The software circles of hell: Java, NetBeans, Maven, Tomcat, and about 50,000 vaguely-named Java libraries
-
They used to have their own shitty search function that never seemed to find anything. It looks like they use Bing now.
-
@blakeyrat said:
Look, maybe my criteria for "doesn't work" is stricter than yours. I'd say a text editor where double-clicking a word doesn't select the word 100% of the time "doesn't work". I'd say a web server that returns 404 codes for files that exist on disk "doesn't work". If you disagree with those two assertions, you're welcome to disagree. But I still think you're an idiot who uses these shitty tools that don't work.
Maybe your problems aren't my problems. Maybe some things (like your double-clicking issues) don't happen to me. Maybe other things (Tomcat 404) are eventually diagnosed (assuming I take the time to do so, instead of bawling that nothing works) as bad configuration. And maybe some things (like specific fonts not rendering at the right DPI) just don't get in the way of me doing any real work. Maybe the shitty tools I use work fine for me. Maybe you just can't figure out how to get them to work for you.@blakeyrat said:
Edit: Oh and I will admit that my experience was at least partially due to a bad Maven configuration, because after the project file was updated in source control, Tomcat worked much more reliably.
So it turns out your config was what was fucked?I'll see myself out.
-
@boog said:
Maybe your problems aren't my problems.
My personal theory is that they are you're problems, but you just don't give a shit about software quality. Or at least not much of one.
Just like Linux users will grab their laptop and say, "see? I put Linux on it and everything works!" but what they really mean is, "everything I care about works", which is a completely different claim. (You can usually defeat those idiots by simple reaching out and hitting the "mute" button on the keyboard, which almost invariably does not work.)
Hell, there are Lotus Notes developers who will claim, to their dying day, that it's an excellent development environment. Delusion is endemic in this profession.
@boog said:
And maybe some things (like specific fonts not rendering at the right DPI) just don't get in the way of me doing any real work.
And yet, it's still broken beyond belief. It's unacceptable for a text editor in the year 2011 to be unable to correctly display fonts. Hell, that was unacceptable in 1990. The difference between me and you is:
1) I know what "unacceptable quality" is
2) I actually care@boog said:
Maybe you just can't figure out how to get them to work for you.
If I have to figure it out, if "working" is not the default case, then the tool is broken.
@boog said:
@blakeyrat said:
Edit: Oh and I will admit that my experience was at least partially due to a bad Maven configuration, because after the project file was updated in source control, Tomcat worked much more reliably.
So it turns out your config was what was fucked?I'd be lying if I said I was an expert in Maven, but no. It turns out that the 404 errors were the result of a bad config, apparently. (Either that, or Tomcat just needed a reboot of my computer to work correctly-- might have had nothing to do with the configuration at all.) I know reading is hard, but try to bear with me here.
However, the "the .JS files don't get updated about a third of the time" problem remains. Which is still a problem, it's just not affecting my job as severely as the before-mentioned problems.
The real problem is that a build system works where the build is non-deterministic! Shouldn't the .JS files either never be copied, or always be copied? What gigantic WTFs are in the foundations of Maven that allow them to be copied sometimes and not other times? (I wager this is the result of overzealous threading-- the file copy happens before the .JS files are ready to be copied.)
-
@blakeyrat said:
My personal theory is that they are you're problems, but you just don't give a shit about software quality. Or at least not much of one.
Just like Linux users will grab their laptop and say, "see? I put Linux on it and everything works!" but what they really mean is, "everything I care about works", which is a completely different claim. (You can usually defeat those idiots by simple reaching out and hitting the "mute" button on the keyboard, which almost invariably does not work.)
You're absolutely spot on. This is why I'm convinced that Windows doesn't work, because, in fact, the volume buttons on my laptop work much better when it's running Linux than Windows 7. I totally blame those button down fat cat Microsoft developers who obviously don't care about the sort of shit they push out the door.
@blakeyrat said:
The real problem is that a build system works where the build is non-deterministic! Shouldn't the .JS files either never be copied, or always be copied? What gigantic WTFs are in the foundations of Maven that allow them to be copied sometimes and not other times? (I wager this is the result of overzealous threading-- the file copy happens before the .JS files are ready to be copied.)
I've never used maven, but ant doesn't have a problem with hot deployments like what you seem to be describing. OK, maybe not like this, since I don't really know how similar our situations are. In development, I use an exploded war file, since it makes hot deployment of updated xhtml and js and css easy. My browser, however, loves to cache the old js and css files, and sometimes I forget to do a full refresh to get the new stuff.
-
@boomzilla said:
@blakeyrat said:
My personal theory is that they are you're problems, but you just don't give a shit about software quality. Or at least not much of one.
Just like Linux users will grab their laptop and say, "see? I put Linux on it and everything works!" but what they really mean is, "everything I care about works", which is a completely different claim. (You can usually defeat those idiots by simple reaching out and hitting the "mute" button on the keyboard, which almost invariably does not work.)
You're absolutely spot on. This is why I'm convinced that Windows doesn't work, because, in fact, the volume buttons on my laptop work much better when it's running Linux than Windows 7. I totally blame those button down fat cat Microsoft developers who obviously don't care about the sort of shit they push out the door.
:-) Oh, I'd go one step further. I blame Bill Gates personally. That man doesn't give a rat's ass about quality. What a jerk. Just because those volume buttons on your laptop don't work.
-
@blakeyrat said:
@boog said:
Um, no, they're not. Insist all you want, but I'm still not experiencing all the errors and faulty behavior that you are.Maybe your problems aren't my problems.
My personal theory is that they are you're problems...Maybe you missed the point I was trying to make in the above statement, but then you said it yourself: reading is hard.
-
@JasonSykes said:
@boomzilla said:
@blakeyrat said:
My personal theory is that they are you're problems, but you just don't give a shit about software quality. Or at least not much of one.
Just like Linux users will grab their laptop and say, "see? I put Linux on it and everything works!" but what they really mean is, "everything I care about works", which is a completely different claim. (You can usually defeat those idiots by simple reaching out and hitting the "mute" button on the keyboard, which almost invariably does not work.)
You're absolutely spot on. This is why I'm convinced that Windows doesn't work, because, in fact, the volume buttons on my laptop work much better when it's running Linux than Windows 7. I totally blame those button down fat cat Microsoft developers who obviously don't care about the sort of shit they push out the door.
:-) Oh, I'd go one step further. I blame Bill Gates personally. That man doesn't give a rat's ass about quality. What a jerk. Just because those volume buttons on your laptop don't work.
You make a good point. But again, you're not going far enough. In case you still think that the proprietary software model is worthy, I give you...the null empty string!
-
@boomzilla said:
You're absolutely spot on. This is why I'm convinced that Windows doesn't work, because, in fact, the volume buttons on my laptop work much better when it's running Linux than Windows 7.
I'm sure you're trying to be sarcastic or whatever, but yes: I agree, Windows is broken on your laptop.
@boomzilla said:
My browser, however, loves to cache the old js and css files, and sometimes I forget to do a full refresh to get the new stuff.
I already added anti-caching to the JavaScript file's URL. It just genuinely does not copy the updated .JS files to the server.
@boog said:
Um, no, they're not. Insist all you want, but I'm still not experiencing all the errors and faulty behavior that you are.
Ok, let me try to rephrase it. I can't trust you when you say "it all works" because I can't be sure you even have the ability to know when something is broken.
It's the difference between looking at NetBeans' incorrect font rendering and saying, "hm, the font rendering is wrong but I can still do my job." And looking at it and saying, "this font rendering looks fine." I don't know which of those two people you are.
Like I said above, there are people who genuinely believe that Lotus Notes is a quality product. For all I know, you're one of those types of people. (Except with NetBeans/Maven/Tomcat instead of with Notes.)
-
@blakeyrat said:
@boomzilla said:
You're absolutely spot on. This is why I'm convinced that Windows doesn't work, because, in fact, the volume buttons on my laptop work much better when it's running Linux than Windows 7.
I'm sure you're trying to be sarcastic or whatever, but yes: I agree, Windows is broken on your laptop.
@boomzilla said:
My browser, however, loves to cache the old js and css files, and sometimes I forget to do a full refresh to get the new stuff.
I already added anti-caching to the JavaScript file's URL. It just genuinely does not copy the updated .JS files to the server.
@boog said:
Um, no, they're not. Insist all you want, but I'm still not experiencing all the errors and faulty behavior that you are.
Ok, let me try to rephrase it. I can't trust you when you say "it all works" because I can't be sure you even have the ability to know when something is broken.
It's the difference between looking at NetBeans' incorrect font rendering and saying, "hm, the font rendering is wrong but I can still do my job." And looking at it and saying, "this font rendering looks fine." I don't know which of those two people you are.
Like I said above, there are people who genuinely believe that Lotus Notes is a quality product. For all I know, you're one of those types of people. (Except with NetBeans/Maven/Tomcat instead of with Notes.)
So, you feel comfortable telling someone else that Windows is broken on their laptop... but you can't imagine that maybe Maven, Tomcat, NetBeans (and I don't know how many other products you've been yelling about) are also simply broken on your laptop and not generically broken everywhere?
Got to admit, I'm following this discussion and occasionally commenting because I find you to be such an interesting variety of a COGI (common or garden idiot).
-
@JasonSykes said:
COGI (common or garden idiot)
The fuck?
Look, call me an idiot if you want, but at least I can hold my head high and say, "I've never released a text editor that can't render fonts correctly!"
-
@JasonSykes said:
So, you feel comfortable telling someone else that Windows is broken on their laptop... but you can't imagine that maybe Maven, Tomcat, NetBeans ... are also simply broken on your laptop and not generically broken everywhere?
You don't grasp the concept that saying Windows is broken on someone's laptop would mean that it is generically broken everywhere? Let me point you to the big distinguisher that you appear to be trying to call blakey on:
@JasonSykes said:
@blakeyrat said:
It's the difference between looking at NetBeans' incorrect font rendering and saying, "hm, the font rendering is wrong but I can still do my job." And looking at it and saying, "this font rendering looks fine."
What is being said here is that there are two kinds of wrong, blakey tend to take the first type and go into entertaining blakeyrants. Some people seem to take offense to the amount of vitriol that he spits out as it is still "good enough" in their mind (meets the "I can still do my job" clause). However due to trolling (or possibly poor communications skills) they tend to come across as the second type in the looking at font rendering example. I'm guessing that you fall into the "I can still do my job" category of wrongness reaction (which is probably the healthier camp to be in), but you are coming across as the other.
DELETED OFF TOPIC RAMBLE-Y BIT
-
@blakeyrat said:
Look, call me an idiot if you want, but at least I can hold my head high and say, "I've never released a text editor that can't render fonts correctly!"
To be fair, though, that's only interesting paired with the statistic of how many text editors you've released that do render fonts correctly.
-
@boomzilla said:
@blakeyrat said:
Look, call me an idiot if you want, but at least I can hold my head high and say, "I've never released a text editor that can't render fonts correctly!"
To be fair, though, that's only interesting paired with the statistic of how many text editors you've released that do render fonts correctly.
I should have pre-empted that snark. One.
-
@blakeyrat said:
@boomzilla said:
Hmmm....@blakeyrat said:
Look, call me an idiot if you want, but at least I can hold my head high and say, "I've never released a text editor that can't render fonts correctly!"
To be fair, though, that's only interesting paired with the statistic of how many text editors you've released that do render fonts correctly.
I should have pre-empted that snark. One.
@http://blakeyrat.com/jamespad/ said:
WHAT!!! <mock outrage> It can't print? No drag and drop support? What is this, the 1980s?Known Issues:
- No printing support
- No drag&drop support
@http://blakeyrat.com/jamespad/ said:
Download it and give it a try. It’s under the MIT license, so do whatever the hell you want with it– source is available.
Oh great! I suppose he thinks that being open source is some kind of excuse for producing shitty non-functional software? Demand better, people, Christ!
-
@DaveK said:
Oh great! I suppose he thinks that being open source is some kind of excuse for producing shitty non-functional software? Demand better, people, Christ!
I hope you're joking and that you don't need me to explain the difference between "pretend that everything works like a boss" vs. "I know there are some issues", right?
-
@blakeyrat said:
None of this shit (the shit in the subject line there) works. None of it. Maaaybe Java, if sane developers used it. (Java's shit for a desktop app, but for a web server its disadvantages pretty much disappear and it's not nearly as awful.)
NetBeans:
* Double-clicking to select a word works maybe 50% of the time
* The Build icon doesn't grey-out after you press it, meaning at one point I accidentally had three simultaneous build processes going
* The aforementioned "it can't render fonts correctly", and "it gives vague, stupid Subversion errors for no reason"Maven:
* About a third of the time after a build, I find it hasn't bothered to copy over the most current .js files.
* After even slightly touching my build profile, it decides it has to re-download all (seemingly) 36 terabytes of libraries Java developers apparently think they need to build a website
* Maven and Tomcat aren't talking to each other in some way that boggles my mind, see belowTomcat:
* Loves to return 404 errors when I know *know* the file exists, almost always after rebuilding the project
* So I've adopted the following rebuild process:
1) rebuild
2) bounce Tomcat
3) rebuild again
but despite doing that, it *still* returns a 404 error maybe 25-33% of the time (and Maven decides not to update my JS files a similar percentage of the time) and I have to do it again
* Pops up an error every time I log in, when its little system tray app fails to start the Tomcat service-- apparently they never bothered to test with a non-Admin account, or on a computer with UAC enabledThis is the only office in our company doing web dev in JSP. Why would anybody pick this shit software stack? Is it like Stockholm Syndrome or something?
Sorry you are having a bad day. For what's worth, I've worked with complex maven builds for almost 5 years now without a problem. Ant builds since 2001. We are talking builds that, for example, builds (incremental or full) wars and ears to then deploys them remotely (as in another data center across the continent), as well as initiating database deployment scripts, restarting the servers (also remotely), all without any problems. Builds that perform unit tests. Builds that fetch static content from a content respository server (like one running with TeamSite) to deploy them either on application servers or on HTTP server front ends. Builds for Tomcat, JBoss and WebLogic (and ATG Dynamo many, many, many moons ago) on Solaris, Linux, HP UX and Windows environments.
Heck, I've never done this myself, but I've seen colleagues using builds that actually roll-back (also remotely) in case of a failure. I'm not going to speculate why you are having problems. I can only tell you why I use these stacks you are having problems with, and how I (and many, many others) have been successful with them. In the most successful environment I worked with, we pulled daily builds for about a dozen different systems, some running on Tomcat, others on WebLogic, with deployments for dev, UAT, Systems Integration, pre-production and production environments.
I wouldn't use Maven if I had my pick (too much under-the-hood magic compared to Ant), but it is what it is, and it is something that I can get to work with a reasonable expectation of success and cost-effectiveness.
As for NetBeans, I've only used it casually, so I cannot say I've encountered the problems you are having. But I've used its Matise GUI tool for real-work GUI development without a problem.
As for Eclipse, I've been using it for a while now, for Java, Groovy, Spring Framework, SOA/Web Services and C/C++ development (lately just for C/C++ and CORBA development with different tool chains.) Never had a problem with it. I've used it very little for JSF development, so I cannot attest to its quality for JSF development.
In reference to your comment of having never seen an Eclipse-based product that works, I've never used Aptana, and the UML-tools in Eclipse are in their infancy (in particular the Papyrus distro).
However, I know that Omondo works well. I've also worked with Rational Enterprise Architect (based on Eclipse) that works like a charm. I've used an Eclipse-based rich admin interface to a OMG Distributed Data Service for Real-Time Systems, and I never saw a problem with it either. I know for a fact of people successfully using Eclipse Modeling toolkits for building domain specific languages as well.
I've also worked with MS tools and environments, and I think that the MS stack is absolutely fine for the jobs in the hands of competent developers. And this holds true for a Java stack, or a Python or Ruby stack. Same with a PHP stack, but for some reasons, there is a greater tendency for shitty developers to gravitate towards PHP.
I've seen complete wtf'esque train wrecks (developmental and operational) with a Java stack (and with a MS stack). And I've also seen very complex projects successfully executed with it.You would not see large financial institutions, Google, eBay, Amazon and IBM using a Java/JVM stack if it was so totally a bag of shit, wouldn't they? I know, rhetorical question, but it is one worth considering.
So that has been my experience with Java, Tomcat, Ant, Maven and Eclipse. Not trying to convince anyone, but simply stating facts that I've observed throughout my career. YMMV obviously, so you have to ask yourself, objectively, why you are having so much trouble with this specific stack.
-
@C-Octothorpe said:
@DaveK said:
Oh great! I suppose he thinks that being open source is some kind of excuse for producing shitty non-functional software? Demand better, people, Christ!
I hope you're joking and that you don't need me to explain the difference between "pretend that everything works like a boss" vs. "I know there are some issues", right?Pretty sure that was meant more in a calling blakey a hypocrite way. Of course this means that we will probably get a Whitman quote from blakey in response (I think that's the one he uses the most, or maybe it was someone else).
-
@locallunatic said:
I was just giving him a gentle ribbing.@C-Octothorpe said:
@DaveK said:
Oh great! I suppose he thinks that being open source is some kind of excuse for producing shitty non-functional software? Demand better, people, Christ!
I hope you're joking and that you don't need me to explain the difference between "pretend that everything works like a boss" vs. "I know there are some issues", right?Pretty sure that was meant more in a calling blakey a hypocrite way. Of course this means that we will probably get a Whitman quote from blakey in response (I think that's the one he uses the most, or maybe it was someone else).
Next: Blakeyrat attempts self-justification by:
1) in a single post, saying that software from him is great, as long as you don't download the software that is shit
2) implying that my bad experience is my own fault for downloading the shit software from an otherwise great coder
3) dispassionately discusses this as if he has absolutely no problem with the software being shit because, after-all, he knows the "secret" to find the non-shit software, so it's not his problem, right?
-
@luis.espinal said:
stack
You mention this word 10 times in your post, FYI
Clearly there is something wrong with you.
-
@luis.espinal said:
(extremely long and boring boot-licking of Java that is "not trying to convince anyone")
Yeah but you're a cretin who quotes the entire OP, so what do you know.
@luis.espinal said:
YMMV obviously, so you have to ask yourself, objectively, why
you are having so much trouble with this specific stack.Gee, I dunno. Maybe because it doesn't work, like I've said in this thread 47 times already? Or are we going back to "the font rendering is broken because I don't know enough about Java" with this argument?
-
@blakeyrat said:
@luis.espinal said:
(extremely long and boring boot-licking of Java that is "not trying to convince anyone")
Yeah but you're a cretin who quotes the entire OP, so what do you know.
@luis.espinal said:
YMMV obviously, so you have to ask yourself, objectively, why
you are having so much trouble with this specific stack.Gee, I dunno. Maybe because it doesn't work, like I've said in this thread 47 times already? Or are we going back to "the font rendering is broken because I don't know enough about Java" with this argument?
Ok, whatever you say boss.
-
@luis.espinal said:
STFYSorry you are having a bad day.
I've used those products a lot.
Not trying to convince anyone, but simply stating facts that I've observed throughout my career. YMMV obviously, so you have to ask yourself, objectively, why you are having so much trouble with this specific stack.
-
@blakeyrat said:
The difference between me and you is: I know what "unacceptable quality" is
What, is your opinion now the global reference standard for what is considered "acceptable" versus "unacceptable" software quality? When did that happen? I must have missed the press release...
-
@DaveK said:
Hey, blakeyrants - want some help on printer support? It's something I'm quite qualified to do.@http://blakeyrat.com/jamespad/ said:
WHAT!!! <mock outrage> It can't print? No drag and drop support? What is this, the 1980s?Known Issues:
- No printing support
- No drag&drop support
@DaveK said:
Oh great! I suppose he thinks that being open source is some kind of excuse for producing shitty non-functional software? Demand better, people, Christ!
Hey, look, opensores software that doesn't include developers overreaching their own talentsto do things they aren't really qualified to do.
-
@blakeyrat said:
... However, the "the .JS files don't get updated about a third of the time" problem remains. Which is still a problem, it's just not affecting my job as severely as the before-mentioned problems.
The real problem is that a build system works where the build is non-deterministic! Shouldn't the .JS files either never be copied, or always be copied? ...
I don't know anything about Maven, but I do know that Visual Studio's build system has options other than "Never" or "Always". Maybe you just never learned how to use them.
In particular, the first thing I would check is whether it's deciding whether to copy or not based on file modification times, and whether clock skew between the two computers is a factor.
-
@pjt33 said:
I don't know anything about Maven, but I do know that Visual Studio's build system has options other than "Never" or "Always". Maybe you just never learned how to use them.
Objection! Relevance?
Is your point there, "Maven isn't buggy because Visual Studio has options that might also exhibit the same bugs (but I haven't bothered to check whether they do) and since I know that argument's weaksauce I'm going to call you ignorant too?"
@pjt33 said:
In particular, the first thing I would check is whether it's deciding whether to copy or not based on file modification times, and whether clock skew between the two computers is a factor.
What second computer? Now you're making up an entire fictional computer to defend Maven's fucking broken-ass brokenness? There is only ONE computer involved in this process. You can't just pull shit out of your ass and call it an "argument" in a debate, Christ.
Ok, let's be a devil's advocate. So Maven doesn't work because a mysterious imaginary ghost-computer somewhere in reality has a different clock than mine? And that explains why the files are copied X% of the time (instead of 100% or 0%)... how?
Wait a fucking second, you just said you don't know anything about Maven. What the fuck, man. Maybe you should bow to my superior Maven experience and just trust me on this one, huh? Instead of writing the dumbest debate arguments I've heard in a long time. Because the latter option just makes you look dumb.
-
@blakeyrat said:
@pjt33 said:
In particular, the first thing I would check is whether it's deciding whether to copy or not based on file modification times, and whether clock skew between the two computers is a factor.
What second computer? Now you're making up an entire fictional computer to defend Maven's fucking broken-ass brokenness? There is only ONE computer involved in this process. You can't just pull shit out of your ass and call it an "argument" in a debate, Christ.
Some people use something called a "server." It's usually a different machine than your development machine. He was suggesting a reasonable scenario as to what might have been going on. Of course, such a suggestion was meant for a reasonable person, so sadly his effort was for naught. Watching your reaction, I wonder how you could ever debug much of anything, since you are so unable to consider that the bug might be your own.
-
@boomzilla said:
Watching your reaction, I wonder how you could ever debug much of anything, since you are so unable to consider that the bug might be your own.
Ok, first of all: 1) the Maven project isn't mine, I just use it because the loser web devs working on this site use it and I don't have any choice in the matter. So even in the best-case scenario, its not "my" bug
Secondly, 2) I didn't write Maven, although you apparently think I did despite obviously hating Java, so I think you're an idiot, and also it can't possibly be "my" bug
Thirdly, 3) If the build system is non-deterministic, that is, if the behavior doesn't happen 100% of the time, there is obviously a bug. So I don't know why people are defending it.
-
@blakeyrat said:
@Seahen said:
When you discovered that you had problems with 2 very different IDEs did it not occur to you to think that your basic problem might not be the IDE? Really?Try Eclipse.
-
@blakeyrat said:
@boomzilla said:
Watching your reaction, I wonder how you could ever debug much of anything, since you are so unable to consider that the bug might be your own.
Ok, first of all: 1) the Maven project isn't mine, I just use it because the loser web devs working on this site use it and I don't have any choice in the matter. So even in the best-case scenario, its not "my" bugSecondly, 2) I didn't write Maven, although you apparently think I did despite obviously hating Java, so I think you're an idiot, and also it can't possibly be "my" bug
Well, yours in the sense that it's in the code / configuration that you're working with, not the build system or java, etc. Fuck, I hate I had to write that.
@blakeyrat said:
Thirdly, 3) If the build system is non-deterministic, that is, if the behavior doesn't happen 100% of the time, there is obviously a bug. So I don't know why people are defending it.
Clearly, deterministic doesn't mean what you think it means. For instance, the scenario presented above (regarding different system times) would be an example of a deterministic process where everything didn't happen the way you thought it should.
And now you're justifying my assertion that you are poor at figuring out software bugs by assuming that something is not deterministic instead of trying to figure out which of your assumptions is incorrect. It's certainly possible that the bug is with maven, but given that you've already discovered problems with the way maven was being used, I'd lean toward another issue with the settings. This is at least partly based on my experience, where the problem (yes, even with java!) is almost always with my code / configuration rather than the 3rd party stuff. Which isn't to say that you haven't necessarily found a bug in maven. But assuming that it's there and not with your (and I don't necessarily mean what you have personally written, you asshat pedantic dickweed) use of maven is likely a way to make you frustrated and prevent you from finding the actual problem.
But it's not my problem, so I don't really care. It's somewhat amusing to watch a craftsman blaming his tools.
-
@Rick said:
@blakeyrat said:
@Seahen said:
When you discovered that you had problems with 2 very different IDEs did it not occur to you to think that your basic problem might not be the IDE? Really?Try Eclipse.
Read that linked thread. I outlined exactly what Eclipse is doing wrong to make it incompatible with my computer. And yes, it is a bug by any reasonable definition of the word-- in fact, NetBeans has the exact same bug, the only reason NetBeans is usable is because it writes less data to my Roaming AppData folder.
What's up with this thread? Where are you people all coming from?
-
@boomzilla said:
Well, yours in the sense that it's in the code / configuration that you're working with, not the build system or java, etc. Fuck, I hate I had to write that.
So when you say it's my bug, what you actually meant was it's a bug. Well, duh, we can agree on that point. That still leaves the mystery of why you said it was my bug when you know I didn't write Maven. I guess I'll just assume you were high on cough syrup and totally out of your mind when you typed it.
@boomzilla said:
This is at least partly based on my experience, where the problem (yes, even with java!) is almost always with my code / configuration rather than the 3rd party stuff. Which isn't to say that you haven't necessarily found a bug in maven.
If it's even possible for a build system to produce different results when run on the same files, using the same target, then it's a bug. Period. I don't see any point debating that further.
Now, maybe something in my company's configuration exposed the bug. But that doesn't magically mean Maven isn't buggy. Maybe the bug is rare. Maybe we're the first to ever discover it. Maybe it works for everybody else in the world except our little team of web developers. But that doesn't magically mean Maven isn't buggy.
@boomzilla said:
But it's not my problem, so I don't really care. It's somewhat amusing to watch a craftsman blaming his tools.
It's kind of different when it's a tool I didn't get a choice about, isn't it?
If I'm nailing in carpet tacks and you hand me a screwdriver (a broken screwdriver at that!) and force me to use it, then yes I'd complain. Wouldn't you?
-
@blakeyrat said:
@boomzilla said:
Well, yours in the sense that it's in the code / configuration that you're working with, not the build system or java, etc. Fuck, I hate I had to write that.
So when you say it's my bug, what you actually meant was it's a bug. Well, duh, we can agree on that point. That still leaves the mystery of why you said it was my bug when you know I didn't write Maven. I guess I'll just assume you were high on cough syrup and totally out of your mind when you typed it.Yeah, clearly I meant something other than what I wrote.
@blakeyrat said:
@boomzilla said:
This is at least partly based on my experience, where the problem (yes, even with java!) is almost always with my code / configuration rather than the 3rd party stuff. Which isn't to say that you haven't necessarily found a bug in maven.
If it's even possible for a build system to produce different results when run on the same files, using the same target, then it's a bug. Period. I don't see any point debating that further.
Yes, as in the case above (with different server times) it could be a bug with the way you were using the build system, which was the point, namely, that you don't understand the build system and aren't using it correctly. In that case, you were assuming that the only files that mattered were the ones that you've been working with, not the files on the server. That's not the case here, but it demonstrates the sort of incorrect assumption that can cause issues like this, whether you like them (or believe in their existence) or not.
@blakeyrat said:
Now, maybe something in my company's configuration exposed the bug. But that doesn't magically mean Maven isn't buggy. Maybe the bug is rare. Maybe we're the first to ever discover it. Maybe it works for everybody else in the world except our little team of web developers. But that doesn't magically mean Maven isn't buggy.
Yes, I definitely agree that this is a possibility. But I wouldn't assume that to be the case, especially when I don't have a good understanding of the tools involved. Also, since you've already discovered bugs in the way the maven stuff was set up, I wouldn't assume that the rest of it is correct, just because it doesn't spit out errors.
@blakeyrat said:
@boomzilla said:
But it's not my problem, so I don't really care. It's somewhat amusing to watch a craftsman blaming his tools.
It's kind of different when it's a tool I didn't get a choice about, isn't it?Not really. You could reasonably bitch about the other idiot devs who chose this, but clearly couldn't set it up properly. Again, you might be right that this is a maven bug, but experience and the story so far point back to those knuckleheads you work with. Look, it's not my fault you don't know what you're doing, and neither, apparently, does anyone else there.
@blakeyrat said:
If I'm nailing in carpet tacks and you hand me a screwdriver (a broken screwdriver at that!) and force me to use it, then yes I'd complain. Wouldn't you?
Yes, but I'll counter with a more relevant analogy. If I gave you an air nailer and air compressor, but you forgot to turn on the compressor, you would assume that the nailer was broken, because the problem can't possibly be the way you're using the tools.
@blakeyrat said:
Read that linked thread. I outlined exactly what Eclipse is doing wrong to make it incompatible with my computer. And yes, it is a bug by any reasonable definition of the word-- in fact, NetBeans has the exact same bug, the only reason NetBeans is usable is because it writes less data to my Roaming AppData folder.
Blakeyrat, foiled again by a shitty default configuration, and an inability or unwillingness to debug the situation.
-
@boomzilla said:
Blakeyrat, foiled again by a shitty default configuration, and an inability or unwillingness to debug the situation.
But why should he? They aren't [i]his[/i] bugs!
-
@boomzilla said:
Yes, as in the case above (with different server times) it could be a bug with the way you were using the build system, which was the point, namely, that you don't understand the build system and aren't using it correctly. In that case, you were assuming that the only files that mattered were the ones that you've been working with, not the files on the server. That's not the case here, but it demonstrates the sort of incorrect assumption that can cause issues like this, whether you like them (or believe in their existence) or not.
I said I won't debate it and I won't. It's Capital-T Truth as far as I'm concerned.
I will re-emphasize that there is only one computer involved here. The second computer only ever existed in pjt33's warped and fragile little mind.
@boomzilla said:
@blakeyrat said:
Read that linked thread. I outlined exactly what Eclipse is doing wrong to make it incompatible with my computer. And yes, it is a bug by any reasonable definition of the word-- in fact, NetBeans has the exact same bug, the only reason NetBeans is usable is because it writes less data to my Roaming AppData folder.
Blakeyrat, foiled again by a shitty default configuration, and an inability or unwillingness to debug the situation.
I'm not debugging Maven, because:
1) The existence of the bug is beyond doubt
2) Now that I know the bug exists, it's not impacting my work hugely
3) I don't need any further details because I'm sure as shit not going to break out the Maven source code and fix the problem for themAs for Aptana/Eclipse, if you're trying to tell me that was only a "misconfiguration" and not a bug, you're dead wrong. There's really no point in talking further about that because, again, it is beyond doubt. (Plus I know you don't genuinely believe any of this shit, you're just trolling to keep me posting-- no way you're that dumb, Boomzilla.)
-
@blakeyrat said:
@pjt33 said:
No, my point was that you're saying that any tool which has build operations other than "Never copy" and "Always copy" is shitty; you claim that VS is not shitty; and yet VS has build operations other than "Never copy" and "Always copy".I don't know anything about Maven, but I do know that Visual Studio's build system has options other than "Never" or "Always". Maybe you just never learned how to use them.
Objection! Relevance?
Is your point there, "Maven isn't buggy because Visual Studio has options that might also exhibit the same bugs (but I haven't bothered to check whether they do) and since I know that argument's weaksauce I'm going to call you ignorant too?"
@blakeyrat said:
@pjt33 said:
In particular, the first thing I would check is whether it's deciding whether to copy or not based on file modification times, and whether clock skew between the two computers is a factor.
What second computer? Now you're making up an entire fictional computer to defend Maven's fucking broken-ass brokenness?
I'm not defending anything. But you're the one who said @blakeyrat said:It just genuinely does not copy the updated .JS files to the server.
It is unreasonable to read that and assume that you are, in fact, complaining about it not copying files to a server?
-
@blakeyrat said:
Ok, let me try to rephrase it. I can't trust you when you say "it all works" because I can't be sure you even have the ability to know when something is broken.
Two things:First of all, I never said "it all works," nor would I, for three reasons:
- I don't know if that's the case.
- It's a lot harder to defend the claim that "it all works."
- I don't care.
However, I can surely appreciate how much easier it would be for you to refute such an argument, and so I fully understand your tactic of rephrasing my comment. Had it worked, I would have had a hard time defending a claim I did not make. It was very clever of you, and while it didn't succeed this time I do look forward to future attempts by you to put words in my mouth.
Second, while you can't be sure of my ability to know when something is broken (an uncertainty that seems unwarranted, as I don't believe I've given you any reason to doubt it yet), I can't be sure of your ability to properly configure the tools you use, especially after you admitted your config was shit. Even if I am some kind of [i]expert[/i] in telling when things are broken, it's much harder to tell if something is really broken when it isn't configured properly.
-
@pjt33 said:
No, my point was that you're saying that any tool which has build operations other than "Never copy" and "Always copy" is shitty;
... I was?
Can you write a SINGLE POST without putting words into my mouth? Let's call it the DailyWTF Super Mud Truck Racing Challenge! Next up: "pjt33". Time to beat: "constantly lying about what Blakeyrat says, because his pants are continually on fire"
Can you win the Super Mud Truck Racing Challenge?
(And no, I'm not going to bother to reply to your post, since you'd prefer the version in your own mind to whatever I typed anyway.)
-
@boog said:
It was very clever of you, and while it didn't succeed this time I do look forward to future attempts by you to put words in my mouth.
I guess I should take the direct approach, like pjt33.
@boog said:
Second, while you can't be sure of my ability to know when something is broken (an uncertainty that seems unwarranted, as I don't believe I've given you any reason to doubt it yet),
I've talked to enough Linux users and Lotus Notes fans that it's basically my default position at this point. The real problem is you were trying to refute the bugs in NetBeans, which are significantly more obviously bugs than the Maven thing. I can accept "it is acceptable that Maven writes updated files a random percentage of the time" (although I don't have the power of imagination to figure out what could possibly cause that in a build system)
But I can't accept someone looking at NetBeans displaying "Consolas 14" and not getting upset that it's not actually drawing Consolas 14 on the screen. There are limits, man. Limits.
@boog said:
Even if I am some kind of expert in telling when things are broken, it's much harder to tell if something is really broken when it isn't configured properly.
If it's possible to configure it in such a way that it is no longer fit-for-purpose, (for existence, Maven in this case) then it's broken. Unless you actually think every application should have a "break me" checkbox in it somewhere, again, I see no point in debating that.
-
@blakeyrat said:
As for Aptana/Eclipse, if you're trying to tell me that was only a "misconfiguration" and not a bug, you're dead wrong. There's really no point in talking further about that because, again, it is beyond doubt. (Plus I know you don't genuinely believe any of this shit, you're just trolling to keep me posting-- no way you're that dumb, Boomzilla.)
I'm not disagreeing that the default configuration is a bug. However, it's a bug with an easy workaround. So, while I agree that they got it wrong, I'm saying that using this particular issue as an excuse to not get work done is wrong (I'm assuming here that Getting Shit Done is ultimately more important than satisfying your personal software standards, since this is all happening in the course of your job, not a personal project). Using is as an excuse for a java rant is OK, however, since the root problem appears to be that java's user home detection method is not nuanced enough for the way that Windows works.
It just occurred to me that both threads demonstrate the same trait in you, is all. They kind of remind me of "A Bug's Life," when the leaf (or whatever) falls in the middle of the line of ants marching along, and the ants behind the leaf begin to panic because they can no longer find the line due to the leaf.
-
@blakeyrat said:
@pjt33 said:
No, my point was that you're saying that any tool which has build operations other than "Never copy" and "Always copy" is shitty;
... I was?
Can you write a SINGLE POST without putting words into my mouth? Let's call it the DailyWTF Super Mud Truck Racing Challenge! Next up: "pjt33". Time to beat: "constantly lying about what Blakeyrat says, because his pants are continually on fire"
It's true, he didn't directly quote you, but please explain how he is misrepresenting the following earlier statement of yours:
@blakeyrat said:Shouldn't the .JS files either never be copied, or always be copied? What gigantic WTFs are in the foundations of Maven that allow them to be copied sometimes and not other times?
@blakeyrat said:
If it's possible to configure it in such a way that it is no longer fit-for-purpose, (for existence, Maven in this case) then it's broken. Unless you actually think every application should have a "break me" checkbox in it somewhere, again, I see no point in debating that.
If you can't possibly misconfigure it, then it's probably not powerful enough to be useful.
-
@blakeyrat said:
@pjt33 said:
No, my point was that you're saying that any tool which has build operations other than "Never copy" and "Always copy" is shitty;
... I was?
I'd love to hear your alternative interpretation of @blakeyrat said:The real problem is that a build system works where the build is non-deterministic! Shouldn't the .JS files either never be copied, or always be copied? What gigantic WTFs are in the foundations of Maven that allow them to be copied sometimes and not other times?
-
@boomzilla said:
I'm not disagreeing that the default configuration is a bug. However, it's a bug with an easy workaround.
I'm actually starting to think you're really that dumb, and not just pretending.
Go back, read the bug again. Read it. Now tell me what the "easy workaround" is. I'd love to hear it, since apparently at the time everybody involved (yourself included) was simply too dumb to find the "easy workaround" and post it in that thread.
@boomzilla said:
(I'm assuming here that Getting Shit Done is ultimately more important than satisfying your personal software standards, since this is all happening in the course of your job, not a personal project).
Since I'm losing PTO every hour I work on this (via a complicated story I really don't want to get into), I really have no incentive to work on this at all, other than as a huge personal favor to the person whose job is to actually do this work. And I can already virtually guarantee he won't appreciate it at all.
@boomzilla said:
Using is as an excuse for a java rant is OK, however, since the root problem appears to be that java's user home detection method is not nuanced enough for the way that Windows works.
Not just the detection method, the very fundamental concept. Like I posted in that linked thread, Windows:
1) Does not have a single "Home" folder
2) Has no requirement that various user-specific folders are on the same drive, or in the same path
3) Has no pre-defined name for those user-specific folders (that is, assuming the documents folder is named "My Documents" or "Documents" is wrong)
4) Has no requirement that the various user-specific folders stay in the same place from login-to-login, or even minute-to-minuteWindows NT has always been this way, and since it's older than Java, that also means that Java has always been conceptually wrong on this point.
Why am I retyping this? I don't know. I'm going to go make a latte.
-
@pjt33 said:
I'd love to hear your alternative interpretation of @blakeyrat said:
The real problem is that a build system works where the build is non-deterministic! Shouldn't the .JS files either never be copied, or always be copied? What gigantic WTFs are in the foundations of Maven that allow them to be copied sometimes and not other times?
Fucking pedantic dickweeds. I left out the word "changed."
You lose the Super Mud Truck Racing Challenge and I want you to die in an extremely painful matter.
-
@blakeyrat said:
@boomzilla said:
I'm not disagreeing that the default configuration is a bug. However, it's a bug with an easy workaround.
I'm actually starting to think you're really that dumb, and not just pretending.
Go back, read the bug again. Read it. Now tell me what the "easy workaround" is. I'd love to hear it, since apparently at the time everybody involved (yourself included) was simply too dumb to find the "easy workaround" and post it in that thread.
< a href="http://forums.thedailywtf.com/forums/p/24771/274309.aspx#274309">OK:
@boomzilla said:File -> Switch Workspace -> Other. Then browse to where you want it to store your workspace, and tell it to use that place.
-
@blakeyrat said:
I can't accept someone looking at NetBeans displaying "Consolas 14" and not getting upset that it's not actually drawing Consolas 14 on the screen.
Who said I wasn't upset? Ever since reading about your font problem, it's all I can think about. I haven't had much of an appetite lately, and I assure you that I've been losing sleep every night. Sometimes my mind drifts to horrible, terrible-looking fonts in a text editor and so I stare blankly out into the distance. Or I start sobbing uncontrollably. My wife is starting to worry about me, and asked that I seek help. I have a meeting with a Font Misrendering Paranoia specialist next Thursday. But maybe that's not the appropriate level of upset to win your approval.Still, as upset as I clearly am, the font rendering issue does not qualify the statement "NetBeans doesn't work" because unless it's rendering Wingdings, it doesn't get in the way of actual work. You see, my not-as-strict-as-yours criteria for whether something works or not is as follows:
- Does it work? If so, then it "works."
- Otherwise, it "doesn't work."
- That is all.
So far that's the only argument I've made about the issue (that I recall). Not once did I argue the font rendering was of good quality.
@blakeyrat said:
If it's possible to configure it in such a way that it is no longer fit-for-purpose, (for existence, Maven in this case) then it's broken.
Hahaha, good one! Oh wait, you were serious? Well I guess we're back to the "everything sucks" discussion then. Unless you can identify for me one product that cannot be configured in such a way that it Does Not Work, that is? Minesweeper doesn't count.
-
@boog said:
@blakeyrat said:
If it's possible to configure it in such a way that it is no longer fit-for-purpose, (for existence, Maven in this case) then it's broken.
Hahaha, good one! Oh wait, you were serious? Well I guess we're back to the "everything sucks" discussion then. Unless you can identify for me one product that cannot be configured in such a way that it Does Not Work, that is? Minesweeper doesn't count.Check the work "ideal"
One can rant and work, multitasking is easy.
-
@boomzilla said:
OK:
@boomzilla said:File -> Switch Workspace -> Other. Then browse to where you want it to store your workspace, and tell it to use that place.
Except that doesn't work because the program doesn't even run. You obviously didn't bother to read the bug. Please don't waste my time with this bullshit.
-
-
@frits said:
@serguey123 said:
Nope. He's totally entertaining, and makes good points.
My thinking is obviously incompatible with people who can tolerate Java.
Oh here's an added fun bonus: because JIRA has decided I no longer exist*, I'm totally blocked on this project. And since I'm sick as a dog I'm just sitting here in my office chair wrapped in a shawl and basically the most pathetic thing ever. Fun.
*) the error message doesn't say "that user account doesn't exist" it literally says "that user doesn't exist". Even the tiniest details in Java products are broken.
