techMusings

ActiveGrid.

Via Mr Udell's blog, saw this on ActiveGrid. I noticed it the last time Jon mentioned the firm, but didn't pay all that much attention -- the fact the name conflicts with ActiveMQ / LogicBlaze products is not good, who knows who got there first.

On the technology page, a few things struck me.

• it only seems to be competing with Java infrastructures. No mention of .NET or ASP at all.
  
• it seems to be trying to spread FUD about Java, by comparing apples to meringues
• Java cannot manipulate Strings directly -- what, it shells out to perl, or something?
• Java means you have to use lots of different connectivity solutions (JMS, RMI, JCA) -- err, they are VERY different technologies in intent and purpose
• java has a lot of overheads for xml because it is strongly typed -- err, these guys don't seem to know much about java, SAX anyone?
  
• old transports bad (ok, they work, are high performance and are well established)
• xml transports good (meaning? REST? WS-*? Indigo? whatever BEA are calling silverstar-mahoogy these days? working? with what database? raw xml to the db, are you serious? I could go on ... )
  
• language neutral services -- great, now i need to train a team to write and debug in a variety of languages? that's a help? I can see that in practice you're going to choose one only for a project, but it's not a huge selling point imho
• database queries in XPath, great
• web forms in XForms... supported by all known browsers

I gave up at this point. Either they don't know what they're doing or, more likely, they got a grad / non-techie to write the docs.

I'm a little disappointed Jon didn't give a more thorough going over. It's great to tout the emergence of LAMP on the enterprise landscape, but this implementation won't help improve it's profile the way it is/is presented.

Update from the Conservatives' Charles Tannock.

Not so good on this one. Looks like the Tories think software patents are ok in general and the legislation for Europe is better than the US anyway. I'll show a few quotes (with permission):

Starts off not too bad:

The objective of this Directive is to clarify existing EU Patent Law and provide patent inspectors with a common framework within which to examine and if appropriate, grant patents for genuine innovations involving digital technology. An explicit objective of the proposal is to ensure that computer software or business methods that do not involve new innovative concepts, making a technical contribution are excluded from patents. This will give the EU a distinctive and different position from the US and Japan.

However, it's been shown that the barrier to qualify is already much lower than the current UK system. Now, reading his bio, Charles is a bright spark, but it also shows a distinct lack of software development knowledge (funnily enough, being a doctor and all...). From here the response goes into freefall..

We need to consider the potential effect of the Directive on software development. I think that the problems here can often be exaggerated. There is little evidence from the USA that software development has been slowed down by the US patent regime. If the EU Directive is passed, it will be more restrictive than the current US patent environment. There is little sign from the USA, of large companies pursuing small companies for patent enforcement -evidence suggests that the opposite is the case. Also, patent specialists consider that the passing of the EU Directive will exclude the attempted enforcement of existing US patents across the EU. We may need to tighten up the proposal to ensure that this happens.

Yep, no evidence of the growing mexican standoff in the US between largely large companies. For a system that is alledgedly there to protect the little guys, it's not a very good showing.

We are also especially concerned to protect innovative companies, especially small firms, using digital technology to produce genuinely original technical solutions. We have been contacted by many of them and they are very concerned that they may be excluded form the patent regime by inappropriate amendments to the EU proposal. Patent royalty income is very important to these companies and is a major incentive to innovative research.

Lastly it is important to consider the impact of the Directive for the European Union to remain competitive in global markets. If Europe's capacity to protect innovation in the field of technology is reduced, compared to other regions of the world, we may in the long run no longer be able to sustain our standard of living by innovation.

For example, European car manufacturers will increasingly compete with low cost Chinese car makers. It would clearly be a disadvantage for European manufacturers, if Chinese producers have access to innovations without incurring research and development expense, because there is insufficient patent protection. We must have a fair and balanced internationally competitive framework.

Between mixing software and non software and throwing in a little scare mongering I think this position is pretty awful. I replied to Charles with a number of points to try to explain my concerns. I have to say, for a party that is looking to help small business this doesn't appear to be a good first move. Charles does mention that various amendments may be necessary, but then fails to specify when -- presumably well after the new rules are in place and damage may wel be done. I suspect, not knowing the procedures that well, that once in place (ie 'agreed'), getting changes will be like pulling teeth.

Overall, Labour 1, Conservatives 0 -- 5 more MEPs still to respond...

I did a quick search for more background, finding this is very scary:

As a final observation, it should be noted that the trend in favor of patentability, started in the U.S. in part via of the instantiation of the Court of Appeals for the Federal Circuit, appears to continue. European Judge Steinbrener, who wrote that "it is legitimate to have a mix of technical and ‘non-technical’ features (i.e. features relating to non-inventions within the meaning of [§] 52(2) E.P.C.) appearing in a claim, even if the non-technical features should form a dominating part," and his American colleague Judge Clevenger, stating that "[t]oday . . . virtually anything is patentable," seem to agree.

[my emphasis, see orignal for some extra citations etc.]

Platypus.

Don't know how I missed this one, but Platypus is a fantastic extension to Firefox which generates Greasemonkey scripts in a (somewhat clumsy, currently) wsyiwyg type way. I'd seen that Firefox 1.1 (Deer Park) was going to allow per-site CSS, so you can stylise a site to your own preferences -- eg make text larger / etc. This goes way beyond that. it's still a little rough but this is very cool.

Since I need to suggest something for them to add - (I assume the context menu is coming/broken in 1.1a1) is to be able to assemble web pages portal style. For example, can I take two pages from bbc.co.uk, say news and weather and create my own new page? The news page is currently quite narrow -- I know they are thinking of relaxing that for some users -- but being able to add aribtrary page fragments would be very cool. A logical extension to being able to cut and paste elements within a page is to be able to paste on to a fresh new page. This could be local, and use platypus, or a server-side script could be generated to produce instant portals.

Obviously the custom aggregation side will raise copyright / control issues, but that doesn't mean it wouldn't be fantastic to have. Firefox's wave of cool plugins, is definately driving some smart innovations on the web.

OK, first update. I've had a reply from Mary Honeyball, with a full statement, which has to be said, says exactly what I wanted. Essentially, the European Parliamentary Labour Party reckon:

• We are not in favour of the patenting of software as in the US.
• Europe needs a uniform legal approach to stop the drifting towards extending patentability to inventions, which would not have been traditionally allowed, and to stop patentability of pure business methods, algorithms or mathematical methods.
• Software products as such, must not be patented.
• Opensource software must be allowed to flourish and the Commission must ensure that this Directive does not have any adverse effect on opensource software and small software developers.
• Patents and the threat of litigation must not be used as an anti-competitive weapon to squeeze out small companies.

Apparently the vote is expected to be narrow, so if you care about having a European software industry at all, go chase your MEPs now!

EU Software Patents.

In a spurt of almost unseemly energy, I've just emailed (individually, long hand (well-ish)), each of the London MEPs to ask about their stance on software patents. Right now, there is a motion in the European Parliament to give the EU, US style (non-worthy-)patents.

For reference, I included a couple of links:

• http://www.guardian.co.uk/online/comment/story/0,12449,1510566,00.html
  
• http://www.nosoftwarepatents.com/en/m/intro/index.html
  

I encourage you to do the same. The UK lists of MEPs for each region can be found here.

I'll post any correspondence I receive back, here. If nothing else, it will be interesting to see if any of them reply / care.

Another business article to read.. Marshall Brain.

Quite good, with some excellent links in there too. Some of the web stuff on WebKEW is good too.

Another (in)secure webstart demo -- from java.net.

bloged is the latest security mare with webstart. First, the app requires a security acceptance -- to edit a blog, why?; second, the cert on it doesn't verify. Great. Must send them an email to fix that...

Mustang b39.

Just noticed over here that Mustang is going to get a compiler API and a script engine one. While the latter (including bundling Rhino) is pretty cool and is great news for Groovy and Beanshell too, there's a little gem hidden in the former description.

We also plan a tree API that will complement JSR 269 by providing access to the parse trees created when Javac compiles a program.

So hopefully a standardised AST / parser for Java code will help innovative build and testing tools. While it's not hard to get hold of a Java grammar, getting one that deals with every nuance and every new feature (eg generics) can be a fair bit harder. This was something we asked for way back at GroovyOne from John Rose -- maybe he helped push this one through. It does seem pointless that every language level tool should need to build their own trees.

I wonder how similar the scripting host will be to the (defacto standard) BSF?

Another cool addition is (for those on Solaris 10) dtrace probe hooks. One of my first questions to the sun guys when seeing a dtrace demo, was "does it understand Java?". It's a great tool, but the JVM is a big and complicated beast, and the last thing you'd want is to have to search for C++ locations that give you information on GC / Java threads / security checks etc. Nice.

Apple on Intel - Java is the winner.

Yeh yeh, lots of news and opinion. For at least a few years we'll have Apple spanning chipset just like Sun do today. A big winner from all this is Java I think. Yes OS X is still going to be a different beast from Solaris and Windows, but for Java techies, an Intel/x86 box is going to be the obvious choice -- Sun, IBM, BEA all have competitve JVMs on x86, so Apple's porting efforts must be eased and/or competition of OS X JVMs has got to be healthy.

Via ongoing, and Bosworth, a comment from Bill which echoes my previous post about mobile code:

Adam, I'd be less interested in XML v Javascript, than Declaration v Behaviour.

This is starting to sound like what Bill Joy said a few years back about XML - at some point you need to send behaviour around, declarations aren't enough. That, for exactly same reason we end up putting scripting hooks in all our declarative languages.

Sending Javascript notation for client eval will work. Eventually you will want to sandbox that behaviour off for security reasons (and limit the effect of bugs). At that point you're within spitting distance of Java 1.0 and its security model, which was designed with running remote code in mind - a chroot jail for objects.

Round and round we go!

So, now we've come full circle maybe it's worth talking about a Jini browser. It's an idead I've had for ages, but essentially it echoes a web browser except that it's Swing (probably). And I don't mean Swing HTML controls, but fat client GUI talking whatever you want off that back due to Jeri etc. You want to administer a switch in the darkest recess of the network? Then open it's GUI.

Now I'm sure a pure fat client approach is overkill and HTML is fine for many (if not most, I'm a thin client fan), but it'll be interesting to see if the Ecmascript world starts evolving most advanced security models to cope.

This brings me to a bug bear (sp?) of mine. Yes, Java has a nice security model (ok nice maybe isn't the right word, but powerful at least) and applets can and most do bhave within it. But WHY do almost all WebStart demo's request full security access just to demo some fat client behaviour? This makes the security guarantees pointless. Lazyweb, please make people fix demos so no extra priviledge is needed. Sun please make it easier to do this. (I know there are a few bits and pieces out there - a tool from Kevin Jones at DevelopMentor springs to mind). Mozilla.org/Sun, please fix the JVM /browser integration/applet capabilities (eg allow an applet to request simply (with no other privs) to live outside the page lifecycle (either tied to the browser or even longer). This stuff needs to be easier. Maybe a Groovy browser / rhino plugin could help blur some lines... dunno.