From omg-list-errors@omg.org Fri Nov 7 08:32:55 2003 Subject: Re: MDA / Martin Fowler [What is a platform, anyway?] From: Allan Kennedy To: Axel Uhl , CC: Joaquin Miller , Edwin Seidewitz , , , "Stephen J. Mellor" , Andrew Watson , Richard Mark Soley , , "'Martin Fowler'" Hi Axel Most of the comment since my response to Clive Menhinick's original query has been to emphasise that "Platform Independence" is a relative quality exhibited by models/programs. I think that is understood by all including Martin. However Martin's (entirely reasonable) beef appears to be simply that the term is gratuitously misleading since it implies an absolute - dependency on no platform - when it is perfectly clear that the highest degree of platform independence is achieved when a model nonetheless exhibits dependence on just one platform: *a* UML virtual machine. (I will be able to say *the* UML virtual machine when it is standardised ;-). For the record I have attached a copy of the draft MDA Reference Model worked on by you and other participants in the ORMSC. For those who are interested, look at the right hand side of the diagram where the various kinds of dependency that can diminish the "platform independence" of a model are represented. The reference model thus attempts to make explicit the intended relativity of the term "Platform Independent". Allan ******LATEST NEWS******** For full press release on SIAP win go to:- http://www.kc.com/press/press_siap.html ****STOP PRESS***** "Model Driven Architecture with Executable UML" by Kennedy Carter Ltd Due for release in January 2004. Read a review at:- http://titles.cambridge.org/catalogue.asp?isbn=0521537711 ================================================ Kennedy Carter Ltd Hatchlands East Clandon Surrey GU4 7RT U.K. Kennedy Carter: Telephone: +44 1483 226 180 Fax: +44 1483 226 199 iUML Support Telephone: +44 1483 226 190 E-mail: support@kc.com web site: www.kc.com Office Location: http://www.theheritagetrail.co.uk/stately%20homes/hatchlands.htm THE SOLUTION FOR OMG MODEL DRIVEN ARCHITECTURE ================================================ The terms "UML" and "iUML" refer respectively to the Unified Modelling Language and Kennedy Carter's MDA product suite and are used with the consent of the Object Management Group. E3-I: This message has been scanned for viruses and dangerous content by UML's antivirus scanning services. --B_3151056300_459003 From omg-list-errors@omg.org Fri Nov 7 22:40:37 2003 Subject: Metadata... (was RE: MDA / Martin Fowler [What is a platform, anyway?]) Date: Fri, 7 Nov 2003 21:31:04 -0600 Thread-Topic: MDA / Martin Fowler [What is a platform, anyway?] Thread-Index: AcOks0SzrrvOGnMgSqO4e1lXFrxB3AA7vqUg From: "Betz, Charlie" To: "Edwin Seidewitz" , ormsc@omg.org, adtf@omg.org cc: clive@trireme.com, "Stephen J. Mellor" , "Andrew Watson" , "Richard Mark Soley" , kc_staff@kc.com, "Martin Fowler" , "Allan Kennedy" X-OriginalArrivalTime: 08 Nov 2003 03:31:27.0092 (UTC) FILETIME=[CABC1B40:01C3A5A8] X-WSS-ID: 13B2BBE3677635-14-01 Content-Type: multipart/alternative; boundary="----_=_NextPart_001_01C3A5A8.BD42F00F" Status: RO This is a multi-part message in MIME format. ------_=_NextPart_001_01C3A5A8.BD42F00F Content-Type: text/plain Content-Transfer-Encoding: quoted-printable X-NAIMIME-Disclaimer: 1 X-NAIMIME-Modified: 1 $0.0002: =20 This practitioner does not anticipate a rapid uptake for the CASE round I= I aspects of MDA in his enterprise IT shop.=20 =20 This practitioner *is* presently deriving considerable value from OMG sta= ndards that help *describe* the large-scale IT problem domain, and is cha= llenging vendors playing in the internal IT tools space to start thinking= about standards interoperability for purposes of application portfolio/c= onfiguration/metadata management (ITSM). Business process modeling may li= kely be the next area of serious focus, and to the extent we buy into BP = vendors who in turn leverage OMG standards, then we may find ourselves "d= oing" MDA.=20 =20 But UML to code isn't on my radar right now. I'm spending far more time w= ith component and deployment metamodels and working on how to integrate t= hem with CWM and IT financials.=20 =20 With all due respect to Martin, his brand of pure software development is= something I'm seeing less and less of in enterprise IT, with the continu= ing drive to buy not build. Integration and extension of packages are big= themes, as is increasing the rationality of IT management (ITSM again). = Standard metadata has a big story in all of these areas, also in increasi= ng the effectiveness of outsourcing and offshoring (sorry, don't shoot th= e messenger).=20 =20 This makes this whole MDA argument a little irrelevant to practitioners i= n my context (and I realize that along with Dave Smith I'm a minority). B= ut the OMG standards are very relevant. I just wish that more attention w= ere being paid to synching up with DMTF, DCML, ITSM, TMF, and that whole = side of the world. For the record (as mere auditing members) we support m= oving ahead with the Software Portfolio Management Facility. =20 Regards, =20 Charlie www.erp4it.com =20 PS. Several of us spent some quality time with the UML 2 deployment metam= odel today and decided we like it pretty well; it seems simple and elegan= t (although there are some subtleties we're still trying to get our heads= around). -----Original Message----- From: Edwin Seidewitz [mailto:eseidewitz@intelidata.com] Sent: Thursday, November 06, 2003 4:08 PM To: ormsc@omg.org; adtf@omg.org Cc: clive@trireme.com; Stephen J. Mellor; Andrew Watson; Richard Mark Sol= ey; kc_staff@kc.com; 'Martin Fowler'; Allan Kennedy Subject: RE: MDA / Martin Fowler [What is a platform, anyway?] I, too, think the term "platform-independent model" is causing some confu= sion.=20 As has already been pointed out by a couple of respondents on this thread= , "platform independence" must be considered as a _relative_, not an _abs= olute_, term. What I want to point out is that it _is_ defined as a relat= ive term in MDA, and its use in an absolute sense is, in my opinion, a mi= sunderstanding of MDA (perhaps a reasonable misunderstanding, given the c= onnotation of the term itself). As to how this, and other terms, are defined in MDA, I turn to the "MDA G= uide" (publicly available from the OMG Web site at < http://www.omg.org/d= ocs/omg/03-06-01.pdf>). I actually had not taken a good look at the lates= t version of this document (even though I am listed as one of the many co= ntributors) until prompted to do so by this thread, and I am impressed wi= th the comprehensive presentation of basic MDA concepts that it gives (fo= r which a good deal of credit is due to the editors, Joaquin Miller and J= ishnu Murkerji). I will pull a few key quotations from this document, but= I encourage anyone who is interested in MDA to look through the document= itself, if they haven't done so already. The MDA Guide (Section 2.2.7) defines a platform as follows:=20 "A platform is a set of subsystems and technologies that provide a cohere= nt set of=20 functionality through interfaces and specified usage patterns, which any = application=20 supported by that platform can use without concern for the details of how= the=20 functionality provided by the platform is implemented."=20 Some examples given include "generic platform types" such as object, batc= h and dataflow; "technology specific platform types" such as CORBA and J2= EE; and "vendor specific platform types" such as Iona ORBIX and BEA WebLo= gic. Of particular interest, the Guide (Section 2.2.9) provides a specific def= inition of "platform independence":=20 "Platform independence is a quality, which a model may exhibit. This is t= he quality that=20 the model is independent of the features of a platform of any particular = type.=20 Like most qualities, platform independence is a matter of degree. So, one= model might=20 only assume availability of features of a very general type of platform, = such as remote=20 invocation, while another model might assume the availability a particula= r set of tools=20 for the CORBA platform. Likewise, one model might assume the availability= of one=20 feature of a particular type of platform, while another model might be fu= lly committed=20 to that type of platform."=20 For "platform independent model" (PIM), the Guide (Section 2.2.12) then s= tates that "A PIM exhibits a specified degree of platform independence so= as to be suitable for use with a number of different platforms of simila= r type." This is as opposed to a "platform specific model" (PSM), that "c= ombines the specifications in the PIM with the details that specify how t= hat system uses a particular type of platform" (Section 2.2.13). Importantly, a PIM in one context may be a PSM in another context. (Secti= on 6 of the MDA Guide has a good discussion of this.) Thus, it is perfect= ly consistent with the term "platform independent", as defined for MDA, t= o say that, say, a J2EE program is independent of vendor, operating syste= m and hardware platforms, while a UML model of the same application is in= dependent of the J2EE technology platform. In one context, the Java progr= am can, itself, be considered a PIM, while in the other it is a PSM. The question then is, of course, whether it is _useful_ to have an applic= ation model that is independent of the choice of the technology platform = at the level of J2EE. But, MDA aside, many people find it useful to do mo= dels, and many of those models are, indeed, technology-platform independe= nt. However, as has been often noted, it can be very difficult to manuall= y keep such models in sync with the application code as development conti= nues -- leading to the use of models as "sketches" rather than as primary= artifacts. The basic premise of MDA is that it should be possible to achieve greater= power for application development by being able to further elevate the l= evel of PIM that can be a primary artifact. The idea is simply to automat= e the transformation of a PIM, independent relative to one or more platfo= rms, into one or more PSMs, specific to each of those platforms. As has a= lso been mentioned previously in this thread, this is very analogous to t= he transformation from a source program to the machine code for a specifi= c operating system and hardware platform that is implemented in a Java co= mpiler. A PIM that is, say, technology-platform independent can then beco= me a primary development artifact -- just as, in normal practice, source = code has become the primary development artifact, even though it is the m= achine code that "really" runs. One can, of course, question whether the fundamental premise of MDA is co= rrect or even possible. But this is a separate (though important) discuss= ion from whether the very concept of MDA is well conceived. And, note als= o, that none of the above discussion mentions UML at all. It is yet anoth= er (very important) discussion as to whether UML, as currently specified = by OMG, is a "good", "universal" modeling language for MDA. I, of course, have a lot I could say (and have said) as part of this latt= er discussion on UML. But this message is already too long, and my commen= ts on the current state of UML are, of necessity, more personal and subje= ctive then the previous discussion. So I will save them for another messa= ge. -- Ed=20 Ed Seidewitz, Chief Architect=20 InteliData Technologies=20 < mailto:eseidewitz@intelidata.com>=20 < http://www.intelidata.com>=20 Office: +1.703.259.3076=20 Mobile: +1.301.455.3681=20 Fax: +1.703.259.3100=20 ------_=_NextPart_001_01C3A5A8.BD42F00F-- From omg-list-errors@omg.org Fri Nov 7 16:15:31 2003 Received: from ws3.uml.edu (ws3.uml.edu [129.63.1.203]) by earth.cs.uml.edu (8.11.6/8.11.6) with SMTP id hA7LFPW32278 for ; Fri, 7 Nov 2003 16:15:25 -0500 Received: from amethyst.omg.org(192.67.184.64) by ws3.uml.edu via csmap id 5196; Fri, 07 Nov 2003 15:21:58 -0500 (EST) Received: from hobbit.omg.org (hobbit.omg.org [192.67.184.3]) by amethyst.omg.org (8.12.8/8.12.8) with ESMTP id hA7M9fNA021387; Fri, 7 Nov 2003 17:09:41 -0500 Received: from dribble.mel.cme.nist.gov [129.6.71.12] by hobbit.omg.org asmtp(3.0c) id 28591; Fri, 07 Nov 2003 16:19:42 -0500 (EST) Received: from nist.gov (polonius.msid.cme.nist.gov [129.6.77.120]) by dribble.mel.cme.nist.gov (8.12.10/8.12.10) with ESMTP id hA7L9rCk012746; Fri, 7 Nov 2003 16:09:54 -0500 (EST) Message-ID: <3FAC0A20.8020206@nist.gov> Date: Fri, 07 Nov 2003 16:09:52 -0500 From: Ed Barkmeyer Reply-To: edbark@nist.gov Organization: NIST User-Agent: Mozilla/5.0 (Windows; U; WinNT4.0; en-US; rv:1.0.1) Gecko/20020823 Netscape/7.0 X-Accept-Language: en, fr, de, pdf, it, nl, sv, es MIME-Version: 1.0 To: philippe.desfray@softeam.fr CC: ormsc@omg.org, OMG ADPTF Subject: Re: MDA / Martin Fowler [What is a platform, anyway?] References: <002401c3a544$ae944e60$100a4b0a@PHD> Content-Type: text/plain Content-Transfer-Encoding: 7bit X-NAIMIME-Disclaimer: 1 X-NAIMIME-Modified: 1 Status: RO All, I've seen some important and correct observations go by, largely without challenge, and I wonder what will happen to the MDA documents to reflect them: 1. Joaquin Miller wrote: > i certainly do not consider MDA to be a platform. It is an approach. MDA is an "architecture" only for OMG standards. It is not a "software architecture" and it most certainly is not a "platform". I agree fully with the statement that it is an *approach* to software systems engineering. And it is a good one. And if we advertise it as something else, we will confuse ourselves and lose much of our target audience. 2. Axel Uhl wrote: > UML VM is just another platform that is in no way special. YES! I know that the UML VM is very dear to Steve and Allan, and I fully agree that it is an important part of the semantics of a UML model. But at this moment is not clear what parts of UML are suitable for what aspects of what kinds of models that are consistent with the MDA approach. In its efforts to be all things to all men, and to use the same diagrams at 3? different levels of abstraction in a system specification, UML cannot have semantics that are simultaneously useful for and consistent with all three levels of abstraction. UML VM is a model of a kind of platform at a specific level of engineering abstraction. 3. Bran Selic wrote: > Most people DO interpret the term "platform independence" in its absolute > simple-minded sense. Phillipe Desfray wrote: > I think that PIM means Platform Independant Model, and that this has only one > meaning : a model which does not AT ALL depends on platforms. > That kind of animal exist though : a conceptual model, a business process > model, many models at preliminary analysis stages are and should be "platform > independant". I agree. This is the expected parse of the term in the English language. It is fine to invent our own language, but then we shouldn't expect anyone else (like our would-be customers and converts) to understand it. If this is not what we mean by "PIM", then we should discard the term. Alternatively, we could identify the (one) kind of platform we had in mind (and understand) and say we only mean independent of platforms of that kind. Does the word "middleware platform" suggest itself? Not to those who are trying to reach a large audience? What if we just use "PIM" as a term? We can say it means "abstracted engineering model", but historically it was an acronym, derived from the Marketese for "a simple matter of programming". 4. Bran wrote: > I have written a number of articles, given numerous talks, and even helped > produce an OMG standard that deal with the fundamental importance of > understanding the capabilities and capacities of your platforms during > software design. That is, "platform-independence" is not even a good thing, except in a very narrow sense. > If we really want to capture what MDA is about, it is about increasing the > levels of abstraction and automation. That should serve as the main selling > point rather than a catchy-but-misleading marketing term such as "platform > independence". Absolutely. I was going to expand on this, but John Hogg did it first: 5. John Hogg wrote: > Another point we should realize is that transformations from PIMs to PSMs are > not a single chain in the general case. Any system is composed of a number of > technology choices--possibly a large number, and possibly varying in different > places. > The PIM->PSM transformation graph is therefore a lattice. At each level in > the lattice, a number of technology decisions are possible; the overall > ordering between technology decisions is only partial, although there are > typically numerous dependencies. > It's crucial that we properly represent these concepts because they represent > the reality of software and system development. If the model doesn't properly > capture them, we'll change the model, not reality. If I may be so bold, I would change the term PIM above to "conceptual model", and the term PSM to "implementation specification", and everything John says holds. There are n "levels" of engineering decisions between the conceptual specification of a system and the implementation specification. Each level in which some engineering decisions are made (and documented) while others aren't is an "abstraction" of the "levels" below it. UML is useful for representing certain kinds of engineering decisions at some of those intermediate "levels", and totally unable to represent or distinguish others. Moreover, the interpretation of a UML model may depend very much on the engineering decisions that were made one or more levels above it. So we have to be very careful and very clear in two ways: - what kinds of engineering decisions we think MDA addresses - what kinds of engineering decisions can be captured in UML Then we can define a "PIM" as embodying a certain set of decisions and excluding others. I think Richard Soley had a pretty clear set of engineering decisions in mind when he started the MDA activity, and the abstraction of those did seem to match the capabilities of UML. And Domain folk (and some others) who were listening to that message have found useful levels of abstraction and are making useful OMG "MDA-style" specifications. Maybe we could learn from them? Phillipe wrote: > Otherwise, negative minded people (is there an intersection with simple minded > people?) will think that MDA is just a marketing buzzword full of nonsense > definitions. I must be in the intersection. -Ed "The analogy that comes to mind is trying to build a house with jello. When the problem is finally identified, the solution is to use only frozen jello. I keep waiting for somebody to move in." -- Ken Fox -- Edward J. Barkmeyer Email: edbark@nist.gov National Institute of Standards & Technology Manufacturing Systems Integration Division 100 Bureau Drive, Mail Stop 8260 Tel: +1 301-975-3528 Gaithersburg, MD 20899-8260 FAX: +1 301-975-4694 "The opinions expressed above do not reflect consensus of NIST, and have not been reviewed by any Government authority." E3-I: This message has been scanned for viruses and dangerous content by UML's antivirus scanning services. From omg-list-errors@omg.org Fri Nov 7 08:51:46 2003 Received: from ws3.uml.edu (ws3.uml.edu [129.63.1.203]) by earth.cs.uml.edu (8.11.6/8.11.6) with SMTP id hA7DpeW13336 for ; Fri, 7 Nov 2003 08:51:40 -0500 Received: from amethyst.omg.org(192.67.184.64) by ws3.uml.edu via csmap id 26973; Fri, 07 Nov 2003 07:58:13 -0500 (EST) Received: from hobbit.omg.org (hobbit.omg.org [192.67.184.3]) by amethyst.omg.org (8.12.8/8.12.8) with SMTP id hA7EgWNA013861; Fri, 7 Nov 2003 09:42:32 -0500 Received: from unknown [213.164.66.194] by hobbit.omg.org asmtp(3.0c) id 14078; Fri, 07 Nov 2003 08:52:29 -0500 (EST) Received: from io-software.com (animals [192.168.0.185]) by miles.io-software.com (8.8.8/8.7.3) with ESMTP id OAA27581; Fri, 7 Nov 2003 14:42:02 +0100 (MET) Message-ID: <3FABA16B.1030202@io-software.com> Date: Fri, 07 Nov 2003 14:43:07 +0100 From: Axel Uhl Organization: Interactive Objects Software GmbH User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.4) Gecko/20030624 Netscape/7.1 (ax) X-Accept-Language: en-us, en MIME-Version: 1.0 To: Allan Kennedy CC: ormsc@omg.org, Joaquin Miller , Edwin Seidewitz , adtf@omg.org, clive@trireme.com, "Stephen J. Mellor" , Andrew Watson , Richard Mark Soley , kc_staff@kc.com, "'Martin Fowler'" Subject: Re: MDA / Martin Fowler [What is a platform, anyway?] References: In-Reply-To: Content-Type: text/plain Content-Transfer-Encoding: 7bit X-NAIMIME-Disclaimer: 1 X-NAIMIME-Modified: 1 Status: RO Allan, > dependency on no platform - when it is perfectly clear that the highest > degree of platform independence is achieved when a model nonetheless > exhibits dependence on just one platform: *a* UML virtual machine. (I will > be able to say *the* UML virtual machine when it is standardised ;-). I don't think this is "prefectly clear." Most fellows on this thread seemed to agree that a UML VM is just another platform that is in no way special. It may be more abstract than, say, the Java platform or the CORBA platform, but it's nevertheless a platform. And as also has been pointed out in the thread, there may be more abstract platforms than the UML VM, for example some sort of weird business rules engine where you say things like "make money" and a mapping renders everything required for that in the next-lower platform, say a UML VM. Still, somebody may come up with an interpreter for the business rules engine that doesn't necessarily require the UML VM. For that reason I won't agree to your request to make the UML VM the "most abstract" platform we may ever conceive of. This would unnecessarily limit our thinking. -- Axel www.iO-Software.com - Leading the Way in Model Driven Architecture -> The leading development platform for MDA http://www.ArcStyler.com -> Comprehensive services from IT-architecture to implementation http://www.iO-Software.com/consulting E3-I: This message has been scanned for viruses and dangerous content by UML's antivirus scanning services. From omg-list-errors@omg.org Fri Nov 7 09:03:22 2003 Received: from ws3.uml.edu (ws3.uml.edu [129.63.1.203]) by earth.cs.uml.edu (8.11.6/8.11.6) with SMTP id hA7E3FW13716 for ; Fri, 7 Nov 2003 09:03:15 -0500 Received: from amethyst.omg.org(192.67.184.64) by ws3.uml.edu via csmap id 27975; Fri, 07 Nov 2003 08:09:45 -0500 (EST) Received: from hobbit.omg.org (hobbit.omg.org [192.67.184.3]) by amethyst.omg.org (8.12.8/8.12.8) with ESMTP id hA7EwONA014186; Fri, 7 Nov 2003 09:58:24 -0500 Received: from smithers.nildram.co.uk [195.112.4.34] by hobbit.omg.org asmtp(3.0c) id 14978; Fri, 07 Nov 2003 09:08:22 -0500 (EST) Received: from esukpc20 (dial-250-D05.JAO1.equant.net [57.68.5.250]) by smithers.nildram.co.uk (Postfix) with ESMTP id 85041250FA8; Fri, 7 Nov 2003 13:58:17 +0000 (GMT) From: "David Price" To: "'Joaquin Miller'" , "'Edwin Seidewitz'" Cc: , , , "'Stephen J. Mellor'" , "'Andrew Watson'" , "'Richard Mark Soley'" , , "'Martin Fowler'" , "'Allan Kennedy'" Subject: RE: MDA / Martin Fowler [What is a platform, anyway?] Date: Fri, 7 Nov 2003 13:58:16 -0000 Message-ID: <003e01c3a537$37cde2d0$fa054439@esukpc20> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_003F_01C3A537.37CDE2D0" X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook, Build 10.0.4510 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1165 In-Reply-To: <5.1.0.14.2.20031107101843.031eaa00@secure.cnchost.com> Importance: Normal Status: RO This is a multi-part message in MIME format. ------=_NextPart_000_003F_01C3A537.37CDE2D0 Content-Type: text/plain Content-Transfer-Encoding: quoted-printable X-NAIMIME-Disclaimer: 1 X-NAIMIME-Modified: 1 All, =20 Based on discussion after a presentation I gave at an ISO conference last week, the problem is both "What is a platform?" and "Of what is it independent?". I found people quite confused about MDA. They thought only one level of PIM-PSM was allowed and the idea that a PIM in one context i= s a PSM in another was also confusing. There were not people who have not studied the OMG guide. Unfortunately, that is the great majority of peopl= e who hear or talk about MDA. =20 To many people PIM/PSM terminology really is misleading. The underlying concepts are not as much of a problem as are the presentation and terminology. I think the core problem is that speaking of PSM and PIM as nouns models MDA incorrectly. A model that is closer in spirit to the underlying ideas would be that there are models and there are platforms a= nd independent or/specific to are relationships between a particular model a= nd particular platforms. There are also derived from relationships between models and relationships between platforms (for which I'm not sure of a g= ood name). If the MDA guide and presentations were simplified in this way, I believe people would understand MDA more easily. =20 Cheers, David =20 =20 =20 -----Original Message----- From: Joaquin Miller [mailto:joaquin@joaquin.net]=20 Sent: 07 November 2003 02:05 To: Edwin Seidewitz Cc: ormsc@omg.org; adtf@omg.org; clive@trireme.com; Stephen J. Mellor; Andrew Watson; Richard Mark Soley; kc_staff@kc.com; 'Martin Fowler'; Alla= n Kennedy Subject: RE: MDA / Martin Fowler [What is a platform, anyway?] =20 Thanks, Ed, for the fine discussion of the term, 'platform independence.' Edwin Seidewitz wrote: As has already been pointed out by a couple of respondents on this thread= , "platform independence" must be considered as a _relative_, not an _absolute_, term. What I want to point out is that it _is_ defined as a relative term in MDA, and its use in an absolute sense is, in my opinion,= a misunderstanding of MDA (perhaps a reasonable misunderstanding, given the connotation of the term itself). I share your gentle and courteous feelings toward folks who misunderstand what is intended by 'platform independence.' But i don't feel that the connotation of the term itself can stand as an explanation. The obvious (and maybe the only likely) meaning of 'platform independence' is: the quality of being independent of a platform or platforms. I don't expect there is any problem with the meaning of 'to be independent of.' So, if = the misunderstanding has to do with the connotation of the term, it must be 'platform' that is the problem. Let me give a few examples that i feel most folks ought to recognize as platforms: IBM ThinkPad 570 E IBM ThinkPad 570 E + Big Bill Windows 2000=20 IBM ThinkPad 570 E + Linux Linux CORBA J2EE IBM MVS BEA WebLogic Apple System X Smalltalk Even if some folks do not consider some of these to be platforms, they ou= ght to admit most of them as platforms. (Of course, 'CORBA,' as a platform, must mean CORBA plus some operating system(s) and computer(s): CORBA software alone is helpless.) The examples, if they are all (or most) platforms, ought to show that platform independence is necessarily relative to the platforms one has in mind. If i understand Martin (i am not well internet-connected, and have not se= en his original comments), or better: If i understand the explanations of wh= at Martin wrote, he points out that it is impossible for a model or specification to be independent of all platforms. I'll buy that. At the same time, i believe one can write a model that is independent of the cla= ss of platforms that includes CORBA Components and J2EE, and a model that is object-oriented, but independent of a large class of platforms. If the term, 'platform independence' is a malaproprism, that can be fixed= =2E On the other hand, if i misunderstand the explanations of what Martin wro= te and what he points out is instead that the idea we are using that name fo= r is bogus, then I'll have to get better connected and read his wiki commen= ts. =2E...... For my part, i certainly do not consider UML a platform. It is a languag= e. I do believe that Process Technology and many others since provide platfo= rms which support UML models. And i suppose that some, many, most, or all UM= L models supported by one of those platforms are not independent of that platform. For my part, i certainly do not consider MDA to be a platform. It is an approach. [omg/2002-06-01 at 2.1.1] It is correct usage to use 'CORBA' = for both the architecture and the platform. But those are two separate thing= s. If it is not clear from the context, then we need to say either 'the CORB= A architecture' or 'a CORBA platform.' If we say 'the CORBA platform,' we must mean the class of platforms that conform to the CORBA specification. If folks want to talk about an MDA platform, that's fine. There 'MDA' is= an adjective, not a noun. Then, of course, given the way our language works= , 'MDA' will, as 'CORBA' has, come to have two meanings as a noun. For no= w, let's be clear, and use 'an MDA platform,' if we want to mean a platform which is MDA-like, or MDA-suitable, or whatever we intend to mean by that term. 'The UML/MDA platform' (with 'the') can (for me) only mean a class of platforms that use the UML language and the MDA approach. It is certainly true that a model for "the UML/MDA platform" is dependent= on "the UML/MDA platform." It is quite possible that such a model could be independent of some class of "UML/MDA platforms," instead of locked into = a particular "UML/MDA platform." Let's hope so. Down the road, anyway. Please remember that it was some time from the promulgation of CORBA (the architecture) to the possibility of a program that was portable across several different CORBAs (the platforms). These things take many years. The MDA of today is like the CORBA of the first years: a beacon showing a direction. It ain't no platform, nohow.=20 Cordially, Joaquin [Martin, please forgive me for commenting before reading.] [Allan, I'd like to quibble about "the UML virtual machine." I'll believ= e it when i see it. I have no objection to the use of 'a UML virtual machine.' (Yes, i do know what is intended, and, yes, this is a quibble. But: Loose lips sink ships.)] PGP Fingerprint: CA23 6BCA ACAB 6006 E3C3 0E79 2122 94B4 E5FD 42C3 E3-I: This message has been scanned for viruses and dangerous content by= UML's antivirus scanning services. ------=_NextPart_000_003F_01C3A537.37CDE2D0 Content-Type: text/html Content-Transfer-Encoding: 7bit X-NAIMIME-Modified: 1

All,

 

Based on discussion after a presentation I gave at an ISO conference last week, the problem is both “What is a platform?” and “Of what is it independent?”. I found people quite confused about MDA. They thought only one level of PIM-PSM was allowed and the idea that a PIM in one context is a PSM in another was also confusing. There were not people who have not studied the OMG guide. Unfortunately, that is the great majority of people who hear or talk about MDA.

 

To many people PIM/PSM terminology really is misleading. The underlying concepts are not as much of a problem as are the presentation and terminology. I think the core problem is that speaking of PSM and PIM as nouns models MDA incorrectly. A model that is closer in spirit to the underlying ideas would be that there are models and there are platforms and independent or/specific to are relationships between a particular model and particular platforms. There are also derived from relationships between models and relationships between platforms (for which I’m not sure of a good name). If the MDA guide and presentations were simplified in this way, I believe people would understand MDA more easily.

 

Cheers,

David

 

 

 

-----Original Message-----
From: Joaquin Miller [mailto:joaquin@joaquin.net]
Sent: 07 November 2003 02:05
To: Edwin Seidewitz
Cc: ormsc@omg.org; adtf@omg.org; clive@trireme.com; Stephen J. Mellor; Andrew Watson; Richard Mark Soley; kc_staff@kc.com; 'Martin Fowler'; Allan Kennedy
Subject: RE: MDA / Martin Fowler [What is a platform, anyway?]

 

Thanks, Ed, for the fine discussion of the term, 'platform independence.'

Edwin Seidewitz wrote:

As has already been pointed out by a couple of respondents on this thread, "platform independence" must be considered as a _relative_, not an _absolute_, term. What I want to point out is that it _is_ defined as a relative term in MDA, and its use in an absolute sense is, in my opinion, a misunderstanding of MDA (perhaps a reasonable misunderstanding, given the connotation of the term itself).


I share your gentle and courteous feelings toward folks who misunderstand what is intended by 'platform independence.'  But i don't feel that the connotation of the term itself can stand as an explanation.  The obvious (and maybe the only likely) meaning of 'platform independence' is: the quality of being independent of a platform or platforms.  I don't expect there is any problem with the meaning of 'to be independent of.'  So, if the misunderstanding has to do with the connotation of the term, it must be 'platform' that is the problem.

Let me give a few examples that i feel most folks ought to recognize as platforms:

IBM ThinkPad 570 E
IBM ThinkPad 570 E + Big Bill Windows 2000
IBM ThinkPad 570 E + Linux
Linux
CORBA
J2EE
IBM MVS
BEA WebLogic
Apple System X
Smalltalk

Even if some folks do not consider some of these to be platforms, they ought to admit most of them as platforms.  (Of course, 'CORBA,' as a platform, must mean CORBA plus some operating system(s) and computer(s):  CORBA software alone is helpless.)

The examples, if they are all (or most) platforms, ought to show that platform independence is necessarily relative to the platforms one has in mind.

If i understand Martin (i am not well internet-connected, and have not seen his original comments), or better: If i understand the explanations of what Martin wrote, he points out that it is impossible for a model or specification to be independent of all platforms.  I'll buy that.  At the same time, i believe one can write a model that is independent of the class of platforms that includes CORBA Components and J2EE, and a model that is object-oriented, but independent of a large class of platforms.

If the term, 'platform independence' is a malaproprism, that can be fixed.  On the other hand, if i misunderstand the explanations of what Martin wrote and what he points out is instead that the idea we are using that name for is bogus, then I'll have to get better connected and read his wiki comments.

.......

For my part, i certainly do not consider UML a platform.  It is a language.  I do believe that Process Technology and many others since provide platforms which support UML models.  And i suppose that some, many, most, or all UML models supported by one of those platforms are not independent of that platform.

For my part, i certainly do not consider MDA to be a platform.  It is an approach.  [omg/2002-06-01 at 2.1.1]  It is correct usage to use 'CORBA' for both the architecture and the platform.  But those are two separate things.  If it is not clear from the context, then we need to say either 'the CORBA architecture' or 'a CORBA platform.'  If we say 'the CORBA platform,' we must mean the class of platforms that conform to the CORBA specification.

If folks want to talk about an MDA platform, that's fine.  There 'MDA' is an adjective, not a noun.  Then, of course, given the way our language works, 'MDA' will, as 'CORBA' has,  come to have two meanings as a noun.  For now, let's be clear, and use 'an MDA platform,' if we want to mean a platform which is MDA-like, or MDA-suitable, or whatever we intend to mean by that term.

'The UML/MDA platform' (with 'the') can (for me) only mean a class of platforms that use the UML language and the MDA approach.

It is certainly true that a model for "the UML/MDA platform" is dependent on "the UML/MDA platform."  It is quite possible that such a model could be independent of some class of "UML/MDA platforms," instead of locked into a particular "UML/MDA platform."   Let's hope so.  Down the road, anyway.

Please remember that it was some time from the promulgation of CORBA (the architecture) to the possibility of a program that was portable across several different CORBAs (the platforms).  These things take many years.

The MDA of today is like the CORBA of the first years: a beacon showing a direction.  It ain't no platform, nohow.

Cordially,

Joaquin

[Martin, please forgive me for commenting before reading.]

[Allan, I'd like to quibble about "the UML virtual machine."  I'll believe it when i see it.  I have no objection to the use of 'a UML virtual machine.'  (Yes, i do know what is intended, and, yes, this is a quibble. But: Loose lips sink ships.)]




PGP Fingerprint:
CA23 6BCA ACAB 6006 E3C3 0E79 2122 94B4 E5FD 42C3

E3-I: This message has been scanned for viruses and dangerous content by UML's antivirus scanning services.

------=_NextPart_000_003F_01C3A537.37CDE2D0-- From omg-list-errors@omg.org Fri Nov 7 09:06:53 2003 Received: from ws3.uml.edu (ws3.uml.edu [129.63.1.203]) by earth.cs.uml.edu (8.11.6/8.11.6) with SMTP id hA7E6kW13864 for ; Fri, 7 Nov 2003 09:06:46 -0500 Received: from amethyst.omg.org(192.67.184.64) by ws3.uml.edu via csmap id 28266; Fri, 07 Nov 2003 08:13:07 -0500 (EST) Received: from hobbit.omg.org (hobbit.omg.org [192.67.184.3]) by amethyst.omg.org (8.12.8/8.12.8) with ESMTP id hA7F3ENA014303; Fri, 7 Nov 2003 10:03:14 -0500 Received: from rio.mega.com [213.11.73.221] by hobbit.omg.org asmtp(3.0c) id 15347; Fri, 07 Nov 2003 09:13:12 -0500 (EST) Received: from relay.mega.com by rio.mega.com via smtpd (for hobbit.omg.org [192.67.184.3]) with ESMTP; Fri, 7 Nov 2003 15:03:22 +0100 Received: from xor.mega.com ([194.3.176.3]) by relay with InterScan Messaging Security Suite; Fri, 07 Nov 2003 15:02:54 +0100 Received: from rio.fr.mega.com by xor.mega.com via smtpd (for relay.mega.com [192.168.167.9]) with ESMTP; Fri, 7 Nov 2003 15:03:03 +0100 Received: by RIO with Internet Mail Service (5.5.2653.19) id ; Fri, 7 Nov 2003 15:02:58 +0100 Message-ID: From: LONJON Antoine To: "'Joaquin Miller'" , Edwin Seidewitz Cc: ormsc@omg.org, adtf@omg.org, clive@trireme.com, "Stephen J. Mellor" , Andrew Watson , Richard Mark Soley , kc_staff@kc.com, "'Martin Fowler'" , Allan Kennedy Subject: RE: MDA / Martin Fowler [What is a platform, anyway?] Date: Fri, 7 Nov 2003 15:02:57 +0100 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2653.19) Content-Type: multipart/alternative; boundary="----_=_NextPart_001_01C3A537.D8EE8340" Status: RO This message is in MIME format. Since your mail reader does not understand this format, some or all of this message may not be legible. ------_=_NextPart_001_01C3A537.D8EE8340 Content-Type: text/plain X-NAIMIME-Disclaimer: 1 Content-Transfer-Encoding: 7bit X-NAIMIME-Modified: 1 There are interesting discussions about the value of UML/MDA and independant platform. I agree with Joaquin statement: "For my part, i certainly do not consider MDA to be a platform. It is an approach." Platform independence and even UML (at least the superstructure) are just use cases of a Model Driven Approach. In a word, they can be the result of such an approach, not its basis. -----Original Message----- From: Joaquin Miller [mailto:joaquin@joaquin.net] Sent: Friday, November 07, 2003 3:05 AM To: Edwin Seidewitz Cc: ormsc@omg.org; adtf@omg.org; clive@trireme.com; Stephen J. Mellor; Andrew Watson; Richard Mark Soley; kc_staff@kc.com; 'Martin Fowler'; Allan Kennedy Subject: RE: MDA / Martin Fowler [What is a platform, anyway?] Thanks, Ed, for the fine discussion of the term, 'platform independence.' Edwin Seidewitz wrote: As has already been pointed out by a couple of respondents on this thread, "platform independence" must be considered as a _relative_, not an _absolute_, term. What I want to point out is that it _is_ defined as a relative term in MDA, and its use in an absolute sense is, in my opinion, a misunderstanding of MDA (perhaps a reasonable misunderstanding, given the connotation of the term itself). I share your gentle and courteous feelings toward folks who misunderstand what is intended by 'platform independence.' But i don't feel that the connotation of the term itself can stand as an explanation. The obvious (and maybe the only likely) meaning of 'platform independence' is: the quality of being independent of a platform or platforms. I don't expect there is any problem with the meaning of 'to be independent of.' So, if the misunderstanding has to do with the connotation of the term, it must be 'platform' that is the problem. Let me give a few examples that i feel most folks ought to recognize as platforms: IBM ThinkPad 570 E IBM ThinkPad 570 E + Big Bill Windows 2000 IBM ThinkPad 570 E + Linux Linux CORBA J2EE IBM MVS BEA WebLogic Apple System X Smalltalk Even if some folks do not consider some of these to be platforms, they ought to admit most of them as platforms. (Of course, 'CORBA,' as a platform, must mean CORBA plus some operating system(s) and computer(s): CORBA software alone is helpless.) The examples, if they are all (or most) platforms, ought to show that platform independence is necessarily relative to the platforms one has in mind. If i understand Martin (i am not well internet-connected, and have not seen his original comments), or better: If i understand the explanations of what Martin wrote, he points out that it is impossible for a model or specification to be independent of all platforms. I'll buy that. At the same time, i believe one can write a model that is independent of the class of platforms that includes CORBA Components and J2EE, and a model that is object-oriented, but independent of a large class of platforms. If the term, 'platform independence' is a malaproprism, that can be fixed. On the other hand, if i misunderstand the explanations of what Martin wrote and what he points out is instead that the idea we are using that name for is bogus, then I'll have to get better connected and read his wiki comments. ....... For my part, i certainly do not consider UML a platform. It is a language. I do believe that Process Technology and many others since provide platforms which support UML models. And i suppose that some, many, most, or all UML models supported by one of those platforms are not independent of that platform. For my part, i certainly do not consider MDA to be a platform. It is an approach. [omg/2002-06-01 at 2.1.1] It is correct usage to use 'CORBA' for both the architecture and the platform. But those are two separate things. If it is not clear from the context, then we need to say either 'the CORBA architecture' or 'a CORBA platform.' If we say 'the CORBA platform,' we must mean the class of platforms that conform to the CORBA specification. If folks want to talk about an MDA platform, that's fine. There 'MDA' is an adjective, not a noun. Then, of course, given the way our language works, 'MDA' will, as 'CORBA' has, come to have two meanings as a noun. For now, let's be clear, and use 'an MDA platform,' if we want to mean a platform which is MDA-like, or MDA-suitable, or whatever we intend to mean by that term. 'The UML/MDA platform' (with 'the') can (for me) only mean a class of platforms that use the UML language and the MDA approach. It is certainly true that a model for "the UML/MDA platform" is dependent on "the UML/MDA platform." It is quite possible that such a model could be independent of some class of "UML/MDA platforms," instead of locked into a particular "UML/MDA platform." Let's hope so. Down the road, anyway. Please remember that it was some time from the promulgation of CORBA (the architecture) to the possibility of a program that was portable across several different CORBAs (the platforms). These things take many years. The MDA of today is like the CORBA of the first years: a beacon showing a direction. It ain't no platform, nohow. Cordially, Joaquin [Martin, please forgive me for commenting before reading.] [Allan, I'd like to quibble about "the UML virtual machine." I'll believe it when i see it. I have no objection to the use of 'a UML virtual machine.' (Yes, i do know what is intended, and, yes, this is a quibble. But: Loose lips sink ships.)] PGP Fingerprint: CA23 6BCA ACAB 6006 E3C3 0E79 2122 94B4 E5FD 42C3 E3-I: This message has been scanned for viruses and dangerous content by UML's antivirus scanning services. ------_=_NextPart_001_01C3A537.D8EE8340 Content-Type: text/html Content-Transfer-Encoding: 7bit X-NAIMIME-Modified: 1 Message
There are interesting discussions about the value of UML/MDA and independant platform.
I agree with Joaquin statement:
"For my part, i certainly do not consider MDA to be a platform.  It is an approach."
Platform independence and even UML (at least the superstructure) are just use cases of a Model Driven Approach. In a word, they can be the result of such an approach, not its basis.
 
-----Original Message-----
From: Joaquin Miller [mailto:joaquin@joaquin.net]
Sent: Friday, November 07, 2003 3:05 AM
To: Edwin Seidewitz
Cc: ormsc@omg.org; adtf@omg.org; clive@trireme.com; Stephen J. Mellor; Andrew Watson; Richard Mark Soley; kc_staff@kc.com; 'Martin Fowler'; Allan Kennedy
Subject: RE: MDA / Martin Fowler [What is a platform, anyway?]

Thanks, Ed, for the fine discussion of the term, 'platform independence.'

Edwin Seidewitz wrote:
As has already been pointed out by a couple of respondents on this thread, "platform independence" must be considered as a _relative_, not an _absolute_, term. What I want to point out is that it _is_ defined as a relative term in MDA, and its use in an absolute sense is, in my opinion, a misunderstanding of MDA (perhaps a reasonable misunderstanding, given the connotation of the term itself).

I share your gentle and courteous feelings toward folks who misunderstand what is intended by 'platform independence.'  But i don't feel that the connotation of the term itself can stand as an explanation.  The obvious (and maybe the only likely) meaning of 'platform independence' is: the quality of being independent of a platform or platforms.  I don't expect there is any problem with the meaning of 'to be independent of.'  So, if the misunderstanding has to do with the connotation of the term, it must be 'platform' that is the problem.

Let me give a few examples that i feel most folks ought to recognize as platforms:

IBM ThinkPad 570 E
IBM ThinkPad 570 E + Big Bill Windows 2000
IBM ThinkPad 570 E + Linux
Linux
CORBA
J2EE
IBM MVS
BEA WebLogic
Apple System X
Smalltalk

Even if some folks do not consider some of these to be platforms, they ought to admit most of them as platforms.  (Of course, 'CORBA,' as a platform, must mean CORBA plus some operating system(s) and computer(s):  CORBA software alone is helpless.)

The examples, if they are all (or most) platforms, ought to show that platform independence is necessarily relative to the platforms one has in mind.

If i understand Martin (i am not well internet-connected, and have not seen his original comments), or better: If i understand the explanations of what Martin wrote, he points out that it is impossible for a model or specification to be independent of all platforms.  I'll buy that.  At the same time, i believe one can write a model that is independent of the class of platforms that includes CORBA Components and J2EE, and a model that is object-oriented, but independent of a large class of platforms.

If the term, 'platform independence' is a malaproprism, that can be fixed.  On the other hand, if i misunderstand the explanations of what Martin wrote and what he points out is instead that the idea we are using that name for is bogus, then I'll have to get better connected and read his wiki comments.

.......

For my part, i certainly do not consider UML a platform.  It is a language.  I do believe that Process Technology and many others since provide platforms which support UML models.  And i suppose that some, many, most, or all UML models supported by one of those platforms are not independent of that platform.

For my part, i certainly do not consider MDA to be a platform.  It is an approach.  [omg/2002-06-01 at 2.1.1]  It is correct usage to use 'CORBA' for both the architecture and the platform.  But those are two separate things.  If it is not clear from the context, then we need to say either 'the CORBA architecture' or 'a CORBA platform.'  If we say 'the CORBA platform,' we must mean the class of platforms that conform to the CORBA specification.

If folks want to talk about an MDA platform, that's fine.  There 'MDA' is an adjective, not a noun.  Then, of course, given the way our language works, 'MDA' will, as 'CORBA' has,  come to have two meanings as a noun.  For now, let's be clear, and use 'an MDA platform,' if we want to mean a platform which is MDA-like, or MDA-suitable, or whatever we intend to mean by that term.

'The UML/MDA platform' (with 'the') can (for me) only mean a class of platforms that use the UML language and the MDA approach.

It is certainly true that a model for "the UML/MDA platform" is dependent on "the UML/MDA platform."  It is quite possible that such a model could be independent of some class of "UML/MDA platforms," instead of locked into a particular "UML/MDA platform."   Let's hope so.  Down the road, anyway.

Please remember that it was some time from the promulgation of CORBA (the architecture) to the possibility of a program that was portable across several different CORBAs (the platforms).  These things take many years.

The MDA of today is like the CORBA of the first years: a beacon showing a direction.  It ain't no platform, nohow.

Cordially,

Joaquin

[Martin, please forgive me for commenting before reading.]

[Allan, I'd like to quibble about "the UML virtual machine."  I'll believe it when i see it.  I have no objection to the use of 'a UML virtual machine.'  (Yes, i do know what is intended, and, yes, this is a quibble. But: Loose lips sink ships.)]



PGP Fingerprint:
CA23 6BCA ACAB 6006 E3C3 0E79 2122 94B4 E5FD 42C3

E3-I: This message has been scanned for viruses and dangerous content by UML's antivirus scanning services.

------_=_NextPart_001_01C3A537.D8EE8340-- From omg-list-errors@omg.org Fri Nov 7 09:30:48 2003 Received: from ws3.uml.edu (ws3.uml.edu [129.63.1.203]) by earth.cs.uml.edu (8.11.6/8.11.6) with SMTP id hA7EUiW14540 for ; Fri, 7 Nov 2003 09:30:44 -0500 Received: from amethyst.omg.org(192.67.184.64) by ws3.uml.edu via csmap id 30660; Fri, 07 Nov 2003 08:37:16 -0500 (EST) Received: from hobbit.omg.org (hobbit.omg.org [192.67.184.3]) by amethyst.omg.org (8.12.8/8.12.8) with ESMTP id hA7FOqNC014768; Fri, 7 Nov 2003 10:25:01 -0500 Received: from e6.ny.us.ibm.com [32.97.182.106] by hobbit.omg.org asmtp(3.0c) id 17528; Fri, 07 Nov 2003 09:34:50 -0500 (EST) Received: from northrelay02.pok.ibm.com (northrelay02.pok.ibm.com [9.56.224.150]) by e6.ny.us.ibm.com (8.12.10/8.12.2) with ESMTP id hA7EOBVn081056; Fri, 7 Nov 2003 09:24:11 -0500 Received: from d25ml05.torolab.ibm.com (d01av02.pok.ibm.com [9.56.224.216]) by northrelay02.pok.ibm.com (8.12.9/NCO/VER6.6) with ESMTP id hA7EOAh9211812; Fri, 7 Nov 2003 09:24:10 -0500 In-Reply-To: <3FAB574D.5030304@io-software.com> To: Axel Uhl Cc: adtf@omg.org, Allan Kennedy , Andrew Watson , clive@trireme.com, Edwin Seidewitz , "'Martin Fowler'" , Joaquin Miller , kc_staff@kc.com, ormsc@omg.org, Richard Mark Soley , "Stephen J. Mellor" MIME-Version: 1.0 Subject: Re: MDA / Martin Fowler [What is a platform, anyway?] X-Mailer: Lotus Notes Release 6.0.2CF1 June 9, 2003 From: Branislav Selic Message-ID: Date: Fri, 7 Nov 2003 09:24:05 -0500 X-MIMETrack: Serialize by Router on D25ML05/25/M/IBM(Release 6.0.2CF1|June 9, 2003) at 11/07/2003 09:24:09, Serialize complete at 11/07/2003 09:24:09 Content-Type: multipart/alternative; boundary="=_alternative 004EA96A85256DD7_=" Status: RO This is a multipart message in MIME format. --=_alternative 004EA96A85256DD7_= Content-Type: text/plain X-NAIMIME-Disclaimer: 1 X-NAIMIME-Modified: 1 I guess I am glad this issue has finally been brought up by others than myself -- I have complained for a long time about it. Most people DO interpret the term "platform independence" in its absolute simple-minded sense. I am surprised that anyone expected otherwise -- hardly anyone ever reads the footnotes which, after all, never make it into the marketing literature. What really bugs me about it, is that the notion is then used as an excuse to ignore platform issues altogether. I have written a number of articles, given numerous talks, and even helped produce an OMG standard that deal with the fundamental importance of understanding the capabilities and capacities of your platforms during software design. After all, the platform is the raw material that runs software and whose characteristics (processing speed, availability, communications characteristics, etc.) will often impose limitations on what can be done with program logic. For instance, anyone who has worked with distributed systems knows that it is usually hopeless to pretend that the system is not distributed and that it can be programmed just like a non-distributed one. There are even formally proven impossibility results that tell us this. If we really want to capture what MDA is about, it is about increasing the levels of abstraction and automation. That should serve as the main selling point rather than a catchy-but-misleading marketing term such as "platform independence". While the latter may serve the near-sighted goal of getting people interested in MDA, there will be the Devil to pay when it is exposed. And, I'm afraid it will compromise or even discredit the whole wonderful notion of MDA. It's probably too late to get rid of it, but let's see if we can at least agree among ourselves to downplay it. Cheers, Bran Selic IBM Software Group -- Rational Software 770 Palladium Drive Kanata, Ontario, Canada K2V 1C8 ph. (613) 591-7915 fax (613) 599-3912 e-mail: bselic@ca.ibm.com E3-I: This message has been scanned for viruses and dangerous content by UML's antivirus scanning services. --=_alternative 004EA96A85256DD7_= Content-Type: text/html Content-Transfer-Encoding: 7bit X-NAIMIME-Modified: 1
I guess I am glad this issue has finally been brought up by others than myself -- I have complained for a long time about it. Most people DO interpret the term "platform independence" in its absolute simple-minded sense. I am surprised that anyone expected otherwise -- hardly anyone ever reads the footnotes which, after all, never make it into the marketing literature.

What really bugs me about it, is that the notion is then used as an excuse to ignore platform issues altogether. I have written a number of articles, given numerous talks, and even helped produce an OMG standard that deal with the fundamental importance of understanding the capabilities and capacities of your platforms during software design. After all, the platform is the raw material that runs software and whose characteristics (processing speed, availability, communications characteristics, etc.) will often impose limitations on what can be done with program logic. For instance, anyone who has worked with distributed systems knows that it is usually hopeless to pretend that the system is not distributed and that it can be programmed just like a non-distributed one. There are even formally proven impossibility results that tell us this.

If we really want to capture what MDA is about, it is about increasing the levels of abstraction and automation. That should serve as the main selling point rather than a catchy-but-misleading marketing term such as "platform independence". While the latter may serve the near-sighted goal of getting people interested in MDA, there will be the Devil to pay when it is exposed. And, I'm afraid it will compromise or even discredit the whole wonderful notion of MDA. It's probably too late to get rid of it, but let's see if we can at least agree among ourselves to downplay it.

Cheers,
Bran Selic
IBM Software Group -- Rational Software
770 Palladium Drive
Kanata, Ontario, Canada
K2V 1C8

ph. (613) 591-7915
fax (613) 599-3912
e-mail: bselic@ca.ibm.com

E3-I: This message has been scanned for viruses and dangerous content by UML's antivirus scanning services.

--=_alternative 004EA96A85256DD7_=-- From omg-list-errors@omg.org Fri Nov 7 11:06:22 2003 Received: from ws3.uml.edu (ws3.uml.edu [129.63.1.203]) by earth.cs.uml.edu (8.11.6/8.11.6) with SMTP id hA7G6GW18796 for ; Fri, 7 Nov 2003 11:06:16 -0500 Received: from amethyst.omg.org(192.67.184.64) by ws3.uml.edu via csmap id 7973; Fri, 07 Nov 2003 10:11:58 -0500 (EST) Received: from hobbit.omg.org (hobbit.omg.org [192.67.184.3]) by amethyst.omg.org (8.12.8/8.12.8) with ESMTP id hA7GdgNA016126; Fri, 7 Nov 2003 11:39:42 -0500 Received: from mailhost.softeam.fr [194.2.241.2] by hobbit.omg.org asmtp(3.0c) id 28954; Fri, 07 Nov 2003 10:49:39 -0500 (EST) Received: from PHD (ATuileries-103-2-1-136.w193-252.abo.wanadoo.fr [193.252.55.136]) by mailhost.softeam.fr (8.9.3/93idpgr-1.2.17) with SMTP id QAA25124; Fri, 7 Nov 2003 16:38:57 +0100 (MET) Reply-To: From: "DESFRAY Philippe" To: "'Branislav Selic'" , "'Axel Uhl'" Cc: , "'Allan Kennedy'" , "'Andrew Watson'" , , "'Edwin Seidewitz'" , "'Martin Fowler'" , "'Joaquin Miller'" , , , "'Richard Mark Soley'" , "'Stephen J. Mellor'" Subject: RE: MDA / Martin Fowler [What is a platform, anyway?] Date: Fri, 7 Nov 2003 16:34:42 +0100 Message-ID: <002401c3a544$ae944e60$100a4b0a@PHD> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_0025_01C3A54D.1058B660" X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook CWS, Build 9.0.2416 (9.0.2911.0) Importance: Normal In-Reply-To: X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1106 Status: RO This is a multi-part message in MIME format. ------=_NextPart_000_0025_01C3A54D.1058B660 Content-Type: text/plain Content-Transfer-Encoding: 7bit X-NAIMIME-Disclaimer: 1 X-NAIMIME-Modified: 1 Steve Mellor wrote : >>> No one is saying that a PIM is independent of *all* platforms. and Bran wrote >>> Most people DO interpret the term "platform independence" in its absolute simple-minded sense As I do interpret (simple minded I am :=). I think that PIM means Platform Independant Model, and that this has only one meaning : a model which does not AT ALL depends on platforms. That kind of animal exist though : a conceptual model, a business process model, many models at preliminary analysis stages are and should be "platform independant". I share the point that once you speak of "executable UML" you just depend on a "UML execution platform" which is not platform independant. It is important to be simple minded here, because the common mortals who struggle (if they care) to understand MDA will not parse the following assertion : "A PIM is a Platform Independant Model ... which is not independant of all platforms". This is an issue easy to solve : 1 give its natural and formal sense the the acronym PIM; 2 stop claiming that "executable UML is platform independant". Otherwise, negative minded people (is there an intersection with simple minded people?) will think that MDA is just a marketing buzzwork full of nonsense definitions. -- Philippe DESFRAY VP for R& D phd@softeam.fr SOFTEAM - Objecteering Software --------Think Object 144 avenue des Champs Elysees - 75008 PARIS - Tel (33 1) 53968400 - Fax (33 1) 53968401 -----Message d'origine----- De : Branislav Selic [mailto:bselic@ca.ibm.com] Envoye : vendredi 7 novembre 2003 15:24 A : Axel Uhl Cc : adtf@omg.org; Allan Kennedy; Andrew Watson; clive@trireme.com; Edwin Seidewitz; 'Martin Fowler'; Joaquin Miller; kc_staff@kc.com; ormsc@omg.org; Richard Mark Soley; Stephen J. Mellor Objet : Re: MDA / Martin Fowler [What is a platform, anyway?] I guess I am glad this issue has finally been brought up by others than myself -- I have complained for a long time about it. Most people DO interpret the term "platform independence" in its absolute simple-minded sense. I am surprised that anyone expected otherwise -- hardly anyone ever reads the footnotes which, after all, never make it into the marketing literature. What really bugs me about it, is that the notion is then used as an excuse to ignore platform issues altogether. I have written a number of articles, given numerous talks, and even helped produce an OMG standard that deal with the fundamental importance of understanding the capabilities and capacities of your platforms during software design. After all, the platform is the raw material that runs software and whose characteristics (processing speed, availability, communications characteristics, etc.) will often impose limitations on what can be done with program logic. For instance, anyone who has worked with distributed systems knows that it is usually hopeless to pretend that the system is not distributed and that it can be programmed just like a non-distributed one. There are even formally proven impossibility results that tell us this. If we really want to capture what MDA is about, it is about increasing the levels of abstraction and automation. That should serve as the main selling point rather than a catchy-but-misleading marketing term such as "platform independence". While the latter may serve the near-sighted goal of getting people interested in MDA, there will be the Devil to pay when it is exposed. And, I'm afraid it will compromise or even discredit the whole wonderful notion of MDA. It's probably too late to get rid of it, but let's see if we can at least agree among ourselves to downplay it. Cheers, Bran Selic IBM Software Group -- Rational Software 770 Palladium Drive Kanata, Ontario, Canada K2V 1C8 ph. (613) 591-7915 fax (613) 599-3912 e-mail: bselic@ca.ibm.com E3-I: This message has been scanned for viruses and dangerous content by UML's antivirus scanning services. ------=_NextPart_000_0025_01C3A54D.1058B660 Content-Type: text/html Content-Transfer-Encoding: 7bit X-NAIMIME-Modified: 1
 
Steve Mellor wrote :
>>> No one is saying that a PIM is independent of *all* platforms.
 
and Bran wrote
>>> Most people DO interpret the term "platform independence" in its absolute simple-minded sense
 
As I do interpret (simple minded I am :=). I think that PIM means Platform Independant Model, and that this has only one meaning : a model which does not AT ALL depends on platforms.
 
That kind of animal exist though : a conceptual model, a business process model, many models at preliminary analysis stages are and should be "platform independant".
 
I share the point that once you speak of "executable UML" you just depend on a "UML execution platform" which is not platform independant.
 
It is important to be simple minded here, because the common mortals who struggle (if they care) to understand MDA will not parse the following assertion : "A PIM is a Platform Independant Model ... which is not independant of all platforms".
 
This is an issue easy to solve : 1 give its natural and formal sense the the acronym PIM; 2 stop claiming that "executable UML is platform independant".
 
Otherwise, negative minded people (is there an intersection with simple minded people?) will think that MDA is just a marketing buzzwork full of nonsense definitions.

--
Philippe DESFRAY     VP for R& D   phd@softeam.fr
SOFTEAM - Objecteering Software
--------Think Object
144 avenue des Champs Elysées - 75008 PARIS - Tel (33 1) 53968400 - Fax (33 1) 53968401


-----Message d'origine-----
De : Branislav Selic [mailto:bselic@ca.ibm.com]
Envoyé : vendredi 7 novembre 2003 15:24
À : Axel Uhl
Cc : adtf@omg.org; Allan Kennedy; Andrew Watson; clive@trireme.com; Edwin Seidewitz; 'Martin Fowler'; Joaquin Miller; kc_staff@kc.com; ormsc@omg.org; Richard Mark Soley; Stephen J. Mellor
Objet : Re: MDA / Martin Fowler [What is a platform, anyway?]


I guess I am glad this issue has finally been brought up by others than myself -- I have complained for a long time about it. Most people DO interpret the term "platform independence" in its absolute simple-minded sense. I am surprised that anyone expected otherwise -- hardly anyone ever reads the footnotes which, after all, never make it into the marketing literature.

What really bugs me about it, is that the notion is then used as an excuse to ignore platform issues altogether. I have written a number of articles, given numerous talks, and even helped produce an OMG standard that deal with the fundamental importance of understanding the capabilities and capacities of your platforms during software design. After all, the platform is the raw material that runs software and whose characteristics (processing speed, availability, communications characteristics, etc.) will often impose limitations on what can be done with program logic. For instance, anyone who has worked with distributed systems knows that it is usually hopeless to pretend that the system is not distributed and that it can be programmed just like a non-distributed one. There are even formally proven impossibility results that tell us this.

If we really want to capture what MDA is about, it is about increasing the levels of abstraction and automation. That should serve as the main selling point rather than a catchy-but-misleading marketing term such as "platform independence". While the latter may serve the near-sighted goal of getting people interested in MDA, there will be the Devil to pay when it is exposed. And, I'm afraid it will compromise or even discredit the whole wonderful notion of MDA. It's probably too late to get rid of it, but let's see if we can at least agree among ourselves to downplay it.

Cheers,
Bran Selic
IBM Software Group -- Rational Software
770 Palladium Drive
Kanata, Ontario, Canada
K2V 1C8

ph. (613) 591-7915
fax (613) 599-3912
e-mail: bselic@ca.ibm.com

E3-I: This message has been scanned for viruses and dangerous content by UML's antivirus scanning services.

------=_NextPart_000_0025_01C3A54D.1058B660-- From omg-list-errors@omg.org Fri Nov 7 11:11:09 2003 Received: from ws3.uml.edu (ws3.uml.edu [129.63.1.203]) by earth.cs.uml.edu (8.11.6/8.11.6) with SMTP id hA7GB3W19072 for ; Fri, 7 Nov 2003 11:11:03 -0500 Received: from amethyst.omg.org(192.67.184.64) by ws3.uml.edu via csmap id 8618; Fri, 07 Nov 2003 10:17:35 -0500 (EST) Received: from hobbit.omg.org (hobbit.omg.org [192.67.184.3]) by amethyst.omg.org (8.12.8/8.12.8) with ESMTP id hA7H5fNA016654; Fri, 7 Nov 2003 12:05:41 -0500 Received: from unknown [213.78.87.93] by hobbit.omg.org asmtp(3.0c) id 1709; Fri, 07 Nov 2003 11:15:39 -0500 (EST) Received: by duojet.home.tratt.net (Postfix, from userid 1000) id 5E8B55102; Fri, 7 Nov 2003 16:05:51 +0001 (GMT) Date: Fri, 7 Nov 2003 16:05:51 +0001 From: Laurence Tratt To: Branislav Selic Cc: Axel Uhl , adtf@omg.org, Allan Kennedy , Andrew Watson , clive@trireme.com, Edwin Seidewitz , "'Martin Fowler'" , Joaquin Miller , kc_staff@kc.com, ormsc@omg.org, Richard Mark Soley , "Stephen J. Mellor" Subject: Re: MDA / Martin Fowler [What is a platform, anyway?] Message-ID: <20031107160612.GF14224@duojet.home.tratt.net> References: <3FAB574D.5030304@io-software.com> Mime-Version: 1.0 Content-Type: text/plain Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.4i X-NAIMIME-Disclaimer: 1 X-NAIMIME-Modified: 1 Status: RO On Fri, Nov 07, 2003 at 09:24:05AM -0500, Branislav Selic wrote: Dear Bran, > If we really want to capture what MDA is about, it is about increasing the > levels of abstraction and automation. That should serve as the main selling > point rather than a catchy-but-misleading marketing term such as "platform > independence". While the latter may serve the near-sighted goal of getting > people interested in MDA, there will be the Devil to pay when it is > exposed. And, I'm afraid it will compromise or even discredit the whole > wonderful notion of MDA. It's probably too late to get rid of it, but let's > see if we can at least agree among ourselves to downplay it. Amen to that. When talking to personal friends in the industry - people who have no problem speaking frankly to me - they've often heard something about MDA, they like the idea, but what do they all get hung up on? You guessed it - PIMs/PSMs/platform independence in general. This is already costing us some credibility in the outside world, hence the starting point for this thread. And as Bran rightly points out, this problem is only going to increase as the MDA message gets spread further and in more detail. Yours, Laurie -- QVT-Partners: http://qvtp.org/ Homepage: http://tratt.net/laurie/ E3-I: This message has been scanned for viruses and dangerous content by UML's antivirus scanning services. From lechner@cs.uml.edu Sat Nov 8 20:50:10 2003 Received: from saturn.cs.uml.edu (saturn.cs.uml.edu [129.63.8.2]) by earth.cs.uml.edu (8.11.6/8.11.6) with ESMTP id hA91o6W30884 for ; Sat, 8 Nov 2003 20:50:06 -0500 Received: from saturn.cs.uml.edu (localhost [127.0.0.1]) by saturn.cs.uml.edu (8.12.9/8.12.9) with ESMTP id hA91o6oU047679 for ; Sat, 8 Nov 2003 20:50:06 -0500 (EST) Received: (from lechner@localhost) by saturn.cs.uml.edu (8.12.9/8.12.9/Submit) id hA91o5AU047672 for lechner; Sat, 8 Nov 2003 20:50:05 -0500 (EST) From: Bob Lechner Message-Id: <200311090150.hA91o5AU047672@saturn.cs.uml.edu> Subject: Re: MDA / Martin Fowler [What is a platform, anyway?] (Axel) To: lechner@cs.uml.edu (Bob Lechner) Date: Sat, 8 Nov 2003 20:50:05 -0500 (EST) X-Mailer: ELM [version 2.5 PL2] MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Status: RO Forwarded message: > From omg-list-errors@omg.org Fri Nov 7 11:23:48 2003 > To: Axel Uhl > Cc: adtf@omg.org, Allan Kennedy , > Andrew Watson , clive@trireme.com, > Edwin Seidewitz , > "'Martin Fowler'" , Joaquin Miller , > kc_staff@kc.com, ormsc@omg.org, Richard Mark Soley , > "Stephen J. Mellor" > Subject: Re: MDA / Martin Fowler [What is a platform, anyway?] > From: John Hogg > Date: Fri, 7 Nov 2003 11:15:45 -0500 > > Axel has captured the platform class/instance distinction well here. A > class of platforms is a technology. Examples are languages, operating > systems, processors and communication middleware. It's quite reasonable > to envision domain- or user-unique technologies, and these may in > principle be supported by software, hardware or hybrid platforms. > > Another point we should realize is that transformations from PIMs to PSMs > are not a single chain in the general case. Any system is composed of a > number of technology choices--possibly a large number, and possibly > varying in different places. (A single large legacy telecom application > may have over a dozen different messaging systems, all for reasons that > made sense at the time.) > > The PIM->PSM transformation graph is therefore a lattice. At each level > in the lattice, a number of technology decisions are possible; the overall > ordering between technology decisions is only partial, although there are > typically numerous dependencies. > > An actual application of MDA is represented as a chain or stack (in > Joaquin's appealing pi-diagrams), because the actual technology decisions > and transformations are made once for each part of a system. (But note > that different choices and therefore different chains with different > technology orderings will apply in different parts of large systems.) > > With respect to Allen, the Reference Model is still under revision. It's > crucial that we properly represent these concepts because they represent > the reality of software and system development. If the model doesn't > properly capture them, we'll change the model, not reality. > > And there *is* a more abstract platform than a/the/some UML Virtual > Machine. It's the Universal Virtual Machine(TM). It's represented > graphically by a single box with the label, "Good Things Happen" and it > perfectly implements the desired behaviour of all systems. Of course, > some details must be added during reification. > > Cheers -- John > > John Hogg > Rational software > IBM Software Group > 770 Palladium Drive > Kanata, Ontario > Canada K2V 1C8 > Voice: +1 613 591 7053 > Cell: +1 613 797 9332 > Fax: +1 613 591 7020 > mailto:hogg@ca.ibm.com > > Axel Uhl > 11/07/2003 03:26 AM > > To: ormsc@omg.org > cc: Joaquin Miller , Edwin Seidewitz > , adtf@omg.org, clive@trireme.com, "Stephen J. > Mellor" , Andrew Watson , Richard Mark > Soley , kc_staff@kc.com, "'Martin Fowler'" > , Allan Kennedy > Subject: Re: MDA / Martin Fowler [What is a platform, anyway?] > > I think Martin, Laurie, Ed, Dave and a few others have pointed out some > important aspects of platform independence in the context of MDA in this > thread, most importantly: > > "Independence" is a relative term that can only be used together with > the preposition "of", so we have to say what the class of things is that > something is independent of. > > The confusion comes about (and Allan has mentioned that a long time ago > in one of the ormsc meetings) that there is some potential for > misunderstanding in the term and how we use it. That's because you may > be led to think "How can I be independent of a platform if I eventually > need one to run the system on?" > > What most of us think the term should mean is probably related to what > Joaquin wrote, using the notion of "classes of platforms". A model that > doesn't care which element of the class it has to deal with is what we > currently call "independent of that class". When you think about it, > this in fact _is_ a bit strange, because obviously the model is specific > to the _class_, only not to any particular of its elements. I guess > that's what gets many people confused. It's just the fact that we're all > so used to terms such as "operating-system independent" or "hardware > independent" that we tried to re-use the term in a more general context. > > I had this discussion with Jishnu at one TC, and he favored the use of > the term "platform independent" _because_ of the widespread use as in > "OS independent" and the like, where it is used and understood the right > way. > > Maybe it's just a matter of explanation. If we finally can get away from > saying a PIM is independent of everything and all platforms, then we > have made good progress towards a less bullshitty, less waffly > discussion of MDA. > > BTW, I agree with the point of view that xtUML is just another platform. > And Joaquin, could you explain why you name "Smalltalk" in the list of > commonly accepted platforms but then argue that UML is not a platform > because it is a language? > > -- Axel > > www.iO-Software.com - Leading the Way in Model Driven Architecture > > -> The leading development platform for MDA > http://www.ArcStyler.com > -> Comprehensive services from IT-architecture to implementation > http://www.iO-Software.com/consulting > From lechner@cs.uml.edu Mon Nov 10 00:58:29 2003 Received: from saturn.cs.uml.edu (saturn.cs.uml.edu [129.63.8.2]) by earth.cs.uml.edu (8.11.6/8.11.6) with ESMTP id hAA5wMW06080 for ; Mon, 10 Nov 2003 00:58:22 -0500 Received: from saturn.cs.uml.edu (localhost [127.0.0.1]) by saturn.cs.uml.edu (8.12.9/8.12.9) with ESMTP id hAA5wMoU070143 for ; Mon, 10 Nov 2003 00:58:22 -0500 (EST) Received: (from lechner@localhost) by saturn.cs.uml.edu (8.12.9/8.12.9/Submit) id hAA5wMk2070149 for lechner; Mon, 10 Nov 2003 00:58:22 -0500 (EST) From: Bob Lechner Message-Id: <200311100558.hAA5wMk2070149@saturn.cs.uml.edu> Subject: Re: MDA / Martin Fowler [What is a platform, anyway?] (fwd) To: lechner@cs.uml.edu (Bob Lechner) Date: Mon, 10 Nov 2003 00:58:22 -0500 (EST) X-Mailer: ELM [version 2.5 PL2] MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Status: RO >From omg-list-errors@omg.org Fri Nov 7 11:23:48 2003 To: Axel Uhl Cc: adtf@omg.org, Allan Kennedy , Andrew Watson , clive@trireme.com, Edwin Seidewitz , "'Martin Fowler'" , Joaquin Miller , kc_staff@kc.com, ormsc@omg.org, Richard Mark Soley , "Stephen J. Mellor" MIME-Version: 1.0 Subject: Re: MDA / Martin Fowler [What is a platform, anyway?] X-Mailer: Lotus Notes Release 5.0.11 July 24, 2002 From: John Hogg Message-ID: Date: Fri, 7 Nov 2003 11:15:45 -0500 X-MIMETrack: Serialize by Router on D25ML06/25/M/IBM(Release 6.0.2CF1|June 9, 2003) at 11/07/2003 11:15:48, Serialize complete at 11/07/2003 11:15:48 Content-Type: multipart/alternative; boundary="=_alternative 005936A985256DD7_=" This is a multipart message in MIME format. --=_alternative 005936A985256DD7_= Content-Type: text/plain X-NAIMIME-Disclaimer: 1 X-NAIMIME-Modified: 1 Axel has captured the platform class/instance distinction well here. A class of platforms is a technology. Examples are languages, operating systems, processors and communication middleware. It's quite reasonable to envision domain- or user-unique technologies, and these may in principle be supported by software, hardware or hybrid platforms. Another point we should realize is that transformations from PIMs to PSMs are not a single chain in the general case. Any system is composed of a number of technology choices--possibly a large number, and possibly varying in different places. (A single large legacy telecom application may have over a dozen different messaging systems, all for reasons that made sense at the time.) The PIM->PSM transformation graph is therefore a lattice. At each level in the lattice, a number of technology decisions are possible; the overall ordering between technology decisions is only partial, although there are typically numerous dependencies. An actual application of MDA is represented as a chain or stack (in Joaquin's appealing pi-diagrams), because the actual technology decisions and transformations are made once for each part of a system. (But note that different choices and therefore different chains with different technology orderings will apply in different parts of large systems.) With respect to Allen, the Reference Model is still under revision. It's crucial that we properly represent these concepts because they represent the reality of software and system development. If the model doesn't properly capture them, we'll change the model, not reality. And there *is* a more abstract platform than a/the/some UML Virtual Machine. It's the Universal Virtual Machine(TM). It's represented graphically by a single box with the label, "Good Things Happen" and it perfectly implements the desired behaviour of all systems. Of course, some details must be added during reification. Cheers -- John John Hogg Rational software IBM Software Group 770 Palladium Drive Kanata, Ontario Canada K2V 1C8 Voice: +1 613 591 7053 Cell: +1 613 797 9332 Fax: +1 613 591 7020 mailto:hogg@ca.ibm.com Axel Uhl 11/07/2003 03:26 AM To: ormsc@omg.org cc: Joaquin Miller , Edwin Seidewitz , adtf@omg.org, clive@trireme.com, "Stephen J. Mellor" , Andrew Watson , Richard Mark Soley , kc_staff@kc.com, "'Martin Fowler'" , Allan Kennedy Subject: Re: MDA / Martin Fowler [What is a platform, anyway?] I think Martin, Laurie, Ed, Dave and a few others have pointed out some important aspects of platform independence in the context of MDA in this thread, most importantly: "Independence" is a relative term that can only be used together with the preposition "of", so we have to say what the class of things is that something is independent of. The confusion comes about (and Allan has mentioned that a long time ago in one of the ormsc meetings) that there is some potential for misunderstanding in the term and how we use it. That's because you may be led to think "How can I be independent of a platform if I eventually need one to run the system on?" What most of us think the term should mean is probably related to what Joaquin wrote, using the notion of "classes of platforms". A model that doesn't care which element of the class it has to deal with is what we currently call "independent of that class". When you think about it, this in fact _is_ a bit strange, because obviously the model is specific to the _class_, only not to any particular of its elements. I guess that's what gets many people confused. It's just the fact that we're all so used to terms such as "operating-system independent" or "hardware independent" that we tried to re-use the term in a more general context. I had this discussion with Jishnu at one TC, and he favored the use of the term "platform independent" _because_ of the widespread use as in "OS independent" and the like, where it is used and understood the right way. Maybe it's just a matter of explanation. If we finally can get away from saying a PIM is independent of everything and all platforms, then we have made good progress towards a less bullshitty, less waffly discussion of MDA. BTW, I agree with the point of view that xtUML is just another platform. And Joaquin, could you explain why you name "Smalltalk" in the list of commonly accepted platforms but then argue that UML is not a platform because it is a language? -- Axel www.iO-Software.com - Leading the Way in Model Driven Architecture -> The leading development platform for MDA http://www.ArcStyler.com -> Comprehensive services from IT-architecture to implementation http://www.iO-Software.com/consulting From omg-list-errors@omg.org Mon Nov 10 06:14:33 2003 Received: from ws3.uml.edu (ws3.uml.edu [129.63.1.203]) by earth.cs.uml.edu (8.11.6/8.11.6) with SMTP id hAABENW12496 for ; Mon, 10 Nov 2003 06:14:24 -0500 Received: from amethyst.omg.org(192.67.184.64) by ws3.uml.edu via csmap id 15585; Mon, 10 Nov 2003 05:21:15 -0500 (EST) Received: from hobbit.omg.org (hobbit.omg.org [192.67.184.3]) by amethyst.omg.org (8.12.8/8.12.8) with ESMTP id hAAC8HNA002702; Mon, 10 Nov 2003 07:08:18 -0500 Received: from eurmta01.london.eur.slb.com [134.32.26.55] by hobbit.omg.org asmtp(3.0c) id 15122; Mon, 10 Nov 2003 06:18:41 -0500 (EST) Received: from conversion-daemon.eurmta01.london.eur.slb.com by eurmta01.london.eur.slb.com (iPlanet Messaging Server 5.2 HotFix 1.12 (built Feb 13 2003)) id <0HO400701ULD7Z@eurmta01.london.eur.slb.com>; Mon, 10 Nov 2003 11:01:26 +0000 (GMT) Received: from mailrelay2.sema.co.uk (mailrelay2.London.eur.slb.com [157.203.41.14]) by eurmta01.london.eur.slb.com (iPlanet Messaging Server 5.2 HotFix 1.12 (built Feb 13 2003)) with ESMTP id <0HO40015SULDF6@eurmta01.london.eur.slb.com>; Mon, 10 Nov 2003 11:00:50 +0000 (GMT) Received: from lones1.sema.co.uk (lones1.London.eur.slb.com [157.203.40.186]) by mailrelay2.sema.co.uk (8.11.6+Sun/8.11.6) with ESMTP id hAAAxqY19407; Mon, 10 Nov 2003 10:59:58 +0000 (GMT) Received: by lones1.sema.co.uk with Internet Mail Service (5.5.2657.72) id ; Mon, 10 Nov 2003 11:00:21 +0000 Content-return: allowed Date: Mon, 10 Nov 2003 11:00:10 +0000 From: "BERRISFORD, Graham" Subject: RE: MDA / Martin Fowler To: "'Martin Fowler'" , Allan Kennedy Cc: clive@trireme.com, "Stephen J. Mellor" , ormsc@omg.org, adtf@omg.org, Andrew Watson , Richard Mark Soley , kc_staff@kc.com Message-id: MIME-version: 1.0 X-Mailer: Internet Mail Service (5.5.2657.72) Content-type: text/plain Content-transfer-encoding: 7BIT X-NAIMIME-Disclaimer: 1 X-NAIMIME-Modified: 1 Status: RO Yes. Surely the priority is to present/agree the benefits/problems that the notion of a PIM is intended to yield/resolve, rather than debate the notion of PI in an abstract way? It appears to me that the current immensely flexible definition of PI is encouraging people with radically different ambitions to cloud the market for better UML. Is there a concensus within the OMG about what a better UML would be? Personally, I want a UML that my (non-programmer) systems analysts can and want to use. Is this compatible with people turning UML into a compilable/executable language? valid for .Net and J2EE? valid for OO and non-OO programming languages? declarative? formal? And why do people want these things? Do we need more than one UML? e.g. My analysts would like a simpler UML (say, excludes features that map only to OO programming languages), with better ways to express business rules in an informal way. Such a UML would be useful in specification, though probably useless to those who want a compilable language. Then again, if, as Alan might hope, a majority in the OMG hope to see a standard that defines a UML virtual machine, then isn't that a more pressing goal than defining extra modelling/specification notations? Graham P.S. my company is being taken over and I may soon disappear from OMG lists. -----Original Message----- From: Martin Fowler [mailto:fowler@acm.org] Sent: 06 November 2003 19:25 To: Allan Kennedy Cc: clive@trireme.com; Stephen J. Mellor; ormsc@omg.org; adtf@omg.org; Andrew Watson; Richard Mark Soley; kc_staff@kc.com Subject: Re: MDA / Martin Fowler > But to respond to Martin's comments more seriously, particularly the > "platform independent malapropism". I myself said during the panel > discussion that "the most platform independent (executable) UML model is > dependent on just one platform - the UML virtual machine. The problem is > that there is no standard defining the operation of that machine." So Martin > is right, its just that he is avoiding the point that the UML machine is a > more abstract (and therefore possibly easier to program) machine than any > 3L level machine. But this is exactly the point of the Platform Independent Malapropism. The issue of whether UML/MDA is a better platform is a completely separate issue from the merits of UML/MDA. Even if I was convinced that UML/MDA was the best thing around, I would still object strongly to the talk of platform independence. I would suggest that you should argue about whether and why the UML/MDA platform is better rather than talk about a false platform independence. Otherwise I think you're just adding confusion and undermining your argument with a red herring. (And cynics would say you're trying to cover up the weakness of the UML/MDA with this red herring.) -- Martin Fowler http://martinfowler.com E3-I: This message has been scanned for viruses and dangerous content by UML's antivirus scanning services. From omg-list-errors@omg.org Mon Nov 10 12:01:32 2003 Received: from ws3.uml.edu (ws3.uml.edu [129.63.1.203]) by earth.cs.uml.edu (8.11.6/8.11.6) with SMTP id hAAH1SW25239 for ; Mon, 10 Nov 2003 12:01:28 -0500 Received: from amethyst.omg.org(192.67.184.64) by ws3.uml.edu via csmap id 14589; Mon, 10 Nov 2003 11:08:00 -0500 (EST) Received: from hobbit.omg.org (hobbit.omg.org [192.67.184.3]) by amethyst.omg.org (8.12.8/8.12.8) with ESMTP id hAAHsgNA007361; Mon, 10 Nov 2003 12:54:42 -0500 Received: from palrel13.hp.com [156.153.255.238] by hobbit.omg.org asmtp(3.0c) id 11870; Mon, 10 Nov 2003 12:05:11 -0500 (EST) Received: from xparelay2.ptp.hp.com (xparelay2.ptp.hp.com [15.1.28.65]) by palrel13.hp.com (Postfix) with ESMTP id 61EA11C0075D; Mon, 10 Nov 2003 08:55:08 -0800 (PST) Received: from hp.com (dhcp-bridgewater-bjf003.esr.hp.com [15.1.195.3]) by xparelay2.ptp.hp.com (Postfix) with ESMTP id D81271C000B7; Mon, 10 Nov 2003 08:55:06 -0800 (PST) Message-ID: <3FAFC2EA.5080506@hp.com> Date: Mon, 10 Nov 2003 11:55:06 -0500 From: Jishnu Mukerji Organization: Hewlett-Packard SGBU/MSO User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.4.1) Gecko/20031008 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Allan Kennedy Cc: Axel Uhl , ormsc@omg.org, Joaquin Miller , Edwin Seidewitz , adtf@omg.org, clive@trireme.com, "Stephen J. Mellor" , Andrew Watson , Richard Mark Soley , kc_staff@kc.com, "'Martin Fowler'" Subject: Re: MDA / Martin Fowler [What is a platform, anyway?] References: In-Reply-To: Content-Type: text/plain Content-Transfer-Encoding: 7bit X-NAIMIME-Disclaimer: 1 X-NAIMIME-Modified: 1 Status: RO Allan Kennedy wrote: >Hi Axel > >Most of the comment since my response to Clive Menhinick's original query >has been to emphasise that "Platform Independence" is a relative quality >exhibited by models/programs. I think that is understood by all including >Martin. However Martin's (entirely reasonable) beef appears to be simply >that the term is gratuitously misleading since it implies an absolute - >dependency on no platform - when it is perfectly clear that the highest >degree of platform independence is achieved when a model nonetheless >exhibits dependence on just one platform: *a* UML virtual machine. (I will >be able to say *the* UML virtual machine when it is standardised ;-). > And that would be the case for only the subset of humanity that wants to think that UML is the end all and be all of modeling or Model Driven Architect-ing.;-). Another source of confusion is the fact that the term UML means different things to different people. Some are happy to think of only the UML Core, which is shared with MOF etc. while it would appear that a predominant group of people think of UML as the whole ball of wax Infra- Super and all, even though I am yet to find anyone that has implemented the whole ball of wax. A quick test to clarify which meaning one is using might be to ask the question "Is CWM based on UML?", and depending on the answer decide where the user of the term UML is coming from. >For the record I have attached a copy of the draft MDA Reference Model >worked on by you and other participants in the ORMSC. For those who are >interested, look at the right hand side of the diagram where the various >kinds of dependency that can diminish the "platform independence" of a model >are represented. The reference model thus attempts to make explicit the >intended relativity of the term "Platform Independent". > Yes. I guess the problem that we have been facing here is the "too much of reading between the lines (or letters as the case may be)" that folks have been doing with the term PIM, giving it all sorts of absolutist connotations:-). Jishnu. E3-I: This message has been scanned for viruses and dangerous content by UML's antivirus scanning services.