World Trade Center, Tower 4, in blue skies [6000x4000][OC]

danwin · 2012-05-24T22:23:10.048389+00:00

No argument from me there. But if a casting director is short on money or is doing a low-margin job, I can see why they'd opt for just a basic camera setup.

zeno · 2012-05-24T17:09:55.740754+00:00

Awful.

I usually look down the other way on a one-way street...because those are the drivers you really need to be afraid of.

ToxtethOGrady · 2 days

I viewed the image from my iPad so the whole Image fills the screen...i thought the punchline was in the strange symbol next to Penn...I gave up trying to decipher it after a minute

danwin · 2 days

Thanks...I do mention that a wifi system is a solution, albeit a non-trivial amount of money. I did a quick search for tethered photography and am going to assume that there are indeed usable free solutions http://tethertalk.com/2010/08/26/free-tethering-software-for-your-canon-dslr/

That said, tethering imposes a couple of restrictions, namely, being tethered. This can be alleviated with a long enough USB cord and movable furniture, but the former can be a hassle (especially if the shooting station is moved to compensate for daylight) and the latter is not a guarantee at a given studio.

The big thing to realize, though, is that I intended the code solution to be seen from the perspective of the director, not the photographer. Though I get paid, not all directors see it necessary to hire a photog who has a real setup and workflow. An assistant/friend who can competently shoot is all that's needed....remember that for the director, the deliverable is model candidates to the designer, not sellable photos.

Should a casting director invest in Lightroom/tethering/wifi transmitters? I don't know, I honestly don't know the scope of their day to day work. So, assuming "no", then we need a solution that assumes you have a laptop, a amateur photographer, and an assistant. Even with the tethering solution, the data entry is still more complicated than the spreadsheet (for example, injus figured out difference between Active and Selected..and the context do Grid mode...when doing mass-metadata ops in Lightroom) ...though you may argue that not having to enter a time stamp makes up for it.

My code solution requires no training beyond what any amateur photog and data entry assistant can do. The extra work, of course, is that the director knows some coding...Is it an ideal solution for managing a casting call? Againm that all depends on the number and size of calls a director does in a season...and the logistics may change significantly between them...

danwin · 2 days

well let's agree to disagree on the philosophical points...and I agree that photography isn't the best use case, I picked it only because the use cases are so much more obvious in virtually every other field.

But I still don't get your sign in process and just want to make sure im not missing something obvious...How does the data relating to the models name get associated with the photo files if I don't exchange memory cards until well after they've left?

danwin · 2 days

Well, I'm going to take the pedantic-ness of this to the next level :)

I am the author so I'm at fault for not describing the situation more accurately.

The photos are snapshots and don't have to be "good", just good enough for reference shots. With that said, it's not ten minutes per model, it's more like a minute per model. If it were ten minutes per model, then this solution wouldn't be needed because the 1 minute it takes to pop out the memory card and transfer the files is relatively short compared to the 10 minutes of shooting time.

That said, there's also no time/necessity for model releases. I don't use the photos for commercial reasons. If the process were to include model releases though, there's still the problem of associating photo to signed release (which would most definitely require more error checking, though not that much)

(since the photos just need to be usable, there's never been a case when I've had to do a retake. I might take an extra shot if I see that the eyes were closed in the previous shot. But it's amazing how that happens only once or twice out of literally a thousand takes.

The critical point that I could've made clearer: the sign-in happens at the photographer's station. You're completely right that if the time is recorded at a different point, such as when the model first arrives, then there's no linkage between photo and model.

So the devil is in the details, which kind of underscores my point. I'm able to come up with a simplified, low maintenance algorithm because I've done this before. If you gave this problem to an outside programmer, he/she might introduce logical, yet unnecessary steps. There are enough unique, bespoke batches of menial, easily automated work that don't need much code at all, just a mindset that recognizes that a problem could be solved by code.

While it seems ridiculous, even from my standpoint, to think it useful for all photographers to learn code...who knows? It definitely would've been silly 10 years ago. But now all files are digital. The problem I describe here is a microcosm, from what I understand, of the photo industry in general, which wastes considerable amount of time of repetitive, error-prone manual processes.

So it's not necessary for every -- or any -- photographer working for a photo company to know how to code a solution here. but again, think of how much time is wasted coding in features/fallbacks because of the gap in communication between programmer and the clients who actually have to execute that workflow, every day. More frequently, in virtually every kind of company, these problems are never recognized and/or fixed in the first place.

You and I would agree that this, again, does not require coding skills from everyone to fix. I differ in that from what I've seen, programming is almost the only way to ground these concepts -- this is an inference from the many, many very logical scientists, analysts, and researchers that I deal with who do not get simple binary/boolean concepts. I'm arguing that if there was a coding curriculum, even if it were just a single class just as most students have a typing/wordprocessing class, then who knows what overall benefits and knowledge might come from it.

kategardiner · 2 days

And in that forum I just discovered/re-remembered that Daniel Schorr passed away.

danwin · 2 days

Ok, point by point:

I just replied to bifftradwell; don't know if I'm just missing something obvious in his solution, or if he missed the obvious point that my solution deals with auto-tagging. If bifftradwell's solution requires typing in tags by hand (which, how do you do that if the memory card is still in the camera...?), then that is not a solution to the problem....that is the problem
You're right, there is the possibility that the assistant can't read off the system timer and makes a mistake. I'm not going to go into the obvious ways to mitigate this but this is one key point that I didn't feel the need to mention: even if this script utterly fails, then the sorting process goes on exactly as it did before. You have a list of models (in the spreadsheet) in rough chronological order and after all the digital photos have been printed out, there's some handiwork to assemble them together.
I agree with you with what Atwood was trying to say. And as I said in my post, I said I believe that Atwood overestimates the ability for people to understand these kinds of problem solving techniques. I have almost never seen anyone, in any of the logical/data professions get these concepts unless they've coded before. The fact that Atwood compares programming to plumbing -- when programming, even as just a profession has such wider applications beyond even programming itself (whereas plumbing skills apply mostly to the domain of plumbing) -- indicates that he overestimates the non-coders' thinking process.
And yes, I agree that it's important to teach people effective problem solving techniques. I said it quite clearly in my post that I don't care if someone can run gcc or not. I just am skeptical that anyone who has never done even the simplest for-loop plus conditional comparison has any real understanding of datatypes, for instance. Ask your scientist friends about it. Or ask your friends who have to program for scientists types, writing such code as "if x == true || x == 'yes' || x=='Y' " for their clients.

danwin · 2 days

OK... I actually do have Lightroom, but can't say I'm a pro with it.

How do you "tag" the photos with the models' names/agencies after the files are transferred from the camera? I mean, how do you do it without doing it manually. Because that's the whole point of this exercise, to not do it manually, which is very difficult when the memory card is still in the camera.

danwin · 2 days

Not among small operations. I can't say how often a director does a casting call that requires some logistical planning (i.e. more than just him and an assistance), but I would guess not enough to invest in a wifi-camera setup with a photo-management script.

danwin · 2 days

What was the strawman? I didn't argue that he wanted to prevent people from coding. I argued that he had a very limited view of the use of programming among non-professionals, and provided an example.

danwin · 2 days

Is that programming, per se, or using computer-aided design?

I have a good friend who works at a movie special effects studio. Even though the design/rendering/modeling is done with their proprietary multimillion software and workstations, he's able to be far more productive than his non-coding coworkers because he writes his own Python scripts that handles routine effects, the logistics of moving one scene to the next stage in rendering, and error/quality checking.

danwin · 2 days

Well, I provide a bunch of examples related to journalism (which would all be applicable to all fields of research and analysis and publishing) in the first grafs, and then in the closing grafs, a bunch of other examples (mostly in corporate office settings) where even simple Excel-VBA scripts saved people time but were incomprehensible to their managers.

So no, it's premature to drop this yet. Atwood's article relegates programming to the margins, which it does not need to be.

ketomine · 2 days

I had always assumed he was approved (until a few posts recently in which he made reference to being incognito)...Yeah, representing an organization, especially a large governmental one, is a big no-no without the official approval.

Still, it was nice to have his perspective.

danwin · 2 days

Well at first I took the picture near WTC but then I'd be considered too patriotic.

(both are next to the office and where I walk past :) )

danwin · 2 days

I don't necessarily disagree with your approach, but keep in mind that for an existing operation, you may add too much to the mix. Taking out the memory card after each shoot, even when there's time to do so (usually there isn't) adds one or two more steps to screw up -- because either you ahve to delete the photos (so that you don't accidentally include them again) or switch to another card (and make sure you picked up the right one)...these seem trivial but are likely mistakes during a repeated task....even surgeons forget to wash their hands before operations, resulting in grievous injuries to patients.

My proposed use of a spreadsheet and time-stamp recording is meant to reduce the need for the assistant to get into the file management part...so yeah, fewer moving parts, the better :)

Two-Year Club	Verified Email
ComboLinker 2010-01-25

danwin

TROPHY CASE

you'll need to login or register to do that

create a new account

login