The ideas about Spindl3top were born on the pages of Advogato, but they solidified and grew up on Kuro5hin. Quickly. It's what drew me to the Kuro5hin community; the fact that people could take an idea that I had milled about in my head, make helpful comments on it, and give publicity to it by posting so that others can learn from it and have a discussion.
For me, it was a fantastic discussion which assisted me in planning... just as I hope this article will produce some interesting perspectives that I can learn from.
Soon thereafter, it was picked up by Slashdot and my network connection and server were pretty much useless for 24 hours as over 200,000 hits logged up. Inoshiro helped out by posting to Slashdot and redirecting the traffic over to Kuro5hin. I had hoped that the Slashdot article might produce interesting ideas; instead, RMS' affiliation with goat.cx and the communist party were discussed.
A day later, press approached me, users called wanting to buy something... anything... they left messages on the Spindl3top voicemail system. One person was so incredibly frustrated that all he could do was shout. "Dammit! How are we supposed to buy anything if you don't have anything for sale!"
Keep in mind: this was only an idea.
I had unknowingly walked into the frenzy of the pre-collapse "dot-com world", where speculation about the "Next Biggest Thing" was an obsession. Yipes. I was a little angry, actually, because if all of this attention could be paid to me and I didn't have anything, it seemed to say something about the hype and propaganda from other projects or enterprises.
I didn't know what to tell the press, so I did my best fielding their questions. You can see the results in this Wired article or in the Boston Globe. I e-mailed Bob Young of RedHat to tell him thanks; I was misquoted as saying they were the "bad guys" in one article and here was Bob Young, President and CEO of RedHat, vouching for me in Wired.
I formed an ad-hoc mailing list to deal with the inflow of mail, eventually switching to PHP-Nuke. It was a good decision, but I didn't have anyone to assist in sysadmin'ing. It was too much of a response, too soon.
So, I did what anyone else would do and began to put my money where my mouth was, one day at a time. I had been planning to wind things down by the end of the year and get ready to visit my family...Instead, I was busy trying to figure out how the heck we could implement this, work out the kinks, and get things off the ground. People involved were extremely patient with me, and I thank them for this, because I was completely overloaded.
I decided to build some testing Blackbirds and finalized the source, which was one of my ideas -- make some sort of "shopping list" for people. I personally assembled testing Blackbirds which were bought and tested by Misha Dynin and Carl Soderstrom. I named the first model Blackbird-DYM as thanks for Misha's help (DYM are his initials, spelled backwards), as he gave a lot of assistance.
After I shipped these final Blackbirds off, I took a much-needed vacation. It wasn't without some degree of struggle, however. My ISP, Flashcom.com, went under... leaving me with two options: UUNet or <snip> <snip> (otherwise known as, "your cables will be cut and you'll have to wait another 6 weeks for another DSL provider").
The day after Christmas, they switched over Spindl3top's lines to UUNet while no one was in the office.... there are times when you could just *kill* to be at the console. UUNet claimed I wasn't in their databases and that I wasn't even a customer. A week later, a tech calls from UUNet, giving me the new IP address.
When I arrived back home, the real work began. Josh and I had to write an entire backend for PHP-Nuke that would allow things like item shopping, shopping-carts, and realtime FedEx shipping calculation. I couldn't have done it without free software, since I was able to access the PHP-Nuke code and change certain things around, interfacing it with the shop program.
This past week, we completed (and are using) the code. It took us a month to write and test hundreds of lines of code before placing it "live" and having people entrust their information with it. This was the final and most complex hurdle to overcome. Currently, we have stuff on order and have put what we have right now into the inventory catalog.
We've come so far, and yet there are still many avenues that I would like to explore aside from cooperative buying, including forming something of a "services" guild (or informal union) of people who have certain specialties that can contribute to an integrated whole. My example is Beowulf clustering; if someone wants to cluster Blackbirds (which are great for clustering), I'm not the greatest at this. The guys at MIT's DeBUG (Debian Beowulf User Group) love to do this and Spindl3top can pass the job over to them. It's the same thing with Synecdoche, a cooperative, non-profit webhost and ISP -- if someone wants to colocate their BB, we pass it over to them and they are able to work with the user to support it because they are a part of the project and know how to work with the BB and Debian. Certain people are great coders, experts in security, etc.
Another idea I have is to create an open certification program which is very low-cost and affordable. Having seen the expensive nature of RHCE and how employers salivate over certs, I think this would be a Good Thing if we can get together as hackers and come up with our own, freely published standards that we expect people we work with to have. My preference, of course, is that we might certify based on Debian GNU/Linux, but it probably should be distro neutral.
My central goal is to keep Spindl3top end-user focused, fair, and of a continually progressing nature. As in any open-source project, of course, it's not utopic and there are bugs that have to be worked out. Some ideas suck and just won't work no matter what. It is and I hope always will be an experimental project supported by hackers and advocates of free software and the open flow of ideas.
I want to thank the Kuro5hin community once again for making something positive like Spindl3top a reality. The "K5 way" does work and, if nothing else, the existence of Spindl3top is living proof of this. I can only hope that I can repay your generosity by continuing, day by day, to make these ideas into a reality while still keeping them low-profile, uncluttered, and un-hypeadelic.
& Fellow K5'er