Rejecting a whole batch

[Mr Standfast]

On 20/05/2019 at 18:07, dustydingo said:

 

I can't for the life of me imagine why any library would suddenly exclude itself from accepting images with selective focus. In fact, I'll wager London to a brick on that that will never happen.

 

As long as the main focus of the images you quote is "properly" placed and sharp, of course images like these will continue to be accepted. What automated process are you refering to?

 

May we have a look at a 100% crop of some of the number that were not accepted?

 

DD

 

 

Yes, let's have a look.

[John Morrison]

4 hours ago, Marb said:

Obviously its automated.

 

You're - obviously - mistaken. The QC humans examine some of the pix in every upload (apart from the first upload); if any pic fails, the whole upload is rejected...

[Marb]

16 hours ago, Joseph Clemson said:

 

Alamy have stated that their QC is done by real people. Why would we disbelieve them? If it was automated, why would it be switched off during the weekend, unless they were being deliberately deceitful?

 

The situation you have posited here is that a real human inspector would have recognised the duplication. This isn't necessarily so. If the duplicates are in the same batch it would require the inspector to be examining  the entire batch, looking for this kind of problem. We know they only inspect a sample so duplication might easily be missed - it's certainly happened for me.  

I would have thought that the whole batch would be looked at otherwise what is the point of QC ?

[wiskerke]

17 minutes ago, Marb said:

I would have thought that the whole batch would be looked at otherwise what is the point of QC ?

 

In case you've missed it those 5 years + 3 days you've been here:

 

A. We receive over 100,000 images a day so it’s simply not possible for us to check every image. We check a small sample of your images and if all images in that sample are ok then we’ll pass the whole submission. If we find one failure then all images awaiting QC will fail.

We take the view that every image you submit should meet our QC standards so when we look at a random sample we expect it to represent the quality of all images submitted. Our top tip is to check all images at 100% (actual pixels) before you submit.

 

(https://www.alamy.com/contributor/faqs/quality-control/?section=4)

 

wim

[Joseph Clemson]

36 minutes ago, Marb said:

I would have thought that the whole batch would be looked at otherwise what is the point of QC ?

 

I thought I had addressed this point somewhat comprehensively in the fourth post of this very thread. However, for convenience I can reiterate the point here 

 

Hello Marb,

I'm genuinely surprised you aren't familiar with Alamy's QC system, it seems you've been around donkey's years and the subject crops up in forum discussions on a regular basis. You can see their explanation of QC here.

 

Basically, Alamy use an industrial type sampling system for QC. Once the contributor has passed their initial submission test, it is assumed they have the skills and workflow to consistently submit acceptable images. The only only check Alamy  make is to examine a small number of images from each batch submitted. It the sample is OK, they assume the rest are OK too. They rely on the professionalism and skill of the contributor. 

 

If an image fails they assume a problem has arisen in the photographer's workflow. They reject all images in the queue to allow the photographer to examine whether the problem is a one off or something which affects several images, and re-submit as required. There is usually a short period of suspension from uploading to prevent contributors from simply removing the bad image and submitting all the rest without checking them further first.

 

The system is the same ass it widely used in industry. No company inspects every nut, bolt and widget they receive at goods inwards. They inspect a small sample and if they find a bad one, they reject the whole batch  and return it to the supplier for them to sort out their problem.

[John Morrison]

42 minutes ago, Marb said:

I would have thought that the whole batch would be looked at otherwise what is the point of QC ?

 

QC can even have a beneficial effect, in raising the standards for both individual photographers and the entire collection. However, it does require snappers to 'buy into' the QC procedure. What seems draconian at first (for those who keep failing) quickly becomes routine. Those who don't understand this will either carry on grumbling... or give up...

Edited May 22, 2019 by John Morrison

[Charles Stirling]

On 20/05/2019 at 15:50, Charles Stirling said:

 “Soft or lacking definition” when the main part of the image was sharp but part was out of focus as is often the case with close-ups.

 

I asked members services about possible changes in what was acceptable with reference to  part being sharp with parts being out of focus. They responded that criteria haven't changed. They went back and looked at my recent fails on these grounds and decided had been too harsh and reinstated them.

 

My guess is that its a very overworked team checking and with a quick glance they see out of focus aspects the whole image isn't checked but instead a quick reject happens. Anyway thanks to QC team for reviewing.

[dustydingo]

4 hours ago, Marb said:

I would have thought that the whole batch would be looked at otherwise what is the point of QC ?

 

All the discussions about QC's sampling methodology and "the point of QC", including in this very thread (see Joseph's posts above), and you still think every submission is looked at?

 

Someone's having a lend, surely?

 

DD

[Betty LaRue]

I had a brother-in-law that loved nothing else but to stir up trouble for his own entertainment.  He’d pretty much ruin a holiday gathering for everyone with his against-the-grain remarks, (and that’s me being kind describing what he said) and leave with a smug look upon his face after starting many arguments. We finally learned to ignore him.

I'm thinking we’re being had.

Betty

[Colin Woods]

18 hours ago, Marb said:

I would have thought that the whole batch would be looked at otherwise what is the point of QC ?

In case the OP is still not convinced, there's this. When I worked as an analytical chemist we were involved in a big study in fish lifecycles. It was studying the effect of chemical pollutants on the sex of fish, and it was a study that lasted 450 days. It was a big deal as it was the first time that this type of study had been carried out over several generations of fish. At the end of those 450 days the fish taken at different stages of the study and frozen were histologically analysed. The lab that carried out the histology was not fully used to working in a highly regulated environment and despite many entreaties to follow the required standards, they did not. When the file (when I say file I mean of course a small van full of folders) was QC'd they found a few errors in the paper trail of the histology samples. Out went the study. Rejected in its entirety. When QC find one error they are properly entitled to assume that the rest of the unchecked product/data will also contain errors. To say that the clients were annoyed is to miss their emotions by a country mile. Luckily our lab was in the clear as all my chemistry data and all the biologists data were accepted - because we were used to being QC'd. Without inspection standards inevitably deteriorate. 

[Joseph Clemson]

4 hours ago, Betty LaRue said:

I had a brother-in-law that loved nothing else but to stir up trouble for his own entertainment.  He’d pretty much ruin a holiday gathering for everyone with his against-the-grain remarks, (and that’s me being kind describing what he said) and leave with a smug look upon his face after starting many arguments. We finally learned to ignore him.

I'm thinking we’re being had.

Betty

 

I did wonder if we were simply being wound-up for a laugh. Being a charitable sort, I took the view that it was simply the difficulty some contributors have of moving a mindset from microstock's QC methodology  where every aspect of every image is scrutinised, to Alamy's QC where the emphasis is on the responsibility of the contributor for their own QC. The two are so fundamentally different that it is easy to see why an adjustment is hard to come to terms with. 

 

The sad thing about QC in microstock land is that it produces a them-and-us mentality where some contributors see submitting images as a battle to get past QC, producing multiple re-submits where they are allowed, in the hope of finding an easier inspector. Such a mindset is fundamentally alien and self-destructive in submitting to Alamy, where QC is a light-touch guardian, and where, as John Morrison pointed out earlier, the tendency is to produce photographers with better photographic, processing and self-critical skills, to the benefit of both individual and agency.

[Betty LaRue]

7 hours ago, Joseph Clemson said:

 

I did wonder if we were simply being wound-up for a laugh. Being a charitable sort, I took the view that it was simply the difficulty some contributors have of moving a mindset from microstock's QC methodology  where every aspect of every image is scrutinised, 

He’s been here over 5 years and hasn’t got with the program? Seems a stretch, Joseph. 😊

Whateveeeer, I’m done. (With beating my head against a brick wall)

[Shergar]

In your opinion Marb do Canon and Nikon quality check every camera off the production line or do they randomly check a number of them?

[KevinS]

It's threads like this one that have made me weary of this forum. It was a great help to me in the past, for which I am thankful. 

Edited May 23, 2019 by KevinS

[MizBrown]

  I recently failed a QC for too much noise in one photo.  The submission were all older photos that I'd decided to submit..  Any future photos in lower light situations will be going through one of my full frame Sony cameras, not the a6000 without a stronger flash (pop-up flash died so using that in the future simply is a non-starter, but it was what I'd used in the original photo with one of the little plastic flash modifiers for bouncing).

 

There is a rock band whose contract said that the dressing rooms would be supplied with M&Ms with  no brown M&Ms included.  The average comments on it said what a peculiar and prima donnaish request this was.  Then someone explained that the bank uses pyrotechnics and other high energy special effects in their shows.  If the brown M&Ms were in the dressing room dishes, they knew that they should do all the safety checks over with their own crew rather than trust the venue because if they couldn't get the brown M&Ms out, they might not have done all the contract-specified safety checks either.

 

Know what a given camera can and can't do.  Don't use it for things another camera can do better.