From mboxrd@z Thu Jan 1 00:00:00 1970 X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me X-Spam-Level: X-Spam-Status: No, score=-1.0 required=5.0 tests=BAYES_00,FORGED_HOTMAIL_RCVD2, FREEMAIL_FROM autolearn=no autolearn_force=no version=3.4.4 X-Google-Thread: 103376,e0c23e7a19a435c4 X-Google-NewGroupId: yes X-Google-Attributes: gida07f3367d7,domainid0,public,usenet X-Google-Language: ENGLISH,CP1252 Received: by 10.68.219.170 with SMTP id pp10mr8136333pbc.1.1340478478764; Sat, 23 Jun 2012 12:07:58 -0700 (PDT) Path: l9ni11664pbj.0!nntp.google.com!news2.google.com!postnews.google.com!glegroupsg2000goo.googlegroups.com!not-for-mail From: Austin Obyrne Newsgroups: comp.lang.ada Subject: =?windows-1252?Q?Re=3A_Recapping_on_=93Bug_Sort=94=2E?= Date: Sat, 23 Jun 2012 12:07:58 -0700 (PDT) Organization: http://groups.google.com Message-ID: References: <169bdbcb-cb43-4db9-9d48-3be2a88473eb@googlegroups.com> <77963856-3a25-4477-9510-769df7a9b85c@googlegroups.com> NNTP-Posting-Host: 31.52.108.135 Mime-Version: 1.0 X-Trace: posting.google.com 1340478478 31422 127.0.0.1 (23 Jun 2012 19:07:58 GMT) X-Complaints-To: groups-abuse@google.com NNTP-Posting-Date: Sat, 23 Jun 2012 19:07:58 +0000 (UTC) In-Reply-To: Complaints-To: groups-abuse@google.com Injection-Info: glegroupsg2000goo.googlegroups.com; posting-host=31.52.108.135; posting-account=pmkN8QoAAAAtIhXRUfydb0SCISnwaeyg User-Agent: G2/1.0 Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: quoted-printable Date: 2012-06-23T12:07:58-07:00 List-Id: On Saturday, June 23, 2012 7:05:56 PM UTC+1, Niklas Holsti wrote: > On 12-06-23 13:20 , Austin Obyrne wrote: > > On Saturday, June 23, 2012 8:54:52 AM UTC+1, Austin Obyrne wrote: > >> On Friday, June 22, 2012 9:45:53 PM UTC+1, Jeffrey Carter wrote: > >>> On 06/22/2012 12:55 PM, Austin Obyrne wrote: > >>>> > >>>> I have been told that my program resembles a known sort program call= ed > >>>> =93Counting Sort=94. I would hate to be guilty of plagiarism and I = would like to > >>>> point out therefore that the salient thing about my =93Parallel Sort= =94 is that > >>>> my implementation is geared to capturing data during any unrelated p= rogram > >>>> run-time and assigning the data in such a way that the separate elem= ents > >>>> index their own addresses in the sorting arrays. A similarity with = some > >>>> other existing paper algorithm is simply fortuitous. > >>> > >>> What you have presented is an implementation of counting sort, nothin= g more. > >>> There is nothing new or unique about your implementation. >=20 > Austin, >=20 > While I agree with Jeff that you have rediscovered Counting Sort, this=20 > does not mean that you are being accused of plagiarism. It is common for= =20 > programmers to rediscover algorithms that are basically simple, but very= =20 > good for some special cases -- and perhaps for just these reasons are=20 > not very well known. >=20 > I could imagine posing this sorting problem (sorting a dense set of=20 > numbers in a known, not too wide range) in a class on programming (not=20 > an advanced class, either) and would expect some of the students to use= =20 > this method (i.e. Counting Sort) in their solutions. >=20 > > Update on performance. > > > > 42750 seven-digit positive integers were sorted in between 1 and 2 seco= nds. > > > > Waiting to hear regarding "Count Sort". >=20 > Since your method is an implementation of Counting Sort, you are=20 > actually measuring "Counting Sort", at least in one implementation. >=20 > --=20 > Niklas Holsti > Tidorum Ltd > niklas holsti tidorum fi > . @ . Thanks for that. The very salient thing that everybody is missing is the way the data is col= lected and sorted simultaneously in "Parallel Sort" compared to Count Sort.= Parallel Sort is more a piece of computer science being implemeted in the= Ada programming language. Frankly I am not familiar with Count Sort in practise (being a comparison p= rogram puts it beyond the pale to me ) but can I ask you one question that = will make my point more clearly. If I asked you to demonstrate "Count Sort" being used to sort a specimen se= t of data how would you go about making the data available to the program. Somewhere along the line you would have to prepare an external batch file f= or reading in or you would key in the data interactively - it has to be lik= e that all the time with Count Sort - herein lies the difference - my progr= am captures the data in parallel at the very origin ie as it is computed - = no double handling =3D> huge time and labour saving. The counting part is trivial. It is the capturing concept that is the impo= rtant difference between what I am calling Parallel Sort and what anyone el= se may call Count Sort . The counting is almost benign when the elements o= f data index their own addresses in an array in both cases. =20 There is only a very tenuous connection in this fact however, plus that bot= h sort programs eschew all comparisom methods and simply count the data acc= ording to magnitude as the means of sorting it. This common factor is not s= ufficient in itself however in my view to be grounds for any one to say tha= t they are largely related versions of the same program. They have only a = very small amount in common. It is a huge leap from my parallel program to = the count sort program in doing this - they have far too little in common t= o be lumped together like this. I await your answer on how is the data introduced to a sort program that yo= u would use at home say. Regards, Austin.