Vinyl & Data

I am going through a few hundred 7" to see if they are already on Discogs (which in itself is already a big task) and I am seeing quite a few that are not on there (so setting them apart to make scans, which is even more work). What I see way too often is so called "release hijacking" where an existing release is changed into a different release. This seriously pisses me off and I have written about it before . Some examples: someone posts pictures saying "alternate labels" (which according to the guidelines should be turned into a different release) or "better pictures" which turn out to be for a different release altogether (different texts, labels, etc.). Or, they don't read the notes that are for a release that say something like "This is the release with X, for the release with Y, see Z" and then add pictures for the release with Y. Sigh. What these people do not seem to realise is this vastly reduces the value of the database f...

Like many other websites Discogs allows you to use certain formatting codes that is then automatically expanded. For example, to link to a release you can use [r000000] (where of course "000000" should be a real release number) which is then expanded to a link to the release. There are also other formatting codes . The benefit is that if necessary Discogs can rearrange their infrastructure and then the links will (or at least should) still be working. Unfortunately they don't have it implemented correctly everywhere (yet), as on the iOS app it doesn't seem to be working (bug report has been sent): These codes are not automatically updated when releases are merged, or removed and then the links basically become useless. So I was wondering: how many releases are there with formatting codes to releases that no longer exist: how many dangling pointers are still there? I grabbed the latest data dump and first looked at all the existing releases that were not drafts...

I was going through a box of old 78s that we got quite a few years back (15 years or longer) to put on Discogs (and found quite a few which aren't on Discogs and need to be added). The records were from someone from Leeuwarden (up north in the Netherlands) according to the stickers the record shop put on it. Most of these were jazz, foxtrots, waltzes, some classical and Dutch, but there were three records that I found very interesting: they were self-produced and two of them had a date of October 8 1948. So first of all, I didn't even know people were recording things themselves back then, so that was a bit of a surprise. As said there were three records: one 7" and two 10" records. I could find out what was on the 7", even though I didn't play it: organ music from the Dutch Reformed church in Dronrijp (which is very close to Leeuwarden) and which apparently has the oldest church organ in the province it's in. The record is a so called Simplex recor...

If you search the Discogs forums you will sometimes see posts from people who dream of giving up their regular job and make their money with Discogs and wonder if that is a smart idea. I was asking myself the same question, not because I want to become a full time record dealer, but because I wondered if it actually is possible. After researching it a bit my answer is: most likely it is not possible, unless you are willing to put in a lot of time and effort, accept a cut in income, add uncertainty and perhaps move to another country to cut costs. I live in the Netherlands, which is definitely one of the more expensive countries to live in in Europe: incomes are higher than in most parts of Europe, but taxes are also higher, as are prices for many things, such as housing, restaurants, postage, flights (dynamic pricing, sigh), and so on. So that is already a bit of a challenge. According to the Dutch government the minimum wage (from July 1 2019 onwards) is € 1635.60 per month (bef...

Some people, including me, use Discogs for having an overview which releases they have. Though it is not perfect because people can, and do, change releases so it matches their own copy (even when it should have been a separate release), but it is better than nothing. There are a few things that annoy me in the way that Discogs organizes collections. One of them is the way that releases can be organized into folders. I understand the rationale behind it: you can organize releases into folders and then look at these folders separately instead of at a big list which would work for a few items, but not when having thousands or tens of thousands of items. Except: for me it is too restrictive. Sometimes I would like to have releases in more than one folder, for example I might want have a release in a folder "death metal", a folder "dutch", a folder "coloured vinyl" and a folder "football" at the same time (although I am not aware of any football ...

I am going to deviate from my usual path a bit, mostly to show the power of having a lot of (correct) data. At some point in the 1980s EMI moved its Spanish office from Barcelona to Madrid. I wanted to see if I could pinpoint when this happened. According to the EMI-Odeon label page on Discogs there were a few addresses in/around Barcelona and one in Madrid. What I noticed when looking at labels of EMI releases is that at some point a few things in the label designs and on the sleeves changed: the region for the depósito legal the name of the company the address of the company An example of a release with the old address is Duran Duran's "Hungry Like The Wolf" from 1982. An example of a release with the new address is Duran Duran's "Ordinary World" from 1993. So the change of address must have happened sometime between those two releases, but when exactly? I took the data dump released in June 2019 as a basis, extracted all the Spanish releases...

A new data dump has been released by Discogs, so time to look at some statistics again, although I am certain that it is very similar to previous months . In fact, I can already tell that it is almost exactly the same: the contributions and to and changes in Discogs are very very consistent. Release statistics I looked at the dump file with data covering May 1 - 31 2019. This dump file has  11,231,882 releases, whereas the previous one had 11,123,192 releases. That means 108,690 releases more. Also: 10,619,089 releases stayed the same 500,776 releases were changed 3,327 112,017 releases were added 112,017 3,327 releases were removed from the database 180 releases had status Draft, Deleted or Rejected 11 releases that were not Accepted were in both dumps 14 releases were moved from Draft to Accepted Changes in the data (that is: changes to already existing releases) are distributed as follows:  which is, again, almost the same graph as in previous months....