-=> Sean Dennis wrote to Kevin Lamonte <=-

 SD> Hello, Kevin.

 SD> Why reinvent the wheel?  BBS technology has been working just fine,
 SD> without the Internet's help, for nearly 30 years.

I agree!  BBS technology has done very well - so much so that some
areas remain unsurpassed by the modern world.  But I'm not urging
anyone to create a new competitor within the BBS world, or even
necessarily alter a specific BBS package to speak some new wire
protocol.  Rather I'd like to bring some of BBSes other strengths back
to the Internet world.

In terms of marketing, most people won't be familiar with what is from
the BBS world.  Today's Reddit, Facebook, and Twitter are all
descendents in spirit of BBSes yet we don't call them that.  The BBS
label isn't needed, but how today's sysops use them is.  Sysops can
make their wall look just like they want it; today's Facebook users
have to rely on Greasemonkey scripts to do the same.  Sysops can
choose which messaging networks to be on; today's Reddit users have to
suffer trolls from everywhere.

If you check out the darknets, you tend to see good/secure file
transfers, maybe an OK chat application, and for most other content
you see a WWW site with HTML.  HTML that will happily tell your
browser to go fetch a resource from *anywhere* and try to display it.
The same browser that you need to access the public Internet anyway so
you can't really harden it without breaking everything else you do
online.

Why is (X)HTML the *only* way to broadcast human-readable content
across the modern Internet?  Why not use HTML between your
keyboard/monitor and your local server (where it's fast and pretty and
can use all the Web 2.0 Ajaxy goodness you like), but produce that
(X)HTML from your own local cache that was populated using some very
smart data transfer protocols over sophisticated darknet technology
that never exposed your browser to this week's 0-day exploit?

Storage these days is *cheap*.  The entire English-language Wikipedia
fits on less than $5 of hard drive space, and that's UN-compressed.
Everything you are ever going to transmit and receive on social
networking sites in your lifetime will fit inside $10 of hard drive
space.  Completely saturating a 1.5Mbps DSL line costs $1.30 per day
in hard drive storage.

 SD> I skimmed over that and quite frankly, too much work for too little
 SD> gain. There's better things to do with one's time rather than reinvent
 SD> the wheel. :)

The neat thing is that many of the pieces that are hard to do right
and/or very time consuming -- basically determing the right bytes to
send on the wire -- are already solved by existing libraries.  Pick
any modern language you want and you've got strong crypto, networking,
multi-threading, Web 2.0, etc.  You also have multiple darknet
technologies to choose from: I2P, FreeNet, TOR, ...

The parts I think are missing are BBS things like automatically
scheduling data transfers for convenient times, multiplexing different
kinds of data streams over the same byte channel, smoothly recovering
from interrupted transfers, and knowing how to treat different kinds
of data in transit rather than wait for your browser to hand it off to
third-party code based on its MIME type.

It's 2010, and my computer was safer online using BBSes 22 years ago
despite thousands of DOS viruses in the wild than my Mac is today on
the WWW.  It was also a lot more fun. :)

... MultiMail, the new multi-platform, multi-format offline reader!
--- MultiMail/Linux v0.49
--- SBBSecho 2.12-Win32