My google-fu on financial info breakdowns for publicly traded companies is obviously weak, but Nintendo said they were going to spend $200M on marketing the Wii *alone*, so it's likely that Sony's advertising budget for the PS3 ALONE is on the order of the entire allocation for the DOJ's antitrust division.
Re:Lets invent yet another language!
on
Groovy in Action
·
· Score: 1
I wondered about why someone would go to all the effort of developing Groovy, and I think it comes down to this: there are Java developers who are fond of the JVM, but don't much care for Java (or at least, not in all situations; maybe static typing gets them down sometimes). Groovy (and JRuby, and Jython) gives them (let's say) dynamic typing without leaving the JVM, with its wide range of libraries and pretty damn good platform independence. You can -- in principle -- drop a few jar files onto your classpath, and access all of your Java-based infrastructure from these languages.
Groovy, since it was invented as a "scripting language for the JVM" seems less advantageous in this regard than the other two, since there's an existing pool of Ruby and Python documentation and coders who don't have as far to travel. OTOH, as such it might be a better fit for a calcified Java programmer...
A related reason is Ruby on Rails envy: there's Grails, and the JRuby gang are focusing on making sure they have RoR support. A slightly different use I can think of for these sorts of frameworks is that they give you the ability to whip up a script that generates a report, or fixes some bad data, or (etc.) and hot-deploy it to your servlet container.
My understanding is that while lots (most? alll?) of J2EE application servers use JMX or provide services based on JMX in some way, the core of JBoss's architecture is JMX, to the extent that if you're running a minimal JBoss configuration, what you've got is an MBeanServer and a few MBeans. Every other service is loaded via the services provided by that JMX 'spine.'
I want to emphasize that I know SFA about (e.g.) WebLogic and its architecture, so what I said above may well not be a unique feature of JBoss. But when I've heard JMX touted as a 'feature' of JBoss, it's about the architecture of JBoss, and not the inclusion of JMX-based services, that is supposed to be the Big Deal.
If I'm wrong about that, and as I indicated I may well be, well, then I've been astroturfed into some false beliefs.
Well, Lotus bought the parts from Toyota under a specific agreement; an exchange of money was the very nature of the transaction. Toyota's got their benefit already. But when you or I (or SuSE, or...) put together a Linux distribution, we do it with lots of GNU parts that are licensed under the GPL, and those parts weren't sold to you -- indeed the creator of the parts probably sees little or no direct financial benefit from his or her contribution.
Of course the analogy's imperfect, but suppose you built a house, and the foundation of the house came from bricks that were made by one group of people, and the framing of the house came from materials that were assembled by others. When it comes time to talk about the house you built using all those parts, the only contribution you acknowledge is that of the foundation. Is that fair That's the crux of the case for "GNU/Linux."
And to head off misunderstandings, there's nothing in the agreement there that says you MUST say it's GNU/Linux, but RMS encourages it so as to make the GNU contribution explicit, to give credit where it is due.
You're soaking in it.
(Firefox has supported Atom since at least the first full release of the RSS support; the Sage plugin also supports Atom).
Kids, Atom's not new. It's been developed by lots of smart folks.
Atom's More Than A Syndication Format
on
Atom 1.0 vs RSS 2.0
·
· Score: 5, Informative
Atom is both a syndication format and an API for creation, updating, and deletion of content. It's already in widespread use by Blogger.
What's been (all but) finalized is the syndication format (and rules for extending it). This allows the working group to firm up the details of the publishing API, which, for my money, is the real payoff with Atom.
A pretty good overview of the history of RSS and the motivations behind Atom is here.
Re:Has NetBeans 4.1 Eclipsed Eclipse?
on
Netbeans 4.1 Released
·
· Score: 2, Informative
Netbeans also has C/C++ modules. I don't know how the projects compare, but it's been there in Netbeans for quite a while.
Not that I don't also like Eclipse, but I want to note that Netbeans has had excellent XML/HTML/JSP/CSS editing capabilities since at least 3.5.
Current incarnations are really good with, e.g. JSPs: Netbeans 4.x does tag completion on custom tags, INCLUDING the URIs you need to reference in the taglib directives and the attributes of your custom tags.
Netbeans added an 'auto-import' feature in 4.x that closes the gap with Eclipse somewhat (don't know what package name a class lives in? Alt-Shift-I will bring up a list of candidates, much as with Eclipse's 'quick fix' Ctrl-1)
Netbeans 4 generates an ant build script that will load all your external libraries and 'war' them up for you (Eclipse does not do this out of the box, and I'm not sure why nobody's needed to scratch that itch yet). This means you're not tied into the IDE to build your web app.
Netbeans 4.x already has, in non-beta form, support for the new language features from JDK 1.5.
Truth be told, for developing any moderately complex web application, right now an out of the box Netbeans 4.1 is, IMO and on balance better than Eclipse with MyEclipse (which you pay $30 a year for).
Eclipse's task list and background compilation are, for me, its two best features right now. To be honest, the fact that the functionality I use all the time is available for Eclipse via plugins whose quality is not always topnotch bugs me a little bit.
The WTP will close the gap on the Servlet/JSP side, but Netbeans 4 has a *lot* going for it.
That would be *Commodore* Decker (outranked Kirk, which is why he was able to endanger the Enterprise). His son was the guy in the first ST movie that ended up merging with V'ger and whasername played by Persis Khambatta.
Only the latter two are apples to apples, really. If you mean C# vs. C++, well, considering strictly the language, C# *overall* seems like it is more productive than C++, which shouldn't be too surprising. That's not, of course, to say that you should use C# instead of C++ everywhere. On the other two issues, I can't really comment.
I don't see how refusing an onerous demand made by someone who lacked the foresight to plan for this eventuality makes you a "dick." It sounds to me like it's been a one-sided relationship for quite a while; the abuse has to end somewhere.
It would be exceedingly nice of the OP to do more than four weeks. Being "on call" at "subsidized rates" is just the boss trying to take advantage of him.
Dude, that's already been filmed, in Jay And Silent Bob Strike Back, which is clearly set in the present day (and so takes place long after the stories so far), Luke's hand grows to enormous proportions, and to hide his identity, he changes his name, and he does some pretty mean stuff.
Mozilla == the successor to Netscape Communicator. It spoke HTTP(s), SMTP, POP, IMAP, IRC and other stuff. It was the original "kitchen sink" wrapper around the Gecko HTML/XML rendering engine.
Firefox and Thunderbird were split off as standalone apps that embedded the Gecko rendering component and a few other goodies from the original Mozilla suite, but they've always been their own critters, from an application standpoint.
So, now it looks like major development on Gecko-based products is going to be on apps that do one small cluster of things well, instead of a large app that does lots of things.
The point is that the motivation to spam blogs rests on the assumption that posting links to one's site on blogs elevates the Google (Yahoo, MSN Search) rank for the sites linked to. Once that assumption is invalidated, the incentive to spam goes away. It should actually help quite a bit.
Enron investors didn't lose their investments due to any direct government or civil action directed specifically at holding them liable for the corporation's accounting practices, they lost their investments because they'd invested in a hollow shell and the hollowness was finally revealed. And why wouldn't the officers get punished severely if a system incorporating an end limited liability were put in place?
Also: it's an implementation detail of the current system that the officers of a corporation have disproportionate power (appointing their good friends as directors, etc.) to their degree of investment in the system. That needn't be the case.
In any event, a system could be set up for those who do want profit without legal liability -- something like a proxy setup, say -- where the investor accepts a lower profit share in return for being personally indemnified (the proxy, on the other hand, wouldn't enjoy such protections).
I'm just playin' with ideas here. At any rate, it strikes me that the degree of accountability investors face is too low. That goes quadruple (to make up a multiplier) for officers of LLCs.
Speaking as a programmer, a lot of that depends on the implementation. Read charitably, Badnarik is talking about more radical change than merely removing "limited liability." A restructuring of civil law would also presumably be a goal. Also, if such a radical change is effected, it is also likely that companies would be far less likely to do things that end up harming people.
And, finally, 'liability' has a civil reading and a tort-related one. Maybe if you could go to jail for being a shareholder in a miscreant corporation, you'd be much more careful about who you invest in...
The two main changes are (1) making sure it works with the upcoming JDK 5.0 (formerly 1.5) release and (2) shipping the Eclipse compiler *for compiling JSPs*. Neither of these moves are strikingly fundamental, although the second one is kind of interesting.
Building a case takes time. A suable entity, like a university, doesn't simply go off making accusations it can't prove. A better analogy than yours is a store security officer who sees Winona Ryder on the security monitors clipping the price tags off of articles of clothing on the store racks and stuffing them in her bag, then arresting her once she's on her way out the door with the merchandise in hand. That gave the security personnel everything they needed to prove intent.
It's not like the university officials did anything remotely analogous to "shooting him in the face", either. They imposed a completely legitimate penalty on him, that he already knew about. He's doing the equivalent of the burglar complaining, upon being arrested, that he shouldn't have been arrested JUST THEN because you waited too long to call the cops, thus "robbing" him of the opportunity to run out of your house before you could catch him.
As I pointed out elsewhere in the thread, you seem to assume that the university was making money by keeping this yobbo paying tuition. That's almost certainly not true.
Right, because as everybody knows, (a) the first time you cheat, it's immediately known (b) it's always a good idea to accuse somebody of something unethical on the merest suspicion, and finally, (c) all public universities make money off of their students through tuition, that undergraduate tuition is every university's biggest cash cow. Especially in the UK. Yeah, providing the infrastructure for an undergraduate, paying all of his instructors, etc., yeah, that's sure covered by (roughly) 1800 dollars. Why, I bet Mr. Nightingale probably sleeps naked on the pile of money he's sucked out of Michael Gunn.
However, the problem is not so much with MySQL itself (it's a good, fast, lightweight storage system for simple and small amounts of data). It's with the perception that MySQL is every bit as good as a more robust engine (Oracle, MSSQL, DB2, take your pick) for any application. That is definitely not the case.
In all seriousness, how widespread is this perception? I mean, do you have an example in hand of someone saying they'd use MySQL for an FDIC insured bank's transaction records? For all the reaming those who defend MySQL get (and I'm not saying you're in on that), only a tiny minority who seem to be actually serious about it seem to be saying it's "every bit as good" as [ commercial products ]. Scratch most of the "zealots", and you will find someone saying (e.g.) "well, yes, it doesn't do subselects but I don't use those often and you can always do another select." They won't be as nice about it, of course, but that's the content of what they're saying.
The most common defense of MySQL I see given is like that given in the post you're reacting to:
The features of [ commercial product ] over and above those provided by MySQL are not worth the extra cost / hassle.
JSP is not *designed* to produce crappy code, even if a lot of people produce crappy code with it. Hell, the same goes for PHP, although the latter's genesis as an HTML-embedded language does, I believe, expose it to that sort of charge.
There are reasons for servlet forwarding and tag libraries; they let you do the heavy lifting in servlets and tag handlers, and just handle display in JSPs.
I say this as the author of several JSPs (some deployed in production on reasonably high-profile sites) with buttloads of spaghetti-code, static Java blocks, and their own methods(!). I did it all out of ignorance of the true power of the tools the Serlvet spec designers put in my hands. It's an education issue more than a spec design issue.
Of course, there is that undercurrent among Java developers that suggests if the language permits a certain practice, it encourages it. That's what you get for designing a "B&D" language =)
Well... he saw what relativity predicted you would see. One good explanation for this result is that relativity is a correct description of what's going on in the universe. But that's only one explanation. A natural one, to be sure, and his experiment certainly counts as a nice test of the empirical consequences of relativity, but it's a defeasible inference from Eddington's results to the claim that gravity causes light to bend.
Slashdot Mirror
In all likelihood, buckets of money. Compare MS' or Sony's ADVERTISING budget to the ENTIRE budget allocated to the DOJ's antitrust division:
My google-fu on financial info breakdowns for publicly traded companies is obviously weak, but Nintendo said they were going to spend $200M on marketing the Wii *alone*, so it's likely that Sony's advertising budget for the PS3 ALONE is on the order of the entire allocation for the DOJ's antitrust division.
I wondered about why someone would go to all the effort of developing Groovy, and I think it comes down to this: there are Java developers who are fond of the JVM, but don't much care for Java (or at least, not in all situations; maybe static typing gets them down sometimes). Groovy (and JRuby, and Jython) gives them (let's say) dynamic typing without leaving the JVM, with its wide range of libraries and pretty damn good platform independence. You can -- in principle -- drop a few jar files onto your classpath, and access all of your Java-based infrastructure from these languages.
...
Groovy, since it was invented as a "scripting language for the JVM" seems less advantageous in this regard than the other two, since there's an existing pool of Ruby and Python documentation and coders who don't have as far to travel. OTOH, as such it might be a better fit for a calcified Java programmer
A related reason is Ruby on Rails envy: there's Grails, and the JRuby gang are focusing on making sure they have RoR support. A slightly different use I can think of for these sorts of frameworks is that they give you the ability to whip up a script that generates a report, or fixes some bad data, or (etc.) and hot-deploy it to your servlet container.
My understanding is that while lots (most? alll?) of J2EE application servers use JMX or provide services based on JMX in some way, the core of JBoss's architecture is JMX, to the extent that if you're running a minimal JBoss configuration, what you've got is an MBeanServer and a few MBeans. Every other service is loaded via the services provided by that JMX 'spine.'
I want to emphasize that I know SFA about (e.g.) WebLogic and its architecture, so what I said above may well not be a unique feature of JBoss. But when I've heard JMX touted as a 'feature' of JBoss, it's about the architecture of JBoss, and not the inclusion of JMX-based services, that is supposed to be the Big Deal.
If I'm wrong about that, and as I indicated I may well be, well, then I've been astroturfed into some false beliefs.
Well, Lotus bought the parts from Toyota under a specific agreement; an exchange of money was the very nature of the transaction. Toyota's got their benefit already. But when you or I (or SuSE, or ...) put together a Linux distribution, we do it with lots of GNU parts that are licensed under the GPL, and those parts weren't sold to you -- indeed the creator of the parts probably sees little or no direct financial benefit from his or her contribution.
Of course the analogy's imperfect, but suppose you built a house, and the foundation of the house came from bricks that were made by one group of people, and the framing of the house came from materials that were assembled by others. When it comes time to talk about the house you built using all those parts, the only contribution you acknowledge is that of the foundation. Is that fair That's the crux of the case for "GNU/Linux."
And to head off misunderstandings, there's nothing in the agreement there that says you MUST say it's GNU/Linux, but RMS encourages it so as to make the GNU contribution explicit, to give credit where it is due.
You're soaking in it. (Firefox has supported Atom since at least the first full release of the RSS support; the Sage plugin also supports Atom). Kids, Atom's not new. It's been developed by lots of smart folks.
Atom is both a syndication format and an API for creation, updating, and deletion of content. It's already in widespread use by Blogger.
What's been (all but) finalized is the syndication format (and rules for extending it). This allows the working group to firm up the details of the publishing API, which, for my money, is the real payoff with Atom.
A pretty good overview of the history of RSS and the motivations behind Atom is here.
Netbeans also has C/C++ modules. I don't know how the projects compare, but it's been there in Netbeans for quite a while.
Not that I don't also like Eclipse, but I want to note that Netbeans has had excellent XML/HTML/JSP/CSS editing capabilities since at least 3.5. Current incarnations are really good with, e.g. JSPs: Netbeans 4.x does tag completion on custom tags, INCLUDING the URIs you need to reference in the taglib directives and the attributes of your custom tags. Netbeans added an 'auto-import' feature in 4.x that closes the gap with Eclipse somewhat (don't know what package name a class lives in? Alt-Shift-I will bring up a list of candidates, much as with Eclipse's 'quick fix' Ctrl-1) Netbeans 4 generates an ant build script that will load all your external libraries and 'war' them up for you (Eclipse does not do this out of the box, and I'm not sure why nobody's needed to scratch that itch yet). This means you're not tied into the IDE to build your web app. Netbeans 4.x already has, in non-beta form, support for the new language features from JDK 1.5. Truth be told, for developing any moderately complex web application, right now an out of the box Netbeans 4.1 is, IMO and on balance better than Eclipse with MyEclipse (which you pay $30 a year for). Eclipse's task list and background compilation are, for me, its two best features right now. To be honest, the fact that the functionality I use all the time is available for Eclipse via plugins whose quality is not always topnotch bugs me a little bit. The WTP will close the gap on the Servlet/JSP side, but Netbeans 4 has a *lot* going for it.
That would be *Commodore* Decker (outranked Kirk, which is why he was able to endanger the Enterprise). His son was the guy in the first ST movie that ended up merging with V'ger and whasername played by Persis Khambatta.
Only the latter two are apples to apples, really. If you mean C# vs. C++, well, considering strictly the language, C# *overall* seems like it is more productive than C++, which shouldn't be too surprising. That's not, of course, to say that you should use C# instead of C++ everywhere. On the other two issues, I can't really comment.
I don't see how refusing an onerous demand made by someone who lacked the foresight to plan for this eventuality makes you a "dick." It sounds to me like it's been a one-sided relationship for quite a while; the abuse has to end somewhere. It would be exceedingly nice of the OP to do more than four weeks. Being "on call" at "subsidized rates" is just the boss trying to take advantage of him.
Dude, that's already been filmed, in Jay And Silent Bob Strike Back, which is clearly set in the present day (and so takes place long after the stories so far), Luke's hand grows to enormous proportions, and to hide his identity, he changes his name, and he does some pretty mean stuff.
Mozilla == the successor to Netscape Communicator. It spoke HTTP(s), SMTP, POP, IMAP, IRC and other stuff. It was the original "kitchen sink" wrapper around the Gecko HTML/XML rendering engine.
Firefox and Thunderbird were split off as standalone apps that embedded the Gecko rendering component and a few other goodies from the original Mozilla suite, but they've always been their own critters, from an application standpoint.
So, now it looks like major development on Gecko-based products is going to be on apps that do one small cluster of things well, instead of a large app that does lots of things.
clear 'nuff?
That may be, but to be completely humorless, it's shorthand for MAJOR_VERSION.MINOR_VERSION It's not a decimal number.
I dunno, but why don't we start with Richard Nixon's list and work from there.
Your ass is mine, Paul Newman.
The point is that the motivation to spam blogs rests on the assumption that posting links to one's site on blogs elevates the Google (Yahoo, MSN Search) rank for the sites linked to. Once that assumption is invalidated, the incentive to spam goes away. It should actually help quite a bit.
Enron investors didn't lose their investments due to any direct government or civil action directed specifically at holding them liable for the corporation's accounting practices, they lost their investments because they'd invested in a hollow shell and the hollowness was finally revealed. And why wouldn't the officers get punished severely if a system incorporating an end limited liability were put in place?
Also: it's an implementation detail of the current system that the officers of a corporation have disproportionate power (appointing their good friends as directors, etc.) to their degree of investment in the system. That needn't be the case.
In any event, a system could be set up for those who do want profit without legal liability -- something like a proxy setup, say -- where the investor accepts a lower profit share in return for being personally indemnified (the proxy, on the other hand, wouldn't enjoy such protections).
I'm just playin' with ideas here. At any rate, it strikes me that the degree of accountability investors face is too low. That goes quadruple (to make up a multiplier) for officers of LLCs.
Speaking as a programmer, a lot of that depends on the implementation. Read charitably, Badnarik is talking about more radical change than merely removing "limited liability." A restructuring of civil law would also presumably be a goal. Also, if such a radical change is effected, it is also likely that companies would be far less likely to do things that end up harming people. And, finally, 'liability' has a civil reading and a tort-related one. Maybe if you could go to jail for being a shareholder in a miscreant corporation, you'd be much more careful about who you invest in ...
The two main changes are (1) making sure it works with the upcoming JDK 5.0 (formerly 1.5) release and (2) shipping the Eclipse compiler *for compiling JSPs*. Neither of these moves are strikingly fundamental, although the second one is kind of interesting.
Hmm, which might explain why I movies produced by Jerry Bruckhemier (who hires directors that love to move that camera around a lot) make me sick ... ;)
Building a case takes time. A suable entity, like a university, doesn't simply go off making accusations it can't prove. A better analogy than yours is a store security officer who sees Winona Ryder on the security monitors clipping the price tags off of articles of clothing on the store racks and stuffing them in her bag, then arresting her once she's on her way out the door with the merchandise in hand. That gave the security personnel everything they needed to prove intent.
It's not like the university officials did anything remotely analogous to "shooting him in the face", either. They imposed a completely legitimate penalty on him, that he already knew about. He's doing the equivalent of the burglar complaining, upon being arrested, that he shouldn't have been arrested JUST THEN because you waited too long to call the cops, thus "robbing" him of the opportunity to run out of your house before you could catch him.
As I pointed out elsewhere in the thread, you seem to assume that the university was making money by keeping this yobbo paying tuition. That's almost certainly not true.
OTOH, nice troll.
In all seriousness, how widespread is this perception? I mean, do you have an example in hand of someone saying they'd use MySQL for an FDIC insured bank's transaction records? For all the reaming those who defend MySQL get (and I'm not saying you're in on that), only a tiny minority who seem to be actually serious about it seem to be saying it's "every bit as good" as [ commercial products ]. Scratch most of the "zealots", and you will find someone saying (e.g.) "well, yes, it doesn't do subselects but I don't use those often and you can always do another select." They won't be as nice about it, of course, but that's the content of what they're saying.
The most common defense of MySQL I see given is like that given in the post you're reacting to:
The features of [ commercial product ] over and above those provided by MySQL are not worth the extra cost / hassle.
JSP is not *designed* to produce crappy code, even if a lot of people produce crappy code with it. Hell, the same goes for PHP, although the latter's genesis as an HTML-embedded language does, I believe, expose it to that sort of charge.
There are reasons for servlet forwarding and tag libraries; they let you do the heavy lifting in servlets and tag handlers, and just handle display in JSPs.
I say this as the author of several JSPs (some deployed in production on reasonably high-profile sites) with buttloads of spaghetti-code, static Java blocks, and their own methods(!). I did it all out of ignorance of the true power of the tools the Serlvet spec designers put in my hands. It's an education issue more than a spec design issue.
Of course, there is that undercurrent among Java developers that suggests if the language permits a certain practice, it encourages it. That's what you get for designing a "B&D" language =)
Well ... he saw what relativity predicted you would see. One good explanation for this result is that relativity is a correct description of what's going on in the universe. But that's only one explanation. A natural one, to be sure, and his experiment certainly counts as a nice test of the empirical consequences of relativity, but it's a defeasible inference from Eddington's results to the claim that gravity causes light to bend.
Why yes, I am a pedant.