Sunday March 25, 2007

Byebye I've had enough of JRoller, you're only reading this because you're subscribed via RSS and are a lazy turd. If you were clever, you'd know that you need to change your subscription to http://feeds.feedburner.com/bileblog instead. The site itself has moved to www.bileblog.org (2007-03-25 21:13:46.0) Permalink Comments [1]

Wednesday March 21, 2007

TSSJS: I'ma l33t Arkeetektor I pwn j00! The last talk I went to today is Enterprise App Mashup: Architecting the Future by Eugene Ciurana.

Of course, the talk is as fluffy and pointless as I had expected. I walked in late because the likelihood of anything vaguely interesting being said was fairly negligible.

The one thing Eugene does excel at is name dropping. Whatever he did at walmart clearly involved next to no 'real' world, and a lot of musing and faffing about with whatever toys happened to catch his fancy. In many ways, it's a dream job. Wouldn't it be great to have a job where you get to fuck around with whatever you want, without ever having to achieve or deliver anything? Without naming names, there seems to be 2-3 jobs like this in this industry, held by vocal public personalities. Yet more proof that it's a cruel unjust world.

The core problem with this talk is how nebulous it is. We get the usual ESB flagwaving, that same tired old diagram of a centralized bus and how we're moving into a happy new world of no point to point, anyone and everyone plops their turdy nuggets onto the bus, who will magically and mysteriously only deliver it to other applications that want to gobble up said turdicles. The clever thing is that there still seems to be some people who haven't tried this approach ,and still don't know that if applications aren't written to talk this common language, you end up writing a zillion adapters, thus having the same ball of spaghetti, except that you're now also now beholden to a bunch of dubious shady people who now happen to own your centralised bus.

The ludicrous example we have is making two applications communicate, one of which talks JMS, and the other talks SOAP. Now, has anyone ever come across such a situation? Where they just happen to have message structure that's similar enough so all you'd have to do is drop in an ESB and now the world is magically wired up?

It's just so astoundingly stupid. One of the most impressive cases of the emperor having no clothes since SOA. Speaking of which, I went to the SOA panel which was equally depressing, services are pretty much....anything. It's a testament to the power of salesmanship that anyone is able to actually sell this crap.

It's very clear that Eugene spent a couple of days playing with Mule (ESB), and thought that he'd give them a shout out as some kind of benevolent godfather type at Walmart.

Even more ludicrous, he recommended that people use...JavaSpaces. Good lord, is it 1999 again? He even mentioned it as a sane alternative to JMS. Good thing everyone is going to ignore everything he's saying, so no harm done I guess.

The talk title certainly didn't disappoint, the talk is so full of shit, so pointless, so inapplicable to anyone that it's impressive that a human being can actually stand up for an hour and make noises in vague sentence structure about it.

Gotta give the man credit though, he has nothing to say but delivers it with style and authority, it's easy for people to end up thinking they've attended a good talk, because the speaker was friendly and engaging and spoke well. The fundamental flaw is that HE SAID ABSOLUTELY NOTHING. The talk basically revolved around how great the speaker is, how he's bought expensive hardware because he's so cool, how he knows so many people and vendors, and how some of them lined up for a glimpse of his holy genitalia. (2007-03-21 21:34:07.0) Permalink Comments [9]

TSSJS: Advanced JPA Sometimes I'm a bit of a glutton for punishment. You'd think that after attending so many conferences, seeing so many JPA drivel talks, I'd stop. Alas, I can't seem to help myself. and am dismayed to find myself sitting in Advanced Topics in JPA by Mark Richards.

It's evident that the use of the word 'advanced' was somewhat liberal in this case. All we've done so far (20 minutes in the talk) is discuss the joys of join tables, entity relationships and how to define them. I shudder to think of what Mark would consider to be an 'intro' talk. Perhaps a 20 minute discussion on the benefits of casting? The benefits of coding with eyes open and thumbs out of anuses, to maximise JPA productivity?

Even more worrying, he actually makes JPA config look so much skankier than it actually is. Switching from hibernate to toplink for example in his case seems to require changing 20 lines of Spring xml. Spring? Yes indeedy, he managed to sneak SpringSpringSpring in there! Either Mark or Spring is pretty fucked up if changing implementations is so much work (it's the former, incidentally).

To his credit, Mark is actually a good speaker, he's just cursed with abysmal material. OHMYGOD, he didn't just.....OHMY...

This is a brave move indeed! Totally unexpected. We were happily traipsing about in many to one land when BOOOM! He brought in...MANY TO MANY! There's an inaudible gasp from the audience at his sheer gumption and audacity. What will this crazy guy do next? Drop his pants and moon us? Masturbate into the first row of attendees? I might explode from excitement and anticipation.

As if that excitement wasn't sufficient, this crazy man just tossed in fetch types. I'm going to need a nap shortly to recover from this.

Oh wait, all is not lost, he actually mentioned something interesting (honest). Throughout the demo, he seems to pointlessly switch from Hibernate to TopLink for no reason, but this finally paid off with showing (what he alleges) a non-compliance by Hibernate, which apparently doesn't support fetching multiple collections, which the spec says must be supported. We switch to TopLink and all is well.

Next we're covering compound keys, which I guess is finally venturing to the realm of potentially non basic trivial features, too little too late I'm afraid. The one thing I am enjoying about this talk through is the hibernate bashing. He's showing fairly standard JPA usage, no custom stuff, yet hibernate seems to shart itself pretty regularly as a result. The latest example is that @IdClass is basically broken (and yes, that's been my experience too). I do like the advice that once hibernate takes a big dump, it'd worth switching providers just to see if it's because hibernate developers are cocksucking chozgobblers, or if it's you who is being a muppet.

Finally, we cover stored procedures, also pretty interesting and I'd say something that does qualify for advanced usage.

All in all, I'm not nearly as angry about this talk as I was in the first 20 minutes. It started off with boilerplate JPA but did eventually manage to eke out some useful info. A speaker who seems interested in his subject and is coherent is such a refreshing change as well, even though he does smirk a lot. (2007-03-21 15:41:27.0) Permalink Comments [3]

Another year, another TSSJS Depressingly, despite making a somewhat valiant attempt to get enough sleep, the night before TSS ended with the usual mix of drinking, stumbling about helplessly, and stay up until 3:30am. Last year the question to inflict on poor unsuspecting victims was 'have you masturbated on a plane', which surprisingly 50% of people seem to have done. This year I'm hoping to get a good list going with 'what deviant sexual act would you pay $10 to see if it were socially acceptable'.

The keynote in the morning kicked off with the usual fumbling intro, with the TSS guys clearly looking like they'd much much rather be somewhere else, possibly somewhere very far away.

As usual, the audience polling thing was a total and utter shambles. It's 9am, there's no way I can count down a bulleted list of 7 items to figure out what number I should press. Which genius though that a bulleted list is a good way of displaying numbered options? Even more depressingly, as I finally figure that out and start furiously jabbing the appropriate button, the handheld piddly pile of poop steadfastly refuses to acknowledge my pitiful attempts.

The one mercy is that there were very few of them compared to every year. Interesting it turned out to be yet another venue where they managed to sneak in vendor content, asking a bunch of java devs whether they're interested in DTrace or ZFS is a perplexing choice, certainly.

The actual keynote by Eric Gamma was, basically, enough to bore even the more excited of attendees halfway to death. Maybe it's because I'm not an Eclipse user, but really, 10 minutes of discussing the benefits of OSS and what it means to have a community, how it's all so wonderfully transparent, and how it's better for your genitalia than a penis pump is a bit...twee, at this point. How many times do we need to hear? Yes, we get it, OSS is as satisfying as openair hot lesbian twin live action with many toys; lets move on already.

Part of the problem is that Eric, while I'm sure is a frighteningly smart guy, is a fairly abysmal speaker. I'm having to work very hard to actually listen, but I've found that it's way too easy to let him become a mildly annoying background noise. Coupled with the miniscule font on the slides, this ensures I haven't a hope in hell of figuring out what's going on, why it's going on, and when's it going to stop.

The talk is supposed to be about Jazz, and we're halfway through the talk and I still have as much information about what Jazz is as I do about the contents of Bill Burke's panties (eeeurgh, even I winced at that lurid image).

The one interesting thing I guess is the release management arm waving, and all the boring crap that goes with it, its impact on development velocity, marketing, metrics, blahblah. Still no clue what Jazz is, but maybe we'll find out soon!

We finally find out that Jazz is some kind of collaboration thiingyboggy. Unfortunately, many people have already chewed off their own heads in sheer boredom, so it might be a little too late. Still, maybe something can be salvaged!

From what I can gather, it sounds like a more pretentious and happy agile noises version of TeamCity with better role and team management (at least to the marketing blurb we're being subjected to). Team collaboration junk, better isolation, scalable, blahblahblah. I'm still angry that yet another hour of my life has gone down the drain.

Ultimately I guess the only people who don't want to kill themselves just to end this talk are probably Eclipse developers (not users), who likely could manage to keep themselves sufficiently entertained by masturbating in the stage's generation direction for this infinitely long hour.

(2007-03-21 15:21:53.0) Permalink Comments [4]

Wednesday March 14, 2007

XML wiring is for girls The fact that XML is, basically, a steaming pile of goatshit is not news. Many many people know this now, yet you have an awful number of people eager to grab a hold of some xml and perform deviant sexual activities in it, around it, and in between its elements.

The problem is that there are some good ideas out there that happen to use xml. Even though xml has nothing to do with the quality of the idea, xml somehow gets credit.

This has become quite apparent lately when participating in some Guice vs Spring debates. The fact of the matter is, Spring's XML sucks unbelievable amounts of ass. Don't buy the bullshit hype of Spring 2.0 improving its XML so it's now all great. All it's done is made it plausible that one might not jab oneself in the eye with a big black ribbed for his pleasure dildo when confronted with it. I cannot believe that a sane person would think that the Spring 1.x syntax is fit to deposit a big dogturd on, let alone use in any application you'd care to be associated with.

I like Spring. I like the integration features it offers. The XML however is revolting. The IoC bits are about as pleasant as using a bucketful of Rod Johnson's moobjuice as anal lubricant. You either have to autowire (which is convenient but too magical as your project grows, not to mention fragile), or explicitly wire stuff in xml, miles away from the source code. Lets not even get started on how abysmal performance is once you're using a real project with more than 3 beans, or heaven forbid, wiring up actions at runtime. You could play many a game of soggy biscuit while Spring figures out who needs what, when, and why.

Of course, the situation is conveniently disguised by the fact that tools such as IDEA natively grok Spring's hellish configuration.

No doubt someone will counter with the fact that thousands of people use Spring and all is well. Yes, thousands of people also used EJB2, oddly enough the very same places that switched to Spring. What are the chances of all these venues suddenly becoming full of wise intelligent developers who from now on will only make sensible sane well considered decisions?

As a result, Spring is hopelessly fucked due to its success. It can't really drop support for crappy old JDK's (contrary to what everyone tells you, banks ARE using JDK 5 now), so will always be lumbered with its current idiotic configuration and wiring approach.

That problem makes it somewhat understandable that the Spring guys, in order to defend their fuckeduptheassness, deride annotations and proclaim that declarative xml goop is Rod's Holy Word. Go forth and write xml, Rod proclaims, and the unwashed masses go forth and positively ooze angular brackets. Stupid fucks. Still, all is not lost, in cases where it's possible to shoehorn annotations into Spring's archaic innards, they'll enthusiastically proclaim it Useful and Recommended, such as the @Transactional annotation.

The other common argument is 'you don't spend too much time in Spring's xml compared to time spent in your code'. Sure, I also didn't spend much time writing ejb2 descriptors either (xdoclet did it), yet everyone bitches about those. Why can't we apply the same standards to Spring?

Fundamentally, Spring has a tough time accepting that the world has moved on, and that its approaches are starting to look a wee bit outdated. Even EE 5 in many cases looks more modern and usable than some of its approaches. Thanks to its widespread use, they can, much like the JBoss people, maintain an insular world view where they're surrounded by sycophants who do nothing but bend over with a shiteating grin plastered over their sallow filthy faces.

So do yourselves a favour Spring guys, and stop being so obsessed with where you are now and with how much you can do with AOP (which nobody cares about, it's not 2004 anymore so move on), and instead look around with an open mind and embrace annotations like you'd embrace a Rod Johnson penis replica. Your configuration bean bullshit still doesn't cut it, dependencies are best expressed where they belong, in the damn source code itself.

(2007-03-14 18:45:17.0) Permalink Comments [37]

Monday February 12, 2007

Good riddance, Marc Fleury. Well well well, it's no surprise to anyone, but Marc Fleury has left the building.

Good fucking riddance, that's what I say. Regardless of what he did or didn't do, he certainly made out like a bandit, and I for one, like many many others, am very irritated (and jealous) about it.

Ultimately though, Marc achieved very little that is concrete. It's very debatable whether JBoss' success had much to do with him. Certainly, it did succeed, and a lot of people got conned out of a lot of money to make it happen, but such is the way of the world. The credit due it is more a matter of timing and of being the only OSS player in the field at the time, not that it's a great product or was particularly suited for any given task. Witness the abysmal failure that is Geronimo, a product that is arguably better developed, yet the world at large doesn't give a flying fuck about LGPL vs Apache licensing. Thus Geronimo remains a laughing stock and a head-scratching cautionary tale of OSS gone insane (not to mention that it's owned by IBM, and has developers that are about as capable of developing a product as Sun is of marketing one).

What Fleury contributed to the world of Java is a personality; love him or hate him, the man certainly deserved to be hated. His incoherent ramblings, the perplexing capitalization, the weak and tenuous grasp of the English language gave us all a lot of ammo, and made for some very entertaining drama.

The fact that he's also a shady, dodgy, and thoroughly despicable scam artist also helped tremendously. Who can forget the great astroturfing story? It's nice to see a company bigwig who will get his hands dirty, and openly misbehave in ways that one only expects particularly handicapped children to.

So sure, it is one less personality in Java land, but it's one we're all better off without. A testament to how bullshit and angry penis waggling can get you all the material success you want; it's a depressing lesson for any hopefuls to learn, and shows that even in this most technical of fields, a good con is just as easy to execute as anywhere else.

I wish him failure in his next endeavour, whatever that may be. Here's hoping that the cosmic balance is finally restored and the justice is served by Bad Things happening to him.

My one wish though is that he'd stop blogging. His blog is so painful to read that I have yet to complete a whole entry. It basically consists of him ranting in his usual incoherent style about whatever happens to have crawled up his ass and died that day. Now, I of all people am quite appreciative of aimless degenerate ranting, but for the love of god, have a care for the poor English language! I don't know how much of a second language English is to that man (though judging by his style, it might be more like a distant third), but can someone please explain to him that randomly capitalising sentence fragments is not a literary tool? It's like trying to read with someone punching you every few seconds; you eventually decide you're better off chewing your arm off than letting some random schmuck repeatedly hit you for no apparent reason.

Even funnier though is the army of JBoss sycophants who comment. You have Andy Oliver still struggling for relevance; nobody has told the poor bastard that he's as much of a nobody as anyone can be, and that it's not 2004 anymore. Random other JBoss halfwits chime in with their inane tuppence, presumably in the hope of garnering an acknowledgment from their divine leader, perhaps in the form of an incoherent collection of words and punctuation directed in their general direction.

Perhaps the most astonishing thing about the thread discussing this on TSS however is Bill Berk claiming he has 15 friends. How on earth did that happen? It's probably him astroturfing again, the reality is more likely to be that he knows 15 people who smiled politely at him as he frothed and mumbled his way through a sentence, as they tried to pointedly ignore his nipple tweaking in an attempt to gain their attention and approval via a well aimed squirt of moobjuice.

(2007-02-12 17:47:06.0) Permalink Comments [24]

Wednesday January 31, 2007

Dodgy benchmarks For anyone following JavaBlogs, you've probably skimmed over a bunch of Sri Lankan shirtlifters cooing recently about a highly dodgy benchmark that somehow proves that Axis2 is faster than XFire. These turds are of course the same bunch of clowns who are very busy plopping their tagnuts all over the WS-* racket at Apache. The benchmark is by none other than WSO2; the company that effectively owns Axis2.

What's interesting about this benchmark is that it manages to compare apples with oranges, and yet has its proponents drawing some truly perplexing conclusions.

There are some, ahem, issues with this benchmark. Like any vendor driven benchmark desperate to compete, It manages to compare wildly different things, just to find a case whereby they 'win'. Instead of comparing the stacks fairly, they focussed on comparing the data binding parts. What they proved in fact is that ADB (a proprietary Axis2 data binding framework) is faster than JAXB (data binding specification). OK, not so surprising. If you wanted to be fair and picked the fastest data binding for XFire too, you could test Axis2/ADB against XFire/JiBX (JiBX being a frightfully fast OSS data binding framework). You'd see that XFire shoves a sharp pointy plunger in the general (or maybe even not general) vicinity of the WSO2's development team's collective asshole.

In fact, buried in the so called benchmark, you'd find the results themselves prove that XFire is faster than Axis2 when the comparison is fair and uses the same binding framework. Somehow though, this rather crucial fact is lost in the conclusions.

What's more impressive is that no sooner than the company issues this highly dodgy report, its employees all start blogging about it and bragging that they finally managed to concoct a benchmark whereby they beat their competition.

Eran Chinthaka's (whoever the fuck he is) blog has a particularly hilarious entry, with such gems as 'We never did marketing for Axis2 telling bad things about other stacks'. Well my friend, I certainly am going to tell bad things about your stack, given your smug, snide, and ignorant post. The real icing though has to be '...then I feel Axis2 is even better than the numbers mentioned in the paper'. Huh? What, you, someone with a vested interest in your product, feel that the benchmark that you published and skewed to highlight your product, is not as biased as it should be? That reality, despite not even conforming to the distortion field imposed by your assumptions, somehow goes above and beyond it? I had no idea that the drugs available in Sri Lanka are so superior to those available to the rest of us.

This is of course, another 'open source' company. I thought that with JBoss being bought out and the old cabal being effectively neutered, we'd seen the end of the snide slimey comments by OSS against OSS, but our Sri Lankan friends seem keen to assume the throne of corporate fuckwittery in the name of OSS.

I hope this benchmark is taken as what it is, a company defending its own product against what is obviously a superior OSS solution. Friends don't let friends use Axis2. Do you want to base your webservices infrastructure on a product from a company desperate to take your money, written as a student project, with a bunch of incompetent weasels at the helm trying very hard not to look like used car salesmen? Do you want something that integrates and embeds well withint your existing infrastructure, or do you want something that hijacks it instead?

Still, good thing we have the goodship Apache; that thriving haven of open communities, that final refuge for corporate flotsam and jetsam, that last bastion against whored solutions foisted on an unwilling and uninterested public, eh? (2007-01-31 01:34:40.0) Permalink Comments [20]

Friday January 26, 2007

What makes for a good JavaOne submission For some inexplicable reason, I'm one of the external reviewers for this year's JavaOne EE and Web tracks. The one thing that's utterly perplexing about it is the dire quality of some submissions.

Tempting as it might be, I'm not going to name names. I'm not going to point out how many vendors pitches there are, or how much they suck. Instead I'm going to try and understand what on earth some submitters were thinking.

For vendors, honestly, what were you thinking? Do you think the people reviewing these things are stupid? Do you really think that JavaOne attendees love to hear about how you solved problem X using your own technology, that nobody can actually use without dropping trow, bending over, and paying for the privilege of being molestered?

The open source crap is just as bad. Here's a hint to you budding open source wankstains. If your project doesn't have a community, has no buzz around it, and is not particularly innovative, then don't bother. JavaOne isn't some whore you can throw your dirty papers at for a quick show and tell. Nor is it a venue for you to idly gaze into your navel and pick out lint in public, musing on its quantity, quality, and what possible use it might have.

Of course, some of the submitters are savvy to this culling process, and it's obvious that they're working hard to try and make the proposal not sound like a product pitch. Unluckily for you though, the reviewers actually do know what's going on in Javaland, and all of them have attended other conferences and know who has talked about what, and how successfully. So be careful, one misplaced vendor pitch could ruin your chances at playing with the big boys.

I don't understand how hard it can be to put yourself in the shoes of the average attendee. The goal of this conference (and ANY good conference) is to help said user, NOT to help the vendor or presenter. The fact that you get to strut your stuff and waggle your genitalia at a few hundred people at a time is its own reward.

So given that the average reader here is a discerning (if somewhat mentally unfit) Java type person, with one finger on the pulse of the community, and another firmly in an orifice, I'd like to you know what you think. What would make a good talk? What talks have you gone to that sucked so much you wanted to shove Gavin Fleury, 3 TDD zealots, and a Websphere consultant into a blender and smear yourself with the resulting goop while making sweet manlove to Howard Lewis Ship's beard?

On and before I forget, HAHAHA VIRTUAS IS OUT OF BUSINESS!!!! TOLD YOU SO! (2007-01-26 18:46:04.0) Permalink Comments [13]

Sunday December 31, 2006

5 things I don't want to know about you The whole '5 things that make me look cool' thing is getting out of hand. It's insidious and evil and idiotic, all at the same time.

For one thing, who on earth would give a flying fuck about all of these irrelevant, tedious, boring, repetitive factoids? This approach works well enough as a social lubricant; it's aimless little drops of pre-cum you can jizz out during a conversation just to keep things lubed up. Online, it makes you look like a tool at best, and more likely a self-absorbed child molesting fruitcake with nothing better to do that masturbate in front of random passers by in the hope that someone will stop and gawp at your pathetic flailing.

The real genius of it though is that it allows the tawdry twats who succumb to it to happily whip out their genitalia in public, rub it against their own orifices and spooge dramatically all under the guise of 'playing along' and doing it for other people.

Almost everyone agrees that a personal blog where one randomly defecates their incoherent and tedious ramblings is a Bad Thing, and most people have learnt (usually through being mocked horribly, especially in our industry) that nobody really gives a flying fuck about how many cats you have, how good you are at fucking your significant other and producing more boring people in your mold, or how excited you find penis slapping. This new approach though enables you get rid of any possible guilt, common sense or simple social graces you have with regards to regaling the world with this tedium, since someone else asked you to do it and it's for some obnoxious and incredibly not funny 'cause'.

Of course, it works well in nerdy culture. Who could resist the chance to brag or boast about themselves? It's considered fairly rude to do so in conversation unless it's a contribution of some sort, but online, you can regale people with your exploits with nary a care in the world, and even better, 'tag' (a polite way of helicopter peeing on) other equally fuckwitted chozgobblers so that they can also participate in this sick and twisted pyramid scheme of self-gratification.

Let it end. Those of who you have already gotten your jollies off, well done. You got to spooge onto the world's face, you felt better about yourself, and took your fill of the public buttplug all in one go. Good job. The rest of you who are vaguely pondering succumbing to this evil, resist. Resist long, and resist hard. Don't become another statistic, don't for one second assume that you're special or clever or a 'someone' just because some random pillowbiter diddled your sensitive spot. You're still a nobody, and far more importantly, people still couldn't give a flying fuck about random trivia concerning your tedious, 'I'd rather have my orifices plugged with my own feces while forced into a grouphug with 3 rubyists, martin fowler's beard, and 2 thoughtwankers than be this guy' mockery of a life.

Here's hoping people spot this pathetic agonising plea for attention for what it is. It's not funny, it's not interesting, it's just sad in that way that only a mentally handicapped 56 year old woman masturbating in public can be.

Happy new year! (2006-12-31 10:40:32.0) Permalink Comments [28]

Tuesday December 26, 2006

Open sores scams It's funny how whenever there's a whiff of money in the air, all the shysters and scamsters come out of the woodwork to try and make a quick buck. What's more depressing is how often it actually works. Witness Marc Fleury, for example, or Gluecode's sale to IBM.

The Java world has a good few companies like these, companies which don't really offer much, but have somehow managed to waggle the open source penis sufficiently that others have thrown money in their general direction for a quick stroke and lick.

It's delightful however to see that this particular scam doesn't always work. Two brilliant examples that come to mind are Mergere and Virtuas.

For the unintialised, both of these companies have no actual offerings, beyond brand. Mergere has a fairly inxplicable business plan; to build a company around maven. Not the actual product, but the idea. Suffice to say, they've flushed away a ton of money down the toilet and gotten absolutely nowhere, just like I predicted a year ago.

Virtuas is a really special one though. The basic premise of Virtuas is....well, it's hard to tell. They seem to have something to do with OSS, but what, nobody is quite sure. They do throw nice parties though, and I fairly gleefully drank their alcohol and wasted their food at various conferences. A fool and his money are easily parted.

The three 'brand names' that Virtuas has (well, had) are Jeff Genender, Matt Raible, and Bruce Snyder. Matt is famous for being the webmonkeyest person in Javaland. Matt to the average Java developer is what George Bush is to the well intentioned republican; an oafish figure that's vaguely embarrassing, but that you can relate to on some level because he's as dimwitted as you are.

Bruce jumped ship fairly early on to join a similar scam with slightly better business prospects, LogicBlaze.

Jeff on the other hand seems to have just left Virtuas to move onto bigger and better (and more commercially viable) things. How do I know? Well, Jeff had posted a superb blog entry airing some of fun stuff going on at Virtuas. Sadly, the article was yanked. I imagine the powers that be sent poor Jeff some mean letters that deprived the rest of us of his literary genius. The article wasn't even that bad, but it did point out that Virtuas spent a ton of money, and achieved nothing whatsoever. Really folks, if you're going to force someone to yank an article, someone somewhere is going to notice, and the negative publicity you get from THAT is likely far worse than if you had let the article stay and ignored it. After all, I wouldn't be writing about what turdburglaring arsebandit chozgobbling pillowbiting chocolate log miners Virtuas are if Jeff's original article wasn't yanked.

I remember being aghast actually a year or so ago when I first heard of Virtuas. The word on the street was that basically, the principals had managed to scam some Indian fund (the native kind, not the Asian kind) and that they had, in effect, unlimited funds to spooge over and so could do stupid shit like throw hundreds of thousands of dollars at JavaOne parties that generated no income or leads whatsoever.

Honestly, when I hear about these people, I'm tempted every now and then to draw a big chart outlining all the incestuous relationships between all these companies. It's like one of those mob charts you see in cop movies. We have Gluecode, which was headed by Winston. Winston has a Philippines based sweatshop where most of the coding is done. He sells Gluecode, and is immediately astounded at his own business acumen. He goes on to form Mergere and Logicblaze through some more money he's scammed, in return for gobs of shares. The rest of the gluecode people disperse to one of those two companies or IBM. On the periphery are people like Bruce and Jeff, who go on to form Virtuas, but also to work on Geronimo. Geronimo uses a bunch of products which happened to be commercialised by Mergere and LogicBlaze. Employees between all of these companies seem to happily hop between them, with IBM at various points no doubt injecting money to keep the whole thing sustainable. Matt Raible grins idiotically throughout the whole thing.

Another brilliant example of the OSS marketing approach is Terracotta, who have recently open sourced their flagshit product. Now, let me ask you this, if you had a product that cost 10k/cpu, and you had a ton of customers, why would you opensource it? The answer is actually pretty simple. You'd do so if the loss of income from license purchases was insignificant, which is the case if in fact you DIDN'T have many customers.

Terracotta also has a fantastic burn rate. Fancy offices, highly competent developers that likely cost a fair few limbs, and one of the stupidest marketing plans ever seen outside of Virtuas. Ignore all the spin, the OSS gesture is a last gasp effort, which might or might not pay off. Don't make the mistake of assuming it's a choice they had to make; it was a choice forced on them to try and salvage something out of what they've built. Ultimately, choosing terracotta or coherence boils down to whether you are building toys, or enterprise applications, at least for the near future.

So do yourselves a favour and avoid all these companies, their shelf life is highly suspect, and they're all a bit too blatant about trying to scam you. (2006-12-26 08:57:43.0) Permalink Comments [23]

Wednesday November 15, 2006

GPL Java: who cares? Sun dual-licensing Java under the GPL is a pretty canny marketing move. Does it have any actual impact on any Java developers? Contrary to what every other blog will tell you, nope, it won't.

Believe it or not, you've been able to download the JDK and 'hack' on it for a while now. Some people have even blogged about getting their patches into the JDK sources, or how they got a bug fixed, and so on. There's absolutely nothing new there.

Yet every little fuck and his penis are waggling furiously about how this is a new era of something or the other. Even that perennial turdburglaring hippie Stallman is on board. How low do you have to sink to need an endorsement from him?

Kudos to Sun though for finding a new crowd to suck up to. It's taken close to 10 years for them to switch to the Slashdot cock from MS cock, but it's nice to see that they're still just as desperate to have their mouths and bottoms filled by alien objects.

At least the slashdot crowd has nothing tangible to copy beyond reams of empty promises and communist idealogies. Hopefully things like jsp (asp wannabe) and jsf (winforms wannabe) won't happen again, instead all we'll get is idiotic licenses and endless debates from non-lawyers about what legalese actually means.

What's perplexing is how ignorant most people are of what this actually entails. So many idiotic blog bleatings about how patches can be contributed, about how javac will be speeded up, about how there's going to be a 'super performance' team that churns out amazing JVMs. Yeah, just look at what a great success classpath and kaffe is, and how blazingly fast they are. The only people who now have access to the JDK sources that didn't before are these nutjobs, who in again, close to decade, have done nothing beyond waggle their fingers for endless self-administered prostate massages.

Still, plenty of silver lining. Nice to see classpath and kaffe and all those people fade into irrelevance. Nice to see another nail in Apache Harmony's coffin. Why would you go with a less idealistic and faggotarsey license, and use a product that promises everything, delivers nothing, and continues to exist purely on corporate charity?

IBM's reaction is also pretty hilarious. They're rightfully annoyed that the project isn't an Apache one. It's pretty obvious that the Java branch of Apache at least is more or less owned by IBM. If a project moves to Apache, you know there'll be some IBM people figuring out how to make money out of it and flog it to the hapless websphere molested masses. Sun keeping control means that IBM's spastic global services are going to have a hard time selling the idea that THEY in fact own Java.

So move along folks, there's no real news here. Just a clever marketing move and an appeal to the idiotic spastic masses. (2006-11-15 09:59:11.0) Permalink Comments [25]

Thursday October 26, 2006

JetBrains' TurdCity When renewing a whole bunch of IDEA licenses, I was pleasantly surprised to find out that this also gave us a bunch of free TeamCity licenses.

TeamCity is the new build server from the wise and wonderful people who brought us IDEA, so surely it'll be the first CI tool that's usable and functional.

Turns out there's some good news, and some bad news. Lets get the good news out of the way first (this won't take long). Unlike every other CI tool I've tried, TC mostly works. It's managed to build stuff (though you'll have to grab the latest EAP for that, 1.0 didn't know what to do with my build.xml). This already puts it in contention for the perplexingly small list of contenders for the coveted 'CI tool that does stuff' category. It also is free for IDEA 6.0 users (though why anyone would pay for it in its current incarnatio is a question best left unexplored.)

Sadly, that's where it ends. It looks like the people who develop this tool have a fairly serious identity crisis. Their father seems to be a desktop developer, and their mother is a cheap whore.

The traits inherited from their paternal genes are the wonderful ignorance with regards to long running processes. TC will happily run out of memory, it'll periodically hang and refuse to play until you kick it in the face, and generally performs as well as Swing 1.1 does.

From its mother, TC inherited that trashy superficial ignorant love of Ajax pioneered by ajaxian.com. That idiotic shiteating grin towards any and every usage of ajax. The goal is to arouse sexual curiosity in the developer, even if it ends up shoving a big black dildo up the ass of the user as a side effect.

In practice, this means that TC has some truly perplexing uses of Ajax. There are tabs on the top, for example. Some of which function the way that God intended (you click on a tab, you switch to that tab), and some of which mysteriously popup a ajax inline window. Huh? What school of UI design advocated THAT approach?

Similarly, when you go to view a logfile, you get the main page which shows notmuch, and you get its body loaded in via ajax. In practice, this usually means a browser hang, and more often than not, the page crapping out.

Of course, when you dig deeper in, you start to become very very scared. Little girl molested by a guy offering candy kind of afraid. Suffice to say, there is much soiled underwear at this point.

For one thing, TC decides to write to your home directory. Yes, it's a server process, but I guess they think it's a desktop app, so even if you're installing it in /usr/local, it'll try to create its db in /home/poopy. God help if you if your home dirs are on an NFS mount (where file locking just doesn't work).

Looking at the horrorshow that is the buildagent runtime, you'll note that for some reason, it specifies all this extra emma gunk (why the hell they chose such a shitty product to integrate it is beyond me) even though you don't have code coverage enabled. Looking over the environment variables it specifies is also going to cause a good deal of eye gouging/ear bleeding, depending on your disposition.

What app is complete without unhelpful error messages? TC continues the fine tradition already well established by other Java apps of providing cryptic unintelligible error messages, when things do go wrong. Fear not, things go wrong often enough that you'll soon be enjoying these messages, and puzzling over them for days to come.

The IDEA integration is nice, but of course, suffers from the same half baked approach that the whole product has. When something is being built, the icon in IDEA does not denote that, instead it shows a broken build image. Running the default inspection profile on the server on a medium-sized project results in 'too many errors', where you can't actually see the inspections in IDEA short of visiting every class, thus making it all rather pointless. 1.0 didn't even let you schedule builds (which is fixed in the latest EAP).

I hope the product gets better. I'm actually quite optimistic, as I don't see it getting much worse. There are a lot of very clever ideas in there, they're just badly done and have about as much polish as a turd. The 1.0 version is about as productive as a conversation with a JBoss employee, so avoid that at all costs, and grab the EAP. (2006-10-26 15:01:22.0) Permalink Comments [15]

Wednesday September 27, 2006

The death of Agile It's a little disturbing seeing the agile crowd at work. In a relatively short period of time, an energetic group with potentially something new to offer has quickly sunk into a oft-derided group of greedy consultant used car salesman types.

I've spoken to a lot of people about this, I have yet to meet a single person who thinks agile, as sold by the agile crowd, is a good idea. I've ensured that everyone I spoke to had at least tried it, in some form or another. Pair programming, TDD, XP, little bits of paper, incremental releases, smug turdy devs, all experimented with and eventually discarded like a used tampon. Pair programming is inefficient and wasteful when compared to individuals who don't slack off, the little cards more often than not end up with the wrong things scribbled on them, the incremental updates result in a big badly thought out ball of mud with no coherence, and the tests end up testing the wrong thing more often than not.

Is it possible to fix all that? Sure, but agile isn't the way to do it, because the practices it espouses do not lend themselves to easy adoption. It's a high barrier that continues to punish, and never rewards its participants beyond that air of smugness and that perplexing 'I just shoved a big dildo up all my orifices and its strangely alluring' look.

The reason for this disillusionment isn't that hard to find. As many have noted, it's rooted in the feeling of incredible disappointment when you realise that no time has been saved, your love life has not improved, and your customers are no happier when you follow this crap.

Genuine techies don't react well to religion, usually. The agile crowd has committed the cardinal sin of stepping over the pragmatism line into the realm of faith. We're surrounded now by the debris and detritus of less than successful agile projects. Instead of questioning the agile practices that might have contributed to the failure, agilists will instead scream out that the flaw is in the implementation, not the principles. Whatever happened to the scientific method? Why are the principles now held to be sacrosanct?

It's that sort of attitude that makes normal people think that agilists are, on the whole, a bunch of greedy fuckheaded navelgazers more intent on group teenmasturbation than concern for fellow man. The irony of their very name is becoming apparent to all; there's nothing agile about their thought processes or acceptance of external input. Either you're with us, or you're not doing agile 'properly', and you can hire us or attend our seminars for the cheap price of a few thousand dollars.

Just say no to agile. Say yes to sane practices that work for your particular need. In the real world, it's not about doing either waterfall or agile, both are silly extremes that in practice never happen. The agilists initial point is well made, one should adapt and be prepared to think outside the box. Taking that advice in earnest involves discarding the modern day agilists and their newly discovered snake oil. (2006-09-27 15:54:35.0) Permalink Comments [45]

Monday August 28, 2006

Please, no more logging! What is it with opensores and logging? I am utterly perplexed by why so many popular frameworks think that everyone cares about their startup cruft.

Maybe the whole problem can be traced to the linux kernel bootup console. We all probably remember the thrill of seeing that, of feeling faintly hackery and l33t, even though we're not kernel hackers and most of it was in fact gibberish. It didn't matter that what the messages actually said had no real impact on the end result (either the system came up or it didn't, and which one happened had nothing to do with the messages). What mattered is that it made the average nerdy fuckface feel cool and hackerish.

Fast foward a few years, and those linux chocolate log miners stumbled bleary eyed into the enterprise world and had to learn Java. Being linux communallovefaggotarse types, they had to bring in the same spirit of assholeness to everyone else.

So, what do we have as a result? We have JBoss, with 6-7 pages of output when things go well, and a (not very) abridged version of War and Peace when they (usually) do not.

We have Hibernate, which feels it's important to inform you of every time a Fleury inserts anything into Gavin's ass, or every time Christian (he's the asshole right? I always confuse him with Emmanuel, one of them is an utter turd, the other is not) defecates on a homeless person.

Not to be outdone, Spring jumps into the fray as well. Reams and reams of utterly worthless output, all about how its beans are busy molesting one another with gay abandon. Thank fuck for the full list of every worthless little bean that's defined, because, you know, it's important to be told of what your own XML files say. It's not like you wrote them or anything.

It's like there's a whole generation of developers that's developed collective amnesia over the difference between DEBUG and INFO.

I beg of all you opensores people, please please think of us users when you poop out your logging. Put messages that WE think are useful, not that you like to masturbate to. When I start up a library or framework, I expect it to start, and don't need to know the details. I want to know when things go wrong, not when things go right. Maybe being mentioned in log4j.properties files everywhere makes you feel special. The person mentioning you though is far, far more likely to think you're an inconsiderate self-aggrandising shithead.

Hell, I'm extending an offer to go over your app for free and send you a patch to remove all the suprious startup messages (hint hint, spring/hibernate people). Just holler if you're interested.

Oh and apologies for the long hiatus, I got married. (2006-08-28 17:26:50.0) Permalink Comments [43]

Monday July 31, 2006

Another googleturd Surprise surprise, google release yet another half baked idea, and techies everywhere bend over and demand that the biggest black object in sight be crammed up their orifices so they can ooh and ahh and generally behave like a bunch of desperate teenagers aching for a fisting.

This time it's google code, and I am astounded (though I should be used to this by now) that anyone in their right mind thinks that this is an improvement over anything equivalent that already exists.

Granted, it does have one tiny benefit over sourceforge, which is that you can access its source control and get up to date files. It's far worse than every other project hosting facility there is out there in every other respect (java.net now offers svn, so the svn support is not such a differentiator anymore.)

What's odd about this particular offering is that while google stuff is generally useless and good for eye candy, it's usually reasonably well executed. In this case it seems like they just rounded up a bunch of apache hippie types and let them futz about with this idea just to stop them from damaging anything important.

For a start, there are some rather important features missing that make this repository, at best, a good backup solution and nothing else. There's no file repository, so you'd have to go elsewhere to actually host your releases. In fact, there's no release mechanism or notification support either, so even if you did by some miracle get some morons interested, you have no way of keeping them updated. The list of project users is also what you'd expect an intern to come up with; it's just a list of logins, with no real names or any way of seeing who 'flibbity.gibbet' actually is, or any way of contacting said person.

Even more surprising, there isn't a single innovative feature here. What's so bad about having a nice searchbar at the top of the svn browsers, instead of the skanky faggotarse default apache svn view? Why can't I use google pages to write docs? Why can't it host my documentation for me and make it searchable via googlemagic?

Trying out this pile of worthless gunk reveals even more flaws. Really basic stuff that shows that Google apparently has a severe QA engineer shortage, or thinks that for trivial toys like this, it doesn't matter if it's halfassed. For example, if a project has 'Apache License 2.0' specified, the link doesn't go to the 2.0 license, but to the generic Apache licenses page.

This sloppiness is prevalent throughout the app. For example, all 'home page' type links go to code.google.com, but nothing pointing to the hosting home, code.google.com/hosting. You'd have to go all the way to the top, then drill down to get to the main entry point.

The form validation is also bizarrely crap. On the project creation page, the create project button is disabled unless you have a description and summary > 3 characters. All good and well, but if your project name is just one char, that's fine, the button is enabled. When you submit however, you're told that the project name is too short. What's so evil about having the same validation mechanism for all the fields? Maybe I'm a dimwitted user, but it's not entirely intuitive to me that I MUST specify a summary and description, and that they must each be more than 3 characters long.

Of course, if you're the kind of guy who likes to create dummy projects just to write bile entries about how much google sucks, then just use Safari. You can happily hit enter there and sneak in 2 character summaries if you so wish.

The issue tracker is somewhat interesting, I do like the freeform label support, but of course, for the sake of consistency with the rest of the app, it's useless for any real world projects. There's no way to add custom tags, so you can't for example add tags for your specific versions. This of course means that for every single issues posted, the first comment you'll get back from the developer is 'err, so what version is this again?'

There's also the issue of stupid defaulting in the issue tracker. I can click new issue, then click submit. There's no detection for the default content being specified, so it's very easy to spam a project with a ton of boilerplate issues.

Editing an issue is equally badly done, there's a 'Add a Comment and Make Changes' title, but all that is under it is a 'enter your comments' textfield. Where's the make changes bit? Ah, you click on the textfield, and the rest of the form magically appears! Yet more bad jarring UI.

All in all, an abysmal effort. Shame on you google, but the real blame here is for all the google fanboys who allow them to get away with such tawdry offerings. In any other company, this sort of half assed effort would never be released, and someone would be held accountable if it was. Maybe google developers pride themselves on not being in 'that sort of company', but in their place, I wouldn't feel so smug about being so sloppy. (2006-07-31 17:37:15.0) Permalink Comments [14]