I hope we are too busy finding good things to build, good ways to build them, and good people to build them with (and for!) to sit around gossiping about who's over the hill and who sucks.
The entire spirit of HN is to upvote what you like. We don't need to engage in meta-discussion about which blogs are worth reading because we simply upvote the specific posts that are worth reading. Prolific and consistently good writers simply appear more often.
If you, I or anybody else doesn't like what Joel writes or Jeff writes or Reg writes, the solution is definitely to write a blog post. But not this post. Instead, write the kind of blog post you want people to read. Be a force for positive change.
And if you don't feel like writing, don't get sucked into pointing fingers and sneering at others. Spend your time reading the good stuff.
Once there was a horse tied up on the side of the street. Whenever someone tried to pass, the horse would kick them. Soon a crowd gathered around the horse until a wise man was seen coming close. The people said “This horse will surely kill anyone who tries to pass. What are we going to do?” The wise man looked at the horse, turned and walked down another street.
I beg to differ with you. It's highly articulate criticism coming from a regular JS blog reader and with an incredibly high reputation on StackOverflow (Cletus seems to be the 4th top ranked user.)
Cletus's post raises very valid points and expresses the feelings of betrayal of many people who contributed to StackOverflow many hours of work for free. And he shows several contradictions in what Spolsky preaches and what he practices.
Ironically your comment doesn't add much to the discussion and criticizes Cletus' blog post for being a negative post. Also you seem to confuse the author of the post with the submitter in HN:
"Cletus's post raises very valid points and expresses the feelings of betrayal of many people who contributed to StackOverflow many hours of work for free."
I'm kinda curious about this sentence. Are there people out there who didn't know that Joel runs his website/forums/StackOverflow for Joel's benefit and not necessarily for the benefit of those who contribute?
I read a bunch of Joel's articles in college, then decided that his site wasn't the place for me. I felt that I wasn't really learning anything new from him, and that the forum participants were rehashing really basic stuff that I'd already learned and wanted to move on from. When it came time to get a job (which was right around Joel's big "Great Programmers Work for Fog Creek" push), I kinda laughed and figured that if I were really a great programmer, I'd rather go build search engines or financial models than bugtracking software. So I just quietly applied to financial software startups, stopped reading JoS, and didn't make a big stink about it on the Internet.
I thought about not writing this comment because, as Raganwald notes, I'm being part of the problem and not part of the solution. But I found my life got so much easier when I started worrying about my goals first and other people's second, rather than assuming other people were looking out for me and then feeling betrayed when it turned out they weren't. That doesn't mean being an asshole or never lifting a finger to help someone. It means that when I help someone out on the Internet, it's because I want to, and I don't expect Reputation or Karma or a job offer because of it. They're nice if they happen, but not something to expect.
I don't care who Cletus is, I was responding to what he(?) wrote. If his post was good but he had a low SO reputation, would I discount his opinion? And I have no idea there is a cletus here as well as there. I wasn't paying any attention to who he is for the same reason I was dismayed by how many words he(?) used to discuss who Joel is.
Finally, I never said his post was not truthful. There is dreck in the world. I know that and so do you.
p.s. But yeah, I hesitated before commenting on account of the irony. And then I removed the first paragraph, which originally read:
(but also lists a fourth in the body of the posting)
4. Employee referrals
Obviously, Stack Overflow is based on building a community and now they are trying to leverage that community to generate money from matching companies with developers.
In his post, Joel states that "[t]he average great software developer will apply for, total, maybe, four jobs in their entire career." Paraphrasing the next two paragraphs, he asserts that the truly great (as opposed to average great) developers "...show up on the open job market once...".
Maybe I'm jaded, but I cannot think of a single truly great developer that got hired through the standard recruitment methods. Recruiters and, especially, HR departments filter out all the truly great developers when sieving resumes because the truly great developers tend to be modest and honest, not traits that get them through a buzzword filter. The truly great developers I've worked got in by bypassing the HR system via an inside person.
I would assert that truly great developers pick the companies they wish to work for. As a result, the only way to catch them is to have a great company so that they find you and you still need to get lucky. As Joel writes, "The great software developers, indeed, the best people in every field, are quite simply never on the market."
I'm sorry but this kind of rhetoric ("they're so good you never even see them..") smacks more of "rock star" mythology than realism.
The best programmers, say Ward Cunningham (to name one more name than you or Joel do), are methodology gurus. The best methodologies admit that all programmers make mistakes and are limited. Perhaps we could learn something...
The assertions (granted, unsubstantiated) were that the gurus aren't putting their CV on web sites for HR departments to find. They are targeting companies they want to work for.
Ward Cunningham is a good name. Lets look at some legendary software engineers...
From Ward Cunninghams's linkedin page, you will see he has held two jobs for a long time: Tektronix for 10 years and his own company for 18 years. Tek was his initial hire and he had two short stints at Microsoft and the Eclipse Foundation. He probably applied traditionally to Tek (first job). His titles at Microsoft and Eclipse Foundation make me wonder if he went through the HR department for those jobs. Ditto for AboutUs. Obviously, he did not interview for Cunningham & Cunningham, Inc.
Linus Torvalds worked at Transmeta for 6 years, OSDL for 3 1/2 years, and Linux Foundation for 6 1/2 years. I'm sure he never had a traditional job application/interview encounter.
Wikipedia lists Ken Thompson as working for Bell Labs, Entrisphere, Inc, and Google. I'm willing to bet he didn't have to run the standard Google gauntlet to get hired there.
Wikipedia lists Dennis Ritchie working for Lucent Technologies (currently) and Bell Labs. Not much resume shoveling going on there.
The Wikipedia page on Brian Kernighan is somewhat vague, but lists Bell Labs and the Computer Science Department of Princeton University.
Guido van Rossum went through more companies than the other gurus listed here. Again, I'm sure the Google interviewers were more nervous than he.
Yukihiro Matsumoto has only worked for NaCl according to LinkedIn and Wikipedia.
David Heinemeier Hansson is another example: he founded and built a Danish online gaming news website, which he ran until 2001. I believe he went from there to 37Signals. He did some contract work for them and was hired on the basis of his obvious talent. Again, non-traditional.
Point wasn't that famous programmer don't have an easy getting hired. Their fame, by itself, would allow this whatever else is happening.
My point is that a theory that bases itself of "the qualities of great programmers" when you mean famous "rock star" programmers, is fundamentally flawed. I think the ideas of the great programmers, at least their great ideas, support this.
The point that charging for finding a job is, in some locations illegal, ought to be cause for concern. The point those entities which charge job seekers for exposure have historically been cons should be considered also.
Perhaps there is are very good answers to these concerns but I don't think these question can be dismissed as gossip.
I read the entire thread and didn't see anyone else agreeing with you. So I wanted to go on record and say that I wish I had more points to upvote you with. Excellent thought and request. I will try to publish more and submit the good things I'm reading to HN.
I got this same feeling while reading it, that the author strung together thousands of words to illustrate two strange points: that what Joel and Jeff have done is different from the existing model and thus, might not work and that Joel is past his prime even though he was once really good.
Are those two points really worth thousands of words, whether they are true or not? It doesn't seem like it to me. I wish that at least 500 of the words would have been used to introduce me to something new that the OP thought was better than StackOverflow/Careers.
Of course, the irony is, here I sit writing a comment about an article while producing nothing. I think I'll go write some code.
Stackoverflow Careers deserves serious criticism. I think it was delivered too soon. They should have waited until StackOverflow became more established.
I've used Stackoverlow on occasion but I don't think the point system means much, and shouldn't be used for an associative career site. People that are competent aren't on stackoverflow all day answering questions for people that didn't read the documentation.
I don't think that they're universally too busy to blog, but I do think that the truly great programmers blog very infrequently because they have better things to do. Paul Buchheit's blog shows about one entry per quarter while he was working on Friendfeed, then it went up to about one per week after the FaceBook acquisition. Marc Andreesen has gone dark since starting his venture fund. I know many really awesome programmers at Google who have zero Internet presence, because they're too busy coding. If you want to know what Rob Pike thinks, you read his code, because he doesn't write much otherwise.
I've found that when I'm working on something really cool, I tend to go relatively dark on the Internet too, and there's an inverse correlation between my comments and my productivity. In other words, when I talk I don't know what I'm talking about, and when I know what I'm talking about, I don't talk. ;-) But I cycle through doing/talking phases several times a year, so I'm continually refreshing my experience. Someone who's been doing nothing but blogging for 3-4 years (which seems to be the time it takes to build a major audience) is probably pretty rusty technically, which makes their advice suspect.
I do think that the truly great programmers blog very infrequently because they have better things to do.
This is really presumptuous, and makes a leap from correlative to causal on the back of a couple of blanket assumptions.
The first is that there are better things to do than blog: based on what? If someone feels the itch to write about something they really care about, and doesn't want to bother with more formal methods of publication, I think that blogging is one of the best things they can do with their time. It is not like they are producing nothing; they are producing something that will -- hopefully -- propagate ideas and critical thinking. This sounds like a very good thing. Not all problems that are looking for an answer will have that answer found in a block of code; sometimes the wetware needs to be refactored.
The second is that people who blog more than infrequently are apt to be technically rusty, and should be subject of suspicion. This make me suspicious of those who would assert that. Again, based on what assumption? I agree with the claim that if they blog more they have less time for other things besides blogging; rollover minutes only work on some cellphones. But to claim that because they blog more they know less; this is quite an extraordinary claim. It wants to lump everyone who blogs a lot in with the blowhards. What if someone has a lot to say, and thinks it is important to get those ideas out? It sounds like you would tell that person to shut up and get back to work. I worry about the consequences of this.
We're talking in the context of "truly great programmers". It makes sense that truly great programmers get that way by programming. That doesn't mean blogging is a useless activity, or that bloggers are bad programmers. Actually, I think most bloggers are good programmers, better than the average non-blogging population. But by and large, they aren't great programmers, because the jump from good to great takes that extra bit of concentration and attention that's consumed by blogging.
Think of programming ability as a pyramid. There're a vast amount of programmers that show up at work, put in their 8 hours, and never really think about what they're doing. These people generally don't read or write blogs. Then there're the people that are passionately involved in their craft. They blog. Then there're the people at the very top of the field - the Guy Steeles, Rob Pikes, Jeff Deans, etc. They usually don't blog, because they put all their concentration into their code, and when they take a break, they do things like photography or swing dancing that take them completely away from programming so they can recharge.
The interesting thing about this distribution is that the average blogger is a better programmer than the average non-blogger, but the average great programmer does not blog.
FWIW, I consider myself a good programmer, but not a great programmer. And I've suspected for a couple years that the reason I'm not a great programmer is because I talk too much and code too little. So I dunno if I'd tell the general blogger to shut up and get back to work, but I certainly tell myself to shut up and get back to work, usually with lukewarm results.
I agree that great programmers tend to be focused on programming, to the exclusion of most other productive intellectual activities.
It's ironic - I feel exactly the opposite of the above poster. I just find it too easy to slip back into programming, instead of blogging about my work or spending the time to open-source any of my projects. I wish I was blogging more (or at all, for that matter).
So you definitely wouldn't want to hire someone with say, the highest stackoverflow score (Jon Skeet)? While I definitely agree that a high score isn't a black and white for a "good programmer," a SO score/account is yet another metric that may be available to try to find the perfect fit for your company. I see no problem with a person making more information about themselves available, especially if that information is in the form of technical Q&A.
It's not quite what I said though. The thing with SO careers is the "tied into SO" feature. That is making your SO score a big metric; but isn't a very good metric at all because I am sure there are as many SO programmers with 2K score as good as, say, Jon Skeet just not as active.
Additionally if someone promoted themselves as "I have X score on SO" it would worry me.
(that's also before we discuss whether the ability to answer questions well on SO marks you as a good programmer or not :))
Additionally if someone promoted themselves as "I have X score on SO" it would worry me.
You have stated this twice, but the reason why this should invoke worry seems to have been left as an exercise to the reader. Perhaps there is some sort of underlying assumption you make about someone pimping their SO score that I am not making?
Help me understand why this should worry me. There are plenty of good questions and good discussion that I've seen on SO, and the reputation people have earned has seem more often deserved than not. I don't think a good SO reputation deserving quite the heated bandwagon snarking this article produced.
There are 2 points it would worry me on; firstly SO score is a poor test of how good a programmer you are. I have a reasonable score here: but I'm not really an entrepreneur and compared to some I'm only an adequate hacker :) but I am active and I usually hold an opinion about everything.
Secondly it would set alarm bells: why are they so excited about their SO score? Is it intended to impress me so that I skim over the rest? (it's not the only flag I've noticed in interviews; there are lots of similar ones, you pick them up). [it'd also worry me that they haven't figured out it's not a great metric]
So what I am saying is: having a good SO score is definitely something great to mention to me. But only in passing or I would get suspicious :)
(on the other hand if I were hiring, say, an API community manager then it's an awesome metric! :) so I guess it depends)
This reminds me of a time I interviewed a well-known blogger for a programming position - and his resume was filled with tons of mentions to citations his blog had received - and very little work experience. The interview, as you might expect, was awful.
I believe having a high reputation score on StackOverflow tells at least one thing; it tells that that person is very good at writing and honestly, that's a very important skill for a programmer.
There are two issues here: the business idea (of the jobs site), and Joel's puffery.
I'll make no comment on the latter point, but I have to wonder about the former-- if the hiring process is broken (and I accept the argument that it is), what's a better angle for a start-up than what Jeff and Joel have done?
With respect to programming jobs startups, it seems like there are at least two ways forward. One is to find a new way to match programmers with existing jobs. The other is to find new kinds of jobs entirely.
YCombinator has funded at least one company finding new ways to match programmers with jobs. And I think YCombinator itself is a new kind of job for programmers :-)
I think the main issue here is that Cletus is in Australia. Assuming things haven't changed since January, Australia accounts for only 3% of the traffic to Stack Overflow[1]. It's hard to expect a critical mass of employers less than a month after the employer beta started[2], especially given how few potential candidates are in the area. Jeff has already floated the idea of offering a coupon[3] to interested employers, demonstrating that he's aware of and working to correct the issue.
Besides, it's way too early to say that Careers has "jumped the shark", considering that the public wasn't even aware of its existence two months ago.
Offtopic: I went clicking through your profile then your Careers CV and see that you did the modos rep tracker. I'm addicted to that thing, so thank you very much. It would be interesting if you added # of visits by user ID then told me that I was visiting way too often because I was a top X% user of the tracker :) Great tool, it has a permanent spot on my bookmarks bar. Thanks again.
Thanks! Sorry it's been broken for the past day or two, the profile page had some major markup changes. As soon as I get some time I'll get it working again (hopefully today) and I'll incorporate a little stat that tells you how many times you've hit the tracker in the past day/week/month/lifetime. And, um, if you know anyone that's hiring, feel free to pass on the link to my CV. Thanks. :-)
OK, I think I fixed it - let me know on Meta or via email if there are any more problems. I also added a counter at the bottom so you can see how many times you've checked it today (EST), this week, this month, and all time.
Does anyone else think that while Fogcreek and 37signals have done something cool, a whole generation has tried to imitate them with little success and the constant posting of Joel + Jason does this site little/no good?
Always interesting how blog posts critical of Spolsky always create so much chatter here....or, almost always. If you happen to include Paul Graham as a target in the same blog post as Spolsky (Blogs Are Godless Communist Bullshit), it seems people here aren't as interested in the article for some reason. Funny that.
You seem to be implying that, on news.yc, it's okay to criticize Joel Spolsky, but not Paul Graham. I think it's that you don't have enough controls in your experiment.
I read this longish blog post in its entirety. I don't agree with all of it, but I think the author made some valid points, and that he's making a good-faith effort to be constructive. I also read the longish Giles Bowkett blog post you're talking about. It was very much not constructive, and in fact was all the way across the scale to conspiracy-theory, tinfoil-hat-style nutbar-ism. I'm not surprised it didn't score very well when it was posted here.
Have you noticed how Paul usually reacts to criticism? Say somebody posts a comment here that is vaguely critical of something he wrote. Paul's usual response is along the lines of: "Are there specific things you can point out that I was wrong about?" That doesn't sound to me like a guy who considers himself beyond criticism.
I'm not saying Paul considers himself beyond criticism, I am saying that the community here jumps at the chance to heap criticism on Joel (there are many historical examples of that here)....the only exception I am aware of is the article in which Paul was also mentioned...for that article, hardly a peep.
The second half was a very verbose way to say "it's hard to start a marketplace."
Fair enough. It's hard, but there are success stories all the time.
Karma + pay-to-be-seen + leverage-blog-fame is one set of infinitely many jump-starting gimmicks. Many potential job hunters won't bite, but the subset that remains might be exactly what certain employers want.
I've met Joel in person, and he's bright, but not visionary. Some of his writing his interesting, but I'm generally unimpressed.
Disclaimer: I'm not impartial. I interviewed with Fog Creek in the summer of 2008. I was rejected, which didn't upset or surprise me. However, when I called to ask why, I got a non-answer (HR: "we'll call if we can release this information"). I'd been unemployed for 4 months due to health problems, the economy was falling apart, and they wouldn't even help me out a little bit by telling me why they said "no". Amazing. If he'd had the courage to tell me why I was turned down, my opinion of him would probably be neutral-to-positive.
Almost no company will tell you why you weren't hired. In the US, as soon as a company gives you a reason for why you were not hired, you can sue and argue that the grounds for rejecting you were frivolous. Worse, if the exact phrasing used in any way possibly implies that you might have been rejected for a protected class (even if that's not actually the case in reality) then you can sue for truly ludicrous amounts of money.
Neither of these are hypotheticals; one company I worked for had had a couple of lawsuits due to the former, and last I heard, a friend's company was halfway through a lawsuit due to the latter.
This is another area where you can thank the lawyers for making it a mess for everyone.
Care to elaborate? His guidance seems to be reasonable and consistent - while it should be taken with a grain of salt (some of it feels anecdotal rather than factual), I don't find it unhelpful, as you posit.
Joel isn't outright dishonest, but when he presents his views he usually does it with the intent of improving his business, or rationalization for his business decisions - so there's a strong slant to his talks. You can see this in a lot of places -
Everything from his talk about 'great work environment' - "and that's why every programmer at Fog Creek has two monitors and their own office!" to his suggestions of how to hire the best programmers concluding with "That's why you should use StackOverflow Careers". When he's not outright pitching his company or products, he's rationalizing decisions. Clearly, Ruby on Rails was too immature for FogBugz - that's why his team developed a custom programming language to power BUG TRACKING software.
The entire spirit of HN is to upvote what you like. We don't need to engage in meta-discussion about which blogs are worth reading because we simply upvote the specific posts that are worth reading. Prolific and consistently good writers simply appear more often.
If you, I or anybody else doesn't like what Joel writes or Jeff writes or Reg writes, the solution is definitely to write a blog post. But not this post. Instead, write the kind of blog post you want people to read. Be a force for positive change.
And if you don't feel like writing, don't get sucked into pointing fingers and sneering at others. Spend your time reading the good stuff.
Once there was a horse tied up on the side of the street. Whenever someone tried to pass, the horse would kick them. Soon a crowd gathered around the horse until a wise man was seen coming close. The people said “This horse will surely kill anyone who tries to pass. What are we going to do?” The wise man looked at the horse, turned and walked down another street.