2
$\begingroup$

I have become an avid observer of the site, and I have learned a ton so far (still mustering up a few questions, but that's another story...). I've been trying to keep an eye on things from the perspective of someone that's had a moderate amount of experience on the Exchange, with a fair handle for what constitutes on- and off-topic questions.

Now, I'm not calling anyone out, I'm just trying to edify my own skills here, and maybe get a bit of clarification, but could someone point out why What is the growth rate of the world wide web? is on topic here?

Knowing that this site wants to be the go-to spot for CS questions, how does this fit into that framework? It seems like it could be something that's on topic for the Webmaster's site, or a similar venue. I fail to see how it advances algorithms, data structures, optimization, Theory of Computation, or any other sub-areas. Perhaps I am overlooking something obvious.

Again, a fine question, and one that might otherwise be constructive, but I'm not completely confident it belongs here.

$\endgroup$
  • 1
    $\begingroup$ I thought so, too, but the other mods were not. The question is one of the kind that is phrased badly, but contains a good question somewhere inside. Here, namely, how do you measure data growth in a huge network? It is not a steller question, but it might admit good answers. $\endgroup$ – Raphael May 20 '12 at 9:19
  • $\begingroup$ @Raphael A fair point, but that's injecting quite a bit of someone else's interpretation into the question. I mean, I understand that you want to hold onto all of the questions you can, but there should be a line somewhere. $\endgroup$ – jonsca May 20 '12 at 9:24
  • 1
    $\begingroup$ I agree. I would have preferred to close and wait for the OP to update the question. Let's see what others have to say; maybe we end up closing the question, anyway. $\endgroup$ – Raphael May 20 '12 at 9:34
  • $\begingroup$ It seems like a reasonable question to me. Data and the Internet/WWW are sibjects that I think everyone'd agree are on-topic here. It is applied and empirical and, as such, quite different from the questions the site has usually seen, but that's not a problem. There are papers in the ACM digital library that (probably, based on the titles) address the question. Basically, I wouldn't be comfortable closing the question as a moderator, nor would I vote to close it as a regular user. Close enough, for me. $\endgroup$ – Patrick87 May 20 '12 at 15:16
  • $\begingroup$ Open-ended and off-topic are distinct reasons for closing a question. Which is it? Regarding what makes this a computer science question... it asks about how an entity of great interest to many computer scientists changes in a way that is important to those computer scientists. Might computer scientists provide unique perspective? I think so. $\endgroup$ – Patrick87 May 20 '12 at 15:37
  • $\begingroup$ I disagree. How is the Texas power grid of interest to computer scientists? If it's of interest to any computer scientists, my guess is that it's a vanishingly small group of them. I do think the question "how many nodes are in the Internet" or some similar would be on-topic. People publish CS papers about that. $\endgroup$ – Patrick87 May 20 '12 at 15:50
  • $\begingroup$ You continue to confuse reasons for closing a question. You might think the question is unanswerable. Under certain interpretations, you are probably correct. However, being unanswerable has nothing to do with the question's being on-topic. $\endgroup$ – Patrick87 May 20 '12 at 15:57
  • $\begingroup$ @jonsca, I doubt the author is asking for exact number, in empirical studies a good enough estimate would suffice. This kind of question are studied in the context of computer and more recently social network. People are very interested in properties of the graphs underlying these complex structures both from theoretical and practical side and the number of vertices is one of the simplest parameters about these graphs. Nonetheless I agree the question can and should be improved. $\endgroup$ – Kaveh May 20 '12 at 18:51
  • $\begingroup$ Adding some motivation to the question would clarify what the OP is really looking for. For example, are they designing a crawler and want to know how much storage they need to store the links? Are they looking for storing the content? etc. At its current form it looks like out of ceriousity question where the OP has not put enough thought and effort to write a good question, it is under-specified. ps: when starting a meta thread about a question it is customary to add a comment below the question linking to the meta discussion. $\endgroup$ – Kaveh May 20 '12 at 18:55
  • 1
    $\begingroup$ @jonsca, the right thing to do in these cases is to ask the OP to clarify the question. In this case the OP replied to my comment, I would suggest that you ask the OP to further clarify the question by commenting. $\endgroup$ – Kaveh May 20 '12 at 20:09
  • $\begingroup$ How is the Texas power grid of interest to computer scientists? — Sometimes we computer scientists get tired of using social networks and citation networks as our only examples. It's nice to have a solid physical example to kick our students and colleagues out of their mental walled gardens. $\endgroup$ – JeffE May 22 '12 at 7:49
5
$\begingroup$

Your reproach is that there exists a reasonable interpretation of the question that makes it off-topic. This does not make the question off-topic. The real tests is whether there exists a reasonable interpretation of the question that makes it on-topic.

Even taking the original wording of the question (“Is there any way to estimate how much data is added to the Internet each second? Are there any studies about this?”), there is indeed a reasonable interpretation of the question that makes it on-topic. “Is there any way to estimate” is a type of question that comes up often in experimental sciences. It's about the Internet, which makes it computer science. This site is not limited to theoretical computer science.

I expect an answer to cite some publications about the subject and summarize the methods used (and not just state a number). The answer is likely to tangentially involve algorithms (more specifically distributed algorithms) and not involve data structures or theory of computation at all, but these are not the only fields of computer science.

$\endgroup$
  • $\begingroup$ I accept what you, Patrick87 and Kaveh have put forth, but not without much reservation. I do appreciate that my concerns were listened to, and I hope that there are no hard feelings in your camp. $\endgroup$ – jonsca May 22 '12 at 9:50
6
$\begingroup$

The WWW has been the topic of research for computer scientists (and physicists, statisticians, social scientists, etc). At its core it is a (very) large graph, and graph theory is central to (theoretical) computer science. Behind this particular question is a significant challenge, one which certainly falls within the realm of what computer scientists do. The challenge is: how does one measure such a thing? Better still: how does one estimate such a thing? This is a challenge for two reasons: one because the WWW is extremely large and two because it is highly dynamic. These are issues that computer scientists (with skills in statistics) are capable of addressing.

In short, I think the question is on-topic.

$\endgroup$
  • 1
    $\begingroup$ Now, if you had presented that as a question, I'd likely have no problem with it. However, as it stands, everyone is reading their own interpretation into the original form and extrapolating. It's not about what a question could be, but what it is. $\endgroup$ – jonsca May 20 '12 at 19:42
  • $\begingroup$ For lack of a totally made up example, if I asked "Why did I spill my milk?" over on Philosophy, it's not going to go over as a profound question on existentialism. $\endgroup$ – jonsca May 20 '12 at 19:46
  • $\begingroup$ I admit that the version I read was the most recent one, not the original one. The most recent version is much more reasonable. $\endgroup$ – Dave Clarke May 20 '12 at 19:46
  • 2
    $\begingroup$ But why did you spill your milk? $\endgroup$ – Dave Clarke May 20 '12 at 19:46
  • $\begingroup$ I didn't, the milk wanted to spill. $\endgroup$ – jonsca May 20 '12 at 19:49
  • $\begingroup$ Deep. Very deep. $\endgroup$ – Dave Clarke May 20 '12 at 19:50

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .