From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jeff King Subject: Re: [PATCH 2/2] gitweb: Show trailing slash when listing tree entry in tree listing Date: Tue, 10 Oct 2006 02:58:49 -0400 Message-ID: <20061010065849.GA2413@coredump.intra.peff.net> References: <20061010054643.GA565@coredump.intra.peff.net> <20061010064117.86409.qmail@web31813.mail.mud.yahoo.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Junio C Hamano , Petr Baudis , git@vger.kernel.org X-From: git-owner@vger.kernel.org Tue Oct 10 08:59:03 2006 Return-path: Envelope-to: gcvg-git@gmane.org Received: from vger.kernel.org ([209.132.176.167]) by ciao.gmane.org with esmtp (Exim 4.43) id 1GXBZs-00016g-95 for gcvg-git@gmane.org; Tue, 10 Oct 2006 08:59:00 +0200 Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S965050AbWJJG6w (ORCPT ); Tue, 10 Oct 2006 02:58:52 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S965046AbWJJG6w (ORCPT ); Tue, 10 Oct 2006 02:58:52 -0400 Received: from 66-23-211-5.clients.speedfactory.net ([66.23.211.5]:44758 "HELO peff.net") by vger.kernel.org with SMTP id S965050AbWJJG6v (ORCPT ); Tue, 10 Oct 2006 02:58:51 -0400 Received: (qmail 23402 invoked from network); 10 Oct 2006 02:58:49 -0400 Received: from unknown (HELO coredump.intra.peff.net) (10.0.0.2) by 66-23-211-5.clients.speedfactory.net with SMTP; 10 Oct 2006 02:58:49 -0400 Received: by coredump.intra.peff.net (sSMTP sendmail emulation); Tue, 10 Oct 2006 02:58:49 -0400 To: Luben Tuikov Content-Disposition: inline In-Reply-To: <20061010064117.86409.qmail@web31813.mail.mud.yahoo.com> Sender: git-owner@vger.kernel.org Precedence: bulk X-Mailing-List: git@vger.kernel.org Archived-At: On Mon, Oct 09, 2006 at 11:41:17PM -0700, Luben Tuikov wrote: > Then let's universally underline absolutely _every_ link in gitweb > which is clickable, regardless of where it appears, the font, typeset > and size. Instead, let's make a strawman argument! Though I agree that it would be nicer for ALL links in gitweb to be consistent, I think there is an argument to be made about look. However, the specific example I mentioned is a single list in which some elements are underlined and blue (which has been the classic user interface hint for a link for a decade), and some aren't. Do you see why I think that might be inconsistent? > Unless you have "a priori" knowlege of "underline means clickable" there Which was my argument in the first place (note that I was talking about people with a high degree of computer exposure). > is no chance of thinking that "not-underlined means not-clickable". There is clearly a non-zero chance. Here's a relatively ridiculous argument. Look at the 'summary' page for a project. For each commit, there are blue and underlined 'commitdiff', 'text', and 'snapshot' links. The date, author, and message text have no such decoration. I click on the underlined things and see that they are all links. I click on date and author and see they are not links. The pattern of underlining links has held for five out of six elements. Do you think it's unreasonable to guess that the sixth element is not a link based on that pattern? Look, I agree that not underlining everything might make the page look nicer. And if we want to balance consistency against aesthetics, that's fine. But please don't argue that there isn't an inconsistency. -Peff