|
|
Saturday, June 20th, 2009
| |
4:45 pm - Blue Vector is hiring
|
Blue Vector, in the rfid management software space, is hiring. They are looking for an experienced C++ engineer. Details:
Profile: Software Engineer /Senior Software Engineer
Experience: 3-7 years.
Desired Skill Set:
- Strong C++ and OOP in unix/linux development environment
- Strong in data-structures and algorithms
- Client server systems, multi-threading, RDBMS a plus
- Network Programming (exposure to frameworks like boost, ACE etc is a plus)
- Basic RDBMS, SQL skills, and some scripting language like Perl, PHP, Python
- Basic Web technologies (HTTP, HTML etc)
They prefer Bangalore based candidates. Send your resumes to ajay at bluevector.com
|
|
(comment on this)
|
| Thursday, June 11th, 2009
| |
2:02 pm - WisdomTap updates
|
My company, WisdomTap, has quietly been rolling out a lot of features, and I wanted to write about them.
- We have a new Home Page which hopefully does a much better job of introducing the new user to WisdomTap, what it does (product search), and how it does it (based on expert user comments across the best web communities). The product snapshot gives a quick overview of the features product - overall rating, an expert opinion, and the 3 best uses for this product. We have highlighted a selection of top ranked, new, and products with deals. We also show 3 opinions from a random selection of expert users we crawled from across the web. This is to tell new users that we're tapping the wisdom of the web's best experts. Finally, the navigation is much improved.
- As part of the new home page design, we changed the look and feel of all pages - specifically, the header and the footer. The colours are much more bright and attractive (in my opinion, of course!)
- We enhanced our IWannaBuy Facebook social shopping app. You now have the ability to message a specific set of friends to ask for their opinions on what to buy. For instance, I might want to make sure that
![[info]](http://l-stat.livejournal.com/img/userinfo.gif) yathin and code_martial help me choose a dslr. The other change we made is that on the "your polls" page, you can see which friend recommended which product. The thinking behind both these features is to instill more trust in the recommendations.
Do give both our product search site and our social shopping app a whirl, and let me know your feedback!
|
|
(comment on this)
|
| Tuesday, May 26th, 2009
| |
11:02 am - Saying no to no
|
I'm closing in on two years since starting my own company (WisdomTap, providing product search). As I was thinking about certain recent events, something struck me. As part of building the company, providing unique technology, we've faced a lot of questions from a variety of people, from technical advisors, from VCs, from potential clients and from potential partners.
- Is the problem technically tractable?
- Is your market big enough? What's your market, actually?
- How is it different from product xyz?
- Why will someone use it?
- Why will they pay for it? How much will they pay? Why will they pay so much?
This is just a sample of the questions we faced. Notice how negative everything is?? That's my insight - so many of the people seem to look at the downside, rather than at the upside. i.e., what are the various ways in which the new venture may fail? This was common across all class of people we interacted with, *except* advisors and users! Advisors are the only ones who balance the business downside by looking at the upside. And, of course, users of the products also give us a lot of feedback - both positive and negative.
I wonder whether this "these are the various ways you can fail" attitude is an Indian thing. I'm especially struck by how negative potential investors (VCs, angels, etc.) are. We have had tremendous support from our original angels, though.
So, my big takeaway as a startup founder - find the conviction and the inner strength to say no to the nattering nabobs of negativism. As someone starting a new business, you're very aware (or soon become extremely aware!) that things will not go according to plan. There are a lot of unknowns. While its good to be aware of potential pitfalls, you've got to remain optimistic, always believe that you will hit the upside.
|
|
(2 comments | comment on this)
|
| Saturday, May 9th, 2009
| |
11:52 am - Leadership
|
Peter Roebuck has a great article on the importance of captains in cricket. After watching the IPL this season, and of course, based on my own stupendous knowledge and experience of the game, I have a few observations to make. First, I think that you can write off the Mumbai Indians. I love Tendulkar the batsman. I don't think that there's any better that I have seen. But, he's a terrible captain. In the past few matches, he's been getting pretty much everything wrong. How can you not bowl out Bhajji's overs? And, in the last match against Delhi, when attack was their only chance, the field was too defensive. And his body language said that the game was lost before they started bowling.
Take another useless captain - KP. At least he was trying hard on the field, and never seemed to be disheartened. But, tactically, he was poor, to say the least. Why keep bowling Praveen Kumar at the death, when he was getting murdered? Why keep playing Uthappa, and in the same slot? When things are not going well, you've got to change tack. What a huge change Kumble has made!
Delhi is the strangest team of the lot. I don't think that Sehwag is a good captain - he's very aggressive, but doesn't know what to do when things don't go his way. However, they've been winning pretty much all the close games...
I'm sure that there are lots of lessons we can learn and apply in the business world.
|
|
(comment on this)
|
| Tuesday, May 5th, 2009
| |
2:11 pm - Scaling software
|
I recently went through the very interesting description of Facebook's photo hosting stack. It reminded me, yet again, how scaling web software is both simple (in concept) and difficult (to practice) at the same time. Here's a simple conceptual framework to follow which will solve an estimated 80% of web scaling problems:
- Identify your bottleneck. Is it cpu, disk, memory, network bandwidth?
- Fix it!
- Repeat
:)
Some tips while fixing bottlenecks -
- Memory access is orders of magnitude faster than disk access.
- Your memory should be enough to hold your working data set (which typically follows extreme power laws - say, 99-1 - compared to actual data set), so its important to determine your working data set size.
- Design your system to be able to scale various layers independently.
- Watch out if you've configured your web server (apache is the one that I use) to be able to use too many server processes. Context switching and swapping will kill you under high load
- Do high load tasks as infrequently as you can. For instance, load some back lookup table once at startup time. Or, open a MySQL database once per http request, and reuse it for all db access during that request.
Of course, there are probably lots of unique cases which can't be solved so easily conceptually.
More resources - Cal Henderson - ex-flickr, yahoo - has a great presentation on web scalability. There are lots of others available as well.
Comments welcome!
Of course, there are probably lots of unique cases which can't be solved so easily conceptually.
|
|
(1 comment | comment on this)
|
| Saturday, April 18th, 2009
| |
1:01 pm - Out of touch?
|
In the upcoming elections, my constituency is strongly middle class - I want personal safety, good roads, electricity, no corruption, efficient government, and an opportunity for good jobs. I don't care (*really*!!) about the caste, age, or religion of the person who can deliver.
In the limited time that I get to see TV, and read magazines (outlook), I start to wonder if I'm out of touch with much of the electorate in India. For instance - almost everyone states that much of the voting is on a caste basis first, and religious basis next. This boggles my mind. Is it because voting on caste/religious lines has largely proven beneficial to those? Or, is it that those voting on caste/religious lines are scared that if they vote outside their caste/religion, things will get worse? Do many people realize that their lot is quite pathetic (look at our pathetic performance on the millennium development goals, and can be easily set right? Having interacted with some people from a completely different social background, albeit in cities, I do know that they have high aspirations for themselves and their children. How is it out in villages? Slums? Small towns?
Are we (you know, educated, well off, blog reading desis) too elitist? Are we completely out of touch? Too relatively comfortable to do anything about this? (Yes, these are rhetorical questions).
What about people who follow the vernacular press? Is their coverage of issues any different from what we see?
If we want to fix a problem, we must first recognize what it is.
|
|
(1 comment | comment on this)
|
| Tuesday, April 7th, 2009
| |
10:17 am - usability frustrations
|
I have a nokia 6300 and an airtel connection. While the builtin browser works, I can't get any other apps requiring internet access - like the gmail app or yahoo go or opera mini - to work - they error out saying "subscribe to packet data first". In my connection settings, there are 3(!) entries for something called mobile office, whereas the builtin browser uses some connection called Airtel Live! (with the ending quote). Any idea what's going on, and how to get apps like opera mini to work?
The Nokia menus are quite indecipherable. There are multiple options dealing with connectivity. Its extremely confusing, and quite badly done. Shame on them!
|
|
(4 comments | comment on this)
|
| Saturday, April 4th, 2009
| |
4:01 pm - guest post on WisdomTap blog
|
|
Ace photog and fellow birder yathin has written a kick ass article on how to start building a dslr collection - check it out on the WisdomTap blog. Thoughts?
|
|
(comment on this)
|
| Friday, March 13th, 2009
| |
10:03 am - Adwords gripes
|
My company - WisdomTap - is doing some search engine marketing for our conversion engine - and I have run into some major annoyances with the Google Adwords tools.
First, I get a 404 for following some internal links like this one!!! WTF???
Next, if there are a set of topics - say, all about campaigns - its a set of links. I HATE HATE HATE this style of creating web pages where you have to unnecessarily click through to topics, when it could have/should have been all on one page. Terrible usability!
For a product which is driving your company's revenues - its mind boggling that such lack of attention has been paid to such trivial issues. Ah well, the joys of a monopoly, I guess.
|
|
(comment on this)
|
| Wednesday, March 11th, 2009
| |
1:51 pm - New camera index
|
|
| Monday, January 26th, 2009
| |
7:09 pm - Happy Republic day!!
|
Dear Denizens of Middle Class India, have you forgotten 26/11 already? Has your mind been numbed so soon? Has the wax of a million candles - and the lifeblood of around 200 people - been shed in waste? Are you going about building up an extra layer of fat on your esteemed middle class arse - or are you actually doing something about it? (I'm in the former category, I'm afraid).
Don't know what to do? Never fear! That's what I'm here for!
- First and foremost, support, and insist on police reform. If you don't put the fear of breaking the law into people - politicians, bureaucrats, drunken speedsters, etc. - all hope is lost. Create a law abiding and law fearing society first.
- Look to civic activity organizations, such as Janaagraha, for more ideas on how to participate.
- You *are* registered to vote, right? Start interacting with your elected reps.
- Consider an alternate party, such as the Lok Satta
That's it for now! Please go back to sending your outraged sms'es to the TV channels (btw, both the phone company and the TV are making money out of your outrage).
|
|
(comment on this)
|
| Wednesday, January 14th, 2009
| |
4:58 pm - More useless startup gyaan
|
I was at Headstart 09 last weekend, demo'ing my company, WisdomTap's, product search capabilities. We even got nominated as one of the top 10 product startups in India! While we personally had a great time interacting with the visitors, and made some important contacts, a few things about startups in India stood out:
- The number and variety of startups was mind boggling. Its just fantastic to see such energy and innovation out of India. And, to think, that this is just a fraction of the actual Indian startup scene. Heartening indeed!
- Venture Capitalists are revered and worshiped as though they're stars.
- Almost universally, the UI was "raw". I can't remember any company who's UI had the necessary pop.
Let's take these one by one. First, the number and variety of startups - surely, its a matter of time before we have a monster hit out of India? However, there is a palpable lack of experience in selling and marketing (my company included). The other big problem is lack of an audience in India. In general, there are more technical guys than sales/marketing types at a startup. Hope that this changes, and fast.
Next, the way people are awed by VCs. It was almost as though VCs were expected to be the enablers of startups in India. All the VCs I interacted with seemed like nice guys (i.e., not the kind of pricks you read about at the Funded), and some of them had some excellent knowledge to share (Ashish Gupta of Helion had a kick ass keynote which was very poorly attended). However, my impression after trying to raise money is that VCs want to invest at later stages, when the risk is less, and expect a bigger share of the company, than expected by most startups. i.e., there was very little chance of any of the startups at headstart getting funded by a traditional VC. So, why do Desi startups still look at the VC community with stars in their eyes? I think that part of it is the excellent PR the VC community worldwide has received. The other part is probably the lack of marketing, sales, and finance people in the startups - so perhaps we tend to believe that getting a VC to fund us would somehow magically enable this aspect of the company...let's stop worrying about VCs, and start worrying about customers and how to acquire them, people!
Another important aspect which must be made very public is that there is a big mismatch in the expectations of an Indian startup to get funded, vs a traditional style VC funding them. There are just too many more mature organizations - not necessarily in technology - competing for this money, and they are less risky from an investors point of view. Also, the amount of money required to run a startup is far less than one thinks. I think that there is a huge opportunity to create some kind of Angel Exchange in India.
Lastly, the UI fiasco. Given the talented graphical designers, and the ability to learn about UI best practices from across the world, its a shame that we're not seeing something better. Oh well.
Lastly, more power to the headstarters to continue helping the nascent tech startup scene in India.
PS: My company, WisdomTap, provides a product search engine which helps you decide which products to buy. If you're looking for reviews, ratings, and recommendations for digital cameras and lenses or mobile phones, do check us out. We also provide this same service to ecommerce retailers to help them increase conversion rates on their site.
|
|
(4 comments | comment on this)
|
| Wednesday, January 7th, 2009
| |
6:11 pm - Neighbourly "love"
|
What does one do with a failed state like Pakistan as neighbours? Take this statement by their FM, Quereshi, from this article:
"On the other hand, God forbid, if war is imposed on Pakistan, then Pakistan will fight back. And every Pakistani will fight back with full commitment and force of conviction. But we do not want a war."
Does this mean that the act of terrorism in Mumbai - if supported through acts of omission, or worse, commission by another Nation - is not to be construed as an act of war? What, then, is an act of war? Suppose some "mercenaries" or "local malcontents" hit terrorist camps in PoK - would that be an act of war?
Consider, also, the Pakistani cry for "proof", and calling whatever is presented is "insufficient" - classic stonewalling tactics. Also, Pakisitani statements indicate that they seem to be convinced that India is going to attack it. Does anyone believe that India actually wants to go to war with Pakistan? Even those fringe loonies on the right only want reactive war, i.e., to a provocative act (or set of acts) - whether actual or real. Its downright surreal that one country - India - is attacked by people originating from another country - Pakistan, and the attacking country claims that its the one being violated!!
Given that both are nuclear states, with Pakistan seemingly quite desperate, war is out of the question. I wonder what the answer is, though? How does India get out of this mess? What do you do with neighbours who want to keep fighting you, even if all you want to do is ignore them? What kind of stick does one use without risking nuclear annihilation? One clearly does not Israeli style of lunatic retribution (anybody paying attention to the photos of the dead Palestinian children??) Any ideas?
|
|
(comment on this)
|
| Thursday, December 18th, 2008
| |
8:24 am - The sine wave
|
Its been just over a year since my company formally registered as an entity with the government. Its been 6 months prior to that since I quit my last job and started up. I have received many (0) queries from eager followers (2.5) who want to know, what are my learnings? Should they start a company too?
In no particular order, here goes:
- I like to write code. I like python. I am ok with using java. I miss c++ :(
- I totally get Hofstadter's Law. It especially applies to getting paid customers. Even this frigging post is a month late!!
- Its tempting to spend money. However, its better to postpone any spending decision as long as possible. For instance, waiting a couple of days will get you a cheaper deal on an UPS. There are NO exploding offers when you're the buyer.
- Rent, don't buy.
- Its cheaper to buy a built up computer with branded components than a branded computer.
- Power situation sucks. Extremely.
- An outcome can have unexpected consequences, which become apparent only later. Were not able to hire someone? More money in the bank, longer runway. Didn't get funded? Focus on building a product which people will pay for. Launched in alpha? UI sucks.
- It takes an inordinate amount of time out of your life. I feel that I'm mortgaging the present for the future. If you're founding, find an office within walking distance. If you're joining one, make sure that you can either commute quickly, or wfh at nights/days.
- Its important to just start it - once you jump in, you will adjust to many things - including a lack of money.
- Don't buy ulips!! They are a f***g money sink
- I'm still a better hacker than code_martial. I am pysching myself up to a stage to get my code reviewed by hitesh. The alert guys still wake up crying at night from memories of his last code review!
- My favourite camera in the whole world is the Nikon Coolpix L1. That's a separate post in itself.
- Its easy to waste money on useless stuff, like contractors and bad UI designers. When it comes to outsourcing work, either go with someone whom you know and have worked with in the past, or insist on seeing their work and make sure that the quality will meet your standards.
- Good webdevs are worth their weight in gold. I highly recommend the team at Hummaa.
- Ideas are worthless. Execution is gold. Don't worry about someone "stealing" your idea.
- Its easy to write crappy, unmaintable, shitty code.
- Nobody listens to me, even when I'm right all the time.
- Release early, release often.
- Free software (of the Gnu kind) rocks. Thanks to all the coders around the world, you really are making our world a much much better place.
- Machine learning is a cool tool. Its surprising how many things it can be applied to
- Its also surprising how much information is out there on pretty much anything - say, like SEO and SEM - and yet, how few people do it well.
- Mathematics is also immensely useful and applicable. I'm ashamed at how much I've forgotten, and have had to go back and relearn. I'm waiting for my kids to grow up so that I can relearn with them.
- You'll hear a lot of pointless advice - "make money!" (how??), "If I were you, I'd frobnicate the doodad and sell to the grand poobah" (well, you're not me, and why don't you do it if you're so sure?), etc., etc.
- Its tough to separate the pointless advice from the useful ones. Hey, who knows, maybe grand poobahs are dying for a frobnicated doodad!
- When you're building a complex, new, technology, its essential to set measurable quality goals for every component, churn out the first version very very fast, and then to obsessively go over the results. We lost a lot of time because we were late with our V1, and we didn't pay the kind of close attention to the quality of the output.
- Test driven development rules. But, nobody wants to write tests. People just want to churn out code and throw it over the fense. Insist on sane development practices like coding standards, source control, packaging, right from the beginning.
- For heaven's f****g sake, comment your damn code already!
- Sales is tough. I have immense respect for salespeople and marketers. Engineers don't realize their value.
- You've got to ask first. If you ask, you may get - if you don't, you definitely will not get. Leave your introvert nature at your last workplace.
- The number of people who are willing to help you is astonishing. Thanks guys!
- Reciprocate by helping others wherever and whenever possible.
- Experiment as fast as you can. Make sure you have systems in place to measure the outcome of the experiments. Its attention to detail which will help you execute efficiently.
A plug - if you know a camera or a mobile retailer, ask them to check out our products. We can help with conversion rates, increased sales, and search engine optimization (SEO).
|
|
(10 comments | comment on this)
|
| Saturday, November 22nd, 2008
| |
9:53 pm - mandatory checkbox with FormEncode
|
If you're developing web apps with pylons - a very cool wsgi compliant framework - you probably use formencode for validating forms. Formencode is very useful, but it takes a bit of bunging around the code to understand it well. One of its really useful philosophies is to throw an error if the form is not filled as expected - say, a format mismatch, or a mandatory field missing.
I recently ran into an issue where I have a form where the submitter needs to agree to the terms of service to continue. It should have been fairly simple - all that was needed was to specifiy that the tac_agree field was not empty. But, it took a fair bit of doing. The *only* thing which worked for me was the idea that there are other mandatory fields, and I can therefore use the RequireIfPresnet chained validator. So, my code to ensure that the checkbox is checked looks like this:
chained_validators = [validators.RequireIfPresent('tac_agree',
present = 'email',
messages = {'empty' : 'Please agree to continue'})]
Hope you found that useful! I spent a good bit of time head scratching to get this to work.
|
|
(comment on this)
|
| Thursday, November 20th, 2008
| |
5:23 pm - Going live
|
I quit my last job at Yahoo more than a year ago to follow my dream of starting my own hi tech software company. I thought that it would be easy. I thought that it would be fun. I thought that I'd have world singing my praises. I thought that - actually, I don't know WTF I was thinking!!! :)
I did want to build a product which would push the existing boundaries in some area, something which would help a lot of people, and something that I'd be proud of. Along the way, I even convinced an equally crazy dude to drop his comfortable life and jump in.
After a year and a half of slogging, we are very proud to present WisdomTap product search and the  WisdomRank Conversion Engine. "Say what??", did you say? If you cut through the attempt at marketing speak, the product search is a search engine which has crawled the best communities on the web - like DP Review and POTN, analyzed this to find who the expert contributers are and which products they like, and present this using a search interface. You can use product search to find which is the best product for your particular need. You can search based on what you want to use the product for - say, to find a lens for wildlife photography. Or, you can search by product search, for instance, a wide angle lens. You can even research a particular product.
Everybody researches what to buy on the web these days - we're helping them find the best product for their needs based on the wisdom of the experts in the web's best communities. We believe that we're on the way to opening up a very large portion of the web - discussion forums and other web communities - to the wider web audience. While we've fine tuned our algorithms on the digital camera domain, we're in the process of rolling out this search for a whole slew of products.
Besides helping the web user decide what to buy based on the opinions of the best web communities, we are also providing this as a service for e-commerce retailers to provide this decision making data on their site. We help convert confused browsers into decisive buyers who know exactly what they want to buy!
Do try out the product search for finding your next digital camera related equipment, and give a shout out to your friends if you like it. If you feel strongly about a camera or lens that you own, get a badge for it to show on your site. And, of course, send me your feedback!
Last but not least - thanks to all my innumerable supporters along the way. If I have seen so far, it is because I stood on the shoulders of giants.
|
|
(4 comments | comment on this)
|
| Monday, October 6th, 2008
| |
1:11 pm - Holiday!
|
After a long time, finally went on a holiday with the family. We went to GRT Temple Bay in Mamallapuram - which is 40km south of Chennai. The resort itself is fantastic - its so close to the sea, that they have had to build a retaining wall. The ambiance is out of the world and the weather was perfect for soaking in the pool or the sea. The rooms, and the service, is great. The food was ok - not gourmet, but not bad either. If you're a fish eater, the choice is much wider - although i was surprised at the lack of south Indian dishes on the menu. They have a swanky specialty restaurant specializing in Tandoor, grills, and pasta. Go figure.
I had always been a hill-station holidayer, but this holiday has converted me. Plus, my daughter had such a grand time at the beach!
We drove to get there. Driving on Indian highways is - interesting. The cretins are out in full force. Mofos change lens at the drop of a hat without checking if someone is coming fast down that lane. This is exacerbated by the cleaners on lorries and buses, whose main job is to stick their fucking hand out before, after, or during the lane change. Idiots cross the road in front of you, even when there's no one behind. Particularly dumb in-bred SOBs drive on the fast lane of the highway with their headlights on - but going the wrong way. This is so that they can avoid going an extra km or two to make a u-turn. Avoiding death is of course not important.
Net-net, it was a BP neutral trip - what came down in the resort came right back up on the roads.
|
|
(5 comments | comment on this)
|
| Monday, September 8th, 2008
| |
3:54 pm - Bihar
|
|
Any ideas on how best to help the flood victims in Bihar with contributions? Is it worthwhile to contribute to the Prime Minister's relief fund, or is it another corruption sink? Any other charities/NGOs/GOs to donate to?
|
|
(2 comments | comment on this)
|
| Friday, September 5th, 2008
| |
10:10 pm - Twittering
|
|
| Friday, August 8th, 2008
| |
3:45 pm - just say no!
|
|
to hiring freshers if you're running a startup. OTOH, if you want to mold some clay (aka chew out some ass), *and* are a masochist, give it a try.
|
|
(comment on this)
|
|
|
|
|
