Re: Some software development by Samskritam.Net presented for your evaluation...

[विश्वासो वासुकिजः (Vishvas Vasuki)]

+ sanskrit-programmers - Please see sktProcess4Intro.docx for an guide to what is being presented - seems to mainly be scripts to convert sanskrit text from one orthographic scheme to another, detecting Chandas (again) and "cleaning" the text in various ways.

dear shrI kRpAlu.

My feedback is as follows:

* It is true that much of what you've developed so far is already more or less available as open source software (oft published at https://pypi.python.org/pypi ) - for example indic-transliteration 1.5.2 with source code published on github.com/sanskrit-coders/indic_transliteration . Using it is as simple as adding two lines to one's code. Similarly with Chandas and possibly sandhi.

* That said, I don't claim to be the sole (or even a major) contributor to such public software - it is something that has been developed by a small, variable but active community of interested people. Please consider joining groups.google.com/forum/#!forum/sanskrit-programmers so as to join them - at the very least you won't waste time reinventing the wheel or if you do, you'll get better and broader feedback.

* Your work is in the right direction and very heartening. But, unless you develop it in an open-source manner (which is simple by using github.com as a code repository) and publish various versions of the library at pypi, it risks joining the ranks of the hundreds of pieces of code developed as part of various Indian university labs - never to be heard of or reused except perhaps on some obscure website.

* I did not check your output files - I'll just take your word that the programs listed in sktProcess4Intro.docx  work as intended.

On Fri, Nov 3, 2017 at 6:08 AM, ogeti krupalu <okru...@samskritam.net> wrote:

Dear Atmiya Friend,

Namaste. Over the last couple of months or so, I have been programming in Python to demonstrate some of my ideas of what would help Samskrit community.

Currently, my test input is based on 2 text files. Attached is a ZIP file containing one DOCX file (SktProcess4Intro.docx). Please open this first. This explains how various files enumerated in the ZIP were 'generated' by our software. With some Samskrit knowledge, we hope you can understand the 'why' and 'what' of this work.

I request you to spare some of your valuable time and evaluate this work and see if you think it is relevant. I also request you to write-up some lines of your fair and free evaluation for helping us to decide our future course of action.

This is only a first-cut version. If you face any problem in the course of your evaluation, I will be more than happy to assist you in any manner you like. Please do not hesitate to call me.

From all I know, each of you is a busy person. Still I hope you may be able to give me some time for this 'divine' work. Please don't feel bad if you can not spare time for any reason.

Friends like Viswas Vasuki might already have all of this functionality and much more. I lacked the knowledge to use their 'Sandhi Engine', which came to  me some time ago. However, we want to move to that soon, either with our own programming or seeking their active 'cooperation'.

Thanks and regards,

Krupalu
7396249650

--

--
Vishvas /विश्वासः

[विश्वासो वासुकिजः (Vishvas Vasuki)]

-removing all others except sankrit-programmers and shrI kRpAlu.

I just wanted to close this thread in the mailing list with a quote from a recent email from shrI kRpAlu which clarifies that they DON'T intend to go open source:

=======

This ZIP is only for users. So, no Python code was included. The expectation is that the users & well-wishers in the address list compare input & output; and approve or comment on what we are doing. (One person already wanted code. I have not given it yet. I will give it after I work out his share in our organization.)

Our work is not FREE. Let Samskritam.Net retain IP rights. This is not amatuer(?) work. Definitely not like what Vishvas is predicting like about universities. We will move into cloud funding, if we get a chance.

I will get volunteers (or paid workers depending on my money) work on this and enhance its functionality year after year.

Let us wait for a week & see who is seeing value in our work. Then, we will decide what to do next.

As the next step, I want to attract someone who can help us to make this functionality available on our site (& enhance the need for a site on SNet.) Text file is too old-fashioned!

Thanks & regards,Krupalu

On Nov 4, 2017 7:05 AM, "vempati Sarma" <vsrk...@gmail.com> wrote:

Dear Krupalu,

My comments:

1. On word file:

a) Change he to Krupalu. Otherwise, it could mean me instead of you!

b) General and punctuation requirements such as spaces etc.

2. Unless I am mistaken, SktProcess4.py file is missing in the zip file. All txt files are received ok. If you are facing problem in sending Gmail you could convert in to text file and send it

3. I also agree with Shri. Vishvas Vasuki, that preserving files in git will be useful both others  as well as you. Convenience is it saves only changes, occupies less memory for modified versions. It also saves older versions at whoever and wherever they have downloaded your main version on particular date. So you will be able to get back to the old code even if it got deleted from you computer. And GitHub graphical version is easy to learn and use.

I will comment further after testing your Python program.

--

Best Regards,

vsrk sarma

[ogeti krupalu]

Dear Vishvas,

My blessings to you.

I am amazed at your speed of work. I am sure it is for the benefit of mankind. I really appreciate your spirit. Some time ago, I have looked at your 'heroes' & really appreciate you.

Having said that, I am keen to go beyond bits & pieces of voluntary good work. I am a contended soul. All my accumulations will be to build a frame-work for bringing Samskritam more forcefully to the world.

Here is my last request to this young group. Can any of you educate us on what great developments are kept by you (& others whose work impressed you as a Samskritist) in 'open source' environment? We would like to evaluate all such work & move forward from there on. Just to save the trouble of re-inventing the wheel. If there is no such ready-made document or presentation (on the lines of what I sent you initially), can I request you to let me know the name of one volunteer, who can be our contact point before we decide to do development ourselves for our needs.

Please understand our seriousness (& not just our 'greed') & help us. It is my pleasure to talk to any young Samskritist. Feel free to call me.

अन्यत् सर्वं भगवतः कृपा!

भवदीयः सदा,

कृपालुः ओगेटि-कुलजः

 +91-7396249650, 9866140406

[विश्वासो वासुकिजः (Vishvas Vasuki)]

+ shrI sai (who, like your planned sanskrit website, is trying to realize vedavaapi.org ) and shrI ramaNa who separately showed interest.

Dear shrI kRpAlu,

Humble thanks for your blessings.

With this mail I would just like to point out a few misconceptions:

* Open source work is often not "amateur" as your earlier mail implied - rather, it tends to approach (or surpass) the best professional standards. This is because people who build open source products are themselves seasoned professionals working on projects close to their heart.

* The programming language you yourself use (Python) is open source software - surely you realize this? Google runs on linux (also open source). And I have not touched (for professional work) proprietary closed sourced software for a long time. What I am trying to point out here is that if the open source mode of operation can produce Python, Java and Linux - you should think hard before presuming that it will not suffice for you.

*

​It is also a misconception to think that open source software is only produced by ​bits & pieces of voluntary good work.​ Not at all. Heard of companies called Red Hat, Suse or Ubuntu? Or non-profits like Apache? Or open source software released by for-profit companies like Google, Twitter an Facebook? Or university-centered products like Scala? Closer home shrI sAi is leading an effort similar to yours: vedavaapi.org , which works within the "open source" mode but finds funding from contributions, grants etc.. (and not from selling closed source software).

* I recognize and respect your best intentions (by no means did I attribute ​'greed' of any sort). I only suggested an alternative approach, without any presumptions. In that you hold some strange views about open source software - I only attribute it to your being the product of a different time and generation. I disagree with your present approach, not because I doubt your good intentions or marked leadership abilities, but because I think it is suboptimal and a product of poor knowledge in certain regards.

​
* Regarding your request that: "​ If there is no such ready-made document or presentation (on the lines of what I sent you initially), can I request you to let me know the name of one volunteer, who can be our contact point before we decide to do development ourselves for our needs."​ - Open source sanskrit software is not under the purview of any single person or organization. Further, the sanskrit programmers mailing list members are not my employees - it would be disrespectful of me to "volunteer" their time. I doubt that you will find a single such point of contact - you can just browse threads in this mailing list and our github repositories to see what is being done; and post questions here. ​

[विश्वासो वासुकिजः (Vishvas Vasuki)]

​+dhaval​

On Sun, Nov 5, 2017 at 8:35 AM, ogeti krupalu <okru...@samskritam.net> wrote:

Dear Vishvas,

Many thanks for taking time to explain. You guessed it right! I am surely an old timer (now 65 years), of nuts and bolts technology. I left IT about 15 years ago and much seems to have changed. While Ramana is my close friend, I know Sai Susarla too personally. (In fact, I spent some time with him when he presented his ideas in Bengaluru for SB's website. I fondly remember his hospitality when I stayed at his home that night.) But for the confidence reposed by Vamshi Krishna Ghanapathi of Mysuru in my abilities, I would not have come this far with my efforts in this direction. Let me try with Sai's help now.

​Great to hear it!​ I appreciate and adore sustained enthusiasm in advanced age - at a time when people sadly tend to waste their skills and time.

 

Also, some time ago, I thought that Shri Dhaval Patel was very seriously working on software of our interest. (I even reported an issue and he came back then with 'express speed' like you.)  I would request any of you to bring him too into the loop, if you happen to have his email-id.

​He is probably reading this thread alread as he is a member of sanskrit-programmers mailing list. But cc-ing him above.​ dhaval (an IAS officer by profession) is a great example of a self-taught programmer.

 

I look forward to all of you giving us at Hyderabad, a helping hand to be of service to Samskrit learners.

​We should help each other​ 🙏

 

Coming back to the OCR issue, has anybody here converted (or supported somebody to do) any sizeable printed book into Unicode?

People routinely OCR big public domain books on the sanskrit wikisource (which is again open source, but uses google OCR which is planned to be open-sourced but is not yet ). I myself motivated the OCR and digitization of https://sa.wikisource.org/wiki/%E0%A4%85%E0%A4%A8%E0%A5%81%E0%A4%95%E0%A5%8D%E0%A4%B0%E0%A4%AE%E0%A4%A3%E0%A4%BF%E0%A4%95%E0%A4%BE:ADictionaryOfSanskritGrammarByMahamahopadhyayaKashinathVasudevAbhyankar.djvu for example. After OCR, a lot of correction is required - for this samskrita bharati's wikisource unit kindly stepped in and did the needful by gathering volunteers. Using this data, we were later able to make that dictionary available for search using stardict apps in a variety of devices (including mobile phones).

​

 

Thanks,

Krupalu

[ogeti krupalu]

Dear Vishvas,

Many thanks for taking time to explain. You guessed it right! I am surely an old timer (now 65 years), of nuts and bolts technology. I left IT about 15 years ago and much seems to have changed. While Ramana is my close friend, I know Sai Susarla too personally. (In fact, I spent some time with him when he presented his ideas in Bengaluru for SB's website. I fondly remember his hospitality when I stayed at his home that night.) But for the confidence reposed by Vamshi Krishna Ghanapathi of Mysuru in my abilities, I would not have come this far with my efforts in this direction. Let me try with Sai's help now.

Also, some time ago, I thought that Shri Dhaval Patel was very seriously working on software of our interest. (I even reported an issue and he came back then with 'express speed' like you.)  I would request any of you to bring him too into the loop, if you happen to have his email-id.

I look forward to all of you giving us at Hyderabad, a helping hand to be of service to Samskrit learners.

Coming back to the OCR issue, has anybody here converted (or supported somebody to do) any sizeable printed book into Unicode?

Thanks,

Krupalu