Monday
Jun082009

"Value cannot be null. Parameter name: encoder" when importing with Toad for MySQL

I've been working on importing some data sets for a project tonight and keep coming across the error message "Value cannot be null.  Parameter name: encoder" when using Toad for MySQL v.4.1.  The files I'm importing are comma separated text files where strings (text) are not enclosed in any quotation characters.  As an aside, I've only rarely seen strings not encapsulated in either single or double quotes but that's another topic.  In the file format window of the import routine there is an entry for string quote character which I set to null (default is ") because this reflects the reality of the files I'm working with.



To make a long story short (yes, I spent far too long figuring out what caused this information message on import), Toad for MySQL appears to want some value in this entry even when nothing is appropriate.  I ended up leaving the default double quote in place and the imports work fine.  Of course, the target table already has all columns typed.

The moral of this story is 'don't leave the string quote character null when importing with Toad for MySQL'

All the best,

Michael
Wednesday
Dec242008

Excel Sparklines

About a month ago I upgraded to Vista Enterprise and Office 2007 Professional on my desktop computer and boy howdy have I been seeing this message frequently!

Grrrrr...another shutdown!

Grrrrr...another shutdown!

This has not been a fully random experience but occurs when modifying data series and axes in graphs.  Some days I don't experience a shutdown at all while other days I will have 5 or more.  Irrespective of how frequently these occur it's extremely frustrating because my primary tool for work is Excel (on a typical day I will be working from 10 or more spreadsheets).  I've become so accustomed to shutdowns and recovery now that the filename append of '(recovered)' is now a regular part of my filesystem.  Another outcome of these errors is that it's greatly improved my skill at stringing together expletives.




On a more serious note, the silver lining in this is that I've finally imaged a Ubuntu machine; my laptop is now set-up with Intrepid (8.10).  I'll have more to say on this later but so far am impressed with the capability of the OS and available software such as Open Office.  In addition to the simplicity of a clean OS is the cost ... FREE.  I've got some learning to do but right now I'm wondering why  I've been paying $$$ to Microsoft.




But lest I forget, this post is titled 'Sparklines' so I'd better get to that.  I've long been a fan of sparklines as well as the larger visual display concepts forwarded by Edward Tufte.  If you don't know what sparklines are here's a description from Dr. Tufte's site:
[S]mall, high-resolution graphics embedded in a context of words, numbers, images. Sparklines are data-intense, design-simple, word-sized graphics. ... Sparklines are wordlike graphics, with an intensity of visual distinctions comparable to words and letters.

Here's an example of a sparkline showing the Dow Jones Industrial Average from 1928 to present:Dow (LN) 10-1928 to 12-2008

Now, thanks to Rob Van Gelder over at dailydoseofexcel you can use a user-defined function (UDF) to generate your own Sparklines in Excel.  Check out Rob's post, In Cell Charting, for the necessary VBA code.  I recommend you store the code in a special custom function workbook and make this available to all your workbooks as an add-in (instructions for creating an add-in of user-defined functions in the Microsoft knowledge base).


Well, that's it.  Enjoy.

MD









Sunday
Nov302008

Blogging, rainmaking, and relationship building

No posts in over a month...now that's a stale blog.  The issue goes back to one of the original questions, to blog or not to blog.  The risks are numerous: frequency of posting, relevance of topics, saying something stupid, blog forum or specific topics may undermine appearance of professionalismm.  Yet, in the face of these risks, David and I chose to adopt the blog in order to promote transparency in our work.  Of course, not blogging in a month isn't very transparent.  OK, I'll work on that.  Actually, I'm going to keep working on a bunch of things.

So then, what's been going on over the past few months?

In short, LOTS, and also, in honesty, not enough, and yet, in fairness perhaps, the right things.

Among other things, the last three months has included the fall semester at the Atkinson Graduate School of Management at Willamette University.  This semester I've had the opportunity to teach my first course, Research for Marketing Decisions.  There's much to say here, much that I've learned.  Let me save this topic for a complete post following finals (two short weeks away).

In addition to teaching, I've been developing a course for the spring semester in the area of databases and data access for managers and research analysts.  Preparing for this course has been a good way to solidify what I've learned over the past decade regarding databases from a user perspective as well as incorporate some of the fantastic tools and work in the area of open source database and data analysis technology from such organizations as the R project, OpenOffice, Palo, and others.  This should also be another complete post, so let me put this aside for the moment.

All of this to finally get to the real reason for this post, the topic of rainmaking and relationship building.  In setting this up, let's consider a simple model of the enterprise with only two components (Yes, Dan, I'm stealing one of your constructs).  The first component is production or whatever is needed to make the product.  The second component is marketing and sales or whatever is needed to position, price, and sell the product.  In thinking about launching a data analytic consultancy (that really is what we're trying to launch here even if it's not been made completely clear), the early focus was on the production component and thinking through the necessary processes, chops, tools to get the job done.  Now we get to reality where we're saying "I know this has value but how do we communicate that value and to whom do we try to communicate that value?"

As I reflect on the consultants I've worked with in the past, I can identify two modes of operation with respect to client relationships.  The first model is where the words are used (partnership model, relationship, win/win, etc.) but the actions are not 100% in line with those words.  This is I call the sales model; we sell our services to you in the same way we might sell a car.  The second model is the true relationship model and is characterized by an earnest effort to seek win/win, understand client needs, and the strength to say 'no' when it's not in the best interest of the client.  While I won't name names with respect to the first model, I will say that John Bird of TNS and Scott Laing and Chris Robson of Parametric Marketing are excellent examples of consultants who follow the true relationship model.

Surely I've already tipped my hand and you know that the true relationship model, one in which win/win is the prize, is the goal of tenpoundsledge.  But relationship doesn't happen overnight and so we've) been thinking about how best to build the relationships we have, we've been developing our ability to listen for needs, and we've been telling a few folks about what we have to offer and getting feedback on that.

That's it.  Relationship.

It's late so I'm going to leave it here for now and pick it up tomorrow or maybe later today which would be even later but then, oh well.

MD
Friday
Oct172008

Business haiku

Data cleaned

MLR and correlation my friends

A season of knowledge
Tuesday
Oct142008

Pick

Tradeoffs are inevitable...

You can have it:




  • Cheap




  • Fast




  • Accurate




Pick two


Pasted on Van Bain's monitor back in the day when he was a 'software janitor' and my job title was one of those strange combinations used when they don't want to use the word secretary.