I think we should institute a minimum length for all new stub pages. There are many pages that are only one sentence, but I don't know enough about the topic to add more information. (the one I came across was [[GDM]]). I think we should make a rule that all stubs should be at least three sentences long. That way, we won't have as many really short stubs around. We could build this into the software that there must be three periods (at least) in every entry. To go further, if we really wanted to, we could say that if an entry has no periods (a vandal probably made it) then instead of giving an error message, it would just show the page (once) as if it would look if it were saved but really not save it. Most vandals (not trolls) just write one simple phrase of babble on a non-existant page once, and it is soon deleted, but we could automate this. --LittleDan
__________________________________ Do you Yahoo!? The New Yahoo! Search - Faster. Easier. Bingo. http://search.yahoo.com
On 5/18/03 12:02 PM, "Daniel Ehrenberg" littledanehren@yahoo.com wrote:
I think we should institute a minimum length for all new stub pages. There are many pages that are only one sentence, but I don't know enough about the topic to add more information. (the one I came across was [[GDM]]). I think we should make a rule that all stubs should be at least three sentences long. That way, we won't have as many really short stubs around. We could build this into the software that there must be three periods (at least) in every entry. To go further, if we really wanted to, we could say that if an entry has no periods (a vandal probably made it) then instead of giving an error message, it would just show the page (once) as if it would look if it were saved but really not save it. Most vandals (not trolls) just write one simple phrase of babble on a non-existant page once, and it is soon deleted, but we could automate this. --LittleDan
I disagree. The stub-length preference allows each user to define what they consider a "minimum" stub. I, for one, believe that a one-sentence stub is better than no information at all. You disagree; the current system offers a non-destructive way of making both of us somewhat happy. A classic compromise.
--- The Cunctator cunctator@kband.com wrote:
On 5/18/03 12:02 PM, "Daniel Ehrenberg" littledanehren@yahoo.com wrote:
I think we should institute a minimum length for
all
new stub pages. There are many pages that are only
one
sentence, but I don't know enough about the topic
to
add more information. (the one I came across was [[GDM]]). I think we should make a rule that all
stubs
should be at least three sentences long. That way,
we
won't have as many really short stubs around. We
could
build this into the software that there must be
three
periods (at least) in every entry. To go further,
if
we really wanted to, we could say that if an entry
has
no periods (a vandal probably made it) then
instead of
giving an error message, it would just show the
page
(once) as if it would look if it were saved but
really
not save it. Most vandals (not trolls) just write
one
simple phrase of babble on a non-existant page
once,
and it is soon deleted, but we could automate
this.
--LittleDan
I disagree. The stub-length preference allows each user to define what they consider a "minimum" stub. I, for one, believe that a one-sentence stub is better than no information at all. You disagree; the current system offers a non-destructive way of making both of us somewhat happy. A classic compromise.
How about this: The convention is that a stub *should* have at least 3 sentences, but if that is impossible, then one or two will have to do, but there would be no software changes --LittleDan
__________________________________ Do you Yahoo!? The New Yahoo! Search - Faster. Easier. Bingo. http://search.yahoo.com
On 5/18/03 12:52 PM, "Daniel Ehrenberg" littledanehren@yahoo.com wrote:
How about this: The convention is that a stub *should* have at least 3 sentences, but if that is impossible, then one or two will have to do, but there would be no software changes --LittleDan
The convention already is that all entries should be as complete as possible. It's funny, though, because you're essentially resurrecting the comma criterion (only entries with a "," would be counted as an article when doing a count of the number of articles in Wikipedia), which I thought was a good criterion for the English language.
BTW, do you have the stub-link functionality on? It's quite useful.
--- The Cunctator cunctator@kband.com wrote:
On 5/18/03 12:52 PM, "Daniel Ehrenberg" littledanehren@yahoo.com wrote:
How about this: The convention is that a stub
*should*
have at least 3 sentences, but if that is
impossible,
then one or two will have to do, but there would
be no
software changes --LittleDan
The convention already is that all entries should be as complete as possible. It's funny, though, because you're essentially resurrecting the comma criterion (only entries with a "," would be counted as an article when doing a count of the number of articles in Wikipedia), which I thought was a good criterion for the English language.
BTW, do you have the stub-link functionality on? It's quite useful.
The stub filter doesn't work for some reason for me.
I don't like the comma rule, because it can be easily (and inadvertantly) worked around. For example, what if someone wrote "in Massachusetts, USA." for the Boston article. Then the rule and subsequently software would consider it a full article. --LittleDan
__________________________________ Do you Yahoo!? The New Yahoo! Search - Faster. Easier. Bingo. http://search.yahoo.com
On Sun, 2003-05-18 at 14:43, Daniel Ehrenberg wrote:
I don't like the comma rule, because it can be easily (and inadvertantly) worked around. For example, what if someone wrote "in Massachusetts, USA." for the Boston article. Then the rule and subsequently software would consider it a full article.
No arbitrary automatic rule is perfect. If the rule was three periods, someone could write "in Massachusetts..." etc. They're all dumb rules, essentially equivalent.
On 18 May 2003 20:36:06 -0400, The Cunctator cunctator@kband.com gave utterance to the following:
On Sun, 2003-05-18 at 14:43, Daniel Ehrenberg wrote:
I don't like the comma rule, because it can be easily (and inadvertantly) worked around. For example, what if someone wrote "in Massachusetts, USA." for the Boston article. Then the rule and subsequently software would consider it a full article.
No arbitrary automatic rule is perfect. If the rule was three periods, someone could write "in Massachusetts..." etc. They're all dumb rules, essentially equivalent.
Not to mention that a three sentence requirement only encourages people to add unresearched material which may well be wrong in order to fulfil the requirement. Or they may write short sentences. These sentences may not read well.
--- Richard Grevers lists@dramatic.co.nz wrote:
On 18 May 2003 20:36:06 -0400, The Cunctator cunctator@kband.com gave utterance to the following:
On Sun, 2003-05-18 at 14:43, Daniel Ehrenberg
wrote:
I don't like the comma rule, because it can be
easily
(and inadvertantly) worked around. For example,
what
if someone wrote "in Massachusetts, USA." for the Boston article. Then the rule and subsequently software would consider it a full article.
No arbitrary automatic rule is perfect. If the
rule was three periods,
someone could write "in Massachusetts..." etc.
They're all dumb rules,
essentially equivalent.
Not to mention that a three sentence requirement only encourages people to add unresearched material which may well be wrong in order to fulfil the requirement. Or they may write short sentences. These sentences may not read well. -- Richard Grevers I am a nobody - nobody is perfect - therefore I am perfect.
I guess you're right, but I think the requirement for counting as an article should be 3 sentences instead of a comma, because a comma can easily be left out in a full article or be included (without use of an elipsis) in a short one. --LittleDan
__________________________________ Do you Yahoo!? The New Yahoo! Search - Faster. Easier. Bingo. http://search.yahoo.com
----- Original Message ----- From: "Daniel Ehrenberg" littledanehren@yahoo.com
I guess you're right, but I think the requirement for counting as an article should be 3 sentences instead of a comma, because a comma can easily be left out in a full article or be included (without use of an elipsis) in a short one. --LittleDan
You may want to have a look at http://meta.wikipedia.org/wiki/Article_count_reform which is, I think, the most recent major discussion of what constitutes an 'article'. The link on that page to Article Count Reform Vote shows the options suggested and voted for.
Andrew (Ams80)
On Mon, 19 May 2003, Andrew Smith wrote:
You may want to have a look at http://meta.wikipedia.org/wiki/Article_count_reform which is, I think, the most recent major discussion of what constitutes an 'article'. The link on that page to Article Count Reform Vote shows the options suggested and voted for.
What was decided: * all articles in talk space containing at least one wiki link shall be considered "articles" for purposes of the front-page article count.
I haven't had a chance to make the change yet, but I'll probably go ahead and do it when I figure out why the counter is currently broken. :)
I'm not entirely sure whether the count is what Dan's referring to here though, as there's also the link color display etc mixed in.
-- brion vibber (brion @ pobox.com)
On Mon, 2003-05-19 at 15:22, Brion Vibber wrote:
On Mon, 19 May 2003, Andrew Smith wrote:
You may want to have a look at http://meta.wikipedia.org/wiki/Article_count_reform which is, I think, the most recent major discussion of what constitutes an 'article'. The link on that page to Article Count Reform Vote shows the options suggested and voted for.
What was decided:
- all articles in talk space containing at least one wiki link shall be
considered "articles" for purposes of the front-page article count.
I haven't had a chance to make the change yet, but I'll probably go ahead and do it when I figure out why the counter is currently broken. :)
By the way, while that rule makes sense in general for the international wikis, it's a worse one than the current one for English language, and will cause yet another unfortunate hiccup in our statistical history. I hope that even if we change the "official" formulat, we'll continue to keep track of the comma numbers so as to give a statistically useful history.