In the comments to my last entry, MJR comments:

Interesting opinion, but harmed by some bugs in the basic ideas. For example, “open source” and “commercial” are not opposites, unless you have some particularly strange definition that you use. At least for the GPL this is completely false. The GPL does not give you permission to sell the GPL’d software. The GPL does state:

You may charge a fee for the physical act of transferring a copy, and you may at your option offer warranty protection in exchange for a fee. This quite clearly doesn’t say: you may sell GPL software. Thus, GPL’d software is most definitely not commercial software even though it may come on a commercial CD. Section 4 then says:

With the recent debacle over whether or not the ASF license is compatible with the GPL license, Leo Simmons discusses the arrogance of the FSF and concludes with: So I might as well stick to the MIT license. But the arrogance of the FSF makes me feel very much like giving them the finger and just going with the ASL. Personally I agree. I’ve never liked the GPL as a license, it’s just way too restrictive to be considered “free software”. It’s open software, but not truly free. The Free Software Foundation has redefined the term “free” to suit their purposes, but it’s hard to argue that truly free software is in the public domain. You can literally do anything you want with it without every having to worry about copyright law. The GPL however, uses copyright law specifically to restrict what you can do with the software. Yes, it’s less restrictive the what copyright law allows by default, but that doesn’t mean it’s not restrictive. It particularly annoys me that the FSF feel the need to try and force everyone to use the GPL instead of picking the license that provides the terms they provide. Many people may want to see commercial software using their code – it helps to spread good ideas and opens up job opportunities for them. Personally I think it’s great when companies use code I’ve released, it’s not like I had plans to use the code to make a profit or I wouldn’t have open sourced it in the first place. The other thing that gets me (as Leo pointed out) is that the GPL isn’t compatible with anything yet the FSF expects every other license to be compatible with it. The day I can use GPL code in my ASL licensed projects is the day I give a damn whether or not people can use my ASL code in GPL projects. In the mean time, everyone except those silly enough to use the GPL can quite happily use my code – in open source or commercial work.

Well I learnt something new today. All this time that I’ve been doing some extremely funky stuff with XSLT (and I do mean funky), I never actually had the idea of the context node quite right. I discovered this because of a bug I ran into with our updating of XPath expressions that use position(). Essentially in the XML editor I’m developing, we pick out the XPath expressions in the XSLT used to lay out the data for editing so that as their input values change we can update the result like it was a spreadsheet. When the input values change we identify which XPath expressions need to be recalculated and rerun them (using Jaxen) by passing in the context node from the document we’ve been editing. The trouble hit when we started using position() because it always evaluated to 0. As it turns out the position() function counts the position of the current node in the current context, not necessarily it’s position in it’s parent node. So where I’d been passing in a single node as the context, I should have been passing in a set of all the child nodes from the parent and pointing to the appropriate node in the list as the “current” one we’re processing. As it is, it’s still not actually implemented correctly because we always use the parent nodes children as the context but the actual context may have been different. For instance given the template: <template match="item"> <value-of select="position()" /> </template> The context is actually the list of child elements of the current context node whose name is “item”. It wouldn’t include any text nodes and it wouldn’t include any elements that weren’t “items”. So for the XML snippet: <parent> <item /> <other /> <item /> <other /> </parent> The context would be the two item elements, but not the text nodes and not the other elements. Fortunately, for our purposes the approximation we’re using (all child elements, excluding text nodes) is good enough for what any of our users will need to do and significantly more accurate than anything the visual design accompanying our product will actually produce. I’ll be interested to see if the testing guys pick up on it though.

Weeeeeee! Eclipse is finally usable on my powerbook. M7 hit the streets a few days ago (or that’s when I noticed it anyway) and the performance improvements are quite nice, particularly on OS X. My plans for getting debian installed on my laptop can now take a bit of a back seat as I don’t need a Linux or Windows box to run Eclipse on anymore. This is actually a major relief since my PC has died recently and the state of Java on Linux for PPC is absolutely abysmal. I can’t seem to find a fully complete version of 1.3 let alone 1.4 – in this context fully complete requires a functional JIT. Maybe people should stop winging about Apple taking too long to release new Java updates.

David Starkoff writes:

What I�m getting at is that the primary audience of judgments is not the general public. The general public couldn�t care less. Further, I wonder whether the primary audience�at least at the appellate levels�should be the general public. I think for the most part, they are decisions argued by lawyers intended for lawyers. There is no doubt that the law is becoming more accessible to the general public. The Internet has seen to that, and the Australian judiciaries have embraced it. AustLII is a great resource and media-neutral citations are fantastic. But does it mean that judges should tailor their writing to the general public? I�m less than convinced. I have to wonder, how do you expect people to take an interest in law if all the documents and all the judgements are written in legalese that only lawyers can underst’don’t expect non-computer types to take an interest in the highly technical documents I write, but I might expect them to take the time to read a carefully written, simple to understand manual or overview of the technical information. In other word’can’t expect people outside of your target audience to read and understand your work. If you want the general public to pay attention to legal issues, you need to make them accessible to the general public and that largely involves lawyers learning how to use the English language in more useful ways – ie: in a way in which they can be clearly and easily understood. That is after all, what language is all about.

At long last, HttpClient 2.0 has been released. It’s been a lot of hard work from a few almost completely distinct teams of developers (with only small contributions from myself if you’re wondering). I was originally hoping that HttpClient 2.0 would make it out in time to be used in EditLive! for Java 2.0 we actually released EditLive! for Java 3.0 a couple of weeks ago and it’s too late even to make it into the upcoming EditLive! for XML release. On the plus side, the CVS snapshot version we’ve been using for the past year or so has been running perfectly without any customers running into problems. There is to be a toast raised to HttpClient 2.0 at the Nelson in Zurich and at P J O’Briens in Brisbane Australia. Company at the Australian toast is particularly required since I believe I’m the only HttpClient developer from Brisbane.

I haven’t had much time to write lately, but thought I’d point out a few things that caught my eye recently. First up, a little something from Eric S Raymond. Apart from being a particularly poorly worded letter (hint: try being nice to people when you want them to give you millions of dollars worth of source code), it’s full of some really odd comments.

Open source is hardly a zero-revenue model; ask Red Hat, which had a share price over triple Sun’s when I just checked. Last I checked the pricing of shares on the stock market was only an indicator of the value of the company when taken with the number of shares available in the country. Thus, whether or not Sun’s share price is above or below Red Hat’s is largely irrelevant. Nor is Red Hat a good comparison when talking about Java. Red Hat makes money from selling it’s support services, there’s already a massive wealth of professional Java support services and training available that aren’t provided by Sun, whereas Red Hat is the only support provider would consider for a Red Hat server. This is particularly so because you rarely need to fix a crisis in Java itself, but rather in your Java code – Sun has no better knowledge of your code than any other Java expert.

So I need a new laptop. My current, battle-worn, tried and true PowerBook G4 400Mhz has developed a large crack down own side and while it’s still running fine, it can only be a matter of time before that crack gets bigger and eventually the LCD falls off, possibly taking half the ports on the back of the computer with it. So I’m starting to look at laptop options – almost certainly a new Mac unless there’s some insanely great deal on a PC laptop, but the prices for PCs look to be as much as or more expensive than the macs anyway. As such the question really becomes, do I want an iBook or a PowerBook and how cool a laptop can I afford? Since spending more than $3000 is going to be difficult, the iBooks are ahead from the start, but what’s got me baffled is that they are so feature comparable to the PowerBooks. The iBook I’m looking at: � 384MB DDR266 SDRAM (128MB built-in & 256MB SO-DIMM) � 933Mhz G4 � 40GB Ultra ATA drive � DVD-ROM/CD-RW combo � AirPort Extreme Card � Keyboard/Mac OS � 14″ Screen A$�2,586.00 The PowerBook: � 512MB DDR266 SDRAM – (256MB built-in + 256MB SO-DIMM) � 1GHz PowerPC G4 � 40GB Ultra ATA drive @ 4200rpm � Combo Drive (DVD-ROM/CD-RW) � AirPort Extreme Card � Keyboard & Mac OS (default) � 12.1-inch TFT XGA Display A$�3,374.00 So the PowerBook winds up with some more RAM (both machines have been customized from their defaults a little), 77Mhz extra CPU power and a 2″ smaller display that uses the same resolution (1024×768). It costs $800 extra. By the time you go up to the 15″ ‘ok that I really want you’re looking at $4000. Sigh. Powerbo’ also do dual head, but I’ve hardly ever used th’ay, and considering I don’t actually o’rking monitor anymore, it’s probably not going to be all that much use in the future. So I guess the ’s the best deal for me, I’ll wander down to the Apple Store tomorrow and have a play with both to make sure the different amo’ cache and bus speeds don’t make too big a difference, but c’rom a 400Mhz machine that’s still fast enough for nearly all of my needs, I imagine everything will be fine speedwise. Anyone got comments on any hidden benefits of iBooks v Powerbooks?

There’s been a lot of discussion about Microsoft’s shiny new patent on putting multiple scripts into one XML file – all of it saying how ridiculous the patent is. This however, gets my vote as best retort. My view: it’s crap, but most patents are. Time to make people go back to having competition I say.

Gregor J. Rothfuss asks:

what will be deployed 100% first, IPv6, the metric system or Unicode? My answer: the metric system, it’s got a massive head start. They all require America to pull it’s head out of the sand and play nice with the rest of the world though so probably none of them will ever happen. Sigh. On the subject of Unicode – what is with defining a universal character encoding that has so many different forms??? UTF-8, UTF-16, UTF-16BE, UTF-16LE, UTF-BE unmarked, UTF16-LE unmarked, collapsed accents, expanded accents. Gah!

Well, I thought I’d download Firefox and try it out for a while (this entry is being written using it). The first thing I notice is it doesn’t look right. Read on if you don’t mind hearing criticisms of Firefox.

Dave Hyatt finally got around to posting to his weblog about the release of Safari 1.2. Some really good stuff has gone into it and congratulations are in order for all involved. Sadly, my most anticipated feature, LiveConnect, appears to be broken. I’ve seen it work when going from JavaScript to Java, but never managed to get JSObject working to communicate from Java to JavaScript. Dave’s blog is actually the first place that I’ve ever seen that actually mentions going both ways, or at least that what I assume he means by: