Talk:Ch (computer programming)
![]() | This article was nominated for deletion on 1 May 2011. The result of the discussion was keep. |
![]() | Software: Computing Unassessed | ||||||||||||
|
Dubious
I've marked a number of claims in the article as dubious. The claim that Ch is used in many universities and schools appears unsupported. No one questions that it is promoted for use in teaching and that it's used in that way at some (unknown) number of institutions. But there's no evidence this can be characterized as many; the use of this characterization appears promotional.
The claims that Ch is comparable to a C shell or that, as a shell for running external commands, it supports wildcarding or iteration of external commands appear factually incorrect. Please see the discussion at Wikipedia:Articles for deletion/Ch interpreter (2nd nomination). Msnicki (talk) 15:44, 5 May 2011 (UTC)
- To put it into context, "many" would be a noticeable fraction of the thousands of academic institutions. Perhaps a few hundred would be enough to justify that term. Scaling it down to the apparent reality (several, e.g., no more than a few dozen) would be less promotional. However, without a reliable source, the statement as a whole is promotional. TEDickey (talk) 23:40, 5 May 2011 (UTC)
- here, here, here and here are my findings related to many schools using Ch for teaching, plus the existing references in the articles. — Preceding unsigned comment added by 108.203.138.17 (talk) 22:13, 12 May 2012 (UTC)
Notability tag
I'm a bit confused by the notability tag. First there are many sources that would seem to support the notability of the topic. Secondly, those sources were debated at AfD/DrV/AfD and found acceptable. Could anyone clarify? Hobit (talk) 02:52, 10 May 2011 (UTC)
- There was the apparent WP:COI on the part of Chuser and various IP-addresses which created this topic. A majority of the sources provided by those editors are either the primary source (Harry Cheng) or authors who have been demonstrated to be close associates of Harry Cheng. TEDickey (talk) 08:22, 10 May 2011 (UTC)
- The article needs prose which explains why the subject is notable. That means referenced content which shows why this subject is important. If more articles set out to explain such from the beginning, we wouldn't have lame wars over the deletion of articles which so not. Chris Cunningham (user:thumperward: not at work) - talk 21:13, 10 May 2011 (UTC)
- @TEDickey: The AfD clearly found it to meet our notability guidelines. The issue of sourcing was extensively discussed and found acceptable. @thumperward: The tags in question make the claim that it doesn't meet WP:N or other notability guidelines. That is different than importance. Also, there are _plenty_ of third-party sources. I just don't see an issue. Hobit (talk) 11:45, 12 May 2011 (UTC)
- I did add the main users guide and removed the dubious tags. Hobit (talk) 11:45, 12 May 2011 (UTC)
- I disagree with removing the dubious tags. My comments supporting the tags appear in section above. The statements were not supported by the cited Campbell article, which clearly describes a C language shell that can also run commands, not a csh-style C shell in the section, "Command Shell & Shell Programming". But as an alternative to the dubious tags, I've fixed the language to match the sources. Msnicki (talk) 13:22, 12 May 2011 (UTC)
- Humm, reading the userguide, it looks a lot like a C shell to me. It does have wildcarding, tab completion and access to standard unix binaries (rm, ls, etc.). Apparently some of that is newer than the Campbell review (or he missed it, don't know), but the user guide supports it. That said, I don't really have a problem with your changes so... Hobit (talk) 01:56, 14 May 2011 (UTC)
- I disagree with removing the dubious tags. My comments supporting the tags appear in section above. The statements were not supported by the cited Campbell article, which clearly describes a C language shell that can also run commands, not a csh-style C shell in the section, "Command Shell & Shell Programming". But as an alternative to the dubious tags, I've fixed the language to match the sources. Msnicki (talk) 13:22, 12 May 2011 (UTC)
- It does not do wildcarding. Chuser conceded that it depends on anything it runs being able to do wildcarding, same as cmd.exe, not like any Unix shell but argued that it doesn't matter. On Unix, the utilities and the shells are different things. Adding a lot of Unix utilities (in this case, GNU utilities) to something that wasn't already a Unix shell does not make it one. Ultimately, whether something "looks like a C shell" could be a matter of personal opinion, I suppose. But the facilities for running external commands that you list (and all the rest that Ch provides) are common to almost all command processors, certainly including ALL the Unix shells, not just csh or tcsh, but sh, ksh, bash and all the rest as well, and even to cmd.exe and I don't think anyone argues they all look like the C shell, too. Personally, I think the only part that might make it look like a C shell to anyone (and only if they don't realize that the "C" in C shell is misleading) is the part that in fact doesn't look like any other shell I know of, certainly not the C shell, that in between running external commands, you can also type pure C language statements and whole C language programs, exactly as Campbell shows. Moreover, Chuser stated that he believes that when Gary Wang described it as a complete C shell in the Spectrum article, he meant it was a complete C language shell, not a csh. But as long as we can move on, I suppose that's all that matters. Cheers, Msnicki (talk) 04:14, 14 May 2011 (UTC)
- Fair enough. I think you know a lot more about the topic than I. I just saw that wildcarding was used in examples in the user manual, so I'd say it's supported. And yes, for us folks who use bash/tcsh/ksh fairly interchangeably, it's a lot like csh. But I do very little shell script writing. Hobit (talk) 03:51, 15 May 2011 (UTC)
- It does not do wildcarding. Chuser conceded that it depends on anything it runs being able to do wildcarding, same as cmd.exe, not like any Unix shell but argued that it doesn't matter. On Unix, the utilities and the shells are different things. Adding a lot of Unix utilities (in this case, GNU utilities) to something that wasn't already a Unix shell does not make it one. Ultimately, whether something "looks like a C shell" could be a matter of personal opinion, I suppose. But the facilities for running external commands that you list (and all the rest that Ch provides) are common to almost all command processors, certainly including ALL the Unix shells, not just csh or tcsh, but sh, ksh, bash and all the rest as well, and even to cmd.exe and I don't think anyone argues they all look like the C shell, too. Personally, I think the only part that might make it look like a C shell to anyone (and only if they don't realize that the "C" in C shell is misleading) is the part that in fact doesn't look like any other shell I know of, certainly not the C shell, that in between running external commands, you can also type pure C language statements and whole C language programs, exactly as Campbell shows. Moreover, Chuser stated that he believes that when Gary Wang described it as a complete C shell in the Spectrum article, he meant it was a complete C language shell, not a csh. But as long as we can move on, I suppose that's all that matters. Cheers, Msnicki (talk) 04:14, 14 May 2011 (UTC)
- What it appears to be doing is using the C library
system()
routine in
stdlib.h
to execute anything that's a command, not pure C language. On Windows, this maps to cmd.exe (and I've been able to match the behavior.) On Linux or Unix, the behavior is locally dependent on whatever has been chosen as the default shell, but usually, that's bash. This would be consistent with a claim that wildcarding works on Unix, but only because it's bash that's doing it there, and with the bash-isms in the Ch manual like2>&1
, an idiom that bash supports but csh does not. Msnicki (talk) 18:16, 15 May 2011 (UTC)
- What it appears to be doing is using the C library
- It may help to explain that if you're allowed to use the
system()
routine, here's all it takes to create a complete command processor in C:
#include <stdio.h>
#include <stdlib.h>
int main( )
{
char command[ 256 ];
while ( gets( command ) ) system( command );
}
- Msnicki (talk) 18:47, 15 May 2011 (UTC)
- got it. I'll note the buffer overflow issue just to sound moderately educated :-). Thanks for taking the time to explain what's (probably) going on! Hobit (talk) 01:12, 17 May 2011 (UTC)
- Wikipedia policy requires wp:No_original_research, we need
WP:RS. In Sriranga Veeraraghavan's article, the subtitle is "Ch is a combined C shell and IDE". is it appropriate to remove "c" from c shell? the command "ls c:/windows", "int i=5, *p", "echo $WINDIR", "getenv("WINDIR")", "`ls`", "33+344*333" works fine in ch shell.
For me, C shell refers to the C language shell in the article, not csh. — Preceding unsigned comment added by 99.100.24.205 (talk) 00:59, 19 February 2013 (UTC)
COI
Several editors have raised concerns on Chuser's talk page and on the WP:Conflict of interest/Noticeboard that a major contributor to the page, Chuser, may be a single purpose account with a close connection to the subject. Msnicki (talk) 14:20, 12 May 2011 (UTC)
- Noting that Chuser has been asked more than once, without giving a straightforward answer, I'm curious what procedures might apply to this situation TEDickey (talk) 20:28, 12 May 2011 (UTC)
- The advice offered on the noticeboard was that it's fair to ask about the relationship and to express concern, giving reasons, but not fair to go beyond that to insist on an answer or to hound the individual. Msnicki (talk) 20:51, 12 May 2011 (UTC)
- However, maintenance tags aren't to be removed without a consensus - which would be discussed here. TEDickey (talk) 20:55, 12 May 2011 (UTC)
- Well, I'm planning on removing the notability and sources tag (but leaving the COI tag). No one has really explained why the sources that were found acceptable at AfD are now not enough. The COI on the sources was pretty much rejected as an issue at the AfD. So I'd say we have enough sources... I'm guessing folks object, but I'd really like to hear a reason other than those raised at the AfD (which I think a wider consensus rejected) or I guess some reason why the consensus at AfD isn't enough. Hobit (talk) 01:59, 14 May 2011 (UTC)
- Even though I was the most vocal in questioning the sources in the AfD, I agree with Hobit on this point, which is why, when the outcome of the AfD was keep, I removed the notability hat and was surprised when it reappeared. We use consensus to decide things here and that sometimes means things don't go your way. The consensus at AfD, which turned on this specific question of notability, was that the sources were sufficient. I think we should accept that and move on. Msnicki (talk) 14:09, 14 May 2011 (UTC)
- I've removed the tags. If someone objects feel free to revert, but please address why the AfD doesn't support said removal. Thanks! Hobit (talk) 03:53, 15 May 2011 (UTC)