Activity › Forums › Creative Community Conversations › What Ever Happened to Metadata?
-
Simon Ubsdell
May 19, 2016 at 4:43 pmYes, I’ve been hammering away on it today and doing some back-to-back comparisons with KFP for a point of reference. As I say, the speed factor is a really noticeable bonus because you’re not waiting for it to build a library as such.
From the FAQ answering the question whether or not it’s a MAM: “Well, sort of. Many of the features like tagging, descriptive metadata support and filtering give it a MAM kind of feel but its scope is currently rather a very light-weight support of production processes rather than long-term archival, although people may still find it useful for that as well. The main point is, that is is much more light-weight than typical MAMs because it does not require an import/ingest step before you can do something useful with your material. That means there is not really a concept of “inside” or “outside” of Kyno, which also means there is no global search of all content Kyno has ever seen.”
Simon Ubsdell
tokyo productions
hawaiki -
Bill Davis
May 20, 2016 at 6:22 amFor those content to shift their focus out of the current task – editing – to undertake a different task – searching – this makes great sense. And letting a separate producer(s) work on content markup that feeds one or more editors is a nice feature.
It’s also one that quite a few high level X editors appear to be enjoying these days.
Ref:
And:
And:
That said, I still prefer the X system that slots search directly inside the editing app.I want to search to enable the flow of my editing “in the moment”. Not as a separate task.
Switching focus between separate discrete tasks – for me – adds “friction” into the editing process.
And to me, editing is all about staying “in the zone” as much as possible.YMMV
New signature under construction and coming soon. Please stand by…
-
Misha Aranyshev
May 20, 2016 at 7:58 amTech gets wide adoption when it’s useful and relevant. Timecode with all its limitation is useful and relevant. Slate metadata is crucial for feature and scripted TV but apparently this is too small a market for this metadata to have a working implementation. Location metadata is useful for docs and news but irrelevant for features and scripted TV. In short, there’s no money in metadata.
-
Simon Ubsdell
May 20, 2016 at 10:13 amI hear what you’re saying. I too am one of those editors who prefers to work inside the NLE and I totally agree about the importance of flow.
But I think the workflows we are discussing here are about the preparation of your media before it ever gets to the NLE rather than a process you would be accessing while editing, so I’m not sure it’s fair to say they are disruptive to the editing flow. Given that usually the purpose is to help you find your material faster once you are in the NLE, I’d say the opposite is true – in other words, by managing your media more effectively, they are assisting with the flow once you get down to work.
Your observation does make me think of another major consideration that I don’t think anyone has yet raised here, and that’s the fact that regardless of which editing platform we are using and whether or not we choose to “pre-prepare” our media, most of us are still doing it twice over. Which is to say that we are organising at the Finder level and then again once we get inside the NLE.
Over here at Tokyo Towers we use Digital Rebellion’s indispensable PostHaste utility which allows us to instantly create complex Finder level project organisation whenever we start a new project, using a house style that we have decided upon. But useful as this is, it still means we have to organise our media all over again once we get into the our chosen editing (or other) application.
It would be great to see ways that would eliminate some or all of this redundancy so we could media manage once and only once. But necessarily that would mean organising at the Finder level not the application level since we want our organisation to ripple through to wherever we happen to be working. That’s obviously going to be tricky but that’s where all those clever developers come in 😉
Simon Ubsdell
tokyo productions
hawaiki -
Misha Aranyshev
May 20, 2016 at 10:23 amI quit building elaborate folder structures long time ago. For any given project I have “Original Camera Footage”, “Production Sound”, “Workprint”, “M&E”, “GFX” and “Deliverables” all on the same level and without any subfolders inside. Works just fine for a30 sec spot and a 2-hour feature. I paid Videotoolshed to make an app that timestamps my “workprint” with the same time the original has and I fill in slate numbers in an NLE — it gives me enough metadata to find my way around the footage.
-
Jeremy Garchow
May 20, 2016 at 12:13 pm[Simon Ubsdell] “Your observation does make me think of another major consideration that I don’t think anyone has yet raised here, and that’s the fact that regardless of which editing platform we are using and whether or not we choose to “pre-prepare” our media, most of us are still doing it twice over. Which is to say that we are organising at the Finder level and then again once we get inside the NLE.”
Since I have started using FCPX, my Finder level organization has changed. Because folder and bins and fairly synonymous, I would make folders and subfolders and import those into the NLE. The NLE then reflected the Finder and vice versa. Since fcpx’s method favors searching and doesn’t really reflect a folder/subfolder construction, I find myself putting footage, music, graphics in to large buckets on the Finder and then sort them in FCPX as most recent or last imported, or ‘best’, or date, or whatever. The reason for this is because sharing libraries or portions of libraries with other editors usually involves a process (Fcpxml or transfer library) in which all the metadata stays in tact. In other NLEs, this wasn’t necessarily true. If you sent over a sequence, none of the browser tags came with the media, in FCPX, keywords and organization come with a Project, or Event/Library of course. It’s a different way of working, and I find that I work differently becuase of it.
I’d imagine that some day in a somewhat distant future, all files will be abstracted from a desktop/finder environment. Also cloud based computing will further abstract files from a finder level organization. Most file organization will adopt some sort of tag/metadata based approach and that will be the new folder/subfolder hierarchy, for better or worse.
-
Brett Sherman
May 20, 2016 at 12:59 pmSlightly off topic. I know this is really about editing application, file system metadata. But on the camera front, I find really almost all the metadata generated in camera useless. Frame rate? Not much there to sift through. This is what I would need at the camera level for metadata to actually be useful.
1. Location aware – few video cameras do this. GPS soaks up battery life and doesn’t work inside. Not sure what the solution here is. Tying to cell phone location data might be a solution.
2. Ability to enter metadata on my cell phone on the fly. There is no way in hell I’m going to attempt to enter metadata within the camera menu. Just forget it.
3. Setting a metadata expiration time/date. Worse than no metadata is the wrong metadata. And I can guarantee you once I enter specific metadata for a project. The next shoot I do, I will absolutely forget or not have time to change it.Interesting that all of these would be enhanced by a connection to your cell phone. I wonder why that hasn’t been really utilized. Imagine if one of the camera manufacturers created a really robust cell phone app. It seems to me they would have a huge competitive advantage.
-
Jeremy Garchow
May 20, 2016 at 1:17 pmThe Sony F55 (and F5) has its own wifi network (created by the camera on a separate Wifi chip on a USB) where you can control and enter a lot of metadata via any web based device. GoPros are wifi enabled, DJI products are controlled via an app where you can control and enter all kinds of things.
You do bring up a good point that metadata is highly subjective. FPS and ISO data would be very helpful for me to have in the Browser and when importing straight from MXF files that data is not captured, but obviously that wouldn’t help you out. Most cameras have fairly easy ways to upload metadata sets via SD card.
-
Simon Ubsdell
May 20, 2016 at 1:18 pmYour Finder organisation system is a good one and works because you are primarily working in FCP X.
The assets that we receive and generate are so diverse and they are needed in so many different applications that we really need to be super organised about our Finder level organisation otherwise it gets really messy and painful very quickly.
Simon Ubsdell
tokyo productions
hawaiki -
Jeremy Garchow
May 20, 2016 at 2:51 pm[Simon Ubsdell] “Your Finder organisation system is a good one and works because you are primarily working in FCP X.
The assets that we receive and generate are so diverse and they are needed in so many different applications that we really need to be super organised about our Finder level organisation otherwise it gets really messy and painful very quickly.”
I understand.
I also collaborate with other applications, and none of the organization that I have put in to FCPX transfers. None of it transfers from any other NLE either. I would love it if tags could transfer to any app, it would make things more tidy.
This also goes to show that metadata is highly subjective, and that MAM’s are so hard to develop because one person’s simple is another’s complex.
Reply to this Discussion! Login or Sign Up