>Hi Ed,
>
>thank you for tears of laughter released by your volcano-like outburst of ingeniously shocking inhumanities. Did at any time anybody survive co-operation with you for more than three minutes? I consider three minutes to be critical, because for most people that would be the longest space of time to live without oxygen. It seems to me that even the breathing of other living beings will let you explode like a pool full of gazoline with a torch thrown in. You're giving all honors to your name - which in the german language is an adjective addressing rude or coarse behavior. <*g*>
>
>Seriously. From my point of view one important task of this forum is to help people to improve their developing skills, especially people with less experiance. If everybody in UT was an guru-like expert, there would be the need for the chatter region only. I doubt it that a person, who was raided by your hellish scenario, will dare on asking a question to the UT members ever again.
>
My problems with Mr. Meester are centered around his basic desire to get people to believe that the (IMO) very narrow, PC, business based perspective that he espouses is couched in a deep understanding of automation and information theory. I'm getting tired of being told how SQL is an incomplete representation of the underlying relational data model, which is is not inherently limited to operating against, but is modelled against, and the ongoing, completely absurd notion that ISAM-based procedural behavior based on preconceived models of unknown data behavior by an "adequately saavy" developer are more optimal as a model for development of applications in the datacentric market.
I'm not CJ Date by any means. I think I've got reasonably good founding in automation theory, with a good deal of background in the mathematics that underlie the basis of database theory - things like formal logical, set theory and topology. I do read (voluminously) and do work in a wide range of environments. I have adequate instincts for knowing when someone has a clue about the theoretical underpinnings of our day-to-day environment. I'm tired of the scalable internal bullshitometer sticking over in the red no matter how I scale things when reading Walter.
Sorry, folks, mine is an analog device, and when things peg out, they peg out. Gotta do something about an enhanced digital device that detects and filters out persistent rectocranial insertion on recurrent sources of channel noise. If anyone has a chip that can be added (preferably to one's shoulder) to identify and remove such noise sources from the waveform, I'd appreciate schematics and a part number.
Let's assume my bullshitometer is defective. I'd say that Jim Booth, John Petersen and Christof Lange are credible sources - if my indicators differ from theirs significantly, I want to check my facts carefully. In fact, if the indicators differ, I
do check my facts, and assume I have some basic misunderstanding. I've yet to see Walter be (in his own mind at least) wrong, or check his facts, even when what he says is directly refuted by both theory and empirical sources. It's obvious that the basis for their signal generation is either is as defective and misaligned as mine, or perhaps (clearly, just maybe - I'm really speculating here) my bullshitometer has identified a persistent source of noise.
Walter likes to quote Date at us. I've read stuff by (by this I mean Date - Walter, too) him - I've attended lectures by him (Date, not Walter!) I guess that Date live, or as published and read by me, doesn't match that pure, clean, unadulterated waveform spewing from Walter.
I use the term spewing intentionally. This is meant to represent an uncontrolled and inadequately shielded and directed source of material that can at times be represented as a continuous stream, but is frequently encountered as discrete, particulate ejecta. I would not want my meaning to be misunderstood in this context.
To rephrase the above, genteel language in true Rauh fashion (just one semester of scientific German over 20 years ago), Walter has a very narrow view, spouts theory that doesn't match his sources, differs in theoretical and real-world application from mine, and completely disagrees with major, respectable sources of knowledge in the field who, when I disagree in my perceptions and analysis of a topic, at least cause me to reexamine my own position, and if I don't understand what they're saying, makes me ask them to help clarify things so I see what causes our different takes on an issue.
In short, I think my bullshitometer is working.
My screed preceding your message is excessive, and reactive, and for that, I do apologize to the other residents of UT wgh deserve to view it as offensive.
Mea culpa, mea culpa, mea maxima culpa I fucked up. Sorry, folks. You may take me to task on this correctly and without fear of my arguing that you're wrong.
But that doesn't change the indicators on the bullshitometer, or the premise behind the screed. Walter doesn't have a clue what is involved in "performance", what establishes performance requirements in a system, or what identifies operations that must be instinctively a purely serial and dependent chain of process and processes that, while ordinarily represented in serial fashion, actually are extremely parallel and discrete, and explcit exploitation of the conceptual parallelism in a system and the "divide and conquer" approach of reducing system performance by distribution of a problem domain.
Let's look at it in the real world. Walter says "hardware is expensive - creebing code saves money." Basic premise -
IRON AND SAND ARE CHEAP. FUNCTIONAL BRAINCELLS ARE NOT. Walter spends a week cutting the cost of a loop from 0.04 seconds/iteration to 0.03 seconds/iteration. Cost of a .01 second improvement in performance per iteration, at $100K/year (does not matter the real magnitude of the number), $2K, so it'll run on the existing Pentium 100 with Win95, 16MB of memory and an older 4B hard drive.
Cost of new 500MHz Pentium III, Win98, 128MB RAM, 10GB drive: less than $1200. No code change to endanger the program, instant performance increase of 50% a reasonable assumption, folks? Or does Walter have some mystical capability to rip more performance out of the P100 than I do, perhaps some hidden, vast degree of systems background that makes old, slower, less than entry level systems outperform new, inherently faster, more capable and more stable hardware. With a warranty! Maybe my bullshitomter assumptions hide the fact that he has deeper wisdom in databases and xBase than I do, but I'll bet very large sums of money on my hardware and systems background here.
Oops.
>For me, you are one of the most important persons contributing to this forum, and your replies often take me a step ahead on my way to a more professional developer. Regrettably, the language and personal attitude chosen in your ongoing thread with Walter Meester are not suitable to your professionality.
Criticism for my lack of civility noted anbd accepted as stated above.
I may be a contributor, but I don't run things here. And please don't base your judgement of the objective postings I make, or even the subjective ones, where I make clear the basis of my subjective analysis. Personal opinion, and my need to relieve stress and frustration through personal overreaction is a frailty that I am subject to.
And you would not believe how much toning down that message represented over my initial reaction. No grin. No joke. No humor at all.
>Please, find your way back to a more objective style.
I'll try. But I'd ask that you and the others who find my screed suitably objectionable to please examine your own internal bullshitometers, and let us know what you think of the veracity and credibility of the signal stream that prompted my reaction.
