For Programmers: Free Programming Magazines  


Home > Archive > Fortran > February 2005 > Re: On legitimacy (was: Q: F2k3 Vote Result)









You are viewing an archived Text-only version of the thread. To view this thread in it's original format and/or if you want to reply to this thread please [click here]

 

Author Re: On legitimacy (was: Q: F2k3 Vote Result)
gary.l.scott@lmco.com

2005-02-23, 4:00 pm


Michael Metcalf wrote:
> "Gary L. Scott" <garyscott@ev1.net> wrote in message
> news:111p09kp845b5ad@corp.supernews.com...
moved on[color=darkred]
> to other
>
> Exactly. This battle was lost in the late '80s. The bit intrinsics

were the
> compromise solution.


Since I know not ALL moved on (me for example, others in my area), I
say its never too late to open this programming domain back up by doing
it better than those other languages.

>
> Regards,
>
> Mike Metcalf


Dan Nagle

2005-02-25, 8:59 am

Hello,

On Thu, 24 Feb 2005 22:44:14 GMT, "James Giles"
<jamesgiles@worldnet.att.net> wrote:

>Dan Nagle wrote:
>
>In other words, it changes the requirements. That means that
>at least some implementations will have to change. But, that
>really means that they'll have to have a compiler flag to continue
>to support the old way they're doing things. This means that
>the committee really should fully examine what other existing
>implementations presently do (and whether they're bertter than
>the proposed "rationalization").


Why not read the papers and tell us?
BTW, the vendors generally support this one, IIRC.

>
>Then it should at least have KINDs corresponding directly the
>the character storage units and not require users of the feature to
>manually pack and unpack characters.


Again, read the papers. There're different kinds of typeless.

>Your "all you have to do" is addressed at the users of the language.
>My "all you "have to do" is addressed a implementors. If it costs
>vendors 1000 hours of effort, but saves 10,000 users 10 mintes each,
>it's more than justified. You seem to think the trade-off works the
>other way around. (Eliminating the need to pack and unpack binary
>data in my own programs will save *me* much more than 10 minutes,
>or 10 days even, over the lifetime of my use of a given compiler.)


You've missed the point, I'm afraid.

I never claimed that typeless does everything that bit strings do.
My point is that there's a _substantial_ cost difference between
the two. And the vendors haven't been motivated to add bit strings
as an extension, so there's, apparently, little demand.

Confusing "benefit" with "cost-benefit" is easy to do. :-)

--=20
Cheers!

Dan Nagle
Purple Sage Computing Solutions, Inc.
Sponsored Links







Also available: Server administration forum archive | Web Design forum archive | Software forum archive | Hardware reviews archive

Copyright 2008 codecomments.com