Jason reported that he's having trouble doing trackback to my blog. I've had trackbacks from a few folks, but I want to be sure that the pyblosxom trackback code is working. So I'm asking a favor. If you have a tool that does trackbacks, could you try to trackback to this post. If you get an error, can you let me know via e-mail so I can try to see what's happening?

Thanks!

Jason replied to my posting about the JCP. I like the way that he titled the post, "Ted Leung on organic specs", except for the spec word. The problem is that we have too many specs. I'd like to see the JCP become much more resistant to standardizing stuff, but I doubt that's going to happen. If you think that having the JDK/standardized API set grow organically, then what we can start to do is just ignore the results of the JCP and continue to develop alternatives and use the leverage of blogosphere as a way to help people realize that the blessing of the JCP is often a recipe for poor technology. Just because the JCP and the vendors are pushing the "standards" doesn't mean that we have to use them. But we do need away of protecting ourselves from vendor lockin.

Jason Carreira and Andy Oliver are shooting it out over the JCP again.

The problem here is that the specs are being driven by vendors trying to get levarage over each other. The specs should be done by non-vendor experts after the area in question has settled down.

I think that we're better off with a system where we let many flowers bloom and have a way for the developer community to make recommendations to each other about what works and what doesn't. In a way, that's partially how I view the techie blogosphere. People talk about what they are using and for the most part they tell the pure unvarnished truth. After a while you develop a sense for whose opinions are trustworthy and tasteful. Take a handful (or maybe more) of these people and put their recommendations together, and you have a system for finding the best practices, tools, and libraries. Then you don't need to to running to something like the JCP just so people will use it. Once time has seasoned the recommendations, then the moment for standardization has arrived.

Via Brett Morgan

And who says that language research has died?

Language research isn't dead. But it would sure be nice to be able to write a real Gnome, Windows, or Mac OS X app in Flow Caml...

James Robertson is frustrated with my frustration:

There's actually fairly vigorous competition in the Smalltalk space, because we have multiple vendors - we get various takes on what works and what doesn't, instead of what Sun knows is best or what Microsoft knows is best.

Unless the multiple Smalltalk vendors are actually changing the language, then they are just competing on the tools. I'm not knowledgeable enough about the state of .NET tools, but in Java, there was lots of competition in tools, Visual Cafe, JBuilder, Supercede, NetBeans, Eclipse, and there'll be more. I don't think that it's fair to say that theres no competition in the curly brace tools space.

Why is Eclipse boring? Perhaps it's because - unlike Smalltalk and Lisp - it takes no advantage of being written in itself. You can't modify the environment, or even ask intelligent questions of it. Being able to do so is what leads to productive tools.

I agree that none of the Java IDE's is really taking advantage of being written in itself. The ability to inject new code into the the IDE while its running is a significant advantage. I believe that this is also language specific, and could be done using something like Jython.

Python doesn't quite get there, because - as a scripting language - people tend to use things like vi and emacs to develop. Productivity simply does not lie that way.

Python is the new poor man's lisp. It feels a lot like lisp, but it isn't. Its only recently (or so it seems to me) that its starting to build up some momentum. The availability of a REP loop alone is a big advantage over tools that don't have such a thing. But I certainly wouldn't argue that the tools are lacking. I don't know if Python is going to cross over. I'm still not completely conviced that I want it to.

So what is this collective blind spot in the developer universe? Is it the siren song of Open Source and free tools? Are developers thinking that if its not free, they won't use it? That's part of it, I'm sure.

It's not the developers, its the people who control the purse strings. At every company that I've ever worked at, one of the first things I did after I got hired was to ask my management for the best tools that money could buy, both software and hardware. In some cases I got them, and in other cases I didn't. Not all of my co-workers did that. Not all of my co-workers were aware of the kinds of tools that were out there. I was lucky. I went to school when Lisp Machines were still around. I knew about the PARC Smalltalk hardware. But most people out there didn't, and still don't. And if they do, they can't get their management to approve, or their management doesn't have the budget.

Also, I make a distinction between open source tools and free tools. Free tools are just that. They don't cost any money. You may or may not get the source. I'm not particularly motivated by free, but its nice. On the other hand, having the source code is important, especially for tools. I have modified my tools. I will continue to modify them. And I'm willing to pay to support the people who are actually making those tools. I'd be happy that goes something like this: There are multiple "editions" of IDE X.

• The "free version" is free and has reasonable but limited functionality and no source. After 1 year, you get the source to the free edition.
• The "normal edition" costs $xxx dollars and has full functionality and no source, but after 1 year, you get the source to that version of that edition, under an open source license.
• The "hacker edition costs $xxx+yyy dollars and has full functionality and immediate access to the source. So you pay extra for the privilege of getting the source a year before everybody else does.

I figure that a year (or pick some reasonable time period) is enough time for a tool developer to have a competitive advantage over people who just want to "steal" the code.

This is a lot like the source code licenses that vendors used to have in the past, except during those days, not that may people wanted access to the source, and the vendors that offered this type of license were mostly OS vendors. Maybe its time to re-examine that entire idea now that we've entered the open source era.

The dashboard feed reports that Nat and friends are taking no prisoners. I'm looking forward to seeing what these guys are able to cook up by the time Miguel gives his keynote on Friday.

Juri Pakaste, the author of straw is thinking about adding support for dashboard to his RSS aggregator, straw. While he's at it, I hope that he can change straw to report the version number according to RFC 2616