Saturday, 18 August 2012

Re: [pakgrid] Research Paper while working in Corporate Sector of Pakistan

 

Dear Ali Raza,

You have raised a very significant question and I have been expressing my opinion over it for about several years now. Unfortunately corporate sector in Pakistan would discourage you on that (I am saying this based on my own experience, exceptions of some companies who promote research may exist).

Not many people know that it's mainly the industry (companies like Google, Microsoft, IBM, Hewlett-Packard, Yahoo!) that publishes a lot in the world of Computer Science so much so that there are now debates within Computer Science academic circles to not allow industry and corporate sector to publish. Follow recent SIGIR 2012 panel on use of proprietary data in Information Retrieval. Unfortunately you have to follow tweets with #sigir2012 label as there is no Web page as of now. It's a very recent event held in Portland, USA and is one of top conferences in the field of Information Retrieval.

There is also another interesting paper in July issue of CACM by Peter Norvig (yes, the AI guy whose book we read in our undergrad AI course). Norvig happens to be Director of Research at Google. You can find details of the paper here:

Also Matt Welsh (I guess you may know about him, he was a highly reputed Computer Science Professor at Harvard who left his tenure-track position a year back to join Google research) blogged about this paper:

My point is simply that such models of companies and corporate sectors publishing research papers tends to exist and it's not just established companies that are doing this. Some startups are also following this model now. One may argue as to how companies may benefit from such publishing and it's a simple waste of money and time. Well there definitely is benefit which is why you see so many companies attending and publishing to top venues in Computer Science. This year I had the privilege to attend the largest and most popular conference related to World Wide Web in Lyon, France and a lot of companies had papers there. In fact most of the exciting research came from companies. Companies want feedback about their ongoing products and they want to learn from academia which is why they attend conferences and network with the top brains in the field (the PhD's and the Post-Docs and the Professors). In fact most of the companies (Yahoo!, Google, Microsoft Research and even Qatar Computing Research Institute which is an affiliate of Al-Jazeera) have an industry-academia collaboration which seeks PhD students as interns and during the internship the student publishes his/her work while working on a problem given by the company. Turns out this model is not even known in Pakistan, let alone applied because unfortunately the psyche in Pakistan is that publishing papers is a time waste and just about theory. We follow an out-sourcing model rather than an innovation model, despite all rants of our IT sector becoming entrepreneurial and innovative products coming I really do not see the real innovation that you see in some of the best companies of the world. I admit it's too much to ask for now but at least one must get the priorities right, companies and the seths within the IT sector look at research with scorn failing to realize that is only research through which innovation can be truly achieved.

I have time and again emphasized that companies and software houses in Pakistan (at least the big ones) should opt for this model. It will give them huge leverage in terms of productivity. They may not see its fruits right away but in the long run it will enable them to come up with remarkable products: research is what powers most of the great products of the world. Had it not been for the PageRank paper we would not have known Google as of today. 

Bottomline: It's almost impossible to make companies understand why publishing would be good for them but I would still encourage you to try that and keep on trying. I am hopeful eventually they will understand one day or the other. Try convincing the company management first as Azhar said. If your company does not allow you to publish with their data forget it. Pick up a part-time data mining or business analytic project out of your own interest, it should be your data, your idea and your technique so that there is no hurdle in publishing (open source track can be followed but not mandatory). Do it purely for research, at times you might get frustrated but in the long run a publication can highly boost your chances of getting into a well-ranked university and that too on scholarship. I followed this path after my BS degree in Computer Science and my publication eventually landed me a fully-funded scholarship at the world's 24th ranking university in technology i.e. KAIST. If you don't want to go for technical MS then the option suggested by Azhar of publishing articles by concealing nature of products and consumers (only concepts and turnouts) can be a good one.

Feel free to contact for any help.

Best Regards,
Arjumand.

Overseas Faculty and Visiting Research Fellow,
Web Science and Technology Research Group,
Faculty of Computer ScienceInstitute of Business Administration IBA,
PhD Candidate, SNA and IR, NUIG and Univ. of Milan, Bicocca.

Email: ayounus@iba.edu.pk
http://arjumand-atif.blogspot.com


From: Azhar Hussain <azharhussain1150@yahoo.com>
To: "pakgrid@yahoogroups.com" <pakgrid@yahoogroups.com>
Sent: Friday, August 3, 2012 4:59 PM
Subject: Re: [pakgrid] Research Paper while working in Corporate Sector of Pakistan

 
Dear Ali Raza,

You seem motivated enough but for publishing you must consider following points:
1. Companies data is proprietary, so you must get permission from your company at first hand.
2. Otherwise concealing the nature of products and consumers just based on the concepts and turnouts you can publish articles.
3. Publishing may be your hobby, it doesn't demand funds from either from your organisation or you there are journals which publish for free.
4. It is always good to win you an opportunity in MBA

Regards
Azhar Hussain


From: Ali Raza Anjum <aliraza_anjum@hotmail.com>
To: pakgrid@yahoogroups.com
Sent: Monday, July 30, 2012 6:16 AM
Subject: [pakgrid] Research Paper while working in Corporate Sector of Pakistan

 
Dear All,
                Can please someone guide me that how to write a Research or Technical paper while working in corporate sector Sector of Pakistan?
Whenever I complete my assigned task I think that there can be some sort of paper related to my task(Because my work involves reading of different Whitepapers and Research Papers related to data mining/ Business Analytics) but then there comes some sort of questions in my mind and these questions always become  hurdles in front of my Idea or motivation.
1.       Will my company make an objection that you are exposing company strategy ?
2.       Will it be fair to use company resources and working environment to strengthen my idea or motivation ?
3.       From where will I get funding for my publication ? Definitely my company will not give me funding for research papers and I also don't want to spend it from my salary J
4.       Will it be more beneficial to spend my time on writing research papers instead of passing certifications ?
5.       How technical paper can be transformed into a white paper ?
6.       And will these technical papers be helpful if I am planning to go for a MBA instead of Technical MS ?
       
And hopefully this topic will proof little bit different from currently ongoing discussion on forum (Paaani wali Caar) J .
Hopefully each and every reply will definitely help me a lot to get motivated again.
 
 
Regards
Ali Raza Anjum
http://www.linkedin.com/profile/view?id=52259014&trk=tab_pro




__._,_.___
Recent Activity:
.

__,_._,___

No comments:

Post a Comment