Collapse-able dollar?

For last few months people have been getting on the soap box and pontificating about a new global currency (a gold backed one, at that which the dollar stopped @ 1972). All of a sudden. Out of seemingly nowhere, every economist was talking about a dollar-weakened future (well, not everyone).

This struck me as odd. What exactly was fueling their worries? How was this tied into the recession? And why a lot of learned business gurus were relating about how the US would take more than 10 years to get out of this recession/mess.

Mess? Yup. Mess.

So, I did some digging and here is what I found.

The (US) Mint has been busy. Real busy.
Its been minting. More than normal. Way more. To put things in perspective.
In the last 1 year, it has expanded the monetary base from the, then, 900-ish billion to approx. 1,700-billion-ish (and its still printing (hint: look at the far right) ...). That's right. In less than 1 year, Bernanke has tried to attack the recession by:
1. Printing his way out, and
2. Make firms and banks take money (remember TARP) and refuse to let them pay it back (a nice way to say, I own you!)

So, where is the money? When people are complaining about unemployment and housing slumps, etc..

Its with the banks. The banks are storing those reserves at 0% interest rates with the Fed in, what are known as excess reserves of Depository Institutions. Fancy ways of saying the money is being spread between the fed, banks, and the TARP-stained companies.

The Fed through the IOER rate (interest on excess reserves), will try to control monetary policy. Make sure that reserves deposited by the banks don't make it into the system. Buy time, until a solution can be found.

That is why Bernanke has gone on record to say that the interest rates are likely to be in the 0-0.25% range for "some time".

Just imagine the inflationary pressure, if a tenth of those funds make their way into the market. The question is how long before the dam breaks?

The CBO has projected

... that the federal debt as a share of the economy will double over the next decade, from about 41% last year to 82% by 2019.

(this debt is the by-product of the various stimulus packages)

This begs the question, why the analysts on CNBC go around talking about US coming out of the recession by the end of this year?

Do they not realize what is really happening or are they more PR and analysts less?
Is the dollar really safe or is it headed the way of the Zimbabwean dollar.

Web sense

Ever wondered who exactly uses the web and what are they doing?
Here's the result from a few places - based on region and apps using latest(2009) data (where available).

S.No.	Country	NS	IWS	Remarks
1	China	253m	298m	(density: 19%) [2006 figures: 103m]
2	US	220m	227m	(density: 72.5%) [2006 figures: 209m]
3	Japan	94m	94m	(density: 74%) [2006 figures: 86m]
4	Brazil	67.5m	67.5m	(density: 26% / 22.4%) [2006 figures: 13m]
5	India	60m	81m	(density: 4% / 7.1%) [2006 figures: 40m]
6	Germany	52.5m	55.2m	(density: 64% / 67%) [2006 figures: 50.6m]
7	UK	43.2m	43.2m	(density: 69% / 70.9%) [2006 figures: 37.6m]

They estimate that in 3 years - China will outnumber US 3:1 and along with India - will become the big 3. Asian piece of the pie by 2012 - 490m

Top languages: Mandarin, English, Hindi, Portugese, Russian

Top apps:
Web - 45%
Email - 6% (78% spam)
P2P - 25%
Streaming - 8%
VOIP -3%

Top countries by penetration, according to Swivel are:
Iceland, Sweden, Netherlands, Australia, New Zealand, South Korea, Canada, Luxemborg, & US

(Apparently, when I publish the table it puts in a huge white space header. Its not there 'cuz of my code! Silly blogger!)

Building a Firefox extension

Knowledge info:
-- some programming knowledge -- check
-- javascript -- so-so
-- understand XML files -- check
-- curiousity -- check

Useful tutorials/sites to help get you started.
@MozillaDev, @mozblog
@roachfiend
@lifehacker

Env setup:
some dev extensions

Just noting stuff for now.  Should have something like a beta out, hopefully, in a month's time.

Perks of being close to a VC

Apparently, 50% of Founders/CEO get fired in startups within the first year!

Barnaby Federer of the WSJ, take on this is simple and clear.

“If you ask a VC what value they add, and you get them after a few drinks, they’ll say, ‘We replace the CEO’ “,
he said. And that, he indicated, does not vary with the economic climate.

Here's how it plays out (source):

1. Usually happens over a friendly lunch.
2. VC mentions about a new person who can benefit the company greatly through his connections or industry experience.
3. (VC) Explains this person is not available to serve on the management team, he could probably help out as a director.
4. Would mean enhancing the board but this "star" guy deserves to be on it.
5. Founder, who wants to please his VC - reluctantly agrees
6. The new guy ends up becoming the leverage the VC needs to restructure the board
7. Enter the New CEO

Recommendations:
1.  Don't recommend someone for a board set until you are satisfied that they can make an actual contribution
2.  Have the guy work as a consultant until he makes a tangible contribution

Why do VCs do this?
- They generally prefer "their" people at the helm.  Nothing personal.
- Also, people who are generally "ruthless" (eg. lacking in moral fibre) generally end up being closest to VCs the longest.

Airtel site -- down

Chetan just highlighted this:   Airtel website is down.

Internal Server Error

The server encountered an internal error or misconfiguration and was unable to complete your request.

Please contact the server administrator, you@your.address and inform them of the time the error occurred, and anything you might have done that may have caused the error.

More information about this error may be available in the server error log.

---

IBM_HTTP_Server Server at www.airtel.in Port 80

Amusing.

failover(rackspace) == exception

Apparently, Rackspace, the guys who are supposed to be fanatical about support, didn't have any redundancy in place after power outage in one of their generators at their primary hosting location outside Grapevine, Texas.

They have apologized and are issuing service credits to those affected.. however, as one guy succintly puts it:

"... your lack of redundancy for your own internal operations is highly concerning ..."

Update:  Went down again for bit
(but according to twitter, the bit was more like 1 hr.

3G Health in the US

First some basics:

Am not going to start from the telephone... (but here is its history, for those interested)
2G networks:
-- basically meant for voice traffic - usually TDMA or GSM type of networks
2.5G:
-- GSM: typically enhanced for data via GPRS - which allows -- generally true throughput  ~ 54 kbps (irrespective of reports of 125+kpbs) - at 125 kpbs - 1 mb file would take 90 seconds to download
-- GSM: EDGE, the so-called "2.75G" does offer upto 384kbps, but thats asymmetric, with 100 kpbs on the uplink side.
-- CDMA:  typically around 50kpbs - with a burstable option upto 144kbps.
3G:
-- built for 1's and 0's -- but not a full IP stack - something like at the ethernet level - to adopt a networking stack paradigm
-- GSM: avg speed 800kbps -- 14 sec download for 1 mb file -- provided it supports HSDPA/USUPA - otherwise, you're looking at 384k sysmmetric.
-- CDMA: typically 155kbps at 1xEV-DO -- going upto 1.8mbps with Rev A
4G (as per LTE):
3G + QoS services + full multimedia support + full-IP-Stack support (AIPN) built-in + support for other interesting things like femtos, etc.
-- GSM:  expect twice the speed of 3G (~ 1.5 mbps -- should take 7 seconds to download a 1 mb file) - even though people claim upto 50 <b>m</b>bps.
-- CDMA:  ~ 27mbps
(for the sake of simplicity - am not getting into 5G and 6G, just yet)

A 3G pulse test was conducted in the US across the major telcos.  Report here

Summary:
Verizon ~ avg 951 kpbs d/l - with good reliability and consistent speed in 89.8 test cases
Sprint 3G ~ avg 808 kpbs d/l - with 90.5 % reliability in test cases
AT&T ~ 812kbps d/l - with 660 kbps upload and 68% reliability - although they claim to be the fastest 3G provider in the US

-- Overall - signal strength indicators did not positively correlate with quality of connection
-- Verizon is generally fast and consistent.
-- Tap Sprint on the West coast.