X-Received: by 10.224.165.130 with SMTP id i2mr5127714qay.2.1367782355670; Sun, 05 May 2013 12:32:35 -0700 (PDT) X-Received: by 10.49.41.1 with SMTP id b1mr1350396qel.18.1367782355626; Sun, 05 May 2013 12:32:35 -0700 (PDT) Path: news.cambrium.nl!textnews.cambrium.nl!feeder3.cambriumusenet.nl!feed.tweaknews.nl!85.12.40.138.MISMATCH!xlned.com!feeder5.xlned.com!news2.euro.net!news.glorb.com!l3no2450379qak.0!news-out.google.com!y6ni0qax.0!nntp.google.com!m7no2479473qam.0!postnews.google.com!glegroupsg2000goo.googlegroups.com!not-for-mail Newsgroups: comp.databases.theory Date: Sun, 5 May 2013 12:32:35 -0700 (PDT) In-Reply-To: <51861409$0$6057$e4fe514c@dreader36.news.xs4all.nl> Complaints-To: groups-abuse@google.com Injection-Info: glegroupsg2000goo.googlegroups.com; posting-host=77.109.123.188; posting-account=-nQufgoAAABsreOCZNqo2Uyh8O-fYVPT NNTP-Posting-Host: 77.109.123.188 References: <992824d5-5a8a-4a63-906b-7d3eff2a0b47@googlegroups.com> <51445299$0$582$e4fe514c@dreader34.news.xs4all.nl> <3bd262e6-f9b2-4861-8327-fe31af438cd3@googlegroups.com> <51803a2d$0$6337$e4fe514c@dreader35.news.xs4all.nl> <279c7cc9-3e6f-490f-a692-af98fc134fc3@googlegroups.com> <36bb212d-47ae-4cb0-bf2b-a926ff2173d4@googlegroups.com> <54934075-6dd1-4f96-9c6f-50db406a7586@googlegroups.com> <518268bb$0$6086$e4fe514c@dreader36.news.xs4all.nl> <51861409$0$6057$e4fe514c@dreader36.news.xs4all.nl> User-Agent: G2/1.0 MIME-Version: 1.0 Message-ID: <1bfb1fe5-85f2-4cc9-af50-a88a667b601e@googlegroups.com> Subject: Re: How to normalize this? From: Erwin Injection-Date: Sun, 05 May 2013 19:32:35 +0000 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Xref: news.cambrium.nl Op zondag 5 mei 2013 10:10:49 UTC+2 schreef Jan Hidders het volgende: > On 2013-05-04 23:32:18 +0000, ... said: >=20 >=20 >=20 > > On Thursday, May 2, 2013 6:23:07 AM UTC-7, Jan Hidders wrote: >=20 > >> But most> textbooks I know are correct and clear on this and will stat= e=20 >=20 > >> that the> goal is for example a lossless-join and=20 >=20 > >> dependency-preserving> decomposition. >=20 > >=20 >=20 > > Can you name some? Because I don't think books are clear about=20 >=20 > > normalization within design, which involves generated predicates and=20 >=20 > > constraint preservation. Normalization per se assumes the generated=20 >=20 > > components are projections of the original, ie are suitably=20 >=20 > > constrained, without addressing the management of those constraints (eg= =20 >=20 > > introducing FKs for lost FDs). >=20 > >=20 >=20 > >> 1. Apply the 3NF normalization procedure we discussed earlier >=20 > >> 2. Include with each component only the FDs that generated that compon= ent >=20 > >=20 >=20 > > It's not clear to me what you are describing or how it relates to=20 >=20 > > arguing against stopping at 3nf. >=20 >=20 >=20 > I've described to you a normalization algorithm that seems to produce a= =20 >=20 > schema in BCNF that is also lossless-join and dependency preserving,=20 >=20 > but also has no inter-component dependencies. That's not supposed to be= =20 >=20 > possible according to the usual normalization explanation. So what went= =20 >=20 > wrong here? >=20 >=20 >=20 > -- Jan Hidders That's simply not true is it ? You are referring to the three-table decomposition, right ? That one does have the inter-component rule that R1 JOIN R2 =3D=3D=3D R1 JO= IN R3 (as I pointed out), no ? "What went wrong" imo is that your decomposition did not match the original= "spirit" of the operation. In which relation schemas get split in _exactl= y_ two, with the dependent attributes of the "small" FD (bc->e) moved _out = of_ the schema. your three-way split is either not a split in exactly two,= or it is two subsequent splits-in-two where the first does _not_ move any = attribute "out of the schema" (precisely because it is still needed for the= second split).