Quote from: icefloe on September 08, 2013, 14:02
A generic talent release will work for sfx!

If it's not too much, can I ask you to send me (in private) a generic TR that worked for you?
Or at least to see what do you write at "Description of sound, musical compositions or master recordings" and what/who is "Talent 1" in the case of audio recordings...
Thanks

Quote from: Beppe Grillo on September 08, 2013, 16:58
Question:
What is a talent release?

http://www.istockphoto.com/help/sell-stock/training-manuals/audio/legal-requirements-talent-releases

Although for music pieces or singing it's somehow normal to have a talent release, it's overcomplicated to complete/scan/upload tons of writing just for a bicycle ring or cricket sound... As always, IS likes to complicate the simple things.

Quote from: luissantos84 on September 07, 2013, 21:54

Quote from: Mantis on September 07, 2013, 21:48
Shutterstock is becoming pickier these days. I no longer bother with textures with them because of those rejections.

he is talking about CanStockPhoto but yep that is correct

Yeah, forgot to mention CanStockPhoto in the post - sorry about that...

Hi folks,
somebody uploads sound effects to IS?
Just wanted to upload several hundred wav's and I sadly saw, a talent release must be uploaded with every audio. This is ridiculous, why they need talent release for bicycle rings, door opening/closing, etc. sounds??? For sure it's more work to complete, scan and manage talent releases than the actual work with the audio recording...
I'm missing something?
Or they accept a generic talent release and that's it?

No, now it's not DT...

I'm thinking, when will these agencies close their uploads due to the fact that everything has been uploaded and they no longer want to accept new submissions of the existing topics...

I'm experiencing these rejection reasons on (seamless) textures and backgrounds. I have no problems selling these at other agencies.

Examples include:
http://www.shutterstock.com/pic-152405483/stock-photo-seamless-green-grass-and-clovers-texture.html?src=BBaRKvNWwfl6vJTzp89E1g-1-16
http://www.shutterstock.com/pic-152405477/stock-photo-seamless-concrete-brick-wall-texture.html?src=BBaRKvNWwfl6vJTzp89E1g-1-21
http://www.shutterstock.com/pic-152405480/stock-photo-seamless-grey-rough-wall-texture.html?src=BBaRKvNWwfl6vJTzp89E1g-1-22
http://www.shutterstock.com/pic-152405489/stock-photo-wooden-wall-texture-fir-wood-background.html?src=BBaRKvNWwfl6vJTzp89E1g-1-20
etc.

1. Levels
2. Curves
3. Shadow/highlights -> shadows
- or if none of the above works -
4. Selective levels/curves on the forté areas of the image.

Please show the original file name for audio and video files. Especially for audio is important. Having versions and then shuffled files, one must listen again the audio to find it's pair on the HDD (to complete the not imported description). You may know, audio files aren't so... visual...

Don't confuse the file size with the image size.

The file size in bytes does not matter (in this context does not carry any information about the image resolution).
The total number of pixels in the image does matter. Count the flattened image's pixels by multiplying width by height.

Saving a jpg image of 1 megapixels (10000000 pixels = an image of 1000x1000 pixels) can result in 5 kilobytes or in 100 kilobytes, depending on the image content, since the image data is compressed and the compressed size depends on the image content. The more details the bigger will be the resulting file size.

Upsampling (enlarging) the image by interpolation does not introduce any new details.

Hope that helps...

Thank you for that link!

Exactly I'm interested on the ability to reproduce something similar they have on their site. My version would be a 3x3 or 4x4 matrix, the selected squares meaning almost uniform colors (space suitable for text).

Regarding the content on the link http://tsdr.uspto.gov/documentviewer?caseId=sn78766484&docId=APP20051208085810#docIndex=16&page=1 says "The mark consists of standard characters, without claim to any particular font, style, size, or color."
If I understand correctly, the name "copyspace" is trademarked, NOT the functionality (?).

So I can build a similar search functionality, just name it otherwise than "copyspace" (??)

I'm building my own site some time ago. I thought, I will include the discovery/search feature on image copy spaces also. Recently I saw the ™ mark near the CopySpace on iStock. I'm unable to find anything related to this trademark on iStock...

However I see on many sites the term "Copy Space"...

Can somebody enlighten me, will I infringe somebody's copyright by including a similar copy space search functionality or even the term "Copy Space" cannot be used???

Something's wrong with the licenses...

http://help.mostphotos.com/customer/portal/articles/967687-licences

"Royalty free Standard" would be the license for FREE images... With MP's Royalty-Free, there's no need for any kind of Extended License!
I think, there's a problem with the expression "...infinitely larger formats and also enjoy unlimited use in your future projects."

Contact support. Attila is recently employed at 123rf and they don't know. It will cost us at least 1000-2000 rejections to get him kicked from 123rf too...

Seriously... It's sad that some folks are playing with our work. Regardless of the fact that we may submit some crap from time to time...

Free images are good when these are EVERYWHERE free. But the DT system may be tricky to some and there are at least 2 ways to let images unintentionally free (one at uploading and the other when removing the 4+ yr unsold images).
It's not good for the industry - from the point of view of the buyers - to see an image for sale at every agency and for free at DT.

I saw many images @DT being free and the similar ones being sold for credits. I guess the free ones were rejected as similar, but in this case it does not mean that the rejected image does not qualify as stock...

Most of the time inspectors have the bad luck to reject best sellers and in the same batch approve the lowest seller/quality.
As for the "specific customer base" - that simply doesn't exists. It's a theory based on assumptions.
Somebody somewhere had a very good point: inspectors aren't buyers, especially because they spend their time looking at images, NOT buying images/creating designs so how they can be sure what has commercial value??? They are only guessing...

This topic comes back from time-to-time... These sites exhibit interesting content to sustain the download links at the file sharing sites. Those who share the links may earn money from those who pay for "high speed" downloads.
Fortunately the stolen files aren't licensed so the serious usage is limited. Unfortunately, many of these files are used in so closed circles that catching unfair use is almost impossible...
Several times I saw my own images distributed via this sites and I started to report the rar's at the file sharing sites. They deleted quickly the files but guess what... Within hours, new links were available at the comments... After several weeks of "fighting" I gave up.
I guess, agencies also cannot keep a workforce just to report links 8/24.
Only one question remains open for me: how these people "purchases" the shared material???

0 downloads here weeks ago. I don't believe buyers only download business handshakes and vector labels... Is there something wrong with the search engine/contributor placement?

Quote from: goober on March 01, 2013, 00:51
I'd like no inspectors once you prove reliable. No refunds. 80% payout to artists. No public record of sales information to stop the copycats.

I like very much the idea of no inspectors when somebody proves to provide high quality... I was thinking a lot, why agencies doesn't implement this time/money saving feature... Anyway, latest reviews can damage a contributor's acceptance ratio, causing a fallback to inspections. But I know, a lot of contributors do a very good retouching and corrections, also include highly relevant keyword, without (intentional) spamming.

While the public records help the buyers (somehow), it is true, it really inspires the copycats. Good point IMHO, I'm wondering, what can be the real drawback to hide the stats???

80% payout would be nice, but I think, in this case, the contributors may need to spend some money BACK for google adwords and magazine advertising. At least in the first several years... But as the picture shows, after the good times arrive, the shareholders appear...

Quote from: Pixart on March 01, 2013, 18:50
If I have $140 at the end of the month I would like the option to only take my minimum required $100 in case I want to cash out again next month.  Does that make sense?

These payouts-at-end-of-the-month features are created for lazy managers. Also they keep the money for a long time. When the buyer purchases a subscription or credits, the money is transferred to the agency. Then, time passes, the licenses are purchased and the money slowly migrates to the contributors. I never understood why BigStock has the grace periods for payouts (=how this prevents fraud)...
I agree with the lower than $100 payout and also with the requested payout amount.

Quote from: Toon Vectors on March 01, 2013, 03:41

Quote from: luissantos84 on February 28, 2013, 16:54
categories are useless, who goes there anyway? a buyer, a designer, a contributor when looking/licensing a picture they know what they are going after so they type a few keywords, why going over categories and then over 5 million files

Categories are quite useful for search as well (in the sense of a controlled vocabulary).  Keywords on an image can be mapped into categories which in turn can be mapped back into search terms applied to that image - i.e. searches on synonyms, common misspellings, international aliases, etc. can better pull up an image that has been categorized.  Categories are more easily translated into other languages and category keywords can be updated and globally applied to categorized images in the search index.

With that being said, it is a time-consuming process to categorize images and disambiguate keywords, thus probably a bit too much to ask of a contributor to deal with directly.  However, internally managed categories can reduce the work burden on the contributor by allowing them to worry less about cramming every possible keyword variant into their metadata.

For sure I agree with the categories. But I hardly believe, there can be written a fine-tuned algorithm that save the contributors from picking a category for every single image. To deal with category guessing failures I'm thinking about a sophisticated and easy batch editing tool...

Quote from: cuppacoffee on February 27, 2013, 18:10
Here are some popular image titles from several different sites. How would these get categorized?
...

The algorithm uses the title, description and keywords to find the right category. If one level fails, or the guessing is very unsure (due to popular words), will discard the guess. For sure, there will be always room to refine the routine.
With all these, I don't expect more than 95-97% success (guess fail for 1 image out of 20-30) for a fine-tuned routine...

Thank you!
Would be a real discovery to find by myself...

Quote from: Maxal on February 27, 2013, 17:40
I like enough the indexing mode of Fotolia. Their categories are well structured and complete.

I guess, indexing means here the rearranging of the keywords in order of the importance...

To be honest I was always irritated by the extra work to rearrange the keywords and also I never found the categories complete. Too deep, always need to open and a lot of categories missing...

As I said, I want to introduce the categories, but with auto-guess feature. If the auto guessing works well, I want to remove the categories from the public and leave it internal. Another stuff that doesn't need extra attention...

Quote from: vannphoto on February 27, 2013, 09:28
as a contributor I'd like: easy uploading, not requiring categories (or make them optional). If you must have categories then make sure they are broad.

a decent stats page that has all info in one graph, so I can see all my details without having to click between pages.

should I even ask how you're going to review images? (need staff ? )

I want to include SS-style categories but with auto-guess from title, description and keywords. Thus, 90% of the times the guessed category will be correct.
Regarding the stats - of course, highly detailed stats are a must. Witch caching techniques, there will be no captcha or other sh*t when checking the stats.

Review - I must hire reviewers at some point, no question about.
Inconsistent reviews and stupid keyword rejection are killing me - and I know, many of us hate these. I want to introduce some extras from the start:
- multi-level reviews: images, metadata(keywords), additional formats may be approved separately. This way your image may pass but the keywords not. After the keywords are corrected, the image may go (back) online.
- re-upload system for resubmission and upgraded (higher resolution, levels-corrected) versions. Replacement versions replace the original ONLY if accepted.
- less rejections due to internal "quality_factor" and "commercial_factor" parameters. The "commercial_factor" is an auto-adjusting value based on the downloads of the image. These params will help placing the better images in the front of the worse but still receiving exposure when sorted otherwise than 'best match'

Quote from: Toopy on February 27, 2013, 01:08
Yeah this is what I thought, for these cases use the 70-200, which is great.

It highly depends on the photos you take, but IMHO 17-40 f4, 50mm f1.4 and 100mm f2.8 are far the best pieces in studio...
Do you use the 70-200 for outdoor shoots? (I find too long the closest focusing distance)

It's amazing to hear how the agency was started...
But I still doesn't like the referral earnings cut...