1. search is based on first navigating by user using dropdown menus or maps to a focussed area and then search by selecting or inputting information other than name.
2. .Input your name and search from data
3. .Only visually search in given (pdf) files
4. The possibility of not knowing area name exactly makes search difficult.
5. It is not easy to see all the lists which may include deletions or recent additions.
6. .Like many others, the Maharashtra CEO's site depends on visual inspection of part-wise marathi/indian language lists in pdf

In short those sites are 'site driven' and mainly navigational aids rather than user driven 'search' aid.
Primary objectives : Features such as

1. Break the barriers of language computing to access governance data which is mostly in legacy formats like iscii (standard) or totally non-standard like akruti or shreelipi.
2. Minimum infrastructure cost ( like OS and database software )
3. Minimum dollar outgo ( avoid imported software)
4. Full freedom: base on opensource and freely downloadable software
5. Opening up source code completely to be dedicated for for public use
6. Multilinguality across all the Indian languages
7. Adherence to standards for compatibility with emerging global OS standard such as Unicode as part of future proofing
8. CONVERSION PROCEDURE:

1. Single-step conversion: One can convert from existing DBFs to SQL format automatically in a single step.
2. Multi-layer error detection and conversion: DBF files are checked for data errors(syntactic) and DBF -to-SQL convertors are very tolerant of data errors.
3. Fault-tolerant process of loading files into PostgersSQL prevents corrupt records from entering database.
4. Modular importing: Data can be loaded from constituencies individually one-by-one by programs,which check data again.

9.Highlights:

1. Speed: search time in milli seconds
2. Search results are displayed in order of their relevance with the searched name.
3. Completely based on FLOSS. Full freedom and Low infrastructure costs
4. Component technologies (Apache2, PostgresSQL, PHP, Linux ) known to be mature well-tested and widely used.
5. Exclusive SouIndics Phonetic matching: similar sounding word indigenous technology
6. Minimal Dollar outgo on software, hardware, services etc.
7. Multilinguality across the nation: conversion to any language script with the same ease as in ISCII.

Scalable from thin clients to Mega-servers,making it ideal for deployment in centralized as well as remote rural places.
Environment of operation : The idea is to make multilingual software suite for search of voters by name and obtain relevant details from the existing data .
The same software suite may be used at
1. On Desktops
1. Registration desk ( distributed or centralised data entry.)
2. Help line on polling day to be use
Primarily (non-techy) GO ( say Collector's office without highly skilled technical resources) as also
NGO ( with a desire to run help line )

1. On the web
1. Individuals sitting at home or
2. In cybercafes
2. in language environment

1. With marathi/indian language enabled, Unicode compliant OS or
2. withouts marathi/indian language enabled and without Unicode compliant OS ( such as MSWin98 but ready to input and receive output in itrans english)
3. with readiness to download and install a free font like Shusha

Features : Input/Output
1. Input Fields:
Assembly-wise Search can be made using any one, any combination or all of the following input fields . All input fields would either be in marathi/indian language or english (itrans) to be converted by program to marathi/indian language .
Assembly name (or number) From Drop down menu

1. Name ( zero,one or more than one of any of the following in any sequence) : either full or part with a * to indicate wild character
2. First Name (Either english (itrans) or marathi/indian language (inscript) but not mixed)
3. Middle Name (Either english (itrans) or marathi/indian language (inscript) but not mixed) Last Name (Either english (itrans) or marathi/indian language (inscript) but not mixed)

Area (any one word) ( zero,one or more than one of any of the following in any sequence)

1. Area name
2. Locality (any one w
3. Valid 6 digit Pincode
EPID No.
Note: As far as EPID no is concerned, it is in English, however, it should be noted that this is not a very reliable entry in the data and should be used not as determining but as only additional information.
2. Options: User is able to exercise any of the following options before search.This is done by offering an opening screen which detects user's configurartion and depending on which display is acceptable or desirable, user can switch to page that will offer the output.

1. Assembly Name ( with code number ) from a dropdown list (select one) or select all to search all assemblies in the database created. (We can have any number subject to hardware limitation)
2. marathi/indian language input or English inputs

marathi/indian language output

1. Unicode : (of course font independent) HTML format
2. Non-Unicode: Shusha font , HTML format installed font
3. Non-Unicode : akruti dynamic font ( subject to CEO's choice)
4. English output
5. Whether to include names even if deleted
6. Max number of result records to be displayed ( admin mode removes restriction of number of results

3. Output : Results which match the input fields, and options.

1. Results include similar sounding words ( 'similar' as explained below.
2. Results are ordered according to relevance compared with input string.
3. No of occurances of each input string in the database record is also separately displayed.
4. If repeated names or repeated set of Ids (State code, Assembly constituency, part number and section ID and serial no in Part) , a table is output for diagnosing why the name got repeated.
5. Results include mechanism (hyperlink) for accessing the needed information for a voter i.e. Booth name and number where she/he may have to go to vote etc.
6. Following outputs or combinations will be made as per choice in opening page.
7. Output is in Unicode marathi/indian language if chosen
8. English and Multilingual output: Itrans english ( made more readable by relaxing standard ) output is given if that option is chosen.
9. Output in Shusha font is given if that option is chosen.
10. Output will be in dynamic font ( as of august 2004 in Akruti) if chosen.
11. Output will be colour-coded to indicate status information, i.e Normal(blue)/ Modified(blue)/ Deleted (Red with strike through).
12. Output will be restricted to first n records where n= 25 to 50 as max number as option. In admin mode,this restriction can be overridden.
13. Number of results and search time is reported if search results is non-zero. Output of booth information for individual will be sent to a text file for storage and subsequent printing
14. Out put of booth info can be in pdf format with embedded fonts.
15. Note: For batch of names search can be conducted in batch mode and results of booth info can be pooled in bulk for bulk painting. This feature is not implemented as yet.

4.Input Method:

1. Keyboards : Search strings has to be in marathi/indian language as the data is in marathi/indian language . However facility has been provided to input in english phonetic as in itrans standard with some relaxation. All in small letters and no capital . Fault tolerant to some extent. Help provided on website. marathi/indian language input may be inscript.
2. Switch between languages/scripts is not needed.
3. Only english input needs to be converted to marathi/indian language by clicking once.
4. Copy-paste from file: It is possible to use names pre-entered in a data file (in Unicode marathi/indian language or english itrans). Thus it is possible to copy and paste from a file to the text box for search.

5. Special features:

1. imilar sounding word assistance: Support for search for similar sounding words : similar sounding is a fuzzy concept and it means here : all words with same consonants, all words with consonants in same sequence, all words with similar sounding consonants in same sequence, all words with matras irrespective of 'rhaswa' or 'deergha' , etc and all words with some similar consonants as per 'varga' say 'pa' varga i.e. 'Pa' 'pha' etc. This will help smartly catch pathak and phatak and fatak etc.
2. Webbased software : - Online utility has been uploaded on website and tried for a single constituency .
3. Scale: The software is able to handle the assembly wide to district wide data and even state wide data. .
4. Support to legacy OSes : For the webbased search, then the devanagari output is available on the user computers which may have MSWindows 98 or MSWindows 95 in a possibly diluted fashion (dilution means only reading of devanagari, and input only in Roman script). The browsers in these software may be enabled to view the devanagari with the assumption that either they can download and install Shusha ( or similar other 8-bit ) fonts in their systems or they use dynamic font.( explained above in output.)
5. Cheaper alternative OS : If users need to get the full flavor of modern technology( unicode based fonts for viewing and interacting) alternative OS(Operating System) along with applications to use Devanagari and other indian languages at a cost of not more than Rs 100 + courier charges, through bootable CD ( gnubhaaratii ) is available. Assuming the user is ready to provide a free partition of say 5 GB in existing hard disk , the OS can even be installed on it.

6. Deliverable: Mode of software distribution Complete application along with data is available in following deliverable forms

1. Capsule harddisk:- in the form of a assembly-wise OR district-wise Live i.e bootable harddisk (gnubhaaratii) with OS Debian 'Sarge' GNU/Linux, Mozilla Browser for interaction and viewing, enabled with Devanagari Unicode and non-Unicode Shusha fonts and input and rendering, open source database ( PostgreSQL 7.4, apache2, and data for the assembly all preinstalled. The Harddisk will have wordprocessor and free opensource office software with indian languages along with free GPL fonts preinstalled.
2. Application CD- A separate CD for application software with scripts . The software will accept as Input a set of .dbf (iscii) files in prescribed file folders, on harddisk .