Re: One Ring to Bind Them

From: Gene Wirchenko <genew_at_mail.ocis.net>
Date: Wed, 16 Jun 2004 09:42:18 -0700
Message-ID: <sos0d09n5vgl23q1e52vtvhuhe6o9856ud_at_4ax.com>


"Dawn M. Wolthuis" <dwolt_at_tincat-group.com> wrote:

>"Eric Kaun" <ekaun_at_yahoo.com> wrote in message
>news:h2Zzc.1822$Pt.1440_at_newssvr19.news.prodigy.com...

[snip]

>That was said in a convoluted way, but my point is that language standards
>are not enough to protect my investment as soon as I opt for a tool from
>Oracle or IBM or MS or whomever.

     Particularly so when some companies consider it their duty to break standards.

[snip]

>and the others who spoke up were right that a declarative language (such as
>SQL) does have declarations that are like sentences (SELECT this, that FROM
>arelation) while a language that is spec'd -- options chosen -- is a set of

     A select is not a declaration. A declaration would be, for example, a constraint:

          accttype in ("A","B","C")

>parms (kinda RPG-like for those who don't think that means role-playing
>games). And I do agree that English as we speak it is not the goal, but

     Help, help! Acronym poisoning! For a few minutes, I was trying to figure out what Rocket-Propelled Grenades had to do with the situation. Taking out bad implementations?

     Then, I realised you probably meant the language whose acronym expands to "Report Progam Generator".

>just like the Palm Pilot taught us to write so it could understand, I do
>think that is a reasonable way to approach a language. COBOL has its charm.

     There are rules. Some are explicit, some are not. Some help, some hinder.

>Java, along with other modern languages are unnecessarily cryptic to the
>seasoned IT professional picking up the language for a first time.

     I do not like the corner cases. The switch statement in C and C-derived languages is a nearly useless thing that puts the corner case of falling through a case onto a pedestal. Yuck!

[snip]

>No, rather the opposite -- data and functions are two sides of the same
>coin. So, clearly I wasn't writing clearly. My point is that you either
>write a function directly in procedural or OO code or you spec them and spec
>when they are to be used -- 6 of one, half-dozen of the other. Having seen

     Maybe by declaration.

>so many languages without trying to collect them over my career, I have not
>yet seen one that makes an order of magnitude break-through in productivity
>for the developer. It seems to me that Java and OO languages had a good
>chance of providing large chunks for reuse, but they aren't there. It isn't
>a silver bullet, but I'm thinking the services strategy, which is language
>independent, has a chance at helping to get some of those bigger gains.
>Knock on wood.

     Or a compatible substitute? <g>

Sincerely,

Gene Wirchenko

Computerese Irregular Verb Conjugation:

     I have preferences.
     You have biases.
     He/She has prejudices.
Received on Wed Jun 16 2004 - 18:42:18 CEST

Original text of this message