Path: ...!goblin1!goblin.stu.neva.ru!news.mixmin.net!aioe.org!.POSTED.bb6wASuMjd0LOtbtueCHsA.user.gioia.aioe.org!not-for-mail From: "Apd" Newsgroups: alt.computer.workshop Subject: Re: FORGERY Date: Wed, 17 Jun 2020 22:40:26 +0100 Organization: ad hoc Lines: 171 Message-ID: References: <3498b3a9-2a15-47d2-9b60-91099ce06c8co@googlegroups.com> <1421306b-b952-492e-89f3-186d153cbff6o@googlegroups.com> NNTP-Posting-Host: bb6wASuMjd0LOtbtueCHsA.user.gioia.aioe.org X-Complaints-To: abuse@aioe.org X-MSMail-Priority: Normal X-RFC2646: Format=Flowed; Original X-Priority: 3 X-Newsreader: Microsoft Outlook Express 6.00.2900.5512 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.5512 X-Notice: Filtered by postfilter v. 0.9.2 Bytes: 8136 "Snit" wrote: > Apd wrote: >> "Snit" wrote: >>> On 6/16/20 6:08 PM, Apd wrote: >>>> You were talking about the output the flood-bot generates. You said >>>> one had to see the code. Diesel said you don't need to see the >>>> source code to "determine what the program most likely is, if the >>>> program's output can be sampled". He then gave a 'not-seeing-the- >>>> source-code' example of how a program disassembly, despite looking >>>> nothing like its source, will reveal what the program is doing. >>> >>> And example that was not relevant to the topic (assuming he did not >>> have the executable program). That is what I think happened, too... >>> he just moved from saying what he could tell about Carroll's flood >>> bot code to speaking about general methods even ones that were not >>> relevant. >> >> It was relevant in that it was an example of being able to discover >> things about code (the bot code or any other code) without having the >> original source. > > Specificity, at least in part, with having the executable to > disassemble. Sure. Diesel was quite clear on that. In reference to > the bot code I did not think he had the executable, but his comments > suggested otherwise. They didn't suggest that to me. And there's no reason to think there's an executable. It's more likely to be a script. > Why else bring up the executable in reference to Carroll's bot? As an analysis example. > I think he just went off topic and wanted to brag a bit about what he > thought I would not know. Now while I might not be able to disassemble > code I do understand the basic concept. > >>>> I would probably have used a different example to do with examining >>>> the output. However, what he wrote was all in context and there was >>>> no topic change or implication of having the flood-bot source. >>> >>> But if he was not sticking to the topic of the bot as I was, as I >>> was, then he changed the topic to not be just about the bot and its >>> code. >> >> It was relevant to the topic which was not changed. > > The topic of Carroll's flood bot code. That was what I was speaking > of. I know. > Maybe the whole misunderstanding is he missed that? Not really that > important to me but clearly it is to him. I don't believe he misunderstood. >>>> See Diesel's post: >>>> >>>> http://al.howardknight.net/?ID=158278349300 >>> >>> He split things up a bit there. Here it is with more complete >>> context: >> >> You've posted that several times already. The complete context is in >> the link I gave which is his reply to you and includes this text >> (yours) and more. > > My point if he snipped my comments and that may have led to him not > seeing > the context. I saw no snippage. He commented on all you quoted and more. >>> >>> ----- >>> I will grant that it seems like an obvious "next step" to >>> have the bot break apart sentences and respond to >>> keywords, but that is more my thing that Carroll's (I do >>> it with my chat bot). So if I were to make such a bot, >>> yes, I would want it to do that... but does Carroll even >>> want it to? I think the main purpose is Google seeding... >>> and it does that well. >>> >>> Without knowing more of the purpose we cannot say if the >>> output shows the code to be good or not. One has to see >>> the code to know that. >>> ----- >>> >>> I am CLEARLY speaking about Carroll's Usenet flood bot and that code >>> alone. No other. I am being specific. >> >> Yes, and you are saying the code rather than the output has to be >> seen in order to say if the code is up to scratch. > > Close. One needs more then JUST the output for that specific code. One > must know the goals. If one does not know the goals one cannot know if > the output reaches those goals or how well or how poorly it does so. > > But, sure, if you know the goals you can speak to how well it works to > match them. And if you have the code or even the executable you can > learn more. I think we all agree on that. Maybe not. I'll accept knowing the goal(s) is one consideration. > But the focus by me was solely on Carroll's flood bot. With that we > can infer the goals: > > * Google seeding > * Carroll playing victim > * Carroll trolling me > * Carroll manipulating others to argue with me > * Carroll controlling conversations and pushing discord. Substitute "someone" for "Carroll". > I listed others elsewhere. And on those I think his bot is rather > successful. If those are the goals then it works. > > If the goals are otherwise then perhaps it does not. > >>> Diesel responded with (in part) -- the post you pointed to: >> >> Before that, and in response to you saying the code has to be seen, >> he wrote: > > Wait: what quote where I said the code had to be seen? The quoted text in this very post where you said: "One has to see the code to know that". >> "One doesn't have to see original source code to be able to determine >> what the program most likely is, if the programs output can be >> sampled". >> >> Then he provided the example of not seeing the source... > > Where he spoke of disassembling the code... with the context being in > response to me speaking of Carroll's flood bot code. > > One cannot disassemble code one does not have the executable for, as > far as I know. Correct. It was an example of seeing how something works without having the original source code. >>> >>> ----- >>> Do you think when you disassemble something that you're >>> provided the original source code that was >>> compiled/assembled by the author? You aren't, what you're >>> given looks nothing like the original source code, but it >>> still tells you *everything* about the program. >>> ----- >> >> ...which is an example of determining what the program does without >> having the source code. > > Specifically in response to Carroll's code. And disassembly. Just as an example of analysis. >> The topic is the same and what he says is a response to you saying >> the code has to be seen. > > Remember I was speaking of Carroll's bot code. Nothing else. Yep, but he addressed your point about needing to see the code. How well he addressed it is another matter.