Interpreting run out information 4. Identifying problem areas

1̶.̶ ̶G̶a̶t̶h̶e̶r̶i̶n̶g̶ ̶I̶n̶f̶o̶r̶m̶a̶t̶i̶o̶n̶
2̶.̶ ̶M̶a̶k̶e̶ ̶a̶ ̶r̶o̶u̶g̶h̶ ̶s̶t̶a̶t̶e̶m̶e̶n̶t̶ ̶o̶f̶ ̶c̶u̶r̶r̶e̶n̶t̶ ̶m̶e̶t̶h̶o̶d̶s̶
3. D̶i̶s̶c̶u̶s̶s̶ ̶b̶e̶s̶t̶ ̶p̶r̶a̶c̶t̶i̶c̶e̶s̶
4. Identifying problem areas
5. Discuss resolutions to problem areas based on best practices
6. Formulate new guidelines

Following on from Forum Thread #52151d4d9469733cfcfc9580, this thread is for Identifying problem areas regarding entering and interpreting run out information.

Sorry for the delay in getting to this stage, this project was delayed due to other items becoming a priority. Hopefully we can move forward now a bit quicker. Please feel free to PM me.

Problem Areas - General statement;

This is a quite specialised area of the site data entry with a number of reasonably complicated usage cases and opinions. When it comes to names (people or companies), making a consistant set of guidelines to cover all eventualities seems quite hard when considering the points marked "Hard to resolve" below. The other points seem easy.

Noted problem areas:

Hard to resolve:

* Mastering engineer initials entered with / without symbols and with / without additional information such as studio location / type. Should be Aliases, not ANVs? (Niks note - seems complicated and hard to reconsile with other methods we use on the site. Examples such as "SST (8)" are really open to interpritation as to what was ment in the inscription.)

* URLs - as on release, or interpreted to company name. (Niks note - this is a very polarized argument that a few prople have been having over the course of well over a year.)

(Niks note - IMHO these two above issues are tied together, and we need to get a Grand Unified theory of how to deal with them all in a consistant manner that captures the information correctly and in a useful manner. Unfortunately, I don't feel we are really any further forward regarding these points at this stage, but what I will do is split that discussion out into a new thread.)

Reasonably easy to resolve:

* Repeating strings of information in different BaOI fields. (Niks note - seems to be a desire to cut down on duplication, but maybe need to look at examples and discuss in a seperate thread?)

* Seperating stamped "Matrix/Runout" strings from hand-etched strings. Or require full string to be entered as one? (Niks note - seems to be weighted toward having full string entered)

Simple to resolve:

* Clearer definitions for the roles of mastering, cutting, pressing, etc. (Can do!)

* Ordering and spacing of BAOI string. (Niks note - try and keep it simple!)

* Representing untranscribable symbols - editorial-style brackets are better than using the Description field? (Niks note - AFAIAA, the agreed form is using square brackets in the string. Code to guidelines?)

* Side identifiers - as on release vs as entered to the release info. (Niks note - as on release is better IMHO, simple decision?)

Best Practices

* We enter factual information
* As on release is an important guide
* Artists can use ANVs for variations in their name, aliases for entirely different names
* Labels and Companies can't use ANVs, we tend to want to capture everything other than insignificant punctuation / abbreviation differences as seperte entities

==================

Please note - this thread is only for listing the problem areas. Discussing them in depth is not going to be helpful, and will probably have to be repeated again in following stages. I feel that all the points have been extracted from the previous thread, but if there is anything I have missed, please state it so it can be included in the next stage. thanks!

nik
Representing untranscribable symbols

Similar problem:
• Vague Unicode symbols
For example: is a triangle always △ or could it be also Δ?
Is a square ▫ or □ or ◻?
Etc.

I am not sure if I understand correctly, but if this is about difficulties considering adding matrix information, here's what comes into my mind:

• As an addition of the question of Έλενα Παπαρίζου* - Τι Ώρα Θα Βγούμε; has the second symbol, but shouldn't we just always go for -? I'm sure that in most cases people don't even notice.
• I've also seen different star symbols (* and another one).
• It needs to be easier to add logo information. ['Sony Music' logo] is making things harder than they are, I think. I've also seen s add it like Matrix / Runout (4 Logos). See Björk - Biophilia. I think there needs to be a clearer method, or at least a guideline.
• If IFPI codes are part of the matrix string, should they also be added in the matrix field or just in the corresponding SID code field?
• The ƒ appears in 'ifpi' sometimes. Should it be added or should we just add it as an f?
• Finally, how about mould text? Text like 'Made in by EDC', for example (Caro Emerald - Tangled Up). Should it be added as 'Matrix / Runout (Mould Text)', 'Other (Mould Text)' or in another way?

Hope this helps locating difficulties! ;-)

Ivo94
As an addition of the question of loukash, how about […]

I was just giving a few examples. :)
It applies to any "ambivalent" Unicode character. Even such a seemingly simple form like a vertical line can cause trouble:
I l | ⎸⎹⏐⃒︳｜𝄀𝄅 and even 1 (in some typefaces)

Ivo94
• The ƒ appears in 'ifpi' sometimes. Should it be added or should we just add it as an f?

I recall an older Nik's advise to enter it as plain "ifpi".

loukash
It applies to any "ambivalent" Unicode character.

Actually a quite a good example are vinyl records pressed by Gramofonové Závody:
Ever since they started in 1951, for their runout stamps they were using a typeface which doesn't visually differ between 0 (zero) and O (capital letter O). Both characters look exactly the same.
Yet there is always a contextual difference.
A matrix stamp like "OE 0341/O B" actually appears as "OE O341/O B" or "0E 0341/0 B", depending on how you want to read it.
But we know – and that's an undisputable fact – that 0 (zero) only appears in the numeral part, whereas the prefix characters and the character after the slash are always letters in this particular scheme; the letters after the slash denote the galvanoplasty generation and were assigned sequentially from A to Z (and as confirmed by an engineer, when they ended up with Z for represses of very popular records, they simply started over with A…)

How about permitting the use of non-breaking spaces to separate different markings on run-out etchings?

nik
* Repeating strings of information in different BaOI fields. (Niks note - seems to be a desire to cut down on duplication, but maybe need to look at examples and discuss in a seperate thread?)

yes, this topic concerns me and there is a related discussion ongoing here:
the 'best' way to sort Matrix and SID Code variations ?

For myself I would agree with the explanation from dreeat and ETM there. This 'marry-up-style' + ascending order make things clear.

But there are other opinions/ideas as well.

Here are two examples for each method in question.
Bolt Thrower - Those Once Loyal 'marry-up-style'

Maybe it could be helpful to get an 'Nik's note' there to push this topic.

brunorepublic
How about permitting the use of non-breaking spaces to separate different markings on run-out etchings?

This was discussed some time during the BAOI introduction, but eventually Nik has advised against.

But generally it may also be a problem area:
• Where does a runout begin and where does it end when the data is spread over full 360°?
• How to proceed when there are several lines of data?

loukash
Where does a runout begin and where does it end when the data is spread over full 360°?

Is it possible to always start with the catalogue code then follow around to get some consistency.

ABusDriversPrayer
Is it possible to always start with the catalogue code then follow around to get some consistency.

Yes, but who knows which is which?
You may know that and I may know that (sometimes).
Others don't.
That's why I consider it a "problem area".

loukash
• Where does a runout begin and where does it end when the data is spread over full 360°?

For vinyl, I would suggest starting from the beginning of the run-out groove and following it to the end.

For CDs and DVDs, this can occasionally be a bit more problematic, but I find most discs have a clear beginning and end.

brunorepublic
For vinyl, I would suggest starting from the beginning of the run-out groove and following it to the end.

Huh? Is that something akin to squaring the circle…?

Any more suggestions? Keeping this moving so we can get on to the next step quicker.

Thanks for the bump Amsreddevil. I am planning to move to the next stage next week, I don't think this needs to be a long drawn out step like the last one...

nik
* Labels and Companies can't use ANVs, we tend to want to capture everything other than insignificant punctuation / abbreviation differences as seperte entities

The question for me is: why do we want this? I don't think the lack of ANV is really the issue. The issue is what a link to a label or company page should actually achieve.

There are two essential differences here to the way documentation works for other parts of the db:
- exact documentation is done in the baoi section by default (Runout/Matrix etc.). We do not have this option for catalog numbers (hence we should not tweak them) or for artist names (hence we have the ANV function).
- there is no 1:1 correspondence between what appears in the matrix area and the entity we would like to catalog, nor with its role on the release. IMO this is the one big problem area. The involvement of entities responsible for those actions we can document (Pressed By, Manufactured By, Made By, Glass Mastered, Lacquer Cut At etc.) is rarely ever stated verbosely in the matrix area. There are lots of different indicators though, that can be used to identify an item's manufacturer and allocate it to the correct discography. Among these indicators are brand names, website URLs, matrix numbers, SID codes or other manufacturing details. Many of these indicators are impossible to document individually in the form of linked entities (unless we accept SID codes, matrix numbers or manufacturing details to constitute such entities as well). Often it is the combination of several such indicators that gives away the true manufacturer.

IMO best practice is to document exactly what appears in the matrix area of the release - in the baoi section. Possibly in a standardized way. Taking some part of this info out of the matrix context and converting it into a link, attached to some more or less arbitrary manufacturing role should not be (and factually is not) part of the documentation anymore. It is necessarily some form of interpretation, no matter how close we remain to the name that appears in the matrix. We should try not to confuse or conflate the two aspects, as it will lead to entities that have no meaning outside of discogs.

Nik, on point three which is now closed, can you sneak this idea in:
http://discogs.sitiodesbloquea.com/forum/thread/363471#521517889469733cfcfb41fe
I thought I had already mentioned it in detail but just mentioned it in ing. Tthis recently came up in another thread and here are some good ideas from it:
http://discogs.sitiodesbloquea.com/forum/thread/52550ef5d07b0969693effeb?page=1#526a2ef95e75a77841086efd
http://discogs.sitiodesbloquea.com/forum/thread/52550ef5d07b0969693effeb?page=1#526a32a4a86b6d3c3af02564
http://discogs.sitiodesbloquea.com/forum/thread/52550ef5d07b0969693effeb?page=1#526a39802ec4b3721f87b406

Just to log it for posterity re: Part 3. nik appears to be favouring a method where variant data is sorted by role first, then variant, rather than the variant first, role second format preferred by the majority on that last thread linked above:
http://discogs.sitiodesbloquea.com/forum/thread/52550ef5d07b0969693effeb?page=1#5278ff8bd07b096a00f7d39d

I hope Nik reconsiders as the role first method is causing problems and leads to data loss and is more difficult to edit.

In addition to the posterity post above, I also wanted to log something Amsreddevil has asked I log here.

my method of describing aspects of a CD matrix:
http://discogs.sitiodesbloquea.com/forum/thread/52870e8ba86b6d45f1d63b55

"I avoid using outer or inner. I stick to three main parts and state the item's location rather than how it's entered unless it differes (moulded and lasered etc):

Mirror band. the place where the matrix is normally found.
Mould/hub, the plastic centre where you'd frequently find the mould SID code.
Stacking ring, between the mirror band and the hub - sometimes the mould sid code can be found there, but less frequently.
http://desktoppub.about.com/od/cddvd/a/fritz_cddesign.htm
http://0.tqn.com/d/desktoppub/1/0/9/m/cd-design-disc-anatomy.jpg

Matrix/Runout (Unabridged): 813 242-2 02 / YF MADE IN
Matrix / Runout (Mirror band): 813 242-2 02 /
Matrix / Runout (Mould/hub): YF MADE IN

"

My post received the following constructive comments:

Amsreddevil
Ta. Sounds right. Will this be added somehow in guidelines when nik gets around to rounding of the series of posts on how best to add the matrix info? Would be good to have uniformity in adding this, there are so many different used on subs.

anssisal
That kind of "disc anatomy" picture + description would be great addition to RSG §5.4.

So basically my method above outlines how to clearly, logically and enter variations, without muddied descriptions instead lists holistically in order of Variant. so instead of this (which is currently receiving the affections of nik):

Matrix / Runout (Variations 1 , 4 , 5 & 6): 936248140-2.2 06/02
Matrix / Runout (Variation 2): 936248140-2.4 V01 UIB
Matrix / Runout (Variation 3): W CD 936248140-2.3 V01
Mastering SID Code (Variations 1 ,2 , 4 , 5 & 6): IFPI L011
Mastering SID Code (Variation 3): IFPI L012
Mould SID Code (Variation 1): IFPI 0581
Mould SID Code (Variation 2): IFPI 05Q1
Mould SID Code (Variation 3): IFPI 05P1
Mould SID Code (Variation 4): IFPI 05N5
Mould SID Code (Variation 5): IFPI 05N8
Mould SID Code (Variation 6): IFPI 05R5

We'd be tasked with editing, managing and using this instead:

Matrix / Runout (Variant 1): 936248140-2.2 06/02
Mastering SID Code (Variant 1): IFPI L011
Mould SID Code (Variant 1): IFPI 0581
Matrix / Runout (Variant 2): 936248140-2.4 V01 UIB
Mastering SID Code (Variant 2): IFPI L011
Mould SID Code (Variant 2): IFPI 05Q1
Matrix / Runout (Variant 3): W CD 936248140-2.3 V01
Mastering SID Code (Variant 3): IFPI L012
Mould SID Code (Variant 3): IFPI 05P1
Matrix / Runout (Variant 4): 936248140-2.2 06/02
Mastering SID Code (Variant 4): IFPI L011
Mould SID Code (Variant 4): IFPI 05N5
Matrix / Runout (Variant 5): 936248140-2.2 06/02
Mastering SID Code (Variant 5): IFPI L011
Mould SID Code (Variant 5): IFPI 05N8
Matrix / Runout (Variant 6): 936248140-2.2 06/02
Mastering SID Code (Variant 6): IFPI L011
Mould SID Code (Variant 6): IFPI 05R5

It has been suggested that if the Variant could be used as a sub header for the information, this latter method would be even more clear and useful:
http://i.imgur.com/lRVt9dt.png

Rendering the latter version as follows:

Variant 1
Matrix / Runout: 936248140-2.2 06/02
Mastering SID Code: IFPI L011
Mould SID Code: IFPI 0581
Variant 2
Matrix / Runout: 936248140-2.4 V01 UIB
Mastering SID Code: IFPI L011
Mould SID Code: IFPI 05Q1
Variant 3
Matrix / Runout: W CD 936248140-2.3 V01
Mastering SID Code: IFPI L012
Mould SID Code: IFPI 05P1
Variant 4
Matrix / Runout: 936248140-2.2 06/02
Mastering SID Code: IFPI L011
Mould SID Code: IFPI 05N5
etc...

Thanks for taking the time to sum up those discussions! I agree with you 100%.

Obviously, the variants should be collapsable.

Eviltoastman
It has been suggested that if the Variant could be used as a sub header for the information, this latter method would be even more clear and useful

But the Subheader field would be detached from the actual data fields, whereas the Description field isn't. Shuffle the fields around a bit, e.g. by the requested "autosorting" (see the corresponding thread), and you've lost track completely which is which…

loukash
But the Subheader field would be detached from the actual data fields, whereas the Description field isn't.

We feel the matrix and sid fields should be within the variant field making the thing fixed/immovable within in order of Matrix (often more than one matrix field is required) and SID field, mastering then mould. The description fields would not be purged, they are still vital. If the dev team are talented enough to have a massive useless banner taking up 70% of the help page, I'm sure they can manage this relatively simple task.

Eviltoastman
within the variant field

D'oh, my bad, sorry.
I should have looked at the mockup as well.
Yes, if the fields are hierarchic, then all is well.

Eviltoastman
If the dev team are talented enough to have a massive useless banner taking up 70% of the help page, I'm sure they can manage this relatively simple task.

My self-preservation instinct prevents me from holding my breath for this one, and I can't do a thing about it.

I'm laughing and people might be able to hear me. I wish there was a simple way of expressing this over the internet.

I wanted to move this forward much sooner (again), but this project gets pushed back a lot due to other tasks. it IS on my todo list to move to the next stage.

Eviltoastman
Obviously, the variants should be collapsable.

Yep, thanks Evilt.

nik
I wanted to move this forward much sooner (again), but this project gets pushed back a lot due to other tasks. it IS on my todo list to move to the next stage.

If we just get on with it, this could be rounded off to the end of step 5 by the end of the week, and then all that needs doing is amending the guidelines to suit all of this. The longer this drags on, the more threads, the more arguments in histories, and so goes the domino effect. This could have been resolved half a year ago?

Eviltoastman
I'm laughing

I was almost crying, in despair......

Amsreddevil
I was almost crying, in despair......

I was laughing at Loukash's joke. I was quite frustrated when a practical solution to a genuine data entry issue was overlooked because another method was prettier in spite of its practical drawbacks, particular when considering the errors it currently causes and confusion it generates and also when I consider that we (Discogs) appear to be endeavoring to make things easier for the layperson.

Another reason why my method trumps that used within:
http://discogs.sitiodesbloquea.com/history?release=378824&diff=52

The subber has lost track of the data, missed "none" from variant 4's mould SID code and there's not matrix for 3 and 4 and so on and so on. s get lost as we advised here:
http://discogs.sitiodesbloquea.com/forum/thread/52550ef5d07b0969693effeb

Please can we stop pandering to the aesthete and concentrate on getting data entry right and minimising the risk of problems when entering the data without having to forsake valid data as this example is not an exception, we see it a lot, particularly with larger runs. If we organise by variant blocks the input is cleaner and allows for a more methodical approach, even if the output looks superficially less pleasing on the left column.

Eviltoastman
Another reason why my method trumps that used within:

When you're right, you're right: "real life" examples of failures are usually the best arguments.
(I'm not being ironic here, I really mean it.)

Eviltoastman
It has been suggested that if the Variant could be used as a sub header for the information, this latter method would be even more clear and useful:
http://i.imgur.com/lRVt9dt.png

Rendering the latter version as follows:

^^ THIS! ^^

(In case it gets beyond this proposal stage, it would be nice if some kind of automagic SQL could be done, to move all current data in Matrix / Runout, Mastering SID Code, Mould SID Code to the new fields. Otherwise there will be a lot of manual editing to do...

Having this concluded before 2014 would be nice......
http://discogs.sitiodesbloquea.com/forum/thread/364226#52151d449469733cfcfc944c
14th March, now 29th November.

It should also be nice if we can add "an empty line" (or ½line) for clarity.

Matrix/Runout (CD1)
Mastering SID Code (CD1)
Mould SID Code (CD1)

Matrix/Runout (CD2)
Mastering SID Code (CD2)
Mould SID Code (CD2)

Matrix/Runout (CD3)
Mastering SID Code (CD3)
Mould SID Code (CD3)

instead of

Matrix/Runout (CD1)
Mastering SID Code (CD1)
Mould SID Code (CD1)
Matrix/Runout (CD2)
Mastering SID Code (CD2)
Mould SID Code (CD2)
Matrix/Runout (CD3)
Mastering SID Code (CD3)
Mould SID Code (CD3)

This will make it more visible and maybe more understanding for everyone.
Some for variants of course.
Preferably would then be that the info will be mentioned in the following way

Info CD1 Variant 1
info CD1 Variant 2
info CD2 Variant 1
info CD2 Variant 2
etc.

Just my opinion~!

Also posted in http://discogs.sitiodesbloquea.com/forum/thread/52550ef5d07b0969693effeb

Can we now move to step 5, please.
Because

nik
* Mastering engineer initials entered with / without symbols and with / without additional information such as studio location / type. Should be Aliases, not ANVs?

shouldn't be hard to be resolved, but discussion should be initiated.

Please....

http://discogs.sitiodesbloquea.com/history?release=955325#latest
Looks like a correct way to add is changed into one who's can getting messy

Why are you changing them when the matter is still being discussed. The project is nowhere near a conclusion. Nik gave an opinion, but the project is still on stage 4 and is currenly on the backburner. Leave as you find for the time being unless things are incorrect.

While we are waiting for this to be followed up, could we reach some consensus in the meantime on some of the things that have already been discussed? One point I keep coming across is the way a company logo is added in the matrix. eg's:
Company X [Logo]
Company X [logo]
Company X [Logotype]
Company X [logotype]
[Company X Logo]
[Company X logo]
[Company X Logotype]
[Company X logotype]
['Company X' Logo]
['Company X' Logotype] etc

As this is dragging on and on, can't we as community at least get a consensus in the meantime how to enter certain things so not even more needs editing in the long-run? Some s are doing edits just to 'correct' this type of info when there is no specific agreed way to add it yet.

Amsreddevil
[Company X logo]

This.

"[Company X logotype]" is of course valid as well, but you can't expect anyone to understand the difference between a logo and a logotype.

Amsreddevil
As this is dragging on and on, can't we as community at least get a consensus in the meantime how to enter certain things so not even more needs editing in the long-run?

As it was recently decided to even relegate actual release data to the description field of the baoi section, I would like to point out that something like "logo" or "logotype" is in fact a description, that does not appear on the item itself. This would indeed be something that is better suited for the description field. For example
Matrix / Runout ((S)=Specialty Records Corporation logo) - (S) 3 82873-2 SRC##01 M1S4

strummin
something like "logo" or "logotype" is in fact a description, that does not appear on the item itself

Hence the use of the square brackets [ ].

loukash
"[Company X logotype]" is of course valid as well, but you can't expect anyone to understand the difference between a logo and a logotype.

Still better to stick to one way. When this does get concluded, these types of elements should be clear to all as it would be listed how to add in the final 'guidelines'.

strummin
As it was recently decided to even relegate actual release data to the description field of the baoi section

This whole project by Nik is to decide what is best to do for run-out info that is part of the baoi, so I don't really get this statement. What exactly was decided where by whom? Relevant thread links?

Amsreddevil
What exactly was decided where by whom? Relevant thread links?

http://discogs.sitiodesbloquea.com/forum/thread/356624#52f0c76bc131f32b89057333

xjoxjox
Hence the use of the square brackets [ ].

We are not normally using square brackets for other decriptions of the runout / matrix info. That's what the decription field is there for.

strummin
We are not normally using square brackets for other decriptions of the runout / matrix info.

Pluralis Majestatis? ;)

"We" (as in: "we Bam Bam, etc.)

Besides, may I remind you of the thread title?

nik
Interpreting run out information 4. Identifying problem areas

(emphasis mine)

We are already using square brackets for a similar purpose as an integral part of the credit role syntax. Nothing hinders us to expand the usage to other areas explicitly as well. All we need is to carve it in stone (if necessary).

loukash
"We" (as in: "we loukash") have been using square brackets in BAOI fields for many years now,

We in this case is even including Me (as in strummin), even that [logo] syntax. But part of that use stems from before the introduction of the description fields and the other question is whether that is really a good idea. There are releases where those square brackets in the matrix are for real.

strummin
http://discogs.sitiodesbloquea.com/forum/thread/356624#52f0c76bc131f32b89057333

"Label codes, distribution or price codes are sometimes prefixed with country abbreviations or indicators. Please enter these in the description field only, not as part of the code itself."
does not pertain to the discussion here, these are not price codes and the like. The is not part of the distribution code, the company logo is part of the matrix string info. There is a clear difference imo.

The use of [] was discussed in the previous parts of this project, and is a much-used way of entering the data.

strummin
There are releases where those square brackets in the matrix are for real.

I see your point, but would not the fact that the word logo is added show the [] are not part of the matrix in those instances?

strummin
part of that use stems from before the introduction of the description fields

As far as I , the Description field was part of the BAOI concept from the start.
(Speaking of those: what rather puzzles me is the usage of parentheses or even square brackets in those description fields. I never understood that. The description is already displayed in parentheses, so why nest text even further? I'm simply separating the description with comma(s): "Matrix / Runout (Runout A, machine stamped): XYZ1234")

strummin
There are releases where those square brackets in the matrix are for real.

That may be possible. Although I'd surely like to see a few examples first… :)

loukash
I'd surely like to see a few examples first…

First hit: Brahms* - 7 Danze Ungheresi / Sinfonia N.3 In Fa Maggiore Op.90

nik
4. Identifying problem areas

One very basic problem area is where we accurately record data from the release and where we place descriptions (or interpretations) of those release data. What is a description and what is data that should be accessible to the search engine and not altered with regard to the items that are represented?

Amsreddevil
these are not price codes and the like.

These are items that appear in the baoi section, equipped with a description field, just like the matrix/runout field. In the former we are supposed to remove parts of the release data and place it in the description field. In the latter we are adding our own descriptions and mix them with the data. It just appears inconsistent to me. It's not like I cannot live with either, but what are these discussions for, if we are not allowed to even consider alternatives?

loukash
That may be possible. Although I'd surely like to see a few examples first… :)

Here is an interesting one. Evidently not for real, but still ...
http://discogs.sitiodesbloquea.com/Tommy-Sears-A-Walk-On-The-Outside-Blues-Adlib/release/3309691

nik
* Seperating stamped "Matrix/Runout" strings from hand-etched strings. Or require full string to be entered as one? (Niks note - seems to be weighted toward having full string entered)

Stamped data is clearly an 'action' that is separate from a hand-etched 'action'. i.e. It indicates at least 2 stages of activity - which needs to be clearly understood from the imprinted text that is transcribed to a field.
(Note: We have had no hesitation in isolating the act/s of mastering & pressing on a CD hub data - with dropdowns for SID Codes. i.e. We DO NOT put these 'actions' into ONE STRING. They have no bearing on the other matter).

Considering the ease & 'manageability' of 'field separation' for newbies & experts - Irrespective of anyone's knowledge of the intricacies of runout data interpretation, it is VERY SIMPLE to ask submitters to SEPARATE any data they are entering; a) into matter which is STAMPED and b) matter which is ETCHED by hand.
Thus, any later reader (with some knowledge) can subsequently begin to define aspects of mastering from those of a pressing plant.

nik
* Representing untranscribable symbols - editorial-style brackets are better than using the Description field? (Niks note - AFAIAA, the agreed form is using square brackets in the string. Code to guidelines?)

IMHO, adding editorial parenthesis WITHIN a Matrix is NOT part of the matrix. Why do we have a 'Description Field' if folk are not going to use it?
Matrix/Runout: 4567 YDH R TH5FG (A-side runout, stamped. [Hand-etched floral motif between the 'R' and 'T'])
Matrix/Runout: Sony CRED 56789 02 03 ('Sony' in logo form)
^^^Simple really...
This ISN'T the Matrix: Matrix/Runout: Sony [Logo] CRED 56789 02 03
Or this: Matrix/Runout: Universal [Logo] GAVU 76543 Universal [Logo] Universal [Logo] Universal [Logo]
It's this: Matrix/Runout: Universal GAVU 76543 (4x 'Universal' logos, one at each point of the com in outer ring).
Also, items like "Made in " are part of the pressing process, not the LBR mastering transfer of a Matrix...
'Laser-etched' data is not pressed machine data. (I go back to drawing attention to the way we separate SID matter)

nik
* Side identifiers - as on release vs as entered to the release info. (Niks note - as on release is better IMHO, simple decision?)

If my vinyl release NUMBERS the tracks, then I will not be NUMBERING them. I shall be entering Alpha data. Sides A & B etc.
WHY do I then have to enter in the Matrix field 'As on release?' I have just spent time transcribing the A & B data, now I need to cross-reference that to the Matrix data - USING THE SAME COMMON IDENTIFIERS.
Release Notes are where I can state how the Sides are presented on the release: Roman (ie I, II), Numeric (ie 1 & 2), In full (ie Side One, Side Two), etc.

I shall be quiet now , thanks for getting this far.

Etching vs. engraving as it relates to vinyl (i hope this is an ok place to put this)

Runout text and information is usually very obviously either scribed by hand, or stamped, into the master disk. Yet almost universally in matrix/runout Description fields, and elsewhere (e.g. Forum Thread #529e58cbea621117a73e0170) the term "etch" ("etched", "etching") is used instead of "engrave" ("engraved", "engraving"). It's even used in the relevant guidline, RSG §5.4.

But etching is [wiki] "the process of using strong acid ... to cut into the unprotected parts of a metal surface", where engraving is [again, wiki] "the practice of incising a design onto a hard, usually flat surface, by cutting grooves into it". Hand-writing in the runout area of a vinyl master disk is surely, indubitably not etching.

It's possibly not important in the grand scheme of things, but this inaccuracy bugs me. (This possibly only because as a printmaker there is a HUGE difference between the two which to others is not apparent.)

Does this count as identifying a problem with interpretation of runout information?

nik
Sorry for the delay in getting to this stage, this project was delayed due to other items becoming a priority. Hopefully we can move forward now a bit quicker.

7 months ago (Oct. 2013) wow

Started in March 2013 ...
http://discogs.sitiodesbloquea.com/forum/thread/364226#52151d449469733cfcfc944c

So the estimated date to finish this project would be end of 2015 mid 16 - right? :)

It could be possible the new airport in BERlin will be finished faster.
My bid goes to Berlin :)

Willow.the.Wisp
It could be possible the new airport in BERlin will be finished faster.

Or to paraphrase comrade Ulbricht:
Nobody has the intention of building a music database!

My mother taught me that when you start something, you should finish it too. i.e.: do not start something if you do not intend to finish it.

Leaving this project in 'limbo' is doing nobody any favours. You have a whole load of s who have given input, who want to get this sorted, but have now been let down imo.

This needs sorting ASAP please. If there is no intention of finishing this however, just please be honest about it and lock the thread.

It is a bit like an RPG. Before Nik can complete this quest he must traverse the wastes and get a "Round Tuit" first.

I am identifying this project as a problem area.

Quatroo66
CD1 Variant 1
info CD1 Variant 2
info CD2 Variant 1
info CD2 Variant 2
etc.

Though management is ignoring this, we can continue discussing, right?

The way the variants are added,
CD1 Variant 1
CD1 - Variant 1
CD1, Variant 1

seen this way and others being used, which would be the best way to add?

Amsreddevil
do not start something if you do not intend to finish it.

it could be helpful to finish this project ...
http://discogs.sitiodesbloquea.com/history?release=113613&diff=14

Amsreddevil
we can continue discussing

Please lets get on with this, even without management participation, we can at least decide some things. If they can't be bothered, we can.

Step 5, please!!!

Well, I am probably going to regret saying this:

What I don't get, right, is how there seems to be no time by a certain database manager to carry on with this project, a relevant project to the current state of affairs, but has time for all these 'little' side projects like the video/film version of this site, the equipment site or whatever it was and all kinds of things that are actually irrelevant to the discography that is discogs at this point. I would think the money and man-hours spent on those could be better spent in getting this site working properly in the first place. A novel idea maybe, but not a crazy one. I really don't get the priorities here. But maybe it's just me. The fact that in all this time Nik could not even have the courtesy to comment here at least once makes me really wonder why on earth he ever started this in the first place, what was the point if this is all there is? The rest of the staff are making more of an effort to communicate lately, maybe one of them can be bothered to take over here? Or otherwise just lock this thread and forget it, have the manners to not leave us hanging any longer at least.

I personally find it to be great that energies are expended on the film project, since what we're speaking of here is not critical and tbh a project that can be abandoned without causing anything but superficial damage to the database. I want this to progress, but not at the expense of other projects.

Eviltoastman
I personally find it to be great that energies are expended on the film project, since what we're speaking of here is not critical and tbh a project that can be abandoned without causing anything but superficial damage to the database.

That is your prerogative, I still think it incredibly unprofessional as a manager, don't start something if you aren't planning to finish it, and especially don't leave a whole lot s hanging who already put a lot of energy into this. It may be superficial to you, it is a point of contention with a lot of s, constant arguments how things should be done, constant edits changing the way this information is entered.

I am unsaving this thread and will now just do it the way I think it correct. And if others complain, I will refer them to the fact management does not seem to care, so why should I?

I don't see any signs that Management are not serious about finishing this project. I also don;t see a great rush from them or a great need to rush.Edits of this information is a problem, but a small one in the grand scheme of things. I too would like this to be resolved bt recognise there are more pressing matters, that Filmogs deserves a chance.

Eviltoastman
I don't see any signs that Management are not serious about finishing this project.

THE SIGNS ARE PRETTY CLEAR TO ME! If management was serious about this, this would have been resolved more than a year ago.
How can you claim they are serious? Really?

And yes, I resaved this. I will be bumping this until management finally pulls a finger out and finishes what they started.

BUMP

Eviltoastman
, that Filmogs deserves a chance.

Not on discogs' dime though

Is it one Discogs Dime? What they decide to spend their revenue on is absolutely none of our business. That said it seems that you and my old mucka Syke are of the same opinion. Does this mean if I apply your logic that he and you have conspired against me ;)

Only kidding Syke.

I like this project being delayed indefinitely because it means I can enter BaOI data any way I want in the meantime. The way I enter is specifically designed to annoy the shit out of j____ and k_____ and l______ and s____.

ChampionJames
I like this project being delayed indefinitely because it means I can enter BaOI data any way I want in the meantime. The way I enter is specifically designed to annoy the shit out of j____ and k_____ and l______ and s____.

Indeed. I quite like the freedom and the current rules which mean that if you mess with how it;s entered, you get EI'd or punched in the ovaries.

Eviltoastman
Is it one Discogs Dime?

It's a general expression, dime includes man-hours and such. Filmogg is irrelevant to discogs as it is now, and as said by others, first fix the one thing and have it working well before starting another.

Eviltoastman
Does this mean if I apply your logic that he and you have conspired against me

Its more a question of using your noggin, every person with a bit of common sense sees the problem here, priorities.

nik
Sorry for the delay in getting to this stage, this project was delayed due to other items becoming a priority. Hopefully we can move forward now a bit quicker. Please feel free to PM me.

Maybe someone should just ask nik for a frank appraisal of the prospects for this project. Maybe just talk to him like he's an actual person and see if he feels there's much chance this will get moving anytime in the near future.

Throwing tantrums only works on adults who actually love you.

Another thread that would not have been needed if this has been concluded
http://discogs.sitiodesbloquea.com/forum/thread/540f3b1aaba9e82ea4e4386c

ChampionJames
who actually love you.

Everybody loves me, just not everybody knows it yet.

ChampionJames: k and l? No idea who they are but I get the feeling I know s and j all too well. :'(

EDIT: Actually, I think I know exactly who l is. Funtime!

ChampionJames
... should just ask

done

Sorry folks. I was aware this project was still 'open'. It felt the interest wasn't that high in it, the important / difficult collusions were still not obvious, and the potential gains were not so big that the effort spent pushing it forward was necessarily balanced.

The last time I looked at it, I wrote the following notes:

Problem Areas
General statement;

This is a quite specialised area of the site data entry with a number of reasonably complicated usage cases and opinions. When it comes to names (people or companies), making a consistent set of guidelines to cover all eventualities seems quite hard.

Noted problem areas:

Hard to resolve:

* Mastering engineer initials entered with / without symbols and with / without additional information such as studio location / type. Should be Aliases, not ANVs? (Niks note - seems complicated and hard to reconcile with other methods we use on the site. Examples such as "SST (8)" are really open to interpretation as to what was meant in the inscription.)

* URLs - as on release, or interpreted to company name. (Niks note - this is a very polarized argument that a few people have been having over the course of well over a year.)

IMHO these two above issues are tied together, and we need to get a Grand Unified theory of how to deal with them all in a consistent manner that captures the information correctly and in a useful manner.

Reasonably easy to resolve:

* Repeating strings of information in different BaOI fields. (Niks note - seems to be a desire to cut down on duplication, but maybe need to look at examples and discuss in a separate thread?)

* Separating stamped "Matrix/Runout" strings from hand-etched strings. Or require full string to be entered as one? (Nik's note - seems to be weighted toward having full string entered)

Simple to resolve:

* Clearer definitions for the roles of mastering, cutting, pressing, etc. (Can do!)

* Ordering and spacing of BAOI string. (Nik's note - try and keep it simple!)

* Representing untranscribable symbols - editorial-style brackets are better than using the Description field? (Nik's note - AFAIAA, the agreed form is using square brackets in the string. Code to guidelines?)

* Side identifiers - as on release vs as entered to the release info. (Nik's note - as on release is better IMHO, simple decision?)

Best Practices

* We enter factual information
* As on release is an important guide to transcribing information, but not always the final method
* Artists can use ANVs for variations in their name, aliases for entirely different names
* Labels and Companies can't use ANVs, we tend to want to capture everything other than insignificant punctuation / abbreviation differences as separate entities

Well. it is time to move to step 5 now, nik.

nik
Nik's note - seems to be weighted toward having full string entered

Fast decision needed - People are already breaking strings up in stamped, etched, struck out, etc. bits.

nik
Clearer definitions for the roles of mastering, cutting, pressing, etc. (Can do!)

Adding the credit role Plating By or Metalworks By // Plating At or Metalworks At would help massively. A number of companies and engineers have been identified where the credit / LCCN role can be applied in lieu of the generic Mastered By / At.

Anyway, time to ignite phase 5 for problem resolution....

To anyone who thinks splitting up matrix strings is a viable option: please take a look at this mess and tell me how it's presentable or readable in any way... Live From London
Loving the description of DVD layers in the notes!

Yukabacera
To anyone who thinks splitting up matrix strings is a viable option: please take a look at this mess and tell me how it's presentable or readable in any way... Live From London
Loving the description of DVD layers in the notes!

Just wait until someone decides to add variants from his/her copy. :D

anssisal
Just wait until someone decides to add variants from his/her copy.

...and no matter how absurd, it will happen. Oh yes, it-will-happen

sebfact
Adding the credit role Plating By or Metalworks By // Plating At or Metalworks At would help massively.

+1

sebfactAdding the credit role Plating By or Metalworks By

...will give a field day to the novice s. Ironically, at the same time I received this notification, I received a reply from a newbie telling me off for suggesting "Lacquer Cut By" credit did not exist on a release, which they added, because they're experts. Here's Cheers, for 3 more years!! (LOL)

http://discogs.sitiodesbloquea.com/history?release=587080&diff=27

Mastering SID Code: IFPI L046
Matrix / Runout: 8793272 @ 1
Other: EMI UDEN
Other (Version 1): 1-3-4-NL
Mould SID Code (Version 1): iƒpi 1595
Other (Version 2): 1-1-5-NL
Mould SID Code (Version 2): iƒpi 15BO

This is another of the many reasons this thread needs finishing.

See http://discogs.sitiodesbloquea.com/forum/thread/543cd282ea62115a8daa858e for a role proposal

Just had to bump this thread with this ridiculous example: AIR - 10 000 Hz Legend

What's wrong with that example.

MusicNutter
What's wrong with that example.

Variant 1.2, 1.4, 2 for example, does not even make any sense.

But think they are referring to barcode bit. Typical example of why this thread needs concluding and guidelines updated with more clarity on things like how to add the barcodes.

Ah... Just in about 3, so just seen that.

I've been breaking up runout info into etched and stamped.

http://discogs.sitiodesbloquea.com/Flowerz-Flyte-Talken-About-Love/release/6210378
http://discogs.sitiodesbloquea.com/Fourmost-Why-Cant-I-Have-You-Twist-A-Tast/release/6189596

Want to update a studio on this one, but baoi info Pet Shop Boys - Flamboyant - 'pressing code' - correct term? Should it be split from matrix string?

Amsreddevil
Should it be split from matrix string?

Feels like we're going 'round in circles ;)
- http://discogs.sitiodesbloquea.com/forum/thread/540f3b1aaba9e82ea4e4386c

nik
It felt the interest wasn't that high in it,

s will lose interest if nothing happens for more than a year. Think it is more a case of s giving up waiting than actually having lost interest. Please lets continue this discussion so this can be concluded and next step can follow. More s are now interested in adding this type of info than ever, would be good to have standards set how to.

Look at this:
http://discogs.sitiodesbloquea.com/history?release=2527685&diff=36

starting removing different variations due the fact it's to confusing how Discogs works at this time ...

seriously - maybe it's better to stop ing discussions like this -
I've more and more the feeling it's pure time wasting ... :[

Hi, is there any point in my reversing the change to BaOIs that has been made here: http://discogs.sitiodesbloquea.com/history?release=14008&diff=74 ? I'd split information into each variant following Eviltoastman's suggested scheme since it made the most sense to me for an archival database to give each variant a seperate and distinct entry (trained as archaeologist so have some archival experience) but this has been 'tidied' out of existence.
Also how can we prevent issues like this one: http://discogs.sitiodesbloquea.com/history?release=870#latest where SID codes have been added but with no Variant indicated meaning my copy is now Variant 2 or 3 but with no way of identifying which (Schrodinger's Disc maybe)..

Bump
nik - we need to continue on this topic, please.........

I'm so confused.. machine stamped, hand stamped, engraved, etched, hand engraved, split, ed, runout, matrix... I've read so many threads on this.. please lets get this going so we can get something laid out. Thanks!

2 suggestions whatever labelling you come up (sorry if already mentioned). my thing has been the consistency of runouts/matrix/labels in the identifiers.
1 - remove as much open text as possible and use dropdowns/selects for each combination, seeing repeated terminology such 'runout' cant be good in the db. e.g ..
"Runout + Side A + Etched" OR "Runout + Side C + Machine Stamped"
2 - use tooltip / rollovers as an example for s to show how they should look (people dont always read guidelines)

thx

Runout should remain in one string, IMO. Etched scribing could be in italics and Stamped could be standard type

mossinterest
Etched scribing could be in italics and Stamped could be standard type

The idea is not so bad but unfortunately it is not possible.

kraftberg
unfortunately it is not possible.

You got my curiosity up. Why impossible? Seems the easiest and cleanest solution, with little to no effort on the developers part, which is the top reason for shooting down most ideas.

mossinterest
Why impossible?

Not impossible, but not as easy as it may seem.
Currently the whole BAOI section is parsed as a single ‹ul› (class "ul.list_no_style") with each line being a ‹li› item. Don't ask me why that, it doesn't really make much sense. But at this time they can only apply styles globally to the text block as a whole, unless they change the whole parsing model for the BAOI section.

the concept is good though, but the only way to achieve diff formatting on 1 line is by using scripting/code or having a ui editor. adding html to database tables is probably not what db needs. i would also avoid the use of italics as its can be difficult to read on smaller screens esp mobile. it makes sense to have fields that require different formatting separated to diff fields - style can be applied globally that way. end s should be adding content not style.