Alex Blewitt

Utterly pointless. "Write a method". Gee, if I'd wanted to see a page filled with Google Ads, I'd have gone to the Google Ad page.

Thanks for quoting me - and for searching through the links - but would it have killed you to link to my blog piece at http://alblue.bandlem.com/2010/10/apple-deprecates-java.html or the InfoQ article that spawned it at http://www.infoq.com/news/2010/10/apple-deprecates-java? Heck, I linked back to the JavaLobby discussion in my write-up.

NetBeans does some things better - Eclipse does other things better. IntelliJ probably does more things better. Having competition is good, and having one IDE pull ahead of the other helps spur on competition as well. I wouldn't say that any one IDE is 'vastly' superior than any others - for example, OGSi development in NetBeans isn't as good as the Eclipse one yet. That doesn't mean it won't get there, but these types of articles are pure flamebait.

And yet, thanks to his post, and a few commenters on the subject, he's now decided that he likes Git: http://reprog.wordpress.com/2010/05/13/you-could-have-invented-git-and-maybe-you-already-have/ From the conclusion: --- Where do I go from here? I am not truly sure. I need to give this some time to sink in, and blog about something else for a while. But I think one distressingly likely outcome is that I’m going to buy the book [amazon.com, amazon.co.uk, free online version], learn git properly and then start alienating all my friends by telling them all, in the most patronising possible manner, that they’re thinking about version control all wrong and it’s really change control. Ah, poop. I feel like C. S. Lewis must have felt when he famously wrote “In the Trinity Term of 1929 I gave in [...] perhaps that night, the most dejected and reluctant convert in all England.”

Seriously, "Google for words you don't understand?" And people voted for this link? I wonder if people vote just because of what's in the title, rather than what the content of the article is.

Jigsaw is the modularisation solution for *the JVM*, not for application code. There's a number of constraints which make it inapplicable for general application code, including: * Only works on JDK 1.7, which isn't going to be released until 2011 * Uses a different container format (jigsaw-module instead of JAR), so won't be transparently usable on other JDKs * Uses a flat classpath, so can't handle loading classes with different versions in the same space (either statically or dynamically) * Allows the module system to be escaped using simple calls like Class.forName() In other words, it's really only a compiler-time benefit, and one can achieve that at the moment with tools like Maven.

Coyler did *not* say that. TFA quotes his blog post here: http://blog.springsource.com/2010/01/12/dm-server-project-moves-to-eclipse-org/ The word 'complex' does not appear anywhere on that page. Further, the quote about whether OGSi is mainstream or not comes from a random poster on The Server Side (Reza Rahman). The actual Adrian Coyler quote says: "At SpringSource we know that open source development and community involvement can play a huge role in evolving simple, pragmatic solutions that enable a technology to bridge from early adopter to mainstream usage. We know because it is a path we have successfully taken many times. In creating the Virgo project at Eclipse.org, we seek to accelerate the journey of the dm Server and of enterprise OSGi along this path."

So, you can already use an existing standard that's been around for 10 years, or you can use someone's pet project. Combine this with the fact that it's heavily Spring oriented, yet SpringSource are showing the future (and what they support) is DM Server. Even if this was a perfect, bug-free implementation, the risk associated with using a one-man-band's solution is a significant reason not to use it for anything serious. Impala may be a fun toy project to have implemented, but its practical use in the real world is so close to zero as to be a statistical rounding error.

Who would use someone's pet project for modularity with Spring when you could use SpringSource's own DM Server, which does everything this does any more?

Watching a video is not a proper substitute for a decent synopsis or abstract. In other words, there's no point in watching a video to find out what a video is about; it's got nothing to do with the skills of the audience, it's a question of whether there's in fact any point in making the initial leap into it. What is perhaps more telling is that despite being informed multiple times that the synopsis is so badly lacking that no-one has a clue of what this is about, that you are also unable to concisely summarise here what it is about and instead ask rhetorical questions or insinuate that the lack of understanding must be on the consumer's side. So I have read nothing, from the original synopsis, or your answers, that backs up your belief that this is "the single biggest thing to hit the programming world". Worse, it seems to be based around a single OS/Platform (.Net/C#), which hardly is the 'programming world' as you put it; there are many other languages and platforms which are used globally and which is therefore interesting only to that subset that uses Windows OS, for example. Lastly, this whole thing seems to be dealing with modularity, which is a well-solved problem in Java (OSGi) and other languages. Heck, you can even buy BUGs that are hardware components that fit together modularly. And let's not forget that the holy grail of OO has been reuse over the ages, which invariably has never quite materialised for reasons that are obvious to those who have been around a while. The point is that had you come out with a decent synopsis of what this is (it's a module system for C# on Windows) then you might have stood a better chance at having people review it in a favourable fashion, rather than the vacuous synopsis here. That should really be the takeaway if you want to promote this in a more positive light in the future.

Has anyone else noticed that polterguy=rickdg=pinglang=mswatcher? Strange how they all seem to post/vote on the same things.

"(BTW, it *is* the single biggest thing to hit the programming world ;)" This is a personal opinion which, I suspect, is shared by few. Furthermore, there's nothing here which indicates why it's so important. Does it solve the multiple concurrency problem of multicore processors? Does it intertwine functional programming with software transactional memory to allow concurrent modifications of data in a typesafe way? Does it permit constraints to be proof-checked against required features? Or is it just a pet project?

Is there any way I can help you understand how to write useful prose that people will read instead of taking the mickey out of? What about starting from a position which doesn't make the assumption that this is the single biggest thing to hit the programming world (skinnability or otherwise) and actually explains what it is all about, instead of using hyperbole that conveys nothing whilst saying little?

"I don't know any better ways to describe what this means other than by using words such as revolution, magical, game-changing and so on." I don't know any better ways to respond other by using words such as barfworthy, gibberish, market bullshit and so on. "Its most important feature was obviously a *KICK-ASS* skin." Good grief, people still give a crap about skinnability of apps? Didn't that die back in 1998 when people realised what they actually needed was usable applications? The only revolution that this made was the revolution of the contents of my stomach.

The synopsis is incorrect, as the blog posting talks about other languages re-implementing maven-like processes, rather than using Maven for other languages.

What part of 'totally bereft of information' didn't you understand?

Post totally bereft of information. Did the people who voted for it actually read the post, or just thought the title sounded good? For those that have read the blurb above (first paragraph) here is the entirety of the second paragraph: "You’ll need a really good attention to detail and a strong understanding of your deliverables. IA is not Usability and vice versa. They are sub sets of each other. In terms of deliverables, (being an IA) I deliver wireframes, power-points, word docs, POCs, and mockups. There are at times design aspects to those deliverables. If you come from the design side then you’ll excel, but if you’re coming from library sciences with no understanding of design principles, you have a bit more homework to do. On the usability side, I simply test if the ideas work through Usability protocols"

Alex Buckley has posted that the 'inactivation' is an automatic kick that happens when a JSR doesn't file for 18 months. Here's the post: http://altair.cs.oswego.edu/pipermail/jsr294-modularity-observer/2009-December/000352.html

Yeah, I apologise for that. I tried to put in the InfoQ link directly as the submission, but it wouldn't let it through - so I put a ref from my blog to that one.

I've posted a follow-up to my original article, rebutting the points made re: Firefox/Oracle. A component system needs to track the versions of components, which are orthogonal to the name or numbers associated with an overall product. You can read more at:

http://alblue.blogspot.com/2008/06/components-products-and-fragile.html

Perhaps you've been living under a rock and haven't heard of OSGi. That's exactly what OSGi allows you to do ...

Hey, congratulations on getting the new look and feel up and running. Alex

This really doesn't do justice to Erlang at all. At best, it's a bunch of threads and a one-queue-per-thread model that's fairly trivial to do. It doesn't actually prevent races or synchronization; in fact, it would be easy for a synchronized object to participate in this causing deadlock. An interested reader might like to read the Scala actors design model, which looks into detail at the way in which message-passing concurrency might work at several levels. One that is considered (and rejected) is that of one-thread-per-consumer, and is shown to be both overly heavyweight and far less performant than other approaches.

Bloid; actually, I posted the groovy has an interpreter post prior to my knowledge of this one; I raised it in response to the comments left on my post. I've also replied to Graeme's blog entry, since he specifically misrepresents what I said. Sadly, the existence of an interpreter for Groovy or not is somewhat irrelivant to the point I was making in the original post; and specifically with reference to the other languages that I was comparing it to, they all have combinations of interpreters and compilers. Frankly, I don't know why Graeme and the Groovies are getting so wound up about it; after all, the groovy interpreter is clearly documented on the Groovy webpage as either Groovy Console or Groovy Shell. Having those things is a *benefit* for languages, not a detriment.

I love this part ... "Despite its promise to enable ‘write once, run everywhere’ mobile software for all phones, J2ME is really better described as ‘write once, debug everywhere, sell nowhere.’ "

I've also put some extra content here, for those that are interested: http://alblue.blogspot.com/2007/05/jvm-or-clr.html