Re: info search
Subject: Re: info search
From: Scott Mintz <scott.w.mintz@gmail.com>
Date: 6/23/10, 01:48
To: Clark Robinson <robinsonchicago@gmail.com>
CC: Barrett Brown <barriticus@gmail.com>

I was also really fascinated by the piece (thank you for sharing), although I think his analogy isn't perfectly fitted (pun intended).

Specifically, I'm going to go out on a limb and disagree with his assertion that it would be impossible for someone to put together a billion piece puzzle. I understand it's tiny in comparison but here's a picture of a 13,000 plus puzzle that someone did: http://www.flickr.com/photos/27203838@N02/2540049306/in/set-72157605357244711/. I imagine the art, whatever that may be, to completing a puzzle doesn't change with size. So why should a smaller sized puzzle be possible, but a larger one "virtually" impossible.

I agree with Clark that "tagging" is important. It helps to reduce redundancy as it relates to gathering. Information can be tagged by including meta data as it is done on the WorkshopPM blog, or it could be blanketed around information that is submitted to certain categories. For example, on the investing website I belong to I can make posts with the subject Apple, or more broadly I can choose Technology which includes Apple amongst others. Certain readers may have Apple on ignore while others may request that all posts made to technology get sent to their email. Another approach is to allow for a modification/addition/deletion of a tag. This is one of the many things I wanted to email about with regards to ProjectPM. Specifically, if information is going to be tagged, do we want the ability to have people to modify the tag to make it more appropriate? I believe such mechanisms if properly created are essential to make the network more fluid and self-sustainable, and I do not think they need to be inundating on the bloggers. i.e. we can have one or more drop-down menu with pre-populated tags, and include some means to create new tags. Don't a lot of bloggers already do similar things to make their posts more easily searched?

I know of no great way to explain my thinking on information gathering/processing as it relates to the post, so I too will use the puzzle analogy in order to give my two cents:

1. We all see the final product and know exactly how the image should look (not as easy in real life).
2. We separate the pieces into buckets (i.e. end pieces in one bucket and pieces that look like part of a car in another bucket)
3. We work together in an iterative process to match end pieces in order to form the border, this often involves educated matching along with trial and error since the pieces offer limited information other than that they are end pieces (or maybe we are too lazy to observe other information that is provided, which would enhance our efficiencies, since we believe with enough trial and error we can form the border in little time due to the fact that the number the combination are relatively small).
4. We work individually or in smaller-sized teams to tackle individual buckets.
5. As the information within our bucket becomes more observable, we're concurrently looking at other buckets, taking it in, and finding points of intersections where we can attach two or more buckets.
6. A feedback mechanism is involved where as our knowledge of our bucket and other bucket grows, we can more easily determine if certain pieces are better matched to another bucket and we move it there accordingly.



On Sat, Jun 19, 2010 at 5:45 PM, Clark Robinson <robinsonchicago@gmail.com> wrote:
Actually these information ethic statements I made do not describe anything new or of much use; journalists have a well established and complex information ethic, so do lawyers, academic researchers, scientists, bookkeepers, cops, and practically everybody I can think of except for politicians seeking election and religious nuts trying to reconcile facts with irrational beliefs.  And really you can't even construct a database without an information ethic. So never mind.

I'll think about the issues in the other e-mail, forwarding velvetelvis's message, and give you whatever thoughts I have later or tomorrow.


 


On Sat, Jun 19, 2010 at 3:37 PM, Barrett Brown <barriticus@gmail.com> wrote:
Off the top of my head - and from the bottom of your post - I think we've decided that the disciplinary system Scott proposed is unnecessary in light of the emergent self-policing that will exist by way of information deemed poor not getting very far into the system.

As for this procedure you describe, I worry that we'd have trouble attracting many bloggers to our network if this procedure is anything other than automatically implemented by software. Perhaps there is a way it could be done in that fashion? Otherwise, this ethic would certainly be of value.


On Thu, Jun 17, 2010 at 8:52 AM, Clark Robinson <robinsonchicago@gmail.com> wrote:
Re: Boudreaux article:

"For these pieces to be fitted together so that the result is valuable requires many people receiving feedback at each step of the way . . ."

In his next column he'll discuss the feedback--of course he's talking in general, not specifically about electronic systems of feedback exchange, but one can not help but think about how electronic connections improve information activities.


Based on my own experience here are some conclusions about large scale information handling by multiple users of information in an electronic network:


There needs to be (and often there is) an information ethic, which every handler of the information subscribes to, and  a system which encourages, facilitates and to the extent reasonably possible enforces this ethic.

The information ethic to be applied by each user includes:
--respect for the integrity and pedigree of the information (preserve evidence of its original source, date of creation, context of creation (such as who paid for it), and not altering or distorting the information)
--concern for the next user of the information, to make his work more efficient, which may mean recording additional meta data (information about the information)
--concern for all future users of the information, making it easier for this information to be found and understood literally

The system should support the exercise of the information ethic by
--making it easy for each information handler to add value--by prompting users to attach critical meta data
--keeping track of who is interested in the information
--signaling interested users of the arrival of new information that has meta data indicating interest
--signaling users when additions to the meta data indicate they may be interested
--users should be easily able to manually share and direct the information, but the system should also be directing information independently (like Scott's bypass sampling). 


I think a proponent of any information to be entered in a system has a duty to also enter complete pedigree data, considering not only his own use of the information but the interests of others who may ultimately access the information. The value of this to subsequent users and ultimately to the institutions served by the system justifies the additional labor.

When a user accesses some information there should be a display of the associated data for his benefit. It should be easy to attach more information, particularly in the form of data structured so the system can appreciate it.  I am more interested in users filling in data fields than in attaching prose commentary to the information (though both are valuable), because the system can not use the latter so readily for the benefit of other potential users.

Ironically, some things are worse in the electronic information world: in paper systems each document has its own identity, but electronic information, absent a well-conceived meta data structure, can lose its identity.  This can occur when information dumping is permitted, which happens because it is so easy and the information ethic is ignored.

What I am writing here is consistent with what Scott was describing in his Thoughts to be Pondered post and I guess what I didn't agree with there was not the proposed feedback structure but rather the application of a disciplinary system to the practice of writers' art.







On Wed, Jun 16, 2010 at 12:34 PM, Clark Robinson <robinsonchicago@gmail.com> wrote:
Interesting, I have been thinking about his article as applied to my own experiences in trying to find ways to make the most. meaningful use of vast quantities of information (in government), a lot of it stored like the puzzle pieces in the field, without appropriate meta data, requiring every user of the information to re-discover for himself some of the same things each prior user discovered, sort of similar to his last paragraph.  I am about to drive to Chicago so that is something to occupy me for the next few hours besides passing 18 wheelers.


On Wed, Jun 16, 2010 at 11:18 AM, Barrett Brown <barriticus@gmail.com> wrote:
http://www.pittsburghlive.com/x/pittsburghtrib/opinion/columnists/boudreaux/s_686130.html

--
Regards,

Barrett Brown
Brooklyn, NY
512-560-2302





--
Regards,

Barrett Brown
Brooklyn, NY
512-560-2302