What to catalogue?

Our Gallery has passed 10,000 images and it is time to talk about the next step - Catalogues.
I use the plural, because I don't believe a single catalogue that could satisfy everyone's desires could be made and would prove too unwieldy even if it could.
The greatest value in a Scott catalogue lies in it's identification of stamps and providing a number for each issue and variety. I don't believe we can achieve this with covers, there are just too many variations and unique instances. We might, however, be able to develop a standard way to describe a cover. 
Few would question the use of the Scott numbering to describe the stamps on a cover. There are catalogues of cancellations and slogans. There is a catalogue of Canada Post covers. I am not aware of any generally recognized numbering for cachets. I would like to see this site become the source of a standard numbering of cachets and for descriptions of covers.
To get there we will need to start mining the images provided in the Gallery and developing a series of catalogues. We have started with a couple of volunteers and a new section in the Gallery called FDC Cachet Catalogues. We will be starting with some recognized cachet makers. We have already started running into some interesting questions however.

  • How do we number cachets?
  • What information do we keep?
  • What constitutes a variety?
  • What do we do about generic covers?

I would like to hear from the people who expressed enough interest in FDC to join this site. What do you think?

Some Catalog Ideas and Comments

The assumption that multiple catalogs will be required should not be carved in stone before considerably more extended thought and discussion occurs.
Personally, I do not believe that multiple catalogs are needed.  I believe that we need a single data base with the capacity to generate search results of multiple kinds.  From a single data base, for example, it should be possible to generate a search for all different cachets for a specific stamp issues; all stamp issues for a specific cachet maker; or all cachets (assuming there are multiples) from a single cachet maker for a specific stamp issues.
Without any implied recommendation of the firm or the specific approach to what I am suggesting, members may want to visit www.jamesmccusker.com , and click on On-Line Cover Shopping.  That inventory tool is my idea of what a cachet catalog should provide.
That is to say that I believe that  the catalog should be an electronic one.  The advantages of going in this direction are overwhelming for me.
1.       Avoidance of printing and distribution costs at the start.
2.       Infinite flexibility for updating and correction with negligible cost.
3.       Immediate remote accessibility online or with flash drive.
The first and most important task of our group, I believe, is to design an immediately appropriate data base – what factors (variables) should be included now – and to ensure that the data base has room for additional factors (variables) that may be used in the future to improve the catalog.
For me, the critical variables are four:
1.       Stamp Issue – Catalog Number
2.       Design Number – (and a single cachet maker may have more than one design)
3.       Cachet Maker (and Not Yet Known may be necessary in many cases)
4.       Colour
EACH of these variables needs to be thought through:
1.        It may be best to include all stamps in a series issued at the same time as a single Stamp Issue for catalog purposes.  Stamp errors and varieties do not seem to me to be relevant to our catalog.
2.       Once assigned, design numbers should not be changed even if new information comes to light down the line.  The specific number assigned to a design does not seem to me to be important.  For communication purposes, what is important is that everyone should be confident that a specific design reference remains unchanged over time.
3.       There is a need to distinguish the cachet publisher from the artist/illustrator, and to determine how these will be treated in the catalog.  For example, Gladys Jackson produced cachets that may have been produced by Chickering, Bower, Sweeney or Smith.  Should these ALL be cataloged simply as Jackson?  Should the Chickering designs be cataloged as Jackson/Chickering or as Chickering/Jackson?  Do we need separate variables for publisher and designer?
4.       Simple colour listing may not be definitive.  “Red and Blue” for example could confuse a Red Frame with Blue Picture and a Blue Frame with a Red Picture.  Indeed, the data base may need to allow for multiple colour variables in the form of Colour Variety First, Colour Variety A, Colour Variety B, Colour Variety C, etc.
In my mind, the critical numbering should be Stamp Issue and Design Number.  That is, for example, 379-10 to denote Design 10 for the 1958 Founding of Quebec Issue.  I do not believe it is at all important which design is labeled 1 or 10 or 20, only that it be uniquely identified.
The cachet maker should be part of the catalog information, but not part of the numbering scheme, because cachet maker does not add any unique identifier to the basic number. 
If there is only one known cachet colour variety, the design should be known as 379-10, for example.  If two additional colour varieties are known, they should be designated as 379-10A and 379-10B.  If another is discovered later, it becomes 379-10C.
The only reason that I know of to include uncacheted covers in an FDC typology is to illustrate corner cards for early servicers of FDCs.  It would seem that this might be OK up to a specific point in time, but is not useful as an general approach.
Cachet catalogs have been criticized in the past both for including and excluding generic designs.  I think the catalog user should have the option of including or excluding them.  This implies that generic designs should be assigned high numbers on the design scale.  It also raises questions about whether it is possible to design a typology where the same generic image has the same design number for all stamp issues for which it was used.
Subject matter variables seem to me to be the most problematic aspect of what has been discussed.  Presumably, stamp catalogs do not ordinarily classify by the subject matter of the stamp, and I am not sure why we would want to classify by the subject matter of the cachet.
I do not think that subject matter classification can be handled with a single variable.  I see a moose, you see an animal, he sees antlers, and she sees a photograph.
Nonetheless, if there are those who would find this useful, I suggest that we approach this with MULTIPLE subject matter variables.  Lots!!
In the scheme of things, I think that this is the kind of thing that could be handled as a second priority after the basic catalog information was established.
This is another problematic issue in that while slogans may be usefully illustrated, it is probably not necessary to illustrate every location for which a first day cancel is known.  It surely is not necessary to tie this in any way to a cachet design or maker.
No comment has been posted, as yet, regarding whether or not market values should be part of the catalog.  For my two cents, I suggest NOT.  Nothing wie might do would require mpre time and produce less real value than that.

Doug Holmes included a suggestion that a Character Value like COL be included for Colorano.  I suggest that we avoid such two-step approaches to data base design.  Space is cheap now – and will be cheaper over time.  Let’s allow lots of room for each variable in our database so that we can have Colorano as a cachet maker, and Photo of Red Moose with Blue Bow in Her Antlers as a subject variable.

What is appropriate?

I'd like to focus on your statement of where to start - design an immediately appropriate data base.
I could have done this years ago. It is, in fact, where I started. There is a difficult concept to wrestle with embedded in it however. "Appropriate". If I had done this myself, I could have catalogue'd my rather sparse collection. This would have been helped me when I went shopping, but wasn't going to tell me what I wanted to know from a catalogue. My first question was - So how many RoseCraft cachets are there?
What we have in the Gallery IS a catalogue - of the cover collections of the ten or so people who have uploaded images. It is backed by a database, and is searchable by any or all of the items you mentioned - as long as the people putting them in entered the information. So if all you want to do is search, all we need to do is update all of the descriptions.
Where I have run into problems is figuring out what this is to be a catalog of. My inclination is the envelope with the interesting part being the cachet. The stamps on it are incidental. You referenced this problem when you asked if is was possible to design a typology where the same generic image has the same design number for all stamp issues for which it was used. To my mind the purpose of the catalogue is to provide a unique identifier for each cachet.
I have had discussions with others who feel that it is not a FDC if it does not have postage and a correctly dated cancellation. I agree with this as a definition of a FDC, but not with what the catalogue is able to deal with.
Pricing actually helps with the discussion. A 'Cachet' catalogue can provide a complete listing of all known cachets, and provide unique numbering. It cannot provide prices unless it deals with many other variables - actual stamps used, their configuration, cancellation types, placement, clarity, city and a host of other factors (I saw one cover commanding a premium for being signed by someone relevant to the subjecct). If someone wants to produce a price list or catalogue their own collection, the 'Cachet' catalogue number should be sufficient to describe the cachet and the rest is up to the particular use of that list or catalogue.
Even assuming that we are talking about cachets and not covers, I'm still not clear on "Appropriate". I see one use of the catalogue as being the definitive source for a catalogue number. The basic uses I see are:

  • Look up a catalogue number. This means you have a number from a sales / auction catalogue or your want/have list and you need to find out something about that cachet. Enter the number - see a cover and the associated information.
  • Search for a catalogue entry for a given cachet. You have a cover in hand and want to determine what the catalogue number is so you can put it up for sale, describe it in your collection notes or reference it in an article.
  • Update the catalogue with a new entry. Once you have searched unsuccessfully, it would be useful to be able to add an entry into the catalogue. This would mean providing sufficient information that any other searches would locate the cachet and avoid duplicates.
  • Researching. This can take many forms. Looking for cachets with characteristics to see if it exists for one. Looking for all of the cachets that share some attribute (from maker to a design element).
  • Browsing. If there is enough supplementary information about each cachet, simply paging through the catalogue.

If we want to determine the "appropriate" elements to track in the database, I think we need to discuss the uses of the catalogue. This board is a good place for this and other similar discussions.

FDC Catalogue

I like the genesis of what is being proposed. I also believe that we need to make sure that we have a sure and solid catalogue numbering/identifying system. Indentifiers become useless very quickly if they are too complex, require too many returns to an index to identify/understand them or are unwieldy in lenght. The catalogue is not just for us who may know the information it is also for the general collector, dealer and/or researcher who may have a peripheral interest and might quickly lose their interest in this fascinating area of our hobby if it is too complex.
I also would like us to consider all aspects before venturing out.
Some I have for example are:
1. How do we classify cachets that, although may seem to originate from different cachet makers end up being somewhat the same. JCR and Rosecraft.?
2. How do we catalogue maxicards?
3. Do we separate US and foreign cachet makers from Canadian cachet makers? Do we know if they are foreign, US or Canadian?
4. How do we deal with joint issues? The 1959 St. Lawrence Seaway has so many varieties and possibilities, combinations and permutations it's scary.
5. What do we do with cacheted covers that are used outside of the official FD of Issue city?
6. One of the things that is important is that we use generic terminology and we catalogue using this terminology if we want our south of the border friends to collect Canadian FDCs. This includes separating First Cachets in the catalogue.
7. Don't forget using Scott catalogue numbers in any classification system requires Scott's written permission.
8. How do we deal with hand-painted, hand illustrated cachets?
9. What do we do with add-on cachets?
Chas. (cjgverge@rogers.com)

Uncacheted covers do not

Uncacheted covers do not really need a catalogue - the Scott number(s) is generally enough to classify the cover. Other details, such as the city where it was cancelled, would not be part of any catalogue number.
Cachets that are designed for a particular stamp or series are partially identified by the subject itself. The cachet maker is generally enough to identify the cachet. In some cases, for example Christmas issues, there may be multiple cachets for a single issue, but these can be handled as variations. The problem that arises is that there are many unknown (or unidentified) cachet makers.
There are also general use cachets, not related to a specific stamp issue. These may also come from known or unknown cachet makers.
We will need three sets of catalogues.

  • By stamp issue - with cacheted covers that relate to the issue.
  • By cachet maker - with all cachets by issue and a separate 'issue' for general
  • By subject - for general use cachets - whether the maker is known or not.

Numbering should be consistent. There are three parts to a number - the cachet subject, the cachet maker and something for variations within these two.

  1. Where the subject is a stamp issue - the value would be the Scott number. Where the subject is more general, a value that describes the actual subject - regardless of the stamp on the cover would be used. The format of the value should keep it from being confused with a Scott number. The Dewey Decimal system is good for identifying subjects across a broad range but this is both wider than necessary and confusing with a Scott nujmber (e.g. 921 is either biography or the 10 cent blue Weathercock definitive).
  2. Where the cachet maker is known, a character value would be provided, for example Col for Colorano or NRC for N R Covers. A list of assigned values would be kept. Where the maker is unknown, a number would be assigned. A list of assigned numbers would be kept. If a maker is unknown, but several cachets can be attributed to the same maker, they would use the same number. If a maker becomes known - the number would be retired.
  3. Variations come in two types in this scheme. They may be intentional, such as multiple cachets for a series that are readily distinguishable, or they may be unintentional e.g. colours or corrections.

The target numbering will be:

  • Scott number.Cachet Maker where the maker is known and the cachet is specific. E.g. 494.RoseC
  • Scott number.number where the maker is unknown and the cachet is specific. E.g. 494.152
  • Subject.Cachet Maker for general use cachets from known makers. E.g. BC12.RoseC
  • Subject.number where the maker is unknown and the cachet is general. E.g. ED91.144

Before committing to a general cachet numbering or a variation numbering method, I suggest that we have enough material in our Gallery to make a start on the catalogues and leave the actual numbers until we have material to work with. To start, there will be two sets of catalogues - one set for covers with known cachet makers, and another for all covers for each stamp issue. General use covers will appear in the second set, but in a separate section. Once enough of them have been identified, a separate album will be set up to allow numbers to be assigned.  Keyword tags for the Scott numbers, cachet makers and whether the cachet is general will be used to make assigning the final numbers easier.