Re: Question about conneg on datasets compressed with gzip

On Wednesday, October 25, 2017, Svensson, Lars <L.Svensson@dnb.de> wrote:

> Hi all,
>
> I've been looking at BP 19 (Use content negotiation for serving data
> available in multiple formats) [1] and was wondering if there is an
> acknowledged way of dealing with datasets that are gzipped due to their
> size. As an example use case let's look at the hypothetical dataset GND.
> There are two distributions available in the RDF serialisations turtle and
> RDF/XML. Due to the data dump's size, the files GND.rdf and GND.ttl are
> gzipped before put on the download page so that we have the two files
> GND.rdf.gz and GND.ttl.gz. Technically, the Content-Type of both files is
> application/gzip, but what we really want to negotiate on is the media type
> of the unpacked content (viz. application/rdf+xml and text/turtle). Is
> there a standard way of doing this?
>
> [1] https://www.w3.org/TR/dwbp/#Conneg
>
> Thanks for any insights.


>From https://twitter.com/westurner/status/917411963495469056 :

> How do I, with a browser, download a resource that's only available with
content-negotiation?

.zsync (~rsync over HTTP) support would be great for many applications
(really instead of gzip, AFAIU):
http://zsync.moria.org.uk


>
> Best,
>
> Lars
>
> *** Lesen. Hören. Wissen. Deutsche Nationalbibliothek ***
> --
> Dr. Lars G. Svensson
> Deutsche Nationalbibliothek
> Informationsinfrastruktur
> Adickesallee 1
> 60322 Frankfurt am Main
> Telefon: +49 69 1525-1752
> Telefax: +49 69 1525-1799
> mailto:l.svensson@dnb.de <javascript:;>
> http://www.dnb.de
>
>