February 4th, 2009 — General
The last year or so has been completely dominated by Twitter, There have been lots of articles like this one which speak about Twitters business model and its chances of survival.
I love Twitter and have declared that i would pay to use it. I know a lot of good Samaritans who feel the same way. In this land of a million good tools, how does one make sure there is enough contribution to the ones we care. Wikipedia for instance runs a very successful donation scheme which works beautifully for them but what about commercial applications like Twitter, Facebook etc. Dont those employees deserve some direct reward for the experience they give us?
A lot of you would have shouted “Advertising” atleast a few times by now. I think advertising was an offshoot of not knowing what a person wants, thats changing with intentions galore on the internet. Advertising has moved from a mass consumption/ small conversion medium to one of data mining. This was clearly the last big play. Also advertising is facing the following issues.
- In an opinionated world its hard to force thoughts, This is again debatable. I have been told the number of gullible people on this planet is increasing at an alarming rate. I disagree, the connected world is getting smarter and forcing consumption by clever advertising but bad quality / unjustified need will continue to wane. Seriously, look at the big cars if you need proof
- Advertising is intrusive, it challenges my intelligence. I invent new ways to avoid it
- Algorithmic assumptions are dangerous. A friend once told me he got an ad for “Meet married women” on facebook because he was part of this huge conversation where every girl was married.
Clearly there needs to be a next step in consumption. I would like to believe its through Free will.
- What if you could declare what you wanted to consume?
- You declare that along with your conditions of disclosing your intention. In this case a contribution from the broker who has access to this information to your favorite venture (twitter, facebook, orkut whatever)
- You close the deal, your venture ackonowledges the payment and you get special treatment (privileged members?)
There are numerous challenges for this to work ofcourse.
- Cant trust companies: Seriously, i love these products but i wont trust such information on their site.
- How do i know i wont be taken for a ride?
Also by this time its fairly obvious that the really clued in bunch will use this. Thats ok, 100k such people will make life easier for things we care. This brings us to the next question, where do i upload this and how does one make sure this works?
- It has to be an independent vendor and preferably not for profit
- It has have a thorough listing of vendors and fine grained control for users. Choose who all should get my buying habit, how many should get it, which area etc.
- No history, what i do once is immediately erased. Next transaction is on a clean slate
- An encrypted signature whose keys are given to trackers on a per case basis so this whole thing can scale. For instance if want a new tv i would put up “New tv 24 inch, 20>30k should not be samsung” in predefined format and encrypt it and put it up at a location. The key to unlock is at one organization who send it to 10 vendors who can then unlock it. This way we can publish this info on our own sites.
- Complete abstraction api, a communication service which hides real user info from all vendors but the one who closes the deal. This one needs to scale and has to be fairly thorough.
- Legal guidelines which protect all the interests of the users
If all this is in place then in an ideal world.
- I would decide to buy something, I put an encrypted text on some url. Upload the key and the url to a service
- I declare the benefactors of the purchase if and when it happens
- I fine grain number of vendors who can contact me and companies i am willing to talk to
- I define length of time this is active for
Now the service picks up the packet, decrypts it creates an abstraction to contact me. Email/ Voip etc and sends it to the vendors. The vendors close the deal, i make sure they let the service know and the service in turn pays the benefactors.
There might be numerous such services since it will be an open protocol and each one can charge a small service fee to the vendors to stay alive. The bigger their network gets there might be a possibility to build a valuable business.
What do you guys think? Is this worth pursuing? Will this make your world a little better?
January 14th, 2009 — General
Today Youtube launched a new feature which mutes all copyrighted content (music especially). This is basically to combat all the litigation against youtube streaming copyrighted music. While this is the right thing to do from the point of the law, this could very well end the great music experience on Youtube. The real losers are all those people who built their music applications on top of the Youtube API, people like Audiolizer.
I also sense a great opportunity for Independent Musicians to make an impact now. When Amazon music store first launched a lot of songs which were sold were covers of Famous songs, this did two things.
- Allowed people searching for a song to buy that song
- Allowed artists a way to be discovered for every cover purchased the listener would discover the cover band and their originals.
So what can one do with Youtube, Especially since uploading covers without prior permission is not acceptable by law. Read here and here for more details.
Well, we all agree that playing covers is not the USP for most artists. Original music shows real talent and probably is more satisfying from the artists perspective too. So be smart, Pick 10 songs which are closest to your genre and are most searched for on Youtube (look at the views to get a fair idea). Make a cover of those 10 songs, get the relevant permissions and write them off as marketing cost, at the end of the cover add your original and let the listener know that you make music and would appreciate feedback.
This might be a new way to conduct SEMM (search engine music marketing), Maybe… Just maybe, he will come back and buy your music/ attend your concerts and be a fan.
Get music back into Youtube and this time do it right. Let those IPR folks discover the old business model is done and dusted. People will subvert and find music to listen, while they are doing that i atleast hope the Aritst can win, not the company/ business.
January 5th, 2009 — General
This is the last in the series of posts about the RadioVeRVe. If you have at any point in time enjoyed our music and the artists featured. Here are a few things you can do to help
From the Artist Perspective:
- Go to shows/ gigs. There is nothing more satisfying to artists than a full house. I know the traffic is bad and you work far off, but try to be there.
- Buy cd’s. I know you have the music already. Buy it anyway
From RadioVeRVe’s perspective, Consider donating to the Open Source projects which make RadioVeRVe happen. Their list and donation links are listed below:
- Icecast/ Ices: The streaming server/ client which is the heart of RadioVeRVe. They are a great bunch of people who make amateur streaming possible. Please donate Here
- WordPress: The awesome blogging platform which we use for the RadioVeRVe site. These guys are absolutely awesome, you can donate Here
There are many more like Mysql, Apache and hundreds of others. But these are specific to our requirement and we are glad someone somewhere used their spare time to help us.
Thanks for listening, Do consider donating if possible.
December 31st, 2008 — General
This is a followup to my last post about the new RadioVeRVe. Before i get back to the deluge of things which need to be done, i would like share some of my experiences in releasing the new RadioVeRVe, hopefully there are somethings in this post which help you in your endeavors.
These days i have suddenly become a senior citizen when i go speak at colleges and meet people at conferences, This is slightly unusual and takes some getting used to. One of the common thing discussed is “How do you find people who help you in your projects/ ventures?”.
Too many people are stuck on this question, “If only i had three guys who worked with me, i could release this in 3 months. I am even ready give share/ stock in my venture”.
Today I can safely tell you “Stop waiting, get out there and start working, if you are doing the right thing you will find the right people eventually”.
Barcamp 6 (or was it 7) was the turning point for RadioVeRVe, i had finally given up on finding the right person to implement our(Me, Atul and Gaurav) vision for the website and had started to dive in and implement stuff myself. This was hard, i am not a css junkie and i generally had more things to do than i could handle. But i was open to putting in the nights and the weekends to get this done. And at Barcamp i met Praveen and Jonathan, to my credit i had some sort of new version to show them and explain what i wanted to do with the site, we got talking and soon enough both Jonathan and Praveen were working on the site as part of team RadioVeRVe.
It was slow at times, even painful sometimes. Weekends of rapid action were replaced with weeks of inactivity but during all of it there was a great sense of belonging of doing what we believed in. After 5 months of hardwork when we finally put out the new RadioVeRVe, it was one of the proudest moment for me. I finally felt like i can pull off anything from scratch and even make some great friends along the way. This confidence is new and i hope will drive me through ups and downs in the future.
If there is one lesson i learnt from this, and this applies to Open Source projects and Startup Ideas, Dont speak about the idea and expect it to draw other people. Start working on it, there is nothing like a demo to convince people to join you and make your idea a reality.
I would like to take a step back and thank Jonathan, Praveen, Shilpa and Arjun (Shilpa and Arjun have been part of RadioVeRVe for much longer) for all the work you guys have put in for RadioVeRVe. It just wouldnt have been the same without you guys.
December 23rd, 2008 — General
I am glad to announce a brand new version of RadioVeRVe! Tons of new music and features. Go ahead and give it a spin.
If you have the RadioVeRVe playlist stored, you might want to delete it and use the new PLAYLIST FILE, we had to make a few changes to improve how we stream.
September 5th, 2008 — General
Most of the people reading this blog feel great passion for their work. In that context we are probably different from most of the world. From what i have experienced most of the world treats work as a way to aid *real* life. While this is completely acceptable, there are certain advantages to being passionate about work and life in general.
- Purpose: Waking up with a list of things which need to be completed. Not drifting is a huge plus in life
Pride: We take great pride in what we do, we wear it on our sleeves. I work on X, i maintain Y and i contribute to Z.
- Identity: People who love what they do always have an Identity of their own.
Having said that, We also need to be aware of its disadvantages. To be successful we need to fight these. I will just enumerate one, cant think of too many others.
Pride: Wait a minute? Wasnt this an advantage? No its not. We are all so involved in what we do, the big picture somehow becomes obscure. We loose track of the objective and become involved with the effort. This is a continuous annoyance and requires great awareness to avoid.
The purpose of any task is to reach the objective we set out to achieve and the effort involved is pointless if that objective cant be achieved and we need to continously remind ourselves of that fact.
Pride gets in the way of progress. Its not about the effort, its about the objective!
September 3rd, 2008 — General
I have been reading a lot of reviews about Chrome. Most people seem to be comparing it to Firefox, which i think is underestimating Chromes capabality. I think its has nothing to do with the browser at all, it has everything to do with the platform.
For long we have touted web as a platform, Apple took the first big step when it chose the web as its Platform for the first generation iphone. That approach failed because the web as a platform was flawed, the reasons IMHO were as follows:
- Hard to write stateful web apps
- Lack of per website security (All apps run in the same address space)
- No access to hardware (relates to security)
- No offline support.
When Apple released an SDK it was an admission of defeat. Chris Messina put it beautifully in his blog post Did the web fail the iphone?. With millions of developers and amazing momentum the web has too much to offer for it to not be premium application development platform. So in that sense the browser was not the premium peice, the platform was.
When Google released Chrome, each one of those questions seem to find an answer.
- Hard to write stateful apps/ no offline support: Chrome includes Gears.
- Lack of per website security: Each tab in chrome runs in a seperate process, which also means this process can be given special permissions based on its security certificate! So your bank application can finally update to your harddisk and your passbook can be available offline! Wheee
- No access to hardware: Again comes down to security, read above
All in all the chrome platform is of great signifcance, i think its real competitors are j2me/java/cocoa/wpf and not the browsers. Only time will tell if the web as a real platform will materialize.
July 29th, 2008 — General
The 10th edition of the Ottawa linux symposium was held between the
23rd and 26th July at the Congress center in Ottawa. Linux symposium
is one of the premium linux conference in the world, Linux developers
from around the world meet and discuss various linux projects and the
roadmap for the future.
I landed in Ottawa on the 22nd. The city is a beautiful mix of
historical and contemporary architecture. In the evening all the
speakers were invited to a cruise acorss the canal, the canal flows
next to a lot of historical monuments. The discussions on board were
focussed on linux as expected, with people discussing topics ranging
from “Ext4 file system” to the “ipv6 stack on Linux”.
The first day of the conference started with Matthew Wilcox presenting
the keynote on “The Kernel: 10 years from in Review”. The talk took
the audience through the journey of linux with various anecdotal
references to Linus. Mathew also gave a brief account of his own
contributions to Linux and how being part of the community had
enriched his life.
The talks this year were dominated by topics which delve into the
kernel or system level features of linux. There was also decent
coverage of Performance, Security, Mobile, UI, Web Applications, and
Green Computing. I did a talk on “Mobile application frameworks” on
the 24th, Although i was initially a bit tense to speak in front of a
room full of distinguished hackers, my talk was well recieved.
The most controversial keynote of the event was presented by Mark
Shuttleworth, CEO of Cannonical (Ubuntu) who spoke about the need for
Free software projects to align their release cycles so a well tested
and professional product can be delivered to the users. I personally
thought that this would kill the diversity of Free software which
thrives on people scratching to their own itches versus a clockwork
release cycle which applies to everyone. Irrespective the talk
presented interesting ideas which will be heavily debated by the
communities for time to come.
Craig Ross, Co-Founder of OLS signed off the event by showing a graph
of T Shirt sizes over the last 10 years and how the T Shirt sizes have
grown linearly over time! He also announced that the next years Linux
Symposium would be held at Montreal.
Harald who recently won the Google-O’Reilly Open Source Award summed
up the event well when he said “It is unique because the key people of
major open source project are in
one place, This helps us be aware of each others progress and learn
from each other”
Linux(TM) needs no introducton these days its become main stream but
the people who make linux happen continue to be more than just
colleagues, they are friends who enjoy and thrive in each others
company. They excel New ideas are formed and implemented when they
meet, this is why a event like OLS is so important for this ecosystem.
The conference for all practical purposes is a open source project in
itself and its maintainers Andrew and Craig need to be hailed for
making such a wonderful event happen. Its projects like OLS which make
Linux and Open Source special.
I initially wrote this article for Live Mint but looks like they had word
limitations and had to chop, so this is the complete version
You can read the published article here
July 26th, 2008 — General
We are all smart people who care about our country and its people. Yet there are very few ways we can express opinion in way that matters. I have always spoken out about things are unfair but is that enough?
What more can i do?
Do i go on streets and protest?
Do i form a group which over time might have some political influence?
At this moment i am not sure, all i know is “If i cant do anything about things like this” then i have no right to cry.
I let “this” happen!
May 12th, 2008 — General
Yahoo recently launched a new feature in search called Yahoo Glue. Yahoo Glue shows search results segregated in different sinks like Yahoo Music, Last.Fm, Wikipedia and Youtube. When i first read about it, i thought this was a kick ass idea. Especially since i was once a avid user of the Firefox plugin Googlepedia which combines google and wikipedia results on a single page. While the over all idea and the implementation seemed quite slick, i discovered a few things which needed to be fixed.
1) Category is more important than source:
For a search engine, category of information is more important than the source of the information (discounting spam). For instance having a box called Last.fm Top tracks makes absolutely no sense in the context of the search request, what makes sense is “Top Tracks” which is the semantic identifier. Same thing holds good for Events, Biography and Videos.
2) Combining results is the prerogative of a search engine:
The first thing i tried when i used this was to check search validness, which is by definition what makes a search engine useful. I searched for “Kenny G”, knowing very well he is playing 2 floors below the Yahoo India office on the 12th of May.
And alas, no listing of the Bangalore event. What makes it even more worse, the event is listed on Upcoming (a yahoo property). This completely destroys the product being
i) Local search
ii) Glue across various sites.
It is imperative for a search engine to parse all kinds of sites and provide the user with a unified view of all that content. If i wanted a distributed search result from each site i would use them individually, the strength of search is the ability to combine them coherently, so i dont have to worry about combining them myself.
I think Events, Videos, Music, Biography, Articles, Talks (numerous others) are tangentials worth pursuing but not at the cost of search quality. The core features of ranking and categorizing need to be spot on for Yahoo Local search to be useful.