Discussion on Problem Statement

[ganeshran]

Below is the problem statement that was authored by Vivek Mithra. I am creating a placeholder thread to start that discussion in GitHub

We need to clearly examine different aspects of Problem Statement.

In my opinion, the intention of this project should only be a technology demostration which Indian Railways must adopt. Let's keep it extremely simple, our goal should only be to prove that IT IS POSSIBLE to print multi-lingual tickets.

We're targetting Unreserved ticketing system, which are currently printed only in English and Hindi in almost all railway stations across India. Most of these passengers know only one language, we would be solving a problem. Creating extension to IRCTC to print tickets in our languages is also necessary, but these customers are already familiar in English. There are some stations which are yet to adopt computer based printing systems, hence they still use three languages in tickets.

At the ticket counter, the ERP that Railway staff uses runs on English. based on the starting point and the departure point of the journey appropriate language(s) along with English must be printed on tickets. For example if a passenger is travelling from Bengaluru to Chennai, Kannada, Tamil and English should be printed. TTE (Train Ticket Examiner) must also be able to verify the ticket information, hence we need to encode the ticket information and print it on tickets (preferably QR). IRCTC should also support multiple languages, but that's a different battle which we can take up at a later stage.

We've two options,

1. We can do an elaborate design based on the problem statement and approach Railways through media and ask them to implement.
2. We can develop a working software.

I'm looking this as an extension of pur PLE campaign, marketing and building a narrative on our work here is the key. Citing compatability issues, Railways might reject our work and build it themselves, hence I guess it's OK even if we build a half-baked software. Hope this remains a private discussion, and please pour in your thoughts.

[mmanivannan]

I am excited about this. I would like to take this as an opportunity to show to Indian Railways, other Indian government departments or any corporations in India who may think that supporting multiple languages in India is either unnecessary or difficult. There is a real need for this service and we would want to encourage others to provide this service.

Ideally, I would like us to develop this in layers,

1. a database layer that indexes all the trainstations to their respective real names in various languages as well as any other names like the train names, etc. We can use Wiki volunteers to generate this database and we wil just import these
2. a multilingual 'dictionary' that maps a technical glossary that can be used to generate the printable text.
3. a software library that can read an ERS pdf file generated by IRCTC, especially its QR Code and extract the English text that we need
4. a software library that reads the extracted text from IRCTC QR Code and translates it to a requested language using a template
5. front end code for web browser plugins, iOS/Android app, Windows desktop application, Web portal for on demand conversion
6. we should be able to generate a PDF file with translated ticket in both English and target language (that is not English)
7. as much as possible we should keep the personally identifiable information local to the user machine and make sure that we don't keep them on our hosts, software, or let any potential third party libraries that we may use so that the PII cannot be exploited by anyone. User privacy is paramount.
8. This should be available to anyone such as third party vendors to use so that they can print it themselves

Regards,

Mani M. Manivannan

[tuxnani]

I can help with Telugu glossary, please let me know.

[iamvickyav]

Hi Everyone, Can u please add me in this project. I wish to contribute for this.

[ganeshran]

Invited you @tuxnani @iamvickyav . Please accept the invite to join the organization

[dhandapani-d]

Moving the Discussion from sourceforge, This the the comment from Mr.Sriram Krishanan on problem statement.

Hi All, I look at this in 2 angles.

1. Technical demostration to non-tech audience that multi-lingual printing is possibe(which the software world knows is very much feasible)
2. A tangible outcome which the members of the group and outside world can see and use

The layers put forth by Mr. Manivannan is right on bull's eye. I have downloaded the list of train numbers with names, station names from the irctc website. We should trigger the translation from the community while we work on the software in parallel.

If the layers put forth by him are accepted, we should start discussing the exact details on top of this abstract layer. I can handle android app and chrome extension while Mr. Dhandapani can work on iOS app. So we need to exactly spec out the end goal, which I think Mr. Manivannan seems to be the most suited based on the credentials discussion thread.

Thank You, Sriram K

[dhandapani-d]

Moved from SourceForge

Hi All,

I was about to type the same and agree with Sriram on the 2 angles of problem, both has different potentials,

1) Technical Demonstration (Multilingual Printing) - Agree with Vivek, Will be helpful for many people, who generally use Unreserved service. I have seen several people who can't understand unreserved ticket in current form which is in English in Hindi only. On this, we wont be able to get a way to integrate it into Indian Railways ticketing system easily. ( We should start looking at options for the same). But we can build this as a prototype where we can feed station codes [From & To], train numbers and Ticket No. to print a localised ticket using the database we build as layed out in Item 1 & 2 by Mr.Manivannan.

Once we have this software ready, We can run campaigns in Railway stations to promote our cause, and print free tickets (by feeding data from peoples original ticket) in local language (thought this might not be accepted, people should carry their original tickets). We can show people and governent, If we can why not IRCTC?.

2) An App (Mobile / Desktop / Portal) can get to people who use reserved tickets.

For App, It will be great If we can also build more options like searching fro trains between stations, availability check, train running status, PNR enquiry etc., So that people can use it full fledged for complete flow (atleast except for booking). Explored on this and found erail.in provides free API for all such functionalities. (Ref: http://api.erail.in/terms) But not sure about their reliability and they also don't give any promise on reliability and availability of the API's in their API term. We also need to attribute them in our app / website. We can explore if we can find any other API's available. But this can be considered as a future plan / long term vision.

Can we take these as 2 different projects?

I completely agree with the layers put forth by Mr.Manivannan. We need volunteers to work on item 1 & 2. Builting Multilingual data for Train / Stations / Glossary can be started immediately.

Thanks, Dhandapani

[iamvickyav]

Please add my colleagues Github Id vijayanbu001 & Mayil1993

[aaykay83]

Hi,

Instead of a db layer o would suggest a key based approach for apps

[mmanivannan]

I explained the purpose of this project as follows in the PLE group.

https://www.facebook.com/manivannan.m.mani The purposes of our project are the following:

1. To monolingual Indians provide a ticket that is readable in their languages - With the QR code and English in it, it may be acceptable by the TTE
2. Technology demonstration to prove that this can be done
3. Encourage private entrepreneurs to provide multi-lingual service in India
4. By doing a public relation action we increase the visibility of PLE, get more members to our group and spread the message of language rights

This is more or less a summary of our discussion. Let me know if we need to add/delete anything from this.

-Mani