Re: CBOR Tutorial

I think that may really be a "yes," Leonard.  :-) . As my comment implied
in non-streaming environments, if top level compressed, you'd unzip the
whole thing before processing, and then regain random-access support.

BUT, I do think per-resource compression is a better approach regardless.

Best,
   Garth


On Wed, Jan 31, 2018 at 5:31 PM, Leonard Rosenthol <lrosenth@adobe.com>
wrote:

> No.
>
>
>
> See @Jeffrey <jyasskin@google.com> message about loss of random access
> when doing so.
>
>
>
> Leonard
>
>
>
> *From: *Garth Conboy <garth@google.com>
> *Date: *Wednesday, January 31, 2018 at 10:48 PM
> *To: *Leonard Rosenthol <lrosenth@adobe.com>
> *Cc: *Romain <rdeltour@gmail.com>, Baldur Bjarnason
> <baldur@rebus.foundation>, Ivan Herman <ivan@w3.org>, "Schindler Wolfgang
> Dr." <w.schindler@pons.de>, "Davis, Greg" <greg.davis@pearson.com>,
> Richard Wright <rkwright@geofx.com>, W3C Publishing Working Group <
> public-publ-wg@w3.org>, Jeffrey Yasskin <jyasskin@google.com>
> *Subject: *Re: CBOR Tutorial
>
>
>
>
>
>
>
> On Wed, Jan 31, 2018 at 9:01 AM, Leonard Rosenthol <lrosenth@adobe.com>
> wrote:
>
> Garth – you wouldn’t want to gzip the entire CBOR stream in a
> non-HTTP-based exchange.   (yes, you could, but it would mess up all sorts
> of things).  You would really want to use CBOR object compression which
> doesn’t exist today in the standard…
>
>
>
> Leonard
>
>
>
> Why?  If it was for non-HTTP (like FTP or USB stick) it wouldn't be a
> steam, just a blob, and you'd want to unzip before processing.  No?
>
>
>
> Best,
>
>    Garth
>
>
>
>
>