Monday, July 2, 2012

Zynga API

Should we consider the Zynga API for our social games on VSC?  What's the cost?


The Zynga API will allow third-party game developers to take advantage of Zynga’s technology and servers, which the company said are strong enough to allow the company to release 1,000 new features to its games in a given week. The Zynga API will allow gaming companies to build their own games on top of Zynga’s technology, enhancing online gaming opportunities for smaller startups.
“Now third parties will be able to focus on what they do best. Enjoy and create beautiful games,” said Kostadis Roussos, chief engineer at Zynga.


Zynga has introduced several new components to zCloud, starting with the Active Social Network (ASN), which Roussos characterized as the "true barometer of social," counting how many people are actively playing with each other. He posited that this science helps Zynga build features for its users that are both social and fun.
Other features include "Matchmaking," connecting Zynga's 290 million monthly players, and "Optimized Game Mechanics," which includes a brand new architecture, codenamed "Darwin," that cuts software requirements in half. Roussos exclaimed that this asset, in particular, is like "magic" because it makes the platform more robust and efficient at the same time.

Sunday, July 1, 2012

Apps for education

There is an Apple site with descriptions of apps for education.
Private review sites:
Iear - This site is dedicated to ed app reviews, "A Community Effort to Grade Ed Apps"
Teach Hub - One review of 20.
Reg Swanson's blog about educational apps and how to use them. 827 followers
Interaction Education - emailed 7/1
Apps for Children with Special Needs
Fun Educational Apps

Applicious - General app site, something on education


Help Implementing Grammar Checker / Writing Evaluator


Help Implementing Grammar Checker / Writing Evaluator

I run a small online education company that is eager to implement some level of automated writing evaluation to help with instruction and evaluation on a few of our educational services (
VocabSpellingCity.com, Time4Writing.com, Time4Learning.com).  

Specifically, we would like to implement a technology to assess and provide specific feedback on the quality of student writing such as:
- Grammar mechanics: proper construction of sentences (subject, verb, final punctuation)
- Basic mechanics such as capitalization, subject-verb agreement (tense, plural, etc), consistency on plural and singular
- Advanced mechanics such as prepositional phrases, sentences of appropriate length, nuances of word choice, avoiding awkward, ineffective, or unclear writing
- Writing quality: vocabulary usage, organization, development  

We are looking for an individual or team to provide us with guidance and implementation,  

BTW, I've checked with the Kaggle team as to whether it is OK to post here with this request for contacts and got approval. If you are interested, I'm mayor at spellingcity dot com.  

AWE - automated writing evaluators

I'm on the hunt for some technology and some consulting help to implement a grammar checker for the student writing materials on VocabularySpellingCity.com and for broad use with Time4Writing.com (web-based writing tutorials).

Right now, I'm following up on an HP-sponsored competition on automated writing evaluators over at Kaggle.com.  The question ansd answer there quickly led me to the ETS Section on Automated Scoring of Writing Quality.

My RFP is for technology to assess and provide specific feedback on the quality of student writing such as:

  1. grammar mechanics: proper construction of sentences (subject, verb, final punctuation)    
  2. Basic mechanics such as capitalization,  subject-verb agreement (tense, plural, etc), consistency on plural and singular
  3. Advanced mechanics such as prepositional phrases, sentences of appropriate length, nuances of word choice, avoiding awkward, ineffective, or unclear writing
  4. Writing quality: vocabulary usage, organization, development

ETC says (and I quote from http://www.ets.org/research/topics/as_nlp/writing_quality they have "an active research agenda to develop linguistic features suitable for modeling aspects of meaning structure in essay-length responses, such as:
  • metrics of text coherence
  • the use of supporting facts from external sources
  • the writer's stance toward the material presented
  • the identification of particular topics addressed in the response

Featured Publications

Below are some recent or significant publications that our researchers have authored on the subject of automated scoring of writing quality.

2012

2010

2009

2008

2006

2005

  • Online Assessment in Writing
    N. Horkay, R. E. Bennett, N. Allen, & B. Kaplan
    Online Assessment in Mathematics and Writing: Reports from the NAEP Technology-Based Assessment Project
    NCES Report No. 2005-457
    U.S. Department of Education, National Center for Education Statistics
    The 2002 Writing Online (WOL) study is the second of three field investigations in the Technology-Based Assessment project, which explores the use of new technology in administering the National Assessment of Educational Progress (NAEP). Learn more or download the full report.

2003

Writing Evualators - Old Notes


There is a set of automated checkers which are a standard tool on most word processors, even on phones.

They are useful tools that could/should be made available in courses for people to learn how to use them.  They also provide a huge time savings for teachers to give  good (but not necessarily perfect) feedback to students on their writing.

This is very relevant to T4W and VSC. There are three overlapping types:
- simple spelling checkers. Checks if the word is in the Dictionary.
- grammar checkers. What this checks is an open question in my mind. Word certainly checks if there is a noun and a verb, if they agree, and if a word is repeated. What else?
- automated writing evaluators. Used to grade many standardized tests. Doesn’t statistically do worse than human testers.

Papers on the topic:


Grammar Checkers ( There are probably more!):
-          http://www.link.cs.cmu.edu/link/,
-          http://www.languagetool.org/

Educationally Oriented:
-          WPP from ERB https://www.wpponline.com/Home/Welcome
-          Vantage MyAccess http://www.myaccess.com/myaccess/do/log
-          Criterion from ETA http://www.ets.org/criterion


Consumer Oriented:

Saturday, June 30, 2012

Mobile Games Economics

I chatted about the economics of publishing mobile free games on the Ipad/Iphone, here's my notes.  They are just my repetition of one guy's loose talk.

Developers create their products with no advance. Often small teams working incredibly hard. Perhaps 3 months around the clock for two guys.  Maybe more. Two or four times more. Most developers go from game to game so they develop tools and frameworks that they reuse.  They get half the revenue.

Publishers of free games aim to get volume by becoming a top free game in the app store. To become a top App, they need to have large number of downlaods. Tehse are achieved by spending marketing money on....paying for downloads.  It seem  that there are some companies, called CPI companies, that have drones of people that download games on demand. Actually for money. A quick search of companies and systems for incentivized downloads produces a range of discussions of how Apple is cracking down on it and whether it is good or not for the industry. Here's one from last year on CPI Economics.

A publisher, I was told, might spend $100K or $200K on incentivized downloads at $1 or $2 per download. This could be as few as 50K downloads or as high as 200K. This gets you into the top charts. Being in the charts means that you might get an actual real user to download you.

Apparently, the charts are recalculated daily and one day in teh top of the carts can produce 15K real downloads, still for free.  This publisher makes his money on virtual goods.  So they hope that out of each ten downloads, one or two users will turn out to be a real player. A real player might use the game 30 times. During that time, he'll buy between $5-$20 goods.  Let me see if this works.

Spend $100K for 100K downloads.
Lets say have two days in the charts so 25K downloads.
Get 2.5K real users.
.25K spend $10 each resulting in $2,500 in revenue.
Woops, these numbers don't work. Not at all.They are off by two zeros, two orders of magnitude!

Legal White hat (he said) CPI companies: Flurry.com, tapjoy.com

Top game companies: EA,Zynga (New Toy, who did Words with Friends), GRE (Japanese), Kabam, Backflip, 6Waves, Motion math, duck duck moose,.

Advertising that they could embed it CPM. Perhaps $.01/day / user (assume 15 minutes).
Farmville at peak was growwing $1M per day on virtual goods.

Another issue is that companies with installed bases of apps can drive in app downloads of new apps. So the big guys make money since they have free marketing. And the big gusy mostly got big by being first out on Facebook when you could grow with spam!

Google Play and Amazon have Android apps stores. Good for 10-30% of Iphone marketi.

Shows: GDC, Media Bistro,com

Flurry has good stats.  josh@ was my friend who provided me with these thoughts....
Better Sources:
http://www.onlinemarketingrant.com/ 12 months old
http://tomasztunguz.com/2011/04/21/the-case-for-mobile-cpi-advertising/ 15 months old

Hottest Product Category at ISTE 2012!

I'm launching a new theme for this blog: mobile education. My focus is on the creating and marketing of it.

 I was just at the ISTE 2012 conference in San Diego. My observation is that the hot new product on the floor was....(drum roll please).....Ipad covers and protection devices.

There must have been a dozen booths selling covers and protectors and what not. One booth had little rubber balls that go on each corner of the IPAD and protects them when they fall.

 Here's what amuses me. Actually, it make me cry. The schools were apparently acquiring these covers in bulk. Some were $5 per Ipad, some were $25.

 But, when I talk to them about educational content for their Ipads, they all answered the same way: This year we are only using free Apps. THIS YEAR WE ARE ONLY USING FREE APPS! Great, instead of having a team of people slaving over educational content, I should be importing rubber balls with slits in them to put onto the corners of Ipads at $20 per shot.

You think I'm kidding. Here are a few write-ups from the ISTE Exhibition Hall Guide:


http://www.tridentcase.com/
Trident Case was founded with the mission to create the toughest cases available for the ever-expanding mobile device market. Trident Case provides protective cases for a broad range of mobile devices, and for a variety of different brands, such as Apple®, Motorola®, Samsung® and many more!


Friday, February 3, 2012

Science Standards

I'm working on a very ambitious K-2nd science curriculum and so we are waiting with baited breath for the latest release of the new standards from Achieve, scheduled for this year some time. In the meantime, we try to predict what they will be by studying the tea leaves and history. I just found a great write-up of the science standards history on the McRel website, I'll quote a bit here:


Science
In science, three efforts have contributed significantly to the development of standards. The National Research Council (NRC) published the National Science Education Standards in December 1996. Material related directly to content standards fills over one-third of the work's 262 pages, while additional chapters address standards for science teaching and professional development, as well as assessment, program, and system standards. The science content standards are written for three grade levels: K-4, 5-8, and 9-12. At each grade level, seven general science topics are addressed. Standards related to these topics become increasingly comprehensive at each grade level.

The second effort within the field of science comes from the American Association for the Advancement of Science (AAAS). Working from the foundation they helped build in Science for All Americans (1992), AAAS's Project 2061 provides over 60 "literacy goals" in science as well as mathematics, technology, and the social sciences. These goals are well articulated across levels K-2, 3-5, 6-8, and 9-12. This effort, published as Benchmarks for Science Literacy (1993), includes a useful discussion and presentation of the research base available to those who worked on the project.

In addition to these efforts, the National Science Teachers Association (NSTA) has published the Scope, Sequence and Coordination of National Science Education Content Standards (Aldridge, 1995) as an addendum to The Content Core: A Guide for Curriculum Designers (Pearsall, 1993). This supplement is designed to make the Core more consistent with the recently published NRC standards. NSTA has also released A High School Framework for National Science Education Standards (Aldridge, 1995), developed under a grant from the National Science Foundation. Like the addendum to the Core, this framework builds directly from the November 1994 draft of the NRC science standards. Essential generalizations in physics, chemistry, biology, Earth and space sciences, and other areas organize the framework. Each generalization is described in some detail with a list of the relevant concepts, empirical laws, and theories or models that students will need in order to acquire a solid grounding in the topic. These subsections are presented in grade sequence (9, 10-12) and include a recommended learning sequence. Other useful sources of information come from NAEP, including their Science Objectives: 1990 Assessment, Science Assessment and Exercise Specifications for the 1994 NAEP and Science Framework for the 1996 National Assessment of Educational Progress (since republished as the Science Framework for the 1996-2000 National Assessment of Educational Progress).

Sunday, January 29, 2012

Cyberlearning Science Key Take Aways

I was at a conference two weeks ago  on science and cyberlearning. Both the science of learning and how best to learn science were the topics.  I was warned prior to attending that the conference was probable not for me.  It's "very cutting edge" and focused on "what really matters."

As a company that's very interested in meaningful products, I was a little insulted by those comments.  However, I now realize what it means. These discussions try to look beyond the reality of today's budgets, testing cycles, teacher and school capabilities and so on. Their target is the 5-15 years out and assumes away any real questions of commercial viability, teacher capability, training needs, or the other issues about taking anything to scale.

Cyberlearning Science Key Take Aways 

1.        There’s a lot out there in the 5-15 year range which could/should go mainstream in education. In terms of understanding our own positioning, we’re very much pushing todays world forward, not reinventing.  This means, the future is now. We don’t have years and years to establish ourselves. We need to win now and be ready to shift quickly in the future.

2.       Ipad Ipad Ipad. Key visual was two kids in strollers with ipads. After a minute, one kid is captioned as “this one can’t really talk yet.”   Then the other is captioned. “this one can’t talk, can’t walk, and is wearing diapers.”  Then their apps are shown: they’re both playing learning games!

3.       Games. Gamefy! Games so students explore how things work trying to achieve/survive.  Games also at the LMS level so that they’re motivated to progress through lessons.  Social games for involvement.  Games with rewards for motivation. Learning as a puzzle with trial and error made acceptable (unlike in social classroom situations). Games that make challenges stimulating.

4.       People learn through experiences. Very few learn well from textbooks and lectures.  There wasn’t much discussion about the segment that does learn well from textbooks and lectures. I learned a lot from lectures though the years. I doubt I'm the only one but we weren't the target of this group.

5.       Collaboration and social aspects. Constructivist learning.  Groups and people matter. Students discovering knowledge. This was not an explicit instruction crowd at all. 

6.       Diversity of learning. Different speeds etc are the rule. There is no average student.  On average, men wear size 9 shoes, how is going with size 9 shoes for everyone going to work out?  I’d like to reconcile this with the learning styles publicity where the initial conception of it was discredited as a concept.

7.       Kinesthetic learning and learning experiences are a holy grail. Many exotic technologies and clever approaches to experiences.  Create earthquakes with subwoofers and 4 computers recording the event from their spot. Do top down projections on the floor to create an amazing immersive environment.  To me, one striking simple truth that was cited but not pursued is that when kids explain something, they learn it.  BTW, applying this simple truth doesn’t require technology or investment, just classroom management and making kids responsible for their and others grades!

8.       Researchers, please collaborate! NSF is worried about islands of innovation and baffling array of unconnected lessons and approaches going to market instead of a unified progressive technology science curriculum. 

Lastly, I remain more committed than ever to the importance of Science4Us' capability to transform primary science education.  But that the vision and view must continue to expand to include a nice link to hands-on activities and to new technologies and techniques for community and involvement. In short, it'll take a lot of money. It's time for me to realize that this needs major foundation support and that my hope of self-funding this all the way is not the optimal plan.  Sigh.   

Thanks to Time4Learning science education (first grade science, second grade science) and VSC's science vocabulary (kindergarten science words, first grade science words, and second grade science vocabulary words) for helping me starting to look at this area. I'm thinking of signing up to take a science methods class as the local university. Of course, my professor would ironically also be one of my employees.  Does that sound like a good idea?