Ok, I broke my promise to announce the latest-n-greatest TurtolCMS here. I did do the release (some while ago, now), but didn't make an announcement. Color me lazy, or perhaps just overly busy. And the book review will be forthcoming, some day. Just last night, my cousin Rob asked me something along the lines of "have you ever been thinking about a blog entry for so long that you just couldn't get it done?" He was talking about his own blog, mind, but hit the nail on the head for me. So here's me getting it done, finally. I alluded, in my previous entry, to another blog I would reference for a starting point. That reference is below, but first, a little history. A little historyTurtol, my much lamented previous endeavor, was originally started to be a hosting company, with 37signals-like aspirations to build some cool web-based applications. Prior to Turtol, I had Yet Another Startup, and after I exited that, I took about 18 months off to be a Dad and work on my 50-some-year-old house and whatnot. Some of my previous clients expressed an interest in my continuing to support their web sites (and so forth), so I put together a tiny little hosting operation, just to service them. It gave me something to do to keep my hand in technology and was just self-sustaining; it made me no money, but I didn't care, because it wasn't supposed to. After a while, some friend or other heard about my hosting, asked if he could have an account, and of course I put that together. Then, of course, he asked for some changes, and I did that. Then another guy came along, and pretty soon I needed to hack together some software to manage it all (web-based, of course), and keep me from actually running a hosting company so I could continue to play with my kids and generally be lazy. Then Mike came calling, having heard about what I had put together, and needing a site for his then-current employer. He asked what I'd charge. I didn't know, because I didn't charge, except for those few who had sparked the whole she-bang in the first place. So I put together pricing and Mike's company cut me a check and all was good. But then Mike said something like "wow, that's like the best control panel I've ever used" and went on to tell me how bad most others were and how his pal Libor (who I hadn't, to that point, met) was always hating having to deal with them and how none of his designer friends even COULD use them. I kinda knew that, since I'd looked at using them before I decided to just build my own. Swell, I said. Thanks, I said. And that was that. Until Mike asked "what would it take for me to resell your services?" The idea, originally, was that I'd run the servers (and develop the control panel further) and he'd sell the service. Several scenarios were suggested. Then Mike had an epiphany about the pricing structure and a few lunches later, Turtol was born. The original business plan was we'd spend some time making the control panel totally kick ass and Libor (who had come on board by then) would help us bootstrap by bringing in ten or so of his freelance designer friends who each had at least a couple of current and upcoming sites needing hosting. But what would become the critical piece of the puzzle was largely an accident born of my frustration with some of the work I was doing.
TurtolCMS v0.6 is out. I released the next version of my favorite pet, the TurtolCMS a couple days ago. Ok, let's be fair: WE, Roberto and I, released it. He did at least as much work as I did. So credit where credit is due. This is pretty much just a maintenance release and a version bump. I've changed the copyright to reflect the fact that I own the software now, not Turtol. We completed and added a bunch of the features that have been waiting in the wings since Turtol's demise in January. And Roberto and I mapped out where we want to take the software. When Turtol was around, we had a guy (Will) whose job was pretty much to work on the CMS. He did other stuff, but since we based our operations around the thing, it made sense to have a guy devoted to fixing the bugs and advancing the features. As we took on projects, we'd look to see where their specific feature requests would overlap with our "general" feature wish-list, and we'd get Roberto and Will to hack it together. It was, in my opinion, a pretty good way to run things. If we had enough advance notice of an upcoming feature, the guys could spend the extra time to build a general purpose tool rather than a specific one, and then we could resell the crap out of it to other customers (or, very often, to concurrent customers). If we didn't have that kind of time, we'd do the general design, Roberto would crank out a very specific solution for our customer (at an appropriately higher price, naturally) and some while later Will and Roberto would generalize what had gone before, and THEN we'd resell the crap out of it. There's no doubt that software publishing has the greatest potential ROI of any industry, precisely because you can resell the crap out of it.
"Not Dead Yet" was uttered at about 85 MPH on a desolate stretch of country road, some 15 years ago, by my pal Andrew with a huge grin on his face. He and I had gone, one fine early spring day, to get my '73 Javelin/AMX (don't worry that you don't know what that is, not many do) out of storage. It was, even then, starting to show its age, and I hadn't had the time to really get it back into shape. So, after checking the fluid levels, tires, belts, hoses and other bits, we set out to "blow the carbon out of her." That's car-geek code for flooring it until you're just plain scared of how fast you're going. One moment, we're pulling out onto the road, slowly, carefully, motor shaking slightly, grumbling moderately. The next, we're burning gas, and roadway, at an alarming pace (this was when gas was cheap, mind you), the hissing sound of the V8 swallowing huge volumes of relatively cold air into the cowl-induction system and the outright angry roar of the exhaust pure music to our ears. Yes, it was showing its age, but my Jav wasn't dead yet. Some while back in this blog, I noted that just because my company, Turtol, had gone defunct, our product the TurtolCMS, wouldn't. This is because it is Open Source Software, and as such can survive the death of the parent company, provided there is sufficient energy, somewhere, to keep it going. I'm pretty attached to the TurtolCMS, but because I've been making other plans and pursuing other opportunities (yes, yes, eventually, I'll let you in, when things are more stable), I haven't been able to put the energy in. Recently, that changed. One of my "other opportunities" will be based on the TurtolCMS. To further that, I, along with Roberto (late of Turtol, also), have been working on the next version. We made some significant advances, but most of the work has been focused on taking a slightly worn-out engine and rebuilding it so that it will be even better in the future. We'll do some benchmarking in the very near term, hoping to see quite a lot of performance improvement. We have some plans in place which we expect to provide even larger gains in performance, features and usability. But it's Not Dead Yet. Look for the release of TurtolCMS v0.6 in a week or so in all the usual places.
Programmers say that a lot. I suppose all of us, when looking back at our earlier work, swallow a bit and say "I can't believe I did that! What was I thinking!" This just shows that we're always growing, getting better at what we do. I hope.
I've been spending most of my time with business development matters recently, with the exception of laying the technological groundwork for the next major version of our TurtolCMS software. Well, the first major version really. We've been using the "0.5x" moniker for a while, because we know it's not what we want it to be. But we have 50 or so clients using this tech, and we can't just wholesale replace it, so Will diligently makes improvements and keeps it alive. Meanwhile, I've been doing a clean-slate redesign. It's been fun.
But once in a while, I need to revisit my earlier mistakes (again, 0.5 is good, but not as good as I'd like) to see what NOT to do in 1.0 of the thing. A client requested a search bar in their site, and I figured I could kill a couple of birds with just the one stone. I set about to create a Google toolkit as an add-on to the TurtolCMS, including a site search mechanism. It was pretty easy code to produce, but it was way klunkier than I'd have liked due to some built-in deficiencies in the old codebase. Leading me to that "I can't believe I wrote that" moment.
Here's to continuing to improve at everything we do.
I recently took Shane and Peter's Interview, as you can see, and in an follow-up comment here, Shane asked:
"a bit off topic but - how does an open source company work. I know its not quite an oxymoron, but I am feeling all conflicted when I think of it." A common sentiment, and I'm not sure I actually have answers just yet. After all, we've only been in business for about 18 months at this writing. We haven't proven we can sustain what we're doing yet. Let's give it another 5 years and revisit. But, having studied many different business models and several theories of market and social dynamics, plus a whole slew of ideas on motivators, I can say that at least my considered hypothesis is what we're doing will work out, long term.
The options for making money from OSS (Open Source Software) seem to fall into several general categories, which we'll visit each in its turn.
|
