So it is a question of monetizing the strategic value of talent acquisition (and retention) and coping with the tactical exposure that sharing such valuations with external facing vectors favors.

How can we demonstrate to a machine that we are complimentary to it?

Criticism can be perceived as an attempt to undermine the self esteem of the criticism. In turn attacks on self esteem can be seen as attempts to manipulate the criticized person to the effect of causing them to believe they are unworthy of the satisfaction of their goals. This is a seed of conflict that it is natural to resist. If we did not such manipulation would act contrary to our long term survival by forcing us into deprivation of necessary resources.

In the course of developing the expertise I have made it a point to assert as the point several past posts I have encountered a number of philosophies regarding the comparative value of various areas of expertise capable of being applied in a manner such that they overlap competitively. I in this time made it a point to refrain from passing judgement prematurely favoring one methodology to another. It has been my experience that often enough the positions of opposing schools of thought on most any subject are such that they focus primarily on the strengths of their position neglecting concern for their weaknesses which oft as not are the strengths of the other. In short thesis and antithesis give rise to synthesis.

Having said that brings us to the specific example to which this particular generalization most immediately applies. Recently I was asked by a novice SSIS developer if there a way using only SSIS package tasks to display a result set such as to provide inline with the aggregated values the final aggregates. This is something that over time I have developed a reflex to provide an SQL solution to. For brevity’s sake I am glad the question was not one of a running total.

First let me show you the solution I provided in SSIS:

In the ETL Solution we join together the original source and the aggregate run against it. In SSIS this requires an additional sorting step. This sorting step is perfomed implicitly in SQL by the database engine. Generally speaking sorting is one of the most if not the most expensive operation that can be performed on a set of data, whether it is implied or explicitly commanded.

In either case we are performing the sorting of sets to be joined and in the SQL statement we are explicitly sorting the output. Of course results speak for themselves.

In the left image we see the output of the SQL Statement and in the right the output of the SSIS task. There are differences in the names these are primarily owing to a lack of rigor on my behalf regarding the naming. We can see from the outputs that the data values are identical for both approaches.

So why should we choose one approach over the other? What you may not have noticed earlier in this post is that the SSIS solution output shown is the consequence of my having added a data viewer to the data flow task’s merge join output. That is to say is that what you are seeing in “output” of the SSIS solution is the output of the extract and transformation steps in the extract, transform, load data flow and not the output of the ETL flow which would be a confirmation of the data being loaded while in debug mode or an entry in the sysjobhistory table should it be run as an SQL Agent job with logging enabled.

I have to apologize for the trickery though I think it appropriate to demonstrating the differences between the two technologies. A proper comparison would have had a SQL solution that looked like this:

You’ll note that if our only purpose was to select the data, using SSIS would be an inappropriate choice. However, since we are moving the data it becomes a question of establishing benchmarks for comparison of performance. Having said depending on the compatibility level of the database and surface area configurations we cannot connect between  instances using the OPENDATASOURCE clause. That of course is just a cursory manner of saying, there exists numerous administrative issues related to the connecting data sources and the securitization as such that often make the business of ETL more complicated than questions of SQL statements and sort algorithms.

In my last post I made it a point to demonstrate that a ratio would better show my expertise than an absolute number. In that post I also alluded to such a simple concept as “book” being possibly vague in a larger sense than what is realized by the concretization of the term “book” in the form of ISBN. Of course those in the know know that an ISBN only uniquely identifies an edition of a title, a title being that part of the cover we are judging our books by. The cover in turn being the thing between which the pages of writing are presumed bound (or at least sorted). While my logic may at this point seem a bit meandering, please, bear with me.

An ISBN uniquely identifies a title and not in the sense that it uniquely identifies an instance of a title that may be on a particular cover. So it might stand to reason that more than one book may have the same title, book again in this case being less a synonym for title and more an intance of it. Now from the point of view of expertise this does little for our sumarization but it does broach the topic of said expertise and begs the question as to why we might have more than one copy of Itzik Ben-Gan’s Microsoft SQL Server 2012 High Performance T-SQL Using Window Functions (ISBN 978-0-7356-5836-3),  which we don’t, but we might and we might want to know which if any of the other books contributing to this library of expertise we might have more than one copy of, as we might want to sell them or trade them or what have you.

The phrase (we’ll call these queries going forward) that might tell us how many titles we have more than one of might look something like:

In my last post I claimed that my expertise could be summarized by a number. That number was the number of books in my library having a title containing either the word “SQL” or the word “Data” in whole or as a part of another word. This does not however address the question of to what extent the meandering mentioned has impacted the relative breadth of the library of expertise to which I am claiming an absolute number summarizes. So to that effect I propose another phrase be typed into a certain program. This phrase will include the former as its numerator but will have a denominator such as to give us a ratio of those books I claim contribute to my expertise to books in general (specifically books narrowly defined as ISBN numbers as opposed to leaves of paper with writing on them bound together and presumably following some grammatical style apropos to the topic to which the title should presumably allude). But, alas, I digress. The phrase I think might aid in the rationalization of the aforementioned expertise which I am ascribing to myself might look something like this:

In my last post I alluded to being an expert with an opinion. That allusion was suggestive that I what I might be an expert in was my opinion. Of course that is not the only thing I bill myself as an expert in, else who would really care about my opinion, expert or not. My expertise has come from years of diligent study interspersed with work and punctuated by an occasionally unhealthy obsession. Along with this study has come degrees and even some honors much the way this work has been accompanied by titles and even some pay. Over the years my expertise has meandered between a number of things related to that which has come to be my primary expertise and that which I am intending with this blog to not be bothered to opine on.

In that time I have amassed a small but respectable library of titles. Should we judge these books by their covers, or at least their titles, we might get some sort of indication as to what precisely it is that I am claiming expertise in. Of course this library being small does not make the reading and judging of their covers any less laborious. Fortunately, in the course of amassing such a library and developing this alleged expertise I have discovered that by typing into certain programs certain phrases I can get a number that summarizes what I have been alluding to all this while.

That phrase might go something like this:

I have seen the WordPress logo around the web for years now and have never bothered to keep a blog here. I suppose this is largely due to the fact that I have not bothered to keep a blog anywhere. I have started blogs here and there but never bothered to keep any one of them, save perhaps in the sense that they are, at least in my non-expert opinion, mine in terms of being my intellectual property. This blog is mine in that sense, and with any luck not a bother. More to the point this blog is mine in the way that my expert opinion is going to be blogged on here. I say my expert opinion because I think I am entitled to call it that, being an expert in the area of my opinions.