v0.993 of mod-pubsub re-released.

I noticed a corrupted file in the 0.993 tarball so I re-released the tarball
(with datestamp 10-31) for download at:

   http://sourceforge.net/project/showfiles.php?group_id=66293&release_id=192963

For those of you who love trivia, we're very close to our 500th mod_pubsub
download and our 25,000th web page view.  Congrats, folks!

Also, Joyce made a lot of really good additions to the Wiki today:

   http://www.mod-pubsub.org/moinmoin/RecentChanges

Including adding Kragen's post:

   http://www.mod-pubsub.org/moinmoin/PythonServerWithOnePerlClientAndManyEvents

And new pages for other considerations that will evolve over time:

   http://www.mod-pubsub.org/moinmoin/PerformanceConsiderations
   http://www.mod-pubsub.org/moinmoin/DesignConsiderations
   http://www.mod-pubsub.org/moinmoin/TheOneTrueThinClient

Also, I really like Joyce's recent post to the blog, "Is simplicity sexy?":

   http://www.mod-pubsub.org/blog/archives/344_Is_simplicity_sexy.html

Included below for those (like me) too lazy to click the link.  :)

mod-pubsub.org is now regularly maintaining over 20 simultaneous connections,
   Adam



A bunch of us recently enjoyed a very stimulating (unfortunately offline, or at least I can't find it in the mailing list archive) discussion, led by Robert O. of New Zealand, about whether there is something off-puttingly... too simple about mod-pubsub. That is an amazingly deep and perceptive question with a lot of ramifications, and one that we still don't have an adequate public answer for.

You would think that something both powerful and easy to use would be immediately perceived by all as a good thing. But... well, let me put it this way. If you've ever watched Rohit give a talk, you've doubtless noticed the audience dividing into those who look at him as if he has the answer to all their questions about the Internet, and those who are more or less squirming in their seats at the utter slobbiness of it all (well, and the third group -- those who just have no idea what's he's going on about but have been literally stupefied by the fluency of his discourse). Invariably the skeptics zone right in on what they imagine to be the weakness of mod-pubsub: lack of guarantees. It's quite amusing to watch the total self-confidence with which Rohit blows off such concerns, leaving his critics spluttering "But, but, but" in his wake. I suspect for mod-pubsub developers, what looks to others like poor engineering values and lack of foresight actually appears to be the genius design bit.

Of late I can't help coming to the conclusion that a whole bunch of good programmers are having an awfully hard time coming to grips with the Internet itself -- as opposed to nice neat tightly-coupled object-oriented programming on a private LAN. On the Net, you don't really know who you're dealing with, where they're coming from, what they want, or how they plan to use the resources you give them. Many engineers are loath to begin designing and building anything until questions like these can be answered completely; and if they're forced to do so, they can drive themselves crazy and burn a lot of resources trying to cover every conceivable ambiguity. Instead of building within the Internet paradigm, they fight everything good about the environment -- particularly the ability to move quickly to share resources with others.

So yes, I think in some ways the apparent simplicity of mod-pubsub works against us rather than for us. A lot of the people who are currently interested in what we do seem to be architecturally-minded chaps, and for a lot of people architecture means building Taj Mahals instead of, say, Japanese detached palaces. Perhaps those who love marble mausoleums won't be able to find anything beautiful in a little heap of sticks and paper and bamboo. But the whole point of the detached palace might be how organically and artfully it fits into its environment, how quickly it can be put up with local products, and how a structure made of the most fragile materials can exist for centuries. It's a different aesthetic, and perhaps one that doesn't come naturally to the (over)engineering mentality -- but it's actually a pretty complete statement on Rohit and Adam's part of what it means to build software for the Internet, or possibly to build the Internet itself. I don't think it's too much to ask the world to accept or reject us on those terms, instead of on the entirely specious grounds that we didn't build something else. 


-------------------------------------------------------
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive?  Does it
help you create better code?   SHARE THE LOVE, and help us help
YOU!  Click Here: http://sourceforge.net/donate/