Building The Team

8/31/2005

37 Signals have an interesting take on establishing the right team:

But keep it in mind: three for version 1. Remember, it's better to make version 1 half a product than a half-assed product. Three people will keep you closer to half a product and a cleaner, tighter, simpler base on which you can grow later.

One person is bad, because you'll usually be battling your own sanity when it comes to timely product releases. I think two is enough. Here's my criteria: both must be visionaries, flexible to make changes to the business plan and technology near the 99% completion milestone, and maximize the yield of the Venn diagram with two stakeholders - us (creators) and them (consumers).

New iPods Next Week?

8/30/2005

Apple is having an exclusive music event next week. I think they are going to do an overhaul to their iPod lineup. My predictions are:

iPod Mini with color - they've done this with the larger iPods and it's the next logical step
iPod Shuffle with a screen or better UI - this needs to happen
iPod Video - they'll make a catchy name for this too

I speculate a general increase in storage across the lines and a price reduction for older pods. There might even be an update to iTunes and some firmware updates as well.

Downtime

8/23/2005

It seems that the server decided to take a nap today. Well it was more like a nice long 10 hour sleep. Let me know if you still find any glitches. I'd actually like to re-write this site purely in Python or Ruby with XML/XSLT transformations. But that would require building my own "Blogger" - and I don't know how I feel about that yet.

Posted from MS Word

8/16/2005

I posted this from Microsoft Word. The Blogger developers are my heroes. Download the plug-in to post from Word too. I can already see a couple benefits to this:

I can use my own custom dictionary on my computer (I cannot live without native auto-correcting)
I do not have to worry about converting HTML entities
I do not have to copy and paste anymore

Although I have to re-write my current song/mood feature and make a Technorati tag macro. Oh well, time to preview this live post!

What Does Free Software Have?

8/14/2005

Jonathan Schwartz of Sun Microsystems has a great post entitled Free Software Has No Pirates. Perhaps the most interesting line from the whole essay is:

And so I thought I'd put down, once and for all, why we're committed to sharing, to open source, open standards, and eradicating the digital divide. Ready? Because we're going to make more money. How? It's trivially simple. Why do carriers give handsets away for free? Because they make money on the subscription necessary to receive the handset...

We'll find out how this strategy fairs. Personally, I think he has identified a great opportunity. I'd go as far as saying that free software has loyal hackers and advocates. The subscription model is not dead.

Email Habits and Storage

8/14/2005

While reading some blogs, I came across an interesting post over at 37 Signals regarding email habits:

However, we were tossing this idea around the table: Why not read an email and then instantly delete it? Why do we save emails? Why do we archive them in folders for safe keeping? We don't save phone calls. We have a conversation on the phone and then we hang up. If we need to take notes for whatever reason we do, but 99% of phone calls are completely ephemeral. And if we forget something, or we need it again, we just make another call.

I save all my mail for several reasons:

One day, semantic web applications like Haystack will help me organize and link data (contacts and content) across different formats. Although a part of me believes that this type of semweb app might just complicate things ten-fold. Some of us are already feeling the crunch of information overload.
Friends like Denise and Justin save and archive everything. It has helped them when looking up historical data so I decided to try it out. After a while though, searching just became a hassle in Outlook.

Services like GMail have conditioned into thinking that we really need 2 gigabytes worth of storage. So, like any consumer, we simply decide to try to use up all that space. In otherwords, there is a correlation between habits and storage capacity. But honestly, who has that much saved email? It seems excessive and inefficient. No organization scheme - folders or labels - can optimize searching through a mailbox at that size. However, large mailboxes make sense if we want to share rich content like audio conversations and videos. But in order for that to happen, we need a faster connection or two.

Doostang - A Better Social Network

8/11/2005

I had the privilege of talking to Mareza, co-founder of Doostang, on a long drive home a few weeks ago. For those that don't know, Doostang terms itself as a "trusted network" that enables you to find jobs. Basically, it's like LinkedIn but inherits more social network qualities. I've been playing around with the site and found it both intuitive and interesting. For instance, there are no advertisements (Friendster and MySpace fall here) - I guess I don't need to roll my own Greasemonkey scripts to block ads! When stacked against Friendster and MySpace it is comparatively fast. A typical social network places emphasis on photos - but Doostang encourages you to upload your resume.

Here are some of my recommendations to Mareza:

Fix the problem of opening new windows (lack thereof attributed to what appears to be a .Net quirk) on the listing pages.
Don't expand too fast - you don't want to compromise the quality of jobs with the total number of users (look at Orkut). Review the invite model: maybe require two invites after a threshold for building a bigger subscriber base.
Get a blog and talk about new features, and goodness.

I recommend being apart of this network. If you need an invite, email me at kunal.r.anand@gmail.com and I'll throw one your way.

TiVo for Newspapers

8/03/2005

I'm upset with Bloglines (slow, dropping content, etc.), so I decided to think about the current model for feeds. I asked myself several questions:

Why don't we only aggregate popular feeds like Slashdot and Kottke?
It's simple. Not everything important floats to the top.

Why don't we only aggregate pre-canned search feeds from Technorati?
Those searches are context specific. If you solely choose this route, you blind yourself from a lot of knowledge out there.

Why don't we use collaborative filtering to determine what to read?
I want control over what I read, not an engine. You might loose serendipity.

So at what point does feed reading evolve? I'm not saying we've reached a peak. On the contrary, we're at a low (feed reading is clunky and discovering new content can be as difficult as browsing for hours). I think we need to re-innovate this space at the service layer. Maybe we need better readers that will:

1. Remove superfluous content
2. Dig up things in the long tail
3. Use backlinks to introduce "interesting" content
4. Figure out a way to tie all these tags together
5. Help me put together annotations easily

I view the future feed reader to be like TiVo for newspapers. Archive this content, show me stuff like this, and make it intuitive to talk or research about. Icing on the cake - it has to be easy.