[jdev] Re: Re: The State of Our Code-bases

[Rachel Blackman]

> A second ago we were talking about one implementation of a protocol, 
> but now
> we're talking about a protocol in a sea of closed protocols.  I'm sure 
> there
> were plenty of closed electronic mail implementations when Sendmail 
> reared
> its head too, and their existence is largely irrelevant here.

I wasn't the one bringing Sendmail into the mix of analogies; you did 
by asking how it got support and how it got books about it. :)

That said, no, Sendmail reared its head in the early 1980's when the 
old networks transitioned to TCP/IP, and was one of the first available 
implementations of this new 'SMTP' protocol which was the new accepted 
standard to connect all these varied networks up together.  Prior to 
that, Sendmail had in fact been 'delivermail' in the 1970's, one of the 
ad-hoc systems of ARPANET mail delivery, back when mail headers were 
sort of an informal agreement rather than a standard, and mail was 
delivered via file transfer over ARPANET NCP. :)

So it really was, honest to god, the first game in town.  It did not 
win massive support by being the best at what it did, it won massive 
support by being the ONLY thing which did what it did, namely, SMTP.  
If you wanted to move to TCP/IP and be part of this new combined 
network, you pretty much had to go Sendmail.

Which was my point, in answer to your 'how did Sendmail win such 
support and get all these books written about it' question.  Heck, it 
was included with the Berkeley Standard Distribution, even, so was 
readily available and preinstalled for most people.

Which, actually, is an interesting thought.  Maybe if the popular 
UNIXes had preinstalled Jabber servers which got auto-configured during 
system install like they do with mailserver and webserver setups, 
Jabber would be more widely adopted by folks.

> Why?  Well, if you think it's so hard to set up Jabberd, try setting 
> up an MSN
> server.  What's that, you can't?  Why not?  Too hard?  Maybe the best 
> selling
> point of Jabber is "that you can."

I guess the question is if we're talking about servers -- in which case 
Jabber is more or less the only game in town, though I've seen some 
attempts to make open, free implementations of the other networks -- or 
if we're talking about Jabber as a whole, in which case I thought it 
wasn't just sysadmins we were trying to attract, but end-users.

I may be making the problem scope too large in my argument, I admit. :)

> Jabberd is less than ideal, but only because whatever distributions are
> shipping it, must be shipping it with a configuration which simply 
> doesn't
> work.  If it worked, there wouldn't be so much criticism around here.

*laugh*  I should read further down in what I'm replying to before 
making suggestions like above. :)

Yes, +1 to this.

> So why don't we direct the attention _there_?  If all Jabber packages 
> on all
> distros worked out of the box, would there still be a good excuse not 
> to run
> it?

Are you talking 'a good excuse not to run it' from a server side, or a 
client side?  Not from a server side, no, there wouldn't be.  But you 
still need something now to lure the end-users off of AIM and MSN and 
Yahoo and onto Jabber.

-- 
Rachel 'Sparks' Blackman -- sysadmin, developer, mad scientist
"If it is not broken, give me five minutes to redesign it!"
-------------- next part --------------
A non-text attachment was scrubbed...
Name: PGP.sig
Type: application/pgp-signature
Size: 186 bytes
Desc: This is a digitally signed message part
URL: <https://www.jabber.org/jdev/attachments/20040902/285df20c/attachment-0002.pgp>