Seeking comments on the Data Catalog (DCAT) US standard v3.0 | Hacker News Reader

Seeking comments on the Data Catalog (DCAT) US standard v3.0

github.com

88 pointsmetasemantic19 comments

I sort of applaud the goals of projects like this, but I feel sorry for anybody who has to be compliant with them.

I'm involved with a data collection and archiving project right now. Occasionally I'll get distracted with ideas like "Maybe I should make it compliant with Dublin Core/OWL/DCAT/WhateverTripleFlavorOfTheMonth" and then I go down a rabbit hole of meta(meta(meta)) documents like this for two hours and come back feeling like I know less than when I started.

So I just make up some JSON thing that seems like it will probably make intuitive sense in 100 years, write programs to generate and read that, and get on with my day.

I'd love for there to be some kind of "get to the point" document (complete with examples!) that didn't require reading 10,000 pages of bureaucrat-ese before having any hope of understanding it.

I agree generally, but dcat has been really easy to implement. There’s really only 10-20 required fields and they are all pretty useful and you already know them anyway.

Someone did a good job, I think, in setting this up to be functional and address the main issues that government catalogs should accomplish- what data that the people paid for are available?

It’s also understandable through the json scheme if you’d like to skip the pretty easy to understand web site, https://resources.data.gov/standards/catalog/dcat-us/

My biggest fear with us-dcat3 was that it would be incompatible with 1.1. And this page says it won’t be, so I’m staying optimistic until I review.

For more information about DCAT-US, click through to

https://github.com/DOI-DO/dcat-us/wiki/What-is-DCAT%E2%80%90...

> DCAT-US is the metadata standard associated with the requirements for enterprise data inventories in the OMB M-13-13 open data policy and the Foundations for Evidence-Based Policymaking Act Title II, OPEN Government Data Act (Evidence Act). The Evidence Act applies to all agencies. These federal policies do not apply to state and local governments which may have their own policies. However, state and local governments are welcome voluntarily to contribute their metadata to Data.gov. To do so, they must publish their metadata using the DCAT-US standard while omitting any federal-specific metadata elements as noted in the documentation.

I haven't kept up with National Information Exchange Model (NIEM)[0] since the first Obama term. Is this related somehow?

[0] https://en.wikipedia.org/wiki/NIEMOpen

NIEM is much more complex is was trying to have semantic exchange vs this is just cataloging and doesn’t do much with being able to interpret and use data.

NIEM was really complicated to implement, and I think that contributed to limited use.

DCAT is for cataloging and describing datasets and I think is used pretty extensively (I think google was using it when they would scrape and then describe data in their data commons project, among others).

If I searched the Internet high and low for a whole day, I doubt I could find a better example of this phenomenon for which I don’t have a name: standards that exist to keep standard authorities busy. Standards that are actually seventeen layers of standards, all developed in isolation from the real ecosystem. Standards that will have between zero and at most three implementations, all mutually incompatible.

Standards designed to be so expensive to implement that only a handful of companies will ever do so.

Architecture astronautism applied to standards.

Stratospheric standards?

How is open data these days? I have the feeling it lost a lot of steam?

Open data is dead, long live open data! https://radiant.earth/blog/2023/05/we-dont-talk-about-open-d...

Open data is undeniably a good and important thing, but a lot of people have stumbled thinking that merely making data open would make data useful. It's time to focus on creating useful data products, some of which will be made available under an open license, some of which will not.

- [ ] DOC: The JSON-LD context file links 404: https://doi-do.github.io/dcat-us/#json-ld-context

Also, it says the schema for physical units are specified by this spec?

FWIU, QUDT Quantities, Units, Dimensions, and Types URIs MAY be used with

https://Schema.org/QuantitativeValue; and neither CSVW nor Model for Tabular Data and Metadata on the Web specify how to indicate physical quantities and units with a controlled vocabulary with URIs?

Does anyone know why is there a “-US” suffix in the name? Is it an extension to the standard DCAT v3?

> DCAT-US v3 is not a “new” standard; it is a “profile” of or implementation of the World Wide Web Consortium’s (W3C) DCAT standard.

It was hidden at the top of the list of bullet points in the readme.

Yeah it is. It basically narrows the spec down a bit. For example, it has some specific classes like the dcat-us AccessRestriction class, but it also mandates the publisher of a catalog (collection of datasets), while vanilla DCAT has it as a recommended field.

In Europe, the EU promotes their own DCAT-AP profile a bit more. Same purpose, quite widely used actually by governments, but not completely compatible with DCAT-US even though both are extensions of DCAT. Fun fact, DCAT-AP v1 predates the standardization of regular DCAT v1, which led to some minor inconsistencies. In the subsequent versions that development process is a bit more aligned now.

recommended title update: Replace DCAT with "Data Catalog (DCAT)"

Ok, done. Thanks!

Also, can we update it to be "Data Catalog (DCAT) US profile v3.0" or similar?

As someone who follows open data closely, I thought this was an update to the W3C DCAT Standard, not the US profile & implementation.

Ok, I've put the US back in there now.

Reply on news.ycombinator.com