Redisplay Messages: Compact | Brief | All | Full Show Messages: All | Without Replies| Author: | BrickCompulsion  | | Posted: | Feb 9, 2020 04:58 | | Subject: | Re: Description standardization note | | Viewed: | 45 times | | Topic: | Catalog | |
|
| I totally support this change
Consistency is key
Maybe search and change all you can find and then put it out there on here for
messages about items missed. That’s if the sql person is on holiday or unavailable
|
 |
| Author: | calsbricks | | Posted: | Feb 9, 2020 04:49 | | Subject: | Re: Description Standardization Note | | Viewed: | 44 times | | Topic: | Catalog | |
|
| In Catalog, StormChaser writes:
| | In Catalog, mfav writes:
| | It would be good for consistency in the catalog description if all instances
of numbers were either integers or words, not arbitrarily some of each.
|
I have noticed this before and definitely agree it needs to be worked on. The
problem is (and I'm starting to understand the rearranging the chairs thing)
there are too many problems.
In the case of this problem, you can't even find the problem because you
can't search for it. So it would involve examining every single catalog
entry. All 100K+ of them.
And perhaps, when the next problem comes along, examining all of them again.
It almost makes a person think that there must be a better way . .
|
The answer is really quite simple - talk to a SQL person and they will advise
you how to organise your query to find precisely what you want. It isn't
rocket science. .
|
|
| Author: | FreeStorm | | Posted: | Feb 9, 2020 03:59 | | Subject: | Re: Description Standardization Note | | Viewed: | 46 times | | Topic: | Catalog | |
|
| In Catalog, StormChaser writes:
| | In the case of this problem, you can't even find the problem because you
can't search for it. So it would involve examining every single catalog
entry. All 100K+ of them.
|
What do you want to find?
Something like this?
"* three *" = 153 results
"* 3 *" = 6675 results
"* three *" AND "* 3 *" = 16 results
As example, last search find:
"Technic, Panel Car Spoiler 3 x 8 with Three Holes and Black and Yellow Danger
Stripes and Mud Pattern (Stickers) - Set 7968"
"Primo Brick 1 x 1 with Three Flowers and Number 3 on Opposite Side Pattern"
Of course,I made query to find only for "three" and "3" on same description,
but with a regexp it should be simple to spot every "word digit" with "digit"
on same description like "blah with two legs and 1 arm"
Some other incoherence (digit 3 should be between '' when you have pattern
after):
"*number 3 *pattern*" = 22 result
"*number '3' *pattern*" = 3 result
If you know what you want to find, it should be easy to make a search query,
to spot the majority of incoherence.
-Fred
|
|
| Author: | StormChaser | | Posted: | Feb 9, 2020 03:29 | | Subject: | Re: Description Standardization Note | | Viewed: | 60 times | | Topic: | Catalog | |
|
| In Catalog, mfav writes:
| | It would be good for consistency in the catalog description if all instances
of numbers were either integers or words, not arbitrarily some of each.
|
I have noticed this before and definitely agree it needs to be worked on. The
problem is (and I'm starting to understand the rearranging the chairs thing)
there are too many problems.
In the case of this problem, you can't even find the problem because you
can't search for it. So it would involve examining every single catalog
entry. All 100K+ of them.
And perhaps, when the next problem comes along, examining all of them again.
It almost makes a person think that there must be a better way . . .
|
|
| Author: | hpoort | | Posted: | Feb 9, 2020 03:00 | | Subject: | Re: Description standardization note | | Viewed: | 48 times | | Topic: | Catalog | |
|
| In Catalog, Admin_Russell writes:
| | In Catalog, mfav writes:
| | It would be good for consistency in the catalog description if all instances
of numbers were either integers or words, not arbitrarily some of each.
So:
Three eyeballs, Three earlobes
or
3 eyeballs, 3 earlobes
not
Three eyeballs, 3 earlobes
...that kind of thing.
There are pros and cons for each option. Please investigate how the search mechanism
works to see which option yields better results.
|
Unfortunately, single digits are thrown out of many of the search results. But
I think in general one would find more numerals in item names than the words
spelled out simply because it makes the name shorter.
|
Please drop the keyword exclusion altogether.
It does not make sense to make use of a full text English language keyword exclusion
list, when the contents of the database is actually not normal full text English,
but a specific dialect: Bricklink English (EN-BL) in which it makes more sense
to search for keywords like 'with' and 'without' and for single
digits. It should be easy to adjust the database configuration. No programming
involved.
Hans-Peter
|
|
| Author: | mfav | | Posted: | Feb 8, 2020 22:49 | | Subject: | Re: Description standardization note | | Viewed: | 63 times | | Topic: | Catalog | |
|
| In Catalog, Admin_Russell writes:
| | Unfortunately, single digits are thrown out of many of the search results.
|
Aren't you the fella to talk to to get that changed, Mister Site Manager?
| | But I think in general one would find more numerals in item names than the words
spelled out simply because it makes the name shorter.
|
While that is undoubtedly true, it's not particularly helpful in any way.
If you search, say, legs assemblies...
search for "three" you get two results
search for "3" you get: The following keywords were excluded from your search:
3
...however there are 13 entries containing "3"
Isn't the point of StormBluser's standardization plan to fix this stuff?
To paraphrase the StormBluser: why name something like this when 75% of the search
words are invalid?
Again, I'll make the point you need to distinguish between identification
and discovery. The identification of a thing is irrelevant if it can't be
discovered. People use the search box to find things, they don't browse 180,000
individual items looking for "3". Hope you get my drift.
Aren't you the number one proponent of making the catalog useful? Let's
figure out a way to make it useful.
|
|
| Author: | Leftoverbricks | | Posted: | Feb 8, 2020 22:24 | | Subject: | Re: Sticker # When adding Items | | Viewed: | 35 times | | Topic: | Catalog | |
|
| In Catalog, OmnipresentECA writes:
| | I have come across a few tiles with stickers that are not in the catalog.
, nor in their set inventory.
How do I go about determining the Sticker # when Adding to the catalog.
For example one is a 87079pb###, how do I determine the last few digits?
|
Search for 87079pb*
You will get 725 results. The next ‘free’ number should be 726.
|
|
|
| Author: | mfav | | Posted: | Feb 8, 2020 21:08 | | Subject: | Description standardization note | | Viewed: | 168 times | | Topic: | Catalog | | Status: | Open | |
|
| It would be good for consistency in the catalog description if all instances
of numbers were either integers or words, not arbitrarily some of each.
So:
Three eyeballs, Three earlobes
or
3 eyeballs, 3 earlobes
not
Three eyeballs, 3 earlobes
...that kind of thing.
There are pros and cons for each option. Please investigate how the search mechanism
works to see which option yields better results.
|
|
| Author: | OmnipresentECA | | Posted: | Feb 8, 2020 15:30 | | Subject: | Sticker # When adding Items | | Viewed: | 52 times | | Topic: | Catalog | | Status: | Open | |
|
| I have come across a few tiles with stickers that are not in the catalog.
, nor in their set inventory.
How do I go about determining the Sticker # when Adding to the catalog.
For example one is a 87079pb###, how do I determine the last few digits?
|
|
| Author: | dcarmine | | Posted: | Feb 8, 2020 14:53 | | Subject: | Re: February Roadmap Project: Old and New | | Viewed: | 53 times | | Topic: | Catalog | |
|
| In Catalog, axaday writes:
| | In Catalog, dcarmine writes:
| | According to the new standard for "difference types" the difference has to be
related to the way a part functions, not just cosmetic differences. So with
that rule, there would be only one entry for this hat.
|
I think there you are talking about run change things. Minor adjustments to
a mold to use less plastic or something that don't change its function at
all don't make "a difference". But a change of design on a hat makes it
a new hat.
|
Ok good to know.
Donna
|
|
| Author: | randyf | | Posted: | Feb 8, 2020 14:39 | | Subject: | Re: Questions about Unikitty / Counterparts | | Viewed: | 32 times | | Topic: | Catalog | |
|
| In Catalog, mjwest83 writes:
| | In Catalog, Admin_Russell writes:
| | But for all other sets, we've decided one entry is enough. To allow more
entries like this would create a precedent with huge implications. So your pending
entries will be rejected in a few days.
|
In regards to the two Unikitties in set , if you only want to list
one of them, fine. But which one? The decision is totally arbitrary. The box
itself is effectively two separate presentations. It doesn't have a front
and back, so much as two fronts. If you look at the instructions, they do NOT
have numbers. Instead it has two equal, independent instruction books. Really,
this set is two completely independent sets that happen to use the same parts.
As such, choosing just one of those Unikitties is a completely arbitrary decision,
and it can be argued that either one should be the primary.
|
The box, instructions, and LEGO set description are definitive about the situation,
although it may not look that way at first.
(1) LEGO lists the first instruction manual as the one used to build the house.
(2) The set is named "Emmet's Dream House/Rescue Rocket!", so it is clear
which model is considered the main model and which is considered the alternate
model.
(3) The set description from LEGO.com says "This 2-in-1 construction toy features
Emmet’s home which can be rebuilt into a rocket."
All of the key information is there to determine that the main model is the house
and the alternate model is the rescue rocket. Based on that determination, the
instruction manual for the house takes precedence.
| | This is NOT a case where there is an obvious primary build of the minifigure,
with a side picture showing an alternative build. (Like a character that has
both a helmet and hair, where one is the primary build, and the other is an alternative
for play purposes.) No, this is outright two separate minifigures that just happen
to use common parts. This is a very different case than just about any other
situation.
|
It cannot be two separate minifigures, because there is only one tail for Unikitty
in the set, as well as only one set of ears, one horn, and one headband plate.
Therefore, we have to make a decision about what minifigure gets added to the
set for now. That minifigure was determined based on the above data that I listed.
In the future, when the "Minifigs:" section of an inventory will hopefully just
be an "Assemblies:" section which includes other assembled counterparts, two
different assemblies of parts can be added. However, we are nowhere close to
that.
The admins can discuss this situation some more, but I don't believe we will
come to a different decision at this time.
Cheers,
Randy
|
|
| Author: | mjwest83 | | Posted: | Feb 8, 2020 11:31 | | Subject: | Re: Questions about Unikitty / Counterparts | | Viewed: | 36 times | | Topic: | Catalog | |
|
| In Catalog, Admin_Russell writes:
| | But for all other sets, we've decided one entry is enough. To allow more
entries like this would create a precedent with huge implications. So your pending
entries will be rejected in a few days.
|
In regards to the two Unikitties in set , if you only want to list
one of them, fine. But which one? The decision is totally arbitrary. The box
itself is effectively two separate presentations. It doesn't have a front
and back, so much as two fronts. If you look at the instructions, they do NOT
have numbers. Instead it has two equal, independent instruction books. Really,
this set is two completely independent sets that happen to use the same parts.
As such, choosing just one of those Unikitties is a completely arbitrary decision,
and it can be argued that either one should be the primary.
This is NOT a case where there is an obvious primary build of the minifigure,
with a side picture showing an alternative build. (Like a character that has
both a helmet and hair, where one is the primary build, and the other is an alternative
for play purposes.) No, this is outright two separate minifigures that just happen
to use common parts. This is a very different case than just about any other
situation.
Also, if they can't be listed as counter-parts, how are they supposed to
be sold. It is flat out the case that this minifigure will be sold as a minifigure,
whether it has a specific listing or not. There are only two ways to do so: one
is as a custom figure (which is pointless) and the other is as an incomplete
set (which is what will be done). When listed as an incomplete set, this pollutes
the averages on which so much stuff is based. When it is such a small part of
the set, that can have a very over-sized impact on those averages. By allowing
the counterpart figure, you avoid those issues.
|
|
| Author: | StormChaser | | Posted: | Feb 8, 2020 02:39 | | Subject: | Re: February Roadmap Project: Old and New | | Viewed: | 58 times | | Topic: | Catalog | |
|
| In Catalog, randyf writes:
| | I suggest adding names and titles that have "Type 1", "Type 2", etc. in them
to the project. Those terms are useless.
|
I completely agree.
But I also know from experience that bundling too many things in together is
not conducive to finishing things. So I may make this a separate project. One
way or the other, it is needed.
|
|
| Author: | axaday | | Posted: | Feb 7, 2020 21:26 | | Subject: | Re: February Roadmap Project: Old and New | | Viewed: | 41 times | | Topic: | Catalog | |
|
| In Catalog, dcarmine writes:
| | According to the new standard for "difference types" the difference has to be
related to the way a part functions, not just cosmetic differences. So with
that rule, there would be only one entry for this hat.
|
I think there you are talking about run change things. Minor adjustments to
a mold to use less plastic or something that don't change its function at
all don't make "a difference". But a change of design on a hat makes it
a new hat.
|
|
| Author: | mfav | | Posted: | Feb 7, 2020 21:10 | | Subject: | Re: February Roadmap Project: Old and New | | Viewed: | 43 times | | Topic: | Catalog | |
|
| In Catalog, dcarmine writes:
| | I have about 98% of the Harry Potter sets if you want to use that for year of
change or verification on that hat. I'm willing to look at them and report
what I find.
|
I'm content to let the mad geniuses of catalog control sort out the catalog.
I'm just documenting the differences I can find in the handful of hats I
have.
Regardless of whatever else, the presence of the prong on some hats and the absence
on others I would think constitutes a different mold, and the supercollectors
may want to note the differences. But, again, I'm content to let the catmins
sort all that out.
If you want to get really crazy, the undersides of the brims also exhibit differences
under the magnifying glass. Some are smooth, some have a pebbled inner and outer
ring, some have an inner ring with parallel striations instead of pebbling, a
couple have a narrow band of striations in an otherwise smooth inner ring. The
subtleties are kind of endless.
For whatever it's worth, though, mine are:
Black 90460: fine pebbling overall
Black: smooth cone, pebbled brim, with prong,
Black: slightly pebbled cone, pebbled brim, with prong,
Blue, Purple, Green: slightly pebbled cone, definitely pebbled brim, with prong
Dark Green, Dark Green with dragon decoration: definitely pebbled cone and brim,
no prong
Tan: slightly pebbled cone, definitely pebbled brim, no prong
Light Purple: cone too beat up to tell if it's smooth or slightly pebbled,
definitely pebbled brim, with prong
The dark green ones and the 90460 are pretty much pristine; the others are old
and worn. Your mileage may vary.
My sample size is small, so additional data is probably warranted if the admins
want to try to make sense of it.
I just updated the hats page with closeups. Don't know if you saw that version
or the earlier no-closeups version.
My opinion otherwise though is that using renders instead of photographs for
the parts is misleading for the collector.
Did I mention: The page cannot be displayed because an internal server error
has occurred.
|
|
| Author: | dcarmine | | Posted: | Feb 7, 2020 19:55 | | Subject: | Re: February Roadmap Project: Old and New | | Viewed: | 50 times | | Topic: | Catalog | |
|
| In Catalog, mfav writes:
According to the new standard for "difference types" the difference has to be
related to the way a part functions, not just cosmetic differences. So with
that rule, there would be only one entry for this hat.
I don't know if that rule applies to minifig categories, but I guess it depends
on how far down the "collector's details" rabbit hole you want to go.  .
I have about 98% of the Harry Potter sets if you want to use that for year of
change or verification on that hat. I'm willing to look at them and report
what I find.
HTH
Donna
|
|
| Author: | axaday | | Posted: | Feb 7, 2020 19:10 | | Subject: | Re: February Roadmap Project: Old and New | | Viewed: | 44 times | | Topic: | Catalog | |
|
| I reformatted a BUNCH of road signs with type 1 and type 2 bases. They should
perhaps be (xxxx / xxxx) like tire combinations?
In Catalog, randyf writes:
| | I suggest adding names and titles that have "Type 1", "Type 2", etc. in them
to the project. Those terms are useless.
Cheers,
Randy
In Catalog, StormChaser writes:
| | We are finally making serious headway with the catalog backlog. The catalog
change requests queue is cleared out.
In the spirit of trying to ensure that the Catalog Roadmap is not just something
to look at, I am continuing with the goal of moving through one project each
month. We have essentially had a January project (flesh to nougat) and there
is a new project starting today for February:
https://www.bricklink.com/help.asp?helpID=2476
What is this project?
This project involves accurately describing the differences between early (old)
versions of an item and later (new) versions. We need this for two reasons:
1. Using the words "old" and "new" were easy ways to avoid having to describe
actual differences between items. Set numbers in minifigure titles performed
the same function, which is why they were removed.
2. When a third version of an item appears, the old/new descriptors are useless.
Similar items should be described by their differences (usually the way in which
the later version is different from the earlier version).
There are around 500-1000 items expected to be affected by this project.
How can members contribute?
For the immediate moment, members may contribute by submitting title change requests.
I'm not sure how we're going to do item numbers yet, so hold off on
those. If you're not confident that you know exactly what this project entails
or need clarification on a specific issue, then please ask before submitting
a request.
For the immediate moment, ignore the additional topic bundled with this project
(looking at the use of with/without in titles). We may be able to include that,
or it may need to be made its own project. This project should also look at
the use of early/later descriptors since they serve the same function as the
terms old/new.
The goal is to wrap this up by the end of February. Of course, we also have
catalog work still to do and that takes precedence over projects - so this may
run over schedule.
|
|
|
|
| Author: | randyf | | Posted: | Feb 7, 2020 18:12 | | Subject: | Re: February Roadmap Project: Old and New | | Viewed: | 54 times | | Topic: | Catalog | |
|
| I suggest adding names and titles that have "Type 1", "Type 2", etc. in them
to the project. Those terms are useless.
Cheers,
Randy
In Catalog, StormChaser writes:
| | We are finally making serious headway with the catalog backlog. The catalog
change requests queue is cleared out.
In the spirit of trying to ensure that the Catalog Roadmap is not just something
to look at, I am continuing with the goal of moving through one project each
month. We have essentially had a January project (flesh to nougat) and there
is a new project starting today for February:
https://www.bricklink.com/help.asp?helpID=2476
What is this project?
This project involves accurately describing the differences between early (old)
versions of an item and later (new) versions. We need this for two reasons:
1. Using the words "old" and "new" were easy ways to avoid having to describe
actual differences between items. Set numbers in minifigure titles performed
the same function, which is why they were removed.
2. When a third version of an item appears, the old/new descriptors are useless.
Similar items should be described by their differences (usually the way in which
the later version is different from the earlier version).
There are around 500-1000 items expected to be affected by this project.
How can members contribute?
For the immediate moment, members may contribute by submitting title change requests.
I'm not sure how we're going to do item numbers yet, so hold off on
those. If you're not confident that you know exactly what this project entails
or need clarification on a specific issue, then please ask before submitting
a request.
For the immediate moment, ignore the additional topic bundled with this project
(looking at the use of with/without in titles). We may be able to include that,
or it may need to be made its own project. This project should also look at
the use of early/later descriptors since they serve the same function as the
terms old/new.
The goal is to wrap this up by the end of February. Of course, we also have
catalog work still to do and that takes precedence over projects - so this may
run over schedule.
|
|
|
| Author: | axaday | | Posted: | Feb 7, 2020 17:27 | | Subject: | Re: February Roadmap Project: Old and New | | Viewed: | 51 times | | Topic: | Catalog | |
|
|
So like we would change this to "NBA Jason Kidd, Dark Blue Jersey Nets #5", right?
|
| Author: | mfav | | Posted: | Feb 7, 2020 15:45 | | Subject: | Re: February Roadmap Project: Old and New | | Viewed: | 67 times | | Topic: | Catalog | |
|
| In Catalog, jennnifer writes:
| | Sure! I'll play along...
I am not convinced that 90460 is really a new version of 6131.
|
I don't know which number to ascribe to which hat, but there are hat differences.
http://v4ei.com/brickref/wizard-hats.php
|
|
| Author: | popsicle | | Posted: | Feb 7, 2020 15:42 | | Subject: | Re: February Roadmap Project: Old and New | | Viewed: | 106 times | | Topic: | Catalog | |
|
| In Catalog, jennnifer writes:
| | Sure! I'll play along...
I am not convinced that 90460 is really a new version of 6131.
They look the same and LEGO uses the same thumbnail image for both. The small
image in Dark Green that I submitted for 90460 was back in the days before I
knew what I was doing. I cannot confirm the part number or what set it was from.
(I guess we all start somewhere!)
Anyway, I would like to see convincing side by side proof that they are significantly
different even if they have different mold numbers. I have not found any molded
with 90460. My Witch from Series 2 has been played with so I cannot confirm it
there either.
I am open to hearing any ideas or missing information about this!
|
I've got a question, Jen...
When are you going to arrive at the same conclusion that StormChaser already
has, that the BL catalog needs those like you (passion for, and knowledge of
the catalog process) and rejoin the team?
It is abundantly clear, that like Robert, you care and are impassioned about
BrickLink's catalog construction.
Consider with those already in place, and you (maybe a few others I can think
of) it would be the makings of a BL Catalog Dream Team.
|
|
| Author: | jennnifer | | Posted: | Feb 7, 2020 15:10 | | Subject: | Re: February Roadmap Project: Old and New | | Viewed: | 82 times | | Topic: | Catalog | |
|
| Sure! I'll play along...
I am not convinced that 90460 is really a new version of 6131.
They look the same and LEGO uses the same thumbnail image for both. The small
image in Dark Green that I submitted for 90460 was back in the days before I
knew what I was doing. I cannot confirm the part number or what set it was from.
(I guess we all start somewhere!)
Anyway, I would like to see convincing side by side proof that they are significantly
different even if they have different mold numbers. I have not found any molded
with 90460. My Witch from Series 2 has been played with so I cannot confirm it
there either.
I am open to hearing any ideas or missing information about this!
Thanks,
Jen
|
|
| Author: | StormChaser | | Posted: | Feb 7, 2020 14:45 | | Subject: | February Roadmap Project: Old and New | | Viewed: | 243 times | | Topic: | Catalog | | Status: | Open | |
|
| We are finally making serious headway with the catalog backlog. The catalog
change requests queue is cleared out.
In the spirit of trying to ensure that the Catalog Roadmap is not just something
to look at, I am continuing with the goal of moving through one project each
month. We have essentially had a January project (flesh to nougat) and there
is a new project starting today for February:
https://www.bricklink.com/help.asp?helpID=2476
What is this project?
This project involves accurately describing the differences between early (old)
versions of an item and later (new) versions. We need this for two reasons:
1. Using the words "old" and "new" were easy ways to avoid having to describe
actual differences between items. Set numbers in minifigure titles performed
the same function, which is why they were removed.
2. When a third version of an item appears, the old/new descriptors are useless.
Similar items should be described by their differences (usually the way in which
the later version is different from the earlier version).
There are around 500-1000 items expected to be affected by this project.
How can members contribute?
For the immediate moment, members may contribute by submitting title change requests.
I'm not sure how we're going to do item numbers yet, so hold off on
those. If you're not confident that you know exactly what this project entails
or need clarification on a specific issue, then please ask before submitting
a request.
For the immediate moment, ignore the additional topic bundled with this project
(looking at the use of with/without in titles). We may be able to include that,
or it may need to be made its own project. This project should also look at
the use of early/later descriptors since they serve the same function as the
terms old/new.
The goal is to wrap this up by the end of February. Of course, we also have
catalog work still to do and that takes precedence over projects - so this may
run over schedule.
|
|
Next Page: 5 More | 10 More | 25 More | 50 More | 100 More
|
United Kingdom, England
Switzerland, Vaud
USA, Oklahoma
Netherlands, Groningen
