Rosetta Code

Talk:Idiomatically determine all the characters that can be used for symbols: Difference between revisions

Page Discussion

Talk:Idiomatically determine all the characters that can be used for symbols (view source)

Revision as of 21:30, 17 September 2014

9,899 bytes added ,  9 years ago
m
→‎Arbitration?: added invitational comments.
(→‎Arbitration?: corrected a MAJOR error in my 48 words and still waiting for the first vote :-()
m (→‎Arbitration?: added invitational comments.)
Line 202:
<br>12. Programs that obey a list of restrictions (by not using features no longer existant in ooRexx) yield identical results for both.
<br>--[[User:Walterpachl|Walterpachl]] ([[User talk:Walterpachl|talk]]) 07:53, 17 September 2014 (UTC)
 
-----
 
(I apologize ahead of time to the Rosetta Code admins and others for duplicating the above statements and breaking them up below in order to address the points.)
 
Since you listed so many "true" statements of yours that I thought I'd address them point-by point here. &nbsp; I really wish you hadn't entered this long list of "true" statements here on Rosetta Code which I feel then have to be addressed, not withstanding your invite. &nbsp; This almost endless listing of truisms is very subjective. &nbsp; My singular opinion concerning all of this is: &nbsp; Enter '''ooRexx''' output in the ooRexx language section. -- [[User:Gerard Schildberger|Gerard Schildberger]] ([[User talk:Gerard Schildberger|talk]]) 21:30, 17 September 2014 (UTC)
 
<br>1. You never complained when I added a Rexx program in the REXX section (new or alternate version).
 
:::: Nor why should ''anyone'' complain? &nbsp; I did not do a lot of things. &nbsp; <nowiki> <humor> Shame on you Paddy, you didn't complain either. </humor> </nowiki> -- [[User:Gerard Schildberger|Gerard Schildberger]] ([[User talk:Gerard Schildberger|talk]]) 21:30, 17 September 2014 (UTC)
 
<br>2. You complained when I (once) used there a feature not supported in '''Classic''' Rexx (thanks).
 
:::: Yes, I pointed out some syntax errors in the (Classic) REXX entry that wasn't appropriate for the REXX section. &nbsp; I wouldn't characterize it as a complaint. -- [[User:Gerard Schildberger|Gerard Schildberger]] ([[User talk:Gerard Schildberger|talk]]) 21:30, 17 September 2014 (UTC)
 
<br>3. Modifying all current REXX programs to be accepted by ooRexx would be a huge effort (changing AND testing).
 
:::: The REXX section is for Classic REXX. &nbsp; It would be silly and a gihugeic waste of time to modify all those programs just to make them suitable for another language, not the least of which is that I don't even have an ooRexx interpreter installed. &nbsp; To even suggest that the programs be modified for another language (and then tested) borders on folly. &nbsp; Since ooRexx has it's own language section, it wouldn't matter that much what the definition of a Classic REXX interpreter is if output from ooRexx would be in the ooRexx language section. -- [[User:Gerard Schildberger|Gerard Schildberger]] ([[User talk:Gerard Schildberger|talk]]) 21:30, 17 September 2014 (UTC)
 
<br>4. Entering the modified versions in the ooRexx section would make ooRexx' rank larger than Rexx' but introduce lots of redundancy.
 
:::: There is no need (my weak opinion) to modify Classic REXX programs and then enter them in the ooRexx language section, that wouldn't show anything different from the original algorithm, but I certainly don't want to hinder or deter anyone from adding a <nowiki> {{trans REXX}} </nowiki> and doing the testing when using ooRexx (in its own language section). &nbsp; I don't understand your concern about rankings, why is that even an issue? -- [[User:Gerard Schildberger|Gerard Schildberger]] ([[User talk:Gerard Schildberger|talk]]) 21:30, 17 September 2014 (UTC)
 
<br>5. '''Classic''' Rexx means different things to different people (I know of at least two).
 
:::: Yes, this is one thing I agree with, but I disagree with the number; &nbsp; I believe it to be many. &nbsp; The meaning of what Classic REXX is (and what a Classic REXX interpreter is as well) causes much debate and discussion, that is the major reason that ooRexx has its own language section --- then the subject is mainly moot and needn't be addressed and discussed endlessly (seemingly) on Rosetta Code. -- [[User:Gerard Schildberger|Gerard Schildberger]] ([[User talk:Gerard Schildberger|talk]]) 21:30, 17 September 2014 (UTC)
 
<br>6. Ideally all interpreters give the same result for a program (apart from the interpreter's or compiler's version, of course).
 
:::: There are other differences (whether or not it's a bug is based on some variations of definitions). &nbsp; I've found a whole ****load (wheelbarrow, I meant to say) of differences between Classic REXX interpreters, most fixed as bugs, others dismissed as working as designed (WAD, ... sigh) --- essentially, you believe what you want, we believe what we want. &nbsp; One problem is which Classic REXX standard is being used (or adhered to). &nbsp; I normally roll back to what I consider as the "gold standard", the IBM CMS implementation of (Classic) REXX, nobody can argue that that version isn't a Classic REXX as that's where it was invented and is maintained as such to this day. &nbsp; Not all interpreters belong in the Classic REXX section, ooRexx belongs in the ooRexx language section, NetRexx belongs in the NetRexx language section. -- [[User:Gerard Schildberger|Gerard Schildberger]] ([[User talk:Gerard Schildberger|talk]]) 21:30, 17 September 2014 (UTC)
 
<br>7. Adding output for a program from several interpreters makes sense when they differ (e.g., the RANDOM and JUSTIFY bifs of REXX).
 
:::: '''ooRexx''' output belongs in the ooRexx language section, not the (Classic) '''REXX''' section. &nbsp; Adding output from object-oriented languages in the wrong language section (i.e., not the "base" language) shouldn't be allowed, it just clutters up the "base" language with an o-o version, whether or not that example uses o-o constructs or features. &nbsp; If people want to see what ooRexx has for output, look in the ooRexx section. -- [[User:Gerard Schildberger|Gerard Schildberger]] ([[User talk:Gerard Schildberger|talk]]) 21:30, 17 September 2014 (UTC)
 
<br>8. Rexx programs using features added by ooRexx are entered in the ooRexx section possibly stating the minimum version to be used.
 
:::: I go further than that. &nbsp; ooRexx entries are to be entered in the ooRexx section, no matter what features are used (or not used). &nbps; Furthermore, no matter which program is used, when using the ooRexx interpreter, enter those results in the ooRexx language section. -- [[User:Gerard Schildberger|Gerard Schildberger]] ([[User talk:Gerard Schildberger|talk]]) 21:30, 17 September 2014 (UTC)
 
<br>9. Two implementations of a language are compatible if each possible program yields the same result when run on both.
 
:::: Er, no. &nbsp; Two languages aren't compatible just because they both "execute" identical programs: &nbsp; '''K=1''' &nbsp; (as a simple example) will probably be executed identically on all manner of programs, not just REXX. &nbsp; Two languages are compatible if they have compatible features (or equal, if you will). &nbsp; ooRexx has quite a number of features/options/extensions that aren't compatible or supported by the Classic REXX interpreters, and, of course, vice versa. &nbsp; I believe that since ooRexx can't interpret all Classic REXX statements, it's not a Classic REXX interpreter. &nbsp; I know it can execute ''most'' statements, but like chimps, we share 98% of DNA. &nbsp; Close, but no cigar. -- [[User:Gerard Schildberger|Gerard Schildberger]] ([[User talk:Gerard Schildberger|talk]]) 21:30, 17 September 2014 (UTC)
 
<br>10. Implementation B is upward compatible with A when every program that executes successfully on A runs on B yielding the same result
 
:::: I don't quite understand this statement. &nbsp; So the case of &nbsp; &nbsp; &nbsp; <big>'''upper @'''</big> &nbsp; &nbsp; &nbsp; (which isn't compatible with ooRexx) means that ooRexx isn't upward compatible with the other? &nbsp; I may be getting the wrong idea about your statement. &nbsp; That would make ooRexx not compatible with any Classic REXX. -- [[User:Gerard Schildberger|Gerard Schildberger]] ([[User talk:Gerard Schildberger|talk]]) 21:30, 17 September 2014 (UTC)
 
<br>11. ooRexx is not fully upward compatible with Classic Rexx but to a large extent.
 
:::: Even '''if''' ooRexx was fully upward compatible with Classic REXX, ooRexx still has it's own language section. &nbsp; People go to the Classic REXX section to look and/or use copies of that REXX code to use with their Classic REXX interpreters. &nbsp; They go to the ooRexx language section to see ooRexx output (or programs). -- [[User:Gerard Schildberger|Gerard Schildberger]] ([[User talk:Gerard Schildberger|talk]]) 21:30, 17 September 2014 (UTC)
 
<br>12. Programs that obey a list of restrictions (by not using features no longer existant in ooRexx) yield identical results for both.
 
:::: That would depend on the list of restrictions (is it complete?) --- I say it isn't, as witnessed by a long discussion on this very topic in some older postings of the '''comp.lang.rexx''' newsgroup. &nbsp; It is not my dog, I don't want nor feel to need to list all the differences, I don't care what they are as I deal with only Classic REXX and that is my only concern and focus. &nbsp; Whether or not whatever differences/enhancements/additions/incompatibilities exists in ooRexx, is one major reason that ooRexx has it's own language section. &nbsp; I see no reason whatsoever why Classic REXX programs (REXX entries on Rosetta Code) need to conform (or be changed) to another language's (ooRexx) restrictions or idiosyncrasies. &nbsp; -- [[User:Gerard Schildberger|Gerard Schildberger]] ([[User talk:Gerard Schildberger|talk]]) 21:30, 17 September 2014 (UTC)
-----
 
Since I have to resort to a very short (50 words) synopsis of what I believe about entering Classic REXX programs on Rosetta Code &nbsp; (without having to define what Classic REXX or other terms are):
 
::: '''REXX''' is for Classic REXX interpreters.
 
::: '''ooRexx''' is for ooRexx regardless that it does (or doesn't) use ooRexx syntax or features.
 
::: Programmers can peruse ooRexx to observe ooRexx differences.
 
::: The fact that ooRexx ''may'' execute Classic REXX programs doesn't deter that: &nbsp; output from ooRexx should be in the ooRexx section.
 
-- [[User:Gerard Schildberger|Gerard Schildberger]] ([[User talk:Gerard Schildberger|talk]]) 21:30, 17 September 2014 (UTC)
 
-----
 
== entering ooRexx entries in wrong language section ==
Retrieved from "https://rosettacode.org/wiki/Special:MobileDiff/243917"