|
|
| | Author: | mfav | Posted: | Jan 2, 2019 16:33 | Subject: | Admin Russell, what's with the images? | Viewed: | 252 times | Topic: | Catalog | Status: | Open | |
|
| I'm noticing my "large" images for these items are showing up as being credited
to other contributors.
Kevin1990
- 970c00pb099
- 970c34pb01
swrbricks
- 970c10pb01
- 970c09pb01
- 970c11pb04
jennnifer
- 970c69pb01
- 970c00pb075
- 970c00pb195
Nikilyn
- 970c00pb337
hazelsden
- 970c86pb02
www_mybricks_de
- 970c00 (red)
And I'm credited for these "small" images which aren't mine...and no
idea whose they are.
air023
atl017
gen052
gen053
stu002
cty057
sw016
aqu011
lea005
sw049
twn029
pln100
pln003
pln050
pln057
pln058
pm012
soc034
(I thought the whole "small image" thing was kaput some time last year...no?)
If this is human error, that's one thing. If there are systemic issues or
database corruption or something else going on, then it probably should be looked
into before it gets worse.
|
|
|
| | | | | |
| | | | Author: | StormChaser | Posted: | Jan 2, 2019 16:45 | Subject: | Re: Admin Russell, what's with the images? | Viewed: | 96 times | Topic: | Catalog | |
|
| In Catalog, mfav writes:
| I'm noticing my "large" images for these items are showing up as being credited
to other contributors.
|
Russell would indeed be the person to answer this, but in many cases it comes
down to the way the system is designed. On this, for example:
The legacy image is from hazelsden. The color image is from you. The color
image is shown as the primary image, which is why you are credited with the large
image. The legacy image also exists, which is why it is shown as an additional
image and the small image is credited to hazelsden.
It's all rather confusing, even to me, but I don't believe it is the
result of either human error or database corruption.
| And I'm credited for these "small" images which aren't mine...and no
idea whose they are.
|
Those I cannot explain.
|
|
|
| | | | | | | | | |
| | | | | | Author: | StormChaser | Posted: | Jan 2, 2019 16:49 | Subject: | Re: Admin Russell, what's with the images? | Viewed: | 81 times | Topic: | Catalog | |
|
| In Catalog, StormChaser writes:
| you are credited with the large image.
|
| and the small image is credited to hazelsden.
|
Oops. Other way around. Confusing, as I said, even to me.
|
|
| | | | | | | | | |
| | | | | | Author: | mfav | Posted: | Jan 2, 2019 17:23 | Subject: | Re: Admin Russell, what's with the images? | Viewed: | 70 times | Topic: | Catalog | |
|
| In Catalog, StormChaser writes:
| It's all rather confusing, even to me, but I don't believe it is the
result of either human error or database corruption.
|
It has to be one or the other or both. It may occur in the underlying programming
which does whatever it does when an "old" image gets replaced (or bumped, or
set to primary, or whatever you call it) with a "new" image.
If my "new" "large" image replaced otherperson's "large" image...used to
be I'd get credit for "large" image. Now, in some circumstances which I cannot
define, my "new" image ends up being credited to the previous(?) large image
uploader. Or something.
There's not supposed to be either "large" or "small" image any more. Now
there's just "image" and where there used to be "small" and "large" there
is now "multiple"....at least that's how I understood the Great Image Declaration
from Russell some time back.
If my images are showing up in somebody else's register, and somebody else's
images are showing up in someotherbodyelse's register...well...that's
database corruption.
Maybe it's just programming, but something somewhere is awry. Just thinking
it's smarter to fix the problem now before another gazillion images get posted
and end up in the wrong slot and then need to be "fixed".
|
|
|
| | | | | | | | | | | | | |
| | | | | | | | Author: | StormChaser | Posted: | Jan 2, 2019 17:28 | Subject: | Re: Admin Russell, what's with the images? | Viewed: | 58 times | Topic: | Catalog | |
|
| In Catalog, mfav writes:
| Just thinking
it's smarter to fix the problem now before another gazillion images get posted
and end up in the wrong slot and then need to be "fixed".
|
Agreed and thank you for bringing it up. It could be human error since I approved
some of the images you mention. I'm sure Russell will shed some light on
the issue.
|
|
| | | | | |
| | | | Author: | Admin_Russell | Posted: | Jan 3, 2019 00:07 | Subject: | Re: Admin Russell, what's with the images? | Viewed: | 136 times | Topic: | Catalog | |
|
|
BrickLink ID CardAdmin_Russell
|
Location: USA, California |
Member Since |
Contact |
Type |
Status |
May 9, 2017 |
|
Admin |
|
|
BrickLink Administrator |
|
| In Catalog, mfav writes:
| I'm noticing my "large" images for these items are showing up as being credited
to other contributors.
Kevin1990
|
[p=970c00pb099]
[p=970c00pb075]
[p=970c00pb195]
[p=970c00pb337]
(red)
|
And I'm credited for these "small" images which aren't mine...and no
idea whose they are.
|
[m=cty057]
[m=sw016]
[m=sw049]
|
(I thought the whole "small image" thing was kaput some time last year...no?)
If this is human error, that's one thing. If there are systemic issues or
database corruption or something else going on, then it probably should be looked
into before it gets worse.
|
Thanks for reporting this. We want to hear from people when things don't
seem right!
Regarding the first batch (minifigure legs), all of your images were indeed stored
in the "small image" slot, which is good, because the large image slot will be
going away. But I just went through these and added your images to the large
image slot for the time being so people will see them everywhere.
There is no problem with the system with this first batch. This is a simple question
of temporary assignment. I had planned to completely hide the large image credit
at this point, but as long as it's still around, we'll try to keep both
slots updated.
The second batch is more troubling. Before I took the reigns of the image project
in May of 2017, there were several attempts to solve some image issues an "easier"
way (i.e. not through the regular upload forms). This resulted in some unexpected
and often unfortunate situations, and it is a prime example of why I am very
reluctant to recommend ANY automatic processes when handling BrickLink's
data.
The images you see (all minifigures) are indeed not yours, but I believe your
username was attached to that slot because at one time you did have an image
there, probably a replacement image for the image now present.
I don't know if you keep files of your submitted images. I know many users
don't, but I still recommend it in case there is an issue like this. But
if you do have such a record, please check if you have images for any minifigures
in the list you provided.
If you do have them, please upload and we will correct the situation through
the regular forms. If not, I will try to find the images on an older version
of the site and I'll take care of it manually.
I'll check all these entries again, but I believe I have taken care of all
username issues you have presented today.
|
|
|
| | | | | | | | | |
| | | | | | Author: | mfav | Posted: | Jan 3, 2019 12:57 | Subject: | Re: Admin Russell, what's with the images? | Viewed: | 78 times | Topic: | Catalog | |
|
| Well, now I can understand why the StormChaser is confused.
If the large image slot is going away, and the large images are going to then
slot into what now is the small slot...
...and I get keeping the small slot because the small slot had ALL colors of
a piece where the large slot typically had one "representative" image per piece
and not per color...
...then I think the solution would be to leave the small images as they were
(as those were the large images slotted down into the small slot) and change
the credit to the appropriate contributor.
If I understand the plan, which I may not, and apparently is also not clear to
more than just me, the plan is to replace all the small images with large images
as they become available. You're keeping the small slot because that slot
is already prepped to contain multiple colors of a piece where the large slot
is not...and you have small images of more pieces than you have large images.
I imagine that you'd lose ~70% of the images if you kept the large set and
abandoned the small set...because you don't have all colors in the large
set.
Right? So I can see the logic of that strategy.
Still.
That leaves me with these questions:
1. Why plug my small images back into the small slot? Why not leave the large-image-put-into-the-small-slot
(as this is the goal) and change the contributor? Seems like you're going
to have to go back and do that at some point.
2. All the legs images in question are recent contributions, and while they may
have slotted into the small slot correctly, that doesn't answer why the large
images are getting mis-attributed. And will the mis-attribution continue?
3. Are there other images in the system that are similarly mis-attributed?
4. I probably do have large images of many pieces for which previously there
were only small slots. However, in the case of the small minifig images in question
here, those large images were uploaded to the system at one time and rejected
in favor of another (usually existing) image for whatever reason. If they were
rejected before, why would I bother to go through the effort of finding all these
things and upload them to have them be rejected a second time?
I hope everyone understands I'm dispassionate about having my images included.
It's not an ego thing. I am concerned that I and other contributors are doing
what amounts to quite a bit of work, and there's a bug or flaw in the system
somewhere...or some bit of information hasn't been communicated appropriately...and
this problem persists, which may result in you having to go back to the contributors
and ask for a re-upload, and so on. And we're all chasing our tails. And
each other's tails.
In any event, I appreciate your efforts here. But you'll understand my reluctance
to comb through the back catalog of images I've created (or create new) and
upload them and so on without some assurance that they're going to be managed
efficiently and appropriately. I just see no sense in adding load to an existing
problem and I'd guess you don't want to be constantly manually fixing
all of it.
Thanks for the "fix" but I'm just not sure that the "why"...the reason for
things being out of whack...have been addressed as you shared no comment on that.
|
|
|
| | | | | | | | | | | | | |
| | | | | | | | Author: | Admin_Russell | Posted: | Jan 7, 2019 02:21 | Subject: | Re: Admin Russell, what's with the images? | Viewed: | 130 times | Topic: | Catalog | |
|
|
BrickLink ID CardAdmin_Russell
|
Location: USA, California |
Member Since |
Contact |
Type |
Status |
May 9, 2017 |
|
Admin |
|
|
BrickLink Administrator |
|
| In Catalog, mfav writes:
| Well, now I can understand why the StormChaser is confused.
If the large image slot is going away, and the large images are going to then
slot into what now is the small slot...
...and I get keeping the small slot because the small slot had ALL colors of
a piece where the large slot typically had one "representative" image per piece
and not per color...
...then I think the solution would be to leave the small images as they were
(as those were the large images slotted down into the small slot) and change
the credit to the appropriate contributor.
If I understand the plan, which I may not, and apparently is also not clear to
more than just me, the plan is to replace all the small images with large images
as they become available. You're keeping the small slot because that slot
is already prepped to contain multiple colors of a piece where the large slot
is not...and you have small images of more pieces than you have large images.
I imagine that you'd lose ~70% of the images if you kept the large set and
abandoned the small set...because you don't have all colors in the large
set.
Right? So I can see the logic of that strategy.
|
Yes. Essentially, that is it.
| Still.
That leaves me with these questions:
1. Why plug my small images back into the small slot? Why not leave the large-image-put-into-the-small-slot
(as this is the goal) and change the contributor? Seems like you're going
to have to go back and do that at some point.
|
There are two different banks of "small" images. There is the original bank which
contains your small images, and this is where the credit on those minifig small
images came from. The images from this bank still show up in Forum macro tags
and other older pages. They are fixed at 80 x 60, and we have no mechanism to
change them, other than if we reupload an image as the main image, and then new
thumbnail will be written over the old one.
The other bank of "small" images can be larger than even the original "large"
image, and the original "large" image was copied into this slot in the case of
minifigs, sets, etc. (basically, anything that could not have images in multiple
colors). When this copy script was run (this was a one-time function), the credit
was NOT transferred.
In cases where the large and small image were the same, or the large and small
images both had the same contributor, there was (and is) no attribution issue.
But there is an issue with large/small pairs that had different contributors.
And fortunately that is not typical.
The whole rationale behind this image unification is because the management does
not like the disconnect between a thumbnail image that shows one thing and a
large image that shows another. It is fine to have a thumbnail that shows only
a portion of the large image. But jumping to an entirely new image is considered
non-standard and distasteful.
| 2. All the legs images in question are recent contributions, and while they may
have slotted into the small slot correctly, that doesn't answer why the large
images are getting mis-attributed. And will the mis-attribution continue?
|
There is no attribution issue with the legs images. You are just interpreting
the credits wrongly. However, because your images are in almost every case superior
to what we have, and because we prefer photos for printed parts, I uploaded your
images to both the old and the new slot and put all other images in the additional
images slot.
So right now, all catalog detail pages of those printed leg parts will show your
username under both small and large credits. But when the large image credit
slot is replaced by the additional credit slot, then the person who submitted
the primary additional image (the "default" additional image, if you will) will
have their username there instead, in cases where there is an additional image.
| 3. Are there other images in the system that are similarly mis-attributed?
|
As I mentioned in another post, there are thousands of images attributed to "Admin"
that belong to other people. We change the credit when they are claimed.
Other than that, I am sure there are plenty of cases where someone helped another
upload an image and the credit never got correctly assigned. Or sometimes a user
tweaks an image an reuploads it under their account - normally in that situation
we will credit the original user, but I'm sure there have been cases where
it didn't happen.
In short, in a database this size, I wouldn't be surprised to find a lot
of errors with attribution, just like there are a lot of errors in other areas.
| 4. I probably do have large images of many pieces for which previously there
were only small slots. However, in the case of the small minifig images in question
here, those large images were uploaded to the system at one time and rejected
in favor of another (usually existing) image for whatever reason. If they were
rejected before, why would I bother to go through the effort of finding all these
things and upload them to have them be rejected a second time?
|
It's hard for me to understand why an image of yours would get rejected,
but it does appear that in each of these cases, your small image (80 x 60) was
approved but your full size one wasn't.
These are all minifig images, so perhaps they didn't show the minifig built
exactly the right way, or didn't show front and back. The images could have
been rejected for any number of reasons.
However, a couple things have changed dramatically in the catalog since you uploaded
these images the first time. First, we have an additional image slot and we are
not afraid to use it. So if a good image comes in, we don’t have to completely
throw out someone else’s image to use yours.
The second thing that has changed is that high quality images with a white background
are greatly preferred over standard shots. Nice images were always appreciated,
but the emphasis was on correctness, not visual appeal.
So for example, the first minifig on the list currently has a large image showing
the neckpiece removed from the minifigure so as to show the printing on the torso.
In the past it was considered more desirable to be explicit about all details
of the figure, but now the emphasis is on a presentation of the minifigure that
will help sell it to today’s consumer.
So in short, we think that your previously rejected images would stand a good
chance of being accepted today. And when in comes to parts, you have many, many
80 x 60 images in the BL system that could easily be upgraded to a larger size.
| I hope everyone understands I'm dispassionate about having my images included.
It's not an ego thing. I am concerned that I and other contributors are doing
what amounts to quite a bit of work, and there's a bug or flaw in the system
somewhere...or some bit of information hasn't been communicated appropriately...and
this problem persists, which may result in you having to go back to the contributors
and ask for a re-upload, and so on. And we're all chasing our tails. And
each other's tails.
In any event, I appreciate your efforts here. But you'll understand my reluctance
to comb through the back catalog of images I've created (or create new) and
upload them and so on without some assurance that they're going to be managed
efficiently and appropriately. I just see no sense in adding load to an existing
problem and I'd guess you don't want to be constantly manually fixing
all of it.
Thanks for the "fix" but I'm just not sure that the "why"...the reason for
things being out of whack...have been addressed as you shared no comment on that.
|
I understand the concern about mismanagement of images. But the site has actually
not lost anything or corrupted it. The attribution problems with certain minifig
images are all going to get ironed out anyway as we prepare to deprecate the
large image slot. If anything, we need to update the way credits are shown, and
make sure that contributors know exactly why an image is rejected.
|
|
|
| | | | | | | | | | | | | | | | | |
| | | | | | | | | | Author: | mfav | Posted: | Jan 7, 2019 12:24 | Subject: | Re: Admin Russell, what's with the images? | Viewed: | 56 times | Topic: | Catalog | |
|
| Thank you for taking the time to respond.
| When this copy script was run (this was a one-time function), the credit was NOT transferred.
|
Oi. That makes for some busywork for somebody.
| The whole rationale behind this image unification is because the management does not like the disconnect between a thumbnail image that shows one thing and a large image that shows another.
|
Woo-hoo management! Good call.
| So right now, all catalog detail pages of those printed leg parts will show your
username under both small and large credits. But when the large image credit
slot is replaced by the additional credit slot, then the person who submitted
the primary additional image (the "default" additional image, if you will) will
have their username there instead, in cases where there is an additional image.
|
I don't know what the plan for this business is going forward, but it still
sounds problematic to me. I'd think you'd just have a single gallery
per contributor of images submitted, and credit the image in the popup window
alongside the image and eliminate all these other ambiguous and confusing and
possibly wrong attributions elsewhere.
| It's hard for me to understand why an image of yours would get rejected, but it does appear that in each of these cases, your small image (80 x 60) was approved but your full size one wasn't.
|
No big deal. Can't expect everything to be accepted.
| The second thing that has changed is that high quality images with a white background are greatly preferred over standard shots.
|
Although I'm not sure of the definition of "standard shot" is quantified
here, I believe this was always the case. The language I believe you should use,
if you want to describe this for future photo takers is: "silhouetted against
white". A piece of paper is white, and lots of objects are shot against that,
but the background in those shots I don't believe is what you're preferring.
| Nice images were always appreciated, but the emphasis was on correctness, not visual appeal.
|
As we're seeing in some current threads, "correctness" is subject to revision
and whim.
| So for example, the first minifig on the list currently has a large image showing
the neckpiece removed from the minifigure so as to show the printing on the torso.
In the past it was considered more desirable to be explicit about all details
of the figure,
|
This is not at all defined. What is more explicit: showing a minifig with a decorated
torso covered by a chest piece, or showing the torso? You can't show both
states at once. But explicit detail implies that you do need to know both states.
I'd think the emphasis should be on accurately describing the item in question
without ambiguity and within reason, according to a set of established guidelines.
In the case of minifigs, in the past, I believe that there was a preference for
a full-frontal image, and/or a full-frontal with a head inset if the face was
obscured by a visor or mask or something. In my case, if a figure has a backpack
or air tanks or similar, I try to make a 3/4 view where this element is apparent.
In the case of some divers with flippers, I positioned the figure such that the
flippers were very apparent. I'm sure some of these were not accepted because
they weren't full-frontals, even though they did more accurately completely
describe the figure.
Additional information can be conveyed with additional imagery, presumably accessed
via the image popup. Thumbnail imagery, though, should be simplified to the extent
that it can be while maintaining the most communicative immediacy possible at
a small size. In other words: the busier the thumbnails are, the more difficult
they are to recognize without effort.
| but now the emphasis is on a presentation of the minifigure that will help sell it to today’s consumer.
|
The problem...don't know that that's the right word...maybe I can rephrase.
The definition of consumer is in question. There are lots of consumers. Some
consumers just want to buy a Darth Vader. Other consumers want to be able to
identify which Darth Vader, with a specific pattern and a specific head which
may be obscured by a helmet and so on. To one consumer the detail is irrelevant,
to another it's absolutely critical.
So, that in mind.
1. I don't believe it's possible, in many many instances, to create a
single image that meets the needs of both ends of the consumer spectrum.
2. Head insets and such do detract aesthetically and potentially introduce some
level of confusion.
3. Multiple-side collage images (front and back shown, also sometimes with an
inset) don't read well at small sizes. Take a look at any of the small black
torso images where there are front and back shown together. Those are dad-gum
near impossible to distinguish from one another. As thumbnails they fail to communicate
adequately.
4. Multiple side images may be misinterpreted as a "pair" of items and not "two
sides of a single item" kind of thing.
Trying to create an acceptable image in the absence of an "image codex"...particularly
in the case of minifigs...is challenging. A definitive set of guidelines would
be appreciated. Also, having the multiple images display in the popup is a step
in the right direction for the for the more hyper-aware customer.
In general, aesthetically, I'd like to see a 3/4 angle on minifigures become
the preferred angle. Here's why:
1. As time goes on, LEGO is incorporating more detail into some figures. Side
details on legs and arms become visible in 3/4 view.
2. Around-the-neck items like backpacks and airtanks can be more clearly shown
in a 3/4 view. Example: with full-frontal shots it's impossible, or nearly
so, to distinguish between an M:Tron fig with tanks and the same fig without.
3. From a technical standpoint, there are a lot of images in the gallery which
are shot either with a flash or the light positioned directly in front and slightly
above the image. This creates a strong white highlight on the top side of the
rounded part of the minifig legs (and sometimes in the middle of the face) and
obscures detail in those parts of the image. A 3/4 position would likely reflect
the light other than directly into the lens of the camera. The instruction to
turn the figure 30 degrees counterclockwise is easier to explain to non-technical
shooters than how to properly position lights, adjust exposure times, turn off
the flash, and the myriad other things one might do to improve the image.
4. Lego, as shiny plastic, can be particularly difficult to photograph well because
the surface is highly reflective. TLG itself tends to "paint out" reflections
in their promotional materials because of this. Any things that can be done to
reduce reflection and glare in the imagery is likely to make the image more communicative.
5. The "emphasis is on a presentation of the minifigure that will help sell it
to today’s consumer" here is being considered without context, and the context
should be an improved user interface sitewide, if greater sales are a goal. A
prettier set of thumbnails alone aren't going to be enough.
| So in short, we think that your previously rejected images would stand a good
chance of being accepted today. And when in comes to parts, you have many, many
80 x 60 images in the BL system that could easily be upgraded to a larger size.
|
That comment elicits a chuckle from this end. "Easily" meaning I spend hours
combing through the archive to locate, extract, and upload them to BL. I appreciate
what I believe is the intended sentiment here, but, yeah. No. Not the top of
my priority list this week.
| I understand the concern about mismanagement of images. But the site has actually not lost anything or corrupted it.
|
This is a matter of perspective. Or semantics. From my perspective, having one
contributor's content being credited to another constitutes corruption. When
the script ran that merged the two sets of images and the attribution was not
transferred, that constitutes corruption. The implication that you can correct
the situation eventually may be intended as reassurance, but (I'm going to
use metaphor here) in the immortal words of Cuba Gooding, Jr. "Show me the money."
| If anything, we need to update the way credits are shown,
|
Yes.
| and make sure that contributors know exactly why an image is rejected.
|
That would certainly be helpful and much appreciated.
|
|
|
|
|
|