A few years ago, in Britain, one of the universities finally shut down their "BA Independent Studies" programme. Three years of studying whatever the hell you want, with the ability to go to any lecture courses in any subject, so long as at the end of it, you had a coherent programme of research.
No, I'd say that part of what is scary about zombies is that a loved one might turn into a zombie. Suddenly, the hero's girlfriend becomes a zombie - or, worse, his own mother or father or best friend - and he is tasked with destroying this zombie version of a loved one who still seems to be in some sense living.
Putting on my poncy lit-crit hat: surely, this shows that one cannot simply pine for some Lockean state of nature and imagine that civilization could just fall apart without it affecting you. (Libertarians, I'm talking to you. Heh.)
P-zombies would not differ from human beings - certainly, if you saw two walk past you on the street you couldn't tell the difference. On some accounts, you might be able to tell the difference if you had them wired up in a neuroscience lab. The zombies of philosopher's imaginations are quite different from the zombies of Resident Evil games (etc.): in fact, if p-zombies were like movie zombies, they wouldn't be very interesting in philosophy of mind (although there would be some interesting biological questions we might have about them...)
His point isn't that sometimes you need to write code that solves scalability issues: it is that a lot of Web 2.0 douchebag types think that they need to write their application using NoSQL and Node and all sorts of other fancy scalability shit, when in fact scalability is not their concern yet and that they should shut the fuck up, install a mature RDBMS (like Postgres or The Other Database) and use it until they have a scalability problem.
But they don't. Because geeks love to play with the new shiny toys and brag about it on their blogs.
You should probably be looking at the Internet Encyclopedia of Philosophy too. I've found that roughly my reading pattern when I hit a new topic in philosophy is: IEP, SEP, textbooks, monographs, research papers. Some of the SEP articles - especially in philosophy of language and epistemology - can be a little intimidating and get into the logical symbolism a bit too rapidly. The IEP articles are a little gentler, and are definitely suitable for undergraduates in a way that some of the SEP articles aren't.
Stanford Encyclopedia of Philosophy is pretty great. A lot of young academics and Ph.D's in philosophy are writing stuff up for it. Really great resource.
It isn't really an alternative to Wikipedia though: Wikipedia is about more than just philosophy. Similarly, the Routledge Encyclopedia of Philosophy - the big printed encyclopedia on philosophy - isn't an alternative to Britannica. It is a subject-specific encyclopedia. The two have different roles.
Kurzweil hasn't just staked his reputation on this barmy timeline, but his life too. I mean, seriously, the guy is popping vitamin pills like crazy thinking that if he can just extend his life a decade or so, the nerd rapture will finally happen and he'll get to be absorbed into the giant galactic Googlebrain.
But, no, this isn't religious enthusiasm gone too far. No, this is SCIENCE. I mean, the man has graphs, so it has to be science, right?
Yes, you can build DLLs and EXEs from IronRuby projects. I haven't done so - I just use it as an interpreted scripting language and a familiar REPL when on the Windows platform - but I've heard it is possible to do so.
On JRuby - which I am more familiar with - you get both an interactive runtime and a compiler (jrubyc) which can turn Ruby into Java.class files.
Simple: choice. Lots of people like Python, and lots of people like Ruby. Having choice is a good thing. Plus there are some libraries (not just Rails) that are Ruby only - including things that benefit.NET programmers like domain specific language tools like RSpec, Rake and so on. Some C# users have been known to use Rake on IronRuby as a lightweight alternative to NAnt, for instance.
Yes. I have used IronRuby - it is pretty nice. I don't know much about the Windows platform, and it is really pretty useful to be able to write simple Ruby scripts that can interact with.NET stuff. Scripting languages running on top of the CLR (and JVM) is pretty damn useful for a wide variety of applications and situations.
We have to go back to jailbreaking the old fashioned way with a computer and a USB cable - it'll take ten minutes rather than five now and require you to RTFM. And all because Apple wants to fix a gaping security hole. DAMN THEE DRACONIAN STEVE JOBS!!1!
...their notion of expertise is limited to only the sorts of things Silicon Valley types think are valuable. 'Social media', 'cloud computing', even Apple.
Actually, it'd be quite useful for both business and politics to be able to find if there are people on Twitter who are influencing people on science, who are influencing our democratically-elected representatives, our media figures and so on.
(After writing that, someone from PeerIndex has just responded to me moaning on Twitter and said that they are tracking a wider variety of categories and will be exposing that in the future.)
I have used Mechanical Turk once: during my undergraduate studies, I wanted people to test out a survey for a psychology of religion class. I put it up on MTurk for $0.75 each. I got really great results, but the best bit was some of the responses in the "any other comments" field I included at the end. People saying things like "this was really interesting and has made me really think".
I am really not sure about it. It really is a stark contrast to some of the Web 2.0 love-in mentality: for all the high minded discussion of community and openness, you dig down and there is this small army of people being paid sub-sweatshop wages to keep it all going.
The Turkers are doing a really good job in shit circumstances with really shitty pay. Go be nice to them if you can. Give them something interesting to do and pay them a bit more than the standard shit rates they get.
Yeah, it would be handy if there was some way of moving electricity from one place to another. Some sort of national grid service where power can be routed from the place it is being produced to the place it is required. I'm sure someone is working on something like that...
I agree: Grand Central Dispatch/libdispatch is fantastic. HawtDispatch does similarly for Java and Scala. If you are using Scala, you even get a very, very similar syntax to the C/C++/Objective-C GCD code. Very cool. Much simpler than handling your own threads.
Government agencies can contact me in writing if they must. In fact I find it far more likely that an unkown number calling me and claiming to be the police is actually a fraudster.
Someone I know (rather, someone I know online) got a call out of the blue from the police claiming that someone had made a complaint of harrassment against him (for the grave crime of mocking them online). He didn't take down any details (like the name and number of the officer who called him, for one). He realised it may have been a scam, so he phoned the police back and they denied all knowledge of having called him or their being a complaint made against him.
Strangely enough, it wasn't a scam: someone had complained about him, then the police realised it was a load of bollocks and tried to hush it up by pretending they hadn't called him. Eventually when someone else called them on it publicly, they admitted that they had called him and then realised that it was a mistake.
Very strange behaviour indeed. Not sure if it is stupidity or malice. I want to attribute it to stupidity, but both explanations work.
It's pretty simple. To represent gender, use a string. Have as default values the lower case string "male" and the lower case string "female" and run any input you get through a function to check if, having removed whitespace and lower-cased the input, it matches the string "male" or "female", you normalize it to that. If someone does not wish to specify a gender, store that as nil or as an empty string.
If someone wishes to specify something other than male or female, allow them free text entry. That is all. Then if you need to find people who are neither male or female, you just query your database for * where gender != "male" && gender != "female".
Same goes for sexual orientation. Pick some sensible default strings for your culture - straight, gay, bisexual etc. - and then for all the "I'm queer, not gay", "I'm asexual", "I'm a furry otherkin" or whatever - let them put in whatever they want as a short string-like token.
Google Charts API is complicated? On the rare occasion I use it, I have it abstracted away by a library. There's a pretty nice one in Ruby that I've used, and a few minutes Googling revealed libraries available for Java (javagooglechart), C#/.NET (googlechartsharp), Python (pygooglechart), Perl (Google::Chart on CPAN) etc.
I've built and maintained a web app for use in enterprise. Fortunately, all my users access the app using decent browsers. In the last month, we have only had ONE visitor who uses IE6, even though the majority of users are on Windows XP. I still build in IE6 compatibility using the IE8.js script, but don't go out of my way to support IE6. If someone has a problem, my general line of advice is to upgrade to Firefox or a later IE version.
The browser is the least of the corporate IT failures I see though. Complete incompetence with e-mail seems the main failing. I recently had to patiently explain that if you want to keep e-mail in sync between two laptops and an iPhone, POP3 is not suitable. You'd think that if you had Exchange server, you'd actually let it use the Exchange ActiveSync protocol. Of course not. No ActiveSync. No IMAP. Just POP3. When management were interrogated about this, the problem of keeping three e-mail clients in sync using POP3 didn't arise: they read their e-mail and then just delete them. When they need to read an e-mail that they have deleted, they phone up one of the people who is trying to juggle their three clients who forwards it on.
The persistence of IE6 is only one of many, many ways corporate IT can suck. Unfortunately, it is one which has the negative externality of making the lives of web developers suck. Once IE6 disappears, corporate IT will still suck. It'll go back to being comedy rather than tragedy.
I haven't seen Locale on Android, but I've seen MarcoPolo for the Mac. It's not driven by location but does settings driven on "context", which you can determine on the basis of a whole bunch of properties including networking, applications, accessible hosts, power status and time of day. You can specify how much each rule should count, so it can 'fuzzy match' contexts. When the context changes, the app can change a whole bunch of different things for you (and, for instance, connect to network shares, set status messages in iChat and so on).
Yes, OmniFocus does it - and Remember The Milk does it too, if I recall correctly.
I plan to build all my own location services stuff (possibly open source it if it doesn't suck too much) so it works with both my iPod touch and with my nasty old phone (and any other way I can think of moving data around).
Slashdot Mirror
A few years ago, in Britain, one of the universities finally shut down their "BA Independent Studies" programme. Three years of studying whatever the hell you want, with the ability to go to any lecture courses in any subject, so long as at the end of it, you had a coherent programme of research.
No, I'd say that part of what is scary about zombies is that a loved one might turn into a zombie. Suddenly, the hero's girlfriend becomes a zombie - or, worse, his own mother or father or best friend - and he is tasked with destroying this zombie version of a loved one who still seems to be in some sense living.
Putting on my poncy lit-crit hat: surely, this shows that one cannot simply pine for some Lockean state of nature and imagine that civilization could just fall apart without it affecting you. (Libertarians, I'm talking to you. Heh.)
PROTIP: philosophical zombies != zombie-movie zombies.
P-zombies would not differ from human beings - certainly, if you saw two walk past you on the street you couldn't tell the difference. On some accounts, you might be able to tell the difference if you had them wired up in a neuroscience lab. The zombies of philosopher's imaginations are quite different from the zombies of Resident Evil games (etc.): in fact, if p-zombies were like movie zombies, they wouldn't be very interesting in philosophy of mind (although there would be some interesting biological questions we might have about them...)
His point isn't that sometimes you need to write code that solves scalability issues: it is that a lot of Web 2.0 douchebag types think that they need to write their application using NoSQL and Node and all sorts of other fancy scalability shit, when in fact scalability is not their concern yet and that they should shut the fuck up, install a mature RDBMS (like Postgres or The Other Database) and use it until they have a scalability problem.
But they don't. Because geeks love to play with the new shiny toys and brag about it on their blogs.
What's the problem? That's perfectly readable*!
* For values of 'readable' defined by graduate-level study of philosophy.
You should probably be looking at the Internet Encyclopedia of Philosophy too. I've found that roughly my reading pattern when I hit a new topic in philosophy is: IEP, SEP, textbooks, monographs, research papers. Some of the SEP articles - especially in philosophy of language and epistemology - can be a little intimidating and get into the logical symbolism a bit too rapidly. The IEP articles are a little gentler, and are definitely suitable for undergraduates in a way that some of the SEP articles aren't.
Let me guess: 9-11 truther? ;-)
Stanford Encyclopedia of Philosophy is pretty great. A lot of young academics and Ph.D's in philosophy are writing stuff up for it. Really great resource.
It isn't really an alternative to Wikipedia though: Wikipedia is about more than just philosophy. Similarly, the Routledge Encyclopedia of Philosophy - the big printed encyclopedia on philosophy - isn't an alternative to Britannica. It is a subject-specific encyclopedia. The two have different roles.
Kurzweil hasn't just staked his reputation on this barmy timeline, but his life too. I mean, seriously, the guy is popping vitamin pills like crazy thinking that if he can just extend his life a decade or so, the nerd rapture will finally happen and he'll get to be absorbed into the giant galactic Googlebrain.
But, no, this isn't religious enthusiasm gone too far. No, this is SCIENCE. I mean, the man has graphs, so it has to be science, right?
Yes, you can build DLLs and EXEs from IronRuby projects. I haven't done so - I just use it as an interpreted scripting language and a familiar REPL when on the Windows platform - but I've heard it is possible to do so.
On JRuby - which I am more familiar with - you get both an interactive runtime and a compiler (jrubyc) which can turn Ruby into Java .class files.
Simple: choice. Lots of people like Python, and lots of people like Ruby. Having choice is a good thing. Plus there are some libraries (not just Rails) that are Ruby only - including things that benefit .NET programmers like domain specific language tools like RSpec, Rake and so on. Some C# users have been known to use Rake on IronRuby as a lightweight alternative to NAnt, for instance.
Why use IronRuby when you have IronPython? Because it is Ruby. Duh! ;-)
Yes. I have used IronRuby - it is pretty nice. I don't know much about the Windows platform, and it is really pretty useful to be able to write simple Ruby scripts that can interact with .NET stuff. Scripting languages running on top of the CLR (and JVM) is pretty damn useful for a wide variety of applications and situations.
So, Oracle are suing Google and making the JVM a less viable platform.
And Microsoft are pulling back on resources for IronRuby.
Looks like it may finally be time for the LLVM to step up to the plate and provide an open source alternative. Here's hoping...
We have to go back to jailbreaking the old fashioned way with a computer and a USB cable - it'll take ten minutes rather than five now and require you to RTFM. And all because Apple wants to fix a gaping security hole. DAMN THEE DRACONIAN STEVE JOBS!!1!
...their notion of expertise is limited to only the sorts of things Silicon Valley types think are valuable. 'Social media', 'cloud computing', even Apple.
Actually, it'd be quite useful for both business and politics to be able to find if there are people on Twitter who are influencing people on science, who are influencing our democratically-elected representatives, our media figures and so on.
(After writing that, someone from PeerIndex has just responded to me moaning on Twitter and said that they are tracking a wider variety of categories and will be exposing that in the future.)
I have used Mechanical Turk once: during my undergraduate studies, I wanted people to test out a survey for a psychology of religion class. I put it up on MTurk for $0.75 each. I got really great results, but the best bit was some of the responses in the "any other comments" field I included at the end. People saying things like "this was really interesting and has made me really think".
I am really not sure about it. It really is a stark contrast to some of the Web 2.0 love-in mentality: for all the high minded discussion of community and openness, you dig down and there is this small army of people being paid sub-sweatshop wages to keep it all going.
The Turkers are doing a really good job in shit circumstances with really shitty pay. Go be nice to them if you can. Give them something interesting to do and pay them a bit more than the standard shit rates they get.
"Fantastic for those who live in sunny states."
Yeah, it would be handy if there was some way of moving electricity from one place to another. Some sort of national grid service where power can be routed from the place it is being produced to the place it is required. I'm sure someone is working on something like that...
I agree: Grand Central Dispatch/libdispatch is fantastic. HawtDispatch does similarly for Java and Scala. If you are using Scala, you even get a very, very similar syntax to the C/C++/Objective-C GCD code. Very cool. Much simpler than handling your own threads.
Someone I know (rather, someone I know online) got a call out of the blue from the police claiming that someone had made a complaint of harrassment against him (for the grave crime of mocking them online). He didn't take down any details (like the name and number of the officer who called him, for one). He realised it may have been a scam, so he phoned the police back and they denied all knowledge of having called him or their being a complaint made against him.
Strangely enough, it wasn't a scam: someone had complained about him, then the police realised it was a load of bollocks and tried to hush it up by pretending they hadn't called him. Eventually when someone else called them on it publicly, they admitted that they had called him and then realised that it was a mistake.
Very strange behaviour indeed. Not sure if it is stupidity or malice. I want to attribute it to stupidity, but both explanations work.
It's pretty simple. To represent gender, use a string. Have as default values the lower case string "male" and the lower case string "female" and run any input you get through a function to check if, having removed whitespace and lower-cased the input, it matches the string "male" or "female", you normalize it to that. If someone does not wish to specify a gender, store that as nil or as an empty string.
If someone wishes to specify something other than male or female, allow them free text entry. That is all. Then if you need to find people who are neither male or female, you just query your database for * where gender != "male" && gender != "female".
Same goes for sexual orientation. Pick some sensible default strings for your culture - straight, gay, bisexual etc. - and then for all the "I'm queer, not gay", "I'm asexual", "I'm a furry otherkin" or whatever - let them put in whatever they want as a short string-like token.
Google Charts API is complicated? On the rare occasion I use it, I have it abstracted away by a library. There's a pretty nice one in Ruby that I've used, and a few minutes Googling revealed libraries available for Java (javagooglechart), C#/.NET (googlechartsharp), Python (pygooglechart), Perl (Google::Chart on CPAN) etc.
I've built and maintained a web app for use in enterprise. Fortunately, all my users access the app using decent browsers. In the last month, we have only had ONE visitor who uses IE6, even though the majority of users are on Windows XP. I still build in IE6 compatibility using the IE8.js script, but don't go out of my way to support IE6. If someone has a problem, my general line of advice is to upgrade to Firefox or a later IE version.
The browser is the least of the corporate IT failures I see though. Complete incompetence with e-mail seems the main failing. I recently had to patiently explain that if you want to keep e-mail in sync between two laptops and an iPhone, POP3 is not suitable. You'd think that if you had Exchange server, you'd actually let it use the Exchange ActiveSync protocol. Of course not. No ActiveSync. No IMAP. Just POP3. When management were interrogated about this, the problem of keeping three e-mail clients in sync using POP3 didn't arise: they read their e-mail and then just delete them. When they need to read an e-mail that they have deleted, they phone up one of the people who is trying to juggle their three clients who forwards it on.
The persistence of IE6 is only one of many, many ways corporate IT can suck. Unfortunately, it is one which has the negative externality of making the lives of web developers suck. Once IE6 disappears, corporate IT will still suck. It'll go back to being comedy rather than tragedy.
I haven't seen Locale on Android, but I've seen MarcoPolo for the Mac. It's not driven by location but does settings driven on "context", which you can determine on the basis of a whole bunch of properties including networking, applications, accessible hosts, power status and time of day. You can specify how much each rule should count, so it can 'fuzzy match' contexts. When the context changes, the app can change a whole bunch of different things for you (and, for instance, connect to network shares, set status messages in iChat and so on).
Anyone know of anything like this on Linux?
Yes, OmniFocus does it - and Remember The Milk does it too, if I recall correctly.
I plan to build all my own location services stuff (possibly open source it if it doesn't suck too much) so it works with both my iPod touch and with my nasty old phone (and any other way I can think of moving data around).