Shamil Salakhetdinov
shamil at users.mns.ru
Tue Mar 28 05:43:34 CST 2006
Folks, I just wanted to add, please, don't get me as a "mad scientist", from
my writing like in my previous posting.
Yes, I do realize that what I write may sound "a bit crazy".
Yes, I do know that good solid modern programming can be done without OOP.
All depends on context.
But OOP is helpful in this quickly changing world to create and keep well
harmonic solutions/software architectures.
IMO "class-less" programming is less useful here and results in many
overhead expenses, which OOP allows to avoid in long run.
The only concept of this "class-less" relatively flexible programming I know
are C++ template functions but this is specific C++ language construct,
which doesn't exist in VB6/VBA/VB.NET...
...when programming on RSX-11M macro assembler(PDP-11) I did investigate
this OS assembler sources because I had a task, which needed to "dynamically
patch" one bit in the core of this OS. These assembler sources were a
fantastic software code - so well the parts of their functionality were
packed and reused. I have never seen so well crafted code like that! And
this code was "class-less" of course. But writing OSes is rather stable
area, with a few releases over years - different context I mean, which is OK
for "class-less" programming...
<OT>
And I wanted also to note that I have many other interests in this life like
sports, traveling when money allow, I have a big family, which I work for
rather heavy everyday as many of you do. And I see this quickly changing
world around me, and I'm still quickly loosing my head with a "women in my
hands"(pun intended :)) like it was several days ago when I danced on a
party several times with a nice women and my wife(beautiful women) who was
also on that party asked me after it - "Did you like her?"...
</OT>
Shamil
----- Original Message -----
From: "Shamil Salakhetdinov" <shamil at users.mns.ru>
To: "Access Developers discussion and problem solving"
<accessd at databaseadvisors.com>
Sent: Tuesday, March 28, 2006 2:09 PM
Subject: Re: [AccessD] Access XP forms bound to ADO recordsets
>> Is it simply to force compliance with the selected interface
> John,
>
> For this thread practical testing task
> (http://smsconsulting.spb.ru/download/tests/a2dds.htm)
> interfaces(Implements
> feature) is used to:
>
> - have clearly defined interfaces to bind a form to its datasource(s) and
> to
> have this form communicating with custom class(es) handling runtime
> events;
> - have prepared "socket interfaces"/pluggable test-bed to easily
> substitute
> one solution's code with the others. This substitution can be done
> dynamically(on run-time) or statically(on design time).
>
> I don't have a good web page link to read on interfaces(Implements
> feature).
> This concept of (abstract) interfaces has many practical applications.
> On the other hand there is a lot of hype around it.
> And there are different opinions is this concept good(has more positive
> effect) or bad(has almost only negative effect) in solving everyday tasks
> of
> practical programming especially in custom programming of (relatively
> simple) tasks for small-/middle-size businesses.
>
>>From "rocket computer science" point of view interfaces are the base to
> implement in practice a subset of what is called also "Liskov Principle" (
> http://en.wikipedia.org/wiki/Liskov_substitution_principle) and its
> consequence "don't talk to stranger" OOP principle - the "Law of
> Demeter" -
> http://www.site.uottawa.ca:4321/oose/index.html#LawofDemeter
>
> My own understanding of these concepts and their applicability in
> practical
> tasks fluctuates with time.
> But only in positive spectrum part of opinions.
> And I see more and more practical applications of this concept.
> And especially in programming for small-/middle-size businesses or
> programming shareware tools - this is the market(IMO) which urge for agile
> eXtreme Programming(XP) and Test Driven Development (TDD) as one of the
> most
> powerful and flexible practical methods of XP. And the interfaces are a
> corner stone concept of TDD and code refactoring in its modern
> understanding - http://martinfowler.com/.
>
> The best book (IMO) I have ever read on such advanced concepts is "Object
> Thinking" by David West (http://www.techbookreport.com/tbr0083.html) - "a
> kind of philosophical journey towards an understanding of 'object
> thinking."...
>
> The best practical books to get broad vision on this concept and to see
> real
> samples are collected here I think
> (http://www.awprofessional.com/series/series.asp?st=44117&rl=1) - I have
> just found it.
>
> Here is one of my exersizes -
> http://smsconsulting.spb.ru/patterns/labs/ObserverPatternLab.htm ...
>
> Here is a very quick introduction to OOP concepts including interfaces -
> http://java.sun.com/docs/books/tutorial/java/concepts/
>
> "All that jazz"/buzz of the modern understanding of programming has got a
> lot from the works of Christopher Alexander
> (http://www.greatbuildings.com/architects/Christopher_Alexander.html),
> Smalltalk community and from the work of GoF("Gang of Four" - Erich Gamma,
> Richard Helm, Ralph Johnson, John M. Vlissides) - "Design Patterns:
> Elements
> of Reusable Object-Oriented Software"
> http://www.awprofessional.com/bookstore/product.asp?isbn=0201633612&rl=1
> and
> their followers as well as from the works/books of Kent Beck - "Embrace
> change","Test Driven Development by Example" etc.
> (http://search.barnesandnoble.com/booksearch/isbninquiry.asp?z=y&pwb=1&ean=9780321278654)
>
> That my e-mail may look a little bit more theoretical than practical - I
> wanted to outline by that not my "shift" to theoretical computer
> science/programming but my opinion that a modern advanced programmer has
> to
> have a broad vision/broad context picture of what are all that nowadays
> OOA&D principles, what associations they are based on and where they go...
>
> As a conclusion I must say I didn't read even a half of the sources/books
> I
> referred above - I just compiled their list now as a reply on your
> question
> :)
> If I ever had an opportunity for half an year paid vacations (dreams,
> dreams...) I'd spend the first part in lazy doing nothing and the second
> reading/investigating all that stuff I referred above - I do think (and I
> do
> know from what I read, found by myself or in books/articles and use in my
> practice) that this is a "nuclear power" stuff with a broad applicability
> in
> practical everyday work of every modern programmer.
>
> And I must outline as a final conclusion - reading all these good books
> and
> artciles gives a lot to not "reinvent the wheel" (because they talk about
> natural things every programmer will come to by themselves sooner or
> later)
> but only everyday practice in using these methods will make a developer
> agile. This "nuclrear stuff" is not easy and one without enough practice
> may
> have easily lost themselves in many classes of the solutions created based
> on these principles...
>
> I'm not there yet (where I should have been as I think after I have got
> where these principles lead) - "most practioners would agree that it takes
> a
> few years to really 'get' objects, unlike the weeks or months it takes to
> learn the syntax of a language."...
>
> Shamil
>
> P.S. Of course one can say they can program without objects etc. - yes, I
> also can do that - it all depends on context of a practical task/project
> to
> be developed....
>
> ----- Original Message -----
> From: "John Colby" <jwcolby at ColbyConsulting.com>
> To: "'Access Developers discussion and problem solving'"
> <accessd at databaseadvisors.com>
> Sent: Monday, March 27, 2006 6:02 PM
> Subject: Re: [AccessD] Access XP forms bound to ADO recordsets
>
>
>> Shamil,
>>
>> Can you give me a quick rundown (or a web page to read) about interfaces
>> and
>> why you use them? Is it simply to force compliance with the selected
>> interface (ensure that all methods / properties are implemented) or is
>> there
>> some larger reason?
>>
>>
>> John W. Colby
>> www.ColbyConsulting.com
>>
<<< tail skipped >>>