FF 3.1b3

============================================
RESULTS (means and 95% confidence intervals)
--------------------------------------------
Total:                 1476.2ms +/- 3.8%
--------------------------------------------

  3d:                   221.6ms +/- 11.5%
    cube:                74.8ms +/- 8.7%
    morph:               54.6ms +/- 31.4%
    raytrace:            92.2ms +/- 4.5%

  access:               215.8ms +/- 21.6%
    binary-trees:        48.6ms +/- 3.9%
    fannkuch:            60.8ms +/- 8.7%
    nbody:               27.0ms +/- 7.3%
    nsieve:              79.4ms +/- 59.2%

  bitops:                46.4ms +/- 10.1%
    3bit-bits-in-byte:    1.6ms +/- 42.6%
    bits-in-byte:        11.6ms +/- 50.6%
    bitwise-and:          2.4ms +/- 28.4%
    nsieve-bits:         30.8ms +/- 7.8%

  controlflow:           35.6ms +/- 3.1%
    recursive:           35.6ms +/- 3.1%

  crypto:               154.2ms +/- 4.9%
    aes:                114.8ms +/- 7.9%
    md5:                 27.0ms +/- 11.7%
    sha1:                12.4ms +/- 22.0%

  date:                 216.8ms +/- 3.4%
    format-tofte:        94.2ms +/- 6.1%
    format-xparb:       122.6ms +/- 3.1%

  math:                  43.6ms +/- 5.9%
    cordic:              20.0ms +/- 4.4%
    partial-sums:        14.4ms +/- 4.7%
    spectral-norm:        9.2ms +/- 30.8%

  regexp:                71.2ms +/- 4.0%
    dna:                 71.2ms +/- 4.0%

  string:               471.0ms +/- 1.3%
    base64:              26.8ms +/- 24.6%
    fasta:               97.4ms +/- 4.3%
    tagcloud:           114.0ms +/- 4.0%
    unpack-code:        176.6ms +/- 0.8%
    validate-input:      56.2ms +/- 2.4%

FF 3.5b4

============================================
RESULTS (means and 95% confidence intervals)
--------------------------------------------
Total:                 1209.4ms +/- 1.3%
--------------------------------------------

  3d:                   181.0ms +/- 3.1%
    cube:                55.2ms +/- 6.2%
    morph:               33.6ms +/- 21.7%
    raytrace:            92.2ms +/- 5.9%

  access:               154.0ms +/- 2.0%
    binary-trees:        50.4ms +/- 7.5%
    fannkuch:            61.6ms +/- 5.6%
    nbody:               28.4ms +/- 6.6%
    nsieve:              13.6ms +/- 5.0%

  bitops:                40.4ms +/- 3.5%
    3bit-bits-in-byte:    1.6ms +/- 42.6%
    bits-in-byte:         9.2ms +/- 6.0%
    bitwise-and:          3.2ms +/- 17.4%
    nsieve-bits:         26.4ms +/- 4.2%

  controlflow:           37.2ms +/- 4.4%
    recursive:           37.2ms +/- 4.4%

  crypto:                78.0ms +/- 6.3%
    aes:                 46.0ms +/- 8.1%
    md5:                 23.6ms +/- 20.9%
    sha1:                 8.4ms +/- 22.4%

  date:                 182.0ms +/- 8.0%
    format-tofte:       100.0ms +/- 12.7%
    format-xparb:        82.0ms +/- 4.7%

  math:                  41.0ms +/- 4.8%
    cordic:              19.6ms +/- 5.7%
    partial-sums:        13.8ms +/- 7.5%
    spectral-norm:        7.6ms +/- 14.6%

  regexp:                78.0ms +/- 9.0%
    dna:                 78.0ms +/- 9.0%

  string:               417.8ms +/- 1.1%
    base64:              23.2ms +/- 11.6%
    fasta:               94.4ms +/- 1.8%
    tagcloud:           115.0ms +/- 3.1%
    unpack-code:        143.2ms +/- 2.4%
    validate-input:      42.0ms +/- 4.7%

So there, its faster by almost 270 points, which is almost 18%. So far the next release seems to be very good indeed.

Yahoo bought the company near the peak of the dotcom bubble for more than $3 billion, which, along with the rise of alternative services, quickly spelled the end of GeoCities’ prominence. Today, it appears that the end of GeoCities is being made official, as Yahoo has closed the service to new accounts and posted an FAQ with some details as to how the shutdown will go.

Wikipedia page here

Adios.

But this FUP puts an end to all the goodwill they created. Not only is it outright cheating, it also defeats the very point of unlimited connections. And in  a country like India where internet is slowly expanding, it is a major roadblock.

To view the complete details of the policy, visit this page

To view the draft  of an online petition, created by members of http://broadbandforum.in, visit this page

Sign the online petition against the policy here

The question is - how true is the cost projection? I'm a big fan of such initiatives, but common sense prevails and its pretty damn clear that this is unrealistic. MIT's OLPC was projected at about $100, but finally ended up being priced upto $190. In short, I believe the Indian government is trying to pull off another cheap publicity stunt. Unless the exchange rate of the US Dollar goes upto Rs.1ooo, I don't see this happening.

But then again, the initiative will be heavily subsidised by the government to make it as cheap as possible. But will they take upto 90% of the price remains to be seen. I'm sure in the current financial position, they simply cannot and will not do so.

Until further details - this simply isn't going to happen.

UPDATE: Yea, its a total flop show. The device is actually some sort of external storage, not a laptop. More here