Michele,

The characters “

<" and ">

” were showing up as bookend characters in fully-formed web links copied from Word into the TAS.  They would cause the link not to show up in the HMTL output from the TAS. Annoying, but not the end of the world; they could removed by hand.

I don’t think that’s why the TAS closed, however. We and others were noticing errors where questions were repeated, questions were missing, and what we were *sure* were proper entries were not outputting properly.  It smells like a database error to me.  We were doing lots of checking, double checking, triple checking of everything.

Given how treacly slow the TAS became, a breakdown didn’t come as a huge surprise to me.  I just hope ICANN’s reasonable about it, and gives applicants a chance finish with a system that’s working, and aren’t dogmatic in their scoring if/when errors that might have been caused by this breakdown show up in the final answers.

It still beats me why they insist on this TAS system, which apart from being expsnsive is hell for applicants. Instead of telling the story of your application in a document that reads front to back in one piece, applicants are forced to work in little pieces, using little windows where you can’t see all text, and where attachments aren’t visible, virtually guaranteeing mistakes and misplacements. 

I understand that ICANN wants a standard interface to make it easy on the evaluators and to “ensure fairness,” although I don’t quite understand how TAS is supposed to achieve that.  I understand that with hundreds of applications, and evaluation teams across the globe, document management is an issue.  But there’s nothing like paper (or a PDF) with easy-to-read type, illustrations, clear headings (and so on) to convey meaning, and it strikes me that ICANN has given up a lot of richness merely to enforce uniformity. 

ICANN should have been thinking about the quality of their submissions, not the ease of evaluating them.  That the TAS system, which encourages sterility in order to ramp up efficiency, is also producing delay and confusion, is richly ironic.

The irony continues because now ICANN evaluators are likely to have to do intensive manual reviews to identify and compensate for any errors to applications that may have been introduced by the TAS, thereby losing all the uniformity and efficiency that was to have been achieved by automation.

Minds + Machines has competed in a number of tender bids over the last few months. In every one of those, formatted paper and/or electronic copies were required, sometimes in multiple copies, with lots of rules about packaging and delivery and so on. But at the end of the day, it was a much easier process than the TAS, and a lot less prone to error, with many fewer possible points of failure. 

In future rounds, I would encourage ICANN to use a more human and rational system, allowing for formatted documents, in order to get the best applications, instead of just the easiest to process. I think the applications we are submitting are quite good, but they would have been easier to read, more comprehensive, less prone to repetition, and would have conveyed more of the joy of innovation if we’d been able to compose them in Word (or similar) and submit as such or as PDF files.  And in the end, I suspect it would be easier for evaluators too.

Thanks for the article,

Antony