Disqus - Latest Comments for chinmay

Re: the "FUNDA"

chinmay — Sat, 08 Aug 2009 14:18:47 -0000

This is beautiful!

Re: SQL Databases Don't Scale

chinmay — Mon, 06 Jul 2009 23:04:59 -0000

I don't think this is an SQL issue really. This is a problem with any system that allows global queries and (therefore) may require global updates.

Also 20-years of failing says nothing really. It took longer to get paging working; you could say that "programs don't scale memory-footprint" before they did. Lastly, any particular way you want to turn the question around?

Re: Heroku | Why Instant Deployment Matters

chinmay — Mon, 23 Feb 2009 17:58:22 -0000

Um... I'm not sure there is any additional friction if you know that the
deploy process will occur without fail and will take no more than 10
minutes. My personal experience with deploying tells me that any number of
things could go wrong (especially for a first-timer) and fixing those could
take hours or days; depending on distance-to-nearest-nerd.
Don't get me wrong: quick deploys would be nice to have. I still maintain
though that having predictable, FAIL-SAFE deploys is what matters more. The
time it takes is a secondary issue: shorter the better, of course.

Chinmay

Re: Heroku | Why Instant Deployment Matters

chinmay — Mon, 23 Feb 2009 17:26:16 -0000

I don't want to distract from the main argument; which is faster deployment is of vital importance. However, I still think that the reasons and figures above are misleading: True: deploy time is increasing as a percent of total product development cycle; but that isn't important.

The product development cycle itself is becoming tighter; so any improvements in the deploy-phase will be a very small percentage of the total product cycle: it takes much longer to actually figure out the rest of the business.

Think about it this way: in 1996; if you made a dramatic improvement in deploy-time, you could save two weeks. In 2008; if you make a dramatic improvement you save no more than 24 hours.

Is that worth the trouble?

Re: Binding NOTA & Plurality.

chinmay — Fri, 05 Sep 2008 11:58:18 -0000

I wanted to do such a similar numeric analysis on my blog, but thought it would inflame some already overexcited tempers.

NOTA is really a silly edge-case. No one gets disqualified for campaigning for NOTA, no campaigning laws apply. No anti-campaign laws apply either. the option wins not on its merit, but on the lack thereof in other candidates.

With the situation so dissimilar; why can the option "win" with a simple majority? Would we not want a higher quorum? A higher than 2/3rd majority?

It's really an unfair system; escaping analysis merely due to its rarity.

I think the US President election system is better: don't just say None of the Above. Recommend a candidate instead. Of course, there're people who write in "Mickey Mouse" ;)

Re: the "FUNDA"

chinmay — Mon, 25 Aug 2008 03:11:07 -0000

No one dare question the might of Microsoft! ;)

The car would require you to upgrade the road every couple of years to fully experience the Vista experience, though. ;)

Re: the "FUNDA"

chinmay — Mon, 25 Aug 2008 03:09:07 -0000

He did, man. Don't worry about it. :)

Re: Abhilash Ravishankar - Thirty30

chinmay — Wed, 30 Jul 2008 08:37:15 -0000

Hmm.... This should be a checklist.
Check things off as you go. Or maybe, you can just strike out things already done? Do it now... I'm curious to see how far you've gotten :)

Re: Abhilash Ravishankar

chinmay — Sat, 19 Jul 2008 14:28:48 -0000

Man.... this is one of those things that strike some deep deep chord. Can't agree more.

Re: Abhilash Ravishankar

chinmay — Wed, 16 Jul 2008 09:24:27 -0000

Look what you oil-sucking globe warming companies do to God's little planet! ;)

Btw, have you tried the green thing like Red Bull. It's much better (though it costs some 3.5 euro a can :( )