I first encountered Frontier in 1998, when it was released for Windows. At that point it was free, and UserLand had yet to ship Manila, but I'm certain it was an idea that was brewing deep inside the company.

I found Frontier because I started reading Scripting News. I found Scripting News because one of the pages I was reading daily -- Cafe au Lait, by Elliote Rusty Harold -- kept posting links that he had found from that page. I figured I might as well go to the source.

I found out that Dave had written ThinkTank, which I remembered from the Mac world. I'd always loved outliners, and when I wrote a scripting environment for a company I was working for, I used the outliner metaphor for the code browser. Imagine my pleasure when I found that Frontier did the same thing, only better...

I loved it. I didn't write much code in it at first, I was more interested in this whole "content management" idea that was rearing its head in the industry.

Then came the fateful day when UserLand decided that they couldn't continue giving away Frontier for free, and released version 5.1 as a paid upgrade. I understood the situation -- I've been getting paid to write software for a long time, and can understand that while giving away free software is cool and makes a lot of people happy, we all have bills to pay. Unfortunately, I couldn't afford the license, so I stuck with the 5.0 version, and started looking for alternatives, which led me down the Zope and Python path (incidently, I found out about Zope on Scripting News).

When the Pike/Radio releases came around, I was happy, because I got to use the new Frontier core (with some features disabled, but the base language was there) again, and I also could afford to buy a copy (paying for software makes me happy, oddly enough). Since I never really used Frontier as a coding environment (with the exception of my first run at making Python work in Frontier), I figured that what was in Radio (with the exception of mainResponder and Manila) was it.

Now, with the open source release of Frontier, I discover that that was not all it. The Frontier coding environment is much richer -- where in Radio you needed to create a database entry to write some code, that is not necessary in Frontier. You can just start a new script right away. Fantastic. Anything that gets out of my way and lets me code faster is a plus in my book. You can save the scripts as files in the filesystem, and have them automatically run when you double-click them -- I feature I vaguely remembered, but had never really used.

There's more, I'm certain, because it's a big environment. As Dave said, it's a life's work, so it has to be big.

He gave it away so it could live on. That's cool. I wish more things had been done that way. He mentions the fate of MORE as being one of the reasons he did this. Imagine if MORE had been released the same way, and then kept up to date.

Now that I know that Frontier will survive as long as someone keeps a copy of the code somewhere (I now have it stashed in many different places on my network), I'm more likely to base projects on it. Frontier should become an alternative right up there with Perl, Python, Ruby, TCL, and any other scripting language out there.

12:55:26 PM    comments ()  trackback []  

As I write this, SpaceShipOne is about ready to land after completing the first step for claiming the XPrize.

My heart was in my throat when the ship started to roll on the ascent, but they did it, breaking the record set by the X-15 rocket plane. SpaceShipOne has now gone higher than any winged craft other than the Space Shuttle.

(SpaceShipOne just touched down) All that remains at this point is for the XPrize committee to certify the attempt based on the radar data, and of course for the team to do it again one more time. Currently the next attempt is scheduled for Monday (October 4, 2004).

8:39:12 AM    comments ()  trackback []