />  does not get interpreted correctly. Until now I have
>>>>>>>> tried a number of alternatives from OSX's bash with no sucess.
>>>>>>>> Suggestions?
>>>>>>>> 
>>>>>>>> 
>>>>>>>> Gregorio Garcia Karman
>>>>>>>> +44 (0) 7532 127 645
>>>>>>>> [log in to unmask]
>>>>>>> 
>>>>>> 
>>>>>> --
>>>>>> Mike Hirst
>>>>>> Managing Director
>>>>>> DAS-360
>>>>>> 16 Ocean View
>>>>>> Whitley Bay
>>>>>> Tyne&  Wear
>>>>>> NE26 1AL
>>>>>> 
>>>>>> tel: 0191 289 3186
>>>>>> email: [log in to unmask]
>>>>>> web: http://www.das360.net
>>>>> 
>>>> 
>>> 
>>> --
>>> Mike Hirst
>>> Managing Director
>>> DAS-360
>>> 16 Ocean View
>>> Whitley Bay
>>> Tyne&  Wear
>>> NE26 1AL
>>> 
>>> tel: 0191 289 3186
>>> email: [log in to unmask]
>>> web: http://www.das360.net
>> 
>> William McQuay
>> [log in to unmask]
>> Supervising Audio Engineer
>> Macaulay Library
>> Cornell Lab of Ornithology
>> 
>> Our Mission:
>> To interpret and conserve the Earth's biological diversity through research, education, and citizen science focused on birds.
>> 
> 
> -- 
> Mike Hirst
> Managing Director
> DAS-360
> 16 Ocean View
> Whitley Bay
> Tyne & Wear
> NE26 1AL
> 
> tel: 0191 289 3186
> email: [log in to unmask]
> web: http://www.das360.net