Patterns as Google Docs in a gDrive

[Theo Armour]

Hi Everybody

 

Please checkout:

https://docs.google.com/open?id=0B0g809SWki8tU3pQRklGQkFWdzA

 

All 253 patterns all tidy in their folders.

 

Anybody can edit, attach comments whatever.

 

There's a full revision history, translation capability and more.

 

Perhaps the most important thing is that there is a full API. So, for example, HQ could be notified any time anybody made an edit.

 

Do note that this is just a preliminary effort. The internal links don't work and the summaries are screwed up. Also I have not yet pushed the code up to GitHub.

 

Why do this?

 

To prove that you can take stuff from anywhere, process it to a Fair Use standard and then blast it out to wherever/however the user wants it - all with no human manipulation of the data. And with no need for a server or any significant infrastructure.

 

Theo

 

 

 

 

 

 

[Ward Cunningham]

On May 7, 2012, at 11:38 PM, Theo Armour wrote:

Why do this?

 

To prove that you can take stuff from anywhere, process it to a Fair Use standard and then blast it out to wherever/however the user wants it - all with no human manipulation of the data. And with no need for a server or any significant infrastructure.

Curiously clever.

The first few summaries I looked at looked surprisingly good. However, I chose patterns I was familiar with and could fill in the missing parts. Then I looked at some less familiar patterns and couldn't get a clue as to what they had to offer from the summaries.

This probably says more about how the human mind works than anything deep in the summarizer algorithm.

Perhaps this is best used as a threat: let people use your aging work as intended or we will heap this abuse on it.

[Theo Armour]

Hi Ward:

 

>> Why do this?

>> To prove that you can take stuff from anywhere, process it to a Fair Use standard

>> and then blast it out to wherever/however the user wants it

>> - all with no human manipulation of the data.

>> And with no need for a server or any significant infrastructure.

 

So that was thought #1.

 

Which is about getting and updating *selected* IP from all over the place. [from CORS servers better but not always necessary  - but love that JSON.]

 

Note the use of *selected*

 

There's a lot of stuff out there. So now we need to start talking about apps cherry picking the good stuff.

 

What you want is for the app to bring to you the shoulders of giants to stand on so you can see even further.

 

Hold that thought for a minute.

 

**

 

Every day I look at web sites like BoingBoing, Slashdot and Hacker News. Why? Because they do and excellent job for me. They select, filter and curate new stuff that interests me. I don't have to look through hundreds if web sites or even use an RSS feed.

 

But where are the curators for old stuff?

 

Wikipedia has tons of stuff but no way to tell me which is the good stuff.

 

If I search Google on, say, taxonomy I will get many results – but again – no curation, no idea which are the good ones.

 

OK, so go to eHow or About.com – and learn about how to make things that really suck.

 

**

 

So I want to learn about windows. I want my brain to be exposed to everything there is to know about windows.

 

And I've got, let's see, eighteen minutes for this.

 

I want to know building codes, the etymology, insulation variable, relevant Alexander patterns and whatever.

 

Every time I use Google, I get taken to a web site that's run by a guys that say they know everything about windows.

 

Except they give no sources, the don't consider peer reviews, they provide no evidence.

 

I keep getting  the shoulders of midgets.

 

**

 

Guess what you can just now start to get for free on the web?

 

The building codes for all fifty states.

 

Up to now these codes were only on expensive paper.

 

Within a few years every code, law and statute will be available. No Lexus/Nexus needed.

 

Ditto every scholarly article formerly locked up in JSTOR.

 

Amazon and Google are fighting it out to get you books or samples of books for free.

 

Very cool.

 

But how do I know which one are the giants? and I want just the giants that do windows.

 

**

 

It turns out the giants are hybrids actually. Part computer. Part human.

 

The first pass is probably human.

 

"Hey, the work of that looks interesting"

 

Computer gets it, parses it, makes it available.

 

People read some sections more than others,

 

Computer reorganizes.

 

Computer makes silly mistake choosing key words.

 

People choose better keywords.

 

People are asked to vote.

 

Nice people are asked to become moderators.

 

So the system improves itself simply by recognizing the patterns being generated by the users.

 

**

 

Thought #2

 

Why do this?

 

To prove that we can build an app to help people see further by delivering the shoulders of giants to people by starting off with an app that's built by standing on the shoulders of the people in this group.

[Ward Cunningham]

On May 8, 2012, at 4:00 PM, Theo Armour wrote:

There's a lot of stuff out there. So now we need to start talking about apps cherry picking the good stuff.

Around the same time I was studying APL and related books I came across an amazing paper by Michael Reddy called the Conduit Metaphor. Wikipedia summarizes it in an article that might be longer than the original paper:

http://en.wikipedia.org/wiki/Conduit_metaphor

Object-oriented programming was new at the time and I was studying linguistics and cognitive science thinking that I might put ideas into objects much like a talented author puts ideas into words. Reddy reminded me that words are not vessels that hold ideas. The metaphor we use to talk about language tricks us into thinking that they are. At best words tickle thoughts that might lead to somewhere useful, maybe not. 

I came to especially appreciate Timeless Way's distinction between the Gate and the Way.

(As I write this post I become interested in the community around the above article. I examine the talk page and discover there is no community at all. Hah, I new something was wrong. The article was still-born.)

This leads me to consider the distinction between a community of purpose and the product of their work. Do their works retain any "life" when the community is gone? Perhaps the only life is in the mind of the viewer in the museum of old works.

Best regards. -- Ward

[Michael Mehaffy]

Ward,

Fascinating on a number of levels!  Chris has always had a linguistic quality to his work, and a structuralist quality (what I like to call "symmetric structuralism," the tickling you refer to, or the metaphoric quality (involving overlap,  ambiguity, poetic relational qualities).  But these are still structural.  You guys in the programming world have always kept the structural dimension, but opened to this poetic quality too, which is what makes it so instructive to us, I think. 

By the way, I got to hear a fantastic talk from Brian Goodwin at our ESRG conference in 2006, noting that the genetic code seems to work this way too -- like a language, full of metaphoric relationships that included ambiguity and overlap.  The structural power comes from the reinforcing linkages...  See excerpt from my notes below.  (More at http://www.esrg.blogspot.com/)

Cheers, m

...Brian Goodwin then gave his observations as a biologist and a complexity scientist. He sees a convergence between many fields, and among the perspectives represented within the group. He noted that biology has recently gone through a crisis: the sequencing of the genome, which held out great promise, has collapsed. The code by itself doesn’t really explain what is going on. The analytical reductive methods are not showing us the full story, because information only makes sense in a context. The real question seems to be, how does an organism make sense of DNA? His hypothesis is that “meaning” is a phenomenon deeply embedded in biology, not just a cultural phenomenon. What we learned was that there are only 28,000 genes to generate the vast complexity of an organism – this produced a crisis in the field. It seems that so-called “junk DNA” actually has a very important role to play in coordination of the process, and in producing the whole. As Bill Hillier said (and as Alexander’s work emphasizes too) we are recognizing that the whole co-exists with the parts in an important way.

He sees elements now emerging, that point the way to identifying the processes of creating organic wholes in the built environment. What is now emerging in biology is that networks of molecules are organizing the wholes and making meaning of it. This is an ongoing and irreducible process – there is no “definitive story”, and there never will be – it’s always more mysterious and complex than we thought. But there are similar principles operating in different contexts. For example, self-organizing networks in cells have a particular property of self-similarity or fractal or power laws which describe their structure – the same structural property can be observed in the World Wide Web (and in cities, as Hillier, Batty et al. have shown). Transcription factors made by genes and targeting other genes, causing differentiating, also follow a power law.

The same structure can be observed in language – it too obeys a power law, between the most common word, the next most common, and so on. Is this a statistical artifact, or a deeper issue? Nucleotides in DNA obey a power law also, and forms a scale-free network. This is now a target of investigation. The implication of this is that organisms are using something very much like a language in order to make sense of the DNA.

This takes us beyond complexity as a pure structural phenomenon, and into the realm of language, in which every sentence is ambiguous, and can have multiple meanings. Yet we understand one another, because we hold open these multiple meanings, and let the process flow. The genetic process too seems to work like this.

Thus ambiguity is an essential part of the story. It seems ambiguity and creativity go together. Machine languages are completely deterministic – extremely good for deterministic computation. But you need the creative, context-sensitive, adaptive aspects that depend upon ambiguity. Too much order is a sign of danger – in fibrillations, in arrhythmias, other fatal conditions. Ambiguity is also a fundamental part of quantum mechanics theory – built into Schrodinger’s wave equation, which includes superposition of possibilities. The system holds open multiple possibilities until there is a kind of convergence on an appropriate solution.

This is how coherence seems to emerge in the organism. This is not an easy concept to grasp. Where is this ambiguity? The networks are not in any one particular state – they are dynamically moving. Different combinations of genes can give rise to the same phenotype – the organism holds them superposed because one may be more important than another.

There is a relationship between ambiguity and self-reference, and the way self-referential networks function. A system that refers to itself can be ambiguous (“I am lying to you now”). But there is a principle of least effort coupling the effort of the speaker and the effort of hearer. (cf. Ricard Solé.) A child makes a noise about something distressing, but does not give enough information. Speakers learn to provide just enough information that with least effort they can understand and make choices.

In a sense all organisms use language. There is a deeper structure at work.

So the relationship to planning? We want tools to produce a coherent organic order – a kind of conversation with the stakeholders… and with the natural world.

There is a historic process under way in the sciences, going back several centuries, with implications for planners (as Jacobs noted 4 decades ago). Kant proposed that science exclude quality. This was a momentous change. Goethe proposed that we let quality back in. Quality has been regarded as subjective and not reliable – therefore it can’t be used. But quality is returning to science, and to environmental management. In the UK the Environment Agency is very interested in engaging local stakeholders to evaluate a river – it turns out to be a very useful and reliable way of getting broad knowledge about environmental health.

When engaging the rhetoric of sustainability we must remember that nature is energy-efficient – it eliminates toxins, its forms are functional and beautiful. These are all “natural” products. Beauty is fundamental to health and coherence, not a mere psychological attribute. Functionality is dominant in culture, defined in quantitative terms, and we need to re-balance the quantitative with the qualitative.

Bill Hillier commented on Brian’s remarks. “It’s not often you hear something that changes the direction of your thought – this was beautiful and changes the direction of my thought.”

--
Michael Mehaffy

Structura Naturalis Inc.

742 SW Vista Ave., #42

Portland, OR 97205

(503) 250-4449
www.tectics.com

[Eliezer Israel]

An issue with the auto-summary:

Sometimes, it pulls pieces out of the original text wholesale.

Pattern #1, for example - most of the text is straight out of the original.