>Software is probably too soft to be engineered. Loved this!
Forwarding to my elder son, a Computer&Electronics Engineer by definition and a software engineer by job. Bulent --------------------------------------------------------------------- http://patoloji.gen.tr http://celasun.wordpress.com/ http://www.flickr.com/photos/bc_the_path/ http://photo.net/photodb/user?user_id=2226822 http://www.pentaxphotogallery.com/artists/bulentcelasun 2016-02-11 20:28 GMT+02:00 Boris Liberman <bori...@gmail.com>: > This is very fascinating. > > Indeed, you sounded as if you were compelled to do some optimization no > matter what. As well, partial optimization may be considered as no > optimization at all... Then whatever the engineer in my story did wasn't > optimization by definition :-). > > I don't have my own definition of engineer. I am yet to work with proper > software engineer... Since about two years ago I don't consider myself to be > a software engineer. In fact, people who do software, when confronted with > the idea of rigorous engineering (such as practiced by other technical > disciplines), usually become very non-willing to communicate any further or > erupt in some kind of an argument. > > Software is probably too soft to be engineered. /deep sigh/ > > Boris > > > On 2/6/2016 21:04, Larry Colen wrote: >> >> >> >> P.J. Alling wrote: >>> >>> Engineer joke. An Engineer is a person who will spend two months to >>> figure out how to do a 5 minute task he must preform every other week, >>> in two minutes. (It's only funny to people who actually do the math). >> >> >> My definition of a natural born engineer is someone who will spend three >> hours figuring out how to do a 30 minute job in 20, once. >> >> >>> >>> On 2/6/2016 1:57 AM, Boris Liberman wrote: >>>> >>>> >>>> "I'm an engineer, I have to look for any opportunity to optimize. " >>>> <-- that's fundamentally wrong. >>>> >>>> The rest is fun reading. >> >> >> I might have phrased it that I'm compelled to look for any opportunity to >> optimize. In your story, the engineer didn't optimize the system, he >> nominally optimized one aspect of performance and in the process pessimized >> the system. >> >>>> >>>> Boris >>>> >>>> >>>> On 2/4/2016 0:27, Larry Colen wrote: >>>>> >>>>> I just posted this to my facebook page. I have a strong hunch that at >>>>> least one or two people on this list will empathize with this. >>>>> >>>>> Life in engineer land. >>>>> >>>>> A few weeks ago, a friend of mine who worked in engineering in a >>>>> previous life, got in touch with me. Another friend of hers, also an >>>>> engineer, was about to get a second broadband connection and needed a >>>>> network cable run from his phone box to his server room. Sometimes >>>>> these installations are straightforward and take a few minutes, other >>>>> times, not so much and it takes someone who knows what they are >>>>> doing. So the first order of business was for me to head over there, >>>>> scope out the place and see if I could help, or if it would be wise >>>>> to refer the job to a friend of mine who owns a network cabling >>>>> business, and actually knows what he's doing. The evening I was free, >>>>> I headed over there with another friend who happens to be an >>>>> engineer, on our way to something else. >>>>> >>>>> So, to set the stage. We need to run a 20m (or 60 foot) cable, from >>>>> the outside wall of the condo, across the ceiling of the garage, and >>>>> up two floors to the office. In effect, we are throwing four >>>>> engineers at the job. In the real world, what would happen would be >>>>> that a real business would send their installer out, with a box of >>>>> cable, a fish line, and a drill, who would spend 10-20 minutes >>>>> tracking down the existing wires, another half hour running the line, >>>>> and 10-20 minutes terminating the line. >>>>> >>>>> But, this isn't the real world, this is engineerland. The first step >>>>> is to find out where the cable starts, and where it ends, then to >>>>> figure out if a new cable can be easily run. This process takes >>>>> something like forty minutes. We determine that it can, indeed be >>>>> done. But, I'm an engineer, I have to look for any opportunity to >>>>> optimize. So, I ask the question, "while we're doing this, are there >>>>> any other lines that it makes sense to run or upgrade?". >>>>> >>>>> Now, we start reverse engineering the existing network. Two hours >>>>> later, we've decided to replace the cat 5 of the existing DSL line >>>>> with cat 6, move the DSL modem from the downstairs office in the >>>>> kitchen to the server room, and to upgrade the cat 5 lines from the >>>>> server room to the wall plates in each of the kitchen office and the >>>>> dining room. >>>>> >>>>> In short, it has taken us about two hours to change the scope of the >>>>> job from running a single cable from the phone box to the server >>>>> room, to running two cables, and to replace four cat 5 cables from >>>>> the server room with an effective 1 gigabit bandwidth, to cat 6 cable >>>>> with a theoretical 10 gigabit bandwidth. >>>>> >>>>> One of the most important things I've learned in my engineering >>>>> career is to get a good set of job requirements before you start. >>>>> There are few things more important than being able to know when you >>>>> have actually finished the job. Yes, the requirements may change >>>>> while you are working on things, but it's important to note (for >>>>> billing purposes if nothing else) that they have indeed changed. >>>>> >>>>> The next step is for the customer to get a rough estimate of the >>>>> distances and send me a note, or spreadsheet, that says: >>>>> 2 wires from point A to B, approximately 60 feet >>>>> 2 wires from point B to C, approximately 10 feet >>>>> 2 wires from point B to D, approximately 40 feet >>>>> >>>>> RJ 45 connectors at points B,C, and D. >>>>> >>>>> What I received was a PDF diagram with 15 different locations, color >>>>> coded lines marking each of the different cables, notes on the >>>>> distances between each location, and notes as to which distances are >>>>> to be the installed cat 6, and which are to be patch cables. >>>>> >>>>> At this point we start discussing the drawing over email and SMS, >>>>> considering such vital details as color of the wire, how to mark the >>>>> wire and jacks, running pull string for future enhancements (already >>>>> implicit in the plan), where to get the various items, scheduling and >>>>> just about every other detail except for the color of the electrons >>>>> in the cable. >>>>> >>>>> At this point we have ordered the specially colored jacks, scheduled >>>>> the work for Monday, and have spent probably close to 15 engineering >>>>> hours on a task that would take a technician approximately an hour to >>>>> do. >>>>> >>>>> On the other hand, the customer will be able to surf the web from his >>>>> kitchen on a home network that is more finely engineered than the one >>>>> in an NSA supercomputer lab. >>>> >>>> >>>> >>> >>> >> > > > -- > PDML Pentax-Discuss Mail List > PDML@pdml.net > http://pdml.net/mailman/listinfo/pdml_pdml.net > to UNSUBSCRIBE from the PDML, please visit the link directly above and > follow the directions. -- PDML Pentax-Discuss Mail List PDML@pdml.net http://pdml.net/mailman/listinfo/pdml_pdml.net to UNSUBSCRIBE from the PDML, please visit the link directly above and follow the directions.
