Return-Path: Mailing-List: contact cocoon-dev-help@xml.apache.org; run by ezmlm Delivered-To: mailing list cocoon-dev@xml.apache.org Received: (qmail 92564 invoked from network); 9 Nov 2000 11:14:16 -0000 Received: from cthulhu.ecom.be (HELO cthulhu.gent.ecom.be) (194.7.39.101) by locus.apache.org with SMTP; 9 Nov 2000 11:14:16 -0000 Received: from byron.gent.ecom.be ([172.16.20.7]) by cthulhu.gent.ecom.be (Netscape Messaging Server 3.6) with ESMTP id AAA27D7 for ; Thu, 9 Nov 2000 12:29:55 +0100 Received: by byron.gent.ecom.be with Internet Mail Service (5.5.2650.21) id ; Thu, 9 Nov 2000 12:04:41 +0100 Message-ID: <703B15C33AA4D411A69000508B09061D01B296@byron.gent.ecom.be> From: "Tom Klaasen" To: "'cocoon-dev@xml.apache.org'" Subject: RE: [C2]Action proposal (long) Date: Thu, 9 Nov 2000 12:04:37 +0100 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2650.21) Content-Type: multipart/alternative; boundary="----_=_NextPart_001_01C04A3C.DBB9BD50" X-Spam-Rating: locus.apache.org 1.6.2 0/1000/N This message is in MIME format. Since your mail reader does not understand this format, some or all of this message may not be legible. ------_=_NextPart_001_01C04A3C.DBB9BD50 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable > -----Original Message----- > From: Sylvain Wallez [mailto:sylvain.wallez@anyware-tech.com] > Sent: donderdag 9 november 2000 11:58 >=20 > Giacomo Pati a =E9crit : > >=20 > > Hi all > >=20 > > I'd like to thank you all for patiently waiting for this proposal. > > I know I was very ternse the last days. This is because I'm have > > to finish some projects apart from Cocoon. > >=20 > > Action proposal > >=20 > >=20 > A few thoughts that came to mind when reading the proposal. >=20 > As far as pipeline description is concerned, the sitemap is great : a > pipeline is simple sequence of transformations and the associated > language is compact. >=20 > Actions (which I think are *really* needed to build web apps with > Cocoon) are more complex since they tackle the logic of the=20 > application, > which can have a complicated structure. Action-chains are a kind of > reusable block (or a method ?), but I think very soon new needs will > emerge when people will use actions : if-then constructs, iterations, > etc. >=20 > So why not allow a kind of tag in some sections of the > sitemap to allow people to express complex behaviours=20 > directly in Java ? > Simple things should be simple (use the limited sitemap language wich > covers 80% of your needs), complex things should be possible (do it = in > Java). Moreover, which not allow map logicsheets which would allow to > factorize high-level constructs such as web site structure or=20 > behaviour > (actions) using custom tags ? >=20 > What do you think of it ? My gutt feeling with Giacomo's proposition was that there would be too = much "programming logic" (as opposed to "administration logic") in the = sitemap. And you seem to feel the other way around. I really think the sitemap shouldn't allow too much programming, = because in the end, you would end up programming everything in the sitemap. As = stated in Stefano's slides: "A publishing framework should impose strict yet flexible practices". Let's not forget the "strict" (which in the end is necessary to allow the "flexible", imho) tomK ------_=_NextPart_001_01C04A3C.DBB9BD50 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable RE: [C2]Action proposal (long)

> -----Original Message-----
> From: Sylvain Wallez [mailto:sylvain.wallez@an= yware-tech.com]
> Sent: donderdag 9 november 2000 11:58
>
> Giacomo Pati a =E9crit :
> >
> > Hi all
> >
> > I'd like to thank you all for patiently = waiting for this proposal.
> > I know I was very ternse the last days. = This is because I'm have
> > to finish some projects apart from = Cocoon.
> >
> > Action proposal
> >
> </snip>
>
> A few thoughts that came to mind when reading = the proposal.
>
> As far as pipeline description is concerned, = the sitemap is great : a
> pipeline is simple sequence of transformations = and the associated
> language is compact.
>
> Actions (which I think are *really* needed to = build web apps with
> Cocoon) are more complex since they tackle the = logic of the
> application,
> which can have a complicated structure. = Action-chains are a kind of
> reusable block (or a method ?), but I think = very soon new needs will
> emerge when people will use actions : if-then = constructs, iterations,
> etc.
>
> So why not allow a kind of <map:logic> = tag in some sections of the
> sitemap to allow people to express complex = behaviours
> directly in Java ?
> Simple things should be simple (use the limited = sitemap language wich
> covers 80% of your needs), complex things = should be possible (do it in
> Java). Moreover, which not allow map = logicsheets which would allow to
> factorize high-level constructs such as web = site structure or
> behaviour
> (actions) using custom tags ?
>
> What do you think of it ?

My gutt feeling with Giacomo's proposition was that = there would be too much "programming logic" (as opposed to = "administration logic") in the sitemap. And you seem to feel = the other way around.

I really think the sitemap shouldn't allow too much = programming, because in the end, you would end up programming = everything in the sitemap. As stated in Stefano's slides: "A = publishing framework should impose strict yet flexible practices". = Let's not forget the "strict" (which in the end is necessary = to allow the "flexible", imho)

tomK

------_=_NextPart_001_01C04A3C.DBB9BD50--