Rakesh Agrawal’s Blog

A piece of evidence that Mark Cuban is right that TV won’t be replaced by TV (at least not anytime soon): Even though digital music is pretty popular, 90% of all music is still purchased on CDs. Reed Hastings mentioned this in a recent interview that he did. This is same reason why I think that good old PVRs and TV time-shifting technologies are going to be around for a long time to come, even though TV shows are increasingly available online through channels like AOL’s In2TV and Apple’s iTunes.

I started testing Riya, the much-talked-about face recognition company, and here are my notes so far:

- There are people I don’t care about having Riya recognize — I’d like for them to NOT show up in the manual training. There’s isn’t a way to tell it to stop showing me certain faces so at this point, when I go into manual training, the first three or four rows are full of poeple that I just automatically skip over. And this makes each training “session” less useful.

- The first thing I wanted to do after I did some “training” was see what faces Riya had properly recognized based on my training. The way I see it, face recognition is the reason why people are going to use Riya so it’s the place a user should be taken after they’ve done the work of uploading and “training” (instant gratification). As it is, I had to hunt and peck for this screen, I wasn’t automatically taken there.

- On the “People” tab, I want to have a count of how many photographs it’s found for each person I’ve trained it for… This would be a good measure for me to quickly get an idea of how well it’s done.

- On the “People” tab, the left and right browser arrows don’t appear to work in the “Browse unrecognized faces in your photos” section.

- I’m curious about the percentage of faces (in my photographs) that Riya has recognized on its own. I have the total number of faces recognized, but I don’t know how many of these were because I specifically said this face = this person. On one end, this number could be zero (this is the scenario where I’ve taught Riya 100% of faces in my photographs) and Riya would be adding no value. On the other end, It could be something high and that would indicate (assuming a low rate of false positives) that Riya’s recognizing a lot of faces.

- The uploader wasn’t as effective as it needs to be — it should have had Picasa-style directory selection/de-selection. As it is, I do a lot of exports from Picasa and so I have lower-res duplicates of a lot of my images at My Pictures/Picasa Exports/[Album Name]. And all of my native resolution pictures are at My Pictures/[Album Name]. So as a result, a believe Riya’s uploading a lot of stuff twice.

- I manage all of my photographs using Picasa. It would be nice if Riya integrated with this so that it picked up edits / crops / rotations that I have made to photographs (Picasa doesn’t change any of the source image files, it just stores the changes in an INI file and applies them whenever you try to do anything with the file outside of Picasa like send it out for a print, do an export, email photographs, etc.). If there was integration, I could also tell Riya to only worry about the photographs that are marked with yellow stars.

- Is Riya uploading the high-resolution images (3-4MB per image)? Or is it just doing image processing locally on my hard drive (utilizing the higher-resolution data) and then uploading a lower resolution version of the image along with the results of that image processing to Riya? Does image recognition quality get incrementally better at higher resolutions and then plateau past a certain resolution?

- Summary: So far, it’s been fun and interesting. But what I really need to do is assume that Riya will integrate nicely into my photo ecosystem (ie my multiple cameras, my portable hard drives full of photographs, Picasa, and flickr) and figure out whether this facial recognition stuff is actually useful. It’s certainly cool, but why do I need to have photos referenced by person? Does it improve the photo sharing experience? It’s more often that I want to share photographs from an event or by theme, but it’s not too often that the theme is a person. I can think of some instances where it would be useful, like if I’m putting together a slideshow for someone’s birthday — then having all the photographs of that person is useful. If I’m looking at really old photographs, I think it might be useful to be able to pluck out photographs by who is in them. Maybe. Like I said, this is one that I have to figure out.

1600+ photographs uploaded so far, I’ll continue to report on how well the technology works and, ultimately, how useful I find it.

Tag: riya

While the debate around book search rages on in the U.S. tech community and around the world, witness the power and efficiency of book search in New Delhi, India at Teksons Book Store (GK-1 M-Block market):

1. I execute a query for ‘Lord Ganesha’s Feast of Laughter’ with a store manager

2. The store manager dispatches two resources to search for the book (see photograph below)

3. The book is delivered to me with a response time of 5.342 seconds

Copy of the mail I just sent to the barcampdelhi Yahoo group:

> Hey folks,
>
> I’m organizing a geek dinner (with help from Gaurav @ Tekriti) on
> Monday, March 13th. At Gaurav’s suggestion, we’re going to meet at
> the DLF Mega Mall food court in Gurgaon at 8:30pm. Please RSVP on
> e-mail (rakesh at snapstream dot com) or by phone (9810497266). Please pass
> the word along to other geek friends, even if they weren’t able to
> make it to barcamp. It’ll be a good chance to connect again after
> barcamp this past weekend.
>
> Hope to see a lot of you there!

I spent all of Saturday this past weekend at BarCamp Delhi (a nice coincidence that it was scheduled while I am here in India for a wedding!)

My summary of Barcamp Delhi:

* I love Delhi (lived here for five years during middle school and high school) and while I’ve spent a lot of time here, I’ve never been around the kind of smart, motivated and talented technologists that I spent time with on Saturday. I expect great things from these people!

* Meeting the folks that I did really hit home what a small technology world we live in and how much smaller it is thanks to blogs. It was a bit surreal (Scoble has described the same sentiment before) to be sitting in Adobe’s office in NOIDA and hearing folks in the audience dissect and discuss technologies like Google News, gada.be, Windows Live and Gmail.

* Adobe’s office in Delhi is really nice — a big thanks to them for giving us their conference rooms, for feeding us lunch, and last, but certainly not least, for the fast and free wi-fi. And while I’m at it, big thanks also goes to the folks that catalyzed and planned the event: Gaurav, Jonathan Boutelle, Manik Juneja, and Amit Ranjan.

* Manish talked about microformats as a solution for knowledge management in organizations and in this context, he had a screenshot of a microformats plug-in for wordpress. I’ve read about microformats and understand what they are all about, but seeing it integrated into a blog publishing platform wasn’t something I had seen before and it was very cool. Manish’s is a great speaker, but I felt the content was a bit jargony and maybe too academic in places — I could tell from the Q&A that Manish gets this stuff so my only criticism here is that IMO he should have communicated it in clearer simpler language. Anyways, I’m still not sure about how structured user-generated content is going to come into existence, but microformats deployed through blog publishing platforms seem like one viable approach. There are other approaches like Edgeio and Google Base.

* Want to befriend an Indian engineer / technologist? Buy him a beer. In particular, buy him a Foster’s.

* Gaurav (one of the barcamp organizers) had a great presentation on bootstrapping a company — it was succinct, to the point and engaging. There were a lot of things in his presentation that I’ve learned myself through personal experience or from others, but I particularly enjoyed how everything was contextualized to India.

* Abhishek Agrawal had a presentation entitled “Why startup?” that preceded Gaurav’s. I missed most of the presentation but caught the tail-end of it and I could tell it that it was a good session. Abhishek carpooled with me to and from barcamp and he struck me as a well-informed, analytical and motivated technologist. I look forward to following what he does with his new company.

* Some of the presenters were talented speakers — VeerChand Bothra and Anuj Khurana were two guys that I’ll single out.

* There were no female presenters, though there were several women in attendance.

* Sorry I couldn’t join everyone for drinks later, but I plan to organize a geek dinner in the next week or so before I head back to the States. If you were interested in joining, drop me a line at rakesh at agrawal dot org.

Barcamp Delhi was great and I hope to see this technology community in Delhi really grow and flourish over time!

• BarCampDelhi tagged Flickr photos
• barcampdelhi

Just finished up the day at Barcamp Delhi and it was great. More on the conference later, for now check out the photographs that I took (or check out all flickr images tagged with ‘barcampdelhi’)

barcampdelhi

As I’ve mentioned earlier I’ve recently begun to use a Windows Mobile 5.0 phone, the Cingular 2125 aka the HTC Faraday.

The phone’s EDGE/GPRS connection has proven to be very useful here in Delhi where one can end up spending a lot of time in the car driving from one part of the city to another (though the traffic gets better every year as the infrastructure improves). So I’ve been using an Airtel SIM chip with an unlimited GPRS subscription and one incredibly cool and useful service that I’ve been using on this trip to India is Shozu.

Here’s the usage scenario for Shozu:
1) snap a photograph with your phone’s camera
2) tell Shozu whether you want to upload the photograph to flickr
3) if you answer yes, it automatically uploads it to flickr

(And if you want, you can configure it so Shozu skips step #2 so every photograph you take automatically gets uploaded… not a feature you’d want to turn on if you were, say, Paris Hilton.)

I used this feature all night at the wedding last night so my parents, who are sorely missing Ananya back in the States, could see near real-time photographs of her:

• Ananya eating some grilled corn on the cob
• more corn on the cob
• Shonali and Ananya
• Ananya chills out at the wedding

Shozu installed pretty easily on my 2125 and linking my Shozu and flickr account into the Shozu software on my phone was pretty easy too. I particularly like the way Shozu seamlessly integrates in with the camera software on my Windows Mobile phone. Immediately after I snap a photograph, the Shozu screen seamlessly pops up to ask me if the photo I just took is something I want to upload. The uploading is also seamless and in-the-background and I could easily check to see whether images I had chosen to “Shozu” had actually been uploaded or not.

The only downside is that the camera on my 2125 (like most phone cameras) isn’t that great. Actually, it’s pretty terrible for anything other than, well, photos where I don’t care about image quality. So, what I really want is to have this built into my digital SLR (like Jason Calacanis put on his Christmas wishlist). Nonetheless, I see myself continuing to use Shozu a lot and it’s a great pointer to the future of Internet connected consumer electronics devices.

The other thought I have is this: considering that early reviews of Wi-fi digital cameras haven’t been that hot, the people designing these products would do well to carefully study the Shozu experience.

I had to download some large files (1gb in size!) and my home broadband connection just wasn’t cutting it. So in search of a faster internet connection, I ended up at the Reliance WebWorld internet center / cafe in New Friends Colony. The experience was terrible — someone at Reliance, hire some real programmers! The sign-up process was four steps but each step took about 5-7 minutes of processing. I asked one of the attendants why this was the case and he gave be the bizarre explanation of “Well, it has to check your desired user ID against the entire national database of Reliance WebWorld user IDs.” Huh? And then, once I was logged in, it was this very crippled kiosk shell on top of Windows with the filesystem locked down. I was eventually able to get around the filesystem being locked down and when I did, to my disappointment I was only able to get download speeds of about 8-9KB/s.

I love the Reliance WebWorld concept and the fact that they are all over the place here in Delhi. These shops will prove to be very valuable to Reliance as it sells more and more consumer services. But they have a lot of work to do on the user experience.