The Web-based Office will have its day

The Web-based Office will have its day

Summary: I first profiled a Web 2.0 office in early September and since then more web-based office products have surfaced.

TOPICS: Microsoft

I first profiled a Web 2.0 office in early September and since then more web-based office products have surfaced. Peter Rip posted recently that he's now "bumped into an alpha or beta Web-incarnation for every Microsoft desktop product". He says most are AJAX, but some are Flash or Flex-based (both Macromedia products).

Peter thinks desktop apps are on the way out, because "no one works at their desktop anymore." While he rightly points out that office people are spending more and more time on the Net, I don't agree with his conclusion that this necessarily signals the end of desktop apps - yet. A recent Computerworld article states that laptops will soon replace the desktop as the preferred business PC, citing mobility as the main reason for this trend. Microsoft office products - and increasingly OpenOffice and other open source desktop programs - will continue to be used on both laptop and desktop PCs for the forseeable future. The reason is the advanced and wide ranging functionality they offer, together with perceived stability in an office environment.

However I do think that long-term, the writing is on the wall for desktop office applications. Once the current crop of alpha and beta web-based office products reach a level of maturity, they will be ready to challenge Microsoft for the minds and pockets of consumers. One of the keys is achieving the level of functionality that Microsoft Office undeniably has. But there are also issues of online security and reliability that web-based apps will need to address, in time. Office apps are just too important to corporate productivity for CIOs and IT managers to entrust their businesses with web-based apps, without complete confidence in their functionality (ability to do the job efficiently) and performance (security and uptime).

The time for the web-based office will come, mark my words. When broadband is ubiquitous, web functionality is richer, issues of security and reliability have been put to rest, and most importantly of all - when Corporates are ready to make the jump. It may be 5-10 years down the track, it may be longer. For now, let's take a look at the current crop of promising web-based office apps and hope a few of them last the distance.

Ones I mentioned in my initial post:
Writely - "The Web Word Processor" (note that for creating documents, it uses an HTML editor and then converts to Word format)
FCKeditor is also an MS Word-like web app. It's open source too.
gOFFICE - "a browser-based online word processor and desktop publishing program"
Num Sum - web-based spreadsheets - except only the author of a spreadsheet can edit it.
Kiko - Online calendar solution powered by Ajax.
Gmail and now the new Yahoo! Mail (Microsoft is rumored to be working on a Hotmail upgrade, codenamed Kahuna)
called S5 - web-based Powerpoint  
Webnote - web-based version of Microsoft's OneNote 
thinkfree - online Office suite
Openomy - online file-system

And ones Peter has added to the mix:

Feel free to add more web-based office apps in the comments below. Better yet, tell me if you agree with my prediction that web-based office apps will eventually displace desktop office apps.

Topic: Microsoft

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
  • Web Office

    I do not think that desktop apps will completely disappear but I sure would like to see the Web-Office succeed. As a software developer I see this as a good thing since some of my application would work best if I hosted them. The main reason being the absolute nightmare of users installing software on their own machines. Some do not have rights to do this and are upset when the software will not install or they forget where they installed the software and install it again in another place after which they do not remember which is the correct version and get data in both.

    I have an automatic way to get updates over the Internet but some companies do not allow employees to update software this way so they have to wait for the IT person who maybe too busy.

    So they are always calling my tech. support for issues unrelated with the product. I would prefer in these cases to make the application a Web Application and host their data and everything.

    I have been asking the browser manufacturers to get on board and offer a much richer UI but they have not seen fit to do so.

    So the sooner this becomes main stream the better. There are many small businesses without tech. support that could benefit from paying a subscription and having their office software hosted by some one else. They do not need the advanced feature of most desktop software so it should be possible to build slimmed down software that works well for this type of use.
    • From what I've seen the solutions are still kind of rough

      I haven't looked into ready-made hosted applications, only web controls, at least one of which were mentioned in this article.

      One of the challenges I noticed from introducing a web-based rich HTML editing control into an application I was building was in managing the different elements of the document. For example, with the editor I put in, if the user uploaded a graphic to include in the document, since the editor rendered HTML just like a browser would, the graphic was not included physically in the document, but rather just linked to it. Incidentally, MS Office used to do this as well, though it offered the capability to embed the graphic in the document. In this case though the graphic had to reside separate from the HTML document on the server somewhere.

      What I've heard some places have done is they've managed this by storing this information in a database, as opposed to files on a server. This way the HTML document could be stored in a field of a database table, and the graphics could be stored as BLOBs in a different table, and associated with the document via. a foreign key. When the document gets rendered, the IMG tags reference a procedure on the server, passing in an image ID (created when the graphic was uploaded), which gets the image and returns it as a binary stream to the client, which gets rendered.

      I think this is the preferred method, because otherwise you end up having to save everything as files, and when you try to manage it, it's difficult to tell which images go with which file (since, to keep image files from colliding you've assigned them all unique IDs for filenames). One of the problems I noticed was that if a user uploaded an image, and then deleted it out of the document, only the link was removed. The image stayed on the server...

      So with the technology right now, it gets complicated.
      Mark Miller
  • Not on my Network

    I wouldn't want it. I'll stick with Open Office, its free, and easy to use.
  • Did you try using any of these ?

    Other than the Word Processing apps, all the others are slow as a dog. Especially the spreadsheet. Can you image if this spreadsheet is of any real size what performance will it have ?
  • Useless!

    Until I have an internet connection that never goes down I'll never use a product like this.
    • Why is that the standard?

      My Internet connection has gone down. It hasn't happened recently. Over the past year, it may have happened for a few moments once or twice, but it was never a matter of having to wait a significant amount of time. The technology is there. I don't remember losing productivity because my Internet connection in my office was down.

      A concern with the current way of doing things is that computers crash. Data gets lost. Software needs to be reinstalled. It may be due to a crash, an incompatibility, corruption, a new release, or a bug fix. We need not be subjective here. The effects of all of these things can be measured. It's ultimately the overall loss in productivity that will be the issue.

      If the advantage of a web based application is that the latest version is available instantly, any bug fixes are part of your system right away, your application software cannot get corrupted, data backup and recovery can be built into it in a way that cannot be done on a PC alone, and you save the time of having to work as a systems programmer to maintain your installed products, then all of those factors have to be weighed too.

      There's also the issue of perception. People used to talk about how if a mainframe was down for a few minutes, it cost millions of dollars in lost productivity. So they switched to Windows, which typically crashed daily for many users. If you add up the time lost by users, it went up. But it was spread out, so nobody noticed. But people argued, with a straight face, that only one person was affected if Windows went down, not the whole company. Obviously if the total productivity loss for the user and for the company was higher overall, the argument was pure nonsense.

      If people wanted to be honest, they would have admitted that PCs were more expensive than mainframes when everything was factored into it. But perception is what people want it to be. Now we are talking about another paradigm shift. People will look at the facts, pick out the ones that suit them, and make an argument for their method of choice. But that's not honest. These things can be studied, and all the factors can be weighed.

      And by the way, if you think that until you have an Internet connection that never goes down, the product would be useless, you would have to say the same thing about this forum or anything else you use on the Internet. If you were correct that these things are useless, then why are you here?
    • Worse than useless!!

      Until I have a computer that will never go down I refuse to even try using software at all. I currently only use my machine as a space heater and it works just fine for that and gives people the impression that I am a modern intelligent person. Fooled them didn't I?
      Still Lynn
  • Interesting ideas but desktop apps will always be better

    Web-based applications absolutely have their place as proven by the popularity of Hotmail (something like 300 million users), GMAIL and others. But relying solely on Web-based applications is just plain dumb for numerous reaons. I'll list a few that come to mind:

    1. Internet connections are not always reliable or even possible. While connections will become more and more common, they'll never be 100% ubiquitous. Some airlines have net access but will all of them have it someday? I doubt it. What about when you're out in the proverbial boonies where there's no access at all? What about the impossible to prevent outages? Do you just wait until the connection is back up to start doing stuff again?

    2. Relying solely on Web based applications is an incredible waste of processing power. There are literally billions of processors in "edge" devices that can do amazing things. Why use them only as "dumb terminals" for Web-based applications? If you can buy a high powered PC for about $300 with a speedy Intel or AMD processor, why would you not use it?

    3. Web based applications will always have poorer performance for several reasons. Even with super high speed connnections there will always be latency when the user on the device has to send a request to a server that could be half way around the world and then server has to send the response back.

    4. Even with some of the interesting technologies like Ajax, Web-based applications will never have the richness and interactivity that full blown PC or phone/pda based applications can have. Think of the coolest Web applicatoins you've ever seen and compare it to what you experience when using an application like Word or Excel or Quicken or any number of other client applications. There's just no comparison. Can you imagine a Web-based version of Photoshop?

    5. No matter how trustworthy the service provider, users (whether in companies or individuals at home) will always have data that they don't want to be stored on the Web. Even if web application providers can find clever ways to allow users to keep the data on their PC's or other device but have the Web-based application do the processing, users will be suspicious.

    6. Finally, the PC is as popular as it is because it's an empowering device. You cna customize it they you want. You can control your data. You can do almost anything you want with a PC. Play games, Write a book. Edit photos and home movies. Manage your finances. Communicate with your friends and family. Relying on Web based "service providers" to provide all of that is a step backwards for many people.
    • Let's look at that.

      Let's look at your points:

      1. That's correct, but will not be an issue for most offices. For places where people work at their desks, and can dial in from home, or go directly with DSL, it won't matter. But there will be a need for a desktop version for those who need it. As long as they are compatible, it's not a problem. Chances are that the laptop version would fall behind since upgrades would not be instant, but it would still be workable. It used to be that people had versions of WordPerfect for their PC, their Vax, and their IBM mainframe. The products were compatible, even if the interfaces differed, but they both did the same thing to the documents. Users could switch off if they were in a place where one was not available. A person who uses the web version at work can still use a PC version on a laptop as easily as he could now.

      2. Using a PC is an incredible waste of processing power. As I type this, 99% of the potential power of my PC is being wasted. So is most of the memory, since most of what's there is not being accessed at the time. Since I have many computers, I need to have many times the processing power that is needed, and lots of redundancy. I have duplicate copies of software on different PCs, which wastes disk space. I have the same applications loaded into memory in different computers, which wastes memory.

      In the old days, a systems programmer would install software on a mainframe. He/she would pick a specific region of memory, and it could be shared among users. It would hold the portions of code that were static, and the areas that changed based on user activity would be in the user's virtual memory. Since things only needed to be in memory while they were being used, the information could be swapped out very efficiently. The OS was also good at predicting what was likely to be used next and getting it back into main memory before it was needed, so the user never saw the difference on a properly administered machine. Furthermore, if several users were using the same application, it actually increased the chances that the reentrant parts of the code were already in memory since all the users were running the same code at the same memory addresses. That's not the case with PCs today if things get swapped out.

      If you don't want to waste resources, then this new idea seems like the ideal solution. There would no longer be the need to buy and maintain all that hardware that you mentioned, which is already being wasted most of the time. If you can buy a high powered PC for about $300, and most of it is currently wasted, you'd be able to buy a much cheaper PC instead and waste nothing.

      3. Always is a long time. Yesterday I saw a presentation on a technology that is literally 1000 times faster than the current Internet, and could replace it in the next 5 to 10 years. Latency is not a problem for me now. I have a dual processor Pentium 4 3.2 gHz machine. I don't have problems waiting for things on most Internet sites. But if I bring up Word, it takes far more time for things to load into memory from my hard disk. If I need to have Excel do a complex calculation, it takes time. The amount of data that goes from the user interface to the application in these cases is minimal. I have far more things on my PC that take time to load or slow down when something is running then I have problems with well maintained websites such as my bank or large eCommerce sites.

      4. I can imagine a Web based application of Photoshop. It would not run remotely, just as Java applets don't run remotely. But it would not need to be installed on my local machine, I would not need to install patches or upgrades when I need new capability, and I would always have the current version. As long as the bandwidth is there, the limiting factor would be how fast my computer could load whatever became the equivalent of the applet. But since the same thing is a factor with local applications, it's a moot point. That is, I'm currently limited by how fast my computer can load something, and that would still be the case.

      5. I have never had data on my PC that was as secure as what I used to have in the mainframe days. A remote host can be configured with technology that precludes viewing or touching the data without a key that only the local user has. Plus, it's backed up instantly, can be mirrored, stored on RAID5 drives to begin with, and copies of the encrypted data can be sent instantly and automatically to off-site locations in case of a catastrophe. You can't store data "on the web." The web is just a transport mechanism. You store it on a trusted computer, and you use TCP/IP as a way of linking up with that computer, instead of needing a dedicated line to connect your private network. None of this is truly "on the Internet." It's just that the Internet provides a way of getting data over a network from your computer to another. As long as that channel is secure, and HTTPS does a good job, then it's less of a risk than somebody tapping your phone line with a different network.

      6. That was the point that people made when the PC first came out. MIS groups were slow to act, arrogant, and they thought they knew better than their users. A PC provided freedom. But the reality became that in a corporate environment, the MIS group got into the PC business. They dictated corporate standards for what software you could use or even install. They took care of connectivity, so you were still stuck relying on them. As we moved from DOS to Windows NT or above, the OS took over enough of your machine that you no longer had direct control over it, or over what portions of real memory you could play with at any time. A lot of the things you are talking about may work remotely, but I'd agree we are a long time away from editing movies on a remote machine. But for business use, those issues are not there. Communicating with my friends and family was actually easier back in the mainframe days. We had email long before most people ever heard of SMTP. My dad had an account on my mainframe, and could communicate with me from the other side of the country, back when a phone call was 35 cents a minute on nights and weekends. The issue was not the technology but how it was used. I had a class A account on the mainframe and could do whatever I needed to do. I could change my priority, force off virtual machines that were not responding, install whatever software I wanted, or write my own. I could do many things that a PC was simply not capable of. When people told me about how their PC could manage a database and even support several thousand records, I just laughed. How much any user is restricted in any environment or on any platform is a corporate decision, not one dictated by the technology. Try installing a game on a PC at your local library and see how far you get.

      It's not the platform or the location that matters, but the capability it provides. When I wanted Internet email, everybody else in the company ended up having it too. And they didn't even need to change applications or install anything. They just typed in a different type of email address. When I wanted to support MIME attachments, I wrote my own code. MIME was nothing more than a proposal back then, but some vendors such as Netscape jumped on it. So did I. And again, every single user on my mainframe got to send documents, photos, or anything else with a few mouse clicks. (Yes, mainframes had mice decades ago, even though most users didn't know they existed. They also had color terminals and graphics that beat the PC in terms of resolution.) So the amount of power has little to do with the paradigm but with what's offered. I can play far more games on the Internet simply because I don't own many on my PC.

      If you are paying somebody else for these applications, and you need not buy all of them from the same vendor, then it's up to them to meet your needs and give you the power you need. Otherwise, you will go elsewhere. I've never used Quicken on-line but lots of people do use TurboTax that way, and would not want to bother installing it on their PC. With this new paradigm, everything will be available instantly. Once you sign up for it and pay, you have it. There's nothing to install.
  • SimDesk

    Another web based word processor.
  • I did say this was the future...

    The responses I've read all confirm why I don't the web-based office is coming any time soon. It will be at least 5-10 years away, precisely because of the broadband, reliability, functionality etc concerns people have voiced here.

    But imagine a time when all those network concerns we worry about today have been solved - 100% reliable and fast broadband, web-based functionality that puts AJAX to shame, complete customization control. Given all that (and admitedly it is a dream right now) is there any reason to restrict an app to a single computer? What if there were disposable PCs in the future, that you can buy to read on the train just like you buy a newspaper now. OK it's getting sci-fi now, but I am thinking long term here.
    • Predictions of the future are often wrong

      Remember the films froms the 40s and 50s predicting what the year 2000 would be like? Flying cars, square trees, no pollution, no hunger, little or no disease, robot servents ...

      Of course there will be some application work on the web, but there are too many social issues why people will want to keep work close to them. Furthermore, history shows that technology doesn't always improve as fast as we would like: 100% reliable high speed broad-band is a likely example. Additionally, some technology may come along and make the whole issue moot.
  • Wireless network reliability is key

    Maybe I'm biased, but I've been using wireless networking for a while on my laptop, and I've found its reliability is only so-so. I've run into the irritating situation where I've tried to post here, or somewhere else, and unbeknownst to me my wireless connection cut out for a few minutes. I hit "submit" and then my screen blanks out and I get the "can't find website error". Now, maybe my wireless network isn't configured properly. That's possible, but if I'm running into it, surely others are too. Imagine the horror that would set in for an average user if this situation happened while using an online office app: "ALL MY WORK IS GONE!!!!!" Think of it this way. A network outage in this situation is equivalent to the computer crashing or the power going out in the midst of your work.

    I think for an online office app. to really gain trust it's going to need to either do frequent online saves, so that recovery is possible in case of a network outage, or be capable of saving locally as well as online.
    Mark Miller
    • I agree

      I've been saying this and looking for a way to do this for at least 2 years and Ajax if finally the solution. As you say, wireless is Key. With WiMax, however, days of intermittent connections may be out the door.
  • I've worked with FCKeditor

    Well, for a web app. project I wrote several months ago. I tested it out a bit, and it was impressive. I could even copy-and-paste a Word document into it. That's a feature that a lot of rich HTML editors have. And I could paste (upload) and format graphics.

    I wouldn't say that the experience was equivalent to Word, nor were any of the other web-based HTML editor controls I tried. It was a little rough around the edges. The thing is, FCKeditor isn't really an online app. in and of itself. It's designed to be a web control, a component that can be included as part of a larger application. There are several commercial web controls that do the same things it does. I forget whether FCKeditor supported printing, but I'm sure some of the commercial controls do. It takes some development work to get it into a web app. where it can be accessed by users.

    I was using it in an ASP.Net app., so maybe others who have used other versions of it (like for JSP or whatnot) will have seen something different. I was poking around in the javascript code to see if I could fix a tiny annoyance it was causing and noticed that the editor logic was using the the COM HTML rendering component in Windows to render the formatting and fonts (possibly the graphics too), and the rich features basically only worked in IE. Likewise I think a lot of the web-based HTML editing controls do the same thing. If a different browser was being used with it, it would fall back to using the generic textbox control like what we use to post here.

    So even though these solutions are web-based, they're not all cross-browser compatible.
    Mark Miller
    • cross browser compatibility

      There are plenty of WYSIWYG text editors out there that are cross-browser compatible (and some are pretty cool). I don't think the cross-browser thing is going to be a real issue.
  • Unplugged Applications

    I think the issue is that with the enabling technology of AJAX we are able to bring web applications to the desktop and take desktop applications to the web. We need to work offline with Gmail and online with an MS Access based inventory system. We have used AJAX enabling technology and Morfik JST ( which is a design methodology to create a desktop Gmail and web-enabled MS Access Northwind. We will be showing these at the Web 2.0 Conference next week to break away from the concept of either/or perception of the world.
  • Convea!

    I really feel for these guys, completely ignored. They have done more than write a web- they have the whole WebOS thing going on there, 40 web applications in a multi-user, tabbed framework. Check it out and be amazed.

    This thing could do more than provide a little competition to windows it could become the standard for deploying integrated web applications.
  • SimpleData - AJAX Billing and Inventory Software

    I work for a company that develops and maintains an AJAX invoicing, and inventory control application for the web called SimpleData. You can learn more here:

    They also have an AJAX business framework for business database apps kind of like Access, called Sysbotz Enterprise available here:
  • Yeah, right

    That's just what I want to do, spend hours writing something in a web app only to have it crash or lose it when I click on a link in an email.

    >>Webnote - web-based version of Microsoft's OneNote

    Yep. Just like OneNote, except without the tabs, the categories, the handwriting recognition, the text-to-speech, the search function, the audio recording function, the toolbars, the UI, and everything else in one note.

    You might have gotten away with comparing it to notepad, but come'on, OneNote?