While using Attribyte to do some research for my next post, I found an article on Real Tech News that looked relevant. Unfortunately, when I clicked through I was presented with this "precondition failed" error page. What an arrogant error response. Because my HTTP request didn’t meet certain vague preconditions I should…

• Scan my computer for viruses, Trojan horses and spyware?
• Mess around with the settings of my personal firewall?
• Turn-off or bypass my proxy server?
• Turn off any download accelerators? (“they actually run slower!”)
• Switch to Firefox?

I should try these things so that I can read a blog post? How much time might someone less tech-savvy waste after seeing this error message - thinking they had been infected with a virus?

I’m sure I’m getting these errors because Attribyte’s bot is running behind the same router as my home network. Sure enough. Three of the hourly checks in the past day have failed with the HTTP 412 (Precondition Failed) error. Not all. Three. Why? After a bit of research, I’m pretty sure this message is generated by a PHP plug-in, BadBehavior. It seems to think Attribyte is, well, behaving badly.

For the record:

• Attribyte’s bots send an appropriate User-Agent string with all requests.
• The bots do run behind a caching proxy server – one that is very common. One of the reasons they are behind this proxy is to try to reduce remote server bandwidth.
• The robots.txt file is checked before requests, but it may be cached for up to 24 hours. Because Attribyte also functions as a personal feed-reader, directives excluding feeds are ignored unless the agent is explicitly excluded. However, entries are not displayed on public pages when the feed is excluded. I think this is fair. Anyway, the RTN robots.txt does not exclude any of the resources that are being requested.

Date: Feb 28, 2009 4:06 AM Number of Comments on Photo:0 View Photo...

Shel Silverstein wrote the lyrics for many of Dr. Hook’s early songs. The writing credits are too small on these newfangled cassette tapes.

Did you travel to Austin to attend the SXSW music festival, hoping the “badges” wouldn’t fill up the Tina and the B-Side Movement show? While you were there, did you take blurry concert photos with a digital camera that looked like this? (1.3 MILLION pixels!)  Did you get a CueCat in the mail because you were a Wired Magazine subscriber? If you have answered yes to any of these questions, you may be a Mth·ology “legacy” reader. After an eight year hiatus, I’m blogging again.

So long sad times
Go long bad times
We are rid of you at last
Howdy gay times
Cloudy gray times
You are now a thing of the past
Happy days are here again!

After just a few months, the SSD in Adrienne’s Dell Mini 9 failed. Annoyed, but thankful it was still covered by warranty, I called the Dell small business customer support number. After a short hold, I was talking to a competent tech to whom I described the problem: “I’m fairly sure the SSD in my Mini 9 died.” After a brief conversation about the error message, he agreed to send me a replacement with an OS image pre-installed . The following morning, less than 12 hours after my call, it arrived along with an empty box to send the whole thing back if the drive didn’t fix the problem. I’ve dealt with Dell’s customer service four or five times. Each time, I’ve been completely satisfied. I suspect it is because I order and deal with them only through the small business division, never the consumer side. In addition to having access to better service, your system won’t arrive pre-loaded with the trial-ware and “special offers” that plague consumer systems. I happen to own a small business, but if you don’t, their only requirement seems to be that you specify a “business name” with the order. Fred’s Bank?

HexLex, the blog, has been retired. The entries from the old blog were mostly off-topic, so they have not been migrated. The most interesting entries are summarized here.

September, 2004 - I wondered why HTTP proxy servers weren't discussed more often as mechanism for minimizing the bandwidth consumed by the proliferation of RSS readers and aggregators.

October, 2004 - EContent published a nice article about Kinja peppered with quotes from yours truly. Later that month, I went on (a much needed) vacation, with a USB "drive" loaded with portable apps.

November, 2004 - I wrote about hashcash , a spam-busting protocol based on SHA-1.

December, 2004 - I purchased TurboTax and it was broken on install due to JavaScript errors. After hand-editing a file, it worked. (Intuit's applications are horrible about requiring certain versions of IE to be installed.)

March, 2005 - I discovered the Mozilla Web Developer Extension . (I still find it indispensable.)

May, 2005 - "In addition to table elements, please feel free to use blink , and font when you insert ads into your RSS feeds. That whole "separate content from presentation" thing was a pretty stupid idea."

July, 2005 - The Google RSS reader was unveiled and, unlike most RSS applications, respected robots.txt (as do the Attribyte agents - I think it is important. I also questioned the validity of the stats quoted in Technorati's state of the blogosphere .

October, 2005 - Beta Bitter? My My .

First, 37signals called out Kinja for being perpetually beta, and now Mark has done the same. Mark claims that "companies are using the beta label to excuse flaws in their web based software beyond just getting initial development feedback. What else can explain beta versions of web applications that have been in production use for over a year."

Really? Do the GMail and Flickr support teams dismiss bug reports because there's a little "beta" on their logo? I personally respond to every Kinja bug report (there aren't many these days) and I have never cited the beta status of Kinja in my response, or refused to fix any real problem because Kinja is "beta." Some perpetually "beta" web applications are that way because the "beta" label is attached with no objective criteria for removal defined. Others may remain "beta" because completion of the "1.0" version is perpetually delayed. Maybe "beta" is a CYA forced upon developers by the legal department? One thing is pretty clear: The "beta" mark on a web application is meaningless to most users . Does my Mom know what the "beta" on the Flickr logo means? If "beta" scared away, say, 50% of the potential users and revenue, you wouldn't see it very often.

The decision to "release" Kinja as "beta" was made, I think, before I joined the project. If I thought "beta" caused scores of potential users to avoid our simple, free, and essentially risk-free application, I'd lobby to have it removed immediately. However, Kinja 1.05-Build 2073 is just around the corner and I don't have time to answer support email with anything more than "TS Kinja is Beta." Ugh. I feel obligated to add this - :)

My view is that the "beta" term is rarely applicable to web applications and is mostly meaningless in this context. After all, web-based applications are effectively "released" as soon as the pages appear on a public server. The risk/reward decision to use "beta" Kinja or GMail isn't quite the same as, say, developing an application that depends upon new features in OS/2 Warp Beta only to find them changed or removed in the final release.

princess lasertron: big omaha/girls in tech

For Debt Collectors, the Dead Are a Healthy Bet...

Wealthy Idiots Meet Idiot Reporter - The Plank

Goodbye Dubai | Smashing Telly - A hand picked TV channel

How To: Tether the iPhone or G1 To Your Laptop For Free 3G Broadband

Rigney Graphics | Lunch Meat | Snake Oil

How Theresa Hatt Caused The Financial Crisis

Pre-War Exakta 66 Instruction Booklet

Download - Ubuntu Pocket Guide and Reference