ags@s.cc.purdue.edu (Dave Seaman) (10/26/88)
In article <1284@buengc.BU.EDU> art@buengc.bu.edu (A. R. Thompson) writes: >Modula-II has some drawbacks. I haven't looked at it for a while, so what >follows is a bit shaky. MII defines a "word", a sixteen bit pattern of >bits. >This is unfortunate since in a very real sense it welds a bias >toward sixteen bit architectures into the language. I have crossposted to comp.lang.modula2 and redirected the followups to that group. First of all, MII does not define a "word". I believe the system module may define a word, but its size is implementation-defined. There is nothing in the MII language itself that is "biased" toward 16-bit architectures. >The biggest sin >MII commits is, however, its definition of "modules" inside of procedures. >This is unfortunate since it makes modules subject to the dynamic comings >and goings of procedures. [long tirade deleted] Only if you choose to do it that way. You can have modules inside procedures, or you can have procedures inside modules. You can even have both at once, if it pleases you. Actually, I believe the concept of "local modules" is being removed or already has been removed, though it was present in the last implementation of MII that I actually used. It seems that few people were making use of local modules. If all modules are required to be separate compilation units, then it is no longer possible to have modules inside procedures. Does that make you happy? -- Dave Seaman ags@j.cc.purdue.edu
frode@m2cs.naggum.se (Frode Odegard) (10/31/88)
In article <3609@s.cc.purdue.edu>, ags@s.cc.purdue.edu (Dave Seaman) writes: > > Actually, I believe the concept of "local modules" is being removed or > already has been removed, though it was present in the last implementation > of MII that I actually used. It seems that few people were making use of > local modules. If all modules are required to be separate compilation > units, then it is no longer possible to have modules inside procedures. > Does that make you happy? Local modules are *NOT* - repeat *NOT* being removed from Modula-2. Even Wirth's one-pass compiler supports them. The fact that Wirth doesn't have them in Oberon doesn't affect Modula-2 at all. The removal of local modules from Modula-2 would render millions of lines of Modula-2 code non-standard. One of the major goals of the ISO committee is to avoid disasters like that. - Frode
seitz@cory.Berkeley.EDU (Matthew Eric Seitz) (11/02/88)
In article <102@m2cs.naggum.se> frode@m2cs.naggum.se (Frode Odegard) writes: > >The removal of local modules from Modula-2 would render millions >of lines of Modula-2 code non-standard. One of the major goals of >the ISO committee is to avoid disasters like that. > Aren't Modula-3 and Oberon both Wirth-approved replacements for Modula-2? If so, what's the point in developing a standard for a language which its own author doesn't support? Matthew Seitz seitz@cory.berkeley.edu
nagler%olsen@CERNVAX.BITNET ("unizh.UUCP Robert Nagler") (11/03/88)
> If so, what's the point in developing a standard for a language > which its own author doesn't support? Here are my best guesses: 1) The author has gone on to new things while the people who are creating the standard are trying to tie up the loose ends that the author feels are not very important. 2) The standards people want to see "their ideas" come to fruition. (The "me too" factor seems to have a strong influence in the current Modula-2 standard effort IMHO.) 3) A de facto standard emerges (viz. UCSD Pascal and VMS FORTRAN) which is more useful (or cheaper) than the original language. 4) People think they have the backing of the author when in reality s/he is just given them lip service. Rob Nagler olsen!nagler@uunet.uu.net [Usual disclaimers apply.]
stan@dbi.UUCP (11/04/88)
In article <7021@pasteur.Berkeley.EDU> seitz@cory.berkely.edu (Matthew Seitz) wrties: > > Aren't Modula-3 and Oberon both Wirth-approved replacements for >Modula-2? If so, what's the point in developing a standard for a language >which its own author doesn't support? > Modula-3 and Oberon are still research vehicles. The initial reports describing these languages appeared in the last fourteen months. These new languages are not considered replacements for Modula-2, rather they are the results of recent research. Modula-3 was developed in jointly by Luca Cardelli, James Donahue, Lucille Glassman, Mick Jordan, Bill Kalsow, and Greg Nelson at the Systems Research Center of Digital Equipment Corporation in Palo Alto, California and the Olivetti Research Center of Ing. C. Olivetti and C., SpA in Menlo Park, California. Oberon was developed by N. Wirth at ETH Zurich. Just this week I got a copy of the M3 report from SRC. From the first sentences of the first paragraph of the first chapter: "Modula-3 descends from Mesa[8], Modula-2[12], Cedar[5], and Modula-2+[9,10]. It also resembles its cousins Object Pascal[11], Oberon[13], and Euclid[6]." (The first papers on Modula appeared in 1977 with the 3rd edition of "Programming in Modula-2" by N. Wirth appearing in 1985. Extensive use of Modula-2 started in 1982/1983 with the distribution of the RT-11 compilers and the production of many Lilith workstations.) Professor Wirth has supported the current Modula-2 standardization activity by replying to inquiries from committee members. While he and all the other programming language researchers around the world lead the way to better tools for writting programs the rest of us must make do with what is already available. To a very large degree standards are based on already accepted practice. Current users and manufacturers of Modula-2 are working to make a standard Modula-2. Perhaps by the time Modula-2 becomes an ISO standard many places will be using Modula-3 and have an interest in cooperating to develop a standard for this new object-oriented programming language. Stan Osborne, ana-systems, Foster City, California usenet: uunet!dbi!stan Phone: (415) 341-1768
poole@forty2.UUCP (Simon Poole) (11/04/88)
In article <7021@pasteur.Berkeley.EDU> seitz@cory.Berkeley.EDU.UUCP (Matthew Eric Seitz) writes: ...... > Aren't Modula-3 and Oberon both Wirth-approved replacements for >Modula-2? If so, what's the point in developing a standard for a language >which its own author doesn't support? Wirth has never `supported' any of the languages he has designed. Generally he has added or removed features to fit his current needs (see the history of Modula-2), without even thinking about what this could mean to other users/implementers. -- ---------------------------------------------------------------------------- UUCP: ...mcvax!cernvax!forty2!poole Simon Poole BITNET: K538915@CZHRZU1A ----------------------------------------------------------------------------
frode@m2cs.naggum.se (Frode Odegard) (11/04/88)
In article <7021@pasteur.Berkeley.EDU>, seitz@cory.Berkeley.EDU (Matthew Eric Seitz) writes: > In article <102@m2cs.naggum.se> frode@m2cs.naggum.se (Frode Odegard) writes: > > > >The removal of local modules from Modula-2 would render millions > >of lines of Modula-2 code non-standard. One of the major goals of > >the ISO committee is to avoid disasters like that. > > > Aren't Modula-3 and Oberon both Wirth-approved replacements for > Modula-2? If so, what's the point in developing a standard for a language > which its own author doesn't support? > [Disclaimer: I'm a member of the M-2 ISO committee] Modula-3 and Oberon are nothing more than research projects. They're nowhere near of becoming Real, Commercial languages. It's no point standardizing something which isn't widely used, and not even finished.. Modula-2 on the other hand is widely used in the Real, Commercial world. Never mind that OS/2 wasn't implemented in Modula-2, but at least some SERIOUS projects have been done using M-2. When something is widely used, it might be worth standardizing. Modula-2 was pretty stable from the beginning, and we never really cought the Pascal disease. A lot of hard cash has been invested in Modula-2, and the users found a need for an international standard. So..the reason we're standardizing Modula-2 is that there is a NEED for Modula-2 standardization. I'm also very happy about the approach we've taken with regards to defining the language. VDM is used, which opens up enormous possibilities for program-proving. - Frode
ACPS2971@RYERSON.BITNET (PATRICK WINGERT) (11/09/88)
Can someone give me information on what This VDM that the ISO commitee is using is and what it can do....A referance would also be helpful.
bowen@prg.oxford.ac.uk (Jonathan Bowen) (11/10/88)
In article <8811090902.aa06534@tgould.doc.ic.ac.uk> you write: >Can someone give me information on what This VDM that the ISO commitee is using > is and what it can do....A referance would also be helpful. VDM = Vienna Development method. It is a formal specification language and method based on predicate calculus and set theory. See "Systematic Software Development using VDM" by Prof Cliff Jones (pub Prentice-Hall) for more info. -- Jonathan Bowen Programming Research Group Oxford University
frode@m2cs.naggum.se (Frode Odegard) (11/11/88)
In article <INFO-M2%88110812332319@UCF1VM>, ACPS2971@RYERSON.BITNET (PATRICK WINGERT) writes: > Can someone give me information on what This VDM that the ISO commitee is using > is and what it can do....A referance would also be helpful. Go read "Systematic Software Development Using VDM", by Cliff Jones (Prentice-Hall). Also, you should all subscribe to the ISO/BSI Modula-2 papers. That'll give you plenty of info. Contact Real-Time Associates in LONDON (+44-1-656-7333) and talk to Steve Collins. I think it's 40 Pounds Sterling per year. - Frode PS: I'm surprised there isn't a comp.lang.vdm group! Check out comp.lang.misc for info. on VDM.
mct@PRAXIS.UUCP (Martyn Thomas) (11/11/88)
In article <INFO-M2%88110812332319@UCF1VM> you write: >Can someone give me information on what This VDM that the ISO commitee is using > is and what it can do....A referance would also be helpful. The Vienna Development Method (VDM) uses discrete mathematics to build mathematical models of the semantics of programming languages (or computer systems). It originated at the IBM Vienna Labs in the mid 70s. See: Systematic Software Development Using VDM; CB Jones; Prentice-Hall (Publisher). -- Martyn Thomas, Praxis plc, 20 Manvers Street, Bath BA1 1PX UK. Tel: +44-225-444700. Email: ...!uunet!mcvax!ukc!praxis!mct
randy@m2xenix.UUCP (Randy Bush) (11/21/88)
frode@m2cs.naggum.se (Frode Odegard) writes: >Also, you should all subscribe to the ISO/BSI Modula-2 papers. Heck of a lot of good that will do. I am supposedly a member of the BSI committee (and am chair of the US committee), and despite begging, do not receive the significant BSI working papers. [sorry to air dirty linen in public, but a year of private has been to little avail.] -- ..!uunet!oresoft!m2xenix!randy randy@dawggon.fidonet.org FidoNet 1:105/6 {..!mcvax!uunet,..!tektronix,..!sun!nosun}!oresoft!randy randy@oresoft.uu.net
frode@m2cs.naggum.se (Frode Odegard) (11/25/88)
In article <117@m2xenix.UUCP>, randy@m2xenix.UUCP (Randy Bush) writes: > frode@m2cs.naggum.se (Frode Odegard) writes: > >Also, you should all subscribe to the ISO/BSI Modula-2 papers. > > Heck of a lot of good that will do. I am supposedly a member of the BSI > committee (and am chair of the US committee), and despite begging, do not > receive the significant BSI working papers. [sorry to air dirty linen in > public, but a year of private has been to little avail.] > -- > ..!uunet!oresoft!m2xenix!randy randy@dawggon.fidonet.org FidoNet 1:105/6 > {..!mcvax!uunet,..!tektronix,..!sun!nosun}!oresoft!randy randy@oresoft.uu.net Hey, is it THAT bad? Well, I'm not a member of the BSI committee myself, so I get my copies from Real Time Associates in London as most people do. I've never had any trouble with getting BSI papers from them. - Frode