lcs@... wrote: > Hi all, > > As Alain (following Derek) has tried to muster more > convincing arguments against the use of type suffixes % & ! # $, I > am (gasp, gasp) replying once more. > Hi Larry, I wouldn't say I am against the type suffixes so radically. Sure I hate to see them in my code and as I told you I don't want them to reappear in my programs when they are not obviously needed. I understand that others will still use them, while I could have some doubts on their reason they might give, because I was first against their omission. Like you, I used to think that they were quintessentially BASIC. But today, I'm forced to see that my experience contradicts the original statement you made that started the thread. However, I'm glad the type suffixes still exist, because as far as I know, FB has remained the best tool to run zillion of BASIC programs with little adaptation work. This is a selling argument in a certain way, but I have the feeling that this very argument is not suitable for the new generation of novice programmers. > But the consistent use of & does remind me that FB > treats handles and pointers as long integers. > That tells me a lot of useful things about manipulating > handles and pointers. Sure FBII had to be consistent, so is FB^3 (à la FBII if you wish), and it provides more means to give you the necessary insights, if you ask me. > Lets face it, programming ain't easy, and familiarity with > syntax can easily treble one's production. Thus consistent > programming pays off. That does not mean that the use of % & > ! # $ is inconsistent programming; rather it says that using > them (or not) is a committment, at least for the duration > of a project. > I agree, consistency is probably helpful to succeed but also to the novices who start to learn the language. There is an issue here because FB^3 allows now so many styles of coding to coexist in the same program. For beginners a certain predictability is preferable, I guess. > > > Not so. I presume we both use the French keyboard, on which all > five are clearly engraved. Only % and # are in uppercase > position, so the typing of & ! $ is perhaps more convenient > than with the US keyboard on which all 5 require shift > modification! > I pity them! > > Another reason I find the design of FB > instructive and mnemonic is worth mention here. The > representation of variables etc. in terms of sequences of > octets can be seen in many notations, indeed many more > than necessary. Alain gave a nice > example of that: > > theRect;8 = [resHndl&]+_bounds > > which says: move 8 octets beginning at the address > [resHndl&]+_bounds to the 8 octet addresses following the address > @theRect of the structure theRect. The fact is that much of > FB programming, no matter how subtle, reduces to the idiotically > simple operation of moving about sequences of octets. Possibly, but to say the truth, I wouldn't have coded for myself the thing like I did in reply to the post. I would have used in the case mentioned a CIconHandle which is normally a predefined type of the Toolbox. Finally, that would give something like that: theRect = resH..iconPMap.bounds It appears less cryptic to me. > > > FB, on the other hand, seems to me to be *unashamedly > concretist* in its architecture and syntax. I like this > because the programmer gets to see how FB code really > functions at the hardware level; that is some sort of > "understanding" of the code and a useful bunch of clues for > optimizing it. > I believe you are right. > > What can I muster as firey conclusion? Maybe this; which I > address to to Joe Wilkins in particular. Basic in general, and FB > in particular, have quite a few prominent features that are > more-or-less rare and more-or-less unfashionable in today's > programming world (type suffixes and thinly disguised use of byte > sequences are two). If you do not like these features you have an > increasing number of alternatives that let you ignore them. On the > other hand if you deprecate them, rather than recommend them as > precious aids for for beginners and part-time programmers, then > you will force them out of use through peer pressure. At that > point, I fear FB will lose its raison-d'^etre and quietly fold. > > Larry S (exit to the sound of the Salvation Army Band) > Now I'm sure, you will love FB^3. -- Cheers Alain ----------------------------------------------------- FB^3 in Europe: http://euro.futurebasic.com/ FB II Pouch: http://www.pixmix.com/FB/outils.html -----------------------------------------------------