Ari Siletz

Divaneh, MM

Divaneh: Thank you for bringing up the advanced topic of  Vaafoor factor. I believe (CNTRL-ALT-V) on a Farsi computer keyboard puts the text in Vaafoor mode where, among other changes, "zh" takes the place of "zeh" and "jeh,"  and "vaav" replaces "b".  

MM: Don't know if there's been a study comparing post 79 with pre 79. But there is a noticeable difference between Farsi and Arabic letter frequency even though pretty much the same alphabet is used. For example in Arabic "Laam" is a close second to  "Aleph" in frequency and "Reh" isn't as prominent. So a reasonable guess would be that post 79 certain language circles have more  "Laam"s and fewer "Reh"'s, relativley speaking. Of course the fancy new Farsi technical words may have offset this.

Interesting blog,

Interesting blog, Ari.

Has there been a comparson of the usage of the "Farsi" alphabets before and after 1979 since the IRI official tone has been leaning towards Arabic words?

BTW, Divaneh is correct in regards to the incidence of "Zh" in Farsi, as our tongues tend to loosen while.....

مژخرف نگو

این شه مژخرفیه میگی؟ آخه ژانم، عژیژم، این همه ژ توی ژبون ماست. باژ میگی ژ نداریم؟ ژل الخالق. 

Aynak

The paper isn't right or wrong at this point.  It is just a theoretical presentation (referencing previous works, as any good paper should) without any empirical data to support the idea that the proposed new keyboard arrangement would work better than the current standard. My complaint is that this emprical data could have been easily obtained but wasn't. It doesnt'  take a CERN to run the experiment, just a software reassigning of computer keys and some undergrad test subjects. I'm actually curious to know if the author's algorithm produced a better keyboard, why wasn't the author  just as curious to know ?!

Ari, I missed the Behbahan URL

  in my original reading of your post and could not figure what you were complaining about given the graph :) (therefore my original post).

After your explanation, I went back to your original post and saw the URL and read the paper,  :)   (missed it the first time as it was not highlighted on my cell phone).

 Still however, (respectfully) this did not make your critique any more valid :)   In my speculation (I could not  access the references, even the one with URL), but these 3 papers, are the source of Behbahns paper.  If not a direct copy :).   Actually optimization of keyboard is the subject of a Persian paper as well in his reference.   So he has really just copied this stuff.   Now as far as methodology, you could say it is wrong, and say why you think it is wrong, but in reality "optimal" means based on certain common criteria which is spelled in the paper.  (like how close the top 3 or 5 most frequently characters have to be for easisest finger movement).   That is why a "," is not in the middle of a keyboad.  So adding 2 or 4 people to test this, in and itself is not going to alter the research result.   This is really not so much about scientific thesis/speculation, but a mathematical approach with embedded certainty.  (i.e doesn't matter if 2 people of 10 or even the entire popluation says 2+2=5, it never is :)).

 Gotti, J. S., & Brugh, A.W., & Julstrom, B. A. (2005). Arranging the Keyboard with a Permutation-Coded Genetic Algorithm. In Proc. Of the 2005 SCM Symposium on Applied 

computing, Volume 2, pp. 947-951.

Klausler, P. (2005). Available at www.visi.com/~pmk/evovled.html, Sep.

Light, W. L., & Anderson G. P. (1993). Typewriter keyboard via simulated annealing, AI Expert, September.

Malas, T. M., & Taifour, S. S., & Abandah, G. A. (2008). Toward Optimal Arabic Keyboard Layout Using Genetic Algorithm. In Proc. 9

th Int’l Middle Multiconference on 

Simulation and Modeling, Aug 26-28, Amman, Jordan

this paper is direct copy/combination of papers on keyboard optimization  

An observation

Aynak, Pendar,

the main drawback of Samimi Behbehan's paper may be culturally revealing. His claim is that the Farsi keyboard letter arrangement arrived at through his algorithm will perform better than the standard Farsi keyboard, saving time. It is an easy task to software reassign the keys on a computer so that his proposed arrangement can be qucikly tested. But his paper doesn't show that he ran this easy and obvious experiment which would have proven that his theoretical result is true in the real world.  Key reassigning and retagging will  take a few hours. Then it would be another couple of weeks to train two groups of people who don't know how to type (so there's no bias) The control group would train on a standard keyboard and the experimental group trains on the new keyboard arrangement. May the best group win! 

Without verification, this interesting paper is only good for getting credentials. The author may be displaying a traditional mindset where concreteness lowers the status of intellectual work, which should remain ethereal and abstract to be respected.  

ارتباط این نمودار و استفاده بیش از حد از "الف" با بی سوادی؟

 آری جان:

 حالا
داری مورچه تو تنبون  افراد می اندازی که تعداد "الف" ها رو قبل از ارسال 
نگارش بشمرند که خدای نکرده در زمره بی سوادان قرار نگیرند؟

ه مثل هرکی‌ بیاد نمی‌تونه از شاه بدتر باشه...

به این میگن بلاگ، که بهت یاد میده که توی دنیا، حتا اینکه انگشتت
کجا باید بره از روی فرمول و حساب و کتاب هستش........وگرنه که روزگارت
میشه مثل من که تا بیای به خودت بجنبی حجت الاسلام هویج لگاریتمش رو تا
دسته چپونده توی تمام ذرات ( هسته، اتم) آرمانهای غنی نشده و سانتریفیوژ
افکارت........

double post

%#$!

Spelling correction

The lower frequency "h" is heh as in hojat ol eslaam not heh as in havij as I have incorrectly written in the graph.