On Tue, 7 Jan 2003 08:38:56 -0700, Jeremy Wadsack wrote: >Ok, but just for the sake of argument, I don't see anything about the >*formatting* in the new samples that can't be done in the current HTML >2.0 with Stylesheets. (Well, except for the General Summary >right-aligned values.)
This is probably true. I haven't added anything specific to analogs analysis of the data, nor did I want to actually change the reports as such, but just give the user the ability. Initially I did have one or two things in mind, but I quickly got sidetracked into looking at separation of formatting and content. I believe analog should generate the content, but leave the formatting to the browser. By having most if not all formatting specified by CSS, the user is totally free to choose how he/she wants to display the data. On Tue, 7 Jan 2003 16:06:35 +0000 (GMT), Stephen Turner wrote: >On Tue, 7 Jan 2003 [EMAIL PROTECTED] wrote: >> >> XHTML using CSS gives the standards-orientated web developer much more >> control - way beyond, for instance, what the bars look like - over the >> look of the page as well as potentially reducing the download size, by >> moving style instructions to the stylesheet. > >The separation of markup and style is not really a valid argument, becuse >analog already allows the user to specify a style sheet. But with very limited functionality. To change the bars for instance, I need to change the config file, then rerun analog. Just to change the formatting, not the content. (not that I have any particular desire, but for the sake of argument :-) On Tue, 7 Jan 2003 16:06:35 +0000 (GMT), Stephen Turner wrote: > >What I'm most worried about is people who have to use text browsers over a >slow connection, or speaking browsers because they're blind. Text browsers - lynx has no problem with the demo-page. Slow connection - until around 1630CET today, jessen.ch was on 64kbit/s. (it just now got upgraded to 352kbit/s) Speaking browsers - if a properly enabled CSS style-sheet is used, this can work far better than without. (granted, I don't know what the current standards support is like outside the scope of lynx/netscape/msie/opera/konqueror). >Although I >still know a surprising number of people who are stuck with old browsers >because their machine won't run newer ones -- maybe I just know too many >people in cash-strapped academic departments. 99% coverage isn't good enough >as far as I'm concerned, at least not when I don't see any compensating >improvement from changing to XHTML. I think the compensating improvement is: the user is almost 100% in control of how the report is presented. Be it in a speaking browser, in black&white or whatever. Maybe this could be achieved with HTML2.0 too, I can't say. But, who really maintains websites in HTML2.0 only ? I agree it IS good for compatibility etc., but at some point it's necessary to move on. (even IBM with their legendary backward compatibility dispensed with full S360 compatibility in the zSeries). Anyway, keeping HTML2.0 support along with what I'm proposing isn't THAT difficult. I've introduced a new output format anyway (configuration- wise), but as the two are far from separate (code-wise), it'll take a lot more testing. (I was merely trying to save myself some regression-testing ...) /Per regards, Per Jessen, Zurich http://www.enidan.com - home of the J1 serial console. +------------------------------------------------------------------------ | This is the analog-help mailing list. To unsubscribe from this | mailing list, go to | http://lists.isite.net/listgate/analog-help/unsubscribe.html | | List archives are available at | http://www.mail-archive.com/analog-help@lists.isite.net/ | http://lists.isite.net/listgate/analog-help/archives/ | http://www.tallylist.com/archives/index.cfm/mlist.7 +------------------------------------------------------------------------
