And a day later...
Hey, where’s that documentation? Yeah… I’m working on that. ;-)
Actually, over the past day I’ve just been getting things more “settled” if that makes sense. Lots of early feedback which was nice, and some stupid easy bugs that needed to be cleared up. I think I’ve gotten the encoding to a reasonable level right now - though Windows-formatted characters (like on cnn.com) are still giving the Content Adapter process grief, it’s a lot better than it was. I swear though, that stuff is like wack-a-mole… as soon as I get the characters to look good on one page, the next one pops up that’s almost unreadable. Again, if you have a favorite site that’s giving you issues, please check out the forum and drop me a note.
Thanks to everyone who linked to me yesterday! It’s been the mobile crowd that’s really given a lot of support so far, and it’s great. These guys are the pros and so grok what I’m doing and give me spot-on feedback. Once I start getting word out to the rest of the world (pinging other blogs, doing some ads on AdMob) then I’m sure I’ll start getting more general feedback along the lines of, “It don’t work, you suck!” which is valuable as well, but less focused. :-)
Ahh, blogging… been a while since I did this. A year exactly, actually! So nice to have a spot to brain dump, hey? But it’s also been nice to not have a blog so that I kept all those ideas I have bottled up until I did something with them. I can’t believe it took me a year though!
So what’s coming next - well, the documentation and screencast is a must, but I’m more concerned about the spots in the service that aren’t where I want them yet. I had a great email from a Blackberry user today that was wondering why I was splitting up the pages into such small chunks, when in my presentation I called out the other transcoders for exactly that same problem. I’ve been erring on the side of caution to make sure the pages display at least, but I definitely want to get to the point where I deliver as much of a page that a mobile browser can handle. My theory is that there’s no latency in scrolling - users will much rather scroll down a page than click and wait.
But what’s the best way to make sure that the page sizes are optimal for a specific phone? The device databases that are out there are notoriously bad, first of all, so relying on them isn’t going to work. I could allow the end-users to choose… But I think that personalization is a crutch that mobile developers use when asked about this - it’s a hard problem, so they say, “oh, we’ll let the users enter in their page size.” Uh-huh. Do you know how few people actually do that? So you end up screwing the vast majority of your users for the benefit of a few power users, simply because you didn’t want to tackle the problem.
I’m trying to keep Mowser as stateless as possible actually and here’s why: First, I didn’t feel like writing the code for Yet Another Signup Page - it hurt my soul just thinking about it (and yes, I know about OpenID, but it’s not viable now for mobiles). But more importantly, I think the benefits from signing up aren’t worth the time and effort for the end-user. I don’t want to create Yet Another Bookmarking Service, nor do I want to create a system that’s dependent on a group of enthusiasts for revenue (especially since the revenue right now is from advertising and affiliation stuff). What I want to do is build a system that provides several services - a directory, a content adapter and shortcuts - and has business model an inherent part of that stateless system. That doesn’t mean I won’t have user IDs sometime in the future - or even tomorrow if it strikes me as needed - but right now it’s not on the big list. Eventually though, as we all know, all applications evolve until they can read and write email so it’ll happen. :-)
The big list is definitely big. In the short term I want to get condensed URLs out there and the ability to send them to your phone or IM/Twitter buddies. I need to keep implementing the features I’ve already talked about - like parsing the handheld stylesheet - and I need to keep tweaking the site in general to make sure users can navigate around quickly enough as well as discover new content. I also need to go through the 500 or so links in my mobile index and clean them up. A command-line SQL mishap a while ago destroyed all the descriptions in the DB… DOH! so I need to go through and re-enter them, which of course means making the admin pages to manage that data, etc…. All while keeping an eye on the logs to make sure nothing catastrophic occurs. Lots of fun! :-)
Okay, back to the trenches…
-Russ