Re: [clo-devel] Re: [lists] Re: Re: Payment Past Due, account
Anthony Ventimiglia <anthony@ventimiglia.org> writes:
How does mailman work, If it recieves mail the same way user mail is delivered, then it should be trivial to pipe it through a spam filter and procmail along the way. Then procmail could isolate spam and send normal messages along their merry way.
That is true but one of the things I do like with the solution I came up with for the place I work is that the spamchecking is done at SMTP-time and if it is judged to be spam it gets rejected there; no collateral spam via bounces etc. Erik.
Erik Enge writes:
Anthony Ventimiglia <anthony@ventimiglia.org> writes:
How does mailman work, If it recieves mail the same way user mail is delivered, then it should be trivial to pipe it through a spam filter and procmail along the way. Then procmail could isolate spam and send normal messages along their merry way.
That is true but one of the things I do like with the solution I came up with for the place I work is that the spamchecking is done at SMTP-time and if it is judged to be spam it gets rejected there; no collateral spam via bounces etc.
The problem I have with that is the posibility of false positives. My filter is at a 98.17% success rate, but that still leaves the posibility that mail I want will be seen as spam. I've been thinkin about writing a script to send a reply to all spammers that their mail was rejected as spam, and possibly telling them a way to resend so it doesn't get rejected. While this seems a little silly, it's like giving the keys to a burglar, I don't think any spammers will actually read a response, but someone who is a false positive will read the response. Using this method, all spam could be replied to and sent to /dev/null. Now back to the subject since we talked about it, I'd like to start a project "cl-bayes" that will be a Bayesian Patter filtering library. If we can get someone to write a mail parsing library, putting the two together to make a CL spam filter will be trivial. Thanks, Anthony -- (incf *yankees-world-series-losses*)
On Fri, Nov 14, 2003 at 03:23:56PM -0500, Anthony Ventimiglia wrote:
about writing a script to send a reply to all spammers that their mail was rejected as spam, and possibly telling them a way to resend so it
Of topic, but... don't do that! Spammers routinely forge their addresses, usually using a live spammee's address as their own. (So it passes reachability tests, etc.) So if you reply to spam, you're only generating *more* spam: spam that goes innocent people. Cheers, -- Nikodemus
On Fri, Nov 14, 2003 at 03:23:56PM -0500, Anthony Ventimiglia wrote:
Now back to the subject since we talked about it, I'd like to start a project "cl-bayes" that will be a Bayesian Patter filtering library.
Oops, sorry. I missed this in my first reply. ;) I must echo the questions of Erik and Mario. Also, IMHO cl-bayes is really not quite kosher as far as names go: there are a zillion things a bayenesian filter can do, and a zillion ways it can be implementead. CL-BAYES and BAYES both suggests a certain naturality. How about some name that contains "bayes", but is not "it". bayesir, for example Cheers, -- Nikodemus
On Mon, 2003-11-17 at 06:54, Nikodemus Siivola wrote:
On Fri, Nov 14, 2003 at 03:23:56PM -0500, Anthony Ventimiglia wrote:
Now back to the subject since we talked about it, I'd like to start a project "cl-bayes" that will be a Bayesian Patter filtering library.
Oops, sorry. I missed this in my first reply. ;)
I must echo the questions of Erik and Mario.
Also, IMHO cl-bayes is really not quite kosher as far as names go: there are a zillion things a bayenesian filter can do, and a zillion ways it can be implementead. CL-BAYES and BAYES both suggests a certain naturality. How about some name that contains "bayes", but is not "it".
I wonder if it's really a good thing for clo admins to exercise this much control over projects, including their names. Perhaps it would be better to just dispense names on a first come, first serve basis? I'm worried that this will come across as heavy handed. -- Miles Egan <miles@caddr.com>
Miles Egan <miles@caddr.com> writes:
I wonder if it's really a good thing for clo admins to exercise this much control over projects, including their names. Perhaps it would be better to just dispense names on a first come, first serve basis? I'm worried that this will come across as heavy handed.
I cannot say I disagree with you. Maybe we should make clear in a case like this that it is just a comment and that approval won't depend on accomodating it - at least this is what I understand was the intent of Nikodemus. Regards, Mario.
On Mon, Nov 17, 2003 at 06:06:22PM +0100, Mario Mommer wrote:
like this that it is just a comment and that approval won't depend on accomodating it - at least this is what I understand was the intent of Nikodemus.
Yes. That was the intent, though I was rather an ass about it. Cheers, -- Nikodemus
Miles Egan writes:
On Mon, 2003-11-17 at 06:54, Nikodemus Siivola wrote:
On Fri, Nov 14, 2003 at 03:23:56PM -0500, Anthony Ventimiglia wrote:
Now back to the subject since we talked about it, I'd like to start a project "cl-bayes" that will be a Bayesian Patter filtering library.
[snip]
Also, IMHO cl-bayes is really not quite kosher as far as names go: there are a zillion things a bayenesian filter can do, and a zillion ways it can be implementead. CL-BAYES and BAYES both suggests a certain naturality. How about some name that contains "bayes", but is not "it".
[snip]
I wonder if it's really a good thing for clo admins to exercise this much control over projects, including their names. Perhaps it would be better to just dispense names on a first come, first serve basis? I'm worried that this will come across as heavy handed.
Miles, I can understand your point, but I think this is kind of a special case. The proposal kind of came out of a thread I had with Nikodemus about spam filtering for clo, so I don't really feel like he was overstepping any bounds by making the suggestion. And really that's all it was, a suggestion, no more no less. In this case (as I have posted in the reply to his suggestion) I really didn't have much of an attatchment to the name, and in fact I welcome any input if someone has ideas for a more applicable name. Cheers, Anthony -- (incf *yankees-world-series-losses*)
On Mon, 2003-11-17 at 14:12, Anthony Ventimiglia wrote:
Miles, I can understand your point, but I think this is kind of a special case. The proposal kind of came out of a thread I had with Nikodemus about spam filtering for clo, so I don't really feel like he was overstepping any bounds by making the suggestion. And really that's all it was, a suggestion, no more no less.
In this case (as I have posted in the reply to his suggestion) I really didn't have much of an attatchment to the name, and in fact I welcome any input if someone has ideas for a more applicable name.
Ok. Glad to hear it. I think I'm particularly sensitive to these kinds of issues because of the authority the clo domain will confer in the minds of a lot of Lispers, but it sounds like you're all working things out. :) -- Miles Egan <miles@caddr.com>
Miles Egan <miles@caddr.com> writes:
Ok. Glad to hear it. I think I'm particularly sensitive to these kinds of issues because of the authority the clo domain will confer in the minds of a lot of Lispers, but it sounds like you're all working things out. :)
I think we are all trying. And your comments are welcome help. Thanks! Regards, Mario.
On Mon, Nov 17, 2003 at 08:40:03AM -0800, Miles Egan wrote:
I wonder if it's really a good thing for clo admins to exercise this much control over projects, including their names. Perhaps it would be better to just dispense names on a first come, first serve basis? I'm worried that this will come across as heavy handed.
Ouch. Miles, very good point. And somewhat embarassing as I've agitated for first-come, first-served paradigm myself... ;) I think that we might add naming guidelines to the FAQ, along with the note that tey are not compulsory. Here's a draft: --snip-- Naming Guidelines You're free to name your project almost any way you like. Names are given out on "first come, first served" basis. Common-lisp.net reserves the right to veto any name, but we really don't expect to need to use that right unless there are potential legal issues involved. More likely we just tell you if we don't like the name, but you're free to use it anyway. However, here are some guidelines: * Informal namespace sharing with http://www.cliki.net/. If there is a Cliki page with the same name as your project name but is not related, please reconsider. * Bindings to foreign libraries should generally be named "cl-foo", where foo is the foreign library's name. Eg. libcurses binding would be cl-curses. * Common Lisp implemntations of standards or semi-stadard foreign libraries should be named "foo", where foo is the common sobriquet for the standard. * As an alternative to the to above, various puns that insert "l" or "cl" somewhere in the name are also popular. * In all other cases "namelike" names are preferred. Eg. Osicat, Cells. --snap-- Comments? Cheers, -- Nikodemus
Nikodemus Siivola writes:
Miles, very good point. And somewhat embarassing as I've agitated for first-come, first-served paradigm myself... ;)
I think that we might add naming guidelines to the FAQ, along with the note that tey are not compulsory.
I think that's a good idea, one of the outcomes I'd like to see from common-lisp.net is some sort of standardization recommendations, somewhat analgous to GNU. This should of course be recommendations, but they should be made nonetheless. For instance one of the great things about Lisp is Docstrings, but unfortunatley too many popular libraries have virtually no documentation, I'm amazed when I read the source of some libraries and barely find any comments. I think making a recommendation that all exported functions should be well documented is a food idea.
* As an alternative to the to above, various puns that insert "l" or "cl" somewhere in the name are also popular.
I always like pun names, so how about I change it to "bayescl" (pronounced "basil"). -- (incf *yankees-world-series-losses*)
Nikodemus Siivola writes:
On Fri, Nov 14, 2003 at 03:23:56PM -0500, Anthony Ventimiglia wrote:
Now back to the subject since we talked about it, I'd like to start a project "cl-bayes" that will be a Bayesian Patter filtering library.
Oops, sorry. I missed this in my first reply. ;)
I must echo the questions of Erik and Mario.
For all of you LGPL, sorry I forgot to mention that, sorry I guess my request was a bit informal.
Also, IMHO cl-bayes is really not quite kosher as far as names go: there are a zillion things a bayenesian filter can do, and a zillion ways it can be implementead. CL-BAYES and BAYES both suggests a certain naturality. How about some name that contains "bayes", but is not "it".
I have to say I don't see your point, but I also haven't given much thought to the name. My original Library I did in C++ is called libbpfl (Bayesian Pattern Filtering Library), and it is a specific goal of mine to make it useful as a general purpose Bayesian filter, as I've wanted to experiment with different applications for Bayesian Filters (most interesting to me would be image and sound pattern recognition). Now I pretty much intent to make the Lisp library very similar in it's use to my C++ library, so maybe I should stick to something similar. But to tell you the truth I wasn't wild about bpfl, I think Bayes and Bayeslib had already been taken at sourceforge. So the name cl-bayes was really not much of a big deal, to be honest I kind of have an aversion to the whole cl- prefix that predominates, but on the other hand it's kind of like the lib prefix for C packages, when you see cl- you know it's a Common Lisp (or library application package). But as far as the cl- has to do with the name the only reason I had that was that you would knoe it is a common lisp library for bayesian filtering. I don't see how that doesn't reflect the zillion potential uses for Bayesian filters. I have no deep love for cl-bayes, but I don't really understand your point about why its a bad name. Thanks Anthony -- (incf *yankees-world-series-losses*)
On Mon, Nov 17, 2003 at 05:01:02PM -0500, Anthony Ventimiglia wrote:
For all of you LGPL, sorry I forgot to mention that, sorry I guess my request was a bit informal.
Are you familiar with LLGPL? http://opensource.franz.com/preamble.html Plain vanilla LGPL is always a bit confusing with Lisp, since it depends on the disinction between dynamic and static linking. Ob name: sounds like there's a miscommunication somewhere. Either I didn't make my point clearly, or I had originally misunderstood something about the project -- in which case my comments are void. Cheers, -- Nikodemus
Nikodemus Siivola writes:
On Mon, Nov 17, 2003 at 05:01:02PM -0500, Anthony Ventimiglia wrote:
For all of you LGPL, sorry I forgot to mention that, sorry I guess my request was a bit informal.
Are you familiar with LLGPL?
http://opensource.franz.com/preamble.html
Plain vanilla LGPL is always a bit confusing with Lisp, since it depends on the disinction between dynamic and static linking.
I'd never seen the LLGPL before, I think that's obviously more applicable, but I don't see it written out fully, is there a place I can cut and paste the full body of the LLGPL ? -- (incf *yankees-world-series-losses*)
participants (5)
-
Anthony Ventimiglia -
Erik Enge -
Mario Mommer -
Miles Egan -
Nikodemus Siivola