A tongue-in-cheek look at Data Services.

It's a rare occurrence these days, but if one has been around an organization for a long enough amount of time, one may acquire a certain "long-view" of that organization. One becomes sort of an unofficial oral-historian, a keeper of the traditions or simply one who- like Roger Miller's famed "King of the Road"- knows "every engineer on every train, all of the children and all of their names... and ev'ry handout in ev'ry town, and every door that ain't locked when no one's around". In my organization, I fill this role. I'm also the department's Scrounger. If you've ever seen "The Great Escape", James Garner's character is my model. The interesting thing about our department is that although I've been there for ~nine years, I'm not closest to being the "old-timer".

That's me.

Still in addition to that, I have a few odd things that distinguish me from everyone else in the department.

First, I'm the only person in our department who was rehired after a RIF (reduction in force- and to be fair, our facility has only has one RIF). Next, I'm the only one who was hired without a background in our industry or experience in the job. Finally, I'm the only person (quite possibly in the entire plant) who has worked almost every possible shift: 1st (~0800-1700), 2nd (1600-0000) 3rd (2000-0500) and 3 day/4 day (nights 1900-0700 three days per week with alternating Sundays).  In other words, I've been around. I'd like to introduce you to some of my coworkers- past and present.

Now, before I begin, I have to say that not everyone has a nickname, and not all nicknames will be universally recognized, even internally. Also, not all of the names are real... some have been changed to protect the innocent- and the guilty! And, to completely protect everyone, there will also be a few dummy coworkers and nicknames.

For starters, there are Papa and Auntie. This pair is not related in any way, except possibly for their nicknames. Then there would be The Princess, E-Man and J.R.- their connection is their school. And then, there's Banana, The Colonel, The Dancer and the Man with Three First Names, along with The Golden Child, the Gnome and the Fresh Prince of Data Services. We also have The Lurker, the Latin Lover, Cube, Z-Man, Chaz-Baby, My Bookie, Buttercup and Nameless.

And, The Mole.There are several others, but decorum and space prohibit them being discussed.

In conclusion, I apologize for saying next to nothing about anyone, but please consider this whirlwind listing something of a dramatis personae of  my department. I think I have a bit more license than Nameless gives me, but just to be safe, I'm respecting their wishes. Keep this handy, though: in future blogs, you might want to refer to it!

I'm betting YOUR organization also has a similar group of individuals- friends, coworkers or associates- who make or break your day. Mine have been more blessing than challenge. I'd love to hear about yours.

As always, I am hochspeyer, blogging data analysis and management so you don't have to.

Dude, you're getting a Dell, or, Resistance is futile.

We don't have many Dell computers in our home. Longtime readers may nod in agreement at this statement, but for the benefit of the uninitiated, we have many computers in our home. And while not all are in service, all ARE serviceable; that is, with just a bit of TLC, I believe we could have no fewer than thirteen PCs online simultaneously.

Granted, I would NOT do this for a number of reasons.

Firstly, most of these boxes were "designed for Windows XP". Now, I have nothing against XP- I have one machine that is an XP machine- and it's physically OFFLINE.

Next, most do not have green power supplies. Although all of the monitors in the SUL are now energy-efficient flat panels, the power supplies aren't really worth upgrading, as most of the PCs that are not in service are pretty much waiting to be cannibalized.

Finally, it's just not practical. Four individuals running 13 PCs in a SOHO environment?

The reason Dell came up is that a certain Dell PC crashed one of my external HDDs recently. I know for certain that it's a particular Dell because every time I plug the external drive into this Dell, it asks if I want to repair the drive. Mind you, this drive normally is hooked up to another Dell or a Lenovo or an HP or Compaq machine with no problems. This particular Dell has issues, though, and the last time I use the drive, I ignored the repair message, and in turn it made the drive unusable. I don't know why or how, but this machine is now persona non grata.

End of story.

Lego data news-

I'd been using Peeron's database for some time, but recently it came to my attention that BrickLink has a more complete database. Not only does it seem more complete, but the part numbers are properly formatted, they are already in a separate column and the descriptions do not succinct and do not seem to require editing. The data will still require a bit of formatting to fit my particular usage. That's the good news.

The bad news is that unlike Peeron's single page text file format, BrickLink displays its data on webpages. Not bad, as its still copy-and-paste. What's really from my perspective is that each page displays fifty part numbers.

There are 837 pages... a relatively small price to pay for superior data.

As always, I am hochspeyer, blogging data analysis and management so you don't have to.

Men Without Hats, and Data.

"In other data news, I've got what appears to be a workable solution for my internal Lego part number. It's fairly lengthy at seventeen characters, and from all appearances, this should be sufficient. I've begun the data entry on this, and tried a few trial sorts. So far, everything looks good, and this is officially stage 2 of the Peeron normalization."

As noted over a year ago in this space, No plan of battle survives first contact with the enemy, the best plans often don't get past the first round or two of testing. It's unfortunately true, and I'm referring here to the Lego portion of Forty-Two. I'm sitting on a challenge right now that's more conundrum than impasse. 

With only sixty-five entries, I ran into one of the proverbial "straw that broke the camel's back" records. My format had been "ANNNNNNNN.XXXXXXX", where A=alpha character, N=numeric character, the "."is a placeholder, and X=alpha or numeric character. It's large and unwieldy, but that doesn't matter, as I only need it to force order onto the Peeron data. So, I need to add a few more characters to the right of the "." placeholder. Still, I hate to redo stuff I've already done. So now, it appears that I'm up to nineteen bytes for the primary key. Hopefully, it doesn't grow more than that.

Before I forget, the Men Without Hats reference is, of course, to The Safety Dance, which I was humming when it looked like nineteen would, indeed, be a safe number. I just skimmed the entire list, and I'm fairly certain I can stick with nineteen.

As far as the rest of the database is concerned, I've been pounding away at my least favorite activity- data entry. It's still a really small database, and not even relational as yet. Three tables, containing in total 753 records.

I think that's a wrap. The internal issue I always have when doing these types of updates is keeping this more blog than change log. Oh well, speeds and feeds are gonna be speeds and feeds....

Lastly, a big shout out to my friends in Russia and Portugal- I hope you enjoy reading this as much as I enjoy writing it. 

As always, I am hochspeyer, blogging data analysis and management so you don't have to.

Richard Strauss, and Data.

This past week, I've managed to get in to work fairly early almost every day so far (Monday- Thursday); I think Wednesday was the only day I came in later, and that was intentional. I've also been getting up earlier, and the time before work has been used to work on the database. I've made some good progress on the few boxed sets of audio CDs that we own, which is where Richard Strauss comes into the picture.

Before I go any further, I should give you a bit of my musical backstory, kind reader. You may have surmised from previous posts (especially the most recent one, Axl Rose, and Data,) that I'm more Rocker than Opera-Goer. This is 100% correct. However, my musical tastes are fairly diverse... Yo-Yo Ma, The Beatles, Chevelle, C.W. McCall, Air Supply, Deep Purple, Newsboys, Sibelius, 80's hair bands, choral, etc. I firmly believe that any music that is good should be played loud when possible.  I like some blues, classic Motown, and a smattering of jazz. I'm okay with the various forms of trance, and dance music- if it's something that catches my ear. I can even deal with disco these days. Things I pretty much have no interest in are rap, hip hop, opera and death metal.

So, what's with Strauss?

Well, the album pictured above is a boxed set I picked up at a library sale for a solitary U.S. dollar. Cheap-value-SCORE! It's a three disc set, with a booklet nearly as thick as the CD case. The 330 page "booklet" has all sorts of details, not merely about this opera, but about this particular recording, its cast and conductor. It also has the complete text and lyrics of the opera in French, English and German. When I purchased it, I had decided that the weight of the boxed set alone made it a good buy... little did I know that this was a "reference" recording, one by which all others are to be judged. Its also conducted by Herbert von Karajan, a legendary conductor.

Still, what's with Strauss?

Data entry, pure and simple. As this past week has seen a renaissance of Forty-Two, I decided to tackled boxed CD sets. The problem is that this particular set has sixty-two tracks, all of which have German titles. I'm slow enough at data entry without having to import special characters, so I did what any reasonable human being would do: I looked up the recording on Amazon, copied the track list and pasted it into Excel. From there, I copied and pasted each track into Access. That's where I stopped with music- I still have three boxed sets to go, and then it's on to albums.

In other data news, I've got what appears to be a workable solution for my internal Lego part number. It's fairly lengthy at seventeen characters, and from all appearances, this should be sufficient. I've begun the data entry on this, and tried a few trial sorts. So far, everything looks good, and this is officially stage 2 of the Peeron normalization. I still need to add dimensions and clean up the text descriptions before importing it into a table.

As always, I am hochspeyer, blogging data analysis and management so you don't have to.

Axl Rose, and Data.

It's been busy at work lately. And the rain has been frequent. And... it's Springtime in my little corner of the world. Consequently, our lawn got a "little" out of control.

(Spoiler alert: this post contains references to Guns 'N' Roses and their songs... 

and a few other musical references!)

Jennifer was kind enough to help out and mow the front lawn, which is what everyone passing our home sees. The back... well, that's another story- and my bailiwick.

My time had arrived. It had not rained for over a day. I looked out the back window and saw the grass blowing like Dust in the Wind. As my gaze landed upon the compost bins, I had to rub my eyes because I thought I saw a black tophat sitting on top of one of the compost bins, and a bandana on top of the other. I shook my head, blinked and looked again. Without warning, the yard had gone from broad daylight to a starless night. The yard's verdancy had turned to a monochrome with rough-cut video quality- I kid you not!. Our neighbor's white fence had been replaced by a gaggle of Marshall stacks, spotlights were illuminating the compost, and Slash, with trademark shades, ciggie and Les Paul, and partially obscured by the output of several smoke machines was furiously spewing out riff after riff, while Axl crooned, "You're in the jungle, baby" as he seemed to float over the grass. I covered my eyes, and shook my head. When I looked again, the moment was over. My yard was green, and back.

Seven Marshall stacks- fourteen cabinets and seven heads 

 Sheesh! Less Monster, more cowbell... maybe.

So, I cut the grass. Yes, it was long- long overdue for cutting. I'm not certain how long it took to complete the task, but I had to make several trips to the compost to empty the mower's bag. When that was done, I decided to tackle another yard task: the ivy. And, to be fair to the myriad of horticulturalists, botanists and assorted green thumbs in my audience, I'm not really certain what this plant is. It has dark green leaves, a woody stalk, and will sprout roots along the stalk. It's also quite capable of climbing. So, we've decide that it must go.

I got most of it off of the red mulberry. Now, technically the mulberry should be a bush, as it has multiple trunks. This one, though, is ~30' (over 9 meters) tall, so I call it a tree. Many websites also call it a tree.

Our Mulberry last Winter

I put the mower back in the garage, and grabbed a pair of leather gloves, a few yard waste bags and some clippers, and proceeded to tear into the ivy. For reference, the bags are constructed of a double walled heavy brown paper, and are approximately 16" x 12" x 35" (40.6cm x 30.5cm x 88.9cm). I filled five of them, and I'm probably only about one-third of the way done. This was sometime last week, and since then, we've had rain at least four times, so the next time I tackle this project, the stubble from the first round should be visible, which should make the next phase a bit easier.

Finally- data! (Sorry, no data pictures!)

On Saturday morning (May 14th), I FINALLY completed the 1st phase of the Lego Peeron data normalization! Speeds and feeds are appropriate here, so here we go!

The original Peeron database I'm using is from March 19, 2012. It contains 18,510 parts (rows). After the first passes through the data- in which the data started out as a text file, and then was converted to an .xlxs file, and then the data had its initial cleansing where a "base" part number was created- the data ended up totaling 16,218 parts. It should be noted that this not a fixed number, as there are more parts that need to go to the "Stickers etc" worksheet. I also want to emphasize that I am cleaning data and not normalizing; after all, all of this data still resides in an Excel file, so it is still a flat file. My next task is to impose order onto this file, so that will require the creation of a unique part number. This part number will be the primary key once it is imported into Access.  

Before any of this happens, I need to come up with a standardized format for my unique part number- currently called the "DB_Tracking_Number". Ugly, but good enough for now. I can't really automate this, once again because of the lack of standardization in the Peeron data, and my attempt to impose my own personal spin on Lego organization.

So, for now, I have ~ 16,000+ part numbers to create.

As always, I am hochspeyer, blogging data analysis and management so you don't have to.