Modifying Class Object
in [Python]
|
From: Alf P. Steinbach on 11 Feb 2010 17:26 * Terry Reedy: > On 2/11/2010 1:37 AM, Alf P. Steinbach wrote: > >> Consider just the >> assert( t is not s ) >> t = s >> >> Does this change anything at all in the computer's memory? > > By 'computer', do you mean 'anything that computes' (including humans) > or specifically 'electronic computer'? In this context I mean the virtual machine that a Python language assumes. Doesn't matter if it's electronic or a pen-and-pencil simulation. >> But since it does have an effect, a memory change has been effected. > > Agreed, in whatever 'memory' the 'computer' is using. > >> You describe that memory change as that t has been "bound" to the same >> object as s. > > I prefer to use the word 'associated': namespaces are a many-to-one > association between names and objects. > >> By which you mean that henceforth, until the next assignment to t, t >> *refers* to the same object as s. > > T and s are both associated with the same object. > >> That explanation in terms of "refers" is necessary. > > I disagree > >> No beginner knows what it means that a name is "bound" to something >> means, until it's been >> explained. > > I agree, which is why I am trying to avoid 'bound', etc, in favor of > 'associated'. One problem of 'bind' is that it sometimes raises the > question of which is bound to which. 'Associated' avoids that. > >> The explanation is necessarily in terms of "refers to". > > I have given an alternative, even if you still prefer yours. "Associated" is just less precise than "refers". "Associated" is two-way. Anyway it's just a term, and if you define it to mean a one-way reference, then nothing substantial is changed except more room for misunderstanding. >> When something A refers to something B, then by definition A is a >> *reference* to B. > > I presume you agree that the name 'Alf P. Steinbach' refers to you. Do > you then consider it to be a 'reference' to you? Yes, and that's irrelevant, because you can't change a name. It's a slightly different meaning. But a name edit field with my name in it most probably refers to me. > In either case, the > Python definition uses 'refers to' in the same way that names refer to > people, and did even before names were used in electro-mechanical > computer programming. That's so subtle a distinction that it appears meaningless to me. It says "refers to" but doesn't mean "refers to"? > >Steven D'Aprano: >>> My version describes what happens at the level of high-level Python >>> code, which is the defined semantics of the language. It makes no >>> assumptions about the implementation, completely unlike yours which is >>> entirely implementation- >>> specific. My description is equally valid whether Python is being >>> executed by the CPython virtual machine on an Intel-compatible >>> processor, or hand-simulated with pencil and paper, or something >>> completely different. Yours is not. > > About 13 years ago, I noticed that electronically executable Python was > very similar to some of the designed-for-human-reading algoritm > languages (pseudocode) that were not. I then coined the oxymoron > 'executable pseudocode' for Python. I see the difference between the > descriptions as reflecting the difference between Python, the executable > algorithm language and Python, the machine programming language. > > >> I describe the high-level language, you describe one implementation. > >> Neither view is *wrong*, per se, but one describes the semantics of > >> the language while the other describes the implementation. > > I think anyone writing books using Python should at least understand the > abstract view even if he prefers to write from the more concrete view. It seems to me that you lack an understanding of the abstract here, going into imagined and not concretely discussed differences between "refers to" and "refers to". Until or if (but I think it unlikely) you can explain clearly what that difference between "refers to" and "refers to" is, it's just wordplay. Cheers & hth., - Alf
From: Martin P. Hellwig on 11 Feb 2010 18:46 <cut all> Well at least you are well written and more subtle than Xah Lee. Though I find him also quite amusing, I do like a good flame-war every now and again, and in that perspective I solute you. -- mph
From: Steven D'Aprano on 11 Feb 2010 20:09 On Thu, 11 Feb 2010 17:11:17 -0500, Terry Reedy wrote: > About 13 years ago, I noticed that electronically executable Python was > very similar to some of the designed-for-human-reading algoritm > languages (pseudocode) that were not. I then coined the oxymoron > 'executable pseudocode' for Python. That was yours? Nice one! -- Steve
From: Alf P. Steinbach on 11 Feb 2010 21:47 * Martin P. Hellwig: > <cut all> > Well at least you are well written and more subtle than Xah Lee. > Though I find him also quite amusing, I do like a good flame-war every > now and again, and in that perspective I solute you. The technical discussion is now at point where one poster maintains that references don't exist in Python, and another poster, in support, maintains that "refers to" in the language spec doesn't mean "refers to" but instead means "refers to", whatever's that meant to mean. As a technical discussion it's meaningless drivel. And as an argument in that technical discussion your allegation of trolling is just yet another fallacy, and meaningless. But in a social context, declaring support or placing oneself within a group, or for that matter "Poisoning the well", it can make sense. This group has an extraordinary high level of flaming and personal attacks, and it's the only group I've seen where there are threads (I think I've seen 3, within the last two months, not participating in them) with the regulars reitererating how "friendly" the group is, how difficult it is to get flamed here. In other groups it's not necessary for the regulars to point out how friendly the group is. But then, in other groups personal attacks are rare. And so when you mention Xah Lee I'm now wondering what is cause, and what is effect. I was informed that he'd done extensive cross-posting, and his own web site seems to confirm that his ISP at one time reacted to a complaint about such cross-posting. I've also seen directly that he has employed pretty foul language in characterizing Python in an article, and even without that language negative loaded characterizations without any evidence to back them up (like, for example, yours above) must be considered intentional flame bait. But did it start that way for Xah Lee? I'm not going to check the archives. It's enough, wrt. the point I'm making about some regulars of the group, that even with such evidence of active trolling at hand -- ISP reaction to cross-posting, needless foul language, unsubstantiated characterizations -- based on my experience here I think it's just as likely that it started out by Xah Lee being flamed, perhaps repeatedly, with personal attacks and by group action. Or perhaps it didn't, but at this point it would not surprise me in the slightest. Cheers & hth., - Alf
From: Bruno Desthuilliers on 12 Feb 2010 04:46
Alf P. Steinbach a �crit : (snip) > This group has an extraordinary high level of flaming and personal > attacks Oh my... (snip remaining non-sense) Mr Steinbach, I bet you'll count this as another "flaming" and "personal attack", but nonetheless : you might have happier time if you were able to understand the distinction between disagreement and "personal attack". (now back to more interesting readings) |