I agree that this proposal has no real meat to it. It's just playing around with names. I do think, however, that it is valuable nonetheless. The idea of v2 is to make DRMAA more accessible and useful to a wider audience. If now isn't the time to clean up ugly parts of the API, when is? We renamed the drmaa_job_ps() operation in the IDL spec for exactly the same reason. I fail to see how the cost is more than some time in one meeting and a few copies and pastes in the spec, and that's a price I'm willing to pay to have a respectable API that looks like it was designed by professionals. Daniel Peter Tröger wrote:
Am 23.01.2009 um 17:15 schrieb Roger Brobst:
My preference would be to not change drmaa_control(), and not worry too much about consistency with drmaa_control() when developing subsequent APIs.
Same for me. We somehow agreed to focus only on real improvements in DRMAA2, since everything else unnecessarily slows us down and creates a new API instead of an update.
/Peter.
-Roger
----Original Message---- From: Daniel Templeton <Dan.Templeton@Sun.COM> Sender: drmaa-wg-bounces@ogf.org To: DRMAA Working Group <drmaa-wg@gridforum.org> Subject: [DRMAA-WG] Another proposal Date: Thu, 22 Jan 2009 13:38:24 -0800
A customer just pointed out the awkwardness of the drmaa_control() operation. As a rule, I don't like operations whose behavior is controlled by a flag. It's generally easier to learn and easier to read is the name of the operation is descriptive of exactly what it does. What would people think of replacing the drmaa_control() operation with drmaa_suspend_job(), drmaa_release_job(), drmaa_hold_job(), drmaa_resume_job(), and drmaa_terminate_job() operations. I don't see the downside, myself.
Daniel -- drmaa-wg mailing list drmaa-wg@ogf.org http://www.ogf.org/mailman/listinfo/drmaa-wg -- drmaa-wg mailing list drmaa-wg@ogf.org http://www.ogf.org/mailman/listinfo/drmaa-wg
-- drmaa-wg mailing list drmaa-wg@ogf.org http://www.ogf.org/mailman/listinfo/drmaa-wg