Discussion List Archives

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Revised CIF syntax guidelines

It is not my intention to create an absolute code, rather to create a
framework for discussions and promote a certain prejudice in favour of
the positions outlined in the guidelines.  COMCIFS voting members
still remain the final arbiters.

Moving on to the particular comments about points 1(i) and (ii), note
that the phrase "domain level" is defined in the preamble to include
DDL dictionaries.  Insofar as a particular DDL can be used across all
domains, changes that do not satisfy all of the criteria in 1, but do
satisfy 1(ii), would logically be implemented using DDL mechanisms (ie
at the "domain level").  So, for example, while fancy syntax could be
introduced to indicate more detail about the relationships among data
items in a data file, which would certainly be broadly useful, this
has been done at a DDL level instead. Does this adequately answer this
aspect of your concerns, Brian?

The intent of 1(i) and 1(ii) is to shelter the syntax from unnecessary
complexity.  Simplicity is both a philosophical and a practical goal.
 It is up to COMCIFS to decide on whether or not it supports
simplicity as a philosophical goal; there appears to be no fundamental
mandate on COMCIFS to support such a philosophy, so it is entirely
possible for COMCIFS to reject these goals.

The practical outcome is that a simpler syntax makes implementation
easier. Consider the conservation principles put forward by Peter
Murray-Rust many years ago (see
http://www.iucr.org/__data/iucr/lists/comcifs-l/msg00115.html)

* You cannot hide complexity, you can only move it around
* for everything you define in a specification, someone has to write code
* it is far easier to write a specification than to implement it

The draft guidelines move complexity to the domain level wherever
possible, thereby reducing the workload on those who need to write
software that needs only to understand CIF syntax and a set of
specific datanames/values (a large class of software).

On Mon, Apr 4, 2011 at 7:22 PM, Brian McMahon <bm@iucr.org> wrote:
> Where matters of principle are involved, I have a certain reluctance
> to work from an absolute written code. Perhaps it stems from cultural
> familiarity with British polity, that does not operate from a written
> constitution (though be aware that the modern Irish state, founded on
> a written constitution, also features large in my cultural baggage).
> In any case, unyielding adherence to formal principles is not wise;
> and a code of principle should always, in my view, be seen as guiding
> rather than prohibitive. Of course, the better formulated the
> principles, the less will they be tested, on the whole.
>
> Having said which, I'm in broad agreement with the thrust of the
> CIF 2.0 statements. Where I have some concerns:
>
> Treating (i) and (ii) as exclusive does seem dangerous. Where it is
> seen that a base-line change could facilitate operation across many
> domains, it might make more sense to implement it at that level,
> rather than have the different domains separately (and possibly
> incompatibly) implement it in dictionaries. The merits would need
> to be considered case by case; but I would be happier if these two
> statements were rolled into an OR complex:
>
>  (i) Implementation of the desired behavior by changes at the domain
> level is not feasible, or else such changes, while feasible,  would
> significantly reduce human readability; OR the change provides
> significant new functionality that is widely applicable to those
> scientific domains where CIF is used
>
> As a principle, point 3 ("Non-technical issues should be dealt with in
> non-technical arenas") seems to me tautologous. As such, I do not
> contest it; but neither would I be troubled if it were omitted from
> the list. In practice, of course, it's fair to insist that the
> scope of particular discussions be constrained to technical issues
> (or, in other circumstances, strictly non-technical issues) where
> it is clear what these might be.
>
> (Re-reading that, it seems to me rather opaque. What I mean is that,
> broadly speaking, I support the way we have been approaching things,
> with the establishment of working groups to consider proposed
> technical developments, and occasional referral upwards to COMCIFS
> for advice when progress seems to be getting hung up on concerns
> over whether we're missing wider implications. Perhaps that referral
> might at times come in at a much earlier stage.)
>
> Regards
> Brian
>
> On Thu, Mar 31, 2011 at 03:20:57PM +1100, James Hester wrote:
>> Dear COMCIFS,
>>
>> Please find below a slightly revised version of the guidelines for
>> developing base CIF syntax and semantics.  I have taken the version
>> most recently posted by John Bollinger, and following discussion with
>> John Westbrook, have added a couple of points relating to maintaining
>> compatibility with previous versions of CIF (1 (vi) and (vii)).  Note
>> also that I have also adjusted the text in 1 (ii) to refer to
>> scientific domains in which CIF is used, rather than scientific
>> domains in general.
>>
>> I plan to call a vote on accepting these guidelines following a short
>> further period for discussion of the revision below.
>>
>> ==============================================
>> Principles guiding development of Base CIF 2.0
>> ----------------------------------------------------------------------
>>
>> Preamble
>>
>> CIF is a framework for exchanging and archiving scientific data,
>> featuring a human-readable, machine-parseable, file format designed to
>> serve as an exchange and archive medium.  'Base' CIF comprises the
>> definitions and constraints that underlie CIF and apply to all CIF
>> files; those aspects defining the CIF file format are documented in
>> the CIF Syntax specification and the CIF Common Semantic Features
>> specification.
>>
>> Base CIF aims to remain as simple as possible by delegating
>> considerations such as ontology, vocabulary, data relationships, and
>> complex and rich data types to domain dictionaries and the DDL
>> formalisms by which those dictionaries are defined.  In the following,
>> the phrase 'domain level' refers to such documents (though it is
>> anticipated that only dictionaries, not DDLs, will be
>> domain-specific).  Definitions and constraints at domain level apply
>> to a particular CIF file only as declared by that file or as required
>> by a particular CIF processor in a particular context.
>>
>> Principles
>>
>> The design of base CIF 2.0 is guided by these principles:
>>
>> 1. A feature should be added to or changed in base CIF only if all of
>> the following are satisfied:
>>
>>  (i) Implementation of the desired behavior by changes at the domain
>> level is not feasible, or else such changes, while feasible,  would
>> significantly reduce human readability;
>>  (ii) the change provides significant new functionality that is widely
>> applicable to those scientific domains where CIF is used
>>  (iii) reliable transfer and archiving of data is not compromised
>>  (iv) there is no simpler way of achieving the desired behaviour
>>  (v) it has been shown possible to implement the change at a cost
>> commensurate with its benefits, as demonstrated in part by a rough
>> consensus and running code.
>>  (vi) Where possible, any new CIF syntax features should be developed
>> as an extension to the current standard, and thus not change the
>> interpretation of archival files that conform with previous versions
>> of the CIF standard.
>>  (vii) Where it is impractical to provide for full backward compatibility as
>> described in (vi), the relevant archival repositories and software developers
>> should be consulted to arrive at a solution that will minimize the impact of
>> such changes.
>>
>>
>> 2. As long as the requirements in (1) are satisfied, base CIF should:
>>  (i) behave in a way that is consistent with common usage
>>  (ii) align with pre-existing standards where those standards provide
>> the required behaviour. CIF 1.1 can be considered a pre-existing
>> standard for CIF 2.0 in this context.
>>
>> 3. Non-technical issues should be dealt with in non-technical arenas.
>>
>> 4. Draft changes to base CIF will be made available on the IUCr
>> website for public comment for a period of at least 6 weeks, following
>> which COMCIFS voting members, after consideration of any objections
>> raised, can vote to accept the change. A change will be accepted if
>> 3/4 of COMCIFS voting members approve it.
>> ===============
>>
>> --
>> T +61 (02) 9717 9907
>> F +61 (02) 9717 3145
>> M +61 (04) 0249 4148
>> _______________________________________________
>> comcifs mailing list
>> comcifs@iucr.org
>> http://scripts.iucr.org/mailman/listinfo/comcifs
> _______________________________________________
> comcifs mailing list
> comcifs@iucr.org
> http://scripts.iucr.org/mailman/listinfo/comcifs
>



-- 
T +61 (02) 9717 9907
F +61 (02) 9717 3145
M +61 (04) 0249 4148


Reply to: [list | sender only]