Path: csiph.com!usenet.pasdenom.info!aioe.org!eternal-september.org!feeder.eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail From: Unknown Newsgroups: comp.lang.forth Subject: Re: Software's Weak Link is Code Date: Sat, 26 Jul 2014 08:17:27 +0000 (UTC) Organization: A noiseless patient Spider Lines: 82 Message-ID: References: <3df01668-b1e3-42ff-8edd-d3bdbc13b02d@googlegroups.com> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Injection-Date: Sat, 26 Jul 2014 08:17:27 +0000 (UTC) Injection-Info: mx05.eternal-september.org; posting-host="7500eadc011d87baa0f8cc360c6fc80e"; logging-data="446"; mail-complaints-to="abuse@eternal-september.org"; posting-account="U2FsdGVkX190s8QzSGGac7VTx40KwPoYbfjArUsD8Y0=" User-Agent: Pan/0.133 (House of Butterflies) Cancel-Lock: sha1:BTWXp9YL/Xlzs6T8T1Hgrm5Rp00= Xref: csiph.com comp.lang.forth:32674 On Thu, 24 Jul 2014 03:26:58 -0400, Rod Pemberton wrote: > On Wed, 23 Jul 2014 12:59:42 -0400, wrote: > >> Software's Weak Link is Code... - this is the title of an 'electronic >> design' article which I just read: >> [link] >> >> Peter Thorne writes about a paradigm shift in which we are living now - >> leaving 'hand-coding' behind us, switching to automated code generation >> using system-diagram-based tools and the like. > > Someone must still code the "system-diagram-based tools" via > 'hand-coding'. > The fact that someone had to make the stone-ax to..to dig the coal, to make the electricity, to make the aluminum..to ;doesn't stop you from useing the airliner? > Once the "system-diagram-based tools" are coded correctly, then anything > compiled using them could solve the problem of code being the weak link, > if code really is such a problem, which I doubt. > > The big advantage isn't the elimination of code, but the elimination of > coding skills. You are just rephrasing the issue. > They've replaced skilled programming with > point-and-click diagramming where you don't need an education or skills > in programming, math, engineering, problem solving, or much intelligence > to code, etc, although you may need some skills of the specific domain, > say PLDs or EFT or OLTP or whatever. > "point & klik" sees only what-is-physically-done. "recognise & select, instead of DUMB/inefficient remembering" is the ADVANTAGE. > Using diagrams would offer tremendous advantages to those who are > especially visual as compared to those who are verbal. Unfortunately, > verbal people account for 75% of the populace. So, you diminish the > number of people who would work as effectively with the new tools. Yes: the econamy doesn't NEED them; like the 95% population that used to be farmers. Now they can all be politicians, lawers and Maidox. > I'm > not saying verbal people can't use visual tools. Visual people use > verbal tools every day. I'm just saying that one group is likely more > effective with specific tools than the other. > Yes. > Another advantage is that an image or "visual code" can represent far > more information than some lines of written code. If anything, the > "image is worth a thousand words" belief is comparable to a language > with a high, average level of "function points" per code statement. > >> This is a chance for Forth to stay alive. Using Forth is an excellent >> choice for building such kind of tools. Forth has all the features >> needed for automatic software generation. > > Compilers, including C compilers, have "all the features" that are > needed for "automatic software generation" too. There are also many > programs which do automatic code generation, e.g., Lex, Yacc, etc. > >> VisualFORTH uses automatic software generation, and that's the way to >> go. > I like that idea. Where did color-forth "go". I've been using 'very visual' [can work without a keybrd] ETHOberon, since 90's. For me, anything else is like tying your shoes-laces wearing boxing-gloves. But the kiddies don't use ETHO. > Ok. But, it seems that Forth would be very "late to the party" to me, > especially considering that VisualBasic has been around since 1991 ... > I.e., timing, advantage, etc issues. > > Wasn't VB very popular? > Rod Pemberton