Recently the couchsurfing organisation changed the couchrequest system. For some this might be an improvement but in my experience it just creates problems. For starters one has to respond in time, before the date that the person requests for. This might be good for the guest requesting hospitality but for the potential host this means you have to reply sometimes instantly, if the date is for tomorrow for example. Otherwise the system doesn’t even allow you to respond!
A related issue is that my percentage of “CouchSurf requests replied to” dropped from one moment to another from 100% to 80%. And going back into my archive I seem to have only missed three requests that were send to me while I was traveling and that were short notice. This obviously doesn’t explain the sudden drop, since I have received hundreds if not a thousand requests over the past 4 1/2 years. And responding to the last one pending only made the percentage increase with 1 percent…
Weird, especially since again there is so little communication about the changes made in the system (this news-item of June 2010 seems to be the only communication). And this is not the only issue, as suddenly I seem to have a bunch of “neutral references” which before used to be positive. Or maybe the explanation is that the amount of bugs is increasing, as in my history of couchrequests there seem to be only a hundred e-mails. Of which the latest dates back to 07/11/2008 (obviously missing a couple of years).
How many full time contributors does it take to do this right?
From “the lead user interface designer”: “We have 30 full time contributors spread over 15 time zones in every continent but Antarctica and have been working this way for over a year.”
http://www.freshbooks.com/blog/2011/01/31/collaborating-without-an-office-part-2-habits-and-strategies/#more-9366
Related and interesting: http://www.couchsurfing.org/group_read.html?gid=7621&post=7918477#post7919321
BTW, I personally don’t care so much about the %age in my profile.
The CS reqest manager isn’t a problem, as you won’t be punished with a lower %reply rate unless you are asked a week in advance and don’t respond. Anyone asking at shorter notice is just taking a chance and have to accept that the host may not be in a position to track their messages every day.
Your problems with the %replied and too many neutrals also sound like bugs. Of course there are too many bugs, but at least it’s not a design flaw.
got this from a newsletter i got on the first of february:
“Announcements
Changes to reply rate calculation
As announced in this newsletter several months ago, when we rolled out the new CouchRequest system last April, we also made some changes to how the reply rate on profiles works. With the new system, responding to a request that’s already expired doesn’t improve a member’s percentage. We also made sure not to penalize people for missing requests received with short notice by specifying that requests received less than seven days before a CouchSurfer’s arrival do not affect the recipient’s percentage.
Now, we’ve made another change that we think will make this number more useful to surfers and more forgiving to hosts. It used to be that the percentage was calculated based on all of the requests that a member had received since they joined CouchSurfing. That meant that if someone was away from the internet for a few months several years ago, that would still be reflected in their percentage today.
In order to give a number that is a better representation of how actively a member responds right now, the percentage is now based only on the last 25 requests. We think this new system strikes a balance between encouraging people to respond to requests and preventing occasional periods of inactivity from marking a member’s profile permanently. “
Newsletters, what newsletter? Ah, the one “send to team-members” http://www.couchsurfing.org/group_read.html?gid=7621&post=7903200
Anyway, I also noticed I was quick with my conclusion since I did miss 5 requests while I was traveling… Which would make it into 80% but still not 81%, which is weird if you know math (76% > 80% > 84%). Or is this done because otherwise the percentage wouldn’t increase at all, as the number of ‘not-responded to’ didn’t go down nor up in this case. Would be nice to get some explanation.
Again this is another ‘feature’ that is implemented without communication
beforehand. Will CS ever learn?