This is the sixteenth episode of the StackOverflow podcast, wherein Joel and Jeff discuss the following:
- Programmer design is scary -- We now, thank goodness, have an actual designer design! Thanks to Jeremy Kratz for assisting us with our design woes. We also appreciate input from Tim Almond, Rob Allen, and Nathan Bowers.
- We plan to have a smooth transition from cookie to authenticated user. I thought Jan Miksovsky had a great post on login barriers. I feel a lot of sites get this wrong by throwing up an abrupt login barrier too quickly. Tear down that login wall!
- you have your choice of login methods: either OpenID or the traditional three "name, email, URL" input boxes. You will be able to fully participate as a non-authenticated user -- both answer and ask questions. You may, however, be capped on reputation score and a few advanced abilities.
- We will not be seeding the site with the data from the Joel on Software .NET forum, as we feel it will bias the site too heavily towards that particular audience. Stack Overflow is intended to be non-denominational. We will be seeding the questions and answers on the site with the content generated by the private beta users.
- One of the great advantages of the new Web 2.0 economy is that there are so few barriers between programmers and the world -- assuming you're comfortable building a web site. Contrast this with the bad old days of distributing software on floppy disks or CDs.
- Can you remember the first time you used Google? When and where did you discover it? How has Google resisted succumbing the portalitis disease after so many years? It's admirable that they've pursued simplicity this long.
- I am mightily impressed that the iPhone can render Stack Overflow, even the jQuery and WMD editor parts. Kudos to Apple for an (almost) no-compromises mobile web browser that delivers a desktop browser experience.
- Joel loves his new Nokia E71 phone, and he cites the physical keyboard as the primary differentiator -- along with the superior third party Exchange integration.
- Revisiting my programming chair article. Although I loved my 1998 Aeron, I felt I could do better -- and the Mirra I'm sitting in now is a distinct improvement. I'm with Jason Calacanis: buy a cheap desk, and the best chair you can afford. Joel recommends the Chadwick Chair which I didn't get a chance to try, unfortunately.
- Don't forget overall computer workstation ergonomics; Joel also recommends variable height desks.
- On McConnell's cone of uncertainty, and the importance of keeping a list of what is to be done. You can't estimate without a list of tasks -- and you probably need to break the tasks down to very minute steps to really understand what those tasks entail, too.
- As Joel points out, there is the risk of the "Objects In Mirror Are Closer Than They Appear" illusion -- your mind's eye tends to fill in the gaps. You gloss over the details and presume things will be simpler and easier than they actually are.
- I believe in two principles during this project. First, having public artifacts that everyone can objectively see and judge. The team should develop a concensus opinion based on that reality. Second, have a plan -- but start on that plan as soon as possible! The sooner you get started, the sooner you will discover all the details and weirdnesses you could not anticipate or plan for.
- Joel proposes having major plans for version two, and just getting a small version one out the door quickly. The actual usage of the app may not be at all what you imagined, and you can change your approach more rapidly to accommodate those real world uses for version two.
We also answered the following listener question:
- Josh Parris: "Why did the Stack Overflow schedule blow out?"
If you’d like to submit a question to be answered in our next episode, record an audio file (90 seconds or less) and mail it to email@example.com. You can record a question using nothing but a telephone and a web browser.
The transcript wiki for this episode is available for public editing.