|
From: | June Finch |
Subject: | Re: [Myexperiment-discuss] Workflow fact sheet for OAI |
Date: | Wed, 11 Apr 2007 11:42:47 +0100 |
User-agent: | Thunderbird 1.5.0.7 (Windows/20060909) |
Antoon Goderis wrote:
I understand your point - however, there is a distinction between what is really mandatory in delivering a workflow and what is extremely desirable from a best practice point of view - these are two different objectives. We do have to consider barriers to adoption in our system design (and particularly on interface definitions where we want to encourage the future possibility of distributed repositories that are outside of our control).We need to be extremely careful about what we do include asmandatory(while we may be able to derive some of this from Tavernaworkflows,would that necessarily be the case for other workflow systems?)I couldn't think of counter examples from the systems I do knowI want to add that one of the few systems that managed to make inroads with end users in terms of them voluntarily registering services, BioMoby, set the bar quite high when it came to minimal information required. It's not a bad thing to have high standards sometimes. Or at least to have the ability to separate work in progress from the good stuff - cfr the joys of debugging others' code.
- We should discuss this tomorrow. June.
Antoon
[Prev in Thread] | Current Thread | [Next in Thread] |