Talk:Recordings

Return to list of all Recordings     See all Categories    Spreadsheet: Recordings-Source-List

Metadata repository: https://data.direct-mind.org/

Some of this is out of date

Table formatting in Recordings

1. Table widths are hardcoded

2. CSS is here: MediaWiki:Common.css

To Do List

Need to eventually reconcile the parallel Excel spreadsheet and the recordings.

Best to wait until more transcriptions are done.

Rose repository

For the editors a directory is maintained for various versions (1st draft, 2nd draft, etc)

TRANSCRIPTIONS

This can be seen from your browser here:

http://distribution.direct-mind.org/ User name name: distribution Password: distribution

AUDIO

Audio recordings are here:

http://audio.direct-mind.org/ User name: audio Password: audio

Stages in the process

Conceptually we have two simultaneous efforts:

1) doing the transcriptions and 2) keeping track of data associated with the recordings.

How to Transcribe

See this page for discussion: Transcriptions http://www.direct-mind.org/index.php5?title=Transcriptions

Archiving the Data

T So far there are a couple issues:

1) Deciding on the proper "authoritative" data source. Let's say that a tape or mp3 file has the wrong date on it, as did the published version of Zen and Common Sense. So where do you go for the best date? Probably the original cassette with the date scratched on it? But how do you find the original?

2) The first issue (above) is only a real problem when there are conflicting dates, which hopefully doesn't happen much. But the next issue will be different versions (different MP3 from the same tapes, and different tapes from the same lectures). To narrow these differences the recordings almost all need to be transcribed. Because you're not going to know wheret the holes are (when a cassette is changes) unless you have it all written out.

3) You can see where this is going and it could get too complicated if we try to collect too much data. But for example when we get the bulk of them done, then we can do a couple different things.

Talk to the old members and try to get their old cassettes, if mp3s are not currently in our possession, convert them to MP3 and return them.

Identify tapes that are low-quality - these will be high-priority in finding better copies.

The reason the recordings have to be transcribed before we do this (I think) is that if people have unidentified or poorly identified tapes, we can just do a search on the website for the content (e.g., a mention of Edith Cavell or whatever) to see if we already have it. Then we can compare audio quality and look for missing pieces. Otherwise we'll get too many duplicates.

4) It would be nice to make a special effort, later down the road, to keep a better account of people present (there's a space in the grid above for this). For example, I'm doing this one now (1/2 done)

http://www.direct-mind.org/index.php5?title=1974-1025-Pittsburgh-Meeting

and I don't know who half the people are, because they're from the Pittsburgh group.

.

Original ReadMe file contents

The following is adapted from a readme file on the lecture DVD - applicable to DVD #2 only

A) TECHNICAL AND SUBJECT MATERIAL

These mp3 files were made from audio tapes, generally recorded on simple cassette machines, now converted to Mpeg layer 3 at 128 kBits/sec and 44,100 Hz. 128 was chosen as a compromise between best audio fidelity and file size.

Audio tapes had a range of quality from poor to excellent.

Normalization or other enhancements may be beneficial for some recordings.

However, at this point in the preservation effort they are in a 'raw' state.

Each side of a tape is in a separate mp3 file, so a two-cassette lecture will have four mp3 files, numbered #1 - #4.

Exception: If a lecture begins with an introduction, the introduction is placed in a separate file, file #0.

I.e., tape side 1 will be broken into 2 mp3 files, #0 and #1.

Naming conventions.

Each talk is placed in a separate directory.

Generally the directory and file names will have the year, the title of the lecture, and in some cases the location.

The directory name begins with the year of the recording, for more convenient sorting.

For the mp3 files, the date is moved to the end, and the file number is placed at the beginning, also for easier sorting.

Each tape converted in this collection has a summary sheet in "RTF" format which will open in any word processing program.

Summary sheet information

1. Number of cassettes in set, and times of each side. 2. Notation whether other versions of the recording may exist. 3. Number of mp3 files created. 4. Subject matter on each side, with times. 5. Transcription priority recommended (1 = low and 5 = high). 6. Technical information/audio quality. 7. Identifiable voices other than RR (may be helpful in locating other recordings, or tracking down historical information).

Summary sheets were made with certain time constraints. Please feel free to add information to the summary sheets and send them to a historian. If you discover defects in the any of these audio files, please notify someone connected with the preservation effort.

Issues

The most difficult and tedious part will be the cataloging: collecting data from various sources and merging the information.

This will have to be an iterative process - i.e., back-and-forth between cataloging and transcribing.

Catalog

Meta-procedure at this point:

1) assemble all information we have currently
2) make sure the Wiki reflects the most accurate information (dates of recordings, locations)
3) bring the Wiki to a state where it will be easy to read and understand, in order to ...
4) find out where the holes are, gather missing recordings

Transcriptions

This is the rewarding part. This work has to be done and we are the people who have to do it.

Transcriptions are necessary for identifying fragments and determining whether or not we already have a copy of the recording, and whether the new copy is a better or worse version than what we already have. Obviously we want to have best the audio quality with the least missing pieces.

Stuff moved from main recordings page

Date is the most convenient sort order. If no Title - may need to come up with a title. Location may be omitted if obvious in title (KSU is Kent; OSU is Columbus; Case Western is Cleveland). If no Location - try to identify voices when questions are asked, or in introduction, or to references within the talk.

This list is one of many and needs to be scrubbed, but scrubbing may not be possible before talks are partially transcribed (to find duplicates).

To avoid confusion let's refer to the talks by Mr. Rose as "recordings" or "transcriptions" and the associated Talk pages as "discussion pages."

Short notes can be put in the Remarks section of the Data grid of the Recording-Transcription page; longer notes can be put on the Discussion page, with a note in the Data grid referring to the Discussion page.

Subheaders on first pass will be File 1, File 2, File 3, etc for the mp3 files (this is used in the initial Data Grid). This will help for editing, especially if someone else picks up the talk midway. As the transcription nears completion, subject headers can be substituted for file number headings.

As material on this page increases, discussion can be moved here: Talk:Recordings (Discussion page for this page).

Data template is here --> Data-Template-For-Page