On Fri, Jan 22, 2010 at 7:59 PM, Arik Baratz <[email protected]<list%[email protected]> > wrote:
> 2010/1/22 Amit Aronovitch <[email protected]> > > > >> How one expects things to behave strongly depends on who that "one" is. >> People with different expectations or different knowledge might have hard >> time understanding (not to mention explaining) the other one's POV, so *any* >> implementation could be considered "a bad idea". >> >> The meaning (purpose) of the two lists can be understood as: >> (1) Python-IL: all things python in Israel, (maybe better named >> python-all-IL), vs PyWeb-IL: web related python stuff in Israel. >> (2) Python-IL: all non-web-related python stuff in Israel (maybe better >> named python-nonweb-IL) vs PyWeb-IL: as before. >> >> > Or it can be understood as "Python discussion in Israel" and "Specialized > discussion about python use for the web in Israel". > Sounds like (1) (at least like what I ment by 1). My phrasing was aimed to highlight the difference between 1 and 2. > I know there might be some crossover, but that's what cross-posting is for > (if done in moderation!) > The problem is not with over-cross-posting, but with under-cross-posting. Every python-IL subscriber will subscribe to both (see my poll). Given that and the fact that more than 60% of current contents is web-related, people will post to pyweb only (you will NOT be able to convince them otherwise, and will not bother to forward their posts yourself for long). After a short while people will not bother subscribing to the "dead" Python-IL at all, so people with general python questions etc, will stop posting here too (well, some of them might post to pyweb instead, so it would become python-IL de-facto, losing its specialization). With this in mind, maybe the best option is to unite the lists completely. This has two downsides: (1) web-oriented people with no general interest in python, might not find/not want to subscribe to a group with no "web" in its name. (2) Subscribers of pyweb will not be willing to give up their beloved google interface, whereas some python-IL subscribers are googlephobic and will object to a list requiring registration in Google Groups. > >> If you think the lists are described by option (2), no doubt your >> reasoning was right, but note that to think that you must know in advance >> that Pyweb-IL exists, and even then this does not seem to be the most >> straightforward interpretation. >> >> > I definitely think an announcement about another list is appropriate. > > >> If you think option (1) describes things best, then it makes sense that a >> person who subscribes to Python-IL expects to get all python contents, >> including web stuff. He *does not need* to subscribe to Pyweb as well (and >> might not even know it exists). If he does subscribe to both, he might >> expect some duplicate contents. This is the most simple and direct >> expectation based on the interpretation of the group's name. Whether this is >> implemented by forwarding the other list or somehow convincing the denizens >> of Pyweb to cross-post is a matter of implementation. >> Now, experienced net-denizens such as the posters in this thread typically >> know about the dynamics of competing mailing lists, so they settle into a >> practice of searching for all groups containing python and subscribing to >> all in order to avoid missing contents. But in this case (given that they >> also typically know about cross-posting) they should also expect duplicate >> contents. >> >> And now we come to the question of practicality. So far, my suspicion >> seems correct - NOBODY seems to want to limit themselves to non-web-related >> python stuff (though some wish to reserve the option for that in the >> future). Some other people (I guess) subscribe to all kinds of web-related >> groups, but don't care much about python in general. How do we provide the >> most usefulness to people interpreting the purpose of the groups correctly >> and least damage to those who do not? >> Since it is harder to convince people (some of which might not be aware >> that python-IL exists) to cross-post than do a mass forward, since I >> consider loss of information worse than *expected* and easily removable >> duplication of messages (I am still having a hard time understanding why >> some people subscribed to both groups did not expect to get duplicate >> messages - see above), I thought that the current arrangement is the most >> practical. >> >> Now, all this is just trying to explain and to understand. I am not trying >> to convince anyone this way or the other (in fact, my personal interest is >> to unsubscribe python-IL from pyweb, so I'll get a little less spam in the >> moderator queue), just trying to clear up misunderstandings. >> > > I've been on-line since 1993. Perhaps due to that I expect mailing lists to > behave in a certain way and expect a certain etiquette regarding exploders > and cross-posting. Yes I could write filters, but it definitely behaves in a > way I personally do not EXPECT a list to behave. > > Well, seems like I've been around a bit longer than you. However, back in the (late) 80's and most of the 90's I never subscribed to automated mailing lists (with Usenet around, mailing lists had seemed a waste of precious bandwidth and disk space to me), so you probably have much more experience with these :-) re: expectations: but if you subscribe to both lists, you do expect to get repeated messages, don't you? (and given the considerations above, you'd probably even think that more than 60% of current contents SHOULD be cross-posted to both lists) So if you care about such things you'd write those filters anyway (and with current configuration they will be a little simpler to write). The vote count reflects my position, even if you take the absentees and > enumerate them as 'keep subscribed'. > Don't worry, I will not count them this way or the other. Nevertheless, since the list-forwarding was not my idea, it would only be fair to give them that extra week to respond before I undo it. > > -- Arik > > Thanks for taking the time to explain... AA
_______________________________________________ Python-il mailing list [email protected] http://hamakor.org.il/cgi-bin/mailman/listinfo/python-il
