[Pattern-language] Domesticating Pattern Languages

Helmut Leitner leitner at wikiservice.at
Sat Apr 18 01:27:07 PDT 2015 

Doug,

Am 17.04.15 um 00:41 schrieb Doug Schuler:
> I'm interested in your points because it will be certainly easier for the
> pattern language community to collaborate if we can agree on some
> definitions.
>
> Aldo and I have been using the expression meta-patterns because the
> idea (at least ultimately) is to write a pattern language whose domain
> is pattern languages. (Alexander's domain was architecture, the  built
> environment).
I've recenty started to make a difference between domains and sub-domains.
Domains, like "architecture", "software" or "medicine" are typically too
large to have just "a pattern language". They have thousands of patterns
and dozens or hundreds of pattern languages.

Pattern languages published are for "sub-domains" or concrete design 
targets,
maybe typically 10-200 patterns in size. Something a person can handle.
e. g. "patterns for building schools" (sub-domain of architecture)
e. g. "patterns for cloud computing" (sub-domain of software)
e. g. "patterns for large scale desaster management" (sub-domain of 
medicine)

In that sense I would even argue that Alexander's "A Pattern Language"
is not a pattern language book, because it contains parts of various PLs
"patterns for regional development"
"patterns for urban development"
"patterns for buildings"
"patterns for interiour design"
but each is not complete enough to be used as a practical design repertoire.
This might explain a part of its lack of success in architecture.

Therefore I consider APL as a "pattern showcase book" and suggest
that architect should go about finding and describing the hundreds of
pattern languages that probably exist in architecture.
> Our use of "meta" is in keeping  with examples such as metamathematics
> (mathematics about mathematics) and metatheory (theory  about theories).
it seems that "meta-" is not consistently used:

The one you describe as meta="X about X"
e. g. metadata, metatheory, metaknowledge

A slight variation meta="theory about X".
e. g. meta-ethics (is not ethics about ethics)
e. g. metalogic (is not logic about logic)

meta="beyond X"
e. g. metaphysics is beyond of physics. (neither "physics about physics" 
nor "theory about physics")
e. g. metacommunication, metamaterial, metaprogramming

meta="higher level"
e. g. meta-analysis, meta-heuristic

there are even other uses:
e. g. Metafiction is a literary device used to self-consciously and 
systematically draw attention to a work's status as an artifact. (maybe 
a beyond-case)
e. g. Metagenomics is the study of genetic material recovered directly 
from environmental samples. (inherent genetic diversity)

> If we were going to write a pattern language on pattern language research,
> I'd think it were just an "ordinary" pattern language.  (/first order/ 
> pattern language?)
>
And is this something you suggest or that you are doing?
Or do you consider this as something trivial?

Could you give some examples of meta-patterns?

> To my mind the example you use for a BUILDING isn't convincing. To my
> mind BUILDING could be a pattern that "solved" the problem of the need for
> enclosure.

Mmmh, this may be different from English to German.
In German the corresponding word for BUILDING ("Bauwerk")
includes everything that is build, also BRIDGE, MEMORIAL,
or HIGHWAY, clearly not enclosing. Maybe in English
ARCHITECTURE would be the vage meta-pattern (in my world).

Sometimes languages are astonishlingly different,
like in "quality without a name" (English) and
having the name "Lebendigkeit" (German).

> True it's more general than SCHOOL BUILDING but a SCHOOL
> BUILDING FOR ART is more specific than SCHOOL BUILDING but that
> doesn't mean that SCHOOL  BUILDING couldn't be a pattern.
I agree on that.
But one can always create a categorial tree of arbitrary fine 
differentiation.
and the problem is not eliminated by that argument.

The problem-solution character is lost when abstracting,
maybe even gradually, at certain levels of abstraction.

> I extended this conversation to  the group.  I'm assuming that's OK with
> Helmut.
>
Yes, of yourse.

Cheers!
Helmut


> Thanks!
>
> — Doug
>
>
>
>> On Apr 14, 2015, at 11:30 PM, Helmut Leitner <leitner at wikiservice.at 
>> <mailto:leitner at wikiservice.at>> wrote:
>>
>> Doug,
>>
>> I'm very looking forward to Krems, and to your corresponding 
>> contribution.
>>
>> I see that we use different terminology, though.
>>
>> So, this posting tries to clarify our terminologies,
>> maybe as a step to unify them.
>>
>> What you name meta-patterns is to me "patterns of pattern research".
>> I see a different design target, e. g. a PROJECT PATTERN DISSERTATION.
>> There are different patterns used, but I see no serious difference in 
>> the
>> quality of patterns or methodology that would justify a new concept.
>>
>> For example, there is a PATTERN REPOSITORY (as a dirty research workshop,
>> containing patterns and pattern candidates, mature and not, and all 
>> kinds
>> of background information; as a basis for collaboration and supporting
>> the creattion of many different pattern products).
>> I assume PATTERN REPOSITORY is a meta-pattern in your sense.
>>
>> Corresponding is the PATTERN PRODUCT, a meta-pattern in my sense.
>> This is an abstract categorial construction.
>> It has has real patterns as instances:
>> e. g. PATTERN LANGUAGE BOOK
>> e. g. PLOP PATTERN PAPER
>> e. g. PATTERN CARD STACK
>> ...
>>
>> Meta-pattern is a pattern abstraction,
>> where the problem-solution-relationship is lost.
>>
>> For example,
>>   RAILROAD STATION
>>   SCHOOL BUILDING
>> are clear problem-solution-patterns.
>>
>> If we abstract from these concrete patterns, and talk about BUILDING
>> as a pattern (which shouldn't this be possible?), the problems are lost.
>> Although we may safely assume that each BUILDING solves
>> a problem, we can't name it, and can't reason about it.
>> Therefore, BUILDING is not a real pattern, it is something different !!!
>> Therefore, currently I talk about BULDING as a meta-pattern,
>> lacking a better word.
>>
>> Often the discussions in communities seems move to this "meta"-level
>> and loose contact to real implementation issues. And one has
>> to try to bring it down to the concrete level to find solutions.
>>
>> Do you see these phenomena too?
>> Have you named them?
>>
>> Best,
>> Helmut
>>
>> Am 15.04.15 um 05:57 schrieb Doug Schuler:
>>> In case anybody is interested, here's the  abstract that Aldo and I 
>>> submitted to the pattern language conference this summer in Krems, 
>>> Austria.
>>>
>>> We'd love any thoughts you may have.
>>>
>>> — Doug
>>>
>>>
>>>
>>> Domesticating Pattern Languages
>>> Meta-patterns and the Development of Pattern Language Use Methodology
>>>
>>> Douglas Schuler
>>> Aldo de Moor
>>>
>>> In the 2014 PURPLSOC workshop Schuler presented eleven “seeds” as 
>>> challenges that must be addressed if pattern languages are to 
>>> effectively promote social change. Here we focus on two 
>>> seeds: meta-patterns and methodology. Meta pattern languages are 
>>> pattern language whose domain is pattern languages. In theory 
>>> this pattern language could be used to help design workshops and 
>>> games, construct sub-languages using existing and new patterns, and 
>>> manage ongoing pattern-based projects. Existing pattern languages 
>>> capture essential lessons and linkages between those patterns. 
>>> However, those patterns are somewhat idealized and "frozen." To use 
>>> them in actual situations, stakeholders must select the right 
>>> patterns, ensure their suitability and customize them, and develop 
>>> and execute a plan. In practice, such appropriation leading to 
>>> effective use does not happen by itself: stakeholders need 
>>> additional guidance. We identified candidate meta-patterns themes 
>>> that include all aspects of possible pattern use life-cycles, 
>>> including pattern creation, evaluation, planning, problem-mapping, 
>>> project generation, etc. In the context of teaching a college 
>>> program using the Liberating Voices pattern language we gathered 
>>> information at various stages of 10 quarter-long group projects, 
>>> including what types of projects they are proposing, which patterns 
>>> they selected for use in the project, and how they used the 
>>> patterns — and why. We believe that a closer look at how patterns 
>>> are used (in the fullest sense) will help identify meta-patterns and 
>>> how these meta-patterns are used together will help identify 
>>> methodologies and other approaches to more effective pattern use.
>>>
>>> Douglas Schuler
>>> douglas at publicsphereproject.org <mailto:douglas at publicsphereproject.org>
>>> Twitter: @doug_schuler
>>>
>>> ------------------------------------------------------------------------------
>>> Public Sphere Project
>>> http://www.publicsphereproject.org/
>>>
>>> Mailing list ~ Collective Intelligence for the Common Good
>>> http://scn9.scn.org/mailman/listinfo/ci4cg-announce
>>>
>>> Creating the World Citizen Parliament
>>> http://interactions.acm.org/archive/view/may-june-2013/creating-the-world-citizen-parliament
>>> Liberating Voices!  A Pattern Language for Communication Revolution 
>>> (project)
>>> http://www.publicsphereproject.org/patterns/lv 
>>> <http://www.publicsphereproject.org/patterns/>
>>>
>>> Liberating Voices!  A Pattern Language for Communication Revolution 
>>> (book)
>>> http://mitpress.mit.edu/catalog/item/default.asp?ttype=2&tid=11601
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>> _______________________________________________
>>> Pattern-language mailing list
>>> Pattern-language at scn9.scn.org
>>> http://scn9.scn.org/mailman/listinfo/pattern-language
>>
>>
>> -- 
>> HLS SOFTWARE/WIKISERVICEwww.hls-software.at   www.wikiservice.at
>> leitner at hls-software.at                       leitner at wikiservice.at
>> DI Dr. Helmut Leitner   MOBIL: +43 676 7221638  TEL: +43 316 383820
>> Rosenberggürtel 41, A-8010 Graz, Österreich
>
> Douglas Schuler
> douglas at publicsphereproject.org <mailto:douglas at publicsphereproject.org>
> Twitter: @doug_schuler
>
> ------------------------------------------------------------------------------
> Public Sphere Project
> http://www.publicsphereproject.org/
>
> Mailing list ~ Collective Intelligence for the Common Good
> http://scn9.scn.org/mailman/listinfo/ci4cg-announce
>
> Creating the World Citizen Parliament
> http://interactions.acm.org/archive/view/may-june-2013/creating-the-world-citizen-parliament
> Liberating Voices!  A Pattern Language for Communication Revolution 
> (project)
> http://www.publicsphereproject.org/patterns/lv 
> <http://www.publicsphereproject.org/patterns/>
>
> Liberating Voices!  A Pattern Language for Communication Revolution (book)
> http://mitpress.mit.edu/catalog/item/default.asp?ttype=2&tid=11601
>
>
>
>
>
>
>
>


-- 
HLS SOFTWARE/WIKISERVICE www.hls-software.at www.wikiservice.at
leitner at hls-software.at leitner at wikiservice.at
Helmut Leitner MOBIL: +43 676 7221638 TEL: +43 316 383820
St. Peter Hauptstr. 33d, A-8042 Graz, Österreich

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scn.org/pipermail/pattern-language/attachments/20150418/694ba94c/attachment.html>

More information about the Pattern-language mailing list