Endorsement Abuse

I review endorsements of the agile manifesto and have seen a variety of abuses.

# Responses

At one time I would respond to submitters that misunderstood the nature of contributions we sought. Eventually I prepared three canned responses that handled most cases.

Your post seems to be more an endorsement of yourself than an endorsement of the manifesto. I encourage you to post again with a focus on our work, not yours.

Your post includes a long url which will not be linked and will interfere with the text layout on the manifesto's endorsement pages. Please consider repeating the endorsement process without including this url.

Your post seems to be an inquiry better directed to AgileAlliance.org. I will leave it to you to follow up with them.

# Spam

Eventually the spam industry worked its way through our email exchange and started posting comment-oriented complements. "nice job, you may find this link interesting".

These appeared in bursts. When hit by twenty variations all promoting the steel industry association I took the website field off of the submission form. This must have broken someone's script because they stopped immediately.

I now reject any post with even a hint of a url.

# International

Since we have hosted translations of the manifesto in numerous languages I now rely on google to translate the 5-10 percent non-english submissions.

I've revised my review infrastructure to link directly to translate. The auto language detection feature worked well until recently. Now its choices are laughable.

# Length

Today I received an endorsement from an italian email domain that appeared in earnest but was easily ten times longer than what I would consider long.

After a pleasant enough correspondence the author sent me a shortened but still long version. I wondered, how long is too long? Would too long break the aging column layout logic used to display endorsements?

This script word counts all previous endorsements and orders them by length.

for i in *; do wc $i/*; done | sort -n -k 2,2

Here are the ten longest at this time. data

9 280 1732 000000041/0026 19 282 1799 000000027/0017 9 287 1650 000000005/0043 19 294 1730 000000245/0047 7 295 2100 000000050/0039 24 324 2010 000000015/0040 15 328 1917 000000009/0018 22 328 2060 000000061/0036 19 373 2562 000000341/0016 28 460 2876 000000199/0048

The author's shortened version will still be the 16th largest endorsement which, for now, I judge acceptable.

The nine digit numbers are page numbers leaving lots of room for growth. page