From Keith.Bierman@eng.sun.com Wed May 5 02:49:33 1993 Received: from Sun.COM by dkuug.dk with SMTP id AA26194 (5.65c8/IDA-1.4.4j for ); Wed, 5 May 1993 18:49:43 +0200 Received: from Eng.Sun.COM (zigzag-bb.Corp.Sun.COM) by Sun.COM (4.1/SMI-4.1) id AA13462; Wed, 5 May 93 09:49:39 PDT Received: from chiba.Eng.Sun.COM by Eng.Sun.COM (4.1/SMI-4.1) id AA10932; Wed, 5 May 93 09:49:38 PDT Received: by chiba.Eng.Sun.COM (4.1/SMI-4.1) id AA26072; Wed, 5 May 93 09:49:33 PDT Date: Wed, 5 May 93 09:49:33 PDT From: Keith.Bierman@eng.sun.com (Keith Bierman fpgroup) Message-Id: <9305051649.AA26072@chiba.Eng.Sun.COM> To: SC22WG5@dkuug.dk (SC22/WG5 members) Subject: re: (SC22WG5.352) X3H5, HPF and F90 Content-Length: 1119 X-Charset: ASCII X-Char-Esc: 29 There is overlap between HPF and X3H5. However, the X3H5 crowd asserts that they do serve different needs. HPF distributed memory data parallel directed decompostions X3H5 shared memory control based decompositions In addition, X3H5 was handed a charter to do Fortran and C. However, it is far from clear that the C crowd doesn't think that this is part of X3J11.1. X3H5 is at least part of the Standards universe, and can be directed to do least harm (or at least one can try ;>). HPF is basically a done deal, and there is nothing we can do about it. They avoided the Standards process entirely. >on parallel architectures. If an architecture independent optimisable language >is not yet possible then it is too early to standardise any of them. hear! hear!. However, there is the problem that unless there is some way to do the same thing on several platforms there may never be enough code to make critical mass. More interestingly, how do we *ever* get something stopped? LIA marches on (its committee voted to advance it to DIS status despite ordering a complete rewrite, from goals to specifics ...).