[8069] in Perl-Users-Digest
Perl-Users Digest, Issue: 1688 Volume: 8
daemon@ATHENA.MIT.EDU (Perl-Users Digest)
Tue Jan 20 22:17:41 1998
Date: Tue, 20 Jan 98 19:00:21 -0800
From: Perl-Users Digest <Perl-Users-Request@ruby.OCE.ORST.EDU>
To: Perl-Users@ruby.OCE.ORST.EDU (Perl-Users Digest)
Perl-Users Digest Tue, 20 Jan 1998 Volume: 8 Number: 1688
Today's topics:
Re: Can unix Perl Modules be used under Win32 (Martien Verbruggen)
Re: Detecting the presence of a sub declaration <rra@stanford.edu>
Re: Directory Listing... (brian d foy)
Re: FAQ too big (Was: Re: PERL forking) (Martien Verbruggen)
Re: grabbing email adresses from webpages (Abigail)
Re: Hashes w/ case-insensitive keys <yary@apicom.com>
Re: I need a man to lick me in to a frenzy!!! <corky@ultranet.com>
J-PERL5 for WIN32 morten_beyer@bigfoot.com
Re: Odd syntax problem (Charles DeRykus)
Re: ok what am I doing wrong? <metcher@no.junk.mail>
Re: Perl 5 and ORACLE 7 on Windows 95 (Martien Verbruggen)
Re: Problems installing perl5.004_01 on HPUX10.20 (J.M. Davitt)
Re: Searching a database (Martien Verbruggen)
Re: source into binary code (Matt Heffron)
Re: source into binary code <jason@primal.ucdavis.edu>
Re: source into binary code <jason@primal.ucdavis.edu>
Re: source to binary pt. II (was Re: Simple example of <danboo@negia.net>
Re: Using perl to validate links <metcher@no.junk.mail>
Re: Why warn? (WAS: Re: Critique My Code! (Please)) (Martien Verbruggen)
Digest Administrivia (Last modified: 8 Mar 97) (Perl-Users-Digest Admin)
----------------------------------------------------------------------
Date: 21 Jan 1998 01:10:50 GMT
From: mgjv@comdyn.com.au (Martien Verbruggen)
Subject: Re: Can unix Perl Modules be used under Win32
Message-Id: <6a3huq$rrt$3@comdyn.comdyn.com.au>
In article <01bd25f3$bb5a9b80$6b9cc4ce@hsegal.dnvr.uswest.net>,
"Kevin Tonkin" <kjt111@hotmail.com> writes:
> To the best of my knowledge, it's not a problem. You may have to change
> your "/"s to "\\"s though. I don't seem to have had to do this with my
> scripts but I've read somewhere that you have to.
Not necessary. perl can deal with that.
It will depend on the module. If the module has not been tested on
win32, it might not run there. On the other hand, if it's written in
perl only, not using any system specific stuff, it might just run. It
will vary for each module..
Martien
--
Martien Verbruggen |
Webmaster www.tradingpost.com.au | In a world without fences, who needs
Commercial Dynamics Pty. Ltd. | Gates?
NSW, Australia |
------------------------------
Date: 20 Jan 1998 17:39:38 -0800
From: Russ Allbery <rra@stanford.edu>
Subject: Re: Detecting the presence of a sub declaration
Message-Id: <m3en227gh1.fsf@windlord.Stanford.EDU>
Tom Phoenix <rootbeer@teleport.com> writes:
> I'm not trying to be rude or anything here, but whatever reason could
> you have for needing to know this? It sounds as if you're poking into
> the internals of Perl, and those internals may well change in a future
> version of Perl, as I'm sure you know. You must have a good reason for
> wanting this.
That's a fair question. :) Basically, I'm trying to reimplement
UNIVERSAL::can() in a fashion that will work under 5.003. I believe
UNIVERSAL::can() actually does what I want, but I also don't believe it
worked right under 5.003, so I have to do this until 5.004 becomes widely
deployed.
The problem is this: I have a module that does autoloading of methods
using a modified version of code from AutoLoader (modified because the
module has a number of submodules for specific functions, and all methods
in a certain namespace come from one file rather than having one method
per file). I'm trying to get better handling of errors in the case of a
few common mistakes. Right now, if someone specifies arguments to a
submodule when it's loaded, the module autoloads the submodule and then
tries to call submodule_init(), which fails with an extremely unintuitive
error message if there is no _init method.
If I generate a set of predeclarations for all methods which may be
autoloaded (the same way that AutoLoader does), then if I have a way to
check to see if a method is predeclared, I can check beforehand and give
the user a much more informative error message (and possibly catch the
error and do something better than carping about it).
--
#!/usr/bin/perl -- Russ Allbery, Just Another Perl Hacker
$^=q;@!>~|{>krw>yn{u<$$<[~||<Juukn{=,<S~|}<Jwx}qn{<Yn{u<Qjltn{ > 0gFzD gD,
00Fz, 0,,( 0hF 0g)F/=, 0> "L$/GEIFewe{,$/ 0C$~> "@=,m,|,(e 0.), 01,pnn,y{
rw} >;,$0=q,$,,($_=$^)=~y,$/ C-~><@=\n\r,-~$:-u/ #y,d,s,(\$.),$1,gee,print
------------------------------
Date: Tue, 20 Jan 1998 20:11:01 -0500
From: comdog@computerdog.com (brian d foy)
Subject: Re: Directory Listing...
Message-Id: <comdog-ya02408000R2001982011010001@news.panix.com>
Keywords: from just another new york perl hacker
In article <01bd2601$49079b60$59f682c1@gateway>, "David Reid" <abb37@dial.pipex.com> posted:
>Hi. I need to capture a directory listing so I can mess with the files
>(move them after sorting and cataloguing them) and would like to do it with
>a perl script.
see opendir() and friends in the perlfunc man page.
--
brian d foy <comdog@computerdog.com>
CGI Meta FAQ <URL:http://computerdog.com/CGI_MetaFAQ.html>
Comprehensive Perl Archive Network (CPAN) <URL:http://www.perl.com>
------------------------------
Date: 21 Jan 1998 01:21:18 GMT
From: mgjv@comdyn.com.au (Martien Verbruggen)
Subject: Re: FAQ too big (Was: Re: PERL forking)
Message-Id: <6a3iie$rrt$5@comdyn.comdyn.com.au>
In article <01bd25e8$5f0a68c0$a041f9a5@dyyfc1.aca.arcochem.com>,
"Kevin Hawley" <cbpkxh@arcochem.com> writes:
> I don't have 40 years to read the FAQ. I use lots of languages
> not just perl, if I read all the FAQ's I would be one smart dude,
And if you don't even brwose through the FAQ of the language you're
working with, you will be doing yourself and your employer a big
disservice. It saves a lot of time to at least have an idea of the
questions and answers in there.
The FAQ is for me mostly the first or second document I read.
Martien
--
Martien Verbruggen |
Webmaster www.tradingpost.com.au | Make it idiot proof and someone will
Commercial Dynamics Pty. Ltd. | make a better idiot.
NSW, Australia |
------------------------------
Date: 21 Jan 1998 01:17:44 GMT
From: abigail@fnx.com (Abigail)
Subject: Re: grabbing email adresses from webpages
Message-Id: <6a3ibo$932$1@client3.news.psi.net>
Pascal Hos (paashaas@rice.edu) wrote on 1603 September 1993 in
<URL: news:34C51730.ABBB7B9B@rice.edu>:
++ I'm not familiar at all with perl, so I would like to know if it is
++ possible to collect the email addresses of webpages. Say one uses a
++ spider search and a webpage containing url's is displayed. How could one
++ obtain all the email addresses without actually opening the url's
++ themselves
Grabbing email address. Hmmm. Now, that's easy, but why don't I
feel like explaining it?
Abigail
--
perl -wle 'print "Prime" if (1 x shift) !~ /^1?$|^(11+?)\1+$/'
------------------------------
Date: Tue, 20 Jan 1998 17:44:06 -0800
From: Yary <yary@apicom.com>
Subject: Re: Hashes w/ case-insensitive keys
Message-Id: <34C552E6.404E@apicom.com>
Forcing it one way or another is the easyist way to do it,
and forcing to all upper or lower is easy:
$insensitive{"\U$key"} = $value
\U forces the following to uppercase, and $key can be any variable deref
(complex exprs like "\U${$foo{'bar'}}->[$i]" work too)
If you really want to implement a case-insensitive hash, take a look
at the docs for "tie".
-y
------------------------------
Date: Tue, 20 Jan 1998 20:13:40 -0500
From: Bob Trieger <corky@ultranet.com>
Subject: Re: I need a man to lick me in to a frenzy!!!
Message-Id: <34C54BC4.ACB@ultranet.com>
Mike Stok wrote:
>
> In article <6a2a08$1s1$1@gaia.ns.utk.edu>,
> Jay Flaherty <fty_no_spam@utk.edu> wrote:
>
> >while(!($frenzy)) {
> > lick_me();
> >}
>
> Might that be clearer as
>
> &lick_me until $frenzy;
>
You are still missing the last line:
unlink($man);
------------------------------
Date: Tue, 20 Jan 1998 19:01:29 -0600
From: morten_beyer@bigfoot.com
Subject: J-PERL5 for WIN32
Message-Id: <885344086.1525547020@dejanews.com>
I am looking for Japanese version of Perl5 for WIN32. Is this available?
Morten
-------------------==== Posted via Deja News ====-----------------------
http://www.dejanews.com/ Search, Read, Post to Usenet
------------------------------
Date: Wed, 21 Jan 1998 00:46:09 GMT
From: ced@bcstec.ca.boeing.com (Charles DeRykus)
Subject: Re: Odd syntax problem
Message-Id: <En3zGx.8Ks@bcstec.ca.boeing.com>
In article <6a11pi$2m7$1@usenet48.supernews.com>,
System Administrator <root@atrium.cardima.com> wrote:
>
>b)
> $_ = join('/', splice(split(/\//, $_), -1, 1)) if (/=/);
>
>However, in practice, (a) works, and (b) generates a syntax error right
>before the split() call:
>
splice's array requirement can be satisfied by
a block eval to generate an anonymous array:
See pg. 252 of the Blue Camel.
$_ = join '/', splice @{[split /\//, $_]}, -1, 1 if /=/;
HTH,
--
Charles DeRykus
------------------------------
Date: Wed, 21 Jan 1998 12:44:47 +1100
From: Jaime Metcher <metcher@no.junk.mail>
Subject: Re: ok what am I doing wrong?
Message-Id: <34C5530F.74CD9D9C@no.junk.mail>
Matthew Cravit wrote:
>
> In article <34b2d995.0@poobah.olywa.net>, Cliff High <chigh@vallier.com> wrote:
>
> >compiler/interpreter variable name ambiguity with which to cope. You create an
>
> That's why values in perl have prefixes - to remove the ambiguity. There's
> in fact no ambiguity where you indicate; @name is always an array context,
> and $name is always a scalar context. ($name could also be an object, but
> that's a separate problem). There is an ambiguity to the reader of the code
> in that @name is singular but refers to a value that is plural, but Perl has
> no problem with that. It's only a bad idea (as I pointed out in my other
> post in this thread) for readability purposes for the humans maintaining the
> code.
>
>snip<
> It's poor programming practice because it's confusing to the reader, but
> Perl has no problem with it.
>
.02c
IMHO it's not at all confusing - $name and @name are just as distinct
identifiers in perl as array_of_names and single_name would be in C.
Although I usually do use the plural for arrays:
$name = $names[0]
Jaime Metcher
------------------------------
Date: 21 Jan 1998 01:08:38 GMT
From: mgjv@comdyn.com.au (Martien Verbruggen)
Subject: Re: Perl 5 and ORACLE 7 on Windows 95
Message-Id: <6a3hqm$rrt$2@comdyn.comdyn.com.au>
In article <01bd25f7$309bab20$1a2bcbd0@mzelmanov.qquest.com>,
"Mikhail Zelmanov" <mzelmanov@qquest.com> writes:
> Hi.
> I installed Perl5.004_02 on WIN95 which included DBI-0.89.
> Then I installed DBI_0.90 and DBD_0.47.
> I am trying to connect Perl5 to ORACLE 7 from WIN 95.
> When I compile a test program I get this massage.
>
> C:\perl\test>perl test.pl
> DBI object version 0.89 does not match $DBI::VERSION 0.90 at
> C:\PERL\lib/DynaLoader.pm line 172.
> BEGIN failed--compilation aborted at test.pl line 5.
>
> How can I fix this problem?
How did you install DBI? It looks like you didn't do it correctly. I'd
suggest reinstalling it, and this time do it as is described in the
README.
Martien
--
Martien Verbruggen |
Webmaster www.tradingpost.com.au | A Freudian slip is when you say one
Commercial Dynamics Pty. Ltd. | thing but mean your mother.
NSW, Australia |
------------------------------
Date: 21 Jan 1998 02:06:12 GMT
From: jdavitt@ripco.com (J.M. Davitt)
Subject: Re: Problems installing perl5.004_01 on HPUX10.20
Message-Id: <6a3l6k$dh8$1@gail.ripco.com>
Anu Manne contractor for Carl Kiger (manne@nortel.ca) wrote:
: I Couldnot install perl on HPUX10.20, got following warning messages
(snip)
me too! perl 5.04_03 built on an HP/UX 9.(something) okay but 10.2 had
problems. the compilor got over the optimization problems but i
had to link it statically to get it to go. e-mail me if you need
the CFLAGS (or LDFLAGS or whatever... I'll have to look.)
------------------------------
Date: 21 Jan 1998 01:26:00 GMT
From: mgjv@comdyn.com.au (Martien Verbruggen)
Subject: Re: Searching a database
Message-Id: <6a3ir8$rrt$6@comdyn.comdyn.com.au>
In article <01bd25f3$67ce5ce0$6b9cc4ce@hsegal.dnvr.uswest.net>,
"Kevin Tonkin" <kjt111@hotmail.com> writes:
> Hi! I'm pretty new to perl and have been thrown in the deep end here!
>
> I'm trying to write a perl script that will ultimately become a cgi. The
> aim here is to search a database containing a bunch of URLs to see whether
> a visitor to my web page from a particular site is in the database. I
> guess what I'm going to be comparing is the environment variable
> HTTP_REFERER and checking to see if it's listed in the database.
>
> Does anyone have ANY idea how I would accomplish this?
open the database
check to see if the entry is there
close the database
The three above steps will greatly depend on what sort of database you
have: A real relational database server? have a look at DBI. Some text
file? Have a look at section 7 of the modules list on CPAN:
http://www.perl.com/CPAN/modules/00modlist.long.html
Martien
--
Martien Verbruggen |
Webmaster www.tradingpost.com.au | Little girls, like butterflies, need no
Commercial Dynamics Pty. Ltd. | excuse - Lazarus Long
NSW, Australia |
------------------------------
Date: 20 Jan 1998 17:34:43 -0800
From: heffron@falstaff.css.beckman.com (Matt Heffron)
Subject: Re: source into binary code
Message-Id: <6a3jbj$pbi@falstaff.css.beckman.com>
In <Pine.OSF.3.95.980120125356.31207S-100000@primal.ucdavis.edu> Jason Christian <jason@primal.ucdavis.edu> writes:
>On 20 Jan 1998, John Stanley wrote:
[... much elided ...]
>On another of those tangents...your MRI software.
Not so tangential. There are other issues w.r.t. software in healthcare
situations. Much of it is regulatory. The US FDA requires that it be
IMPOSSIBLE for the software running a system to be modified by the end
user (at least it must be very hard, so that accidental modification is
impossible. I'm not sure of the exact wording, that's not what I do here.)
> I think it possibly
>unfortunate, and likely dangerous, for the medical profession to move
>away from the standards that have served it so well for so long: peer
>review. Medical procedures have historically required full disclusure
>and review by the profession. Now "you" (or whoever) can take some
>procedures out of that system...but it makes *me* nervous.
I believe that FDA requires full source code for their review when
approving a computer system that lies within their "sphere-of-influence."
(I recall being warned that the FDA does NOT like "cute" or "humorous"
comments in code.)
This begs the question of trusting the regulatory agency. :-)
Peer review does not require full public disclosure. Scientific papers
are reviewed prior to publication. Would a system of peer review under
non-disclosure agreements make you *less* nervous? The reviewers would
have to be competitent and "respected", which means that they are quite
possibly employees of "the competition". Reviewing *my* software could
well advance their ability to compete *against* me.
Going back to people who sell their software development services (vs.
the software itself). Releasing my source code to the world advances the
ability of my competition to bid for work that is similar to work I've
already done. I can bid at the cost of revising the existing work, but
my competition must bid at the cost of full development (unless they have
*my source code*). This does not argue against providing source to a
client under the restriction that THEY do not release it without some
additional, agreed upon, compensation.
[...]
Matt Heffron
--
Matt Heffron heffron@falstaff.css.beckman.com
Beckman Instruments, Inc. voice: (714) 961-6742
250 S. Kraemer Boulevard, Brea, CA 92821 fax: (714) 961-6781
I don't speak for Beckman Instruments (or CRFG) unless they say so.
------------------------------
Date: Tue, 20 Jan 1998 18:07:41 -0800
From: Jason Christian <jason@primal.ucdavis.edu>
Subject: Re: source into binary code
Message-Id: <Pine.OSF.3.95.980120170623.31207c-100000@primal.ucdavis.edu>
On 21 Jan 1998, John Stanley wrote:
> In article <Pine.OSF.3.95.980120125356.31207S-100000@primal.ucdavis.edu>,
> Jason Christian <jason@primal.ucdavis.edu> wrote:
> >I will *certainly* pay much less per hour of your
> >time if you obfuscate your code (in whatever fashion).
>
> You will either pay the licensing fee or not. You don't get to set the
> price for my code. I do.
No you don't. No I don't. There are various levels and amounts of
software services that I am willing to buy at various prices. You
have a similar relationship for your supply of software (services).
Your unwillingness to release source code, while your right, influences
my willingness to buy. I will pay a lot more for the same
functionality if I have access to the code. *I* am willing to accept
a covenant against reuse of code by others other than myself or my
agents...but I do need to be able to show the code to programmers who
come for me later.
> >On another of those tangents...your MRI software. I think it possibly
> >unfortunate, and likely dangerous, for the medical profession to move
> >away from the standards that have served it so well for so long: peer
> >review.
>
> Yep. Except I don't think the term "peer review" applies to the
> software. The interpretation of the results perhaps, but not the
> software itself.
>
> >Now "you" (or whoever) can take some
> >procedures out of that system...but it makes *me* nervous.
>
> I really do wish you would read what is written. I am pointing out the
> dangers of "free software" when applied to this application, not trying
> to remove any procedures to make it happen.
By "procedure" I mean "methods for making-people-better". I include
in that writing code to drive apparatus. I see no reason why computer
code is different from any of the other elements of modern science and
technology, including medical science and technology: the best way to
make it foolproof is to use peer review aggressively. That is part of
the power of the internet, and the rapid development of Linux is
testimony to the applicability of that approach to large-scale
software development. If *I* were in charge of the software to drive
a big piece of medical machinery, I would want it to go through the
same kind of review as the machinery went through. Evidently your
mileage varies.
> >patents)...but I stillam *really concerned* about the inability to
> >review exactly how a machine works.
>
> There is a big difference between a knowledgable professional reviewing
> how a machine works and giving the code to the users because it will be
> "ethical" to let them debug and modify it themselves.
>
> >No doubt. But hey, if you're running the company you can ignore
> >everybody's brother's suggestions.
>
> I see you have abadoned "the real world" you sought to bring into the
> discussion earlier. No, if someone rings your telephone, you can't
> ignore them. It may be someone who wants to buy your software. It may
> be someone reporting a real bug. Every time someone calls up telling
> you how to make your code better because they know the source and can
> program, too, you lose time and money. And you can't just ignore calls
> from everyone's brother because you have caller ID and know it is one
> of them, they may be calling to order something. Every letter you get
> has to be opened to make sure it isn't a purchase order. In the real
> world, either YOU do it, and thus lose time from programming, or you
> hire someone.
Poor Linus. It Must Be Hell.
> >As to Dr. Nursely fixing the
> >software, this is another straw man.
>
> It's doctors and nurses, not Dr. Nursely. But anyway, a benefit of
> "free software" is that users can fix and modify the code. You can't
> tell me the benefit is that and then say it isn't going to happen. If
> it isn't going to happen, then there is no need to have "free
> software".
Y'know, I have a program that you can use to fix your humor filter.
>
> >The hospitals, and medical profession
> >in general, have long experience with coordinating review and
> >improvement by a large and relatively open system of experts.
>
> And a long experience of small experiments that lead to marvelous
> discoveries. The doctor has the source. He thinks he knows how to fix
> it. He tries. The hospital learns about it after the fact.
You bet. Why just last month one of the pediatricians at our local
hospital had a great idea about how to cure Altzheimers. Well, the
experiment didn't work, but what the hey, the old boy didn't really
notice.
Naah, doesn't work that way. Now, if the pediatrician *happened* to
have a bit of neurology and geriatrics in his portfolio, that might be
okay. More generally, though, the system works tolerably well at
kicking problems over to an appropriate specialist. In the case at
hand, the specialist is...someone with specialized knowledge in
computer programming. If the client (in this case the hospital) has
the code, it is a hell of a lot easier to find someone to
fix/improve/extend it than if all they have is the shrinkwrap.
> >Fine. I stand corrected. The rules for behavior within a corporate
> >environment, and in the production of a standardized product, are likely
> >to be different in many cases from those in an environment of
> >professional-service agreements. That is all I am saying.
>
> So what? What is your point?
That my statement of a code of ethics for independent PSEs does not
mean that software engineers in a corporate environment are unethical
when they do not follow all elements of it. I am answering your
criticism of my position. My point is, no doubt, not very
interesting. Nor was the criticism to which it responded.
>
> >*I* am approaching the question from the buyers side, and
> >explaining how *I* think a free-software, in the sense of readable and
> >editable source code, benefits both buyers and sellers.
>
> And I am trying to show you that this is not always the best thing.
I never argued that it was. There is a market for geegaws, and a
corporate form is often a fine way of exploiting that market.
> >> >I do not believe that someone
> >> >with training and talent in program ought to be restricted to being a
> >> >Professional Software Engineer.
> >>
> >> This is from left field. Who said this?
> >
> >Why don't you deal with the whole paragraph?
>
> I did. You left the whole paragraph, and my replies to it, in your
> quote. Don't you look at what post before you post it?
>
> I also asked you who said that someone with training and talent in
> "program" ought to be restricted to PSE. If nobody is debating this, or
> has said it, why are you arguing with ME about it? Aren't you the fellow
> who just told me that I was arguing with you over somethign you had not
> said?
Fortunately, in addition to my HumrFiltr (TM), I also have my famous
patented RantParsley. :.-10,.-3!RantParsley gives the following machine
translation:
}Well, then, on that at least we don't disagree.
That's cleared up then. Good.
>
> > I am trying, with no
> >evident success, to clarify my position, which is that
> >training-expertise-and-talen in computer programming can be, and ought
> >to be, useable either as an employee, within a corporate environment,
> >perhaps even producing shrink-wrapped geegaws (you all have spoiled
> >"widget" as the economist's standard term for an anonymous product),
> >using whatever obfuscation schemes the corporation requires, or as an
> >indepentent PSE, selling programming services to clients much like an
> >attorney sells legal services to clients.
>
> So I'll ask you again: who has said that this ought not to be the case?
> If you feel you need to clarify your position, you must beleive that
> someone else holds an opposing position. Who is it?
I have taken the liberty of running my answer through my geegaw. From
STDOUT we get:
Nobody, I guess. I'm glad that's cleared up.
> >> >Physicians and attorneys are often
> >> >employed by businesses, or are entrepreneurs in their own businesses; so
> >> >long as they do not claim to be "practicing medicine" or "practicing
> >> >law," such activities are not unethical.
> >>
> >> Back the horses up, Ramone.
> >>
> >> Let me get this straight. You are saying that you think it is not
> >> unethical for a doctor to run his own business, as long as he does not
> >> claim he is practicing medicine? You are saying the same for lawyers?
> >> By that claim, you imply that it IS unethical for a doctor to run his
> >> own business (or work for someone else) and claim he is practicing
> >> medicine. Ditto for lawyers.
> >
> >The corporate form that a medical or legal practice takes is not my
> >concern. Kaiser is a medical practice, so was the single-physician
> >practice that I visited as a child. Both are in the business of
> >providing medical services to patients, and are bound by the physicians
> >code of ethics.
>
> And in both cases those doctors make the claim that they are practicing
> medicine. You claimed it wasnt' unethical if they didn't make that
> claim, which implies that it IS unethical if they do. Now explain that.
>
> >Sometimes physicians go into the business
> >selling...geegaws. Dr. Spademan, an orthopedist, invented and sold a
> >fine ski binding. My dentist had a patent on a high-speed drill. Many
>
> Much yammering about "human ethics" and doctors who open non-medical
> businesses, deleted.
>
> What's the point? So they can start another business. How is it
> unethical?
Fortunately, we can run my reasonable response backwards through
RantParsely, to get the unreasonable version:
NOBODY EVER SAID IT WAS UNETHICAL, OR AT LEAST NOT ME, WHY DO YOU KEEP
MAKING THINGS UP?
>
> >Free Software, in the sense of readible and editable source code, has
> >the status of a proposed requirement for the code governing independent
> >PSE's.
>
> Now nice, a code that will get roundly ignored. Or not make it into the
> final document.
>
> >As a client of that profession, I support the requirement.
>
> As someone closer to that profession, I do not. But once again we see
> the propensity for people who support "free software" to push it on
> those who do not. You support it, so you want it to be a requirement.
'Requirement' is a big word. I don't think, for example, that the
state ought to require membership in the Bar. *I* insist on
membership for attorneys that I hire. *As a consumer* of programming
services, I am telling you that I very strongly prefer to receive the
source code. I am also suggesting that I am not alone in this, and
that the "free software" "protocol" can work for a lot of people.
>
> >> It is not unethical for a doctor to form a business to practice medicine
> >> and announce that he is practicing medicine. Who he works for has no
> >> bearing on the matter. As long as he is licensed in the state he is
> >> working in, he can claim that he is practicing medicine whether he works
> >> for WidgetCo as a staff doctor or for himself in his own office.
> >>
> >> Ditto for lawyers.
> >
> >That is true. If, on the other hand, he takes a job as a medical
> >examiner for Adverse Selection Insurance, Inc., it *would* be unethical
> >for him, after he has performed my medical exam for life insurance, to
> >prescribe me some medicine.
>
> Can you cite specific language from the Professional Doctor's Handbook
> that says this? As long as he has a license to practice in the state he
> practices in, he can prescribe medicines.
I'll take your word for it.
> >What about a physician who formed a
> >company for the manufacture of therapeutic geegaws? There things are
> >tricky. If there is no substitute on the market, then it is hard to
> >argue prescribing the product.
>
> Hard to argue what about prescribing the product?
Hard to argue against it.
> >If not...well, the good physician has
> >placed himself in a conflict of interest, which is the sort of thing
> >ethics codes are supposed to prevent.
>
> So you are arguing that a doctor cannot create a company for the
> production of anything that he might suggest a patient purchase? That is
> the only way to prevent a conflict of interest.
The physician has an ethical duty to prescribe the product that best
suits his patients. He may well have a fiduciary responsibility to
his stockholders to sell his geegaw. No doubt most physicians have
the ethical guts to place their physician ethics over their pecuniary
motives or fiduciary duty. But there is a conflict. That is *precisely*
what is meant by "conflict of interest," and the avoidance and resolution
of such conflicts is one of the primary functions of ethical codes.
> >I hire a professional to help me through the intricacies of bits and
> >bytes. If you are taking my dollars to fund the development of a
> >geegaw...you have placed yourself in a conflict of interest.
>
> Excuse me? So now ANYONE who takes money to write code is in a conflict
> of interest? You are a hoot.
Nooh. If the development for later sale of a geegaw is your business,
or if you are employed by an enterprise who is in that business, then
there is no conflict. If you are taking money from a client to
perform services, then you must be careful not to let your dreams of
future jackpots interfere with your provision of services. An ethical
code can help. Others have dealt with these issues before you. "Re-use
their code."
> >I bought
> >your time and your expertise, which you ought, under any reasonable code
> >of professional ethics, to put to the best interests of your client,
> >while you are serving him.
>
> You paid for certain services. You do not own me. You do not own
> everything I create, unless I agree to that before we start doing
> business.
Indeed. As a matter of sound professional practice, systematically
agreeing that I can *at least fix the damned code* is not a bad idea.
All a code of professional ethics is is a systemic prior agreement,
which the profession has determined to be in its mutual joint
interest.
> >Locking up the code, so that after you flip
> >out to follow Rush Limbaugh on tour in a beat-up BMW van I cannot fix
> >the bugs that appear, does not, in my book, count as looking after the
> >best interests of the client.
>
> And at the time I am following Rush, you are no longer paying for my
> services. I am no longer required to look out for your best interests.
No you're not. But handing over unmaintainable code is a curious definition
of "best interests".
> >If you want to be paid more for being a PSE...fine.
>
> Oh, this is a dilly. You can now produce anything where I said I wanted
> to be paid more for being a PSE? No? I didn't think so. A fine straw
> man.
Not a straw man. Mine, mine, all mine.
> >I pay well for a good attorney, too. But the ethics of
> >the geegaw manufacturer, while undoubtedly laudable, have little to do
> >with "client-server protocols,"
>
> I guess you don't understand what "client-server protocols" means, if
> you beleive that just because you once paid me to write a program for
> you we have one going on.
Hey, I stole the words, not the meaning. Alas, most human
transactions are less easily codified and anonymized than the passing
around of little packages. For human transactions, things like
"reputation" and "reliability" are sometimes important. If I made my
living selling software services, I might be concerned about developing a
reputation for fly-by-night operations.
We have employed the same accountant for about 20 years now.
Attorneys we use rarely. I see no reason not to expect to keep going
back to our programmer again and again. If he bails, though, at least
we have the code.
---------------------------------------------------------------------------
Jason Christian University of California, Davis
jason@primal.ucdavis.edu Agricultural and Resource Economics
Office:(530)752-1357 FAX:(530)752-5614 Davis, CA 95616
------------------------------
Date: Tue, 20 Jan 1998 18:15:59 -0800
From: Jason Christian <jason@primal.ucdavis.edu>
Subject: Re: source into binary code
Message-Id: <Pine.OSF.3.95.980120181058.31207d-100000@primal.ucdavis.edu>
On 20 Jan 1998, Matt Heffron wrote:
> In <Pine.OSF.3.95.980120125356.31207S-100000@primal.ucdavis.edu> Jason Christian <jason@primal.ucdavis.edu> writes:
>
> >On 20 Jan 1998, John Stanley wrote:
>
> [... much elided ...]
>
> >On another of those tangents...your MRI software.
> Not so tangential. There are other issues w.r.t. software in healthcare
> situations. Much of it is regulatory. The US FDA requires that it be
> IMPOSSIBLE for the software running a system to be modified by the end
> user (at least it must be very hard, so that accidental modification is
> impossible. I'm not sure of the exact wording, that's not what I do here.)
>
> > I think it possibly
> >unfortunate, and likely dangerous, for the medical profession to move
> >away from the standards that have served it so well for so long: peer
> >review. Medical procedures have historically required full disclusure
> >and review by the profession. Now "you" (or whoever) can take some
> >procedures out of that system....but it makes *me* nervous.
>
> I believe that FDA requires full source code for their review when
> approving a computer system that lies within their "sphere-of-influence."
> (I recall being warned that the FDA does NOT like "cute" or "humorous"
> comments in code.)
> This begs the question of trusting the regulatory agency. :-)
>
> Peer review does not require full public disclosure. Scientific papers
> are reviewed prior to publication. Would a system of peer review under
> non-disclosure agreements make you *less* nervous? The reviewers would
Sure, that works fine.
> have to be competitent and "respected", which means that they are quite
> possibly employees of "the competition". Reviewing *my* software could
> well advance their ability to compete *against* me.
>
> Going back to people who sell their software development services (vs.
> the software itself). Releasing my source code to the world advances the
> ability of my competition to bid for work that is similar to work I've
> already done. I can bid at the cost of revising the existing work, but
> my competition must bid at the cost of full development (unless they have
> *my source code*). This does not argue against providing source to a
> client under the restriction that THEY do not release it without some
> additional, agreed upon, compensation.
The bigger the project, the more likely it is that restrictions on
re-use can be worked out. Enforcement is certainly difficult. Better,
perhaps, to move away from the restrictions, and get the vendors to
bid the initial contract at full cost, rather than lowballing in hopes
of gaining an edge on later work (which is how an economist might
analyze behavior in a multistage auction). If you got full cost, on
the understanding that the code was then free, then you would have no
complaint about later reuse.
> [...]
>
> Matt Heffron
> --
> Matt Heffron heffron@falstaff.css.beckman.com
> Beckman Instruments, Inc. voice: (714) 961-6742
> 250 S. Kraemer Boulevard, Brea, CA 92821 fax: (714) 961-6781
> I don't speak for Beckman Instruments (or CRFG) unless they say so.
No fair bringing facts in, eh.
---------------------------------------------------------------------------
Jason Christian University of California, Davis
jason@primal.ucdavis.edu Agricultural and Resource Economics
Office:(530)752-1357 FAX:(530)752-5614 Davis, CA 95616
------------------------------
Date: Tue, 20 Jan 1998 19:47:38 -0500
From: Dan Boorstein <danboo@negia.net>
Subject: Re: source to binary pt. II (was Re: Simple example of Select needed please)
Message-Id: <34C545AA.561A1EAB@negia.net>
brian d foy wrote:
>
> In article <34C51E82.C4D53499@ixl.com>, Dan Boorstein <dboorstein@ixl.com> posted:
>
> >Dave Barnett wrote:
> >> "Security through obscurity is no security at all."
> >> -comp.lang.perl.misc newsgroup posting
> >
> >not this again! and now it's no longer just a "dangerous miscarriage
> >of the truth", but devolved into being no security. all that ranting
> >and raving and the end result is a misquoted, baseless assertion.
>
> the group did exist before your recent thread, as did this quote.
> Dave's had it in his .sig since at least November of last year.
i stand corrected on the quote. please accept my apologies dave.
however, i still think it is more wrong than ever.
my thread? how on earth do you get that? i was not:
a) the primary author
b) the most frequent poster
c) the largest user of bandwidth
by the same logic we get:
a) Mr. Chow Wing Siu led to the fall of comp.lang.perl.misc
b) this is all larry wall's fault. he started perl.
c) brian d. foy is to comp.lang.perl.misc as ice age is to dinosaurs
(i thought i saw your name in there a few times)
anyway what is wrong with this thread? it is certainly germane to this
news group. made evident by its popularity. are not security issues and
source control issues relevant to a language which is largely delivered
in source format? i say more so.
at one point i was accused of trolling by another reader of this group.
needless to say this was not appreciated, and to be honest i never
intended this to be a fiasco. i was merely stating my beliefs. it
appears to me that because i twice disagreed with a very outspoken
and respected perl teacher/author, i was labelled a blasphemer and
"stoned" for my crimes. i am beginning wonder if the 'culture of perl'
has been supplanted by the 'cult of perl'.
> calm down.
i've been calm from the start. this is an inherent problem of the
medium. as you read my monospaced textual words you are free to
associate whatever emotion you might like (hence the evolution of
smilies). many people read my posts and imagine me jumping around,
upset about what others have posted. if you knew me you would laugh at
the thought. please don't be presumptuous about my state of
excitement. :) <-- (is that better?)
calmly and delightfully responding with rational explanations,
dan boorstein
------------------------------
Date: Wed, 21 Jan 1998 12:07:11 +1100
From: Jaime Metcher <metcher@no.junk.mail>
Subject: Re: Using perl to validate links
Message-Id: <34C54A3F.16855D0A@no.junk.mail>
Martin Vorlaender wrote:
>
> nikkim@senate.leg.state.mn.us wrote:
> : Is it possible to use Perl to check if HTML links on a page are active?
> : I would like to add this capability to our website so users will not get
> : a standard error message when a link is no longer valid and I can redirect
> : them elsewhere.
>
> Have a look at Randal Schwartz' WebTechniques column #14, at
> http://www.stonehenge.com/~merlyn/WebTechniques/col14.html
> for a tool that verifies links.
>
> If your website runs on Apache (or something sufficiently similar),
> look into the possibilities of the 'ErrorDocument' directive.
>
> cu,
> Martin
>
> P.S.: Thanks, Randal, for these columns. They make a de- and enlightening
> reading.
Two things I've added to my link checker (twice the size and half the
speed of Randal's :-( ) is checking for missing NAME anchors and mailto:
hrefs.
The NAME part is easy - every time you see a URL go by that has a #,
split off the last part
and remember it in a hash:
my ($path, $frag) = split (/#/, $anchor);
and every time you see a name anchor go by, remember it as well. Once
you've collected them all, check that the two lists match.
The mailto: is harder because mail servers won't always tell you if a
user exists. About the best I could do was this sort of thing:
.
.
.
# Check mailto links by opening an SMTP channel to the host and
sending the EXPN command
# There are many reasons why the EXPN command might fail - only
codes 550 (mailbox not available)
# and 553 (mailbox name invalid) indicate a bad address
<snip>
if (my $smtp = Net::SMTP->new ($check_url->host, Hello =>
"Yam")) {
$smtp->expand($check_url->user);
$bad_mailto = ($smtp->code =~ /55(0|3)/);
}
else {
$fail_code = " (Couldn't contact server)";
}
<snip>
$check_url is a URI:URL object.
I would have posted the whole script, but after reading Randal's I'm too
embarassed.
Jaime Metcher
------------------------------
Date: 21 Jan 1998 01:19:17 GMT
From: mgjv@comdyn.com.au (Martien Verbruggen)
Subject: Re: Why warn? (WAS: Re: Critique My Code! (Please))
Message-Id: <6a3iel$rrt$4@comdyn.comdyn.com.au>
In article <885315855.97255428@dejanews.com>,
dg50@chrysler.com writes:
> From where I sit,
>
> "warn "foo\n";
>
> is functionally equivelent to:
>
> "print STDERR "foo\n";
It isn't though. If you had had a look at the documentation with
perldoc -f warn
you would have known that warn does a lot more than just dump a message
on STDERR. That is just the _default_ behaviour. You can however set a
handler for $SIG{__WARN__}, which allows you to do almost anything on
a warning.
Martien
--
Martien Verbruggen |
Webmaster www.tradingpost.com.au | I took an IQ test and the results were
Commercial Dynamics Pty. Ltd. | negative.
NSW, Australia |
------------------------------
Date: 8 Mar 97 21:33:47 GMT (Last modified)
From: Perl-Request@ruby.oce.orst.edu (Perl-Users-Digest Admin)
Subject: Digest Administrivia (Last modified: 8 Mar 97)
Message-Id: <null>
Administrivia:
The Perl-Users Digest is a retransmission of the USENET newsgroup
comp.lang.perl.misc. For subscription or unsubscription requests, send
the single line:
subscribe perl-users
or:
unsubscribe perl-users
to almanac@ruby.oce.orst.edu.
To submit articles to comp.lang.perl.misc (and this Digest), send your
article to perl-users@ruby.oce.orst.edu.
To submit articles to comp.lang.perl.announce, send your article to
clpa@perl.com.
To request back copies (available for a week or so), send your request
to almanac@ruby.oce.orst.edu with the command "send perl-users x.y",
where x is the volume number and y is the issue number.
The Meta-FAQ, an article containing information about the FAQ, is
available by requesting "send perl-users meta-faq". The real FAQ, as it
appeared last in the newsgroup, can be retrieved with the request "send
perl-users FAQ". Due to their sizes, neither the Meta-FAQ nor the FAQ
are included in the digest.
The "mini-FAQ", which is an updated version of the Meta-FAQ, is
available by requesting "send perl-users mini-faq". It appears twice
weekly in the group, but is not distributed in the digest.
For other requests pertaining to the digest, send mail to
perl-users-request@ruby.oce.orst.edu. Do not waste your time or mine
sending perl questions to the -request address, I don't have time to
answer them even if I did know the answer.
------------------------------
End of Perl-Users Digest V8 Issue 1688
**************************************
