To me, this sounds a lot like cognitive dissonance. It seems like Joel believes that his blog has primarily served to further Fog Creek and that he does his blog because he enjoys it. To cope with these two competing ideas, he comes up with a new one: he likes his blog because it promotes Fog Creek. Therefore, when the blog quits being beneficial to Fog Creek, he quits.
I actually find this sad. In hindsight, some of Joel's writings might not seem terribly radical, mostly because they've since been accepted. But most of them were when he actually wrote them. I think we underestimate the effect that Joel has had on the software industry.
My take-away was slightly different: I think his goals changed along the way. These days, he is very interested in furthering Fog Creek. But when I started reading his essays (before even "Joel on Software", when he was using Dave Winer's software to manage his blog), I think his goals were less clear and a lot of the essays were just ideas that he needed to express.
As with many similar bloggers, I think his early years were incredible, but the good ideas gradually became scarcer -- compare his last five years of essays with his first five years. I suspect that is because in his early years he could mine the many ideas he'd had prior to blogging. So it may not be a bad thing from him to let that lie fallow for a while. He can always un-retire if so moved.
Anyhow, I wish him well and hope he finds joy in following his current interests.
What are Joel's contributions, exactly? Write your own programming language and quit your job if you don't have an Aeron chair in your own private office?
I've always enjoyed Joel's blog and will miss it. I wonder if the almost-immediate criticisms in places like Hacker News of anything he writes is an unstated part of the reason he's giving up blogging.
Over the years most of what Joel has written has been for me both illuminating and entertaining. I've only disagreed vehemently a couple of times with what he has written (his stance on exceptions, for example).
Back in 2001 I even used his Joel test to dramatically improve the quality of the software produced by the team I was leading at the time.
Maybe I am a minority, but if I discover some new online company, I tend to check their blog to have a glimpse about what they are up to, how their thinking is in general and whether they are able to communicate in an intelligent way. To sum it up, I personally like it when company has a blog, and I see it as a bonus, even if they speak only about themselves.
I do the same. In fact, if it's a relatively young company and I find that the blog hasn't been updated for months, that will influence my decision on buying their product. The blog may or may not be a good indicator of company health, I don't know. But in the absence of other indicators, I'll treat it as a heartbeat signal.
I can confidently say reading company blogs has never factored into my purchase decisions. I also don't read company press releases, most company copy, or most ads before buying. What's the point? That time is better spent reading independent reviews and such.
This seems like an oddly backwards take on blogging. Do you really start a blog solely because it's a way to flog your product? Not because you have things you want to say and want an outlet to say them in, without having to deal with finding a way to get yourself published in a magazine column or something?
I can imagine sometimes it's a good choice even if your only goal is the sales/exposure your blog brings in, but it's not surprising to me that it would often not be. There are plenty of people, including quite a few smart ones, blogging just to say things that they want to say, so it's a fairly competitive market in which you're up against people who are willing to work for free, because they categorize blogging as something other than work.
In fact, I'm not sure I really believe even he originally did it for that reason. Did he really start Joel on Software solely to be "a blog that actually generates leads, sales, and business success"? Or was it because in 2000 blogging was still relatively new, it seemed interesting, and he had a lot of things to say about software? I can't believe it wasn't at least a mixture of those.
I have to believe that what we're seeing is fallout from his posting on Inc. where he (publicly) lamented about how distressed he was by Atlassian's performance, and started demonstrating doubt that his company would survive. He compared Atlassian to Oracle, and Fog Creek to Informix. Even if he had those thoughts privately, to so openly admit to the competition, (and his customers) that he believed that Atlassian's approach might be more effective, and they would end up being the winning team, was unwise to say the least. (And violated several dictums of Art Of War)
I've used both Jira and FogBugz, so I take a little bit of exception at his statement "we have the undisputed No. 1 product among the 5 percent to 10 percent of programmers who regularly read blogs about programming."
I'd like to see the the numbers behind that claim. Perhaps they No. 1 product among people who read joelonsoftware - but "blogs about programming" in general? I'm willing to bet money that Jira has far surpassed FogBugz in that space, if only because every single company I know ends up using Jira for so much more than bug tracking.
You can be certain Atlassian is aware of every step Joel takes:
He's right on just about every account. For most web-based businesses, there are better ways to get users than blogging. (I say this as someone who built a pretty solid one off of a blog in the past as well.)
There's too much competition. It takes too much effort to write high-quality content frequently enough to attract lots of attention. The conversion rate is low, and the price is high.
Sure it works for some people. It worked for Joel and 37 Signals. It worked for me once upon a time. All of us started before the word "blog" had entered the popular lexicon, but it could probably work now with the right niche and enough effort.
But if you're a small startup, you'd be better off programming some viral hooks, maybe doing some basic SEO, testing out Ad Words, a/b testing landing pages and every other point in your sales funnel, etc. Blogs, Facebook Pages, and Twitter just won't give you the ROI those will when developer hours are a big factor.
For me there's a big area around corporate blogs vs personal blogs. With our startup we've always erred on the side of corporate, keep it very much about releases and other important pieces of information we feel our users need to know about our product.
Recently we've wondered whether our next startup should have a much more personal feel - much the same way that say Peldi over at Balsamiq has developed his product and blog (http://www.balsamiq.com/blog/) and in a similar way to what Joel has as well. Peldi in particular has been completely transparent with his numbers and you do feel a genuine personal connection with him when you read his blog - I think this has helped him to connect with his audience (as with Joel) and you can relate to them both much more in a personal capacity.
For me I think it's important to have a blog (even just a coporate blog) - but I do wonder whether you can gain more traction by adding a personal touch to it.
I think it's important to remember that Joel did it the other way around-- the blog preceded the business.
Joel built up an audience, and simultaneously built a product that would appeal to this audience. But even that makes it sound too calculated, as Fog Creek's first product wasn't really aimed at Joel's readership at all.
Joel wrote about something he was passionate about, and wrote well. He also made a product that resonated with decent chunk of his readership.
Here's a good test: would people who have no interest whatsoever in your product be interested in your blog?
If so, it's a personal blog. Cool. (Just remember that your readers are not necessarily your customers, and vice versa.)
If not, it's a marketing tool. Which may also be cool, if that's what you're after.
Is relying on blogging really the reason Fog Creek hasn't crossed over or is it because they write tools that only 5-10% of the programmer audience are interested in?
He is saying that blogging has had a significant opportunity cost and that his blogging time could have been spent on something else that would have led to Fog Creek crossing over.
CityDesk, which was originally supposed to be "the" Fog Creek product, but had the misfortune of being most applicable to an audience Joel couldn't reach.
Copilot (née Aardvark) is doing just fine. Indie shops love it for taking a look at problems on their customer's computers. Not really the market we thought we'd nail when we first launched it, but, again, it ends up reinforcing Joel's statement.
Isn't it a little strange that he announces his "blogging retirement" on a magazine's website inside an article hidden behind an interstitial ad? Shouldn't this have gone front and center on Joel on Software?
I've read JOS since...well, I think the beginning, or dang close. Joel's articles are some of my most-forwarded. The earliest were phenomenal.
Now, though...most of the time I figure he's going to try sell me something. Each post is about a new feature. The Inc articles give me the idea that coders just aren't where he's at anymore. Bigger fish to fry.
"What's more, I have trouble pointing to other successful entrepreneurs who have used the same formula and reaped the same dividends I have."
The first bit of advice given by anyone is to start building content, so that SEO can help your webapp be discovered. While this is good advice, in a way Joel is contradicting this. He is effectively saying that SEO will only take you so far. This has further impact when it is coming from probably one of the most linked to blogs.
The question to ask here is whether this building content stuff so very important? I agree with the thought about building an audience, which is priceless, but what about plain old content found through google?
"Building content so that SEO can help your webapp be discovered" is a great first step. But it will indeed only take you so far.
Joel's interested in selling his products to people who have no interest in reading JOS. And, starting another blog with alternate content is not the best way to reach these people, either.
On the StackOverflow podcast, I believe he mentioned something about not wanting to dominate the blogging world anymore. Someone says "Joel" and everyone else thinks "Spolsky." Yet, he doesn't mention that at all here. I wonder what's up with that?
Also, he says he's going to quit "for the most part" podcasting and public speaking. Is this also the end of the StackOverflow podcast? Seems unlikely that he'd force that onto Jeff, who seems to like it a lot.
I figure the podcast can stay alive at least short term. Riffing off of Jeff's podcast agenda by phone for two hours a month isn't nearly as much work as blogging a full two out of every five work days.
FogBugz has competitors, I think especially Jira, but it's hard to dominate the market if you only target 5-10%.
Isn't pg an entrepreneur who has reaped similar dividends from a similar formula of "blogging"? i.e. a book/essays addressing bigger issues, has indirectly promoted YC.
pg's "job" is to reach smart geeks living anywhere, convince them he knows what he's talking about and then fund the best among them.
Blogging = high return on effort. He's reaching many geeks and selling them on his qualifications all at once.
Joel's "day job" is managing and encouraging the growth of his (NYC-only) company. Blogging has a lower return on effort because only a small percent would be able/interested in moving to NYC to work for him. Local advertising in NYC would probably provide a better return on his time/money.
1) I bet he could sell the same audience on new products as many times as he could make new useful products. He could scale his company horizontally instead of vertically.
2) Something is off if you have the attention of the hardcore programmers in the world, and can't make your programming tool spread. Why aren't these programmers falling over themselves to use and recommend FogBugz?
Just a guess about #2: the programmers might be recommending FogBugz to their bosses with the pursestrings, but the bosses may not like it enough. FogBugz's approach to project planning might be "the only approach that actually works well" (which I believe) but it's definitely non-traditional and takes control of the schedule away from the people who usually like to have that control. So I think there's typically a lot of resistance from outside the development group to adopting FogBugz.
[+] [-] j_baker|16 years ago|reply
I actually find this sad. In hindsight, some of Joel's writings might not seem terribly radical, mostly because they've since been accepted. But most of them were when he actually wrote them. I think we underestimate the effect that Joel has had on the software industry.
[+] [-] adamc|16 years ago|reply
As with many similar bloggers, I think his early years were incredible, but the good ideas gradually became scarcer -- compare his last five years of essays with his first five years. I suspect that is because in his early years he could mine the many ideas he'd had prior to blogging. So it may not be a bad thing from him to let that lie fallow for a while. He can always un-retire if so moved.
Anyhow, I wish him well and hope he finds joy in following his current interests.
[+] [-] jrockway|16 years ago|reply
[+] [-] stevoski|16 years ago|reply
Over the years most of what Joel has written has been for me both illuminating and entertaining. I've only disagreed vehemently a couple of times with what he has written (his stance on exceptions, for example).
Back in 2001 I even used his Joel test to dramatically improve the quality of the software produced by the team I was leading at the time.
[+] [-] lucifer|16 years ago|reply
"Better to keep your mouth closed and be thought a fool than to open it and remove all doubt".
[+] [-] greyman|16 years ago|reply
[+] [-] yungchin|16 years ago|reply
[+] [-] raganwald|16 years ago|reply
[+] [-] edj|16 years ago|reply
[+] [-] username3|16 years ago|reply
[+] [-] _delirium|16 years ago|reply
I can imagine sometimes it's a good choice even if your only goal is the sales/exposure your blog brings in, but it's not surprising to me that it would often not be. There are plenty of people, including quite a few smart ones, blogging just to say things that they want to say, so it's a fairly competitive market in which you're up against people who are willing to work for free, because they categorize blogging as something other than work.
In fact, I'm not sure I really believe even he originally did it for that reason. Did he really start Joel on Software solely to be "a blog that actually generates leads, sales, and business success"? Or was it because in 2000 blogging was still relatively new, it seemed interesting, and he had a lot of things to say about software? I can't believe it wasn't at least a mixture of those.
[+] [-] ghshephard|16 years ago|reply
I've used both Jira and FogBugz, so I take a little bit of exception at his statement "we have the undisputed No. 1 product among the 5 percent to 10 percent of programmers who regularly read blogs about programming."
I'd like to see the the numbers behind that claim. Perhaps they No. 1 product among people who read joelonsoftware - but "blogs about programming" in general? I'm willing to bet money that Jira has far surpassed FogBugz in that space, if only because every single company I know ends up using Jira for so much more than bug tracking.
You can be certain Atlassian is aware of every step Joel takes:
http://jira.atlassian.com/browse/JRA-19746
[+] [-] shin_lao|16 years ago|reply
[+] [-] Tichy|16 years ago|reply
[+] [-] mattmaroon|16 years ago|reply
There's too much competition. It takes too much effort to write high-quality content frequently enough to attract lots of attention. The conversion rate is low, and the price is high.
Sure it works for some people. It worked for Joel and 37 Signals. It worked for me once upon a time. All of us started before the word "blog" had entered the popular lexicon, but it could probably work now with the right niche and enough effort.
But if you're a small startup, you'd be better off programming some viral hooks, maybe doing some basic SEO, testing out Ad Words, a/b testing landing pages and every other point in your sales funnel, etc. Blogs, Facebook Pages, and Twitter just won't give you the ROI those will when developer hours are a big factor.
[+] [-] patio11|16 years ago|reply
[+] [-] Jim_Neath|16 years ago|reply
[+] [-] abalashov|16 years ago|reply
Not that that's bad. Just saying.
[+] [-] brown9-2|16 years ago|reply
[+] [-] whyleym|16 years ago|reply
Recently we've wondered whether our next startup should have a much more personal feel - much the same way that say Peldi over at Balsamiq has developed his product and blog (http://www.balsamiq.com/blog/) and in a similar way to what Joel has as well. Peldi in particular has been completely transparent with his numbers and you do feel a genuine personal connection with him when you read his blog - I think this has helped him to connect with his audience (as with Joel) and you can relate to them both much more in a personal capacity.
For me I think it's important to have a blog (even just a coporate blog) - but I do wonder whether you can gain more traction by adding a personal touch to it.
Not sure what other YC'ers think ?
[+] [-] michael_dorfman|16 years ago|reply
Joel built up an audience, and simultaneously built a product that would appeal to this audience. But even that makes it sound too calculated, as Fog Creek's first product wasn't really aimed at Joel's readership at all.
Joel wrote about something he was passionate about, and wrote well. He also made a product that resonated with decent chunk of his readership.
Here's a good test: would people who have no interest whatsoever in your product be interested in your blog? If so, it's a personal blog. Cool. (Just remember that your readers are not necessarily your customers, and vice versa.) If not, it's a marketing tool. Which may also be cool, if that's what you're after.
[+] [-] javery|16 years ago|reply
[+] [-] nickelplate|16 years ago|reply
[+] [-] ananthrk|16 years ago|reply
Any idea what products he is referring to? CityDesk or Aadvark (or something else)?
[+] [-] gecko|16 years ago|reply
Copilot (née Aardvark) is doing just fine. Indie shops love it for taking a look at problems on their customer's computers. Not really the market we thought we'd nail when we first launched it, but, again, it ends up reinforcing Joel's statement.
[+] [-] smiler|16 years ago|reply
[+] [-] bjplink|16 years ago|reply
[+] [-] phsr|16 years ago|reply
[+] [-] richardw|16 years ago|reply
Now, though...most of the time I figure he's going to try sell me something. Each post is about a new feature. The Inc articles give me the idea that coders just aren't where he's at anymore. Bigger fish to fry.
[+] [-] akshat|16 years ago|reply
The first bit of advice given by anyone is to start building content, so that SEO can help your webapp be discovered. While this is good advice, in a way Joel is contradicting this. He is effectively saying that SEO will only take you so far. This has further impact when it is coming from probably one of the most linked to blogs.
The question to ask here is whether this building content stuff so very important? I agree with the thought about building an audience, which is priceless, but what about plain old content found through google?
[+] [-] yungchin|16 years ago|reply
[+] [-] michael_dorfman|16 years ago|reply
"Building content so that SEO can help your webapp be discovered" is a great first step. But it will indeed only take you so far.
Joel's interested in selling his products to people who have no interest in reading JOS. And, starting another blog with alternate content is not the best way to reach these people, either.
[+] [-] thunk|16 years ago|reply
[+] [-] rudd|16 years ago|reply
Also, he says he's going to quit "for the most part" podcasting and public speaking. Is this also the end of the StackOverflow podcast? Seems unlikely that he'd force that onto Jeff, who seems to like it a lot.
[+] [-] rayvega|16 years ago|reply
http://meta.stackoverflow.com/questions/40602/joel-to-quit-t...
He might have meant no longer doing other people's podcasts.
[+] [-] dpritchett|16 years ago|reply
[+] [-] zavulon|16 years ago|reply
[+] [-] noarchy|16 years ago|reply
[+] [-] 10ren|16 years ago|reply
Isn't pg an entrepreneur who has reaped similar dividends from a similar formula of "blogging"? i.e. a book/essays addressing bigger issues, has indirectly promoted YC.
[+] [-] ekanes|16 years ago|reply
pg's "job" is to reach smart geeks living anywhere, convince them he knows what he's talking about and then fund the best among them.
Blogging = high return on effort. He's reaching many geeks and selling them on his qualifications all at once.
Joel's "day job" is managing and encouraging the growth of his (NYC-only) company. Blogging has a lower return on effort because only a small percent would be able/interested in moving to NYC to work for him. Local advertising in NYC would probably provide a better return on his time/money.
[+] [-] gruseom|16 years ago|reply
[+] [-] staunch|16 years ago|reply
2) Something is off if you have the attention of the hardcore programmers in the world, and can't make your programming tool spread. Why aren't these programmers falling over themselves to use and recommend FogBugz?
[+] [-] DougWebb|16 years ago|reply