-
Time for FCPX to step up – collaboration
-
[Tom Sefton] “However, if you integrate frame.io well and work with exports and uploads of xml files, it can work very nicely.”
I work with Frame and I just don’t see how that’s applicable. PostLab Cloud gets you close, but it’s still not the same, because it’s a check-out/check-in process. My concern is not long-distance collaboration. It’s for multiple editors working in the same facility. That’s what the Adobe solution is all about. It basically gives Premiere editors an Avid-like process.
I agree that current methods with FCPX are functional, but they are a workaround. I’d rather see Apple do something that would work correctly. Not because I can’t work without it. Rather, it’s just another roadblock that affects whether I can get other editors within the facility to have any interest in X at all.
– Oliver
Oliver Peters – oliverpeters.com
-
with how much all of the 4 major NLE’s are borrowing from each other these days – i cant figure out why the other 3 just dont copy Avids multiuser approach?
-
[Neil Goodman] “i cant figure out why the other 3 just dont copy Avids multiuser approach?”
That’s basically what Adobe has done. Avid Bin file = Premiere Pro project file.
– Oliver
Oliver Peters – oliverpeters.com
-
[Jeremy Garchow] “I’d vote for multiuser collab too. It’s seems like it would be doable.”
If well implemented, this is one of the most useful features possible. The value is not restricted to large high end post-production teams who are co-located using a NAS. A two-person independent documentary team which is geographically distributed and intermittently connected to the internet could also benefit. There are more of the latter than the former.
However Adobe’s announced collaboration approach evidently is based around (1) Timeline locking and (2) Shared local storage – IOW a co-located team using a NAS.
With FCPX, lots of work happens in the Event Browser before touching a timeline, so some integrated collaboration of that work phase is needed. Also a collaboration approach is needed for distributed teams which are not co-located.
The problem is achieving this in a reliable, simple, truly effective implementation is really hard. It would essentially be a distributed database capable of reconciling various editor inputs. In the FCPX workflow it should allow multiple people to rate and keyword material, then resolve the conflicts. Picture something like MergeX but more elaborate, fully integrated, with conflict resolution backed by FCPX UI support.
Moving from the browser to the timeline, Resolve already has timeline merge capability. This essentially does a “diff” of two different timelines and graphically displays the similarities & differences. FCPX collaboration would ideally include something like that.
While it may see simple, there are various technical issues like distributed deadlock and livelock. Testing and supporting collaboration functionality is very difficult.
FCPX is based on the SQLite database — but that is a single-user code library, not a multi-user database engine. How to get from that to a reliable multi-user implementation is unclear. Resolve can optionally use a PostgreSQL database server, but their code must be designed for that. You can’t just take a single-user app and move it to a multi-user database.
-
[Joe Marler] “A two-person independent documentary team which is geographically distributed and intermittently connected to the internet could also benefit.”
That’s essentially what PostLab Cloud does.
[Joe Marler] “However Adobe’s announced collaboration approach evidently is based around (1) Timeline locking and (2) Shared local storage – IOW a co-located team using a NAS. “
True. They already have Team Projects, which is cloud-based collaboration. So this is in addition to that.
Honestly, Apple could take the very simple first step, which would be the ability to open a Library in read-only mode when someone else has it open (first person to open a Library gets write permission). Then from that, being able to move/copy material between Libraries. The next level would be to drill just a bit deeper and do this at the Event level. But in fairness, FCP “legacy” could never do this either, so it’s probably never been a priority for Apple and may never be.
– Oliver
Oliver Peters – oliverpeters.com
-
[Neil Goodman] “i cant figure out why the other 3 just dont copy Avids multiuser approach?”
I share your bafflement. When I worked for production companies that made series television, it was all Avid systems and the editors were able to work quite easily together on shows because of the fantastic multiuser function within Avid. This has been a feature in Avid for many years.
Why is this so challenging to implement in other NLE’s?
https://www.linkedin.com/in/tmprods
tallmanproductions.net -
[Tom Sefton] “I’ve seen tens of editors having to roll back because they can’t export anything at all”
These are localized problems.
The vast majority of Adobe Premiere issues that get posted to online forums are system incompatibility related and not software bugs. The posters almost always think that it’s Premiere bugs triggering their problem but after further analysis of their system setup, storage, etc. many of these so-called “bugs” have nothing to do with Premiere itself. That’s not to say that Premiere isn’t riddled with bugs. It is. But it’s misleading to suggest that each new rollout of Premiere is a minefield of new bugs.
I’m currently working at a company with six edit suites all running Adobe CC and for the most part the main problems that have arisen are related to permissions issues on the SAN and not Adobe Premiere bugs. And I would argue that our situation is the norm and not the exception.
https://www.linkedin.com/in/tmprods
tallmanproductions.net -
[Joe Marler] “While it may see simple, there are various technical issues like distributed deadlock and livelock. Testing and supporting collaboration functionality is very difficult.”
Are you saying Apple can’t handle difficulty?
-
[greg janza] “Why is this so challenging to implement in other NLE’s?”
Because from the very beginning, Avid designed a file and database structure that turned out to be conducive to collaboration. Whether this was out of necessity, foresight, or dumb luck is unknown. (Maybe some early Avid folks will see this thread and chime in.)
Unfortunately, this was not the case with Premiere or Final Cut. Adobe has figured out a way to get around that with this new feature. I had also thought the original FCPX format (before Libraries) would make this easier. But the bottom line seems to be that neither Apple nor Adobe started with a data format that made this easy.
– Oliver
Oliver Peters – oliverpeters.com
-
[Jeremy Garchow] “Are you saying Apple can’t handle difficulty?”
I’m just saying (depending on the implementation) this would be a different type of difficulty than Apple is accustomed to. E.g, you don’t see any Apple machines on this list, therefore all the experience obtained in this rigorous environment by shaking out bugs in threading, file system, transaction management, etc. is missed: https://www.tpc.org/tpce/results/tpce_last_ten_results5.asp
For a very simple coarse locking implementation, it’s less difficult. But locking a library doesn’t buy you that much. Locking a project doesn’t do anything for the large % of work that takes place in the Event Browser. Locking an Event is better than a library but that provides less functionality than MergeX does now.
Adding more fine-grain concurrency and conflict resolution makes it more beneficial but it becomes exponentially more difficult to implement. It would vastly expand the required test matrix and entail an ongoing test and support burden.
Maybe there’s some middle ground which gives some truly useful collaboration but is extremely reliable and not so expensive to test and support. Google Sheets seems to work fairly well in multi-user collaboration, but Google has lots of experience doing that. I’m sure Apple is considering the cost/benefit ratio of various FCPX collaboration schemes.
Reply to this Discussion! Login or Sign Up