- From: Imsieke, Gerrit, le-tex <gerrit.imsieke@le-tex.de>
- Date: Sun, 17 Sep 2017 14:02:58 +0200
- To: XProc Dev <xproc-dev@w3.org>
At this week’s XProc Next Community Group [1] meeting [2] in Aachen, Gioele Barabucci proposed some changes to p:pipeline [3] that would make it a bit more versatile and avoid or delay the switch to p:preclare-step proper that many pipeline authors have experienced as inevitable after some evolutionary “steps” of their pipelines. There was quick consensus that the p:pipeline shortcut, contrary to what the WG assumed when they wrote the XProc 1.0 spec, has note been used very frequently, and that the existence of this shortcut made teaching XProc a bit harder than if it were not present at all. So we agreed on removing p:pipeline in 3.0. Some argued in favor of keeping both names as synonyms for p:declare-step, like xsl:stylesheet and xsl:transform in XSLT. The group could not agree on that, again for teachability reasons. But some reasons were brought up why renaming p:declare-step could be beneficial. For example, when p:declare-step has no @type attribute, it declares a pipeline that cannot be used as a step in another pipeline, therefore p:declare-pipeline might be adequate. One argument against this was that if you want to turn a declared pipeline into a step by adding a type attribute, you shouldn’t need to rename the whole thing. Consequently, some suggested to have p:declare-pipeline and p:declare-step as synonyms. Others wanted to make it concise by renaming it p:step (or p:pipeline). Some replied that we had just removed declaration and usage ambiguity for p:input/p:with-input [4] and that we should explicitly state that we are declaring a step rather than using it. Still others weighed in that in case of p:input, p:output, and p:option, there is no 'declare-' in it, either, stating that there can be no confusion between a step’s declaration and a step’s invocation because a step is invoked by an element whose name is the declaration’s @type attribute anyway, rather than by a ficticious p:invoke-step or, gasp, p:with-step element. In the end, there were good arguments for renaming it p:declare-pipeline and also good arguments for keeping p:declare-step, but the overall inclination seemed to be that we should know this element by just one name all the time. Maybe the broader XProc community has some new proposals, arguments, thoughts on this? If you just want to support one of the existing proposals, you can participate in this doodle poll: http://doodle.com/poll/zicuz2tps5ax246f Add your comments by replying to this message or by adding a comment to the poll and/or the Github issue [3]. Gerrit [1] https://www.w3.org/community/xproc-next/ [2] https://github.com/xproc/Workshop-2017-09/wiki/Agenda-and-Minutes [3] https://github.com/xproc/3.0-specification/issues/136 [4] https://github.com/xproc/3.0-specification/issues/106 -- Gerrit Imsieke Geschäftsführer / Managing Director le-tex publishing services GmbH Weissenfelser Str. 84, 04229 Leipzig, Germany Phone +49 341 355356 110, Fax +49 341 355356 510 gerrit.imsieke@le-tex.de, http://www.le-tex.de Registergericht / Commercial Register: Amtsgericht Leipzig Registernummer / Registration Number: HRB 24930 Geschäftsführer: Gerrit Imsieke, Svea Jelonek, Thomas Schmidt, Dr. Reinhard Vöckler ------------------------------------------------------------------------------ Meet us at Frankfurt Book Fair Hall 4.2, L 72. More at https://www.le-tex.de/en/buchmesse.html
Received on Sunday, 17 September 2017 12:03:50 UTC