Quoting Kathy:

Quote:
In this case, copyrighted material is not allowed into the online database.

We all know that, and this is the reason users prefering IMDb data do not contribute to Invelos database.

On the other hand, I hardly understand how IMDb could copyright the name of an actor. Ken would be able to do that, as he uses name variants that are generated by his rules and exist only in his database.

To illustrate what I say:

Here is a non existing variant used in Invelos database, Francois Berleand:



Now, I use this Invelos variant for a google search. I get no result with this variant. All the results use the correct variant, François Berléand. Can IMdb copyright a name used by the actor himself and all the other sites (Wikipedia, Allo ciné, Comme au cinéma, Le Parisien, uniFrance Films...)? I do not think so...

Quoting surfeur51:

Quote:
On the other hand, I hardly understand how IMDb could copyright the name of an actor. Ken would be able to do that, as he uses name variants that are generated by his rules and exist only in his database.

This is one of the misunderstood concepts--made worse by their black-and-white wording on the site--that gets walked around due to too many concerns (same goes for so many other sites), and to keep from complicating things by splitting hairs I assume, Ken just avoids the issue entirely for safety's sake.


I'm not a lawyer, but from what I had to learn through other repeated issues over the years (copyright of material used in fan-games,) you are right in that a list of common names cannot be copyrighted. And even if they could be, the MPAA or similar would have a higher president for copyright than IMDb since they are borrowing the data as well. You can trademark individual names, but you can only copyright original ideas and creations.

That said, about 15 years ago, give or take a a few years, litigation went through in the United States giving an implied (but fair-use) copyright to anyone who creates something original online. This includes artwork, music, software, writings, et al.  This means if I create a piece of software and post it online, it is automatically considered, in the most basic legal form, as copyrighted for being my own, original work. That would be the same for IMDb.

While the names listed may not be directly in the bounds of copyright, it can be argued that the software (particularly the linking system) is protected by copyright and therefore the names cannot be used via the exact same method used by IMDb. Anyone who posts on the site (as well as most forums I have seen,) if they actually read the ToS for the site, will see that by posting on IMDb they agree to give up their own legal claim of copyright turning it over to IMDb as their own property.

Lastly, though, as I said before, it is still only an implied fair-use copyright. Just slapping a © somewhere doesn't mean you are secure. That's more of a "just letting you know" symbol. To be protected from legal action, you have to register any copyrighted material with the legal office of your country.

Quoting Kathy:

Quote:
Despite many requests over the years to make changes regarding IMDb, Ken has not implemented them. I'm not sure why, I would guess that it is a legality issue.

I can tell you why...cost.  IMDb charges a licensing fee to use their data.  Last I checked, the lowest priced plan was $15,000.  If Ken chose to use their data, this program would probably require an annual fee to use the on-line.  Speaking for myself, I prefer it the way it is.

Quoting TheMadMartian:

Quote:
Quoting Kathy:

Quote:
Despite many requests over the years to make changes regarding IMDb, Ken has not implemented them. I'm not sure why, I would guess that it is a legality issue.

I can tell you why...cost.  IMDb charges a licensing fee to use their data.  Last I checked, the lowest priced plan was $15,000.  If Ken chose to use their data, this program would probably require an annual fee to use the on-line.  Speaking for myself, I prefer it the way it is.

I prefer it the way it is as well. I wouldn't hate the idea of using their data... even if Ken decided he wanted to go that route... I would keep my cast/crew lists local then.

Quoting TheMadMartian:

Quote:
IMDb does not copyright the names, they copyright the 'data compilations'.

I do not want to use their data compilation, specially their roles that are often different from credits. Just use the name each time it is obvious that is is the name usually used by everybody, as in my example.

I really cannot see the interest, in the case I gave above, to have four variants in the CLT (as it is the case: François Berléand, François Berleand, Francois Berléand, and Francois Berleand), then open a common name thread to know which one is the most used following Invelos rules. In one year I'll have perhaps an answer that we'll perhaps change two monthes later since this actor is still in activity.

The correct one is obvious, it is easily found with a google search, and it happens to be IMDb name in 99,9% of cases, with no copyright problem. If we just avoided those silly variants, we would eliminate from the database thousands of non linking data with no extra (and useless) work with CLT and common names threads.

Quoting TheMadMartian:

Quote:
IMDb works because they use a unique identifier for each actor and, behind the scenes, link all variations of that actor name to that identifier.  If I type in Francois Berleand, it links back to nm0075710, which has a display name of François Berléand.  Invelos needs something similar.  Not being a programmer, I have no idea how hard that would be to impliment. 

It would likely be a total rebuild of cast and crew from the ground up while at the same time maintaining the current method until all names are transferred over.

Transferring the names to the new system might not be too difficult at first, but maintaining the lists (making sure when the person enters John Doe as the cast member links it to the proper one in the new database) will likely be just as difficult to maintain as the current system unless a new verification method is used which could be tough (think viewing CLT but needing to be automatic with each John Doe group already separated.)

Quoting TheMadMartian:

Quote:
If I type in Francois Berleand, it links back to nm0075710, which has a display name of François Berléand.

And how do you think IMDb choose the display name, which happens to be the correct one in 99,9% of cases ? Only by knowledge of the language of the actor or by google search...

IMDb display name has the same problematic than common name in Invelos database. Dvdprofiler linking system would work perfectly with a correct list of common (display) names. This list could be Invelos based (as is CLT), and filled using the knowledge of all users, Spanish, Swedish, Norvegian,  French,... users giving correct spelling for actors of their country . Common name rule (most used) could be still used for real variants (Rebecca Romijn/Rebecca Romijn-Stamos). A new rule about birth year changed to specific ID for same name/different actor would also avoid that a lion (Shaka in "Flyboys") plays a girl (Shaka in "Flash Gordon")) as we can see in CLT.

This list could be implemented tomorrow without any program change, and would need far less work from users than those useless common name threads. For next program version, Ken would just have to create a link between this list and the online database, without having to change the existing linking system, and the list would be used by contributors as Parsec explained on top of page 3 of this thread.

T!M,

When I contribute profiles I get BYs popping up on almost every single one. In most cases there are several up to several dozen. In almost all of these cases I get a prompt from invelos to document these BYs before submitting them.

So far I have run across thousands of lthese valid/invalid (who knows) BYs so I'm guessing my estimate  of tens of thousands might actually be small.

I don't know how these invalid BYs got into my database but I didn't add them. They must be in the online database.

Another problem is that valid BYs often are not on the BY thread. Although helpful, the BY thread is lacking in the ability to track valid/invalid BYs because there are people who post in it and others do not.