Last Thursday at about 1AM Eastern, the Daily Kos tech team announced a site update that fundamentally changed the way users interact with the site. Previously, information about whether you had loaded (and been able to read) a comment was stored on the server. That meant that, no matter what browser or computer you used, you would be guaranteed to have red [new] flags appear beside the comments you hadn't read. But now, after the update, this information is stored exclusively in your browser.
For people who only use one browser on one computer, this update is, at least in principle (though as some have noticed, not in practice) a transparent change. But for anyone who reads Daily Kos from more than one location (e.g., a computer at work, at home, in the kitchen, or on an iPhone), the change is dramatic, and for the worse.
We are assured that the change was made on an emergent basis, and "that we'll end up with something that will work for you and for us relatively soon, and something that is much better at guessing which are actually [new] at some time after that[,]" but it is obvious that the individuals who decided to make this change were insufficiently familiar with how essential the [new] feature was to those of us who make an effort to read and follow the discussion in the most active diaries on the site. And for us, every day that server side comments are unavailable is another day that Daily Kos remains fundamentally broken.
The tech team seems surprised that anyone thought the change really mattered. But it did. Substantially. Imagine, for example, if your webmail provider forgot which emails you had already read. People used to routinely put up with having to manage read vs. unread email manually. This was how the system worked with POP. Today, almost every email system of significance uses IMAP, ensuring, for example, that when you read your email at home or on the road, you can tell that you did so when you return to the office. This functionality may not have mattered when most of us could expect a handful of emails per day. But now several per hour--or more--are quite common. The analogy to Daily Kos is direct. As one commenter explained:
You can't use the DKE all day discussion threads without [server side [new]]. . . . In order to actually follow the discussions now, you have to decide which computer/phone you are going to use for that thread and only use that device. Otherwise, you have to re-read hundreds of comments trying to find new comments.Another noted:
It's a huge change that is going to impact a lot of people and I am really frustrated that you launched it without any warning or discussion in the middle of the night.
When you go back to a thread you've already read much of, but now that record is erased, the inclination is to give up and disengage from the thread. It's way too tortuous to try to hunt and pick out new the comments you haven't seen.One captured the frustration that many of us feel:
I know that tech support does their best and I won't insult that, but this last system upgrade was NOT an upgrade. It's like going back to the dark ages.On top of all of this, if you regularly clear your cookies, use a privacy feature in your browser, or delete all of your private data every use, the change to [new] will impact you even if you only access Daily Kos from one computer.
My point in writing now is as follows: vague promises of future functionality that gets us back part of what we had before are not going to solve the community problem that was created with the update. I am calling for the tech team to level with us about the timeframe they have for redeploying server side [new] and, more fundamentally, to recognize just how important it was to so many of us. Even granting that the change was done quickly because of an emergency (the traffic spike from the debates), downplaying such a fundamental change in a middle of the night diary that few people would have read leads to an inevitable loss of confidence in the tech team: respect for the users, who are the key to this community and the success of the site, is absolutely fundamental. There needs to be a better recognition of what features matter and why.