Screenshot / Code Snippet

Column 80


Browsing the existing Stack Exchange family of websites was OK in a text-based browser, but when you're stuck in a Linux shell with no way out and no other computers around and you just know you saw the answer on Server Fault last week, this is how you can find it easily.

It assumes you already have existing knowledge of the SE family of sites in order to navigate.


Public Domain. Feel free to do with it what you will. Grab it here (note: This is the old v1 API. I have not packaged v2 yet).


http://www.column80.com (note: looks like rubbish in modern browsers, Firefox, Chrome, Internet Explorer, etc. It has got to be viewed in Lynx to get the full effect). offline as of late 2019 due to virtually no traffic over the past 10 years.


Optimised for Lynx. This sounded pretty niche when I made it, but it looks like it might already have been some use to some users!


Leave some feedback in this question.


I haven't done any programming in about five years, so it was also a way for me refresh my skills. It uses the PHPStack by Thomas McDonald. If you really want the source and you promise not to make (public) fun of me, then I'd be more than happy to share. I've cleaned up the code a bit, and incorporated some suggestions from the public, so I feel comfortable releasing the code. Grab it here.

  • (10th Jun 2010) now updated to incorporate the unannounced API changes
  • (25th Jun 2010) changed front page to retrieve data from StackAuth, and changed URL format to prepare for URL Rewriting (using stack.PHP)
  • (11th Jul 2010) now updated to work with the v1.0 API (the list of Stack Apps sites is looking healthy!)
  • (9th May 2014) updated the site to use the v2 of the API. Removed some never-used features such as tag search and user profiles, as I'm pretty short on time.
  • 79
    +1 for acknowledging the console-lovers and treating them to this! Jun 5 '10 at 0:43
  • 11
    ...and for mentioning Lynx :) Jun 5 '10 at 0:43
  • 3
    +1 And it does look OK in Chrome. The only issue I saw was long unwrapped lines, which makes sense when optimising for a text-based browser.
    – Mark Hurd
    Jun 5 '10 at 1:28
  • 1
    @Mark, one of my plans for down the track is to wrap words nicely at 80 columns which will fix that, and make it look OK in modern browsers as well (thank God for the <pre> tag) Jun 5 '10 at 1:39
  • Yay, something useful came about as a result of my creations. What are you using for the search, considering that was the only method I haven't quite figured out the best way to do?
    – nobody
    Jun 5 '10 at 6:55
  • @Thomas, I used your search function out-of-the-box. It worked quite well, because the API requires the use of the options rather than a string, so I found it worked just fine. Jun 5 '10 at 7:35
  • Since you are using PHP, you could just use the wordwrap function: php.net/manual/en/function.wordwrap.php , no need for any pre tags me thinks?
    – SeanJA
    Jun 6 '10 at 22:13
  • @SeanJA, Good idea, thanks. I'll get that implemented sometime today. Jun 6 '10 at 23:12
  • @SeanJA, thanks I've done the word wrapping. It looks better in modern browsers now. Jun 7 '10 at 9:20
  • 1
    +1 for making me laugh and remember that I work at a company that loves their AS/400 Jun 7 '10 at 12:25
  • Nice idea! Have you tried it with HTML tables though? I've had good experiences with their rendering in lynx
    – russell_h
    Jun 11 '10 at 18:04
  • 1
    @Simon - thanks. One of the MySQL databases went over its size quota, should all work now. Jan 27 '11 at 21:39
  • 1
    Absolutally - Go for it :) Mar 6 '11 at 10:48
  • 2
    @muntoo - if you read the description, it's not about being addicted, it's about being stuck in a console session with no other options and you still need to get access to Server Fault. Apr 11 '11 at 0:14
  • 1
    @hhh - no. The API does not expose these methods, not even the v2.0 API. Feb 28 '12 at 22:09

Just wanted to give you a heads-up...

One of the questions contained <button> in the title, causing the entire rest of the page to be rendered in a button. You might want to run the title through htmlentities() first.

  • 1
    Ooh, thanks. Do you know what the site & question ID is? I'll try it out in Lynx to see if it screws up on there. Sep 1 '10 at 2:31
  • @Farseeker: Here's one: stackoverflow.com/questions/3563753/… Sep 1 '10 at 3:01
  • 1
    htmlspecialchars() is enough, isn't it?
    – nyuszika7h
    Feb 12 '11 at 11:57
  • yes, it is. that escapes all dangerous characters Apr 2 '11 at 19:10
  • Yes, but it seems to not show the \<span\> at all in the title now. Also, the button looks like "\<submit\>"(Screenshot here: imgur.com/EUzzA), also happens in listing.
    – nanofarad
    Jun 18 '12 at 21:43
  • 1
    @MarkHenderson FUBAR is an adjective(Especially the first word) so your comment should read "Yep, it's also FUBAR[or FUBARed] in Lynx."
    – nanofarad
    Jun 19 '12 at 11:34

This is nice, but it gives the "You're not using Lynx" warning for other text-based browsers. It looks and works great in w3m as well, so it should probably skip the warning in that browser as well.


Just pointing out: this system actually violates the content license because it does not link back to the original posts or authors in any way on the questions.

From the blog:

  1. Visually indicate that the content is from Stack Overflow or the Stack Exchange network in some way. It doesn’t have to be obnoxious; a discreet text blurb is fine.


  2. Hyperlink directly to the original question on the source site (e.g., https://stackoverflow.com/questions/12345)


  3. Show the author names for every question and answer


  4. Hyperlink each author name directly back to their user profile page on the source site (e.g., https://stackoverflow.com/users/12345/username)


  • Interesting. I wonder if this applies to remixes from the API, but I guess I can comply. Not entirely sure how useful linking to the originals will be from a plain text browser, but oh well. May 22 '14 at 1:26
  • @Sally Thanks. I had forgotten about this. Ironically for years I did comply with the license but I had to make some changes a few years ago and I suspect I undid the attribution requirements. The source code for this site is, honestly, probably lost. I will see if I can dig it up to make it compliant. Feb 18 '16 at 11:08

Searching with empty string (http://www.column80.com/api.v2.php?a=stackoverflow&s=) gives error:

Fatal error:  Uncaught Error 400: An API error has occurred.
  thrown in /home3/markness/column80.com/stackphp/src/api.php on line 284
  • 1
    Thanks. I've got some stuff I've gotta do on the site in the near future, so I'll address this then. Mar 4 '16 at 22:30

Nice but is there a nice url we can remember that goes straight to:

http://sa.column80.com/?api=0 maybe?

It would be nice to have the search button BEFORE the title and tag options.

Also I get bugged about cookies, do you really need cookies?

  • Honestly, I hardly remember how this little app works. It might be time for a refresh. The cookies were from memory from an earlier revision that must not have been taken out. Sep 27 '12 at 1:00
  • Cookies are gone, URl structure has been changed. May 8 '14 at 23:10
  • Nice. The search is a bit hit or miss though. (E.g: it's order dependent) Have you considered using Google or Bing? May 9 '14 at 4:51
  • I am just using the built-in Stack Exchange search. I believe it's a full string search, but not entirely sure. Problem with google search is that it will redirect you to the full SE site, not my Column 80 one. That might be OK, it might not be. May 9 '14 at 4:57
  • It looks like it's possible to extract the question number from the URL. May 9 '14 at 14:01

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .