Rainy days and Java always get me down


I've Moved My Blog

It's currently located at http://www.urlinone.com/blog

I should say "I'm moving my blog." It's a pretty painful process.

Pebble has blown up on me, and it's been many months since I've been able to blog reliably. I've lost posts. And now I've got to figure out how to migrate my past blog posts from Pebble to my new destination without all the URLs changing, lest external links become 404 Not Founds.

Why does everything in the 21st century have to be a three-day project???

For several months, I have been too swamped to blog for reasons I won't go into now, but I am irritated enough by what just happened that I am making the time to vent.

How many times have you been doing something on a web site that takes a while (or you get side-tracked), you finally go to submit, and you are greeted with the login page. Your server-side session has expired, and your work is gone. If you are lucky, depending on how the site is implemented, you can click Back and recover your completed form. If you are really lucky, you can complete the login, and the site will have maintained your form data and submitted it. If you are really unlucky, your work is simply gone, never to be seen again. Your choice is to recreate it from scratch or just say screw it, which is what I've just done. I'm writing this, instead.

It's enough to make you not want to do meaningful work in a browser. In fact, when I have lengthy text fields to write (like this), I typically do it in a text editor. But what are you supposed to do when it involves all kinds of form controls, such as checkboxes, drop-downs, radio buttons, or text fields? You are at the mercy of the server session to keep your session alive until you click submit. You can't really preserve that in a text editor, without a whole lot of effort.

I can't claim to follow Jakob Nielsen's work, but losing your work because of a session timeout is the ultimate in unusability (if there is such a word).

Either the browser needs to be made smart enough to preserve previous inputs to compensate for this case (perhaps a Firefox extension would do the trick here), or server-side code has to be a little more forgiving of form submissions from expired sessions. This is the kind of awful user experience that makes people, myself included, hate computers.

BTW, Simon, nice job on Pebble. I can write with impunity in the blog entry form, because, even if my session expires, Pebble lets me log in again and submit my blog entry at the same time. That's the way it ought to work...

... which brings me to the other half of my gripe. You never know how a web site works (or doesn't) until it's too late. Maybe there should be some kind of operational policy file you can query that tells you how long a session lasts, what happens to form submissions when a session expires, and so on. At least then you know what you're dealing with.

Even though Pebble works, I'm still doing a Select All, Copy All before clicking Post. One bitten, twice shy.


I replied to your post on my blog: http://rifers.org/blogs/gbevin/2005/3/15/session_timeout_solution
Dude, I totally agree with you. Case in point, Yahoo Email! If you type in a bunch of text in the email, come back a little later and somehow you are signed out (any number of ways for this to happen), trying to submit takes you to the login, but once logged in you can't go back. I think it's BOTH sides that should take this on. First of all, ALL server side apps that have sign-ons of some sort and require input should, across requests, save your input as if you were signed in, allow you to sign in, then take you to the next step. Period. It is horrible usability when this is not the case and I often dont revisit. I also will often write the support people and let them know I am an enterprise engineer and whoever is doing thier site has their head up their arse becasue of this situation and to give me a call, I can fix it. The other end is the browser. Sometimes I can't understand why I hit the BACK button which supposedly fetches a cached page BUT in FireFox at least, all my input is gone. GONE!! I don't remember this being a problem in IE, but I haven't used IE for a couple years now. I really do think that if a browser can cache pages, it should cache all the input values as well.
hey! you are full of spam
I am spam-free, for the time being.
why do people spam so much in other persons sites?
Interestingly, I feel this is partly driven by the trends in geographically distributed engineering teams and outsourced development. There's more of a need to tie it all togther through a product manager or product marketing manager.


Add a comment

Title
Body
HTML : b, i, blockquote, br, p, pre, a href="", ul, ol, li
Name
E-mail address
Website
Remember me Yes  No 

E-mail addresses are not publicly displayed, so please only leave your e-mail address if you would like to be notified when new comments are added to this blog entry (you can opt-out later).

TrackBack to http://www.leegrey.com/hmm/addTrackBack.action?entry=1110844060000