[tdwg-tag] Differences in thinking between TDWG and LinkedData groups about data sharing / integration

Fri Apr 24 04:16:01 CEST 2009

Pete

You should have sent something around on the mailing list, I could have given you an example of a TaxonOccurrence.  Or perhaps you did and I missed it???

Anyway, with Herb IMI, Paul Kirk and I have set up an resolver to provider TaxonOccurrence RDF data,
see for example urn:lsid:herbimi.info:specimens:100069 (or http://lsid.herbimi.info/authority/metadata/?lsid=urn:lsid:herbimi.info:specimens:100069 in your browser).  It also has an example of using Interaction data - ie in this case a host plant (IPNI ID) of a fungus (Herb IMI specimen) with an identification to a taxon concept and name (Index Fungorum name).

Jim - feel free to help improve the ideas and processes of TDWG if you find them that bad.   :-)

Kevin

From: tdwg-tag-bounces at lists.tdwg.org [mailto:tdwg-tag-bounces at lists.tdwg.org] On Behalf Of Peter DeVries
Sent: Friday, 24 April 2009 1:46 p.m.
To: Jim Croft
Cc: Bob Morris; tdwg-tag at lists.tdwg.org
Subject: Re: [tdwg-tag] Differences in thinking between TDWG and LinkedData groups about data sharing / integration

Hi Jim,

Thanks for keeping in good humor. :-)

I was trying to get my head around the TaxonOccurrence standard so I could rewrite my observation records,
and was hoping on finding some examples. That experience, and some related issues, made me a little irritable,
for that I am sorry.

It might be useful to make up a test set. It does not even need to be real. In my experience, this is where
you really start to see if the standard does what you want it to do. Many times I have thought that I had
everything figured out, only to discover after loading data into my triple store that something will not work
or something that used to work was now broken by the addition of a new feature.

I am not bothered by the idea of branding, it just seemed that the it was being confused with the goal of
persistence in what seemed to be a unproductive thread.

Thanks again for being amused, tolerant and insightful. :-)

- Pete

On Thu, Apr 23, 2009 at 7:50 PM, Jim Croft <jim.croft at gmail.com<mailto:jim.croft at gmail.com>> wrote:
Can't let this great opportunity pass...  :)

It is not just us.  It is the regrettable human condition.  See also:
http://www.informationweek.com/news/infrastructure/management/showArticle.jhtml?articleID=216600011

<diatribe_alert/>

One of the intriguing and enigmatic things about TDWG is that it does
not seem to respect its own standards, preferring to invent another
set rather than fix or enhance what is already there.  More than the
'not invented here' syndrome, we have to deal with 'not invented here
this week' and the user is left with the impression of a happy
self-absorbed group chasing a flock of flitting butterflies - 'oh,
that's pretty, I will try and catch that one'.  Yep, just the human
condition.  In another context, Roger H,
http://www.hyam.net/blog/archives/346, has provided the best
description of TDWG and its standards I think I have ever heard: "It
is kind of like the guy selling seagulls on the beach. You give him £5
and he points into the sky and says 'That one is yours'. Yes he is
providing a service but the relationship that counts is the one
between you and the gull."  This is both profound and frightening in
how close it is to reality.

On the complexity and costs of implementation, you are absolutely
right.  Taxonomic database standards have moved out of the domain of
taxonomists, they no longer understand what we are talking about
(hell, I no longer understand what we are talking about.  And the rest
of you?  c'mon now, be honest... :) and we are left with this
relationship of 'trust us' paternalism that I do not think is all that
healthy - we all trust and respect Microsoft, right?. It is getting
increasingly difficult give advice to a bod with a beat up PC in a
developing country on what they should do - the alphabet soup
surrounding GUIDS and the like just does not cut it when the mind set
is an Excel spreadsheet with a bunch of intuitive headings.  Oh, I'm
sorry - you mean it is not just in developing countries?  There is a
widening gap between 'Joe the Taxonomist' and business of taxonomy
data standards that we do not seem to be able to address (do we even
care?).  We used to be able to give our staff in the herbarium a TDWG
standard and say 'this is what we have to do.'  Not anymore...  Ah,
the good ole days...  Maybe the answer lies in the hegemony of a new
and benevolent  'MSOffice for Taxonomy'.  Could work, but I do not
think it is going to be particularly satisfying.

On the reuse issue, Greg W argues that we do not do nearly enough of
this and I agree with him.  He argues that TDWG should focus on the
standard and not the application and implementation of the standard
and has proposed that in our vocabularies we should adopt the
principles of nomenclatural priority, that is, going back to *Dublin*
Core, adding stuff chronologically from other other standards,
including our own, until there is no option other than to invent
another one, or there is nothing in our domain left to standardize.
For taxonomists there is something inherently attractive in this
approach - don't describe a taxon where it already exists, don't
invent a standard where one already exists.  To retrofit this and
untangle all the synonymy and homonymy in our existing standards and
implementations is going to take a lot of work though.  But the
vocabularies and ontologies are a good start.

On the 'branding, issue, it is not so much branding but attribution.
Apart from the moral and legal issues, it is unscientific not to
attribute, source and provide lineage for data.  The is no
optionality.  We have to do it.  Even if the initial supplier
'disappears'.  *Especially* if the initial supplier disappears.
Attribution (branding if you like) is absolutely essential for
credibility.  If someone is not going to do it, they can not have our
data, and we will not use theirs.

On the architectural issue, I can not really get all that hung up on
it.  If a standard is good, it should be able to be implemented in a
number of architectures (isn't that almost a definition of
interoperability?).  Where things get 'interesting' is when
architecture (and the continuum towards application) becomes the
standard or part of the standard.  TDWG needs to constantly ask itself
to what extent it needs to get involved with implementation of the
standards it promotes.  I would argue 'not at all', but this is
another discussion.

And the 'insider' cabalistic nature of TDWG?   What can I say - it has
always been this way.  A standard attracts a champion and the champion
establishes a fiefdom of acolytes around it.  Yep, the human
condition.  And it sort of works.  Some of the time.  (btw - another
artifact of the human condition - your brilliant ideas are never
perceived as such until someone else has them - just ask poor old
Wallace how he is feeling this year).  A downside of this approach is
that the various TDWG standards are very poorly coordinated between
each other - this is something we should be able to do something
about.

We could piss on the TDWG tent from the outside, but you have to
agree, it is much more satisfying to get inside the tent and piss on
and piss off everyone in it...  :)

<disclaimer>None of the above ideas are mine.  I am following the TDWG
standard practice of restating them without attribution :)
</disclaimer>

Ah... that was fun...

jim

On Fri, Apr 24, 2009 at 5:09 AM, Peter DeVries <pete.devries at gmail.com<mailto:pete.devries at gmail.com>> wrote:
> Respectfully,
> 1) Only certain classes of organizations will be able to contribute since
> the standard is requires special skills. Those groups that can pay for
> hardware and
>     a person specific to this standard for perpetuity. I look at this and
> think that a number of groups that could be providers cannot because of the
> way the
>     system is implemented. Why not have a simple RDF tar or zip file format
> that GBIF checks with a crawler every night?
> 2) There is very little reuse of existing vocabularies, geo for instance.
> Similar to the "not invented here mentality".
> 3) Discussions and decisions seem to be too much about making sure that
> providers keep their "brand" on the data even if they disappear.
> 4) Suggestions or alternative ways of thinking are rejected until an insider
> restates them without attribution
> 5) It is not at all clear how some of these decisions are made. It appears
> as if some people disagree, there is discussion. Then years later there is
>     the same discussion. It seems that some smaller group keeps pulling
> everyone back to the same architectural decision.
> 6) Where are the example data sets? We should have some example data sets
> available to see if the standard can be used to answer real questions?
>     Either they don't exist or they are only available to a few.
> I actually have nothing but praise for GBIF and uBio (except for the minor
> encoding thing), this more about trying to work within TDWG and getting
> stonewalled. I am having the same feelings about it that I had a few years
> ago, after which I left to try to make something that worked so I could
> proceed with my project.
> It probably was unfair to imply that the fiefdoms are by design, rather than
> a side effect of the implementation standards, and for that I apologize.
> - Pete
>
>
> On Thu, Apr 23, 2009 at 10:56 AM, Bob Morris <morris.bob at gmail.com<mailto:morris.bob at gmail.com>> wrote:
>>
>> "described by anyone" is not the same as "described by anyone in any way
>> convenient to the describer", so I find this quotation somewhat
>> disingenuous. More precisely, I wonder what TDWG standard or proposed
>> standard you find enables fiefdoms \in ways that are impossible under some
>> other solution to the problem the standard addresses/.
>>
>> Bob Morris
>>
>> On Thu, Apr 23, 2009 at 10:47 AM, Peter DeVries <pete.devries at gmail.com<mailto:pete.devries at gmail.com>>
>> wrote:
>>>
>>> This paragraph below seems to encapsulate the differences in thinking
>>> between the linkeddata community and
>>> some of the TDWG people on how to best share biodiversity data.
>>> "The notion of a fabric of resources that are individually described,
>>> queried, and resolved may seem unmanageable or like science fiction. For
>>> organizations that are used to large, manual, centralized efforts to
>>> standardize on everything, it may seem anarchic to allow resources to grow
>>> organically and be described by anyone. The same people would probably not
>>> believe the Web possible in the first place if there were not already ample
>>> proof of its success."
>>> REST for Java developers, Part 4: The future is RESTful
>>>
>>> From http://www.javaworld.com/javaworld/jw-04-2009/jw-04-rest-series-4.html?page=4
>>> I think that some people may have lost sight of the goal of making data
>>> available to improve the understanding of our
>>> natural world and hopefully better manage our natural resources.
>>> It does not seem that creating a distributed network of fiefdoms will
>>> help us achieve this goal.
>>> - Pete
>>> I was led to this article by @janzemanek on twitter.
>>> ---------------------------------------------------------------
>>> Pete DeVries
>>> Department of Entomology
>>> University of Wisconsin - Madison
>>> 445 Russell Laboratories
>>> 1630 Linden Drive
>>> Madison, WI 53706
>>> ------------------------------------------------------------
>>>
>>> _______________________________________________
>>> tdwg-tag mailing list
>>> tdwg-tag at lists.tdwg.org<mailto:tdwg-tag at lists.tdwg.org>
>>> http://lists.tdwg.org/mailman/listinfo/tdwg-tag
>>>
>>
>>
>>
>> --
>> Robert A. Morris
>> Professor of Computer Science
>> UMASS-Boston
>> ram at cs.umb.edu<mailto:ram at cs.umb.edu>
>> http://bdei.cs.umb.edu/
>> http://www.cs.umb.edu/~ram
>> http://www.cs.umb.edu/~ram/calendar.html
>> phone (+1)617 287 6466
>
>
>
> --
> ---------------------------------------------------------------
> Pete DeVries
> Department of Entomology
> University of Wisconsin - Madison
> 445 Russell Laboratories
> 1630 Linden Drive
> Madison, WI 53706
> ------------------------------------------------------------
>
> _______________________________________________
> tdwg-tag mailing list
> tdwg-tag at lists.tdwg.org<mailto:tdwg-tag at lists.tdwg.org>
> http://lists.tdwg.org/mailman/listinfo/tdwg-tag
>
>

--
_________________
Jim Croft ~ jim.croft at gmail.com<mailto:jim.croft at gmail.com> ~ +61-2-62509499

"Words, as is well known, are the great foes of reality."
- Joseph Conrad, author (1857-1924)

"I know that you believe that you understood what you think I said,
but I am not sure you realize that what you heard is not what I meant."
 - attributed to Robert McCloskey, US State Department spokesman

--
---------------------------------------------------------------
Pete DeVries
Department of Entomology
University of Wisconsin - Madison
445 Russell Laboratories
1630 Linden Drive
Madison, WI 53706
------------------------------------------------------------

________________________________
Please consider the environment before printing this email
Warning: This electronic message together with any attachments is confidential. If you receive it in error: (i) you must not read, use, disclose, copy or retain it; (ii) please contact the sender immediately by reply email and then delete the emails.
The views expressed in this email may not be those of Landcare Research New Zealand Limited. http://www.landcareresearch.co.nz
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.tdwg.org/pipermail/tdwg-tag/attachments/20090424/b35eb5fd/attachment.html 