So true. And one thing that occurred to me recently: I've been reading Slashdot for over 10 years and I've seen editors come and go, but I have never seen an advert for the job of Slashdot editor advertised. Every tech news site I read has posted ads for editors and writers in that period, but not Slashdot. It is the one site that doesn't try to recruit staff from its readership. It's also interesting to see how high the UIDs are for a lot of the Slashdot editors - several of them apparently didn't even have Slashdot accounts until after they got the job. It's therefore not really surprising that they'd be a bit useless.
There are very few interesting posters over the 2M UID mark... but there are some. There are now over half a million accounts in that range, so it would be a bit surprising if none of them ever said anything interesting. That said, there are probably a lot of patterns that you can watch for to spot mod abuse. For example:
Don't allow (silently drop) moderations by one account of another from the same IP address
Look for clusters of accounts that each moderate the others up (or down). Even when this isn't abuse, it just generates an echo chamber. Reduce the probability of these accounts getting mod points.
Or you could just bring back the old metamod system. You know, the one that actually worked, where anyone in the oldest 90% of accounts could do it, but had no control over the posts that they metamoderated.
According to the Hall of Fame, I'm the most active commenter for this quarter (I really need to stop procrastinating), which probably means I will be ignored by the Slashdot overlords more than more people, but here are my inflation-adjusted two cents:
Most Discussed. Don't care. It's not like it's hard to see the comment counts on the front page.
This Day on Slashdot. Might be interesting. Probably not though - lots of slow news days. Sounds like an attempt at recycling old content.
Upgrading Slashdot to modern hardware and new versions of MySQL and Apache. Irrelevant to most users. Upgrading to a real database and a more modern web server might be interesting, if only for the flame war in the comments that it would provoke.
Cleaning up the topics pages. Long overdue. Although part of the problem is that most of the current crop of 'editors' are stunningly ignorant of their subject matter and so routinely file things in the wrong category.
Improving methods for sharing submissions. Yes, Slashdot needs more Facebook integration. And more Twitter integration. And it definitely needs to jump on the Google+ bandwagon! The last changes to this crap meant that my user CSS no longer blocks the share button, as it previously blocked the little 'I am an attention whore so desperate for approval that I want to help companies build a database about me' buttons. Please, please, please, provide a user option to turn off all of this crap, if you must have it.
Thumbnails for articles with videos. Even better would be an option of hiding all articles with videos from the front page. The last ones have all been spam, so I wouldn't even have wanted to read them in text form. I usually read Slashdot while waiting for a compile job to finish or while having a short procrastination break, so things that require 100% attention such as videos are of no interest to me. Stick them all in videos.slashdot.org and make it as easy to ignore as idle.
Flag-a-comment abuse reporting. As you say, a step backwards. Slashdot isn't Slashdot without trolls. Mod them down, but don't delete them.
Removal of old and unused Slashboxes. If they're unused, no one will notice or care, so this is irrelevant to everyone. If, as I suspect, by 'unused' you mean 'some people use them, but I don't' then you're just trying to bill removing a feature that people use as an improvement. I suggest you quit Slashdot and get a job at Apple.
A much overdue overhauling of the FAQ. Again, long overdue. Note that overhaul usually implies improvement and please remember that when you do it.
Fixes to user preferences. Bug fixes are good. Currently lots of this stuff was broken by the Web 2.0!11111eleventyone rewrite.
The launch of the Slashdot Hall of Fame. Dear God No! The 'achievements' section was bad enough. I thought this was an April Fools joke when it was launched, but it stuck around. Now we have more of this crap. Clever people are able to learn from the mistakes of others. Most people can learn from their own mistakes. The sign of total idiocy is failing to learn from your own mistakes. Now we have a hall of fame which is going to promote exactly the same behaviour as the old public karma numbers, a system Slashdot abandoned for very good reasons. Please, learn from your mistakes, don't keep repeating them.
Fixes to the D2 comment system. Maybe next time you could do this before making it default? For the record, I mostly like the D2 system. The biggest bug, however, is that you can type a long comment and then accidentally hit cancel instead of preview and lose it (which you couldn't with the old one, as browsers would warn you if you tried to navigate off a page with a full text field). Fix that first! Slashdot always embodied the ugly-but-functional school of design. The rewrite made it no less ugly, but made it less functional. We're happy with Slashdot being ugly, but please make it actually work. Another example: it still requi
Not always. The first connection to the Internet from the UK was from one guy at University College in London. He got an account on ARPANet via a collaboration with some people stateside and then made his connection available to anyone in the UK university network (not sure if it was JANET then, or just the thing that would later grow into JANET). There were dozens of people all using the same account, with not authentication - as long as they had an account on a university system they could connect to ARPANet, without anyone on that side knowing who they were...
Earlier still was ADVENT, whose magic word was xyzzy . Whole 'nother country, that.
The first time I played ADVENT was on an IBM-made 8080-based machine running (I think) CP/M. There was one big blue box for the motherboard and all of the daughterboards (CPU, RAM, disk controller, serial controller) and so, a similar sized one for the 8" floppy disk drives, and finally a terminal.
The second time I played ADVENT was on my Psion Series 3, a device with more RAM and more processing power, and a GUI, which fitted into my jacket pocket.
The best thing about most technology from the past is that we don't have to use it anymore.
When the use of compression grew in modem transfers, baud often stayed the same, or rose slower than the bit rate due to the compression.
It doesn't have anything to do with compression. The baud rate is the number of symbols per second. The bps rate is the number of bits per second. When you have two kinds of symbol (e.g. beep and silence, high and low) then the baud rate is the same as the bit rate. If you have 4 kinds of symbol then each symbol represents two bits and so the bit rate is double the baud rate. With better ADCs and DACs (and a sufficiently low SNR) you can distinguish a lot more different symbols at the same baud rate. If you could distinguish 256 different tones then a 300 baud modem could run at 300B/s (2400b/s).
I have a projector projecting about a 2m image onto the wall in front of my sofa and a 5.1 surround sound system connected to the (silent) computer that plays DVDs and streams video. I don't think a 'nice big TV' would improve things for me.
It's not just about market share, although that does play a large part. For malware you spread you need a large or sufficiently interesting target for someone to bother writing it (an OS with only a dozen users, all of which were major banks that used it for Internet-facing transaction processing systems, for example, would be an interesting target even though it would have a tiny market share).
Then you need an attack vector. Operating system vulnerabilities aren't that uncommon (check the CVE database for the Linux kernel), but most of the time these attacks come through userspace applications. From there, it depends on what the attacker wants to use. Desktop operating systems tend to be more vulnerable in this regard because very few applications are properly sandboxed, so once you've compromised one you've got complete access to everything the user does. Server software tends to be a bit more careful with privilege separation, so a Linux server may be a lot more secure than a Linux desktop.
Finally, you need some mechanism for it to spread. This is often related to market share. For example, Windows worms used to be very common because if you look at any random IP on the local network you're likely to find a Windows machine. If you've got some Windows exploit, you can spread to every machine on the network very quickly. The same was true of email worms - a worm that compromised Outlook Express could send a message to everyone in the address book, and at least some of them would be running Outlook Express and so it would spread. In contrast, if the lone Mac in the corner of the office is infected then it's harder for it to find another Mac to infect before someone spots unusual traffic patterns and cleans it up.
Not all police are bad, it's just that 95% are giving the other 5% a bad name.
What do you base that on? Sure, it's a glib remark, but what are your experiences with the police? Every time I've encountered them, I've found them helpful, polite, and courteous. Maybe it's just my good fortune to always interact with the 5% though.
Doesn't it bother anyone that unicast is a horrible waste of bandwidth?
Yes and no. It is a waste of bandwidth if everyone is watching the same thing at the same time. A better solution would be more caching at the edges. If I watch a popular show, it should be stored on my router and then if my neighbours watch it they should stream it from me, keeping it on the local segment. That gives you most of the advantages of multicast, but without the limitation of people needing to receive things at the same time. You could perhaps do some speculative caching, so if it predicts that more than one person on a segment will want to watch something it will broadcast it on that segment and they all cache a copy.
250GB is a lot of video. iPlayer HD is 3600Kb/s. At that rate, it's over 150 hours of streaming - or about five hours a day. The SD content is 800Kb/s. At that rate you can stream over 22 hours per day and still not hit the cap.
People are watching less TV partially because there's less TV to watch in an hour.
I stopped watching broadcast TV except on the BBC in the UK because there were too many ads (I then stopped watching the BBC channels, because iPlayer is a more convenient way of getting at the same content). When I tried watching TV in the USA, I was amazed anyone put up with it. You had at least twice as many ads as we did when it passed my tolerance threshold.
I don't have a TV anymore, and I watch a lot more TV shows than I did back when I had one. Between iPlayer and DVD rentals, there's a lot of enjoyable content, and I know that if I sit down to watch a TV show for an hour then I will get an hour of entertainment, not 45 minutes of entertainment and 15 minutes of being annoyed.
I proposed something similar a few years ago: Offer a small discount (say, 50% of the ticket price) on the DVD in exchange for your cinema ticket on the way out. I wouldn't be surprised if at least 10% of cinema goers would impulse purchase the DVD. Ideally, you wouldn't delay the DVD launch at all: as soon as the first person walked out of the cinema after the first showing, they'd be able to buy the DVD. I'd have thought that the people who go to the first showing of a movie are the most likely set to want to buy the DVD and making it available at the point when they're thinking about the film the most would maximise sales.
How about making the DVDs available in a timely fashion? I don't buy many DVDs anymore, but I rent a lot. I often have to wait 6-12 months between a film being in the cinemas or a TV show airing in the USA and it being available to rent. On the other hand, if I wanted to pirate, films are usually available within a few days of cinema release and TV shows within a few hours. If you say 'you can get our product illegally now, or legally in 6-12 months' then you shouldn't be surprised when a lot of people opt for now. Especially when you spend a huge advertising budget on telling them that they want to see it right now...
And so could Google. Viacom's market capitalisation is $27B, with a turnover of about $15B. Google currently has about $30B in liquid assets. If Google, Apple, and a few other companies got together they could easily provide $10-20B to a fund that would finance freely redistributable films.
Compare the string handling of C++ with that of Java or C# or Perl or Tcl or... Well, let's say it like this: C++ gives you the pain of the complexity without nearly enough of the gain of well-implemented features.
Strings are a really good example of how C focusses on microoptimisation. You have very fast character access to std::strings. And you have very fast iterators. And that's it. And because none of the methods are virtual, you can't just slot in a subclass that's more efficient in your usage. Compare this to Objective-C strings, which are implemented as a class cluster with different implementations for different uses. For example, when using ICU it's trivial to wrap its internal unicode string ADT in an NSString subclass. Because the accessor used by anything that needs speed copies a range of characters into a buffer, it's fast. In C++, you end up copying the entire string as soon as you find something that wants a std::string, or a QString (Qt) or a WtfString (WebKit) or whatever. Fine for very short strings, not so fine for longer ones.
The focus on microoptimisation in C++ means you end up with very tight coupling, which makes high-level optimisations (where the real performance wins come from) very hard.
It depends on what you count as human. Modern humans are generally thought to have been around for about 50,000 years, with anatomically similar ones (but without the same behaviour range) extending back 200,000 years. The first members of the homo genus appeared about 2.5 million years ago, so there were a lot of vaguely human-like things around a million years ago, but probably none that you'd invite over for tea. Wikipedia has a nice timeline.
The, ah, problem with the n-body problem is that it is chaotic. Very small errors in measurements can result in large change to the outcome. Fortunately, it's not chaotic over its entire domain[1], and there are regions within the solution space where approximations work fine.
[1] Any mathematicians reading this are probably going to crucify me over my abuse of terminology there, sorry...
With C++, there's a similar problem that everyone is focusing so much on performance
More accurately, everyone is focussing too much on microbenchmark performance. C++ is a language designed for an inlining-happy compiler with lots of compile-time specialisation. This results in very large code, which means that you end up with a lot of instruction cache churn. That's a total performance killer on modern hardware for large programs, but new features of C++ (with the possible exception of lambdas) are designed to make it even worse.
No it doesn't. That was one of the things that the StrongTalk team learned when Java was in its infancy. Type feedback (in class-based languages) provides more accurate information than user type annotations. A modern JVM doesn't even use the source-language annotations, it infers the types based on profiling.
{lots of stuff about how Perl is even worse}
Not really relevant. Yes, there are worse languages than Java. There aren't, however, any languages worse than Java that are anything like as successful as Java.
And this [design pattern fetishism] is unique to Java?
Not unique, perhaps, but Java does it far more than any other language. Design patterns are useful, but they can be taken too far. Occasional frameworks do in other languages. Pretty much everything in Java does.
Slashdot Mirror
So true. And one thing that occurred to me recently: I've been reading Slashdot for over 10 years and I've seen editors come and go, but I have never seen an advert for the job of Slashdot editor advertised. Every tech news site I read has posted ads for editors and writers in that period, but not Slashdot. It is the one site that doesn't try to recruit staff from its readership. It's also interesting to see how high the UIDs are for a lot of the Slashdot editors - several of them apparently didn't even have Slashdot accounts until after they got the job. It's therefore not really surprising that they'd be a bit useless.
That's there already. The ads are the ones posted on Slashdot.
There are very few interesting posters over the 2M UID mark... but there are some. There are now over half a million accounts in that range, so it would be a bit surprising if none of them ever said anything interesting. That said, there are probably a lot of patterns that you can watch for to spot mod abuse. For example:
Or you could just bring back the old metamod system. You know, the one that actually worked, where anyone in the oldest 90% of accounts could do it, but had no control over the posts that they metamoderated.
According to the Hall of Fame, I'm the most active commenter for this quarter (I really need to stop procrastinating), which probably means I will be ignored by the Slashdot overlords more than more people, but here are my inflation-adjusted two cents:
Most Discussed. Don't care. It's not like it's hard to see the comment counts on the front page.
This Day on Slashdot. Might be interesting. Probably not though - lots of slow news days. Sounds like an attempt at recycling old content.
Upgrading Slashdot to modern hardware and new versions of MySQL and Apache. Irrelevant to most users. Upgrading to a real database and a more modern web server might be interesting, if only for the flame war in the comments that it would provoke.
Cleaning up the topics pages. Long overdue. Although part of the problem is that most of the current crop of 'editors' are stunningly ignorant of their subject matter and so routinely file things in the wrong category.
Improving methods for sharing submissions. Yes, Slashdot needs more Facebook integration. And more Twitter integration. And it definitely needs to jump on the Google+ bandwagon! The last changes to this crap meant that my user CSS no longer blocks the share button, as it previously blocked the little 'I am an attention whore so desperate for approval that I want to help companies build a database about me' buttons. Please, please, please, provide a user option to turn off all of this crap, if you must have it.
Thumbnails for articles with videos. Even better would be an option of hiding all articles with videos from the front page. The last ones have all been spam, so I wouldn't even have wanted to read them in text form. I usually read Slashdot while waiting for a compile job to finish or while having a short procrastination break, so things that require 100% attention such as videos are of no interest to me. Stick them all in videos.slashdot.org and make it as easy to ignore as idle.
Flag-a-comment abuse reporting. As you say, a step backwards. Slashdot isn't Slashdot without trolls. Mod them down, but don't delete them.
Removal of old and unused Slashboxes. If they're unused, no one will notice or care, so this is irrelevant to everyone. If, as I suspect, by 'unused' you mean 'some people use them, but I don't' then you're just trying to bill removing a feature that people use as an improvement. I suggest you quit Slashdot and get a job at Apple.
A much overdue overhauling of the FAQ. Again, long overdue. Note that overhaul usually implies improvement and please remember that when you do it.
Fixes to user preferences. Bug fixes are good. Currently lots of this stuff was broken by the Web 2.0!11111eleventyone rewrite.
The launch of the Slashdot Hall of Fame. Dear God No! The 'achievements' section was bad enough. I thought this was an April Fools joke when it was launched, but it stuck around. Now we have more of this crap. Clever people are able to learn from the mistakes of others. Most people can learn from their own mistakes. The sign of total idiocy is failing to learn from your own mistakes. Now we have a hall of fame which is going to promote exactly the same behaviour as the old public karma numbers, a system Slashdot abandoned for very good reasons. Please, learn from your mistakes, don't keep repeating them.
Fixes to the D2 comment system. Maybe next time you could do this before making it default? For the record, I mostly like the D2 system. The biggest bug, however, is that you can type a long comment and then accidentally hit cancel instead of preview and lose it (which you couldn't with the old one, as browsers would warn you if you tried to navigate off a page with a full text field). Fix that first! Slashdot always embodied the ugly-but-functional school of design. The rewrite made it no less ugly, but made it less functional. We're happy with Slashdot being ugly, but please make it actually work. Another example: it still requi
Not always. The first connection to the Internet from the UK was from one guy at University College in London. He got an account on ARPANet via a collaboration with some people stateside and then made his connection available to anyone in the UK university network (not sure if it was JANET then, or just the thing that would later grow into JANET). There were dozens of people all using the same account, with not authentication - as long as they had an account on a university system they could connect to ARPANet, without anyone on that side knowing who they were...
It's like oldfag, except that people over the age of 12 who or who aren't on 4chan say it.
Earlier still was ADVENT, whose magic word was xyzzy . Whole 'nother country, that.
The first time I played ADVENT was on an IBM-made 8080-based machine running (I think) CP/M. There was one big blue box for the motherboard and all of the daughterboards (CPU, RAM, disk controller, serial controller) and so, a similar sized one for the 8" floppy disk drives, and finally a terminal.
The second time I played ADVENT was on my Psion Series 3, a device with more RAM and more processing power, and a GUI, which fitted into my jacket pocket.
The best thing about most technology from the past is that we don't have to use it anymore.
When the use of compression grew in modem transfers, baud often stayed the same, or rose slower than the bit rate due to the compression.
It doesn't have anything to do with compression. The baud rate is the number of symbols per second. The bps rate is the number of bits per second. When you have two kinds of symbol (e.g. beep and silence, high and low) then the baud rate is the same as the bit rate. If you have 4 kinds of symbol then each symbol represents two bits and so the bit rate is double the baud rate. With better ADCs and DACs (and a sufficiently low SNR) you can distinguish a lot more different symbols at the same baud rate. If you could distinguish 256 different tones then a 300 baud modem could run at 300B/s (2400b/s).
I have a projector projecting about a 2m image onto the wall in front of my sofa and a 5.1 surround sound system connected to the (silent) computer that plays DVDs and streams video. I don't think a 'nice big TV' would improve things for me.
It's not just about market share, although that does play a large part. For malware you spread you need a large or sufficiently interesting target for someone to bother writing it (an OS with only a dozen users, all of which were major banks that used it for Internet-facing transaction processing systems, for example, would be an interesting target even though it would have a tiny market share).
Then you need an attack vector. Operating system vulnerabilities aren't that uncommon (check the CVE database for the Linux kernel), but most of the time these attacks come through userspace applications. From there, it depends on what the attacker wants to use. Desktop operating systems tend to be more vulnerable in this regard because very few applications are properly sandboxed, so once you've compromised one you've got complete access to everything the user does. Server software tends to be a bit more careful with privilege separation, so a Linux server may be a lot more secure than a Linux desktop.
Finally, you need some mechanism for it to spread. This is often related to market share. For example, Windows worms used to be very common because if you look at any random IP on the local network you're likely to find a Windows machine. If you've got some Windows exploit, you can spread to every machine on the network very quickly. The same was true of email worms - a worm that compromised Outlook Express could send a message to everyone in the address book, and at least some of them would be running Outlook Express and so it would spread. In contrast, if the lone Mac in the corner of the office is infected then it's harder for it to find another Mac to infect before someone spots unusual traffic patterns and cleans it up.
Not all police are bad, it's just that 95% are giving the other 5% a bad name.
What do you base that on? Sure, it's a glib remark, but what are your experiences with the police? Every time I've encountered them, I've found them helpful, polite, and courteous. Maybe it's just my good fortune to always interact with the 5% though.
Doesn't it bother anyone that unicast is a horrible waste of bandwidth?
Yes and no. It is a waste of bandwidth if everyone is watching the same thing at the same time. A better solution would be more caching at the edges. If I watch a popular show, it should be stored on my router and then if my neighbours watch it they should stream it from me, keeping it on the local segment. That gives you most of the advantages of multicast, but without the limitation of people needing to receive things at the same time. You could perhaps do some speculative caching, so if it predicts that more than one person on a segment will want to watch something it will broadcast it on that segment and they all cache a copy.
250GB is a lot of video. iPlayer HD is 3600Kb/s. At that rate, it's over 150 hours of streaming - or about five hours a day. The SD content is 800Kb/s. At that rate you can stream over 22 hours per day and still not hit the cap.
People are watching less TV partially because there's less TV to watch in an hour.
I stopped watching broadcast TV except on the BBC in the UK because there were too many ads (I then stopped watching the BBC channels, because iPlayer is a more convenient way of getting at the same content). When I tried watching TV in the USA, I was amazed anyone put up with it. You had at least twice as many ads as we did when it passed my tolerance threshold.
I don't have a TV anymore, and I watch a lot more TV shows than I did back when I had one. Between iPlayer and DVD rentals, there's a lot of enjoyable content, and I know that if I sit down to watch a TV show for an hour then I will get an hour of entertainment, not 45 minutes of entertainment and 15 minutes of being annoyed.
I proposed something similar a few years ago: Offer a small discount (say, 50% of the ticket price) on the DVD in exchange for your cinema ticket on the way out. I wouldn't be surprised if at least 10% of cinema goers would impulse purchase the DVD. Ideally, you wouldn't delay the DVD launch at all: as soon as the first person walked out of the cinema after the first showing, they'd be able to buy the DVD. I'd have thought that the people who go to the first showing of a movie are the most likely set to want to buy the DVD and making it available at the point when they're thinking about the film the most would maximise sales.
This is very Linux-centric
No, it's very UNIX Release 6 centric. It hasn't been true of most modern UNIX and UNIX-like systems for about 20 years.
There seems to be a bug. I have the 'Ads Disabled' checkbox ticked, but I still see this big ad right in the top-centre of the front page.
How about making the DVDs available in a timely fashion? I don't buy many DVDs anymore, but I rent a lot. I often have to wait 6-12 months between a film being in the cinemas or a TV show airing in the USA and it being available to rent. On the other hand, if I wanted to pirate, films are usually available within a few days of cinema release and TV shows within a few hours. If you say 'you can get our product illegally now, or legally in 6-12 months' then you shouldn't be surprised when a lot of people opt for now. Especially when you spend a huge advertising budget on telling them that they want to see it right now...
And so could Google. Viacom's market capitalisation is $27B, with a turnover of about $15B. Google currently has about $30B in liquid assets. If Google, Apple, and a few other companies got together they could easily provide $10-20B to a fund that would finance freely redistributable films.
Compare the string handling of C++ with that of Java or C# or Perl or Tcl or ... Well, let's say it like this: C++ gives you the pain of the complexity without nearly enough of the gain of well-implemented features.
Strings are a really good example of how C focusses on microoptimisation. You have very fast character access to std::strings. And you have very fast iterators. And that's it. And because none of the methods are virtual, you can't just slot in a subclass that's more efficient in your usage. Compare this to Objective-C strings, which are implemented as a class cluster with different implementations for different uses. For example, when using ICU it's trivial to wrap its internal unicode string ADT in an NSString subclass. Because the accessor used by anything that needs speed copies a range of characters into a buffer, it's fast. In C++, you end up copying the entire string as soon as you find something that wants a std::string, or a QString (Qt) or a WtfString (WebKit) or whatever. Fine for very short strings, not so fine for longer ones.
The focus on microoptimisation in C++ means you end up with very tight coupling, which makes high-level optimisations (where the real performance wins come from) very hard.
The problem with the Yellowstone supervolcano is that it probably won't wipe out California and the East coast
Don't worry, the flood of refugees will.
It depends on what you count as human. Modern humans are generally thought to have been around for about 50,000 years, with anatomically similar ones (but without the same behaviour range) extending back 200,000 years. The first members of the homo genus appeared about 2.5 million years ago, so there were a lot of vaguely human-like things around a million years ago, but probably none that you'd invite over for tea. Wikipedia has a nice timeline.
[1] Any mathematicians reading this are probably going to crucify me over my abuse of terminology there, sorry...
With C++, there's a similar problem that everyone is focusing so much on performance
More accurately, everyone is focussing too much on microbenchmark performance. C++ is a language designed for an inlining-happy compiler with lots of compile-time specialisation. This results in very large code, which means that you end up with a lot of instruction cache churn. That's a total performance killer on modern hardware for large programs, but new features of C++ (with the possible exception of lambdas) are designed to make it even worse.
It benefits greatly from its static typing system
No it doesn't. That was one of the things that the StrongTalk team learned when Java was in its infancy. Type feedback (in class-based languages) provides more accurate information than user type annotations. A modern JVM doesn't even use the source-language annotations, it infers the types based on profiling.
{lots of stuff about how Perl is even worse}
Not really relevant. Yes, there are worse languages than Java. There aren't, however, any languages worse than Java that are anything like as successful as Java.
And this [design pattern fetishism] is unique to Java?
Not unique, perhaps, but Java does it far more than any other language. Design patterns are useful, but they can be taken too far. Occasional frameworks do in other languages. Pretty much everything in Java does.