Web-Browsers and Web-Apps

While web-apps are being used more and more, one thing remains constant: most users first launch a web-browser, navigate to the app via a bookmark or by typing the app’s address in the browser, then they start working. In other words, using a web-app begins with surfing the web, which is an extra step, and a potentially very distracting one.

The distraction doesn’t stop once the app is showing in the browser. A web browser typically displays a menu bar at the top, and the standard web toolbar with Forward and Back buttons, Reload button, an address bar and a host of other things.

Web browsers were designed to… browse the web, not to serve as a platform for applications that happen to be delivered from the net. As a result, most of the buttons and menu items don’t make a lot of sense when using a web-app. For example, what is the meaning of a “back” button within the context of a web app? This is the subject of much debate, but the simple answer is that it doesn’t mean anything. If it disappeared it would be to the benefit of web-app user and designers alike.

All other buttons are directly related to surfing the net (”browsing”), not using a web-app. The address bar is useful for knowing where you are when navigating between different pages, or as a way to get somewhere quickly by typing in an address. But web-apps are not address-driven, they are action-driven and this makes the address bar far less useful in that context.

The point is that although the web is a wonderful platform for delivering applications, the web-browser is an awkward program for using an application.

Mozilla Prism: The Non-Browser

With all the useless stuff cluttering a browser when a web-app is being used, surely somebody would think of creating a more minimal browser that does away with all navigation-related amenities. That’s exactly what Mozilla Labs has done with Prism, and in its first incarnation Prism can be described simply as a Firefox browser devoid of all menus and toolbars. However there is more.

When you set up Prism to be used with a web-app, you can select an icon that represents the app and place it on your desktop, or in your Windows start menu. Clicking on the icon launches Prism with that web-app. As a result, the web-app user launches the web-based application as if it were a desktop application.

In addition to these benefits, running an app in Prism is likely to be more secure since it runs separately from your web browser, and the app feels faster since the memory footprint is smaller.

What’s Next?

The team developing Prism is exploring future directions for their experiment, and they are looking for feedback from all who might have an opinion on the matter.

So if you have an interest in the future of Web2.0 in healthcare, you may consider visiting the Prism forum and making a suggestion.

The web browser is becoming one of the most important desktop applications now that users are relying on web-apps more, and their performance is key to the success of the more advanced web-apps.

The user’s perception of the responsiveness of a web-based application is in large part dependent on how fast the browser can render heavily styled pages, and execute lengthy bits of JavaScript code.

If browsers are so important, then the question should be asked: “How good are the browsers?”

The short answer: they’re fairly good, and getting better. The browser wars are heated these days with no less than 3 major contenders (Safari, Firefox, and Internet Explorer) so the competition to create a faster, more efficient browser that makes web-apps feel like desktop apps is fierce.

For a longer answer see this JavaScript speed comparison for current browsers.

The link above confirms my observations that Safari is fast and FireFox 2 is dog-slow. However, there is good news for FireFox fans: FF3 (currently in beta) appears to be significantly faster than its older brother according to the article linked above.

Time will tell how the final version of Safari for Windows will perform, but for the time being its Mac-based twin is the smoothest browser to run our Medical Resident Scheduling web-app on.

Could a Facebook-for-healthcare-professionals take off?

The idea would be to allow healthcare professionals to network, communicate and share files such as presentations, and spreadsheets among each-other.

It seems to me this idea, suggested by Nicholas Carr for all industries, is a pretty good one. Social networks are far less awkward to use than email for sharing documents. Think of how easy it is to post pictures to a profile and let those who are interested in viewing them do so at their own convenience. So why not use a Facebook-like system in healthcare?

Bruce Friedman does not think it would take-off:

The reason that there will never be a Medical Facebook, in my opinion, is that most healthcare professionals don’t have the time or inclination to learn yet another application or create personal pages that detail what they are up to in their professional lives.

I’m afraid Bruce Friedman is missing the point of social networking. While kids no doubt spend too much time prettying up their pages and writing poetry in blog entries, the real value of social networking is that it enhances and simplifies the process of communicating and sharing files with a group using the internet. It’s like email2.0 if a cliche helps.

A healthcare-Facebook would be designed with minimal gadgetry so that it would be straightforward to use. I can imagine making such an application a lot simpler than your typical desktop email client. The user could put her contacts together into groups with which she could share files and have ongoing discussions. Each user’s profile need not consist of much more than what is on a business card, with optionally a single picture to help jog one’s memory of who the contact is (we are visual creatures after all).

I do not believe such a system would have to be hard to use or unnecessarily time-consuming. In fact I think the efficiency gained by all would make it an attractive way for healthcare professionals to communicate and indeed it could take off.

While working on the Medical Resident Scheduling Tool I have found that there is no central resource for information on scheduling medical residents. It seemed fitting that H2-O2 would be the ones to create such a resource since we are in the business of making resident scheduling less burdensome.

Resident-Scheduling.com is a directory of links that will be helpful to everybody who has to make medical resident schedules, regardless of how you choose to make them (pencil and paper, Excel, some other software, or H2-O2’s software). The directory is divided into two categories: links to software for use in scheduling residents, and links to useful articles about the topic. There will probably be more categories in the future but it is best to start small.

The directory is human-edited. This means that the links displayed on the site were reviewed by a knowledgeable person (yours truly, as it happens to be) before posting. This ensures that you will only find resources that are interesting and useful.

Lend a hand

In addition to the advantages shown above, a human-edited directory is a lot of work to create and maintain! More links will be added in the coming weeks but your help would be appreciated. There is a Suggest a Link page that I sincerely hope you will use to share with the rest of the community those gems that have helped you along when you had to make schedules.

The time has finally come to launch an H2-O2 blog. As the Lead Developer of H2-O2 there are fascinating things to comment on in three different but related areas that I follow:

1. The growing role of the internet in healthcare.
2. The new internet technologies that revolutionize our daily lives.
3. Starting and running a software business that depends in part on 2) and lives within 1).

I look forward to joining the conversations on these topics with the other bloggers out there.