last day (14 days later) » 

12:12 PM
0
A: Build and strengthen the Stack Exchange community with "crossover questions" between sites

Adam DavisCrossposting is bad. A question which could be answered on two different sites is not a good question. Perhaps it needs to be more reasonably scoped. Perhaps the person asking needs to understand their problem a little better so they know who best to ask for a fix. Perhaps they need to understa...

 
This is simply not true. It used to be, back in the trilogy days, but no more. For example, simple bash questions are 100% on topic on Super User, Stack Overflow, Ask Ubuntu, Unix & Linux, Ask Different, and probably Raspberry Pi as well. If I were to ask "how do I assign a value to a variable in bash", there is simply no way to tweak it in order to make it more applicable to any of those sites. It fits all of them. Their scopes overlap. What we have now are the same basic questions repeated on all sites by different users. That is precisely the waste of effort that the no cross-posting rule tries to avoid.
 
@terdon I disagree. For obviously simple, and very easily "googlable" questions, sure, there's a lot of overlap, but the overlap ends after the trivial questions for the reasons stated in this answer. Still, with Stack Exchange moving to a single login and integrating the communities even further, who knows - maybe they will turn this whole thing into Yahoo Answers or Quora. I hope not, but it's their site.
 
@AdamDavis no, the scopes of many of those sites overlap enormously. Configuring WiFi on Ubuntu, for example, is 100% on topic on Unix & Linux, Ask Ubuntu and Super User (I'm a mod on the former and a >20k user on the other two).
 
@terdon Please quantify "sites overlap enormously." Does that mean we can take 80% of the questions asked on one site, ask them as-is on the other site, and get the same answers? Or are you arguing about a few percent of questions which might need to be modified slightly to really fit the site? If the former, then the sites should simply be merged (and honestly, unix/linux/ubuntu is a weird situation anyway - we should be discussing bikes and opendata, or stackoverflow and cooking) I don't doubt your experience and perspective, I just don't see it as a problem that needs to be fixed.
 
@AdamDavis it refers to a large subset of questions on each site. Basically all bash questions from any of the 6 sites I mentioned before would be on topic on all 6. What percentage of the site they represent depends on the site in question. At least 80% of AU questions would be on topic on U&L as well. All *nix questions on SU would be on topic on U&L. The difference between AU and U&L is i) some things are Ubuntu only (belong on AU) and ii) Ubuntu users tend to be more GUI oriented so the answers tend to be more often GUI based.
My point is that for a certain subgroup of questions and sites, the overlap is really enormous, yes. Not for all sites, obviously, and not for all questions. The scopes are overlapping venn diagrams, not concentric circles.
 
12:12 PM
@terdon I'm not interested in AU and U&L. As far as I'm concerned one should be closed as duplicate of the other. They are, at best, outliers in this discussion. I remain unconvinced that the rest of the sites should allow cross posting just because you want your favorite sites to allow it.
 
@AdamDavis hardly outliers. They're actually the 3rd (AU) and 8th (U&L) sites by traffic on the network. SU is the 2nd and SO is the first. The only site I mentioned that's not in the top 10 is Rpi.se (Apple is 8th). I'm talking about the shared scope of some of the most popular sites on the network. I agree that not all sites have overlapping scopes, but we now have several dozen sites, many of whose scopes do overlap. The result is the same question asked by different users on different sites: precisely what the no CP rule was meant to avoid.
 
@terdon I don't mean outliers in terms of usage or community. I mean outliers in terms of this discussion - their venn diagrams don't just overlap - the U&L circle completely encloses the Ubunto circle. Ubuntu could very clearly be a tag on U&L. So it's really pointless for you to keep using them as a prime example as to why this feature should be implemented. Maybe this is an issue for those sites. It isn't a significant issue for the rest of the network, and further cross posting doesn't solve a big problem so it doesn't need to be implemented.
 
Moved here so I don't need to spam you and you can feel free to ignore. I quite agree that the U&L/AU case is special. The scopes are actually relatively clear to me but I'm obviously atypical being a high rep user of both. They are certainly vague to the vast majority of users.
That's not the case for U&L, AU, SO, SU or Apple though. Their scopes all overlap and they also each have their core topics that fit better there than anywhere else. My point is that we now have many sites whose scopes overlap enough that the same questions are asked on each of them.
For example:
5
Q: assigning the output of a command to a variable in a shell script

user3114665I am new with unix and I am writing a shell script. When I run this line on the command prompt, it prints the total count of the number of processes which matches: ps -ef | awk '/siebsvc –s siebsrvr/ && !/awk/ { a++ } END { print a }' example, the output of the above line is 2 in the command...

12
Q: How can I assign the output of a command to a shell variable?

Nathan G.I want to assign the result of an expression to a variable and concatenate it with a string, then echo it. Here's what I've got: #!/bin/bash cd ~/Desktop; thefile= ls -t -U | grep -m 1 "Screen Shot"; echo "Most recent screenshot is: "$thefile; But, that outputs: Screen Shot 2011-07-03 at 1.55...

0
Q: Capture output of command in variable

IanYou know when you do something like apt-cache search something | less? Well (in a Bash script) how do you get the input and set it to a variable (like less is doing, but less isn't a Bash script)?

0
Q: Variable assigning in terminal

eeecoderis there a way to assign a value to a variable, that value which we get in terminal by writing any command? Example: command: sensors we get cpu temperature. How can I assign this value to a temp_cpu variable. Thanks..

The same question on 4 sites. This is not a good thing. It would be far better to have all those questions somehow merged.
 
1:11 PM
It seems we are covering the same ground here. See my first response above, "For obviously simple, and very easily "googlable" questions, sure, there's a lot of overlap, but the overlap ends after the trivial questions for the reasons stated in this answer. "
Further, "This is not a good thing" != "This is a bad thing, and if not corrected will ultimately hurt Stack Exchange or its users"
The proposal is for a feature that, for some users and some sites, might be useful. It doesn't fix a problem, it doesn't enhance the sites. If anything, it'll make answers more confusing.
Let's say a user asks a simple question on Ubuntu, and the same question on U&L. The Ubuntu distro includes a simple graphical utility to perform the function that isn't otherwise common. Let's say there's only one question between the two sites. Which answer should the user accept?
Besides all this, the scope of this feature request has not been analyzed. How many questions across the network would this actually affect?
Would we be able to close a question on SO as a duplicate of something on Ubuntu, which, later, is found to be a duplicate of something on SU? The network doesn't even follow duplicate chains like this, so are we going to force users to click several times before they get an answer, or just delete more questions so someone asking what they believe is a programming question never finds it because it's showing up in google as some strange site called "ubuntu" and they don't know what that is?
We can't show the same question to google as belonging to all the relevant sites, google doesn't like exact content duplication, so ultimately it would reduce the ability for people to find answers on the network.
 
1:26 PM
@AdamDavis those are very good points. I don't claim to have a solution, I just want to i) make sure that a question has as high a chance as possible to be seen by the expert who can answer it and ii) that at the moment we have many duplicate Q&As on the various sites, so we have precisely the duplicated effort we're trying to avoid. I would like to see a discussion about it.
Personally, I like Jnat's approach but there may well be a better way. I also agree that it may well lead to more problems than it solves. I'd like to look into it is all.
 

last day (14 days later) »