5 ms·

I am fascinated with the trajectory of Joel and his company. Maybe because I followed it so closely from Joel's earliest essays (which are classics and have bec

I am fascinated with the trajectory of Joel and his company. Maybe because I followed it so closely from Joel's earliest essays (which are classics and have become commonly accepted wisdom, the source mostly forgotten) and the Fog Creek forums (the Business of software) which had a great cast of characters and drama (tail of the g!) for a couple of years. One of the interesting things is it seems to me like Joel, for all his wisdom, always made weird (I would say wrong but the man is a legend after all) decisions. Like his whole premise was to hire great programmers, give them an office with a door that closes, and watch great software appear. Which would take at least ten years. It was clear that this software was originally CityDesk (a content management system, say no more), then FogBugz (the issue tracker doomed from the start by its name alone), and even Copilot, to circle back to the documentary. The magic never happened, but lo and behold, out of nowhere comes Stack Overflow and makes all of those other ventures an only slightly relevant footnote. And not from any clever Strategy, but completely serendipitously! Turns out Joel was more right than even he probably thought - just press on being great, something great will come out of it. Though not necessarily what you had planned. There is something beautiful about this, to me. I am sure people will tell me how ridiculously mistaken and misunderstood my take is, but it's how I read the story.

JoyfulTurkey 3y ago
The office with a door thing was how Microsoft was back in the day, if I remember correct. From the stories I heard, some of the offices were ridiculously small/narrow but they still had a door. Joel was a Program Manager for Excel at MS so my guess is he brought that mindset over. Really, I wish more places would go back to that. Being able to work with minimal interruptions is great.
- Calamitous 3y ago
  Ironically, after working in cubicles and hurk open floor plans for most of my career, now I have an office with a door. Ten feet from my bedroom. ;)
- borski 3y ago
  Also how Apple was.
thefourthchime 3y ago
There was a podcast with Joel and Jeff as they talked through the development of Stackoverflow. It seemed that Joel had hired Jeff to code it, and then discussed what it should be like, and then as Jeff had something to show they talked about where it would go. Meanwhile, there is plenty of software physiology sprinkled all throughout. It was a fascinating podcast. At least that's what I remember from it, I haven't listened to it since it came out, what... 15 years ago?? https://soundcloud.com/stack-exchange https://soundcloud.com/stack-exchange
- codinghorror 3y ago
  "hired" is a bit of a stretch! Joel had some design input, and the original "let's replace this terrible thing with something better" idea for sure, and Expert-sex-change was the mimeograph, but it was me, Jarrod Dixon and Geoff Dalgas and then Kevin Dente in the earliest days.
  - thefourthchime 3y ago
    Holy shit! Thanks for the correction Jeff!
hermitcrab 3y ago
>(tail of the g!) I was very active on the BOS forums. But that isn't jogging any memories. Perhaps it was before my time?
- sfjailbird 3y ago
  A forum member started a topic to point out an issue with the Fog Creek website: On the top-left Fog Creek logo (which could be clicked to go to the main page), the tail of the letter 'g' was non-clickable. The poster pointed out that this should be fixed. Joel himself (who rarely participated on the forums) commented along the lines that the guy should immediately check himself into a mental hospital. It was pretty hilarious, a lot of people took issue with Joel's comment and it was the drama for a few days. Standard web forum stuff :) Hmm, now I wonder if that kind of comment would be judged harder in today's climate!
raincole 3y ago
You said it like CityDesk and FogBuz were some miserable failures. Were they? Or they were actually profitable for years? To be honest the whole comment reads very SV vibe. Go big (as big as StackOverflow) or go home.
- stevoski 3y ago
  CityDesk was a miserable failure. FogBugz earned some good money. I suspect it bankrolled them so they could make Trello (which had very little revenue).
- pketh 3y ago
  It depends on how you define failure. Fogbugz didn't dominate the market like Jira did, but it had a dedicated customer base. It's revenue was used to fund the development of Trello in the early days, and to fund the first 2~ years of development on Glitch. When FogBugz was sold off, it was still a modestly profitable business that a better acquirer could have stewarded into a better performer but the bug/issue tracking market is/was really tough. Source: I used to work at Fogcreek, did some light dev/design work for FogBugz, and was the co-creator of Glitch.
ozim 3y ago
Is it really that fascinating? I also followed Joel's essays and was more into it back then. But as I look at it now it is generic startup setup. They pivoted enough times to land gold mine. With getting developers do what they like they were able to pivot quite easily and they were onto hiring smart people. In the end that is what sane business person should do, pivot, pivot, pivot until you land great idea. Only thing you cannot pivot are great people because there is limited amount of those and there is unlimited amount of ideas. What bad companies do is sticking to some bad idea that maybe earns money but then they burn through legions of junior devs to keep train going.
tom_ 3y ago
Everybody I know that used FogBugz seemed to really like it! They also did Trello.
- kilroy123 3y ago
  I definitely liked it a lot more than Jira back in the day.
- nicolas_t 3y ago
  I actually liked Fogbugz a hell of a lot more than Jira... Sadly it's been sold and is barely a shell of what it was but it had a lot of a premise and was a lot more developer friendly.
ekanes 3y ago
I think that's the right take. If you're are legitimately good, and you press on long enough, (and admittedly you have some luck), you'll often win eventually.
- RetroTechie 3y ago
  Persistence. Sometimes being good isn't even required, as long as you keep at it.
chubot 3y ago
I see where you’re coming from, but fog creek was private the whole time I bet they made a killing from both citydesk and fogbugz There’s no way to know, but if they had nice offices in Manhattan and were self funded in those days, that’s a clue Fogbugz was a solid #2 in the market from what I remember There are lots of people making tons of money in areas like CMS and bug tracking The ones you hear bragging about money are usually trying to attract employees to a marginal business The ones who are making money don’t want to attract competitors by talking too much
wvenable 3y ago
> Fog Creek forums (the Business of software) which had a great cast of characters and drama (tail of the g!) for a couple of years. If you know where to look, some of that cast of characters is still going strong on a forum of a similar design to this day. It's a bit crazy. Stack Overflow came right out of Joel's complaints about Experts Exchange and is probably the best example of being with the right people at the right time with the right idea.
- toyg 3y ago
  > If you know where to look, some of that cast of characters is still going strong on a forum of a similar design to this day. Uh, where...?
  - mdemare 3y ago
    You’re on it now.
    - hermitcrab 3y ago
      I think they are referring to another forum, but I guess they have their reasons for not naming it.
      - hermitcrab 3y ago
        And it doesn't look like forum exists anymore. I guess it also tapped out.
        wvenable 3y ago
        It did have a redirect but at some point the domain name expired. It was kind of a fun domain to own though.
  - marktangotango 3y ago
    I would guess this one, but I don't think any usernames carried over. Except maybe the one I can't recall right now (startup related to streaming video over .gif back in late 90s iirc).
mixmastamyk 3y ago
Home runs are a lot fewer in number than fouls or even base hits. Not entirely surprising.
veidr 3y ago
I forget even who the competitors were, but FAUGUEBOGGEZ (as we called it then) was wildly popular for being less shitty than the #1 at that time. (Maybe it was JIRA, or maybe my memories of having to use JIRA have overwritten every other shitty bug tracking system I hated...?) Anyway, it didn't remain so, but IIIRC FogBugz was on top (in terms of being the consensus-better-than-the-actual-market-leader) for at least a few years. Which, if you didn't take 50 or 100 million in VC money, is probably a massive success for 99.99% of people. But yeah, nothing compared to the most-used website in the history of programming.
- ayewo 3y ago
  IIRC, there were several competitors with no clear leader [1] but Bugzilla was pretty popular. Then Trac [2] showed up with a clean UI which allowed it to gain mindshare among OSS projects. JIRA ultimately became the market leader due to having a product that sucked less than what was out there and being infinitely customizable. One of the things Atlassian did to gain mindshare was to gift large OSS projects, often under the ASF (Apache Software Foundation) banner, free licenses to JIRA and Confluence. FogBugz was popular if you were a regular reader of Joel’s blog or if you had heard of the Business of Software community. Outside of that bubble, it wasn’t that well known. 1: https://en.wikipedia.org/wiki/Comparison_of_issue-tracking_systems https://en.wikipedia.org/wiki/Comparison_of_issue-tracking_s... 2: https://en.wikipedia.org/wiki/Trac https://en.wikipedia.org/wiki/Trac
- eichin 3y ago
  We looked at FogBugz as a replacement for RT, but it had some severely opinionated limitation on release flow/version numbers that meant it didn't fit our existing product releases, let alone our future plans. (I think we were pretty interested in it for the UI and the dynamic velocity modelling? But the presentation/demo pretty much ended with the "so don't do that" discussion :-)
- emmelaich 3y ago
  Definitely JIRA/Atlassian. I remember Joel cursing "those Australians" in one blog post. Not literally cursing but you get the idea.
fleaflicker 3y ago
Don’t forget trello

JoyfulTurkey 3y ago

The office with a door thing was how Microsoft was back in the day, if I remember correct. From the stories I heard, some of the offices were ridiculously small/narrow but they still had a door. Joel was a Program Manager for Excel at MS so my guess is he brought that mindset over. Really, I wish more places would go back to that. Being able to work with minimal interruptions is great.

Calamitous 3y ago
Ironically, after working in cubicles and hurk open floor plans for most of my career, now I have an office with a door. Ten feet from my bedroom. ;)
borski 3y ago
Also how Apple was.

thefourthchime 3y ago

There was a podcast with Joel and Jeff as they talked through the development of Stackoverflow. It seemed that Joel had hired Jeff to code it, and then discussed what it should be like, and then as Jeff had something to show they talked about where it would go. Meanwhile, there is plenty of software physiology sprinkled all throughout. It was a fascinating podcast. At least that's what I remember from it, I haven't listened to it since it came out, what... 15 years ago?? https://soundcloud.com/stack-exchange https://soundcloud.com/stack-exchange

codinghorror 3y ago
"hired" is a bit of a stretch! Joel had some design input, and the original "let's replace this terrible thing with something better" idea for sure, and Expert-sex-change was the mimeograph, but it was me, Jarrod Dixon and Geoff Dalgas and then Kevin Dente in the earliest days.
- thefourthchime 3y ago
  Holy shit! Thanks for the correction Jeff!

hermitcrab 3y ago

>(tail of the g!) I was very active on the BOS forums. But that isn't jogging any memories. Perhaps it was before my time?

sfjailbird 3y ago
A forum member started a topic to point out an issue with the Fog Creek website: On the top-left Fog Creek logo (which could be clicked to go to the main page), the tail of the letter 'g' was non-clickable. The poster pointed out that this should be fixed. Joel himself (who rarely participated on the forums) commented along the lines that the guy should immediately check himself into a mental hospital. It was pretty hilarious, a lot of people took issue with Joel's comment and it was the drama for a few days. Standard web forum stuff :) Hmm, now I wonder if that kind of comment would be judged harder in today's climate!

raincole 3y ago

You said it like CityDesk and FogBuz were some miserable failures. Were they? Or they were actually profitable for years? To be honest the whole comment reads very SV vibe. Go big (as big as StackOverflow) or go home.

stevoski 3y ago
CityDesk was a miserable failure. FogBugz earned some good money. I suspect it bankrolled them so they could make Trello (which had very little revenue).
pketh 3y ago
It depends on how you define failure. Fogbugz didn't dominate the market like Jira did, but it had a dedicated customer base. It's revenue was used to fund the development of Trello in the early days, and to fund the first 2~ years of development on Glitch. When FogBugz was sold off, it was still a modestly profitable business that a better acquirer could have stewarded into a better performer but the bug/issue tracking market is/was really tough. Source: I used to work at Fogcreek, did some light dev/design work for FogBugz, and was the co-creator of Glitch.

ozim 3y ago

Is it really that fascinating? I also followed Joel's essays and was more into it back then. But as I look at it now it is generic startup setup. They pivoted enough times to land gold mine. With getting developers do what they like they were able to pivot quite easily and they were onto hiring smart people. In the end that is what sane business person should do, pivot, pivot, pivot until you land great idea. Only thing you cannot pivot are great people because there is limited amount of those and there is unlimited amount of ideas. What bad companies do is sticking to some bad idea that maybe earns money but then they burn through legions of junior devs to keep train going.

tom_ 3y ago

Everybody I know that used FogBugz seemed to really like it! They also did Trello.

kilroy123 3y ago
I definitely liked it a lot more than Jira back in the day.
nicolas_t 3y ago
I actually liked Fogbugz a hell of a lot more than Jira... Sadly it's been sold and is barely a shell of what it was but it had a lot of a premise and was a lot more developer friendly.

ekanes 3y ago

I think that's the right take. If you're are legitimately good, and you press on long enough, (and admittedly you have some luck), you'll often win eventually.

RetroTechie 3y ago
Persistence. Sometimes being good isn't even required, as long as you keep at it.

chubot 3y ago

I see where you’re coming from, but fog creek was private the whole time I bet they made a killing from both citydesk and fogbugz There’s no way to know, but if they had nice offices in Manhattan and were self funded in those days, that’s a clue Fogbugz was a solid #2 in the market from what I remember There are lots of people making tons of money in areas like CMS and bug tracking The ones you hear bragging about money are usually trying to attract employees to a marginal business The ones who are making money don’t want to attract competitors by talking too much

wvenable 3y ago

> Fog Creek forums (the Business of software) which had a great cast of characters and drama (tail of the g!) for a couple of years. If you know where to look, some of that cast of characters is still going strong on a forum of a similar design to this day. It's a bit crazy. Stack Overflow came right out of Joel's complaints about Experts Exchange and is probably the best example of being with the right people at the right time with the right idea.

toyg 3y ago
> If you know where to look, some of that cast of characters is still going strong on a forum of a similar design to this day. Uh, where...?
- mdemare 3y ago
  You’re on it now.
  - hermitcrab 3y ago
    I think they are referring to another forum, but I guess they have their reasons for not naming it.
    - hermitcrab 3y ago
      And it doesn't look like forum exists anymore. I guess it also tapped out.
      - wvenable 3y ago
        It did have a redirect but at some point the domain name expired. It was kind of a fun domain to own though.
- marktangotango 3y ago
  I would guess this one, but I don't think any usernames carried over. Except maybe the one I can't recall right now (startup related to streaming video over .gif back in late 90s iirc).

mixmastamyk 3y ago

Home runs are a lot fewer in number than fouls or even base hits. Not entirely surprising.

veidr 3y ago

I forget even who the competitors were, but FAUGUEBOGGEZ (as we called it then) was wildly popular for being less shitty than the #1 at that time. (Maybe it was JIRA, or maybe my memories of having to use JIRA have overwritten every other shitty bug tracking system I hated...?) Anyway, it didn't remain so, but IIIRC FogBugz was on top (in terms of being the consensus-better-than-the-actual-market-leader) for at least a few years. Which, if you didn't take 50 or 100 million in VC money, is probably a massive success for 99.99% of people. But yeah, nothing compared to the most-used website in the history of programming.

ayewo 3y ago
IIRC, there were several competitors with no clear leader [1] but Bugzilla was pretty popular. Then Trac [2] showed up with a clean UI which allowed it to gain mindshare among OSS projects. JIRA ultimately became the market leader due to having a product that sucked less than what was out there and being infinitely customizable. One of the things Atlassian did to gain mindshare was to gift large OSS projects, often under the ASF (Apache Software Foundation) banner, free licenses to JIRA and Confluence. FogBugz was popular if you were a regular reader of Joel’s blog or if you had heard of the Business of Software community. Outside of that bubble, it wasn’t that well known. 1: https://en.wikipedia.org/wiki/Comparison_of_issue-tracking_systems https://en.wikipedia.org/wiki/Comparison_of_issue-tracking_s... 2: https://en.wikipedia.org/wiki/Trac https://en.wikipedia.org/wiki/Trac
eichin 3y ago
We looked at FogBugz as a replacement for RT, but it had some severely opinionated limitation on release flow/version numbers that meant it didn't fit our existing product releases, let alone our future plans. (I think we were pretty interested in it for the UI and the dynamic velocity modelling? But the presentation/demo pretty much ended with the "so don't do that" discussion :-)
emmelaich 3y ago
Definitely JIRA/Atlassian. I remember Joel cursing "those Australians" in one blog post. Not literally cursing but you get the idea.

fleaflicker 3y ago

Don’t forget trello