Re: [apps-discuss] the URI definition model (was: Fun with URLs and regex)

Bjoern Hoehrmann <> Fri, 30 January 2015 10:29 UTC

Return-Path: <>
Received: from localhost ( []) by (Postfix) with ESMTP id C14CF1A8AB8 for <>; Fri, 30 Jan 2015 02:29:26 -0800 (PST)
X-Virus-Scanned: amavisd-new at
X-Spam-Flag: NO
X-Spam-Score: 1.56
X-Spam-Level: *
X-Spam-Status: No, score=1.56 tagged_above=-999 required=5 tests=[BAYES_50=0.8, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_NONE=-0.0001, RCVD_IN_SORBS_WEB=0.77, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=no
Received: from ([]) by localhost ( []) (amavisd-new, port 10024) with ESMTP id CqsG3kKCAOuP for <>; Fri, 30 Jan 2015 02:29:25 -0800 (PST)
Received: from ( []) (using TLSv1.2 with cipher DHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by (Postfix) with ESMTPS id 595021A8AB7 for <>; Fri, 30 Jan 2015 02:29:25 -0800 (PST)
Received: from netb ([]) by (mrgmx101) with ESMTPSA (Nemesis) id 0Lm7MT-1Xho4S1Mg5-00Zi1z; Fri, 30 Jan 2015 11:29:12 +0100
From: Bjoern Hoehrmann <>
To: Sam Ruby <>
Date: Fri, 30 Jan 2015 11:29:06 +0100
Message-ID: <>
References: <> <>
In-Reply-To: <>
X-Mailer: Forte Agent 3.3/32.846
MIME-Version: 1.0
Content-Type: text/plain; charset="ISO-8859-1"
Content-Transfer-Encoding: 8bit
X-Provags-ID: V03:K0:D8+zuOeNlStRqix7JUG0Dl57ZQqumE4MMlaSZKVD/3P9tA4Ozhl OxtB855R8HPf9YvaVDvsn891EnkPPvH3Pjy039YSJud62qCtUXgLYfgKhoo8ynsdVZYmpez EQNgL8YwUYwgG6QDKK3ceug9nAXKSRxG5hoH/pB7A7HdLxmsb+Api4ZT7xsa0NfVCxTKwNG UrMfkZZNAWracW4WyzYFA==
X-UI-Out-Filterresults: notjunk:1;
Archived-At: <>
Cc: IETF Apps Discuss <>
Subject: Re: [apps-discuss] the URI definition model (was: Fun with URLs and regex)
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: General discussion of application-layer protocols <>
List-Unsubscribe: <>, <>
List-Archive: <>
List-Post: <>
List-Help: <>
List-Subscribe: <>, <>
X-List-Received-Date: Fri, 30 Jan 2015 10:29:27 -0000

* Sam Ruby wrote:
>Permit me to provide a use case for both queries and fragments?
>In a recent discussion with Tim Berners-Lee, he described a use case for 
>file: URIs, namely developing an HTML application locally (using only 
>relative references), and being able to push that code to a server and 
>have it "just work" when served over HTTP.
>HTML applications may include stylesheets and scripts.  Scripts, when 
>running, may do interesting things based on the query and fragment.  The 
>fragment, in particular, is routinely (ab)used by so-called "Single Page 

You can always use fragment identifiers regardless of the scheme used.
As for query strings, I would not be surprised if question marks are
interpreted as path data in many implementations. Either way, the use
case above was a valid one a decade ago, but nowadays it is so often
necessary, and much easier than in the past, to actually use `http` or
in fact `https` in the near future, that `file://` is not a plausible
way to develop, starting with the inability to use `XMLHttpRequest` on
`file://` in some browsers.
Björn Höhrmann · ·
D-10243 Berlin · PGP Pub. KeyID: 0xA4357E78 ·
 Available for hire in Berlin (early 2015)  ·