pan-users
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Pan-users] GNKSA


From: Duncan
Subject: Re: [Pan-users] GNKSA
Date: Thu, 7 Jul 2011 22:31:02 +0000 (UTC)
User-agent: Pan/0.135 (Tomorrow I'll Wake Up and Scald Myself with Tea; GIT 9996aa7 branch-master)

Beartooth posted on Thu, 07 Jul 2011 17:10:27 +0000 as excerpted:

> Yup; and I shudda thunka that, too. I do have experience with a big list
> (ca. 4000 subscribers in my years), though none so humbling & gratifying
> as yours. What's more, 90+ % of them were Alpha Plus when they did break
> Internet silence.

Pardon this attempt to break my arm patting my own back a bit, but I've 
had a couple such "humbling and gratifying" experiences. =:^)  The one I 
remember best, because it was really the first, was back in the IE/OE 
groups (MS server).  By this time I had been on USENET long enough to 
know about repost requests, but those happened mostly on binary groups 
for missing parts (this was before par2 became popular, at least in the 
groups I did), and I was most active in a few select tech-related text 
groups, mostly MS IE/OE (and my ISP's) at that time.  Of course, as 
regulars here can attest, it's entirely possible for some of my longer 
text posts to compete with smaller binaries for size. =:^]

The particular event in question was when I got the first "repost 
request" for something I'd written.  Seems a regular had developed a 
habit of archiving many of my explanatory posts[1], but had gone on 
vacation.  When he came back, he saw many followups, but none apparently 
quoted my post in full (good sense, for my posts, they're not designed to 
quote in full!), and my post had by then expired off the servers (at 
least the main MS server for those groups), but he wanted my original 
post to archive, thus the "repost request"!

I tell you I was walking on air for a WEEK after that!  I had my first 
repost request, for a post of *MINE*!! =:^)  Repost requests for text 
really /are/ rare, too, and I don't think I've gotten many more of them, 
but WOW, that one sure had me riding high for awhile!  That's a memory 
I'm going to treasure for the rest of my life! =:^)

> That list was work-related; and some institutions had a policy of
> telling their people they were welcome to read, but not to post lest
> they bring down bushels of spam on the heads of sysadmins.

Too bad, when you know by the ones that do eventually break thru that 
many would have top quality insights, were they to post.  Oh, well...

> [snipperoo : stuff good even by Duncan's standards, and
> convincing]

Good, because I had to rewrite the thing multiple times.  It kept going 
off in directions that might be very interesting to write and for others 
to read, but there was no way they were going to come back to the point, 
even in the hundreds of lines my posts often become.  So when that would 
become apparent, I'd have to delete a whole section and rewrite, taking 
care to word my thought differently so it didn't head off toward 
Katmandou.  But then it'd head off toward Timbuktu instead, and I'd have 
to repeat the process. =:^P

So I'm glad it eventually turned out well.  I struggled with it enough!

> But I do claim the blind hen's occasional grain of corn : we need
> to optimize against the likely abuses of the twenty-teens, whatever we
> can guess of those.

Good-point, that.  And I enjoyed the metaphor[2]).

> Hesitant, second-hand example : I've encountered intelligent
> posters on a wide variety of topics who say they filter out all posts
> from Google Groups. (I don't go near GG, and wouldn't know how to do the
> filtering if I did.) Ought we to take some account (not flat filtering,
> but some more differentiated response) of that source?

Interesting you bring that up.  Someone posted a request for help scoring/
killfiling posts originating from GG, recently.  It's possible but not 
easy, at present, as scoring on message-id (which contains a unique GG 
identifier) isn't one of the choices directly available in the GUI, so 
one must first set it up as a subthread score, then text-edit the 
scorefile directly, changing the references header to message-id header.

If only the score creation GUI had a choice for message-id header, the 
process would be MUCH easier.

> Better example : are there ways Pan could help with such spam as
> gets past news servers? If I used Alpine as a newsreader, I bet I could
> make filters like the ones I use for Alpine email.

The scoring system allows that to some extent already.  But with the regex 
it's a bit of an arcane art[3] tho I don't think that should be changed.  
The real limit that I see, tho, is that we still don't have automated 
actions (like mark-read, delete, or auto-download) based on scoring, as 
pre-C++ rewrite pan did in the form of rules.  That would make them 
TREMENDOUSLY more useful, IMO.  That and adjusting the GUI to handle 
other headers, already scoreable (since a khaley patch some time ago) if 
you hand-edit the scorefile appropriately, would make scoring far more 
approachable for ordinary users, even while keeping the power of regex.

> And how about links to malware sites? One trick there is to install
> Dillo and make it the default browser. There's a lot it won't do,
> by design; but it's fast, and it usually gives you enough of a look to
> decide if you want to paste a URL into some other browser. Maybe Pan
> could default to using Dillo, and give the user an option to install it
> if need be, or to change the default.

Ugh.  Such a default would invite a pan dependency on dillo, and I don't 
believe anyone wants to go there.  Pan already has a reasonable default 
to the system browser (the user can even choose which system browser 
invocation method they want to use), plus a way to set it to a specific 
browser if desired.  Beyond that, IMO, it's upto[4, since I'm already upto 
three footnotes, might as well make it four] the user to set it as they 
wish.

---
Footnotes:

[1] That's a bit scary to think of on its own, that they're not just 
archived on the usual sites but specially archived by specific users 
archiving specifically me!  I better make sure I don't post something 
I'll regret, but that has always been my policy anyway.  I stand by my 
posts, saying I was wrong when necessary but never denying what I posted, 
tho sometimes I'll write up a big one and end up deleting instead of 
sending.  That used to bother me but then I realized that if I wrote that 
much about it, yet decided it wasn't something to post, I obviously had 
some stuff to work thru, and if that's what it takes to work thru it, so 
much the better.

[2] I spent 6 years in Kenya growing up at a mission school where my 
folks served as missionaries.  My dad helped sponsor a number of students 
by providing the money and housing to raise chickens.  We'd purchase day-
old chicks, flown in from Israel IIRC, which the students would raise to 
almost adult and sell.  My dad would get his money back for the next 
round and the students would have the money to pay for another couple 
semesters of schooling.  Of course there were other chickens around too.  
My sister kept a couple as pets for a year or two, and we enjoyed the 
eggs. =:^)  So your metaphor immediately took me back to childhood.

[3] Not sure what could be done about scoring arcaneness, tho, at least 
without losing its power.  Regular expressions are by definition both 
powerful and arcane, and that's what pan uses.  Plus pan's scoring system 
is a semi-standard, common to a couple other news clients as well, so 
there'd need to be a pretty good reason to change the backend, altho 
making the GUI a bit more accessible, in particular with more header 
options, and changing the backend to handle body and/or whole-message 
too, if so desired, not just the headers, would be a useful endeavor and 
within practical reach.

[4] Upto:  I'm a descriptivist, not a prescriptivist.  IMO, upto is 
sufficiently used as to be accepted as a word in the public vocabulary, 
and has the accepted precedent of into, which certainly must have been 
proscribed at some point as well.  With my usage I'm simply and 
deliberately advancing that practical acceptance, being aware of the 
controversy surrounding it but deliberately making that choice none-the-
less. =:^)

-- 
Duncan - List replies preferred.   No HTML msgs.
"Every nonfree program has a lord, a master --
and if you use the program, he is your master."  Richard Stallman




reply via email to

[Prev in Thread] Current Thread [Next in Thread]