X-Message-Number: 23971 Date: Sat, 24 Apr 2004 02:46:05 -0400 From: "Kevin Q. Brown" <> Subject: Re: new "rating" system > Not a rating system, but a classification system. And a flawed > system, because a single item can have multiple classifications. Hi Kennita, Thanks for checking out the rating system and taking the time to give your feedback. I'll rate^^^^ categorize your message as "Insightful", once it is distributed in the next CryoNet digest. (Your message and my response to it will be in the same digest. Anyone can respond that quickly since the CryoNet queue is accessible from the main CryoNet web page.) > Many is the message I have seen that is insightful, interesting, > informative, and offtopic at once. Or redundant and funny. Or > informative and flamebait. Or... you get the picture. I had > expected the rating to be from 1 to 10. I was inspired by the system used at slashdot.org, which seems to do quite well with a single categorization for each message. But the needs of SlashDot are not the same as the needs of CryoNet and you rightly point out that multiple categories can apply to a single message. Rather than requiring the rating to be just one of the various categories listed, I'll look into supporting multiple categories. Rather than placing one's entire vote as "Interesting", etc. it could split into two 0.5 votes, three 0.33 votes, four 0.25 votes, etc. > On the methods: I'm not fond of needing to open a browser > window to do ratings. What if I could cut and paste the Cryonet > headers into a message to "" ... Sorry, I have no plans for email-based rating/classification submissions, since parsing hand-entered data, with meaningful feedback on errors, can be messy and most people prefer web-based solutions that give instant feedback. Perhaps the web-based system could be better streamlined, though. > Even if we kept the browser window, an eleven-digit rater code > seems excessive. A four-digit number would allow for plenty of > growth (and I would have a chance of remembering it). FYI: The system uses an emailed "rater code" because I don't want to make it too easy for someone to "stuff the ballot box". Yes, the rater code is long. I'll add "shorten rater code" to my TODO list, but I hadn't intended anyone to remember it. Instead, just copy and paste it from your email. If your browser has cookies enabled, then the rate.cgi script now will set a cookie to save your "rater code" for you. If you lose that, then just generate a new "rater code", which will replace the old one. Those codes are disposable; you shouldn't have to remember them. You do need to remember your CryoNet subscription address to get your "rater code", though. If you have multiple email addresses, with some addresses forwarded to others, you might not remember which address you used to subscribe to CryoNet. > On further reflection, I think the whole rating system is a bad idea. > It seems to be an obtrusive, blanket, omnipresent solution to an > intermittent annoyance . . . You probably didn't know that I also use a hammer to crack open peanuts, hire a document destruction agency to shred my junk mail, and save all CryoNet postings on microfiche in a temperature and humidity-controlled environment. :-) Seriously, calling this overengineered is a valid complaint. But it's also interesting and, I think, useful. This isn't the first time that I provided a lot more mechanism than the minimal necessary. In message #23652 I offered to JavaScript-encode one's postings, ** which was a natural extension of the work already done to protect email addresses in CryoNet archives from being harvested by spambots and to protect the entire contents of the CryoNet queue from search engine robots. Offering that service may have been a mistake since it increased my workload and also enabled people to avoid responsibility for their postings, but I offered it because it was technologically interesting and might be appreciated by some CryoNet subscribers. How does that relate to the rating / categorization scheme? Your "rater code" could provide identification for a web-accessible form to JavaScript-encode one or more of your CryoNet postings. Then you wouldn't need to send private email to me with the request. I'm not promising to implement that immediately, but mention the possibility as an example of how one overbuilt tool can dovetail nicely with another to create unanticipated new and interesting services. Getting feedback on one's postings should be interesting, too. Rather than providing just a single number ranging from 1 to 10, as you expected, the multiple rating categories will give CryoNet posters a better handle on the ways their messages are viewed by others. Maybe I should add a "long-winded and verbose" category for my overlong, detailed messages? Kevin Q. Brown ** Note that JavaScript-encoding a CryoNet message should not be confused with any secure type of encryption. The encoding just scrambles the text and does not require a key for decoding. Nevetheless, it thwarts spambots and search engine robots while remaining easily invertible by a JavaScript-enabled web browser. Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=23971