Quoting skipnet50:

Quote:
There's the problem, YOU don't see it, Hal. You missed on you are argumentative in the extreme. But...

Skip<shrugs>

You're right.  I don't see it.

Please enlighten me instead of making accusations.

Quoting EnryWiki:

Quote:

I still mantain Ken should chip in and give us some sort of standardization for mixed case names, either by a statement/rule or by a filter on the Name field in contributions.

I think he already has.  Rather that try and get us all to do it the same way, he has chosen to have the program ignore variations in capitalization.  That is, of course, just my opinion.

Quoting EnryWiki:

Quote:
Quoting skipnet50:

Quote:

[...]
this an issue that Ken does need to fix so that the local will deal properly with upper and lower case entries.

If Ken "fixes" that, as you ask, DeVito and Devito wouldn't link any longer.

Far, far better that they link than that upper and lower case go back out of whack again.  I don't want to deal with the dueling LeVar vs. Levar contributions again.  Ken would have to be nuts to accept the suggestion that he "fix" the upper and lower case entries.  It's working far better the way it is.

Just to add a bit more data to this; If in your local database you have an entry for 'Danny DeVITO' and you download a profile with 'Danny DeVito', what do you get? Hint ... your database does not change to 'Danny DeVito'. If there is an update the program will show the change from 'Danny DeVITO' to 'Danny DeVito', but it will not change it in your local database! The next time you submit the profile the master database will get updated to 'Danny DeVITO' for this profile.

pdf

Quoting skipnet50:

Quote:
And therein lies the difference AND the problem, gard. You want linking, screw the data accuracy, and I want the data accuracy FIRST and Foremost and the linking is secondary, nice to have but NOT at the expense of inventing data.

Skip

The problem with your argument, Skip, is that it focuses totally on data accuracy purely for accuracy's sake to the exclusion of the usability of those data.  It just doesn't sound to me that collecting data is very worthwhile if you can't use that data to do something, like link all the titles a person appears in.  It's all well and good to know that Danny Devito is credited in movie A, but of what use is that information if you don't know that he's the same person as the Danny DeVito who appeared in 100 other movies?

I just don't see any value in collecting data as an end in itself.  Data is merely a tool which we can use to do something.

Quoting kdh1949:

Quote:
Quoting skipnet50:

Quote:
And therein lies the difference AND the problem, gard. You want linking, screw the data accuracy, and I want the data accuracy FIRST and Foremost and the linking is secondary, nice to have but NOT at the expense of inventing data.

Skip

The problem with your argument, Skip, is that it focuses totally on data accuracy purely for accuracy's sake to the exclusion of the usability of those data.  It just doesn't sound to me that collecting data is very worthwhile if you can't use that data to do something, like link all the titles a person appears in.  It's all well and good to know that Danny Devito is credited in movie A, but of what use is that information if you don't know that he's the same person as the Danny DeVito who appeared in 100 other movies?

I just don't see any value in collecting data as an end in itself.  Data is merely a tool which we can use to do something.

I focus oin data largely because i don't need a linking system, MOST of the time. I have as much or more data in my head about film as we do in the database.. But, Ken, if the data is not accurately represented within the context of our Rules and without users continually trying to create their own interpretations, then the linking system will NEVER be of any value to anyone. This is also yet another reason why the Common Name system based upon user input is no good, it is dependent on users inputting what they see NOT what they think it is. The simple association would work much better as a linking system.

Look at the base of this discussion, ken, it is ALL about user interpretation of data, NOT the REAL data that is displayed. it is about what the users imagine it to be.


I am also all about usability, Ken but in order to generate usability one must first have an accurate base to work from, otherwise you get IMDb. That might be good enough for you, but not for me.
Skip

Paul:

Stop being obtuse. According to our Rules LEVAR=Levar, to argue anything else is utter stupidity. Unlike some users I also have absolutely no conceren about what a person's "real" name my be. it is completely irrelevant to Profiler, wea re not constructing a Family Tree for anyone...what is relevant is the Credited data and only the credited data. That is the foundation for the Credit Lookup Tool and we are building the CLT on sand as  is demonstarted by by comments such as Paul's and others, This si one of the reasons the CLT does not yield accurate results and NEVER will, I am surprise that someone has not tried to create Marion Morrison as a Common Name for John Wayne.

Build the foundation on rocks and the house will stand, build it on sand and it goes out with the first tide.

Skip

Quoting T!M:

Quote:
So I do agree with EnryWiki in that I feel Ken should let us know how to deal with this: I think there's quite a lot to gain. Even just the thought of never having to rehash this same debate all over again every few weeks seems like sheer heaven at this point...

I also agree with that. But are we sure we are speaking of EnryWiki's opinion, and not of enryWiki or Enrywiki. In fact the capitalization rules for this user have not been documented, and nobody can be sure it is the accurate, knowing that we have absolutely no concern about what's a person "real" name  may be... An CLT says nothing