| Deutsch English Français Italiano |
|
<100a6d9$e80n$1@dont-email.me> View for Bookmarking (what is this?) Look up another Usenet article |
Path: ...!eternal-september.org!feeder3.eternal-september.org!news.eternal-september.org!eternal-september.org!.POSTED!not-for-mail From: Mike Terry <news.dead.person.stones@darjeeling.plus.com> Newsgroups: comp.theory Subject: Re: How the requirements that Professor Sipser agreed to are exactly met --- WDH Date: Sat, 17 May 2025 15:27:52 +0100 Organization: A noiseless patient Spider Lines: 108 Message-ID: <100a6d9$e80n$1@dont-email.me> References: <vvte01$14pca$29@dont-email.me> <fceb852a146ff7238c5be7a0adf420474a8fb5df@i2pn2.org> <vvuc7a$1deu5$5@dont-email.me> <c5a47349d8625838f1ee2782c216e0ebf9223bc6@i2pn2.org> <vvuj6l$1j6s0$3@dont-email.me> <b78af2e0b52f178683b672b45ba1bc2012023aaf@i2pn2.org> <1000dlc$21dtc$5@dont-email.me> <1000qdb$24gr3$4@dont-email.me> <1000rir$24jh0$3@dont-email.me> <1000rqc$24gr3$7@dont-email.me> <1000son$24sr2$3@dont-email.me> <7947826fb84c9c8db49c392b305d395c3669907f@i2pn2.org> <1002dre$2i4bk$14@dont-email.me> <1002vp2$2mbr6$3@dont-email.me> <10030c3$2mivc$3@dont-email.me> <87h61mang3.fsf@nosuchdomain.example.com> <1003cu5$2p3g1$1@dont-email.me> <10070cl$3mmus$1@dont-email.me> <1007j6b$3qb7l$2@dont-email.me> <1009iu4$agi7$1@dont-email.me> MIME-Version: 1.0 Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 8bit Injection-Date: Sat, 17 May 2025 16:27:53 +0200 (CEST) Injection-Info: dont-email.me; posting-host="7e4f67ee19a027ba8df965182a825f7b"; logging-data="466967"; mail-complaints-to="abuse@eternal-september.org"; posting-account="U2FsdGVkX19xDhYj9BUt80iMeIffkTJDscJpRq10m7U=" User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:91.0) Gecko/20100101 Firefox/91.0 SeaMonkey/2.53.18.2 Cancel-Lock: sha1:vZpQmPnyA0mhX1ef8tEWvjLkVCA= In-Reply-To: <1009iu4$agi7$1@dont-email.me> Bytes: 7161 On 17/05/2025 09:55, Mikko wrote: > On 2025-05-16 14:47:39 +0000, olcott said: > >> On 5/16/2025 4:26 AM, Mikko wrote: >>> On 2025-05-15 00:36:21 +0000, Mike Terry said: >>> >>>> On 14/05/2025 22:31, Keith Thompson wrote: >>>>> olcott <polcott333@gmail.com> writes: >>>>>> On 5/14/2025 3:51 PM, dbush wrote: >>>>>>> On 5/14/2025 11:45 AM, olcott wrote: >>>>>>>> On 5/14/2025 6:20 AM, Richard Damon wrote: >>>>>>>>> >>>>>>>>> And since the DD that HHH is simulating WILL HALT when fully >>>>>>>>> simulated (an action that HHH doesn't do) >>>>>>>> >>>>>>>> *NOT IN THE ACTUAL SPEC* >>>>>>>> <MIT Professor Sipser agreed to ONLY these verbatim words 10/13/2022> >>>>>>>> ���� If simulating halt decider H correctly simulates its >>>>>>>> ���� input D until H correctly determines that its simulated D >>>>>>>> ���� would never stop running unless aborted then >>>>>>>> >>>>>>> That Sipser didn't agree what you think the above means: >>>>>>> >>>>>> >>>>>> If that was actually true then you could provide an >>>>>> alternative meaning for the exact words stated above. >>>>>> >>>>>> I keep challenging you to provide this alternative >>>>>> meaning and you dodge because you know that you are >>>>>> lying about there being any alternative meaning >>>>>> FOR THE EXACT WORDS LISTED ABOVE. >>>>> >>>>> No alternative meaning is needed, just a correct interpretation of the >>>>> words (which appear to be incomplete). >>>>> >>>>> The quoted sentence is cut off, something that I suspect you didn't >>>>> notice.� Here's the full quotation from a previous article: >>>>> >>>>>>> <Sipser approved abstract> >>>>>>> MIT Professor Michael Sipser has agreed that the following verbatim >>>>>>> paragraph is correct (he has not agreed to anything else in this >>>>>>> paper): >>>>>>> >>>>>>> If simulating halt decider H correctly simulates its input D until H >>>>>>> correctly determines that its simulated D would never stop running >>>>>>> unless aborted then H can abort its simulation of D and correctly >>>>>>> report that D specifies a non-halting sequence of configurations. >>>>>>> </Sipser approved abstract> >>>>> >>>>> **If** H correctly simulates its input in the manner you claim, >>>>> **then** H can correctly report the halting status of D.� (That's a >>>>> paraphrase that probably doesn't capture the full meaning; the full >>>>> **quotation is above.) >>>>> >>>>> To put it another way, If H correctly simulated its input in >>>>> the manner you claim, then H could correctly report the halting >>>>> status of D. >>>>> >>>>> I'm not surprised that Sipser would agree to that.� The problem is >>>>> that it's a conditional statement whose premise is impossible. >>>>> >>>>> If an equilateral triangle had four sides, then each of its four >>>>> vertices would be 90 degrees.� That doesn't actually mean that >>>>> there exists an equilateral triangle with four 90-degree vertices, >>>>> and in fact no such triangle exists.� Similarly, *if* a general >>>>> halt decider existed, then there are a lot of things we could say >>>>> about it -- but no general halt decider can exist. >>>>> >>>>> I'm not quite 100% confident in my reasoning here.� I invite any >>>>> actual experts in computational theory (not you, PO) to criticize >>>>> what I've written. >>>> >>>> I doubt that Sipser would be using your interpretation, relying on a false premise as a clever >>>> kind of logical loop-hole to basically fob someone off. >>> >>> The details of H are not known to Sipser, so he can't know whether a >>> premise is false. It is possible that some simulating partial decider >>> correctly simulates a part of the behaviour of some D and correctly >>> determines that the unsimulated part of the behaviour never halts; >>> for example, if the unsimulated part is a trivial eternal loop. That >>> one premise is false about HHH with DDD is a part of what was asked. >> >> Mike explains all of the details of exactly how a >> correct Simulating Halt Decider is derived from >> the exact meaning of the words that professor Sipser >> agreed to IN THE PART THAT YOU IGNORED > > No, he does not. He does not even believe that it is possible to derive > a correct Simulating Halt Decider form the exact meaning of any words. > That's correct. We could build a correct /partial/ SHD though, which I explained. The idea behind an PSHD is ok, and a class of HP inputs could be correctly decided with a PSHD. Obviously a PSHD H could not decide its corresponding H^ input, as the Linz HP proof implies. Since PO's HHH /does/ decide its corresponding DD (incorrectly), it is not a PSHD, since PSHDs are not allowed to decide incorrectly. [A correctly coded PSHD HHH would never halt when given its (HHH^,HHH^) input. PO's problem is that he misunderstands the entire context of Sipser's words. Sipser's words concern how a PSHD H could decide some FIXED INPUT D it has been given. PO wants to interpret them as what happens when H is modified, and D is also modified to reference the new H. So he's modifying what is supposed to be a fixed input half way through his interpretation. Sipser would be holding his head in his hands if he knew (and cared) ... :) Mike.