Web-like applications for the smartphone and other portable devices are a new medium of scientific communication that has arrived. Often called simply “apps,” this medium is an alternative to the familiar Webpage, but with very different design requirements. OSTI has several projects going to explore and develop apps that will provide access to its many resources and collections. It won’t be easy and it won’t be quick but it is happening.
Apps have taken off with the phenomenal growth of smartphones and similar hand held computers. Apple alone has over 200,000 apps for its iPhone; some are free but many are for sale. App development is now a major industry and several federal government agencies, such as NASA, have fielded popular apps.
There are really two very different kinds of apps, although they may look and feel the same. The stand alone app is one that runs on the mobile device, as a piece of native software. Common examples include games, a scheduler, a dictionary or other reference works, photo albums, etc. No outside connection is required to use these apps, although they are typically acquired by downloading from the Web. The content of these apps is typically static, except for user inputs.
Then there is the Web app, which is really a Website designed for the tiny screen of the mobile device. The Web app is viewed using the device’s Web browser, so no special software is required. As with any Website, the Web app requires the user to be on-line when it is used. And as with many Web pages, Web apps can have dynamic content. Common examples are similar to Web uses, such as news, weather or search engines.
Because OSTI’s products are Web based and dynamic, OSTI is probably mostly interested in Web apps. However, there is also a hybrid kind of app, which runs in stand alone mode but which is updated via the Web. Given that OSTI’s collections change slowly and are not especially time sensitive the hybrid app may also be of interest.
The Challenge of Web-to-App Conversion
The real challenge in app development lies in converting OSTI's Webpage content into app content that is fit for the tiny screen. This includes the search results pages, which are very challenging. The generic, abstract challenge is described below. This is preliminary work. Doing a few cases will no doubt refine these procedures greatly, including developing specific procedures for different kinds of content. OSTI has millions of dollars invested in its Web content. Converting even a fraction of this to app size will be a lot of work.
Note that in theory one can view a normal Webpage on a smartphone or other mobile device. It is just a matter of scrolling from side to side and up and down. In fact people do this now but it is laborious and very unsatisfying. Pages designed for 12 or 20 inch screens do not work on 2 or 3 inch screens.
When it comes to converting OSTI’s Web pages into Web app pages there are really two distinct cases, so I will describe a generic method for each. One is for general Web pages like the Home page, the Innovation pages, the Accomplishments product, etc. The other is for search results pages, which have a very different logical structure.
For general Web pages, there are typically specific logical segments of the Web page that can stand alone as mobile device Web pages. Let's say there are 10 such segments and assume that each is the proper size for a smartphone page. In that case one does two things. First make each segment a page in the app. Second create a front end menu page that lists, and links to, all 10 segment pages. The conversion is now ready to go into the app. If a segment does not fit the tiny screen then one may have to rework it into several smaller segments. Of course you can do a lot of smartphone specific tweaking along the way.
As you do this with more Web pages, additional front end or interior menus may be needed to tie them all together. You may also have to give up a lot of interior linking and go with a simple tree structure, because the number of links per segment may be limited by the small space. These limits on linking need to be explored.
For search results, OSTI’s Web pages presently provide a long list of results plus other useful features, like clusters, Wikipedia, news, etc. One can make these extra features into separate segments, where at the front end the user picks a search term plus the feature they want to see. But the basic hit list is still way too big for the tiny screen. However, each hit has a number of features itself, such as title, snippet, authors, MLT button, tag cloud, etc., depending on the product. The solution is to separate these as well.
What should work is to first present just the list of titles, possibly even truncated to fit the screen. Then if the user selects a given title they get the whole bag of features for that title, or as much as can be put on one screen, or another list of choices. The devil is in the details because each product is different. The basic point is that it has to be a highly menu driven system with a lot of little segments. One can also scroll and zoom but that really does not work for blocks of text.
OSTI has a lot of options when it comes to developing apps. Information Bridge might be best, but it might also be the hardest. The interface is powerful but that also makes it complex, a veritable control room of features. News or the Ostiblog might be easiest, but they are not especially representative apps. In any case OSTI will meet the grand challenge of these app content issues, which are different from the technology issues, albeit closely related.
Questions or comments welcome. Onward!
David Wojick
Senior consultant for innovation
OSTI
Seems these days there's an App for just about anything you want to and it's espcially easier to use the App on your mobile device to get the results you want. In fact, I would argue that given the proliferation of Apps for mobile devices that the user isn't so much surfing the web, but targeting specific areas or niches with pinpoint accuracy.
And then there's the SIRI app which uses cloud based speech recognition to answer user questions and it seems to hold unlimited potential in its use. Personally, I see this concept being embraced by other App makers and eventually the keyboard we know will go the way of the eight-track player and other obselete forms of technology.