English > Support

Adding movies, etc.

<< < (4/4)

Anson:

--- Quote from: patch on September 13, 2009, 05:22:39 am ---
--- Quote from: Anson on September 12, 2009, 05:00:41 pm ---I tried creating a new record with only an URL and neither title nor original title, and it didn't work.
--- End quote ---
Agree this is counter intuitive & workarounds using dummy titles to create record without confusing PVD matching logic would just add to the confusion.
--- End quote ---

The problem is probably this "internal index and matching" which rick described somewhere above. PVD has to do something like that and probably needs either title or original title to do it, and thus can't add an url without at least one of those other two fields.


--- Quote ---A good solution / feature request would be supporting url in "add from clip board".
--- End quote ---

That would work nicely for my purpose:
- go to the website and search for a movie
- copy the URL
- either "add URL from clipboard" or add an URL with the "new movie master"

But it also would have to take care of exactly the same problem (not having a title), and thus be no "standalone" solution without other related changes to handling records without titles and/or at least internally use a dummytitle derived from the url or something.

Since i like experiments, i once created a new record with the url as title, but that crashed the IMDB plugin (some endless searches with repeating parts of that url, etc). But maybe that could be a method to "add movies by URL" if plugins would check whether a title looks like an URL. then PVD itself wouldn't have to take care of titles too much and the work would be done by the plugin ... (just an idea; might also be total crap since i have no idea how all this big software package looks internally)

rick.ca:

--- Quote ---The problem is probably this "internal index and matching" which rick described somewhere above. PVD has to do something like that and probably needs either title or original title to do it, and thus can't add an url without at least one of those other two fields.
--- End quote ---

Yes, as I've said several times, it's by design. For the methods of adding a movie we've been talking about, the program needs to check for duplicates and determine how they will be handled, run a search to find the online source, etc. The fact there is nothing in the current design for adding a record by URL alone doesn't mean there's anything complicated about it. In fact, it should be considerably simpler. If there is already an invisible record with the same URL, then it will be made visible and updated. If not, a new record will be created and all the information (including title) filled from the download (since all the user has is a URL, there is no concern about overwriting existing user information).

The point is, until the adding of a movie by URL alone is provided for in the design, you have to respect the design. If you have a URL, by all means use it. But the only way to do so is to add it to an existing record. Existing records, by design, must have titles. If and when the capability to add a record from a URL alone is added, it will not involve the user in creating a new record. The program will use the URL to download the information and create a record. If the download fails, the process will likely be aborted and no record will be created.

Anson:

--- Quote from: rick.ca on September 13, 2009, 05:43:34 am ---What's clearly counter-intuitive is the idea of creating a record with no name. When I used the word "intuitive," the context was not just the program, but record keeping in general. I think most people can appreciate the consequences of putting a paper document in an unmarked folder and filing the folder in a random location in a filing cabinet. Why would their expectations be any different here?
--- End quote ---

it should be different because we work with computers and not with pen&paper and a physical storage. On a computer you have many options to file something and search for it besides a title, eg an advanced search for
--- Code: ---URL IS NOT NULL
and
ORIGINAL TITLE IS NULL
--- End code ---
and then fill that data by using a plugin.

If you have a secretary in your office, it would be counter-intuitive to give him/her a note with a reference number for some document in another office, and him/her telling you that he/she can't get the info and copy it for you unless you go there first yourself and copy the title of the document to the note first, or at least write anything there which looks like a title (which later is not needed at all when retrieving the info by use of the document's reference number). It would be really blond secretaries (sorry for not being PC :-) who can't do that themselves and require the boss to give such a title just to temporarily file the note in their hand until the real document is retrieved (by ignoring that title and looking at the reference number only) and replaces all data on the note.


--- Quote ---I think a polite warning is counter-productive in the cases where a smack in the head is called for. But I suppose that's just me. ;)
--- End quote ---

I just tried ...
- if i input all data for a movie and leave the title fields empty, the record and all the work is lost. that would be a really big smack with a hammer.
- if i already have a record, i can change everything and even delete both titles as well as year and ID, and it will still be saved! Thus: why shouldn't an (optically identical) record be saved and give a hammer-smack when i write all the info for the first time?



--- Quote from: patch on September 13, 2009, 07:39:41 am ---It has a creation date and unique internal id so can readily be displayed and found.
--- End quote ---

although i agree with the rest of the post (see my suggestion above for searching for the record), there seems to be no unique statical internal ID which can be shown to users in any kind of way. that would be what i have looked for, but as i understood from another post, such IDs are generated some kind of dynamically by using info like title and date.



--- Quote from: rick.ca on September 13, 2009, 08:22:54 am ---If you want to do this now, you can open a new movie record, paste in the URL and type an "X" for the title, then save and download. I don't find typing the "X" particularly demanding.
--- End quote ---

when i suggested using dummynames, you didn't like that idea, and the "X" is just that: a dummyname. In addition, using the same name "X" for all movies leads to problems, not only when importing a CSV file with a lot of URLs but also because of endless questions whether the existing entries for "XxX" and similar were meant.
That's why i had suggested in a previous post how to create dummytitles for importing URLs from a CSV: "dummytitleNNNNNNN" with NNNNNNN being the number from the IMDB URLs (of course, this is only practical when i do it automatically, eg with my new "myvovies to pvd" converter which works now :-).


--- Quote ---The idea the program should allow a user to create a record without a name, or warn them they're attempting to do so is silly.
--- End quote ---

when i can delete the title later (i have tested that), it is silly if i can't download info including correct title without first creating a new record with any (even incorrect) title.
When accidentally forgetting to enter a required field (or even more so, when that field seems to be not so required in the end), it is common practise and politeness to not smack users with a hammer and deleting/ignoring the info they have entered correctly, but warn them instead. I consider this behavior in PVD to be a bug and not only a feature. (and i consider the fact that it is required at all to be only a featuire and not a bug)

Anson:

--- Quote from: rick.ca on September 17, 2009, 07:53:26 am ---If you have a URL, by all means use it. But the only way to do so is to add it to an existing record. Existing records, by design, must have titles.
--- End quote ---

not exactly...
you can delete titles from existing records. they only need to have a title to create them.

edit: just did an additional experiment with strange outcome:
import from CSV: only an URL field -> a new record is generated with no title and no URL

Navigation

[0] Message Index

[*] Previous page