Podcast 51Revision #2, 4/29/2009 6:56 PM
Podcast 51Revision #10, 12/13/2009 6:38 PM
184.108.40.206: "Changed erroneous "Runner Radio" to correct "RunAs Radio". Hyperlinked "RunAs Radio" and "Serverfault"."
Spolsky: Edit me!Atwood: Edit me!
Pryor: Edit me!
Atwood: That's true...
Spolsky: And it's not analogue in the sense that there's no..when it goes from 3:12 to 3:13 as it will in a moment, there's no ambiguity as to when it happens, it's not gently shifting it just (popping noise) changes.
Atwood: Well I have a clock..sort of, clock fetish, I have a couple.
Atwood: I have a nixie clock, I have a pong..
Spolsky: What's a nixie clock?
Atwood: Nixie is like those old timey, like they would have on Soviet warships? Digital readout that's basically it's, it's hard to explain, it's basically...
Spolsky: Very analogue.
Atwood: Very analogue but it's as digital as you could get at the time.
Atwood: And that's known as a Nixie tube, basically a tube that has all the numbers in it, it's really cool!
Spolsky: Oh and they turn?
Atwood: No they don't turn! They just light up individually it's like neon...
Spolsky: OH! Oh yeah! It's like, in other words, every position has like a 1 2 3 4 5 6 7 8 9 0 neon thing?
Atwood: Yeah! Yes.
Spolsky: Oh that is awesome!
Atwood: It's really cool!
Pryor: What's it called then when they have the 8 and then it only lights up parts of the 8?
Atwood + Spolsky: Oh that's just LED
Atwood: LED's are good though
Spolsky: We went to, Michael and I just came from the Computer History Museum in Mountain View
Atwood: Yes! That's great, I'm glad you were able to go to that.
Pryor: An interesting part about that trip was the Joel had worked on every single computer in the museum.
Atwood: That's because he's very old.
Pryor: He could tell me how everything worked.
Spolsky: But the ones I had worked on …
Pryor: There was an Enigma machine and he's like
<lots of laughing>
Pryor: (in a silly voice) “When I was just a lad”
<lots more laughing>
Spolsky: They have a giant Babbage computer that they just built.
Pryor: Yes! He worked on one of those too!
Spolsky: Honestly, it looked like a spaghetti maker, it looks like you feed in the pasta dough on one side and it goes “raurourghrourghrough” and spaghetti comes spurting out the other side.
Pryor: It was kind of weird to see all that old equipment and just kinda get excited about it? I feel like such a nerd like “Oh my god! It's an Apple II GS” I remember my Apple II GS.
Atwood: Yeah, they have like every computer ever, which is really cool.
Pryor: He was getting excited about the HP calculators, which is super nerdy.
Spolsky: Yes! I've used an HP35, HP65, and an HP 41C. Although not the 65 so much. But the 41C oh my god it had alphanumerics, it was LCD not LED – it had letters! On the display! It could say “Hello”, and NOT just by typing 07734
<all 3 laugh>
Spolsky: and holding it upside down!
Pryor: What other words can you write...you can write “Boobless” on a calculator
Spolsky: And “Shell oil”.
Spolsky: So what's new in the way of StackOverflow this week?
Atwood: Well you know it's odd sitting here with you guys and actually looking at you whilst we do the podcast, which is weird, for number one. Usually I'm just staring into the computer!
Atwood: And I'm not actually sitting in front of the computer which means I can't actually research
Pryor: You can use my computer if you want
Atwood: No, no, I don't need to, uh
Pryor: You've got StackOverflow memorised, at this point?
Atwood: So, those phrases, if you wanted to find them there was literally no way to.
Spolsky: But who wants to search the comments? They're just comments.
Atwood: Well that's what I thought too but I kept finding these comments that were really funny and really topical, to the point that I would like
Spolsky: You want to link to them.
Atwood: I didn't want to link to them per-Se, but I would actually enter them in twitter, because I thought they were so interesting and amusing
Atwood: And people seem to like them, they responded to them and I just found that the quality of comments was surprisingly high.
Atwood: And it deserved to be sort of
Spolsky: Elevated a little bit?
Atwood: Yeah, elevated a little bit!
Pryor: But technically you can go back to the old way if you want, you said. Are you going to allow that in the settings?
Atwood: Yeah, we're not big on preferences, I'm kinda down on preferences as a design technique?
Spolsky: You mean the user chooses their preferences?
Atwood: Yeah! I think that your default should be good enough for like, more than 95% of the people? Otherwise you're kind of screwing up.
Spolsky: Well anyway yeah..
Atwood: But in this case I actually believe that this is a preference that we will add to the system because you could make the argument that you really don't want to see comments unless they're really really good or maybe not even at all. So you can actually set the threshold at which they get escalated to the question page? Right now the threshold is 0 essentially, the last 5 comments will appear if there are 5 comments, and then..
Spolsky: You mean they don't, you no longer have to click to see them?
Atwood: Yes! You no longer have to click to see them, because what I found was that if I was interested enough to look at a question, I was interested enough to read pretty much every comment in the question, so the normal pattern for me at least was: Click on a question, and then click click click click click – expand all the comments.
Spolsky: That makes sense!
Atwood: As I'm reading and then I'm like, why do I keep doing that? That's just useless clicking!
Spolsky: It's fun! Clicking is fun!
Spolsky: <laughing> Makes you feel like you're in charge!
Atwood: <laughing> Clicking is actually fun! That's the weird thing about clicking! And anybody who has ever played Diablo, they'll tell you, that's pretty much an orgy of clicking, that's pretty much all you do just click click click click click and it is fun!
Atwood: But I felt with comments that it made more sense to escalate them to the page, you have to do it in a way that you're not overwhelming the page with comments, so it's top 5, by date or by voting – if we have enough votes then we'll show you the top 5 by votes.
Pryor: So the threshold is set to 5 by default?
Atwood: The total number threshold is set to 5, we're thinking about adding a threshold variable in that, the comments that are escalated to the main page have to have n votes to even appear at all – in a user preference. Like you could set it to 5 and then the only comments that got voted to 5 or higher would actually be escalated to the main page for you .
Spolsky: What would be the name of that user preference? <laughs> Like what would the little radio button say?
Atwood: I don't know! I don't know that one...
Spolsky: It seems a little too complicated I mean, unless you experience that, you may not know how to set that preference.
Atwood: Yeah, well it's an advanced user setting, but some users really objected to comments being on the page at all?
Spolsky: They just, you know what? You mean you got feedback, somebody sent you an email saying they don't like comments …
Atwood: Yeah! There's not a lot, there's a minority but..
Spolsky: Yeah. Forget it. You could do anything. You could change it so that there's a button you click and it sent you money in the mail, $20 bills
Spolsky: When you click that button $20 bills arrive in the mail? And you would get somebody saying <grumpy old man voice> “Why did you change stack overflow? It was PERFECT”
Atwood: Yes. What's with all this stupid money you keep sending me!?
Spolsky: Yeah, it's like people will, they like the status quo, and no matter what you change there will be people complaining. Which actually makes it very hard to build a site where you listen to your users.
Pryor: Well that's the big thing because, Joel's been having a conversation about all these features every single time so..
Pryor: And consulting with the people through the blog.
Spolsky: Which is good, yeah, which is fine.
Pryor: But now that they've become accustomed to that it's very difficult because ..
Spolsky: They're gonna start, I mean, every time I changed a font, I could change a font on “Joel on Software” and get, you know, 30 angry emails, and then change it back and get another 30 angry emails, 25 of them from the same people that were upset the first time I changed them “<grumpy old man voice> I just finally got used to that font that you changed”. <laughs> And have to change it all over again.
Atwood: I think you can, sort of tune that stuff out though, but I do think there's a level of feedback where you can tell if something is generally positive or generally negative.
Spolsky: I don't know I'm thinking of, what if you were Facebook right? They just did that redesign where they changed facebook into like … what'd they change it into? Twitter?
Spolsky: Tumblr? <laughs> And there were like millions of people complaining, out of hundreds of millions, but still millions, and they just said “yeah, sorry, we're not listening”.
Pryor: Now you have to vote it's like, democratic, they're doing some democratic system.
Spolsky: So how the heck do you decide – they are?
Pryor: For that, for that particular thing because, you know not for design issues or how the site works or the direction of the site but, you know.
Atwood: Well one thing I want to talk about, and it pains me a little bit to talk about this, is: I have a very ow threshold for “Meta” discussion, in other words, a Podcast about Podcasting, a Blog about Blogging.
Spolsky: Hey lets talk about this gear that we have set up here! The coffee table chair.
Atwood: I think a certain amount of it is okay, a certain low amount, I don't have a problem with...
Spolsky: Sure! Somebody should do a Podcast about Podcasting, just not us!
Atwood: Right! And it shouldn't be all the time, so it's really a question of ratios, how much time do you spend talking about yourself basically.
Atwood: And the things that you're doing. And one of the questions you asked me in email was, “Where do people go to discuss StackOverflow if not StackOverflow?”
Atwood: And, on one level I can kinda see that because, everybody in the right place, you know, they're all interested they have this one shared interest: StackOverflow. But you have to really consider the audience like, how many programmers actually come to stack overflow thinking “I want to learn more about StackOverflow”? Like, why would you even come to StackOverflow in the first place?
Spolsky: Huh? Oh yeah!
Atwood: You don't come because you care about StackOverflow, you care about getting an answer to your question.
Spolsky: Well lets divide it into our two communities, we have, that's the 86% that came from google, but the 14% that come because they want to play with it, they want to have fun, they want to try to answer questions, they want to show their knowledge, they just love answering questions for people, it's just kinda fun, they like earning points, they like the badges, the achievements, the $20 bill button that sends them $20 bills to their house..
Atwood: Who wouldn't love that?
Spolsky: I know!
Atwood: I'd hate it but they would like it!
Spolsky: Yeah..it doesn't show up in IE6, you're not seeing it? You've got to upgrade your web browser!
Spolsky: And those..so that second community might actually! But the trouble is by the time they're like, asking... There's this other issue here which is the phenomenon that all social networks will have, is that some people love the social network so much that eventually its core functionality ceases to entertain them, but they love it so much that they kinda wanna hang around and find other games to play with the system.
Spolsky: So they're on StackOverflow and they're like, “Wow this is so cool I'm gonna answer a whole bunch of questions, I know everything about FORTRAN lets see how many FORTRAN questions there are” and they answer all 6 FORTRAN questions.
Spolsky: And there are no, and they're clicking refresh, refresh, refresh on the FORTRAN tag but nothing is coming in. And then maybe they make up a question that they ask themselves <laughs> about FORTRAN and sign up for another account <laughing>
Atwood: Why isn't FORTRAN more popular?
Spolsky: Or they'll just be like, you know, “Does anybody still use (something) strings?” and then they'll go answer that under a different fake ID, and then they'll notice there's not that many other people on the FORTRAN section and they'll start to get a little bit bored, and they'll .. but they love the site so they'll just start kicking other people you know? This is what we refer to as the “Lets stop playing chess and lets start playing throw the chess pieces around the room” <laughs> game.
Atwood: Well I don't think it has to become negative though.
Spolsky: Doesn't have to, but it DOES become meta.
Atwood: It becomes meta, and I personally, I often rant in my blog about, blogging about blogging, and occasionally I will do it just to be clear, I mean it seems hypocritical
Atwood: It becomes meta, and I personally, I often rant in my blog about, blogging about blogging, and occasionally I will do it just to be clear, I mean it seems hypocritical, because I posted that on twitter about how much I dislike meta stuff, and then of course somebody immediately messages me a post I did about blogging
Atwood: And I was like, you're right, I mean occasionally I do do it, it's a question, like I said, of the ratio, how often it happens. And with that in mind, it's been the fact forever that, you know, we really discourage people from asking questions about StackOverflow on StackOverflow.
Pryor: But you guys have just spent 50 podcasts talking about StackOverflow?
Atwood: But this is a separate area!
Spolsky: This is the podcast, this isn't meta. This is the podcast about StackOverflow.
Pryor: I know but what happens is that the consumers of your podcast are so excited about this new thing that you've built, all the features of it, how it's going to work, there's also been some back and forth about should it work this way, should it work that way? You know, it's changed, you know the comments have just changed in the past couple of weeks?
Atwood: Yeah there's lots of changes.
Pryor: The way, the voting and all of the way it looks and .. so they, it's not like a normal site where you've just come to it, it's already been in existence, and you're like: this is how it works, come! Use it! But if you don't like it, go to uservoice and suggest something, but there's been all of this talk about how the site has been built, how it's going to work, and that's very interesting to the type of people that consume your podcasts and read your blogs, so they're obviously going to come to the site and ask questions like that, and that will probably die off eventually, but that's still part of what people are interested in. One of the things people are interested in.
Atwood: Well that's true, you don't want to step on people, I mean, I don't want to be the guy that says “you know what, you're really interested in our site, screw you! Go away!”, but that's not really my goal...
Pryor: <laughs> But you do say that! <laughs more> you do say that.
Atwood: Well, to the extent that I say go away what I'm really saying is, I think it needs to be a separate area? And what I object to in the current system is that we allow a certain amount of it, a certain very low amount of it is tolerated.
Pryor: So just make a new domain name, stackoverflow.stackoverflow.com
Atwood: But..but I think you..
Pryor: Stackoverflowoverflow like ..
Spolsky: Yeah that's a good one! <laughs> stackoverflowoverflow!
Pryor: Yeah Jon Skeet came up with that.
Atwood: Okay so, even if we did that! Now remember Joel, at one point we were talking about the Joel on software discussion boards?
Atwood: And I said, well why not use StackOverflow software for the discussion boards?
Atwood: And you lectured me, rightfully so, about how it's not really conducive to discussion.
Atwood: The system that we've built is very much a Q&A system, in other words, the default sort is voting, so there's no... chronology gets screwed up, and it's about one person asking a question, everybody else reacting to that question, in an answerable way, not talking amongst themselves, and this is sort of counter to what discussion is, and that's another problem I have is that our system, I don't think adapts well to pure discussion.
Pryor: Maybe pure discussion, but there's definitely...people like it to discuss things...
Spolsky: The trouble is as you make it a community, a community – whatever community means, it's an overused term, but, if it's like a village, if it's like a, the people who all work in the, you know, machamony wing of the hospital, is that a... real wing that they have? I don't know..
Pryor: No... <laughs> The marriage wing?
<lots of laughter>
Spolsky: <laughs> I'm trying to think of a wing of the hospital where people don't know each other that work together, you know what I mean. It's a community! It's the, Fog Creek programmers, it's the people who love C#, there are all these different communities, and every – why does my phone keep knocking? Stop knocking!
Atwood: It wants you to buy an iPhone!
Spolsky: Um, that's a, the text message knocking. Whatever community it is, there's certain things a community has to have, or it's not going to be a community it's just going to be boring, so if it's just q&a, at some point it's going to be very very dry and very very icky, but if once in a while somebody on special occasions has a lot of fun with a funny comment, or a funny post about the boat, or programming on a boat, or “what's your favourite programmers cartoon?” is our favourite one, those things, we want to have a small number of those, if you have that all the time the site is degrading, it's changing topic and it's turning into reddit, whereas, you know, if that's what people want to talk about because they're bored of talking about programming, go ahead, and we even can't really allow the regular StackOverflow users who are bored with programming topics, to hijack it and turn it into a site about .. the Bush administration.
Pryor: That's not going to happen, that's not going to happen though, like, the community is already there and you don't have to go in and monitor each question, like on the individual scale. It's like how you were telling me before about how Google, Larry and Serge's
Spolsky: Last weeks podcast?
Pryor: ..didn't sit around and, look at each individual website and decide if it should be in or out, the just built a system and it just worked and, eventually the meta-questions that people are asking are just going to go away, because the site is going to be more static.
Atwood: Well the other thing about the meta questions that's odd is, I occasionally have to go back and actually remove the old meta questions because it's referring to old versions of the site, when we've significantly changed the way the site works, like there's a whole discussion that is essentially obsolete, because it was a meta discussion about some feature that doesn't work that way any more, so somebody would find it and go “what the hell is this”?
Spolsky Well you could just go and hit edit and say “this is obsolete” or whatever. Vote it down.
Atwood: Well I..
Spolsky: That's the way it's supposed to work.
Atwood: Well I'm so down on meta that I just usually delete that stuff.
Spolsky: Down on metaaaa <deep voice>
Atwood: Now, one analogy that I like...
Spolsky: That's like <can't make out> Down on meta!
Atwood: I was thinking about doing a poster, “Meta is murder!”
Atwood: Yeah but, on twitter, Kevin Dente messaged me and said, he said, you know the problem you have is that you have it's like, when your kid are out of school and causing trouble, it's because you didn't give them an after school activity. That makes sense. And I said well, it doesn't really work, that's not really a good analogy, and then he said okay, refining it, it's like: “You have a school, where the students eventually become teachers, and need like, a teachers lounge.” I thought that was a better analogy in that..
Spolsky: So more smoking!
Atwood: … in this you have a teachers lounge, right?
Spolsky: So what we need is like a room with a bunch of teachers smoking!
Atwood: Well you know because it is sort of like, students and teachers, that is what StackOverflow is ultimately supposed to be, people learning about a topic together, where the students become the teachers eventually.
Spolsky: Hey lets have, like a chat room, that you have to have like 5000 points to get in to.
Pryor: There already is an IRC chat though.
Spolsky: Is there?
Spolsky: Well lets have another one, you need 5000 points for it. Jeff could you raise your microphone an inch?
Pryor: Well it's not just for the people that have, like those aren't always the people that want to discuss this stuff. You know, it's not about the people with higher reps, it's not like these moderator people need a place to go, there's people that have a rep of 300 or 400 or 50, that want to talk about the same things though.
Spolsky: No I know this isn't the way of getting that off it's just the way of making a separate teachers lounge, a meeting place. I also want to have... did we talk about that “flag” thing? The idea that – it's kind of weird because somebody, I mentioned on, I was talking to the Microsoft people I think... some Microsoft people, I mentioned that the New Zealand .NET discussion group people wrote to us to complain that all their active users had left and gone to StackOverflow, draining them of all essence and there was no longer a cool New Zealand .NET discussion group, and I thought what we should do is add the ability to turn on a little flag next to your name if you wanted to, with your nationality?
Atwood: We have, Project
Atwood + Pryor: Euler does that
Spolsky: Yeah and so does Airliners.net (?) It's pretty cool! Then you could have a teachers lounge, and a New Zealand lounge, and a .. you keep bringing up that boat question!
Pryor: I love the boat question, the reason is because last week I was sitting at the Fog Creek lunch and I told everyone we should move to a little, like, private island in the Bahamas, and we can make a little programming commune...
Spolsky: The whole company?
Pryor: Yeah! Everyone could live there! And we could write software from there instead of ..
Spolsky: Oh so that's why you were ordering up all kinds of satellite dishes!
Pryor: Yeah! Then we needed to built solar panels in the roof, and we had to figure out how to harvest rainwater, had to figure out how we were going to get satellite internet, those sorts of things. And so the answer to this question has lots to do with, you know, internet access when you're in the ocean, and things like that, and getting power whilst you're in the ocean, and the same answers I was researching.
Pryor: And it turns out that the guy who answered said that basically, un-metered internet access is $5000 per month for satlelite internet, which is a little high...
<Atwood laughs a lot>
Atwood: <laughing> That's relatively high!
Spolsky: And it's probably not that high bandwidth either, and it's got to be high latency...
Pryor: 2 Megabits per second
Spolsky: Euargh, that's crappy – we've got one person watching Hulu and it's over.
Atwood: Well you guys live in NYC! You're gonna move to an island where there's no other people, why the extremes?
Spolsky: Going from an island that's chock-full of people to an island that's got no people.
Pryor: I like the idea of just going somewhere and just, having nothing, having to build everything. You have to like, survive. On your own.
Atwood: It seems more attractive as an idea than an actual..
Pryor: You may very well be correct.
Pryor: You make it out there and a week later...
Spolsky: Why don't you take the summer interns as their inauguration?
Pryor: Well we love solving problems, and I think the people at the office were really excited about it, I mean not the reality but just the idea of like, you know how do we set this up, how would we get all of this stuff there? And then what if we have to build a shelter if there's a hurricane?
Atwood: Well you know it's funny you bring that up, because that's what I kinda objected to in that question, where it was so hypothetical that it was like, “what if we did this?”
Atwood: And I was like, this is so impractical, right? I mean I wasn't convinced that the original..
Spolsky: Somebody's going to do it!
Atwood: But I wasn't convinced that the origional poster was serious, first of all that was my, the root concern, and then the other concern was that..
Atwood: Well it did, actually, once we brought it up on the podcast, somebody posted a really nice response to it, though, you know.
Spolsky: Let me play a listener question that's a little bit related, not a lot, but a little bit.
Peter: Hey Joel and Jeff, this is Peter in Dallas, Texas, my question is related to kinda one of the original goals of the site, in terms of finding answers to questions that are kind of buried somewhere in some old email string or some broken forum or something like that, what do you think about the practice of, that if somebody does find an answer in a place like that, but then they don't find it on StackOverflow, posting that question, and then answering it themselves, to create a new canonical place for that problem to live. Thanks!
Atwood: It's been allowed from the earliest days!
Spolsky: I think it's a great idea, yeah!
Atwood: It's even in the FAQ.
Spolsky: Just don't, you know, don't cut and paste because that's a copyright infringement.
Atwood: Oh he's talking about transferring information from one place to another?
Spolsky: He's talking about, you ask a question, you can't find the answer on StackOverflow, you go to “the website that has a hyphen in the domain name”, you find the answer in a question there, and you say “you know what, I wish this question and answer were on StackOverflow”, but not on the website with the hyphen in the name, so lets take, you basically, bring it into stack overflow by re-asking the question and re-answering it there, and you just have to do it so that it's not a copyright violation, basically, which means you have to rewrite the question and the answer.
Atwood: Wait wait wait, you said two different things there you said “Asking a question you couldn't find the answer to”, but he didn't ask somewhere else.
Spolsky: It was answered somewhere else.
Atwood: It was asked somewhere else too, but not with a good answer?
Pryor: No, with a good answer.
Atwood: He found the answer..?
Pryor: But it wasn't on StackOverflow.
Spolsky: But it wasn't on stack overflow: should he then create a new StackOverflow entry with a question and answer for that topic?
Atwood: Actually, I don't know that seems redundant on some level. Usually the reason you come onto StackOverflow and ask something is you don't really like the presentation of the information that you've found, either it's not good enough, or it doesn't cover the right bases, it's not complete..
Atwood: There has to be some other reason than just copying, I think you want to enhance it in some way.
Spolsky: Well just being on StackOverflow instead of on a site that pretends you have to pay to get the answer you want, even if you don't, that's an enhancement of the answer.
Pryor: The fact that it's a wiki too!
Spolsky: Yeah. It's a wiki, and our site is easier to find, and, you know.
Pryor: It loads faster probably too.
Spolsky: Yep! There you go!
Atwood: So that's, that's defensible. I mean to do that.
Spolsky: Anyway why did I bring that up, it was somehow related to something we were talking about right before that.
Atwood: Well the Cuba reference, apparently one thing I didn't notice was the boat question? Was a reference to “The old man and the sea”?
Atwood: Santiago? And Cuba?
Spolsky: Oh it was really?
Atwood: Yep. Supposedly, somebody brought it up – it's been so long since I read “The old man and the sea” that I'd forgotten, but somebody pointed it out.
Pryor: Did he have a cat named Ender too?
Atwood: Yeah the cat named Ender was a weird addition, that was not in “The old man and the sea”!
Spolsky: Um, I've noticed that “how is babby formed” has been removed from Yahoo! Answers, sadly.
Atwood: Ah! I know, that is sad.
Spolsky: I found it in the google cache fortunately and was able to reproduce it for a slide .
Atwood: You should re-ask it!
<lots of laughter>
Atwood: <laughing> I'm sure that's done every hour!
Spolsky: I think it might be a troll.
Spolsky: Based on the picture of the guy who asked it, who has a shaved head with some kind of logo carved into his shaved head.
Atwood: Yeah. Well you always suspected that “how is babby formed” was never a real question.
Spolsky: A legitimate question, but if you just go to Yahoo! Anwers right now, you'll find questions like that, when I went to Yahoo! Answers, the current question was like, “Who eats the ants after I squish them?”
Pryor: What happens to them?
Spolsky: What happens to them – no it was, “What eats the ants after I squish them?”
Atwood: You know Michael would probably vote to keep that open.
Spolsky: Well that's a great Yahoo! Answers question!
Atwood: <laughs> Well they don't set any boundaries right?
Pryor: You're changing my argument!
Atwood: I'm just teasing.
Pryor: Because it's not that I want everything to be open, I just think, the truth of it is that I was afraid that people were more, they were getting this moderator disease you know, “That's not programming related! That's not programming related! And this isn't even close!” you know, and they were just closing things, and part of posting that question gave me the answer which was totally fine, there's just as many people out there opening questions as there are closing them, and that's fine, that's what's going to happen right? There's always going to be questions that people close..
Atwood: Well let me interject, what's funny about that is I've gotten complaints from both camps, the people who like to close things complaining it's too hard to close things, I swear to god, and then the people who like to open stuff claiming it's too hard to open things. And I had to tell them like...
Pryor: Maybe that's good!
Atwood: Well the person I was talking to actually liked to close things, and he said, you know, it's so much easier ..
Atwood: And he was like, “It's too easy to open stuff! You made it real hard for us, we like to close things, and you made it far too difficult!” And I'm like, you know it's the exact same process in reverse?
Spolsky: To close and open something.
Atwood: The only, in fact, the only advantage they have is that questions start out open, obviously, I'm not going to start out questions closed, that's ridiculous, he's like “oh we have to have 5 people to close something!”.
Pryor: What was that question, it has 2 votes to reopen it, and when I went back and looked at it it was, um, it had 0.
Atwood: They age. The reason we age the votes, is because we don't – the reason that became an issue is when they're offensive, when you flag things as offensive? Stuff will be just offensive enough, to accrue like 1 vote per week, you see what I'm saying?
Atwood: It's not really offensive, but then after 5 weeks...
Atwood: So then, a lot of votes do that, not because we're singling you out.
Spolsky: <laughing> Somebody comes in at 12:30 on Sunday <incomprehensible>.
Atwood: We're not singling you out, we just felt we had to do that because you've got to capture the velocity of these things.
Pryor: So these things they're just like..
Atwood: They're going up and down, back and forth forever yeah..
Pryor: Right, right.
Spolsky: Let's take another listener question here.
Questioner: Hi Jeff and Joel, this is <name>, I'm a computing engineer student at <something> state, I'm wondering about how you guys feel about whether software engineers should know more about how the computer actually works, like micro controllers and... just basically how the computer actually runs, the guts and the internals and the bare metal stuff, whether they should know this or just abstract everything away, I'd like to see what you guys think.
Atwood: That's a good question.
Pryor: They teach that in college, at least...
Spolsky: They do.
Pryor: At least where I went to school
Spolsky: Sometimes they do, at least, there's all kinds of levels, where did you say you went to? San Jose?
Pryor: It's more of like, an electrical engineering type of thing.
Pryor: It's like something you learn and then, once you understand it..
Spolsky: There were these classes where you learnt to make like an <and? N?> circuit, with transistors?
Spolsky: I never took that one.
Pryor: I had to built like, a little computer that read bar codes, I was telling you about that.
Atwood: Well isn't learning C at some level a little like learning about hardware?
Spolsky: Well it's halfway there.
Atwood: Yeah! Because you're going down a level!
Spolsky: Yep, but there's another level below that where you're actually..
Pryor: Like assembly..
Spolsky: You're learning logic circuits and you're learning, how a shifter might be assembled, and transistors...
Spolsky: That is a pretty low-level book.
Atwood: So I think if you're interested in this topic, if you wonder if you should be learning that, you should look at Charles Petzold's “Code”, just look at it.
Atwood: And like, pick it up and browse through it and see if it really appeals to you?
Atwood: You probably should.
Spolsky: I sorta feel like that stuff should be interesting to programmers.
Atwood: It is! It's fascinating!
Spolsky: When Michael and I went to the computer history museum, everything we saw there, like, if you're gonna be a good programmer, you have to be the kind of person that is wondering how that... punch-thing worked..
Atwood: I disagree..
Spolsky: And wondering what all those transistors were doing, and...
Pryor: I don't think you have to feel that way, to be a good programmer, it just turns out that good programmers are..
Spolsky: I'm confusing correlation and causation.
Pryor: Exactly. But I think you don't have to be... to know how a microphone controller works, or be able to solder things together and make circuits.
Spolsky: No, it's true that you don't have to but if you take a roomful of good programmers, and took them to the computer history museum, if they don't all get excited and point jumping up and down saying “Oh my god! That thing only held 100 bytes? That's like 9 closets full off...”
Pryor: Well what about building your own computer? Because I know Jeff was a big fan of building your own server.
Atwood: Yes! I am, and I agree – it's not required, like I don't lose respect for programmers who are totally uninterested in hardware.
Spolsky: But you just have to find that stuff interesting, or there's something wrong.
Pryor: When we first started Fog Creek and we were in an apartment, on John Street, down town Manhattan, and it was just an apartment that we rented, Joel and I were working there because we needed an office to go to, we did everything ourselves back then you know, and it was kinda ridiculous because I convinced Joel that we needed a – I might have told this story before...
Atwood: Well actually Joel has told this story before, you haven't...
Pryor: I was like, i'm going to build our own server!
Atwood: Yes, yes.
Spolsky: No it was build our own computer, it was going to be a computer.
Pryor: Yeah, we built it and then any time it was hot out, the computer would just blue screen..
Pryor: and we never knew why...
Spolsky: Microsoft! <laughs>
Pryor: and eventually we figured out that I forgot to put the, um..
Spolsky: The glue!
Pryor: No not glue, the ..
Atwood: The thermal compound?
Pryor: Yeah the thermal compound, you've heard it before – sorry, audience.
Spolsky: Not everybody has heard every episode it's okay to sometimes repeat things.
Pryor: Okay, and so, I figured out what happened.
Spolsky: Not everybody has heard every episode it's okay to sometimes repeat things.
Atwood: Okay here's the way I look at this is, do you feel like you learnt something from that?
Pryor: Yeah! Definitely, but I would never do it again.
Atwood: But you never have to do it again!
Pryor: And it was fun to do it, but you spend so much time on these ridiculous discussion boards looking up the motherboard that you bought and with the video card that you bought and you have to change the BIOS and it was all these weird, things that you had to know and get different versions of, and it was kind of voodoo magic? To me at the time, but then when you get it working you feel like <happy sigh> that's amazing.
Atwood: Yes, well you feel like you learnt something this is the voyage of discovery that everybody has, right? And actually, I was on another podcast today, RunAs Radio talking about Serverfault, and it was fun to finally talk about Serverfault, because you know, we spend so much time on programming topics, and I love programming – but it's also fun to know about other things. You know, it's fun to be multi-disciplined on some level.
Spolsky: Yeah! That's, this is the whole point, you have to be – I'm not saying you have to know that stuff, I'm saying you have to find it interesting. If you don't find it interesting, you're probably not a serious programmer, and you're not going to find programming that interesting, you're not going to find learning more about programming interesting, you're ..
Atwood: But I don't think that's always true, I mean the puzzle stuff – I, the puzzles drive me crazy, I have no interest in the puzzles?
Spolsky: Okay, you don't have to be interested in puzzles!
Atwood: Well a lot of people like the puzzles! And it's not like I would hold that against them, you know, but I think it varies.
Spolsky: That's okay, I don't think that puzzles... I think like, umm, PDP11's and, old computers, and transistors and..
Atwood: Did you see the live demonstrations where they power one up whilst you were there?
Spolsky: Unfortunately no, we didn't time it right.
Atwood: I got to see them playing space war on that, it was awesome!
Spolsky: Aww, that's cool.
Atwood: It was very cool. But yeah, it's great. But now, one clarification on what I said earlier? I think a lot of times, learning C and learning how the hardware works, it's not that you;re going to go and build a computer, you know, but I think – let me draw a rough analogy. Sometimes we do stuff on StackOverflow, because we feel that if we don't do it enough to understand it, then, we don't understand it enough to even let other people do it for us. Like advertising. Like, I felt like we needed to do all of our advertising internally? Handle dealing with advertisers, how we place ads on the site, actually get that together so we can understand how that works.
Pryor: That's just bootstrapping right? That's what we did when we started Fog Creek, we wanted to do everything ourselves first.
Spolsky: We wanted to, you know how these things worked before we started outsourcing.
Pryor: Yes, but if we did it again I wouldn't do it myself.
Pryor: I wouldn't fix the toilet in our office ever again, even though I did once..
Spolsky: That's a famous thing that happened in the early days of Fog Creek, Michael spent an afternoon ..
Pryor: I had to go get a plumbing part and ..
Pryor + Spolsky: Fix the toilet.
Pryor: And Joel installed the thermostat on our wall once, he put a new thermostat in.
Pryor: Just stupid stuff like that, like ..
Atwood: That's extremely off topic.
Atwood: But this stuff is incidental like, knowing how the hardware works, knowing how advertising works, this is all relevant to me running StackOverflow, right, on some level?
Spolsky: Right, sure.
Atwood: It doesn't mean I should do it forever, and if I continue to do it you should step in, and say, that's ridiculous stop doing that <laughs>
Spolsky: Wait, we have a list of things here I'll bring them up after this podcast...
<lots of laughter>
Pryor: We should make a StackOverflow for people building computers.
Spolsky: Do we have a, um Michael do you have a … oh yeah SystemBuilders.
Pryor: Because there's a definite question and a definite answer to a lot of these questions.
Atwood: Yes, it's very q&a driven, I totally agree with that.
Spolsky: And plus, StackOverflow would be great for that because when somebody asks “What's the best Video card?”, that's something that is going to be constantly updated and constantly – having our wiki ability to have things be constantly up-to-date.
Atwood: Or just have things not be utterly wrong, that's really how I look at it.
Atwood: It's not that they're going to be perfectly up to date, it's just that the odds of them being totally wrong are lower.
Spolsky: The thing, you know what's amazing, is that sometimes you ask a question and the correct answer is “Yep, sorry, this just cannot be done”, and the vendor fixes it! In the next release.
Atwood: And then it can be done.
Spolsky: And then it can be done! But think of all the web pages in the world that tell you it can't.
Spolsky: Don't ever believe a web page that tells you you cannot do something. If you believe you can fly: you can fly. Michael did you have a StackOverflow question that you wanted to talk about?
Pryor: Hmm...I really liked the boat question...
Spolsky: We did that last week!
Atwood: <laughing> And we talked about it already!
Spolsky: <laughing> Talked about it today too!
Atwood: <still laughing> Do you want me to get you a t-shirt that has the boat question on it? You can wear it around?
Pryor: So, there's one I found on the “Most active for the week”, it's called “Black hat knowledge for white hat programmers” .
Atwood: Give us the flavour of the question.
Pryor: So, the question is basically, you know..
Atwood: And what's the number? Give us the number.
Spolsky: The number is 772596.
Pryor: “To what extent do you think the honest programmer needs to know the methods of malicious programmers?” basically is the question so..
Pryor: So how much time as a programmer, you know, is it fair for you to learn about all these kind of vulnerabilities and … because in some ways knowing that information is kind of...
Pryor: Yeah. It's also kind of dangerous because, I'm just saying from other peoples perspective if they came in in looking at you reading some websites...
Spolsky: Or like homeland security would cut you off.
Pryor: ...about how to build a blue box for example you know..
Spolsky: A what box?
Pryor: I'm going way back ..
Atwood: A blue box!
Spolsky: What's a blue box?
Atwood: It's the classic, subverting the phone network free long distance box.
Spolsky: Oh that was a blue box?
Atwood: It's more of an analogue, it's all digital now that would never work.
Pryor: There was one in the history museum. I did that in college, I figured it out and...
Atwood: They have one in the computer history museum?
Pryor: Yeah! Steve Jobs' ones...
Spolsky: Yeah they have Steve Jobs' ones that he built with, whatshisname, didn't he hang out with captain crunch?
Atwood: Steve Jobs built something? Are you sure it wasn't Woz?
Spolsky: No, Steve Jobs and Captain Crunch were hanging out.
Pryor: Okay, maybe it was both of them.
Atwood: Steve took all the credit? <more laughter> No I think that's a great question and my answer would be: The more your code will be facing the world, the more you need to know about that stuff. If your code is going to be running on an intranet in a really protected environment, there's really a limit to how much you need to know about that in my opinion, because you're code is just never going to be...
Spolsky: I'll second that, and I'll add the thing I just said before which is that, if you're not interested in that stuff <laughs> you're probably not a good programmer! You should find it interesting! Even if you don't need it.
Atwood: Well where do you draw the line though? Because you could say, you should be interested in everything.
Spolsky: Yup, well no there's just a class of things that to be a good programmer, you just have to have the kind of mind that is interested in that class of things. Like certain types and certain ways in which problems are solved that are kind of ingenious, that has to appeal to you. Ingenious solving of problems.
Atwood: Well that's true, problem solving is something programmers should be interested in, right?
Atwood: And if they're not then that's definitely a warning sign.
Spolsky: And some people aren't and you know, people are interested in very very different things, like some people are interested in what other human beings are thinking and how to convince them to do what you want <laughs> or understand what they want or, I mean some people are interested in this and some people are interested in modern art! But what programmers are interested in is elegant ways to solve interesting problems.
Atwood: Yeah that's a great way to characterise it, it's very broad but definitely true in my experience. So where would you draw the line? Okay so I would draw the line where, if you're spending all your time on the black hat...what's your ultimate goal with this knowledge? Maybe that's where you have to draw the line. If you're actually going out and attacking sites for essentially, monetary gain, and not just, you know because on StackOverflow....
Pryor: Do you think you went through a stage of that in your life though? I mean I know I did when I was at college...
Atwood: Oh totally.
Pryor: ...it was sort of, I had, I realised I had this power to do something that other people couldn't do and you have to sort of come to grips with it, and use that correctly. I mean I went down to the payphone, I had a little portable mac and you could download all the boxes in one application, I went down to the payphone in my dorm, and I played the tone... because I wanted to see if this worked!
Atwood: Yeah sure, if you want to see if it works that's fine!
Pryor: I wanted to see if it works and I ran the tone, and I dialed my parents, and my mom is at the phone, and I was like, that's ridiculous!
Atwood: Hi mom! Bye! <laughs>
Spolsky: Yeah! You know what I …
Pryor: I was actually scared! You know I was frightened at the time.
Spolsky: You know what I did?
Atwood: What year was this? <laughs>
Atwood: Ok. Wow, it still worked then?
Spolsky: You know what I did in the early 80s? People used to have rotary dial phones, there was like a circular wheel on the front of the phone.
Spolsky: I know most of our listeners will not believe this! And it had 10 little holes on the front of it, you would insert your finger into one of these holes and make a circle until your finger stopped on this little J shaped piece of metal and let go, and it would go click click click click for the number of times based on the number that you dialled, and in order to prevent people for using these phones for outbound calls which costs money, and if you wanted to set up a phone that was only good for inbound calls, there were locks that you put on the 1, you could dial a lot of 1's but that didn't get you anywhere <laughs>. There were no numbers you could dial, just al to of 1's. Except the Israeli emergency number which was kind of handy. So people put these little locks on the 's so that you could only dial up to a 1. And I realised that quickly pressing the Hang-up button .. or what's it called?
Atwood: Yeah I don't know what it's called but I know what you mean..
Spolsky: And if you click it, that did the same thing as dialling a 1 and if you click it twice rapidly that was dialling a 2, and I realised yo could actually dial crazy long distance phone numbers on a phone that was locked!
Atwood: <laughing> You hacked the analogue phone system!
Spolsky: Yeah! And it worked, I was calling all kinds of international..
<lots of laughing>
Spolsky: And at one point I was in the army and there was one of these phones and I phones my parents in England and had a nice long conversation with them and this poor other kids <laughing> who was actually English got in trouble. <laughs> When the army battalion got there about their phone bill! And they were like “who just called England?” You know in those days a 20 minute call to England, was like $8 or $80 or some large amount!
Pryor: I think it would be difficult for parents like, especially if you didn't know enough?
Atwood: You didn't fess up?
Spolsky: But he didn't really get in trouble because they couldn't prove it was him.
Pryor: If you had a kid that was, you know, learning about computers and knew more than you did and you saw them doing things like that you'd be very frightened, you'd be like oh my god!
Atwood: Let me use a concrete example. So we do actually have--even now--security holes on SO, increasing small ones, narrow ones.
Atwood: But some of them can be serious because if you can get a script to execute on our site, there's almost no limit to what you can do? Scarily enough, I mean we, there's a few protections that some browsers afford...
Spolsky: Wait when you say scripts executing on our site?
Atwood: In other words if you can get our site to serve up that you provided.
Spolsky: That you wrote. You can do a lot of nasty stuff.
Atwood: Yes but, you can essentially do that by embedding a script tag in some format..
Spolsky: Of course.
Atwood: which retrieves...
Spolsky: In some place that we display like...
Atwood: Any place that we display information that you have put into the system, it's a classic sanitisation issue but it's hard to get it right everywhere.
Atwood: So if you found that out, what would you do? This is what really determines if you;re white hat or black hat, or grey hat, like if you hate us you would exploit it. And you would quietly tell a bunch of people that hated StackOverflow about this exploit.
Atwood: And encourage them to attack us right and, do things that are...
Spolsky: You must come down to StackOverflow and demand a bribe!
Atwood: But then if you're a grey hat, you'll be like, aaah I don't care, you know like, I did this, it was fun, but I'm not going to do anything with it – I'm not going to publicise it, I just don't care – but I'm not going to tell the owners either. Because I don't care enough for them to fix it.
Pryor: Yeah, or maybe you would even set it up so that it raised your rep or something. Something for yourself.
Atwood: Yeah, or you might do something that's innocuous or just clever, so that we might eventually see it and say “Oh that guy is clever”, and then the white hat would be, to basically tell us about it and explain not only how you did it – but how to fix it, and that sort of thing. Now, we've been very fortunate on StackOverflow in that almost all the security stuff we've seen, people have been nice enough to tell us before doing anything bad with it or, anything at all.
Spolsky: Sure I mean that's what you'd expect, most people are good.
Atwood: Yeah! They kinda like your site and, you know I think it's, is it wrong to do black hat stuff? No, I think it's awesome to know as much black hat stuff as possible, but it's a question really of what you're doing with that information that sets the tone for whether it's good or bad or whatever.
Spolsky: Guns don't kill people, people kill … guns.
Atwood: People kill guns.
Pryor: Bullets kill people.
<all 3 laugh>
Atwood: Bullets kill people. <laughs> So just get rid of the ammo! And then we solve all these problems! That's a good question.
Spolsky: All right, let me ask one. Did we ever do the...this is a random one, developer salaries. This is an old one. This is somebody claiming that there has been a large drop in the number of people graduating with computer science degrees, this is question 400865, and now this drop in people graduating in computer science degrees should lead to developer salaries rising. This person claims.
Atwood: But it's not an immediate effect! It's not like, tomorrow we're going to have this big increase in salaries <laughs>.
Spolsky: Yeah! I don't think it matters that much, the number of people entering the workforce is like 2% or 3%, of the people in the workforce so even if there were no computer science majors next year there will only be a 3% decrease.
Atwood: How many need a computer science degree though? There's tons of people coming into the workforce ..
Spolsky: Without the degree?
Atwood: Yeah! They could be really good programmers too so this is already a fraction of the overall number.
Spolsky: Right. I don't think it's that significant, I haven't noticed any serious change. I mean there's a little bit of a recession going on that is causing nominal starting salaries to be a little bit lower because you can get away with it.
Pryor: But this is indefinitely for many other industries, like lawyers salaries are way down.
Spolsky: Yeah but they'd gone way up I mean, the bubble is over in lawyer salaries, and a little bi in programmer salaries, I mean there was a point where people graduating from Stanford and going to work at google that could honestly expect to be earning 100k in the first year and that was ridiculous, that was just the bubble, and that has ended now. But that's just, we're talking about 1% here, but, I used to get in trouble all the time during the .com crash there were all these nice, competent people who didn't have jobs? But they weren't the super programmers, the super programmers found some place that needed them even in the worst of the crash I think – and then every time I said that people were like <silly voice>”Waaaah you don't think I'm a super programmer? It's hard out there to find jobs!”.
Spolsky: And it kind of depends because look, if you want to stay in Green Bay, Wisconsin, then there may not be a job in Green Bay, Wisconsin, but...
Pryor: I don't think people objected to you want to hire really good programmers, I think they objected to some of the methods that we used to screen, initially screen people.
Spolsky: That's a different subject, the truth is there really were a lot of programmers who lost their jobs in the first .com crash, and there are a lot of programmers losing their jobs now, in this current recession, and some of them are great programmers, some of them are terrible programmers, most of them are – we've said this a few times – the companies that are laying off 3% of their staff? Like Microsoft did about a month ago, are probably using this as an excuse to get rid of people who they didn't feel like getting rid of in some other way.
Pryor: No! But you see today Microsoft's earnings are way down!
Spolsky: Yeah yeah! But did they lay people off? This might be an excuse for Microsoft to shut down something stupid...
Spolsky: like live.com <laughs>
Atwood: AWW that's cold, COLD!
Spolsky: Sorry! <laughing> It's not stupid they really should do....No it is stupid for Microsoft! You know Internet Explorer! Close it down! Microsoft has only spent money on IE and gained nothing.
Atwood: Well that's the thing if you're a company as large as Microsoft, surely you can get rid of something.
Spolsky: Let's talk about how large Microsoft is! Oh my god did we talk about this last week or is this going to be boring?
Atwood: I don't know, I think so!
Spolsky: The campus is gigantic! The building numbers go up to like 130 or something and then there's some buildings with letters. And that's just in Redmond, there's also an Issaquah campus, there's a Silicon Valley campus, and there's a whole bunch of smaller places around the world, in Redmond there are several exits from the highway which are all in the domain of <laughs> the Microsoft campus, it is so easy to get lost, it is so gigantic, the buildings are huge, in the middle of the campus they built like a little mall with like a, bunch of restaurant and every possible type of cell phone from all four carriers....
Pryor: I bet you still couldn't buy an iPhone!
Spolsky: <laughs> Yeah you couldn't buy an iPhone!
Spolsky: Yeah! They built like a little village contemporary mini-mall just so that Microsoft employees could get their hair cut or whatever, or go to a restaurant.
Pryor: I was right next to the visitor centre that we went to see, that was interesting because today we went to the computer history museum and Joel mentioned that the Microsoft visitor centre was really..
Spolsky: They had an X-Box exhibit? They had a couple of surface tables, they had some Zunes, and they had some laptops you could play with.
Pryor: There's no history! I mean if you went to Microsoft...
Spolsky: There was an Altair but that's it.
Pryor: There might be some other place where this is all located but at the Microsoft visitors centre...
Spolsky: No there isn't, this is all Microsoft has.
Atwood: Was this the one at Redmond?
Atwood: Yeah that's the main one.
Pryor: There was a Microsoft store which sold everything you could possibly think of, branded with the slogan “I'm a PC”.
Spolsky: Yeah <laughs> you can get an I'm a PC hat, an I'm a PC pen...
Pryor: I'm a PC shirt, I'm a PC shoes....
Atwood: I'm a PC PC?
<all 3 laugh>
Spolsky: You could put it on a Mac that would be weird, buy the sticker and put it on a Mac.
Pryor: But they didn't have...I'd've expected...They could have had everything the Computer history museum had.
Atwood: Yes they should, I agree totally. I've been to that visitors centre and there is a small area where they have memorabilia of really early early releases of windows, going back to DOS days, it's tiny though, it's easy to miss...
Spolsky: It's gone.
Atwood: Really? You blink and you miss it..
Spolsky: Looks like it.
Atwood: They have the guitar from the Windows XP launch, there was a special guitar made signed by everybody,.
Pryor: Yeah we didn't see that.
Atwood: It's the tiniest space you have to know where it is.
Spolsky: Must be like, a temporary exhibit or something?
Atwood: I don't know if it was temporary, but it's easy to miss. They don't focus on it so you're right Michael, they should, because there's a huge part of history that Microsoft is a key in.
Pryor: But they must .. Joel was saying that they must have made this visitor centre because people will go to Microsoft and there's nothing to see right? It's all a bunch of buildings, so they had to make it so people had a reason and something to do, a destination to go to, and they did – but It was kind of a let-down. Especially after going to the computer history museum today. That was great. Oh and our company, Fog Creek software was in the museum, which was kind of cool. Well, not really. It was in a book. In the store.
<Lots of laughter>
Atwood: Prominently featured! Centre piece of the display!
Spolsky: No it actually was they had “Founders at work” prominently featured in the Computer history museum store, with Fog Creek software on the cover.
Atwood: Did you nudge anybody and say “Hey I'm in that book”?
Spolsky: There was nobody there.
Atwood: Well did you find somebody and bring them over? <laughing> Hey! You! C'mere! I'm in this book! Look! Me!
Atwood: Well I'm glad you enjoyed it, I looove the computer history museum. I almost demand that anybody geeky who comes to visit me, they have to go, like I almost march them there. I feel so strongly about it.
Spolsky: Do they still have that store called “Weird stuff”?
Atwood: I think they do! But that's like, is that in the bay area or Is that?
Spolsky: Yeah it's in Silicon Valley, somewhere.
Spolsky: It was a store I don't know, I haven't been there in about 10 years but they sued to just buy the technical crap of every single start-up that had failed and was going out of business, like, liquidators, so what they had there was often a lot of pretty new gear but it was always weird like they would have, a whole wall of oscilloscopes, because Silicon Valley wasn't only about software and stuff like that, they also had electrical engineering, and it's kind of fun because you're usually ploughing through the detritus of some poor company that had fallen apart and closed down, the dashed dreams.
Spolsky: Of a million start-ups. Did we answer the question? Are we even talking about this question?
Atwood: What was the question? <laughs> It was about salaries and...
Spolsky: You know what the number of computer science majors has not actually gone down, there was a bubble in the number of computer science majors during the first .com boom, and the bubble ended, and we're back to the same number of computer science majors we had before the .com boom, so there's just a natural number of people who wants to study computer science, which in the united states is sadly very very low.
Atwood: Wait wait you say that, then at the same time, I guess you're thinking of PHD level?
Spolsky: No, no no! Undergrad.
Atwood: Oh undergrad.
Spolsky: I mean there's just very few, the year I graduated there were 4 other people graduating with me.
Pryor: Yeah they said that 9,000,000 number? Professional programmers? Doesn't that seem ridiculously small to you?
Atwood: In the whole world? In the world yeah, it's a little small.
Atwood: Because the population of California is like...what? What's the population of the US?
Spolsky: 300 Million in the US.
Pryor: 7 million in New York.
Spolsky: 8 Million in NYC, 20 million in New York.
Atwood: So it'd be like if everybody in New York was a programmer?
Pryor: Yes. Let me ask that question on StackOverflow.
<lots of laughter>
Atwood: That is a nightmare scenario, by the way.
Spolsky: That question is already on StackOverflow
Pryor: Can...we....create....a.... <miming his typing>
Atwood: Don't worry, I'll go in and delete that question later on.