Character count, packet search, txt to rtf

[BuzzerZen]

12-pt Times New Roman, 1-inch margins: 100 characters per line
6 lines = 600 characters
8 lines = 800 characters
10 lines = 1000 characters

10-pt Times New Roman, 1-inch margins: 115 characters per line
6 lines = 700 characters
8 lines = 925 characters
10 lines = 1150 characters

Thank you.

[BuzzerZen]

Here's another thing I did today: I made a Google Custom Search that covers the packet archives of repute, namely Stanford, quizbowlpackets, and collegiate.quizbowlpackets. Enjoy having a marginally easier way to google for questions.

[theMoMA]

Here's another thing I did today: I made a Google Custom Search that covers the packet archives of repute, namely Stanford, quizbowlpackets, and collegiate.quizbowlpackets. Enjoy having a marginally easier way to google for questions.

Hey, that's cool. Could you add Carlo's ACFDB or Arnav's dealie (however you spell it).

[BuzzerZen]

Here's another thing I did today: I made a Google Custom Search that covers the packet archives of repute, namely Stanford, quizbowlpackets, and collegiate.quizbowlpackets. Enjoy having a marginally easier way to google for questions.

Hey, that's cool. Could you add Carlo's ACFDB or Arnav's dealie (however you spell it).

Google doesn't seem to have an index for either site right now, sadly.

[BuzzerZen]

Latest in the "Evan likes doing silly things with computers" files: I have successfully leveraged existing tools to be able to transform plain-text ACF-formatted quiz bowl questions into RTFs. Not entirely existing tools, I guess. I slightly modified this thing called pandoc. Anyway, the point is, if anyone else happens to be interested in doing this, which doing it will probably require moderate-to-extensive software-compilation yak-shaving, let me know.

Edit: the point being that I can write my questions in vim now and nobody will be the wiser.

[Mechanical Beasts]

Latest in the "Evan likes doing silly things with computers" files: I have successfully leveraged existing tools to be able to transform plain-text ACF-formatted quiz bowl questions into RTFs. Not entirely existing tools, I guess. I slightly modified this thing called pandoc. Anyway, the point is, if anyone else happens to be interested in doing this, which doing it will probably require moderate-to-extensive software-compilation yak-shaving, let me know.

Edit: the point being that I can write my questions in vim now and nobody will be the wiser.

I pretty much want this.

[grapesmoker]

Latest in the "Evan likes doing silly things with computers" files: I have successfully leveraged existing tools to be able to transform plain-text ACF-formatted quiz bowl questions into RTFs. Not entirely existing tools, I guess. I slightly modified this thing called pandoc. Anyway, the point is, if anyone else happens to be interested in doing this, which doing it will probably require moderate-to-extensive software-compilation yak-shaving, let me know.

Edit: the point being that I can write my questions in vim now and nobody will be the wiser.

If there's anything I like, it's shaving yaks. Send me info on how you do this, please.

[Maxwell Sniffingwell]

Do you have a Sega Game Gear, Jerry?

[cvdwightw]

12-pt Times New Roman, 1-inch margins: 100 characters per line
6 lines = 600 characters
8 lines = 800 characters
10 lines = 1000 characters

10-pt Times New Roman, 1-inch margins: 115 characters per line
6 lines = 700 characters
8 lines = 925 characters
10 lines = 1150 characters

Thank you.

Independent verification tells me this is wrong. Six lines of "UUUUUUUUUUUUUUUUUUUUUUU" in size 10 TNR with 1-inch margins yields 384 characters; eight lines is 512, ten lines is 640. Six lines of "UUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUU" in size 12 TNR with same margins yields 324 characters; eight lines is 432, ten lines is 540.

However, in actual quizbowl questions, it looks like you're just overshooting the size 12 TNR and just undershooting the size 10 TNR. Remember since TNR is not a fixed-width font like Courier New is, the width of each letter and the space is different. So these are only averages and guidelines.

[BuzzerZen]

Independent verification tells me this is wrong. Six lines of "UUUUUUUUUUUUUUUUUUUUUUU" in size 10 TNR with 1-inch margins yields 384 characters; eight lines is 512, ten lines is 640. Six lines of "UUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUU" in size 12 TNR with same margins yields 324 characters; eight lines is 432, ten lines is 540.

However, in actual quizbowl questions, it looks like you're just overshooting the size 12 TNR and just undershooting the size 10 TNR. Remember since TNR is not a fixed-width font like Courier New is, the width of each letter and the space is different. So these are only averages and guidelines.

My metrics were derived from using a sentence from an actual quiz bowl question with spaces and lowercase characters and everything. Six lines of an uppercase letter with no spaces is clearly going to contain fewer characters than six lines of words. In any event, this is why "lines of type in a proportionally spaced font" is a silly metric, and why I wanted a rough estimate of character-count equivalents. If someone wants to go through ACF packets and do some rigorous statistical analysis on tossup length, that'd clearly be better.

[cdcarter]

My metrics were derived from using a sentence from an actual quiz bowl question with spaces and lowercase characters and everything. Six lines of an uppercase letter with no spaces is clearly going to contain fewer characters than six lines of words. In any event, this is why "lines of type in a proportionally spaced font" is a silly metric, and why I wanted a rough estimate of character-count equivalents. If someone wants to go through ACF packets and do some rigorous statistical analysis on tossup length, that'd clearly be better.

EDIT: Wait nevermind. Yes someone who is not me should do this.

[BuzzerZen]

The 21 tossups in Brown's ACF Fall packet have a mean character count of 661 with a standard deviation of 18.98. The shortest tossup is 630 characters, the longest is 699.

[cdcarter]

The 21 tossups in Brown's ACF Fall packet have a mean character count of 661 with a standard deviation of 18.98. The shortest tossup is 630 characters, the longest is 699.

Now do a regression of char count, and number of lines for this and other packets. That would actually be helpful.

[BuzzerZen]

The 21 tossups in Brown's ACF Fall packet have a mean character count of 661 with a standard deviation of 18.98. The shortest tossup is 630 characters, the longest is 699.

Now do a regression of char count, and number of lines for this and other packets. That would actually be helpful.

You're the one who's taken statistics.

[BuzzerZen]

A guide to doing more or less what I have done. Feel free to ask questions.

[cvdwightw]

The 21 tossups in Brown's ACF Fall packet have a mean character count of 661 with a standard deviation of 18.98. The shortest tossup is 630 characters, the longest is 699.

Now do a regression of char count, and number of lines for this and other packets. That would actually be helpful.

You're the one who's taken statistics.

If you want to help me with this project, go to http://spreadsheets.google.com/ccc?key= ... wjfRo8ZCRA and enter data from some packets.

How to do this:
Packet should have normal 1" margins. If not, fix the margins.

Tournament, Round, Tossup, and Font should all be self-explanatory.

For "Full Lines", count the number of full lines taken up by the tossup. For "Spillover (in)", round to the nearest half inch (in Word 2007, go to Page Layout -> Align -> View Gridlines, then go Page Layout -> Align -> Setup Gridlines and set Horizontal Gridlines to 0.5", then just round by eye if the end of the question falls between two gridlines). Columns for "Total Inches" and "Total Lines" should update automatically, if not, copy the formula in the row above it. For "Characters", use Word Count and use the Characters (with spaces) value.

If you want, you can enter the rightmost two columns into Excel and create a Characters vs. Lines scatter plot. Add a trendline, and make sure to click the box to set y-intercept to 0 (0 lines = 0 characters).

For the 84 tossups I have in there so far, I get Characters = 113.51*Lines for TNR 10 with an R-squared value of 0.969.

[cdcarter]

If you have an iPod Touch or an iPhone, check out http://web.me.com/cdcarter/iApps/acf_score/index.html. It's a simple ACF game score keeping application that will not only track the score, but give the stat lines for individuals as well. If I thought there were enough of these little devices in the quizbowl world for it to be meaningful, it would be pretty easy for me to allow this to send scores to a central server and also track in depth bonus conversion and that kind of stuff. But until the touch technology spreads out, it's just a cute little toy that might come in handy, at least during practices.

[Mike Bentley]

Latest in the "Evan likes doing silly things with computers" files: I have successfully leveraged existing tools to be able to transform plain-text ACF-formatted quiz bowl questions into RTFs. Not entirely existing tools, I guess. I slightly modified this thing called pandoc. Anyway, the point is, if anyone else happens to be interested in doing this, which doing it will probably require moderate-to-extensive software-compilation yak-shaving, let me know.

Edit: the point being that I can write my questions in vim now and nobody will be the wiser.

Do you have a solution for converting the other way? Going from rtf to plain text and then back again would make a lot of the problems in the Question Mixer go away.

[SnookerUSF]

I apologize for the lack of relevance to the thread, but I think Christian's IPhone App is pretty neat idea, if not totally practical, yet. I am emoting approval and you should to, though perhaps privately.

[ihavenoidea]

If you have an iPod Touch or an iPhone, check out http://web.me.com/cdcarter/iApps/acf_score/index.html. It's a simple ACF game score keeping application that will not only track the score, but give the stat lines for individuals as well. If I thought there were enough of these little devices in the quizbowl world for it to be meaningful, it would be pretty easy for me to allow this to send scores to a central server and also track in depth bonus conversion and that kind of stuff. But until the touch technology spreads out, it's just a cute little toy that might come in handy, at least during practices.

Thanks Chris, I'll definitely use this in practice.

[Matthew D]

is there one like this for a palm?