What is plagiarism exactly?

[Chris Whittington]

If you copy from one author, it's plagiarism. If you copy from two, it's research.

[Sean Evans]

If you copy from one author, it's plagiarism. If you copy from two, it's research.

...What is originality? Undetected plagiarism. WILLIAM RALPH INGE

On a serious note:

Misappropriating the coding of another computer chess program or programs is neither a moral nor a philosophical transgression; it is a professional one.

Cordially,

Sean

[hyatt]

If you copy from one author, it's plagiarism. If you copy from two, it's research.

Only in some countries. Not in mine...

Nor yours...

We might talk about China, of course.

[BB+]

What is plagiarism exactly?

I suspect you can check the dictionary, if that is desired, while I will give a working definition/procedure from the standpoint of an academic instead.

Some plagiarism cases are easy (verbatim copying), some are difficult. Most investigations start by trying to be quite explicit concerning what was done, and why it should be considered "unethical" (somewhat of a mystical academic word). Moreover, local knowledge is often important. A law school will differ from a medical school, while maths and literature won't have the same standards either. Additionally, the particular standard to apply in a given case might depend on the nature of the offense. Sometimes a direct precedent is available, but more often than not there will be some novel aspects to the case. What is to be noted, though, is that at an early stage at least some standard (modifiable to some extent as things progress) is chosen for the specific situation, so that the whole process is not endlessly side-tracked into the "meta-discussion" of generalities.

Since most examples from students on assignments (even in CS classes, where they can be quite inventive) tend to be rather perfunctory [and a teacher usually doesn't want to go thru the ordeal if it is not rather clear], I'll take "plagiarism" in a master's/doctoral thesis as my running example. It is quite untypical for an entire thesis to be unoriginal. More frequent would be the copying of a smallish part of it, something like 3-5 pages (out of 100 in total) in a given section/chapter of the thesis -- unsurprisingly, this material is often that which is least well-understood by the student (it might be a side-bar that interlinks the subject of the thesis to something else, or perhaps some technical point that the advisor suggested should be addressed). The student will frequently reword (or even re-order) said "plagiarised" material, but in most cases it will still be apparent the student didn't really understand it, but just spliced it in (for instance, there will be no constructive discussion added). In many cases [depending on the subject matter], simply prefacing this by: "The section follows [...]" as a citation will suffice. Or the work in question can simply be excerpted -- this is, after all, a rather small percentage of thesis.

On the other hand, if such material is not properly referenced, then this is indeed passing off someone's else work as one's own "original" work. One common excuse [and there are various famous persons who have copied parts of their thesis] is that it is a "small part" of the thesis, which I guess could be called "whitewashing" [rather than consider each part of the thesis, white or black, as an individual element for "plagiarism", just throw it all together and get a small percentage of grey]. Note that this is rather different from copyright [at least in the "damages" phase], where questions of percentages more often do arise. Usually those who are found to have plagiarised material in their thesis [or in other work] seek careers outside academia --- combined with a general lack of "half-way" penalties (you either revoke the awarded degree, or you, umm..., umm...), unless a case is particularly egregious and/or the offender ends up in politics, it is not likely to echo too far.

I could discuss how a plagiarism case is handled, but it's probably about how you would expect, with Commitees making subjective assessments, appeals to the Faculty Senate, or beyond, etc.

[BB+]

For "standards" in computer chess, there are various precedents (to be found in back copies of ICCA/ICGA Journal), while Larry Kaufman recently proposed (regarding "values" in evaluation, but his line of thought can be expanded): The bottom line (in my view) is this: if it is obvious to a knowledgeable observer that you could not have come up with the values you used without having seen those in another program you have made a derivative of that program. Of course this leaves room for interpretation but so do most laws. So something like: "If it is obvious to a knowledgeable observer that you could not have come up with X without having seen it another [specific] program", then it's unoriginal.

Here X should be sufficiently specific to rule out that it's merely accidental, or a common idea among many programs. It might be either "a large chunk of code", or so many elements of commonality [in important functionality] that it couldn't occur by accident. It could either be with specific implementation techniques, or more generally in the structural framework to the extent that this is "unique" to a given program (e.g., vanilla alpha-beta with qsearch() is not novel -- but the precise concept of 6 layers of depth-dependent search routines, all having differing pruning conditions at various specific places, etc., would be sufficiently specific to qualify as "protected expression", at least to some extent). For instance, the EvaluateWinner() code from Crafty that appears in pre-Beta Rybka would definitely suffice. Also sufficing (but requiring more analysis) would be the IPPOLIT congruence with Rybka 3 [from his statements, it seems that Rajlich concurs with this, even if one were to use a stronger "copyright" standard].

[Sean Evans]

For "standards" in computer chess, there are various precedents (to be found in back copies of ICCA/ICGA Journal), while Larry Kaufman recently proposed (regarding "values" in evaluation, but his line of thought can be expanded): The bottom line (in my view) is this: if it is obvious to a knowledgeable observer that you could not have come up with the values you used without having seen those in another program you have made a derivative of that program. Of course this leaves room for interpretation but so do most laws. So something like: "If it is obvious to a knowledgeable observer that you could not have come up with X without having seen it another [specific] program", then it's unoriginal

Hello BB+, so what you are saying is that your derivative testing will consist of a comparison of Rybka/Ippo* piece evaluations to Crafty piece evaluations and see if the evals reconcile. If they reconcile you will advise us that Crafty is a derivative program of Rybka/Ippo*.

Is that correct?

Cordially,

Sean

[Chris Whittington]

What is plagiarism exactly?

I suspect you can check the dictionary, if that is desired, while I will give a working definition/procedure from the standpoint of an academic instead.

Some plagiarism cases are easy (verbatim copying), some are difficult. Most investigations start by trying to be quite explicit concerning what was done, and why it should be considered "unethical" (somewhat of a mystical academic word). Moreover, local knowledge is often important. A law school will differ from a medical school, while maths and literature won't have the same standards either. Additionally, the particular standard to apply in a given case might depend on the nature of the offense. Sometimes a direct precedent is available, but more often than not there will be some novel aspects to the case. What is to be noted, though, is that at an early stage at least some standard (modifiable to some extent as things progress) is chosen for the specific situation, so that the whole process is not endlessly side-tracked into the "meta-discussion" of generalities.

Since most examples from students on assignments (even in CS classes, where they can be quite inventive) tend to be rather perfunctory [and a teacher usually doesn't want to go thru the ordeal if it is not rather clear], I'll take "plagiarism" in a master's/doctoral thesis as my running example. It is quite untypical for an entire thesis to be unoriginal. More frequent would be the copying of a smallish part of it, something like 3-5 pages (out of 100 in total) in a given section/chapter of the thesis -- unsurprisingly, this material is often that which is least well-understood by the student (it might be a side-bar that interlinks the subject of the thesis to something else, or perhaps some technical point that the advisor suggested should be addressed). The student will frequently reword (or even re-order) said "plagiarised" material, but in most cases it will still be apparent the student didn't really understand it, but just spliced it in (for instance, there will be no constructive discussion added). In many cases [depending on the subject matter], simply prefacing this by: "The section follows [...]" as a citation will suffice. Or the work in question can simply be excerpted -- this is, after all, a rather small percentage of thesis.

On the other hand, if such material is not properly referenced, then this is indeed passing off someone's else work as one's own "original" work. One common excuse [and there are various famous persons who have copied parts of their thesis] is that it is a "small part" of the thesis, which I guess could be called "whitewashing" [rather than consider each part of the thesis, white or black, as an individual element for "plagiarism", just throw it all together and get a small percentage of grey]. Note that this is rather different from copyright [at least in the "damages" phase], where questions of percentages more often do arise. Usually those who are found to have plagiarised material in their thesis [or in other work] seek careers outside academia --- combined with a general lack of "half-way" penalties (you either revoke the awarded degree, or you, umm..., umm...), unless a case is particularly egregious and/or the offender ends up in politics, it is not likely to echo too far.

I could discuss how a plagiarism case is handled, but it's probably about how you would expect, with Commitees making subjective assessments, appeals to the Faculty Senate, or beyond, etc.

And we have plagiarism from the standpoint of Art and Culture, according to the wiki .....

• Plagiarism and the history of art
  Through all of the history of literature and of the arts in general, works of art are for a large part repetitions of the tradition; to the entire history of artistic creativity belong plagiarism, literary theft, appropriation, incorporation, retelling, rewriting, recapitulation, revision, reprise, thematic variation, ironic retake, parody, imitation, stylistic theft, pastiches, collages, and deliberate assemblages.[3][12][13][14][52] There is no rigorous and precise distinction between practices like imitation, stylistic plagiarism, copy, replica and forgery.[3][4][5][6] These appropriation procedures are the main axis of a literate culture, in which the tradition of the canonic past is being constantly rewritten.[52]
  These appropriation procedures, vital to the whole history of art, have gained more and more importance since the beginning of the 20th century, with the boom of the modernist and postmodern movements; in modernist and postmodernist art, appropriation has been heightened as the central and representative device.[12][15][16]

One of the footnotes .....

• (p.437) There is between 'translation proper' and 'transmutation' a vast terrain of 'partial transformation'. The verbal signs in the original message or statement are modified by one of a multitude of means or by a combination of means. These include paraphrase, graphic illustration, pastiche, imitation, thematic variation, parody, citation in a supporting or undermining context, false attribution (accidental or deliberate), plagiarism, collage, and many others. This zone of partial transformation, of derivation, of alternate restatement determines much of our sensibility and literacy. It is, quite simply, the matrix of culture.
  
  (p.459) We could, in some measure, at least, come closer to a verifiable gradation of the sequence of techniques and aims which leads from literal translation through paraphrases, mimesis, and pastiche to thematic variation. I have suggested that this sequence is the main axis of a literate culture, that a culture advances, spiralwise, via translations of its own canonic past.

so, here we have an artistic/cultural view of plagiarism at variance with the academic view. Which way computer chess? And, a side question .... why are essentially academic standards rather than copyright standards being applied to Vas, a commercial programmer, who must have assumed his obligation was to stay the right side of copyright law, not academic rules?

Seems to me computer chess is more art than science and the artistic model with the obvious proviso of staying the right side of the law has merit in an age where the information available has exploded, in particular with the release of open source codes.

[hyatt]

Does not change the fact that copying source code without permission is (a) a violation of ICGA tournament participation rules and (b) a violation of any license agreement in force at the time and (c) a violation of copyright law. The latter two can be intertwined and are sometimes exclusive of each other, sometimes not...

All the hand-waving, whining, impugning the process / panel / ICGA is not going to change a thing.

Have you ever seen this:

"God, give me the serenity to accept the things I can not change, the courage to change those that I can, and the wisdom to know the difference."

While that is an al-anon idea (the serenity prayer) there is a message there you might want to dwell on for a while. You can only change _your_ actions. Nobody else's. Maybe, if you quit trying, the discussions will die away and this goes into the history books, rather than being dragged back to the present with each futile post you drop here...

[BB+]

And, a side question .... why are essentially academic standards rather than copyright standards being applied to Vas, a commercial programmer, who must have assumed his obligation was to stay the right side of copyright law, not academic rules?

My guess is: the ICGA views itself as an "academic" enterprise [which has been true to at least some extent over at least part of its history, or at least David Levy thinks so ]. An additional point is that if the ICGA did use essentially a "copyright" standard, they would be more susceptible to lawsuits over their competence to adjudicate matters.

And we have plagiarism from the standpoint of Art and Culture, according to the wiki .....

I'm not sure that Art and Culture are all that relevant. It is hard for me to conceptualise that the purpose of the WCCC is all that much artistic. That's not to say that a programming competition couldn't be judged on artistic merit (the canonical example being the International Obfuscated C Code Contest) -- if they held (say) a Parody Competition for Best Re-Fashioning of a previous World Champion on modern hardware, then obviously different standards will apply.

The question of "transformative value" versus "originality" is certainly one of the interest, but the ICGA [for better or worse] seems rather firmly in the latter camp, particularly when the "transformative" process remains concealed. As you note, with the explosion of "open source code" this might want to be re-visited, though I'd still argue that "concealment" would be a major sticking point for most of the historical ICGA decisions [a partial exception being with LION++ in 2006, I suppose, though there again much of van den Herik's decision was based on the fact that they mentioned the Fruit influence in the release notes, but not explicitly on the entry form].

[BB+]

And, a side question .... why are essentially academic standards rather than copyright standards being applied to Vas, a commercial programmer, who must have assumed his obligation was to stay the right side of copyright law, not academic rules?

See also http://www.open-chess.org/viewtopic.php ... =30#p12252 and the post directly following that one. Various arguments against a "copyright" standard are proposed, such as jurisdictional differences, the ICGA desire to make the author (as opposed to a copyright holder) the principal actor, and a few others.