Google: Programming for the Web is the way to go

Google: Programming for the Web is the way to go

Summary: At the Google I/O Developer's conference today, the message has been clear: programming for the Web is the way to go. Sure, native apps are still important but the future is on the Web.

TOPICS: Google, Browser, Hardware

At the Google I/O Developer's conference today, the message has been clear: programming for the Web is the way to go. Sure, native apps are still important but the future is on the Web.

To drive home that point, consider the popularity - as well as the limitations - of netbooks. The miniature notebook computers have compromised on hardware, such as smaller hard drives, to keep prices low. But that doesn't necessarily mean that performance has to be compromised.

Also see: CNET: Google expects scrutiny, likes Netbooks

Google says the power is in the Web browser, the one app in a netbook where most users will spend the majority of their time. And Google, of course, benefits economically when more users tap into the Web. The more users turn to the Web, the more they'll interact with Google - whether through search or other apps - and advertisers using tools like AdSense to reach that growing audience.

Still, the browser's influence and presence can't be diminished - especially as Web connectivity continues to grow in the mobile space. On a mobile device, the Web has grown - in connectivity, user experience, functionality and more. Why would anyone want native apps when the Web experience is gaining ground?

With that said, we can't ignore the 800-pound gorilla in the room: Apple's closed-wall App store. Those aren't Web-based apps; they're native to the devices. And that seems to run counter to what Google is talking about, right? The short answer: yes and no.

It's true that iPhone apps are native. But Vic Gundrota, Google's VP of Engineering, was quick to point to the brilliant minds at Apple, which were proactive enough to allow Web apps to be saved on the home pages of the iPhone, making them appear to be apps on the device. Gmail, for example, was not written as an iPhone app but rather as a Web app for the iPhone, one that uses HTML 5 to offer an app-like experience to the user. But the shortcut to the Web-based app appears will lead the user directly to that Web page. The bottom line: the two can co-exist, Google says.

The beauty of all of this, of course, is that users gain when companies like Google make new innovative tools available to developers who are encouraged to tap into their imaginations and push the limits of what can be done.

Also see: Google I/O kicks off with bigger push into Android, Web apps and more

Topics: Google, Browser, Hardware

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.


Log in or register to join the discussion
  • Yeah, right

    No self-interest there. Program for the web?

    I would venture an easy guess that the most disliked iPhone 'Apps' are the ones that are essentially web-front ends.

    For some things a web front end makes absolute sense, especially where a lot of dynamic data is involved.

    But there is nothing like sitting on an airplane, and playing a game, reading an ebook, or scrolling through a list of restaurants available at your destination.

    The best apps will always provide strong client-side functionality, without requiring a connection the web.
    • Strong client side

      Sure. You're absolutely right, but as the iPhone system illustrates, the distinctions between a web frontend and native apps are diminishing. These web apps can be designed to be fully functional when disconnected and then synced again when reconnected. Just because the "most disliked" Apps (in your estimation, based on what data?) are web apps doesn't mean that excellent web apps can't be designed to supplant native Apps.

      We're not there yet, but at some point in the not-too-distant future I imagine that your computer's OS and application suite will become less important as standards emerge, led by companies like Google and Mozilla.

      Just because Google stands to benefit immensely from this migration doesn't mean that they're wrong.

      I was just daydreaming earlier about a desktop that would do away with the distinctions between web and native. Perhaps Moblin or Android might be a start.
    • I agree... least for mobile platforms. The trend there is just the opposite. Pull the stuff off the web and make use of the hardware. Native mobile apps tend to be way more useful than web apps. And WebOS actually turns web programming into native programming interestingly.

      Sorry Google but you're backwards on this one.
  • A Web front end...

    with a SQL database on the back end can provide a business nearly everything imaginable. But web programming alone isn't worth much.
  • Web programming \Cloud computing are overrated.

    Don't get me wrong. I know the trend and I know that web-base programs have a lot of benefits over native programs.

    Well, it's just that some apps are just better left offline. Why bother trying to bring them on the web if they won't be any better?
    • Simple

      They give more control to developers and software makers and give more options to users.

      One of the reasons I chose to make my character sprite generator an online web application as opposed to something downloaded and used client side was so that I could provide direct updates with new parts.

      Compared to something like say, Microsoft Office, imagine if the software were web based and you used it online via a web connection. They could provide instant updates to everyone, ensuring that *EVERYONE* got, say, the latest malware patch, effectively hammering virus and malware infections back to the stone age of the Internet.

      Or imagine, again with Microsoft Office, instead of having to pay $400, you only have to pay $5 a month in a subscription fee that you can cancel and start up any time you like, always getting the *NEWEST* version of the software, never having to bother with upgrades and shelling out another $400 bucks every two years.

      Such a form also ensures that *ALL* companies are all using the latest version of the software, and as such program incompatibilities will be done away with and you can rest easy not having to wonder whether your client can read the latest PDF or Word file you sent off to them.

      Further, the software will become cheaper since pirating is effectively obliterated using such a form, no longer will software be downloadable and easily transferred from one system to another, you'll have to use it online with only one account being able to connect to said software at a time. So sharing accounts won't do you any good since only one of you can use the software at a time.

      Also, by the time this revolution comes around to such a point, web connectivity will be completely saturated and there will be no point on the planet in which you can't directly connect to the Internet. In fact connecting to the Internet itself will likely become free and companies will make money by charging access to specific parts and networks on the Internet. So you could say get online for nothing, but if you wanted to use say Google you would need to pay a $1 a month subscription fee.

      It's a good model, yes there are a *LOT* of hurdles to overcome before it can become a truly viable option for most software companies, but I think eventually that's where things are headed. It just makes more sense.
  • Yeah right, w/ almighty AJAX

    What a joke. Try achieve what a rich client could do in AJAX and then tell me it's not a big pile of laughing stock.
    • Why Would You Try Using AJAX?

      To make a web application? I mean, you couldn't. Scripting languages like javascript are just that, scripting languages, they're not object oriented programming languages. If you want to develop web applications you need to use either Java or Flash with ActionScript 3.

      The mp3 player on my CB site is a good example of what a web application looks like:

      It's also one of the first *true* web applications that was ever developed. Being able to minimize, close and move the application around on the screen freely.

      I'm also working on the first and only cookie cutter Flash Webbie board template which uses a Windows style interface:

      When released it'll rival other cookie cutter boards like the ever popular UBB (ultimate bulletin board) setup. Mine will be superior though in that it'll rely on a combination of XML and SQL for board entries and accounts in order to maximize server efficiency and loading times.

      Basically using the Flash application itself to do client side what would normally be done server side via PHP and using entries stored in XML files rather than in a SQL database. The expensive server load for such forms will be effectively done away with and you'll be able to run a highly popular web board even on some cheapo $6 a month web host.

      It's also going to be fully skinable via XML and of course, as it's Flash based it'll be uber cross browser and cross platform compatible. The way it works and looks on the developers computer is *EXACTLY* the way it will look and work on all other systems.

      Most of my stuff is currently constructed using a form I call LAMP FX, which is a Linux, Apache, MySQL, PHP backend with a Flash, XML front end. Or just Flash LAMP for when the XML isn't used.

      Flash truly is the be all and end all of web front end solutions at present. Here's my current list of all the reasons why Flash is
      superior to all other user front end environments:

      1. Support of 32 bit JPEGs (more specifically you can import 32 bit PNG files, use JPEG compression on them and Flash will keep the
      transparency layer intact).

      2. Fully cross operating system compatible with all major operating systems.

      3. Fully cross browser compatible with all major browsers.

      4. Support of the On2 VP6+ codecs.

      5. Support of color matrices.

      6. Support of audio spectrum analysis.

      7. Support of alpha transparent video.

      8. Fully cross browser support of alpha transparent image files (possible with a bunch of extra javascript...even then though there
      are compatibility issues and it's far less efficient).

      9. The ability to track the number of bytes loaded on a particular object.

      10. The ability to dynamically control the loading of media files.

      11. Direct connection to ports with the ability to "talk" directly with a particular protocols language. Very awesome.

      12. Ability to analyze and manipulate image pixel data in a whole variety of ways. (only Java is comparable in that respect as far as front ends)

      13. Ability to stop a media file from being loaded during the loading process.

      14. Ability to control the resizing of elements, being able to turn on and off smoothing for specific elements and making use of bilinear and bicubic resampling. (can be done with a combination of PHP and javascript, but it's less cross browser compatible and uses 10 times the amount of server resources and is about 700% slower, even with server side image caching.)

      15. Fully custom cursors (can be done with SOME versions of IE using CSS, but even then yer limited to cur files and can't use larger

      16. Custom context menus.

      17. Custom scroll bars.

      18. FIFTY TIMES FASTER than javascript when dealing with animated elements.

      19. Custom video/audio controls (can be done ~somewhat~ with javascript but ONLY with certain browsers like IE...and even then with only about half the functionality that you can get with Flash.

      20. Direct support, creation and control over vector based graphics.

      There's TWENTY different things that can only be done with Flash. There's a couple that ~sorta~ can be done without it...but only if you want to make your sites so that they only function on Internet Explorer...and even then you can't get ALL the functionality that
      Flash offers.

      Even for just a PLAIN TEXT site Flash is VASTLY superior to using HTML and CSS; mostly cause you can use embedded fonts, you can have liquid sizable text as opposed to just liquid space stretched text, you get proper font antialiasing in Flash and the overall file size will be MUCH smaller. And really, why bother with a deficient form of page styling like CSS when you can use it's big brother, XML?
  • I think its a good supplement...

    not a all out solution. I was hit by hurricane Ike and with the outages of power and internet, a web only solution would not work well some areas were down for weeks. In a perfect world it would work without any outages of any kind.

    In education there is a issue with Https, we had to disable the protocol on our network because kids were using it to hit proxie sites to get around our web filters. Think what you will about web filters but we are required by law to filter the web to protect our kids. I bring this up because that is how you login to alot of web services. I think its a good supplement but you will still need a offline client, for planes, car rides, no wifi areas, etc... in which case isn't it about what we have now??
    • College Education?

      To be perfectly blunt high schools and elementary schools should *NOT* have Internet access and in fact Internet access ought to be something that's illegal for children. There is *NO* plausible reason for a child to be online, there just isn't. There are no shortage of OFFLINE games, encyclopedias, programs, etc for children to use that do not require going online.

      The bottom line is that content on the Internet CANNOT be controlled, not even by the ones offering it. I know more than a few trolls who think it's ~real~ fun to go on Disney websites and the like and to make little subtle alterations and additions that often go unchecked for weeks and months before they're finally removed. So unless you wanna be faced with questions like, "Mommy, what's a dolcett fetish?" ...yeah, probably not the brightest idea to let yer kids online.

      That, btw, at last check, wasn't something blocked by Net Nanny and other supposed child protective software. Go on Google right now and type the word "dolcett" and you'll get fun sites like this coming up:

      ...btw if you go past the warning page I'm not responsible for your therapy bills, just a lil warning for you.

      And of course, beyond all that, there are no shortage of trolls and invective word artists out there who will not hesitate to rip your child's mind all apart. Most invective word artists like myself do not make any assumptions about the online world and a six year old will be treated no differently than a sixty year old in most online web forms (chat rooms, web boards, etc).

      Most parents worry about "online predators", pedophiles and such going after their kids, when what they ought to worry about is the people like me who will verbally rip your child apart in so many different directions you'll be lucky if they aren't slitting their wrists by the end of the week. Children DO NOT belong online...ever. I don't care what your lame ass excuse is.
  • Same battle cry for the eary 1990s

    I've been in information systems for a long time. In the early 1990s, when the web was coming into its own, the "word" was that everything will be on the web and the operating system, and conventional apps, will soon go the way of the do-do. Oracle's Larry Elison was even saying that his Net PC will make Microsoft redundant.

    Not everyone bought into the hype. But they hype still rears it's stupid head, every now and then.

    The fact is that web-based apps are great for some things. Not so great for others. For example, ask Siebel Systems (now owned by Oracle), Oracle, SAP, PeopleSoft, and others, how making their apps all web-based has changed the world. For starters, application performance today, from the user's perspective, is slower on their 2009 web-based apps (and today's zippy hardware) than it was on their year 2000 traditional client apps. A lot slower. Also, a web-based UI provides a subset of the functionality that traditional desktop apps do.

    Web based apps are great for situations where you need a web-based app, like online shopping or online banking. Conventional desktop apps kick butt when you need that sort of thing.

    Today, as in 2000, "do it on the web" is still a case-by-case question, rather than an across-the-board answer.
    • Well said

      I like to do 3D graphics, and eventually hope to make a living at it.

      I'm trying to picture creating 3D graphics through a browser, on an application existing in the 'cloud'.

      Nope. It's just not coming clear...
      • You Likely Won't

        If you're so creatively stunted that you can't even see the incredible potential of an online based 3D graphics platform then you likely don't have enough to creativity to ever make a living in such a field.

        I mean that field is probably the one that makes the *MOST* sense to have web based. Why? Well, if you knew anything at all about 3D rendering you would know that to do anything worth selling you're going to need some pretty hefty symmetric multiprocessing
        clusters to handle the final rendering.

        My own system is a dual Xeon server workstation based setup with 8 gigs of memory and even it chokes to a halt when trying to render out anything with more than a million polygons using high end settings.

        Having a web based 3D application could give a guy like me access to a few Altix 3000 super clusters for say a monthly subscription fee of $10 a month. It'd be sort of like out sourcing the rendering work for your 3D modeling but being able to do it directly within the program itself.
    • Missing The Point

      Sure it's an old dream, but it's one that is slowly becoming a reality. Back in the early to mid 90s much of what was being imagined simply wasn't possible, but in this day in age, it is possible. That's why right now is such an exciting time to be a web developer.

      Flash with ActionScript 3 was pretty much the starting gate and now that it's out there...yeah, it's gonna be another Internet gold least for those who know how to code.

      And as far as web forms eventually replacing whole operating systems, well, some of us are actually working on such things:

      Neat, huh? It'll be fully skinable too when it's finished.
  • I own my own data

    Give me a beefy desktop. Hard-disk space is cheap. I will manange my own data. Not interested in putting my stuff in the "cloud" where who knows who can see my data. We have already had security breeches and long downtimes. No thanks !
    • No Offense

      But your data would probably be better protected online than on your computer. Security has more to do with the individual though than where the data is being stored.

      They make these fun devices now, not sure if you're aware of it, they're usually USB based and plug in between the computer and keyboard.

      What do they do?

      Record every single keystroke made on the keyboard...completely undetectable by the computer system itself, easily put on and taken off in seconds, almost entirely unnoticeable.

      Heh...the fact is, there really isn't such a thing as "secure data"...ever. Data security is an illusion at best and your data is only as secure as the want to try and steal it. If someone wants your data...they have it. It's as simple as that. There's nothing to stop someone determined to get your data. The recent leak of the work print copy of the new X-Men movie was a good example of how that works.

      If you don't want someone to get your data...don't put it on a computer. LOL
  • Web Apps??? Web browsers??? What???

    As a developer who creates both, I can honestly say that when IE8 came out, I lost a lot of faith in the web app... a lot of sites that I had once used without problem, failed to work, and some I had developed had to be changed in the way they used the browser. Until they find a platform that can be used for browsers that doesn't change dramatically, then web apps will be proned to problems depending on what version of a web browser a user has. At least with desktop development, you can design a system to run on a particular OS and have a pretty good idea of the end result. New OS's come once every couple of years, but the web is still changing. Yes, some day I will agree with Google that web apps will take over, but at this stage when deadlines and dollars dictate a project, then web apps will still be harder to write and maintain while things are still changing!
    • There Already Is

      It's called "Flash"...look into's gonna be *BIG*!

    • Not with MSIE8?

      Suggestion: If Microsoft applications fail try an Open Source equivalent or better...
      • You're Obviously Not A Developer

        Or you just think you're a developer.

        We don't live in a perfect world where everyone uses the same browser and despite your hard on for Microsoft *NO* browser maker on the planet currently adheres to all of the *RECOMMENDED* standards put forth by the W3C. And in fact, most sites will work on IE before they'll work on other browsers simply because IE has the largest market.

        That said, the reason why many sites break on different and newer browsers is, simply put, because of BAD DEVELOPERS who didn't bother to make sure their stuff was cross browser and cross operating system compatible. They're often people like you, who are so incredibly narrow minded that they can't think outside of a single browser and OS combination, so their sites effectively fail on everything except their own setup.

        And as pig headed as those poser class developers like yourself are, you magically expect millions of Internet users to just switch over to a different browser just because *YOU* think it's "better", and often only for the most simplistic, asinine, backwards and inconsequential reasons imaginable...usually having nothing more to do with anything aside from a retardation into the old corporate vs little guy mentality, completely ignorant on all levels to any *ACTUAL* facts pertaining to browsers or, more importantly, THE CODE.

        Piss poor coding is likely to break on any number of browsers since piss poor coding will cause a browser to retard back into quirks mode, which is gonna work differently on pretty much every other browser out there.

        Those with weak coding knowledge are far more likely to produce bad code and there are a *LOT* of poser class flunkies out there running around trying to pass themselves off as web developers, often with so little coding knowledge that their entire site is just piecemealed together via spaghetti bitching programing practices; taking various segments of code out of other peoples sites and just copying and pasting them on over into yours without any idea at all as to how they even work or why.

        And then on the opposite end of the fail spectrum you have the idiot developers who think that following the W3C's RECOMMENDED "standards" religiously like it's some kind of a freakin cult is the only way you should code, not comprehending on any level that W3C compatibility effectively means your site BREAKS on, well, basically EVERYTHING, since, again, *NO* browser maker currently adheres to *ALL* of the RECOMMENDED "standards".

        If you're going to be a successful web developer you have to keep *ALL* major platforms and *ALL* major browsers in mind. As well as *ALL* major screen resolutions for that matter and a variety of other factors too.

        Currently the best front end available is Flash as far as cross browser and cross OS compatibility, but then Flash has some pretty steep down sides, in that if you don't know how to use Flash, if you don't ~actually~ know how to code in ActionScript...yeah, yer just gonna make a great big fracking mess with Flash and you'd be better off sticking with something more your speed like HTML. Flash is definitely *NOT* for the amateur or poser class developer.