I'm definitely having problems with this list. It seems postings by me have 
gone through, and I see discussions in the archive, but I haven't gotten any of 
them. Here's some followups. It does means the attributions aren't as precise 
as I would normally have, but it looks like I'm mostly replying to a set of 
posts by Peter Murray-Rust:

Let's see if this posting works....

PMR:
> > In practice I would see an Open standard as one that is:
> > * published
> > * redistributable without permission

Some specifications in this field were published in journals, such as SMILES 
and the original MDL connection table specification. There was also the SMD 
(Standard Molecular Data) Format effort described in

  Barnard, J.M. (1990) Draft specification for revised version
  of the Standard Molecular Data (SMD) Format, J. Chem. Inf.
  Comput. Sci., 30, 81-96.

Because JCICS owns the copyright to these, and prohibits redistribution without 
permission, do you mean to say that these are not open protocols? I'm sure the 
SMD people had in mind that it be to all intents and purposes "open".

(I'm told that SMD was one of several attempts in the 1980s and 1990s to 
develop a '"standard" connection table format'. The failure of these to make 
further inroads may be instructive, but I don't know enough of the details. 
I've wondered if I should go and do some PhD research on the history and 
development of chemistry software.)


PMR:
> If you are using "Standards" in a formal sense then there are very few in our 
> area.

What is a standard?

http://www.bsigroup.com/en/Standards-and-Publications/About-standards/What-is-a-standard/
> Put at its simplest, a standard is an agreed, repeatable way of doing 
> something. It is a published document that contains a technical specification 
> or other precise criteria designed to be used consistently as a rule, 
> guideline, or definition. Standards help to make life simpler and to increase 
> the reliability and the effectiveness of many goods and services we use. 
> Standards are created by bringing together the experience and expertise of 
> all interested parties such as the producers, sellers, buyers, users and 
> regulators of a particular material, product, process or service.
> 
> Standards are designed for voluntary use and do not impose any regulations. 

http://www.etsi.org/WebSite/Standards/WhatIsAStandard.aspx
> ISO/IEC Guide 2:1996, definition 3.2 defines a standard as:
> 
> 'A document established by consensus and approved by a recognized body that 
> provides for common and repeated use, rules, guidelines or characteristics 
> for activities or their results, aimed at the achievement of the optimum 
> degree of order in a given context'.


PMR:
> W3C protocols (XML, HTML, CSS) are not standards, they are "recommendations"
> IETF does not produce standards but drafts (e.g. MIME, URIs)
> The only standards are those that go through ISO (none in chemistry that
> matter) or ECMA. Things like Javascript, MS-OOXML (yes). Not Java.

The history of RFCs is clear that Postel used the term "request for comments" 
in part because the term "standard" was too strong of a word, and because RFCs 
include many things which are not standards (read RFC 1796 - "Not All RFCs are 
Standards"). Some of the RFCs are standards.

As to saying that "IETF does not produce standards but drafts", well, that's at 
direct odds with

  http://en.wikipedia.org/wiki/Internet_standard
  Internet Standards are created and published by the Internet
  Engineering Task Force (IETF).

  An Internet Standard is a special Request for Comments (RFC)
  or set of RFCs. An RFC that is to become a Standard or part
  of a Standard begins as an Internet Draft, and is later
  (usually after several revisions) accepted and published by
  the RFC Editor as a RFC and labeled aProposed Standard. Later,
  an RFC is labelled a Draft Standard, and finally a Standard.
  Collectively, these stages are known as the standards track,
  and are defined in RFC 2026.

PMR [on requirements for open standards]:
> It must have the trust of the community. That does not necessarily mean a
> completely democratic process. CML is a benevolent dictatorship with input
> from the community.

I happen to have trust in the MDL connection table specification. The 
maintainers have over the 15+ years, since its first publication in JCICS shown 
that they (that is, both MDL and Symyx) maintain it and improve it over the 
time, including the updates to handle limitations of the original format. The 
specification document is freely available and well done, though not 
redistributable. I personally have never needed nor requested a change or 
clarification to the spec so I do not know how open they are to that.

Do you concur? If not, can you describe the basis for which one has trust or 
not?

Note that for MDL/Symyx/whomever holds the copyright to that specification 
document, it is to their advantage that they publish the spec so that other 
programs can create data files used as input to their systems.

PMR:
> This is not normally possible with manufacturer formats - I cannot
> influence what Bruker, CambridgeSoft, etc. create. I cannot affect the next
> distrib of Pipeline Pilot.


Why not? Convince Pipeline Pilot that native support for (say) CML would be in 
their interests and they would likely change the code.

In any case, my pointing to software products is actually not relevant - would 
you point to a specific format or specification that they have which is not 
open?

Daylight had their CEX effort, which was

   http://psb.stanford.edu/psb96/sessions.html
   CEX (Chemical EXchange) is an interface designed for chemical
   information exchange and interoperability in today's
   distributed environment. CEX provides a complete, working
   and extensible chemical information interface as commercial-quality,
   public domain software to which vendors can easily and freely
   port. CEX is an object-oriented mechanism for exchanging
   information in a semantically well-defined and program-independent
   manner which is particularly well suited for chemical information.
   CEX was developed by an alliance of chemical information software
   consumers, mainly large pharmaceutical companies with the
   assistance of software vendors, and is hardware operating system
   and vendor independent.

and I've read their meeting notes in some archive somewhere.

Tripos documents their mol2 format - again, because it is to their advantage to 
do so - and I have requested and gotten support for understanding the format 
without being a licencee of their software.

By comparison I can also describe the difficulties I had some 10 years ago 
trying to get the PDB specification changed, where the spec clearly was in 
disagreement with how it was actually used. (One specific example was the X-ray 
resolution field, which was described as a floating point number when the 
number of significant digits actually had meaning, so that a resolution of 
"1.0" was not the same as "1.00". This prevented full round-tripping in my 
parser code.)

I can also point out academic open specifications where the original group has 
long since lost interested in maintaining or developing the specification, 
making it hard to influence any changes.

Best regards,

                                Andrew
                                [email protected]



------------------------------------------------------------------------------
Return on Information:
Google Enterprise Search pays you back
Get the facts.
http://p.sf.net/sfu/google-dev2dev
_______________________________________________
Blueobelisk-discuss mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/blueobelisk-discuss

Reply via email to