Both issues have been corrected and pushed to production. There's a cache server sitting in front of ID so you may not see the change immediately, or consistently, for a little while.
Here's a sample: http://id.loc.gov/authorities/subjects/sh88003539.skos.rdf
Is that what you are expecting?
In other news, except for our desire to remove lang tags from Name resources, which did in fact happen, none of the other "background" details I provided below were factors here. This issue actually pre-dated those changes by quite some time, much older than I anticipated. So let me ask you this: Have you been using the ID SKOS data regularly (say for months or years) and just noticed this today or were you not using the ID SKOS data regularly and noticed it the moment you started?
All the best,
From: LC Linked Data Service Discussion List <[log in to unmask]> On Behalf Of Antoine Isaac
Sent: Tuesday, November 05, 2019 10:50 AM
To: [log in to unmask]
Subject: Re: [ID.LOC.GOV] Questions about SKOS data and language tags
Thanks for the very fast answer! Much appreciated.
All this sounds good, and I'm happy to hear you're working on a fix.
May I ask you the favor of deploying both fixes at the same time? I.e. waiting until the language tags can be added to the skos:altLabels before removing the (tagged) skos-xl:altLabel values? We really need these tags at Europeana and it would be cumbersome if there was a period of time during which they're not available in a simple way.
Very best regards,
On 05/11/2019 16:26, Ford, Kevin wrote:
> Thanks, Antoine, for flagging this.
> Beginning with the second (less important) one: That's just a straight up bug. We'll get that fixed.
> As for the inconsistent use of lang tags between the simple SKOS statements and the SKOS-XL ones, we'll also address that. It's a bug too, but one that will require a little more untangling.
> Background for anyone interested: We decided to remove the lang tags from the Names resources (this was something we wanted to do for quite some time but is a non-trivial change, requiring an entire reload), but wanted to maintain the lang tags for pretty much everything else. The inconsistency you noticed in the Subjects is clearly something that was missed.
> -----Original Message-----
> From: LC Linked Data Service Discussion List <[log in to unmask]> On Behalf Of Antoine Isaac
> Sent: Monday, November 04, 2019 11:55 AM
> To: [log in to unmask]
> Subject: [ID.LOC.GOV] Questions about SKOS data and language tags
> Dear all,
> At Europeana we're trying to harvest LCSH's LOD when our providers use it. And we've spotted an oddity with respect to the use of language tags, which is a key aspect for us.
> For example at
> How come that the simple SKOS skos:altLabel statements there have values without language tags while the extended SKOS-XL pattern indicates language for the same labels?
> PS: and a less important question: how come that there is a handful of skosxl:altLabel statements with simple literal values (with language tags by the way) while this is not allowed in SKOS?
> Thanks for the help,