why is the first quartile (Q1) often larger than Q2?

eg: how many images do you have online?

Q1     2435.5
Q2     612.5
Q3     1500
Q4     13000

seems that Q1 should be the lowest.

Am I the only one annoyed by the rejections that include 2 completely opposite reasons - "too many or too specific", "edges are too feathered or too sharp", or "too much noise or overuse of noise reduction". Well, which one is it?

--=Tom

http://www.stockxpert.com/browse_image/view/29437791



done... thanks.

There are many good things about SS, and them being #1 in sales by a large margin for me are near the top of the list. They do however have some serious issues, #1 I think is spam. They are really bad in that regard, and don't appear to be concerned or doing something useful about it. (DT seems to be both, and IS seems to be concerned at least).
I think that SS is a bit of a lottery with similar pics, some get uploaded at a lucky time and get indexed early and get enough sales quickly to stay up in the search rankings, other very similar images don't, and never see the light of day. If you play long enough you are bound to get some winners, and at least it does appear that it is more of a random thing than who you are that makes an image take off. I am also not discounting that an image has to be good too, but there are a fairly large number of images that appear very high in the search that I think wouldn't ever get there if they were uploaded today. I like their fairly light touch on their forums, but there are some things they could do for contributors that would make things a lot easier for us that they haven't. (like some way to see how many DL and image has without wading through all the others looking for it, and a way to edit keywords more easily). I don't particularly care for the feed the beast need there, but it does help motivate me to make more images.

All sites have good and bad points, SS has more good for me these days.

I'd be very sorry to see SS get swallowed by a macro.

I think that there are a few problems with how IS search works that encourages spamming, or at least discourages what they describe as proper keywording.

Among these are the default keyword mapping. So (and this example is likely to be incorrect) if I search "border" looking for some sort of frame around a blank area but IS defaults to "border-national boundary", then by spamming a frame image as "border-national boundary" it will come up in the search, and if the buyer sees enough of the sorts of image they are looking for, they will search no further. Instead it should bring up all or none of the "border" images and ask for refinements.

IS also does a poor job of multiple word keywords, sometimes keeping them lumped together, sometimes splitting them - this can change on refining a search.

Often even knowing the keywords of an image it can be hard to find it, and the buyers don't have that luxury.

I do not pretend to be an expert at IS search, in fact I am lousy at it, but I bet plenty of people do low level default searches and as long as they get a number of images that appear to be what they are looking for, they just browse through a few pages of results not knowing that they are perhaps missing out on many more images.

Don't even get me started on all of the terms that aren't in the CV, or aren't in the CV for the meaning I mean.

It sounds from Shank_ali's example above like I should have 10-15 default keywords I put in every image (like nobody, photography, landscape (or portrait), color, etc. etc.), but then Valaaami is getting keyword rejections for just that.

I applaud IS for attempting to clear up the mess that is keywording, but they get a D for implementation.

quote:

"Maybe they didn't think it necessary to have the same key phrase in there 4 times?  Huh" 

That is possibly it, I think I might have originally had things like "ski-lift" and "chair lift" that all got matched to the same thing, but then why doesn't IS just kick it to one instance of their CV match? surely that is easier than rejecting a file and wasting reviewer time and pissing people off.

so far they have rejected all of my images keyword rejections for other reasons too, but boy do they have a weird idea of what bad keywords are.

for example this panoramic image of the uinta mountains isn't scenic or natural enough for them...



The following keywords used for this file do not appear to be fully relevant to the subject.

{[ Scenics (Nature),  Scenics (Nature)]}

I ordered some stuff on the 25th so that it would get here before I took off in December, no problems with the discount (except I had to order something else to kick it over 200$ - but that is fine, another toy to play with). The package got here the 26th! wow, nice surprise.

on a second note on Adorama, a friend ordered something and later that day the price was lowered and after she called, they switched her to the lower price. Nice. It sure beats some of those fly by night bait and switch places.

--=Tom

Without really going into exactly how and what they are testing, it seems that having more pixels hurts the test results. Perhaps a full frame 100 pixel camera could ace the tests, but what good would it be?

Still, it is nice to see someone trying to come up with some sort of testing standards.

I had a laptop hard drive die on me, and I managed to read the data off of it by freezing it and hooking it up with an external cable while it was packed on ice. I am not sure exactly why that worked, but I saw it mentioned as a possible fix online, and since it wasn't working I thought I'd give it a try. If nothing else works it is worth a try, just keep it wrapped well so moisture can't condense inside while it is cold. good luck.

From what you can gather these days there is no completely effective anti dust system on SLR's.
That's one good thing about Bridge cameras-that you dont have to worry about this.
But getting a superzoonm to produce the quality needed is another thing!

I have a BIG dust spot in my powershot S3 IS. It is a pain in the butt, but I have gotten used to trying to place it somewhere it won't show up. I am tempted to take it apart, but scared too.

With the SLR I get more dust spots, but so far I've been able to remove all the noticeable ones with either air or a brush. It seems to take 2-3 months to get enough in there that I start noticing them. (My camera claims to be self cleaning, but I don't know how much that really helps).

It is a bit scary to reach into the camera, but if you are careful, it shouldn't be that big of a deal.

--=Tom

perhaps using a polarizing FILTER is what gets you the overfiltered. It seems somewhat random to me, as do the rejections for incomplete MR. I have had them rejected for not having my signature, for not having my address, for not having the subject's birthday... (and the same MR taken before and after that). since I have plenty of images to send them my paltry 15 a week, I never bother resubmitting, but just send the next 15. It almost seems that the ones SS rejects for poor light are accepted at IS and vice versa. I have never had any pics with the bright popping colors accepted at IS, but I have heard if you can manage to get one in, they sell like mad. Since I stopped trying to send them my best sellers, but just send them everything once it gets to the top of the queue, I am surprised by how many images that aren't accepted elsewhere or are poor sellers are accepted. oh well. I suppose they know what they want, although it doesn't seem that way.

I have come to the conclusion that artifact = anything that is associated with a digital picture - digital noise, jpeg artifacts, banding, jagged histograms, tired inspectors, sensor dust... I have no idea what distorted pixels are though.

--=Tom

This COULD be a good thing, or they could mess it up horribly. knowing IS, I'm not placing any bets. If they go after people who have a few weird keywords (mostly due to their weird limited CV mapping), then it will be horrible, if they take out the serial purposeful spammers, it will be a good first step.

They also need to get rid of the weird default terms for multiple choice keywords to remove the huge advantage enjoyed by images spammed with those default terms.

It might be a hopeless battle, but if they can pull it off properly, it will drag the other sites (other than DT, which seems to have it fairly under control) to take action. Good luck to them.

--=Tom