Quoting Kathy:

Quote:
Quoting TheMadMartian:

Quote:
Quoting Kathy:

Quote:
Not only that - who is to say that a space indicates the next letter/word belongs in a different field.

Quote:
Unless you know me, who the hell knows which field any of the letters belong? 

That's kinda my point.  Without actually knowing the person, there is no real way to know so how do we decide how to parse the name?  Rather than guess, I use the fact that the data was presented in multiple parts separated by a space.  If you, the general you not you specifically, can find a better way that doesn't require research by the contributor, I am happy to do it differently 

Which is why I think the most logical way of parsing is to put the first letter/name in  the first field, the last letter/name in the last field and the rest of the letter(s)/names(s) in the middle field.

Of course, that's me...

Same here, the example of "A.A.A. Smith" should be A.A.A//Smith.  The name will display just like the screen credit since a space is injected between fields for display (and CLT) purposes.

Quoting Kathy:

Quote:
I think you misunderstood me Scooter.

Just to clarify what I mean by "first letter/name in  the first field, the last letter/name in the last field and the rest of the letter(s)/names(s) in the middle field", I would parse the A.A.A. Smith name as:

A. / A.A. / Smith

Ahhh....


Well with that,  what would end up in the CLT is "A. A.A. Smith" as the screen credit  and not  "A.A.A. Smith".  Because for display purposes and the CLT a space is injected between fields.  So IMO the way you would do it is incorrect since the exact screen credit is not represented in the online profile.

Quoting SoylentDave:

Quote:
I was asked for a page or two ago, but I didn't have a screenshot handy. Here is the specific example in my database. When I started the thread I didn't have the name handy so I used the generic A.A.A, Smith. Mr Adams is currently in the database as M.A.C. Adams, M. A.C. Adams and M. A. C. Adams for various titles, all of which (as far as I recall are all shown onscreen similar to this screenshot.

Looks to me like that should go in as M.A.C.//Adams, anything else would add spaces that are not there and con-volute the CLT

I agree as well.

Quoting T!M:

Quote:
Quoting Scooter1836:

Quote:
Looks to me like that should go in as M.A.C.//Adams

That sounds lovely, but we need to address the other half of the question as well, which is: then how would we list it if the on-screen credit read M. A. C. Adams? Because if you're entering this as M.A.C.//Adams, while next week entering an M. A. C. Adams credit as M./A. C./Adams, then you'll end up with two separate, non-linking entries for the same person in your database - and that can never be the answer.

This is not a question that should be answered based on one particular example, or while looking at just half of matter - instead we need a general, consistent approach that works on all of these names. We need to decide how we handle three-initialed names, and we need a way that ensures that we don't end up with separate, non-linking entries for the same people if they're sometimes credited with, and sometimes credited without spaces between the initials. That was the sole reason behind the filter for two-initialed names as well: to cut down on double, non-linking entries for the same people. So that should be the focus here, too.

Getting back to this particular example, I could happily go along with entering M.A.C. Adams as M.A.C.//Adams, but ONLY if we're listing M. A. C. Adams as M.A.C.//Adams as well (using the "credited as" field to retain the difference) - again, to make sure we don't end up with double entries for the same person. You can't address one without the other. So then the rule of thumb should be: whenever you see a credit that uses three initials for the first name, put them all in the first name field, while using the "credited as" field - if needed - to retain the on-screen spacing.

And that would be fine. Of course, the joke then is that we'd choose for the exact opposite of the initials filter that we already have. Because the current filter actually boils down to: whenever you see a credit that uses two initials for the first name, split them over the "first" and "middle" fields, while using the "credited as" field - if needed - to retain the on-screen spacing.

Well that is already handled.  First of all when the credit is entered, it is to be entered in exactly as credited.  So however it is entered would reflect what is on the screen. 

The filter does not handle this combination, presumably by design since other initial based combinations are not covered so we have to handle it normally. The "Credit As" field would be used only for common name purposes just like anything else. you would only have "M. A. C. Adams[M.A.C. Adams]" if the credit was M.A.C. Adams and the common name is M. A. C. Adams and the screen credit is M.A.C. Adams.  On the opposite side if the common name is M.A.C. Adams and the screen credit was M. A. C. Adams the credit would be listed as "M.A.C. Adams[M. A. C. Adams]"

It is all handled by the current rules, entering in what is credited and entering in the proper common name when applicable.  It would be just like any credit that is not being manipulated by the filter today.  Just like the several other "initial" conditions that the filter does not address by design.

I am sure that we need to retain the screen as it is on the screen.

Handling it as a normal credit is the only thing we can do because:
a) The rules do not state we should manipulate the results
b) We have no clarification from Ken that we are to handle it differently
c) The contribution filter does not address this condition.

As I stated above, since other "initial" conditions are are not handled by the filter by design we must assume this is another unless explicitly told otherwise by the database owner.  Just like the other "initial" conditions not handled by the filter, this should be handled like any other credit.

One other thing I should point out is that threads have been pointed out with polls on how to manipulate the data for consistency as well as discussions within this thread.

Some of those data manipulations for consistency I may agree with. However these changes cause conflicts with the existing rules so IMO they cannot be implemented unless we get confirmation from Ken via a clarification, a rule change, or change to the contribution filter.  So until one of those things happens the rules need to apply as is.

An example of the contradiction is if you contribute "M. A. C. Adams[M.A.C. Adams]" and "M.A.C. Adams" is the common name I could accurately vote NO to that contribution since "M. A. C. Adams" is not the common name.  And in the opposite is if an existing profile had "M. A. C. Adams[M.A.C. Adams]" and "M.A.C. Adams"  and someone contributed that to be  "M.A.C. Adams"  because  "M.A.C. Adams" is the common name that would be a correct contribution.

So at this time the point of consistency is applying the rules as they exist today.

Polls are very useful in gauging the community opinion, and in some cases those polls are valid for justifications for contributions.  But when a poll generates an opinion in conflict with the rules its use can really be only used to attempt to get something changed, which is this case IMO since we are talking about name manipulation.

Quoting T!M:

Quote:
...we need a way that ensures that we don't end up with separate, non-linking entries for the same people if they're sometimes credited with, and sometimes credited without spaces between the initials.

I do not see how this space problem is different of accents. And you do not care to have non linking variants when same people are sometimes credited with, and sometimes credited without accents

Quoting DJ Doena:

Quote:
That is simply not true and you know it.

If you have CA "Louis de Funès" and CA "Louis De Funes" that doesn't mean they can't link. They just have to have the same common name. Even if in the end that happens to be the (gramatically wrong) Louis de Funes]

I still do not see the difference. For M.A.C. Adams, you can also use the common name you find in CLT.

"M. A.C. Adams" is credited in the following 9 titles (21 profiles)
"M.A.C. Adams" is credited in the following 10 titles (16 profiles)
"M.A. C. Adams" is credited in the following 0 titles (0 profiles)