Information Week did an interview with David Harper, founder of Winksite, and one of the organizers of NYC’s Mobile Monday regarding mobile search engine optimization.

There’s some great tips in the interviews, especially for people who are just starting to set up their mobile sites.

But there’s an interesting trend that may push back against SEOs being effective. Check out the graphic on Yahoo Mobile’s site.

Search engines are taking it upon themselves to get the user to information without leaving the search site itself. If this continues, more and more information could be harvested from sites and presented in a search portal environment without the user ever visiting the original mobile site.

I think you have a lot to worry about now if your mobile site presents basic information like weather or other obvious basic info. You don’t need to leave Yahoo or Google mobile search to get that stuff.

If your mobile site is actually a service, like a mobile social network, you’re probably safe for a while, since search engines won’t be able to offer a quick replacement for your service.

If you’re just some random bit of information, like “Generic drug alternative to Ambien” - you need to worry where the search engines will draw the line. I’m sure all of them want to be able to pop up the succinct answer to the user without linking out, but they probably can’t do that effectively now. So how long until you suddenly see your mobile traffic drop off because they mined your data?

I read the complaints about EDGE speeds here in the states more and more now that iPhone users are suffering with it. I’ve used EDGE for years, and have adapted my browsing and application habits to fit into it, so I tend not to subscribe to most of the complaints. Sure, 3G is much better, but I’m on Long Island, and don’t get 3G out here. Plus, when I used to go into 3G areas with my Blackjack, I often forced it to EDGE to give me more than twice the battery life.

However, it appears that I’ve been remiss by discounting all these complaints. I was in Manhattan today for some meetings and was demoing on the iPhone and other devices over EDGE. Out here on Long Island, things load pretty well, but they fell flat on their face in the city. So I used the DSLReports Mobile Bandwidth Test to compare Manhattan’s EDGE service with the service out here in the suburbs of Long Island.

It’s obvious that loaded networks will be slower, but I thought AT&T mitigated things like that with better infrastructure in dense areas, and the difference wouldn’t be drastic.
Well it’s actually quite sad. In Manhattan, I averaged about 85kbit/sec with a 1.5 second latency. On Long Island, I averaged about 156kbit/sec with a .996 second latency. In practice, that averaged out to a user experience that felt over twice as slow.

So I have a new found appreciation for you urban EDGE complainers. Though we don’t have any HSDPA out here, so it kinda balances out.

What are the speeds like in your city? Is New York a worst case scenario? Do you notice a drastic change in speed based on the time of day?

As a side note, I realized that DSLReports have a new iPhone / JS-enabled device Test. It will probably be more accurate with client-side assistance, so if you’re on an iPhone or have a Javascript-capable device, use that to test instead of the basic one.

Sony released a new Folding@Home client for PS3 today, and it’s the first PS3 app released (in the US at least) that can be used through the PSP’s remote play feature. Check out the video of my PSP connecting and using it below.

I’ve been excited about the idea of using the PSP as a media access device for the PS3 ever since their first vague announcement for it well before the PS3 release, and boy have they made us wait for it.

The functionality has come in dribbles, with the PS3 not supporting remote play on release, then adding it in for 60Gig local WiFi connections only. I have the 20 gig (which I’ve upgraded to 160), so I had to wait for them to release Internet mode for remote play before I could even try it.

When I finally could use it, it would only let me view pics, listen to mp3’s, and watch videos. This was worthless, since I had no intention of copying stuff from my media server to the PS3. I could also choose to use the relatively crap web browser as well. But why?

Then they released DLNA media server support, and I could use the PSP to connect to the PS3 and browse the shared media through it. Here’s where it became useful - I used it to watch some TV episodes from a hotel room while I was away, and it totally worked.

But what I thought would be really cool would be to do things like check in on an RPG, or do MMO-type things like Playstation Home while away. This is technically feasible, but with the PSP’s reduced control scheme, it’s tough for a developer to work in that type of support. The PS3 component library also probably steals a bit of memory away from your game or app for every system feature you use, so developers would be hard pressed to give up memory for a feature with questionable potential.

This Folding@Home step gives me hope that we’ll see more apps and games integrate remote play. Many games like RPGs can deal with a reduced control scheme, and turn-based RPGs and casual MMOs don’t even have to worry about network lag making gameplay impossible. They will have to deal with font and readability issues however. I still believe it could be a well-implemented and useful feature on the right game.

Let’s see where this goes. The PSP platform as a multimedia device is getting long in the tooth (PSP slim - bah), but it’s only because Sony is dropping balls left and right with it. Get me excited again Sony. I’ve had games running for months straight on the device (yeah, I really do love playing games on it), but give me a reason to exit them and use the machine in other ways!

As the conga line of iPhone apps gets longer and longer, I’m beginning to see enormous differences in design philosophy and usability between them. Some of these are tragic, which is why I wanted to compare one of the quickest-loading yet fully-functional and attractive apps, Facebook, with a potentially very useful yet load-time crippled app, Accuweather.

I use the Facebook app frequently - if only for the fact that it takes seconds to pull up and glance at. I would use the Accuwaether app a lot more than the Facebook app, but over EDGE it’s just too annoying to pull up, and the tragedy is that it doesn’t have to be.

Look at this streamlined, pretty, and functional Facebook app. It’s a perfect blend of visual sheen with practical usability with a 5-8 second load time for me. Yeah, over EDGE! It should be a poster child for AJAX apps on the iPhone.

This weather app from Accuweather also looks very nice. Weather that’s more detailed than the iPhone’s built in widget is something I tend to want quite frequently, and I like the idea of this widget very much.

However, it takes about 20-25 seconds on average to load - about three times more than the Facebook site. That’s just a little past the upper bound of my tolerance level for a mobile information app like this. Sure, it needs a lot more image assets than the Facebook site, but it also wastes a lot of load time for its design as I’ll note below. There are a bunch of things that can be done to keep the Accuweather app pretty and make it efficient at the same time.

You’ll see below each image above is a load table produced by Firebug. I’ll reference them below, and you can grok them to see how each asset affects the load time for each page.

Many of these observations will be completely obvious to seasoned JS developers, both on the web and mobile. But they’re worth pointing out because these apps could not be more different in design philosophy, and in this case, Facebook’s optimized design makes the difference between usable and novelty.

Keep the Number of Requests Down

Over EDGE connections, not only do file sizes make a difference in your total load time, but the amount of separate requests to the server as well. If you look in the load tables, the Javascript sources for both apps total around 7kb, however Facebook puts them all in one file, requiring only one request. Small things like this make huge differences over cellular connections.

Don’t Use Images to Style

If you have a taste for design, CSS can be quite powerful by itself to create a very attractive application. Small decisions like choices in color, font size and layout can make a huge difference. You don’t need to resort to textured background images, rounded corner images, and textured fills for tables. Save the images for what need to be images - icons, headshots, etc. Not only do style images increase total KB load, they also violate the rule above and create more requests to the server. Look at the PNG list for the Accuweather app. It’s a mile long - and most of them are for style and not for iconic information.

Optimize Your Asynchronous Loads

Create the illusion of faster loading. After your Javascript loads, you have decent control over the order in which you fill in your data. Fill in, format and display the light stuff first - text. The text is usually what people are after anyway. Facebook does exactly this in all of their lists - text appears quickly, and images follow. Accuweather could do this with alternate image text formatted and positioned nicely in their tables that are gradually replaced with their icons as they load.

Akamize!

Eliminate as much non-mobile network overhead as possible. Akamai, of course, is the industry choice for this, caching your content on many sites all over the world in order to decrease network distance for any request. I know this isn’t an option for most people, and I don’t know if this is the reason why Facebook’s assets are generally just faster loading than Accuweather’s. But I suspect it is.

At least get some solid load balancing and a good host for the content. EDGE produces enough latency, you don’t need to add to it with poor hosting.

By just improving these few aspects, Accuweather could take their app from useful demo to pretty and practical. I do have to say that they were one of the first out of the gate with an iPhone app, so credit is due there. So let this serve as a nudge to give it another revision.

As for the rest of you, I’d like to hear your mobile AJAX optimization tips.

The title and look of this competition discredit it from the start, but who am I to thumb my nose at potential funding for great mobile ideas.

A quote from the into:

We are searching for innovations that will bring the mobile business to a completely new level.

Can a single independent product idea bring mobile to a completely new level? We’ve been looking for that idea for a while now, and I’m becoming more and more convinced that this miracle will likely come from something with an existing web userbase - like a social network - or from some freaky-popular TV tie-in like how American Idol put SMS on the map in the US. If this doesn’t happen, it will come painfully gradually, as the next generation grows up with difficult mobile interfaces and doesn’t mind dealing with convolution to get what they want on their phones. Or everyone will copy Apple and make usable phones (ahem).

For now, ff this miracle product is an app and not an SMS service, it will have to be pre-installed on phones by the manufacturer or carriers so people don’t have to install it. I don’t just mean installed on all Nokia phones either. We all know how much they throw on their phones from the factory nowdays anyway. That’s not enough to gain traction - especially in the US.

Then again, I’m very US-minded. Is the rest of the world poised in a tiger crouch, waiting for some cool app to come down the pipe so they can bend over backwards to install it? Nokia’s international footprint is enormous, so will pre-installing the right app on their phones create an international hit that the American majority will never even hear of? Will this international phenomenon be so huge that the creator will be able to dress like a pimp who looks like he just got lucky in Blackjack and have two prostitutes to hold a phone to his ear while Red Herring takes a pic for their cover?

I’m guessing no. After all, pimpin’ ain’t easy. Neither is mobile.

I’ve been uploading a bunch of vacation photos to Flickr recently, and discovered that in addition to only displaying the most recent 200 photos in your timeline, a basic account will cause badges that you create to stop functioning when photos drop off the timeline.

Specifically, by adding the new photos, I caused the badge in this post to break as those photos have fallen out of the top 200. So to get it working again, I was forced to upgrade to a pro account.

In a Web 2.0 world, this is unacceptable. Sure, limit my uploading sizes and amounts, and remove features. But don’t cause something that worked before to suddenly stop working. If all Web 2.0 services behaved this way, we’d see a ton of broken badges, links and data all over the web - and that’s not very Web 2.0 is it?

Bad Flickr. Bad.