User First Web

Last night I pooh poohed a Slashdot article talking about the frailty caused by relying on TinyURL.

Less than 12 hours after I wrote my post, TinyURL.com started returning 404 errors prompting real concern from netizens.

Steve Rubel wrote, “The thought of an evaporating TinyURL…is all more than a little bit frightening, yet fascinating.” Marshall Kirkpatrick wrote about the dangers of using TinyURL and then jokingly harassed people on twitter who used TinyURL.

Everyone is missing the boat on this story. Less than 24 hours after a Slashdot article raising the alarm, the service that has been reliable for years goes down? Sounds like conspiracy to me.

Could it be that someone felt it necessary to prove a point about the dangers of TinyURL by taking the service to its knees? Or has the most unprobable outcome occurred and the long forgotten Slashdot effect finally returned? :-)

“There was a day when Web development was in a sad state of affairs. The majority of developers laughed at JavaScript, and were focusing on a killer new server-side web framework, or a new ORM library. The consensus was that that browser was dumb. The modern TN3270 if you will…

…We finally feel that the time has come for mobile phones to be the major device that users have for accessing their data and getting things done. You only have to travel around the world to see that already happening. Now, with the iPhone, a decent Java system, and more, we see the toolkits that will allow developers to build fantastic applications on the phone.”

Source: Devphone.com

Some recent news on web site performance:

• I got a chance to present to a class of web developers at PSU on site performance. I gave a variant of the presentation I gave a DevGroup NW. The presentation was video-taped and will be posted online soon. I enjoyed the class and had some great conversations after the presentation.
• Yahoo is using AJAX in an interesting way to speed up their site by pre-loading stylesheets for search results pages once someone starts typing in the search box.
• OReilly has announced a conference focusing on web performance called Velocity. Velocity will take place June 23-24, 2008 at the San Francisco Airport Marriott in Burlingame, California. They are currently soliciting proposals for presentations. I’ve got a topic in mind for a new presentation, and am thinking about pitching a session.

Getting a chance to present again was a lot of fun. I forget how much energy I get from talking to people about web technology. I want to thank Richard Appleyard again for the opportunity.

Slashdot today has an article asking the question, “Do Tiny URL Services Weaken Net Architecture?” The argument in the Slashdot article is pretty hilarious (Short version: TinyURL goes down and the Internet crashes). Instead of chicken-little scenarios, let’s talk about the one significant way that the growth in TinyURL-like services is changing marketing.

Shorten url services are all the rage these days. The key factors in their popularity are the increase in mobile devices, especially text messaging services, and emerging technology like Twitter and Pownce. When people use these systems, a premium is placed on short messages. Shortening a url to save characters becomes a necessity.

The impact of use of TinyURL-like services is that it becomes much more difficult to track the conversation surrounding your company or product. Savvy marketers today have multiple searches set up to scour the web looking for urls that point to their web sites. However, when people use TinyURL and other services, links to your web site are impossible to detect because the TinyURL is random.

This makes it more critical than every to watch incoming referring urls to find Twitter references and other places linking to your site. My testing shows that the referring url will show up correctly in web analytics despite the use of TinyURL.

The gPhone isn’t a Google phone. Instead, it is the formation of an alliance to develop an open platform for mobile devices.

The new consortium is called the Open Handset Alliance. The Alliance is formed around the Android platform that Google has contributed to the Alliance. Andy Rubin, Google’s Director of Mobile Platforms, describes Android as:

The first truly open and comprehensive platform for mobile devices. It includes an operating system, user-interface and applications — all of the software to run a mobile phone, but without the proprietary obstacles that have hindered mobile innovation.

If this is true—and the devil is in the definition of “open” as it always is—this could be a substantial development for mobile devices.

UPDATE: According the Open Handset Alliance FAQs, the platform will be released under Apache v2 Open Source License. The code will have a publicly accessible repository. It sounds very open thus far.

Last summer at Web Visions, I had an extended conversation with Kinan Sweidan of Ximda who had presented on determining location using mobile devices. During the conversation, Kinan talked about how difficult it was to develop on mobile devices.

The primary problem seemed to be carriers who don’t see their phones as a platform for other development. Handset manufacturers are beholden to the carriers because their hardware and software are useless if the companies like Verizon and AT&T decide not to allow the phone on their network.

The economics also favor the carriers because the cost of developing a phone is higher than most consumers will pay which is why the cost of the hardware is often underwritten by signing contracts that lock in services. Apple is rumored to receive another $432 from AT&T for every iPhone that is sells.

With this as context, it is possible to see why the Open Handset Alliance could be a game changer:

1. The price of developing new hardware will presumably decrease because of an open and shared development of the OS. A decrease in handset costs will loosen the hold carriers have on phone manufacturers by decreasing the need for underwriting of phone costs. This does not decrease their stranglehold on their networks. Legislation would be required for change that dynamic.
2. The combination of Android and Apple’s recent decision to release an SDK may mark a turning point in the understanding that the value of mobile devices will increase as they open up to outside developers.

If the second point has actually come to pass—if in the last few months the mobile industry has woken up to the realization that their future is dependent on becoming a platform for a wide variety of developers—then things will get very interesting very quickly.

The possibilities for mobile devices are astounding. 2008 is shaping up to be a very big year for mobile.

With all of the buzz this week about Google’s Open Social, everyone’s attention is focused on the web-based social networks and missing the impact of mobile technology on social media. Per usual, the Communities Dominate Brands blog is ahead of the game on this one.

In Tomi T Ahonen’s latest post on Communities Dominate Brands, Tomi points out that:

Informa’s latest Mobile Industry Outlook report for 2007 reveals that yes, mobile social networking services did continue their dramatic growth for the past 12 months, and are already worth over 5 billion dollars in 2007.

$5 billion dollars! This again dwarves the revenue associated with web-based social networks. Tomi’s post echoes one of his posts from a year ago where he put the then $3.45 billion in mobile social networking in perspective:

3.45 Billion dollars this year! Wow. A bit of context. All of iTunes revenues last year were about 400 million dollars. TV-interactivity (voting for Big Brother, Survivor Island, Pop Idol etc) were worth 900 million dollars. Internet gaming revenues, all multiplayer games etc, were worth 1.9 billion dollars. All internet adult site revenues were worth 2.5 billion dollars in 2005… Oh, just to be clear - that mobile digital content revenue is more than all (non-mobile phone based) online social networking revenues combined. In only two years, the mobile side of digital communities has shot ahead of the online world. Amazing!

There you go. If you weren’t previously convinced that mobile is the next big thing, it’s hard to dispute the fact that today’s big thing—social networks—is already bigger on mobile devices than on PCs.

Someone recently pointed me to this funny blog post entitled “15 Unfortunately Placed Ads” (Some of the content is a bit risqué).

The humor of the ad placements alone isn’t enough to warrant a blog post. Instead what interested me was the online ads were almost certainly placed by software attempting to match keywords.

One of the example is a Folgers Coffee ad placed next to an article entitled “Coffee Might Trigger Heart Attack in Some.” Folgers likely bought ads to be placed on Yahoo’s content network whenever the word “coffee” showed up. However, they probably wouldn’t chose to place their ad next to this story if given the opportunity to chose.

This is the danger of context-based advertising. The current filters are not smart enough to know whether or not the true context of the content is conducive to the advertising.

I wonder whether sites like Facebook will encounter this problem more frequently because it is just as likely that someone is going to be ranting about coffee as they are praising it. I imagine Google and the other leading advertising engines are already experimenting with ways to determine the positive or negative nature of the content.

I think we’ll see a lot more ads with unfortunate placement before the technology progresses to the point where it can be prevented automatically.

Twitter’s Facebook application has recently been updated to allow you to change your status whenever you post to Twitter. This sounds like a good idea in practice, but in reality, it turns out to be simply redundant.

Status updates is one of the few Facebook items that are available via RSS. Now that Twitter updates change the Facebook status, most of the RSS that Facebook provides is now full of updates that I’ve already read in Twitter.

The only updates I can get out of Facebook’s walled garden is something that both started outside of the garden and something that I’ve already read. It’s easy to see why everyone thinks Facebook is so useful. </sarcasm> :-)

Read/WriteWeb asks readers, “Is the mobile web (or, rather, using mobile devices to access the web) finally coming of age?”

In a word: Yes.

For more details, on why:

• It’s a Mobile Web. We Just Don’t Realize It.
• 2.7 Billion Served: Mobile Phone Usage Dwarves Everything Else

The real question is when will it take off. Are we looking at a 6 to 9 month or a 12 to 15 month horizon before the groundswell takes off.

Web Analytics guru Eric Peterson recently came to a similar conclusion about the mobile web. I think Eric nailed it when he said that the coming mobile web is probably less like a wave and more like a incoming Tsunami.

One of the big features in HTML5 has been implemented by the Safari developers and it’s a doozy that I wasn’t aware of: client-side database storage.

From the Surfin’ Safari blog:

The client-side database storage API allows web applications to store structured data locally using a medium many web developers are already familiar with - SQL.

Like cookies, you can store the databases per domain. I’m struggling to determine if the persistent storage in Firefox is the same thing. Firefox’s implementation states clearly that it isn’t available to web pages (only “trusted callers”). IE seems to be (per usual) implementing something similar, but slightly different.

Niell Kennedy wrote a good summary of the different techniques for boosting Ajax performance using local storage and why this would be beneficial.

I’m starting to get excited about HTML5. This database feature could be very useful for web applications.