W3C home > Mailing lists > Public > public-semweb-lifesci@w3.org > May 2009

Re: Question about standardization

From: Bijan Parsia <bparsia@cs.man.ac.uk>
Date: Wed, 13 May 2009 11:21:34 +0100
Message-Id: <415C0928-F770-4DD3-9D20-B3355614C7E9@cs.man.ac.uk>
Cc: Michael Hucka <mhucka@caltech.edu>, Oliver Ruebenacker <curoli@gmail.com>
To: public-semweb-lifesci hcls <public-semweb-lifesci@w3.org>
On 13 May 2009, at 10:48, Oliver Ruebenacker wrote:

>      Hello Bijan, Michael, All,
>
>   The group does not deliver standards, but can submit something for
> consideration to W3C.

Hence my care in distinguishing what the group can and cannot do.

>  I remember this being discussed indetail in one
> of the phone conferences.

Well, if you would read a little more carefully, you wouldn't have to  
remember.

>   If SBML is not deserving of being a standard, the what is?

HTML.

> Something
> no one uses?

Ah yes, non sequiturial rhetorical questions that were obviously  
intended as a crushing dialectical blow, and succeed at being such  
--- just not in the intended direction.

There are a host of reasons to undertake standardization in a formal  
standards body and a host of reasons not to. The best one, in my  
opinion, is ensuring interoperability of existing systems (i.e., to  
reduce competition at one level to enable better competition at  
another level). In my opinion, standardization shouldn't be regarded  
as something akin to publishing a journal article (i.e., a sanctioned  
archival representation of a body of significant work). (BTW, this is  
just to give an example of what I would consider a bad reason, not to  
remotely suggest that this is "the reason" for pursuing SBML  
standardization. I don't *know* what the reasons are for pursuing  
SBML standardization, hence my asking for, you know, a reasonable  
discussion of the rationale. That discussion has to happen as there  
lots of people you have to convince aside from me.)

Standardization has costs that can harm efforts. It drains time,  
energy, money, control. For example, the standard expected commitment  
of an organization to a W3C WG is on engineer day a week (more, if  
you're an editor). A WG is a committee that you do not control the  
membership of. Indeed, that is the W3C...you have to convince a lot  
of people who *properly do not care about SBML* (or OWL, or...) that  
it's worth diverting the *extremely limited* resources the W3C has to  
SBML.

There's a lot of politics since there are limited resources. Your  
sort of know-nothing arrogance will, I suspect, play rather poorly.  
It is so playing with me and I started this conversation fairly  
sympathetic to SBML.

I've been involved in lots of W3C WGs both inside and outside the  
semantic web area and am trying to offer considered, reasoned,  
helpful advice based on my experience and understanding of the  
process. If you don't want that input, then yay. I'll stop now.

Cheers,
Bijan.
Received on Wednesday, 13 May 2009 10:17:42 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 18:00:55 GMT