Date: Wed, 14 Mar 2007 17:17:42 -0800
Reply-To: David L Cassell <davidlcassell@MSN.COM>
Sender: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From: David L Cassell <davidlcassell@MSN.COM>
Subject: Re: sending emails
In-Reply-To: <FD746D3C07FB0C478AB951C18CE266A4A929CB@aaunsw412.au.cbainet.com>
Content-Type: text/plain; format=flowed
David.Johnson@CBA.COM.AU replied:
>
>Actually Dave, I'm sorry to disagree with you after you said such nice
>things about me, but Outlook uses MAPI: the Microsoft Mailer API.
>
>The data exchanged between the SAS session and the application is
>changed by the MAPI client, which could just as easily be Outlook
>Express, Microsoft mail or any other mailer client using MAPI. You can
>demonstrate this in stark reality just by trying to send an email to
>multiple persons. The SMTP definition is that multiple addresses are
>each contained within double quotes, delimited by a space and contained
>as a group within brackets. Outlook will accept this, semi-colon or
>comma delimiters, single quoting, email addresses, address book names
>and all manner of other spurious structures: all in the name of being
>user-friendly.
>
>I seem to remember that I had no trouble attaching multiple files to an
>email message sent with SMTP, but I'd need to check since I deliberately
>avoid sending attachments, and HTML emails.
>
>
>VIM is interesting, and I experimented with that when I first tried to
>use email last millennium. We had an early Lotus email client called
>cc:Mail, and I had an early palmtop computer built by HP with a 9600Bd
>PCMCIA modem which had the cut down version of cc:Mail. I got my leave
>pass to a distant SAS conference by pointing out that I could monitor
>all my batch jobs by email while I was away. (SAS had released the
>email engine for FILENAME with the new SAS version within the previous
>few months.) VIM has some issues, but I'm afraid I didn't have time to
>experiment with these more recently when I experimented with Pegasus as
>an email client for some batch reporting. I still believe that a little
>effort on my part would have made that work as easily as I remember
>cc:Mail being set up.
>
>
>Umm, and since I am being picky (it gives you a chance to add another
>post to your post count): the size of the Proceedings for the conference
>is governed by the Executive Committee. Key SAS employees are included
>in that Committee, but the Executive have repeatedly pointed out that it
>is a User conference. When the paper proceedings were finally laid to
>rest a few years ago, we gained faster delivery of the Proceedings,
>colour for our published papers, a new layout that better handled wide
>images, and some relaxing of the page count that had applied hitherto.
>Since the proceedings are on CD, much of the extra space is now taken up
>by publishing a rolling three years of papers, which was a brilliant (if
>obvious in hindsight) decision. However, with the increase in paper
>imagery, I occasionally wonder when the limit might be imposed on file
>size rather than on page count. I'm sure someone has already looked at
>this, and I suspect we might see either a rationalisation, or a move to
>DVD. The issue might be that DVD doesn't have the market proliferation
>or standardisation of CD on computers. (And the stupid fight about
>blu-ray standards comes to mind here.)
>
>Kind regards
>
>David
>
>/* - - - - - - - - - - - - - - - - - - - - -
>It is a capital mistake to theorize before one has data.
>Insensibly one begins to twist facts to suit theories, instead of
>theories to suit facts.
>-Sir Arthur Conan Doyle
>- - - - - - - - - - - - - - - - - - - - - */
>
>-----Original Message-----
>From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of
>David L Cassell
>Sent: Wednesday, 14 March 2007 2:32 PM
>To: SAS-L@LISTSERV.UGA.EDU
>Subject: Re: sending emails
>
>David.Johnson@CBA.COM.AU sagely replied:
> >
> >You're quite right Ben, the paper doesn't mention that wretched
> >dialogue.
> >
> >It does however make the point fairly strongly that sending email
> >through an email client is likely to cause corruption to the input
>data.
> >As a specific example, the use of the "NUL" option as a Null address as
>
> >documented in the SAS FM will still cause email to be sent if "NUL" can
>
> >be matched to an address in the email client address book. My friend
> >in Praha suffered many such emails from me before I discovered and
> >proved that the client would take the data it was given and try to
> >"help" you to send a complete email.
> >
> >I remain a firm believer in email clients not being used in the path
> >between SAS and email sending because of the changes they can cause.
> >SMTP is very much better, although it does require the correct
> >construction of email address strings. I refer to specific examples
> >including the construction of multiple email addresses and the
> >"corrections" the client will perform, while the SMTP server will
> >simply refuse to process the malformed string.
> >
> >
> >As other examples of client corruption; the string content of the
> >message body can be changed, the encoding value can be altered, the
> >client can "choke" on too many emails which requires patches such as
> >the delaying of second and subsequent emails through a sleep() loop,
> >the sender value can be changed to the default sender value on the
> >client even if a different and valid email account and password are
>provided.
> >
> >
> >Thank you for the link to the dialogue beater. I knew one would be
> >created eventually.
> >
> >Kind regards
> >
> >David
> >
> >/* - - - - - - - - - - - - - - - - - - - - - It is a capital mistake to
>
> >theorize before one has data.
> >Insensibly one begins to twist facts to suit theories, instead of
> >theories to suit facts.
> >-Sir Arthur Conan Doyle
> >- - - - - - - - - - - - - - - - - - - - - */
> >
> >-----Original Message-----
> >From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of
> >ben.powell@CLA.CO.UK
> >Sent: Friday, 9 March 2007 10:19 PM
> >To: SAS-L@LISTSERV.UGA.EDU
> >Subject: Re: sending emails
> >
> >That's a great paper. I'm not sure it mentions the dialogue Outlook
> >throws up, that can be thwarted if you install ClickYes:
> >http://www.contextmagic.com/express-clickyes/
> >
> >Rgds.
> >
> >On Fri, 9 Mar 2007 14:35:20 +1100, Johnson, David
> ><David.Johnson@CBA.COM.AU>
> >wrote:
> >
> > >The conditional answer is "Yes".
> > >
> > >The conditions relate to the configuration of your SAS session and
> > >your
> >
> > >email access. But generally you can, and a detailed tutorial was
> > >presented on the subject at last year's SUGI. This paper might be of
>
> > >some assistance.
> > >
> > >http://www2.sas.com/proceedings/sugi31/256-31.pdf
> > >
> > >A situation that is similar to the one you are writing about is dealt
>
> > >with late in the paper where email directives are discussed.
> > >
> > >
> > >Kind regards
> > >
> > >David
> > >
> > >/* - - - - - - - - - - - - - - - - - - - - - It is a capital mistake
> > >to
> >
> > >theorize before one has data.
> > >Insensibly one begins to twist facts to suit theories, instead of
> > >theories to suit facts.
> > >-Sir Arthur Conan Doyle
> > >- - - - - - - - - - - - - - - - - - - - - */
> > >
> > >-----Original Message-----
> > >From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of
> > >Emanuele Castano
> > >Sent: Friday, 9 March 2007 2:25 PM
> > >To: SAS-L@LISTSERV.UGA.EDU
> > >Subject: sending emails
> > >
> > >hello all,
> > >
> > >does anyone know whether SAS can generate email messages? don't
> > >worry, i am not intending to send spam emails! i score MCQ tests for
> > >my classes in SAS, and it would be great if SAS could automatically
> > >generate an email (through outlook, for instance) so that the content
>
> > >for obs 1 on variable
> > >1 is sent to the address indicated in variable 2.
> > >
> > >for instance, i have:
> > >
> > >10 castanoe@newschool.edu
> > >
> > >and email is generated for this email address, with 10 in the text.
> > >
> > >is it at all possible?
> > >
> > >thanks in advance.
> > >
> > >emanuele
>
>Well, there was hardly room for anything else in that paper! It was
>pretty full of detail as it was. And SAS doesn't give people an extra
>50 pages for their paper. (Yet.)
>
>There are plenty of other points that are relevant, but will have to
>wait until David writes a SAS *book* on using email.
>
>For example, the typical user is at the mercy of the rules enforced by
>the sysadmins. So it may or may not be possible to attach anything. Or
>it may be possible to attach only text/html/xml files but not binaries.
>The use of attachments may invoke MIME, and that can introduce other fun
>mail-like issues. Buy your sysadmins a beer and talk to them about
>this.
>
>Or there may be another MUA (Mail User Agent) besides MS Outlook.
>Outlook has its flaws, but at least it attaches multiple files
>intelligently.
>I had plenty of headaches with Sun Mailtool on this one.
>
>And MS Outlook at least uses SMTP, instead of, say, VIM, as Lotus Notes
>insists on using. That adds a whole extra layer of complexity to the
>process.
>
>So check your MUA and your MTA (Mail Transport Agent), and check with
>your system admins to see what they will let you do without major
>reconfiguration of their mail servers.
>
>HTH,
>David
>--
>David L. Cassell
>mathematical statistician
>Design Pathways
Ugh. I did say SMTP for some weird reason. How embarrassing.
<Jon Lovitz voice>
No, I did it on purpose, to pad my posting count. That's the ticket!
I did it on purpose, and.. and I'm the *president of Pathological Liars
Anonymous. The president, and also the secretary-treasurer. And...
<\Jon Lovitz voice>
My experiences with VIM have been less than sterling. Lotus Notes
may be useful for emailing from external apps, but Lotus Notes is such
a top-down app that if one is not the Notes sysadmin then one probably
is not going to be able to make everything work the way one wants.
Perhaps a tester will get some features going, and be unable to use others.
One of the ones I have seen is a clamp-down on the use of attachments.
So YMMV.
And, rather than a move to DVD, I would rather see a concerted Gutenberg
Project kind of effort to get old hardcopy SUGI papers into electronic form.
Lex has put a couple PDFs of photocopied SUGI papers up on his site (thanks
Lex, I appreciate it!), but it's still not the same as a searchable
resource.
David
--
David L. Cassell
mathematical statistician
Design Pathways
3115 NW Norwood Pl.
Corvallis OR 97330
_________________________________________________________________
Get a FREE Web site, company branded e-mail and more from Microsoft Office
Live! http://clk.atdmt.com/MRT/go/mcrssaub0050001411mrt/direct/01/
| 
