Dear Larry, Thank you for your contribution. I've seen that one in so many newbie tomes about email etiquette, that were written on a first to (new small, alternative) market, by newbie to the net journalists, and originally for decades in internet news groups that were appropriately for the Phds. who predominated then to give us the net itself. Well, who's impressed with what? Wasn't it you who typed "Phd.". You have impressed me, some. Everybody's a critic. I was responded to in public, rather than private, as again here, and in a subtleness missed by many, am responding (again) in (un)kind to demonstrate it's effects: your "public" correspondance (sic), and now my response in (un)kind, rather than traditional "offline" personalized reply email traditions. I went to college. I (like your and my retorts here have) took up space. Now it's traditional for me to feel I'm supposed to answer back things like: "If you don't like what you see, go see something else. Don't whine. Don't impose your personality on others. And always criticize in private. Don't intimidate those who seldom contribute into contributing more seldom, if never. I bet those guys know things we don't know but could sure use, even if they are only confirmations and afirmations, or exactly the opposite, that can broaden the perspective of the subjects at hand for some of us. If they speak to the forum's theme, in any real way or part, I hope to get the chance to see what they have to say." but I just don't have the time to take a shot at you; to return a volley of fire. OK, lets play for just a moment. Rodney King, after taking "the greatest ass whoopin of all time", bounced back from his feelings and said "Couldn't we all just get along". Let's play nice together kids. This isn't technical writing here. It's only email. A "forum" for'em. "We have met the enemy, and he is us" (POGO). "Yeah, and we're beating him back, Sir" (ANON). Personalized expository submissions that attempt to use same to stimulate not only thought, but what is too rarely achieved in "dry" forums; memory that inspires action and change by larger numbers of people, from larger audience participation (short of the Jerry Springer effect, we pray, ... some of us) are what's sought and expected here, IMHO. That's what we were "chided" to do by the creators of this forum, when it was first created, and again when it was revitalized. e.g. Just postings of things like new cool URLs that fit recent threads herein takes a lot of my time (to go and see, lest I miss out) that could be saved by reading nice short reports/revues of their content that is more "sufficient", as reported, to facilitate my just appending these emails to an archive file for keyword searching, to then find those URLs in useful context, on demand. I greatly appreciate such URL notification and posting, but many are soooooooo terse (yet so valuable, in so many cases, so many ways). To those that did "revue" for us, many thanks, for saving so many minutes in the present and the potential to save so many in the future, when I remember one of your phrases and use it to locate the source you cared enough to apprise us of. I won't be appraising creative writing skils tho (sic),. That's bottom-feeding by those who truely are self impressed. I understand typos and keystroe errors that look like misspellings being left in so contributors can find brief of moments to share w/us, (not too) poorly formed or punctuated sentences, and run-ons that are actually stream of thought w/no time to seperate. Fortunately, I type well enough that my doing all those is usually mistaken as colloquial prose (which it will unfortunately soon be, I fear). Is this a forum or an e-bibliography (a running library card). This is new millennium EDI/XML/e-biz stuff. Our pocket protector is now our e-holster. Draw (compose), pardon-ne'r. Consult the primal emails that started this forum. I read many the day published and remember many well, and abide the encouraging requests. Mostly bc. those encouragement's (pronounced "incur rage mants, but not meaning anything like that) were so "wisely" crafted and have now been so successful. My self discipline to immediately follow each "chide" to try something by giving my own real world example may too often be seen as something else, by others with different motivation than what mine really is. When I can't be clear enough to chide clearly bc. the narration is growing enough already, I skip straight to my example's methods. I write what I'm attempting to describe to my target portion of the total audience. I avoid writing to the Phds. in the manner they publish in front of the other Phds.. That's for the journals. This isn't such a forum, nor am I trying to describe chopped liver. Sorry! Thank you very much, Jim Cunningham P.S. Yeah. Soooo? My goal therefrom is unabashedly selfish: to ---->vigorously<---- promote the drastic reduction of source code by the very drastic reduction of all the "asides" and distractions of definitions, initializations, validations, tests etc. The same old keystrokes that have been typed too many times, just because the names were changed to protect the income, and make "hand compiling" or "divining the pith of the program" so distracting and tedious from so many "side-trips", despite the necessity of their function. The power of not only a data-dictionary but of multiple normalized data-dictionary relational records or "appendations of function" to/within such data dictionary tools seems rarely implemented and thus underutilized, if not substantially not understood. Thus I feel compelled to make an attempt to contribute (but let's not tell them about "the power of Progress"? Yeah, they're gonna find out anyway). Finish crafting the tools, I say. Thus, having successfully "removed" all that to the dB, prepending nomalization's keys to trivialize auto-reinsertion to source-code "streams" via dB record driven pre-preprocessor(s), "pattern hatching" of the remainder becomes much easier and much more clear. Then I strive to move actual source code (patterns), first to callable files organized in directory trees of "names" that announce function, then to template files and subsequently to dB records of normalized pattern elements. The tokens left in the smaller compact modules of source code, when wisely "tokenized" themselves with names chosen in appropriate vernaculars (and these "facets" especially, can further be reduced to tokens for overlayment from dB records, for very useful translations/vernaculars/points-of-view, documentations and internationalizations and such) can be further exploited. It's simply a matter of taking all of such work and encapsulating all the code I "keystroked" to achieve all this into tokenized preprocessors, moving any and everything "specific" to a particular invocation of such work to generalized dB records, which I then normalize and make "populable" for the specifics of the next times I want to do such things on other things. Patterns seem then to "pop out" at me, as I realize I never need to stroke the keys in that particular pattern ever again, if I "press on". Then I write a driver to use from now on to interview me for the current specifics so IT, calling a cascade of preprocessing, now populates additional dB record sets to feed another preprocessor to overlay templates (constructed/generated from the other dB record sets by preprocessing) to then generate the new source code suite (which turns out to be not so "new" overall), truly in microseconds. There can be so little code left in the source code, quite reasonable data constructs to get all the rest into the dB become obvious, IMHO. Since, not only can every program be reduced to combinations of the three simple constructs of loop, assignment and comparison, let's not forget they themselves are constructs and can be further reduced to "lists" containing only the token 0 or 1 (since that's all each of the transistors can manage), if laid out in the appropriate patterns (of patterns). The nomenclatures of the new languages, IMHO, run in the opposite direction. It's time to work at higher levels, not great long executable statements with soooo many keystrokes; so much reaching for the punctuation and shift keys with the weakest, least coordinated and shortest fingers (can't we quit QWERTing). I typically ran my prototype documentation generator in multiple passes "by hand" over the source code, creating, by editing, multiple sets of various useful "manifestations" of same, but particularly enjoyed producing English documentation, in outline form, that compiled "directly", by that form being pipelined through a short cascade of preprocessing reconstitution. Getting such to work in reverse is more work at first but you quickly get connected to seeing everything the other way round and it quickly feels familiar. Then that tool set is then quite useful for many things; not just translators, but translator generation in a nicely general, clear, seriously self documenting, easily maintainable system. Thus, "single click regeneration" from updated structure records melded with the particular data of the particular instance/project/package seems an attractive alternative to traditional version control techniques. Long about now, could EDI/XML aficionados help move us all forward to higher level data structures, higher methods of organizing and ordering them, opportunities of reuse, (spoken/written) language independence, etc? e.g. if you've done the work in English, why not have the work, done in a/all other languages, be available, "a keystroke away"? It's not their job; not so much as our friends at XML-PERL-DOM; tho that forum is ""hooked" on PERL" (but you go guys! GREAT work!). I was attempting to empower the office clerks of the world (many sans diplomas at many levels, yet intelligent nevertheless) who had for so many decades excelled at paper form creation (pencil and ruler > copy mach.), organization, implementation and utilization of work, yet lost so many of their number from the writing of programs, and give them back the utility they had provided. They lived and worked among those who actually did the work of industry. They knew the "real" names of "things" in their contexts. A potential loss in the numbers of programmers, having to give those jobs back, is likely. But I predict global business will insist. Why spend scarce hard currency of third world countries on their people learning to keep current on the newest fad programming languages when those languages themselves can be completely obviated, when implemented as well as we can and should. I think that with things like multiple dBs of EDI, XML, records about their structure etc. and auto-generated browsers triggered by record updates, and auto-regenerated generators, why do we keep stroke-in (sic) our boards? We ought to be (strictly) dragging, dropping and clicking. That's what I envisioned many years ago and delivered at the prototype level. But why do most developers of program development workbenches/suites, themselves, still use the keyboard? They didn't finish the work for their own benefit (double entant? I'll never tell). The cobblers children go shoeless. That all this should then be "interfaced" with voice recognition and speech synthesis (simply pipelined) so the computer would simply interview the worker who needed something, so as to then immediately provide it, did not escape me, (many) years ago, as those technologies were around back then and reporting some very good work. We just had to wait for R&D cost recovery, part count reduction, programmed ASIC production lines, surface mount technology, and accumulation of heat sink data by trial and error (so few graduated engineers actually did well enough at heat transfer differential equations for sufficiently accurate/usable modelling/predicting, until programmable automatons for same became widely distributed, so as to break the gigahertz and "real estate" barriers long ago), not to mention marketing channel/distribution-rights conflict resolution, etc. I have a question. Why aren't we there yet? Or better yet: Why are we still just here? Are the continuously repeated tediums of nonautomated code production, and especially nonautomated production of the underlying code production productions, holding us all back? Do those who sponsor/pay_for (I'll try to even work in camel notation, to inspire the readers to take note of how much is actually available to pursue clarity in nomenclaturizations) our work continue to be double and triple billed (at a minimum)? Why do we "hatch" patterns, instead of the other way round, generating them with the purpose of reuse, selecting/clicking/speaking them and ordering/dropping/speaking them? Because we're not doing same on the things they are made out of? Shoeless AND clueless? Barefoot and loving it? Does the emperor have such a small wardrobe? Why isn't the computer asking to interview us to provide us with more and more empowerment? "Generated code is usually unoptimized" is a misstatement of the problem. That's the compiler's job! The compiler writers job, specifically. So let's get the EDI and XML specs. done as well as they need them. But then we really should take those raw languages away from any but the compiler writers, and their compilers should translate optimized code into much more readable and still compilable form. Give me my own version of same, with much renaming of most things, transposition of punctuations and nomenclature, in English outline form. And I don't want to do any more work than scrolling and clicking. They should move as much of the source out of itself and into something like generated "CORBA tables", to reincorporate in not only reconstituted source code, but time-linked/up-to-date "help"/documentation (these should never be separate "things") Things self-referential are only ambiguous. They are seldom intractable over long enough periods of thinking about them in search of clarity of thought about them. Certainly none of the ones we consider here are intractable. They often yield to parallel thinking. We can walk and chew gum at the same time w/o either of a trip and fall incident or foot in mouth event, pretty much w/o breaking a sweat. And, after all, we have forums for'em, too. And journals. P.P.S. If you can't remember it, forget it. ------ XML/edi Group Discussion List ------ Homepage = http://www.XMLedi-Group.org Unsubscribe = send email to: [EMAIL PROTECTED] Leave the subject and body of the message blank Questions/requests: [EMAIL PROTECTED] To join the XML/edi Group complete the form located at: http://www.xmledi-group.org/xmledigroup/mail1.htm
