home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!darwin.sura.net!mips!swrinde!zaphod.mps.ohio-state.edu!sol.ctr.columbia.edu!ira.uka.de!chx400!sicsun!disuns2!liasun7.epfl.ch!baechler
- From: baechler@liasun7.epfl.ch (Emmanuel Baechler)
- Newsgroups: comp.databases
- Subject: Re: Looking for a USABLE database
- Message-ID: <4430@disuns2.epfl.ch>
- Date: 30 Jul 92 17:32:22 GMT
- References: <1992Jul22.180503.27010@homebase.vistachrome.com> <l6v00iINN8qf@jethro.Corp.Sun.COM> <4352@disuns2.epfl.ch> <1992Jul28.140922.8573@sdf.lonestar.org>
- Sender: news@disuns2.epfl.ch
- Organization: Ecole Polytechnique Federale de Lausanne
- Lines: 54
- Nntp-Posting-Host: liasun7.epfl.ch
-
- In article <1992Jul28.140922.8573@sdf.lonestar.org>, abledsoe@sdf.lonestar.org (Al Bledsoe) writes:
-
- |>
- |> > this is totally unacceptable: labs frequently add new tests, or add COMBINED
- |> > tests (based on existing ones), and this means the addition of new columns.
- |>
- |> Uh, have you read anything about relational database theory?
-
- I am definitely not an expert about relational database theory (my own field is
- AI) but I have read quite a bit about that. The fact is that the doctors and the
- people of the labs regulary create new analyses and want to integrate them in
- their result sheets, period. They did not hear anything about relational
- databases, they don't want to hear anything about that, they want to be able
- to enter new columns on their result sheets and these result sheets are represented by tables in most of the labs.
-
- |> I've worked in the health field here in Texas since 1975, doctors are not
- |> standard users!!
-
- Maybe yours have a different attitude, that's happy for you. A few of our doctors
- are quite educated about computers, but this is definitely not the case of the
- majority.
-
- |> The ANSI MUMPS database/language has for years and years accomplished your
- |> task in a very efficient manner. The U.S. VA laboratory system was written
- |> with MUMPS (see DEC's DSM product info). I now work for a laboratory vendor
- |> that uses PICK or UniVerse's PICKeese running on UNIX servers such as
- |> Sequent or IBM's RS/6000 or HP's 9000/857's etc.
- |>
- |> These database vendors are just now discovering the Open world as they
- |> emerge from their vertical niches. Intersystems is a MUMPS vendor that
- |> markets a product called M/SQL, hiding the name MUMPS and flouting
- |> their 4GL-like frontend. UniVerse 7.0 by Vmark is doing the same thing
- |> by embedding SQL and creating a global dictionary to the relational
- |> world.
- |>
- |> So, there is no lack for tools to create efficient and well interfaced
- |> applications for the medical community. Byte magazine May 1992 gives
- |> credibilty to the view that RDBMS's have had their chance. Combining
- |> what was learned through them with what has been demonstrated through
- |> such DBMS's as MUMPS and PICK will find new functional products from
- |> vendors that address the real world of weird users like doctors.
- |>
- |> Hint: try Advanced Revelation in a client/server design with UniVerse.
-
- Thank you very much for yourt answer and sorry for the delay of my answer.
-
- --
- Emmanuel Baechler. | Tel.: ++41-21-693-2732
- Laboratoire d'Intelligence Artificielle | e-mail: baechler@lia.di.epfl.ch
- Ecole Polytechnique Federale de Lausanne | or: baechler@liasun6.epfl.ch
- MA-Ecublens | Standard Disclaimer
- CH-1015 Lausanne Switzerland
-
- Ban the bomb. Save the world for conventional warfare.
-