There's a few assertions that have been made that I'd like to challenge.

There are "good" and "bad" ITGs, and really, we just need to sort them better.
-

One thing that has been brought up repeatedly is that there are "good" ITGs and there are "bad" ITGs.  All we need to do is downvote and close the bad ones, right?

Well, what makes a bad ITG?  Clearly "too vague" ITGs are bad, asking "what's this game that has elves in it" is not going to be constructive.  I'd also argue that some specific details are just useless to us - for example, the specific shade of blue that the sky was in the second level.  Additionally, specific details are at the mercy of the asker's memory, and may be incorrect.

The problem is that **we can't tell the difference between helpful and unhelpful detail.**  We can't objectively gauge when a question contains enough helpful detail to be answerable.  

For instance, I could craft an ITG that was a list of all the colors in the main character's sprite, and it would make a nice bulleted list to create an extremely detailed, specific, and utterly unhelpful question.

However, maybe somewhere, someone has a list of all the colors in every main character's sprite in every game ever, and for them, this is terribly *helpful* information.  They can answer this question in seconds with their database.  

Some might be tempted to vote to close or downvote this question, even though in 10 minutes time, the sprite color wizard will arrive to solve it.  

This is an extreme example, but frequently I get lucky with an ITG where the asker [remembered a specific detail][1] that I also remembered.  

I run into this problem often with ITG questions, on both sides.  Sometimes, I'm the answer wizard, and sometimes I'm the guy saying "this is too vague."  I've answered ITGs (and been *sure* I was right) that got closed shortly afterwards for being too vague, and I've been chewed out in comments for suggesting that perhaps the person could provide us with more detail.  

At what point do we stop waiting for the answer wizard?  Should we just leave questions open indefinitely, even though it has become clear that the detail provided is insufficient?  Do we close them after a year, even though it's still as likely that on the 366th day the answer will arrive as it was on the 365th?  

How can we sort the details provided by the asker into categories of "helpful" and "unhelpful" and make a determination about whether or not we have enough in the "helpful" category to continue?

We can say "well, everyone makes their own call and whatever happens, happens" - but I've seen already that there are enough community members who believe almost all ITGs to be bad that the result is going to be friction, arguments, and fighting, as this is a terribly subjective process.  That's against the SE model and bad for the community.  

ITGs are helpful questions and answers to a wide audience and the site. 
-

A few times now I've heard the argument that ITGs are beneficial to more than just the asker, and that we get a decent amount of traffic from these questions.  Based on the data, I don't believe that ITGs are in a class alongside other questions.

**Views**

One aspect of "usefulness" is the number of views a question gets.  More views means that more people found the question after it was asked.  ITG questions are terrible in this regard, here's a selection of tags and the **average views per question** on that tag:

- ITG - **264** views per question on average
- minecraft - 2,381
- starcraft-2 - 1,419
- skyrim - 4,953
- pc - 1,703
- xbox-360 - 890
- dwarf-fortress - 497
- dungeons-of-dredmor - 806
- terarria - 1,365

(This is just a somewhat random cross-section of the top tags - I ran a few more, but the trend takeaway is that ITG is far below the average in this statistic, even compared to somewhat obscure games that have small but considerable following on the site)

To date, ITG questions have brought 152,191 views to the site.  That's about **one weekend's worth of traffic** for this tag over the life of the site.  

I don't think we should be deciding whether a tag stays or goes based on views alone, but I want to point out that the argument that many people find this class of question useful is not really supported by the data.

**Anonymous Feedback**

I also reviewed the "anonymous feedback" section of the site to attempt to understand if random visitors were frequently pleased with the ITG answers.  **I paged through the top 500 of the "most" and "least" helpful posts based on anonymous feedback, and did not see any ITG questions on either list.**  I found 2 in the top 500 overrated, and 1 in the top 500 underrated.

**New Users**

For 575 questions on ITG, **question askers had a to-date reputation of less than 150 55% of the time.  Almost half of those (or 25% of the total askers) had a reputation lower than 25.**

This isn't perfect, and I'd rather look at how many went on to ask/answer other questions besides their first, but you can see that, on average, ITG questions are not a reliable source of high-quality contributors.  


  [1]: http://gaming.stackexchange.com/questions/42098/please-help-me-to-identify-this-windows-space-game-over-15-years-old