- Joined
- Apr 12, 2011
- Location
- Sydney, Australia
Very acceptable, Colin! Nice quality scan, and the logo does nothing to detract.
G45 GBase Thinktank
- Thread starter bosshoss
- Start date
- Joined
- Apr 12, 2011
- Location
- Sydney, Australia
An update.
Here's the latest version, replacing the black/red layout with a color-coded system that takes up much less space. This avoids the need for continual vertical scrolling.
Naturally there's still a lot of work to be done before it's usable.
http://www.g45central.com/G45dataentryX.html
Here's the latest version, replacing the black/red layout with a color-coded system that takes up much less space. This avoids the need for continual vertical scrolling.
Naturally there's still a lot of work to be done before it's usable.
http://www.g45central.com/G45dataentryX.html
ohiogarage
Ikon Class
- Joined
- Apr 21, 2011
I Know i am late to the dance hear, but would Publisher Info be a good thing to add to the data entries? I sometimes use the publisher info to tie together related labels and such.
- Joined
- Apr 12, 2011
- Location
- Sydney, Australia
Good suggestion. No-one asked for this previously, so I thought it was not needed. However as you point out, it may help in certain cases, such as label discography analysis.
- Joined
- Apr 12, 2011
- Location
- Sydney, Australia
I have been thinking long and hard about the G45 Ratings system. I remember long ago on the old forum, there was a topic about creating a system to calculate an overall G45 rating using an automated algorithm. I can't locate that topic, but I know that Moptopmike favours such a system, and we have had some discussions about it.
After thinking about the existing Song/Rarity/Legend system, I realised there is something fundamental missing from a collector viewpoint - the physical object. I would call this the Artifact Rating. It would apply only to the look and feel of the physical object, from the sensibility of a '60s garage 45 collector.
Consider 3 versions of the same imaginary, legendary record:
Version 1 - first pressing, released on vanity label with cool custom graphic, and a fold-out 2-sided heavy cardboard picture sleeve.
Version 2 - second pressing, released on a different vanity label with only a white label, and a flimsy one-sided paper picture sleeve.
Version 3 - third pressing, released on Columbia promo, no picture sleeve.
All are equally rare with 5 copies known of each pressing. All are the same versions of the same 2 songs.
Currently, the S/R/L system has no way of awarding a higher rating to version 1. That is a fault with the current system, which the Artifact rating would address.
The G45 S/R/L/A system could be used to apply the following ratings:
Version 1 - 5/5/5/5
Version 2 - 5/5/5/3
Version 3 - 5/5/5/1
This additional Artifact rating would also assist in the overall compilation of the G45 list. The additional factor would help to tip the balance one way or the other, for 45s which were otherwise tied on the same score. But a high Artifact rating would not be worth as much as a high Song, Rarity or Legend rating.
Here's the "weighting" that I imagine could apply to the 4 ratings.
Song - worth 45 percent of the total rating, linear proportional weight
i.e. 5 stars=45/45, 4 stars=35/45, 3 stars=25/45, 2 stars=15/45, 1 star=5/45, 0 stars=0/45
Rarity - worth 30 percent, with logarithmic proportional weight
i.e. - 5 stars=30/30 4 stars=15/30 3 stars=7/30 2 stars=3/30 1 star=1/30, 0 stars=0/30
Legend - worth 15 percent of the total G45 rating, linear proportional weight
i.e - 5 stars=15/15, 4 stars=12/15, 3 stars=9/15, 2 stars=6/15, 1 star=3/15, 0 stars=0/15
Artifact - worth 10 percent of the total G45 rating, linear proportional weight
i.e. - 5 stars=10/10, 4 stars=8/10, 3 stars=6/10, 2 stars=4/10, 1 star=2/10, 0 stars=0/10
Having a logarithmic weighting for the Rarity rating ensures that "less than 5 known copy" records will be pushed higher towards the top of the list, provided they are also great songs.
The "Song" rating would be the TBM approved panel rating, including fractional values. The fractional value would increase resolution for the overall results. i.e. an overall G45 rating of 67.45 would beat a rating of 67.44. (The fractional value comes from the TBM rating).
The database program itself could calculate the overall G45 rating every 24 hours, and re-order the list every day according to the results.
Also, the values would be entered as a "10" star system for finer auto-calculations, but displayed to the user as a "5" star S/R/L/A system.
http://www.g45central.com/G45dataentryX.html
After thinking about the existing Song/Rarity/Legend system, I realised there is something fundamental missing from a collector viewpoint - the physical object. I would call this the Artifact Rating. It would apply only to the look and feel of the physical object, from the sensibility of a '60s garage 45 collector.
Consider 3 versions of the same imaginary, legendary record:
Version 1 - first pressing, released on vanity label with cool custom graphic, and a fold-out 2-sided heavy cardboard picture sleeve.
Version 2 - second pressing, released on a different vanity label with only a white label, and a flimsy one-sided paper picture sleeve.
Version 3 - third pressing, released on Columbia promo, no picture sleeve.
All are equally rare with 5 copies known of each pressing. All are the same versions of the same 2 songs.
Currently, the S/R/L system has no way of awarding a higher rating to version 1. That is a fault with the current system, which the Artifact rating would address.
The G45 S/R/L/A system could be used to apply the following ratings:
Version 1 - 5/5/5/5
Version 2 - 5/5/5/3
Version 3 - 5/5/5/1
This additional Artifact rating would also assist in the overall compilation of the G45 list. The additional factor would help to tip the balance one way or the other, for 45s which were otherwise tied on the same score. But a high Artifact rating would not be worth as much as a high Song, Rarity or Legend rating.
Here's the "weighting" that I imagine could apply to the 4 ratings.
Song - worth 45 percent of the total rating, linear proportional weight
i.e. 5 stars=45/45, 4 stars=35/45, 3 stars=25/45, 2 stars=15/45, 1 star=5/45, 0 stars=0/45
Rarity - worth 30 percent, with logarithmic proportional weight
i.e. - 5 stars=30/30 4 stars=15/30 3 stars=7/30 2 stars=3/30 1 star=1/30, 0 stars=0/30
Legend - worth 15 percent of the total G45 rating, linear proportional weight
i.e - 5 stars=15/15, 4 stars=12/15, 3 stars=9/15, 2 stars=6/15, 1 star=3/15, 0 stars=0/15
Artifact - worth 10 percent of the total G45 rating, linear proportional weight
i.e. - 5 stars=10/10, 4 stars=8/10, 3 stars=6/10, 2 stars=4/10, 1 star=2/10, 0 stars=0/10
Having a logarithmic weighting for the Rarity rating ensures that "less than 5 known copy" records will be pushed higher towards the top of the list, provided they are also great songs.
The "Song" rating would be the TBM approved panel rating, including fractional values. The fractional value would increase resolution for the overall results. i.e. an overall G45 rating of 67.45 would beat a rating of 67.44. (The fractional value comes from the TBM rating).
The database program itself could calculate the overall G45 rating every 24 hours, and re-order the list every day according to the results.
Also, the values would be entered as a "10" star system for finer auto-calculations, but displayed to the user as a "5" star S/R/L/A system.
http://www.g45central.com/G45dataentryX.html
- Joined
- Apr 20, 2011
- Location
- The shores of southern CT
I think the artifact rating should only be cited in specific cases to tip the scale in case of decimal ties. Otherwise it isn't necessary, IMO 
- Joined
- Apr 12, 2011
- Location
- Sydney, Australia
That could be easily done. We could award every record a standard Artifact rating of 2 stars. Then it could go up or down only if necessary, or warranted.I think the artifact rating should only be cited in specific cases to tip the scale in case of decimal ties. Otherwise it isn't necessary, IMO
Or even better, it could be reduced to a weight of 1 percent of the G45 total, but increased to a resolution of 100 "stars" (displayed to the user as 5 stars). That way, it would increase the sorting resolution of the listing x 100 , but have a minimal impact on the actual scores.
The weighting would be adjusted to :
Song - 50%
Rarity - 35%
Legend - 14%
Artifact - 1%
Frantic
G45 Legend
- Joined
- Apr 20, 2011
whatchutalkinboutwillis
- Joined
- Apr 12, 2011
- Location
- Sydney, Australia
It's just a way of ensuring that the G45 list has fewer tied scores. If you only have 3 values (Sound, Rarity and Legend) each with 5 stars, then you only have 15 possible total scores per disc. That's not enough to meaningfully rank 20,000 records in G45 order using an automated system.whatchutalkinboutwillis
But with the addition of Mike's decimal TBM Song rankings, Rarity and Legend increased to 10 stars, and the 100 star Artifact rating, the sorting resolution is increased to 150,000 possible scores. Or something.
- Joined
- Apr 12, 2011
- Location
- Sydney, Australia
Some great news.
Aaron has finished his work for the company website. He will start work on the G45 Data Engine next week. He's travelling to the UK tomorrow, and will work on the project from there. Meanwhile, I will continue to work on the Data Entry Console.
You can check out a few new features including Moptopmike's official Pressing Plant dropdown data.
http://www.g45central.com/G45dataentryX.html
Aaron has finished his work for the company website. He will start work on the G45 Data Engine next week. He's travelling to the UK tomorrow, and will work on the project from there. Meanwhile, I will continue to work on the Data Entry Console.
You can check out a few new features including Moptopmike's official Pressing Plant dropdown data.
http://www.g45central.com/G45dataentryX.html
- Joined
- Apr 12, 2011
- Location
- Sydney, Australia
It's finished, sort of.
All the required fields for data entry are present, and ready to deliver data to the php program which Aaron will write to store the data on the server.
If anyone wants to have a say in the final structure of the database, now is the time. Once Aaron starts work next week, he will be pretty quick, and changes will be harder to make when he's finished.
Check out the current format here:
http://www.g45central.com/G45dataentryX.html
There are a few minor things still unfinished. For example, the G45 ratings should disappear if you select anything other than a 45. And the fields should clear each time you select a new "track" (but the entered data should re-appear in its proper place when the track is re-selected). Same for the Personnel section, which allows up to 10 band members to be entered. Also I forgot to add the "comment" field at the bottom. I'll add that tomorrow.
All the required fields for data entry are present, and ready to deliver data to the php program which Aaron will write to store the data on the server.
If anyone wants to have a say in the final structure of the database, now is the time. Once Aaron starts work next week, he will be pretty quick, and changes will be harder to make when he's finished.
Check out the current format here:
http://www.g45central.com/G45dataentryX.html
There are a few minor things still unfinished. For example, the G45 ratings should disappear if you select anything other than a 45. And the fields should clear each time you select a new "track" (but the entered data should re-appear in its proper place when the track is re-selected). Same for the Personnel section, which allows up to 10 band members to be entered. Also I forgot to add the "comment" field at the bottom. I'll add that tomorrow.
- Joined
- Apr 20, 2011
- Location
- The shores of southern CT
I am quibbling, but I think that "repro exists" should appear after quantity known box
as it makes it seem like it refers to the repro exists quantity.
I also think you should have a comment box for repro identification only if the repro exists box is checked - then it would appear.
as it makes it seem like it refers to the repro exists quantity.
I also think you should have a comment box for repro identification only if the repro exists box is checked - then it would appear.
- Joined
- Apr 20, 2011
- Location
- The shores of southern CT
Yes, mine appears "Scrunched" in places as well.
Perhaps the page should be checked in all versions of currently used browsers. I only use Firefox, not the most up to date version because my PC hangs due to clashes with anti-virus running
Hicksville Hogwash
Tennalaga Class
- Joined
- Apr 11, 2012
- Location
- Gauley Bridge, West Virginia.
i have over lay issues too. these lines (urgggh... these lists) are the lines hiding behind whatever should be under them. Scrunched as Mike says...lol:oops:
- Heavy
- Freakbeat
- Pop
- High Energy
- Instro
- Novelty
- Soul
- M. Voc.
- F. Voc.
- Harmony
- Unison
- Kiddy
- Snott
- Wrong Song
- Off-Center Label
- Bubbly Wax
- Misprint Artist
- Misprint Title
Frantic
G45 Legend
- Joined
- Apr 20, 2011
ditto, scrunched in FIREFOX
mr. splendid
Ikon Class
- Joined
- Apr 20, 2011
- Location
- Berlin , Germany
I use Firefox and get the scrunched look as well and the dropdown menus don't work for me , but that might be my javascript blocker - although I don't get the usual alert .
- Joined
- Apr 12, 2011
- Location
- Sydney, Australia
HMMM.
I just installed Firefox, and indeed, everything is a total mess.
That sucks. I will have to figure out why.
I just installed Firefox, and indeed, everything is a total mess.
That sucks. I will have to figure out why.
Similar threads
