Re: MailPush confirmation of off-line e-mail notification

Rohit Khare (
Fri, 06 Feb 1998 07:49:21 -0800

> You must have got these too, yes? Can you FoRK an opinion?
> -- Ernie P.
> Begin forwarded message:
> From: "MailPush Support" <>
> Date: Wed Feb 04 19:31:08 1998
> To: Ernest Prabhakar<>
> Subject: MailPush confirmation of off-line e-mail notification
> [] Was notified of your e-mail message [Rhapsody =
> Product Marketing Manager: the truth comes out]
> By: Phone Call At: Wed Feb 04 19:31:08 1998
> -----------------------------------------------------------------
> Would you like to be notified off-line of new e-mail message ?
> Sign up for FREE MailPush, which can do almost anything you like.
> You'll never have to check for new mail in your mail box again.
> <attachment missing>
> -----------------------------------------------------------------

No, I've never heard of Quite separately from whatever void in
Doron's social life might be driving him to be actively notified *any* *time*
*any* *FoRKpost* *is* *ever* *emitted*, I can imagine this service might sill
be useful. Let's tool over and have a look:

: Would you like to be notified off-line of new e-mail message ?
: Sign up for your FREE M@ilPush, which can do
: almost anything you like. You'll never have to check for
: new mail in your mail box again.

Now, I've seen a lot of internet come-ons and scams in my time, but I've never
seen anything that

1) promised to do "anything you like", which begins with cleaning my desk and
taking dictation and goes downhill from there, ending up somewhere near
"reinstall a fresh copy of Win95 every morning".

2) promised to filter my incoming mail to a manageable zero.

[this page, by the way, deserves some kind of note as home of the first hebrew
banner ad gif I've ever seen. It has a nice US-ASCII smiley in the text, to
boot. However, why isn't the Hebrew smiley (-: ? Ron? Oh, when we need Ron the
most... (actually, it's a fully bilingual site: english or herbrew)]

It works only with the POP service of certain ISPs (about two dozen right
now). Remember, right on the FIRST WELCOME SCREEN, you're supposed to blindly
hand over your POP account and password to this anonymous service, probably
run by the Mossad.

I don't know what networks support this, but it must support a 'fast dial bip'
to work with:

: Signaling box
: Plug it into the telephone socket - The on-line green led will blink. Upon
: receiving a new message, the red Led will blink. Upon receiving a message in
: which in the subject contains the word "chat", the yellow Led will blink.
: Pushing the reset button will turn off the red or yellow Leds , and will
: blink back the green led.

I can only imagine putting this into the loop of Dan Kohn's
cellphone-back-to-Seattle-back-to IP-back to SF-back to NJ-back to roaming
ISP-back to VocalTec laptop-to Dave Crocker scenario.

Or you can use the "Mailpush softwar". Which works with the Win95 dialer to,
um, dial up to the server every so often and check if you have new mail. No,
you are NOT allowed to ask me how this might be an improvement.

It will, however, offer marginal utility by paging and/or allowing voicemail

It's a free service, which means I think they're also gonna be pagig with a
lot of advertisements. But why stop there? I always wondered if advertising
companies would want to do nano-sponsorship: free calls to Aunt Greta in
Minnesota if we both agree to listen to 30-secong Proceter and Gamble radio
ads at random intervals...

And that's it. It's a two-page site. Nothing more to deconstruct.


PS. Today in class, the Geneva-to-Geneva phone call example came in handy to
give lie to the meme "without computers, 2/3 of the population would have to
be telephone operators" -- one of the other revolutions that keeps the size of
the telecom sector in check (at about 3%!) is that *users* are now operators:
ever enter 25 digits to reach your vmail from abroad? That's work. Credit to
John L. King for the example.

PPS. In another class, we discussed dynamic runtime optimization: just-in-time
compiliation, profiling, and re-compilation to acheive even higher
efficiencies. One downside is that the machine code would never be stable: new
chips with extra caches, different clock speeds, etc, might trigger the
adaptive system into writing completely different code, potentially exposing
latent bugs. I pointed out that this software on a Caltech asynchronous
microprocessor could lead to self-modifying code that reacts to the outside
temperature! Async procs react to cooling and level of drive voltage. So with
a lot of cache hits, the compiler might turn up the power to the cache to
drive it faster. Or, you shoot the satellit into outerspace, and the code
morphs because the functional units just got ridulously faster in the vacuum
of space...