The main reason why one should never use frames� is the high development cost. This means you could use frames, if, and only if your budget is� around € 100.000,- and you have a great knowledge about using frames, because you will need both of these to develop and maintain your web application or website.

For each and everyone who still thinks they can or maybe even should use frames for some reason, I want to make clear that almost� anything you would use frames for can be� achieved without the use of� frames. Using inline frames is always good, since these are not accounted for as being a real frame within a frameset, but that is a different discussion.

Now I hear some people wonder already why frames should never be used. Okay, everybody feel free to add more problems with frames to this list, but I will start showing the most obvious problems.

1. Browser incompatibility
   some browsers simply cannot handle frames. Older browsers cannot although this is not really important anymore since they are barely used anymore. But a lot of browsers on mobile devices also cannot display frames, which is a very important problem to think of in the very near future. Also browsers for visually impaired people most of the time cannot handle frames.
2. Search engine incompatibility
   most search engines simply do not index frame sites because they cannot be read. When at some point they magically do find a frame page, they tend to not display the frameset correctly when a users clicks on this page. Usually only one page of the complete frameset is displayed, leaving users without any form of navigation (since this is usually another frame), sometimes even generating javascript errors, and� the page� style usually being messed up.
3. Titles not visible
   A big� issue in web usability is to display a different title for every page a visitor sees. This to improve the awareness of the user of their current location. Guess what? Not possible with frames… Only the title of the frames page is displayed at all times.
4. Many more files
   to really use frames one should create for example 4 files to display one page to a user. The frameset will be loaded who will load another 3 pages making you develop 4 complete pages, and separate unnecessarily them into many files on the server. This makes the file management a lot more complicated, especially when you have 10 different framesets all loading content in different frames. Can you image all the folders and files you need to create!?
5. Bookmarking errors
   some browsers incorrectly bookmark frame pages. This means they bookmark the frameset, when in fact the information people want to save is in one of the sub frames. When visitors then try to find back the page they bookmarked, they will get the standard frameset page again. This also applies to sending someone a link, through ICQ or Messenger for example. A user simply copies the URL from the address bar and gives it to a friend, who will see a completely different page when he or she opens the (frameset) link.
6. New window problem
   many users will open a new page in a new window, not even to speak of the current hype of using tabbed browsing. When a user opens a new page in a new window or tab (using ALT-click or CTRL-click) only this one frame will be visible, leaving all other important frames out of focus. In this case the navigation and for example the page� header will not be visible anymore, one of the� no go’s� when discussing web usability.
7. Scrolling
   when using big frames with fixed sizes, or using many frames, there will be a lot of scrollbars visible which are not intended for use. For example when a window is pretty small, a visitor sees� one horizontal and one vertical scrollbar, but when using frames, a visitor could see 2 scrollbars for every frame,� which results in the mess-up of almost every page. Next to that, try clicking on a menu and then scrolling another frame. To do this, a visitor should first move their mouse to the other frame, and sometimes even click it to make it able to scroll again…
8. Javascript
   when we get to scripting a little bit, we see that javascript has a good interpretation of frames. The frame names can be used to execute a command in a certain frame, or even in multiple frames at once. But� it� would be even more easy� to� not use frames, but just one page? In that case the javascript code can be programmed more simple and much faster!
9. No javascript
   and what about people having javascript disabled? With frames, most of the time javascript is absolutely necessary, for example� to� reload 2 pages with one click on a link. Without the use of frames, the need for javascript is mostly optional, instead of required.
10. Printing
    This subject is so obvious, I think you should try it yourself. Print a frameless webpage, and try to print a frameset. You are bound to get problems, just like a lot of your visitors who do still print certain pages to refer to later.
11. Slow
    slow you say? But I thought frames where supposed to improve speed on webpages? Yes absolutely true, but this concept also has a downside. The fact that only one frame (a part of the page) needs to be loaded every time� is a good advantage for visitors with slow connections, because the data transfer is the bottleneck, not the number of server connections. With high speed internet these days, the number of server connections is more important in defining the speed of a webpage. When using frames, a visitor enters a webpage, for example loading� 4 pages (frameset page + 3 frames; top, menu and content). This means that there� need to� be 4 connections to� the webserver, compared to only one connection when not using frames. The data loaded is almost the same, but the number of connections also has a great hit on general performance. Now when surfing through a frames webpage, every time only one frame needs to be reloaded, just like any non-frame webpage. This means that when browsing through 20 pages the� difference in� connections is almost nothing, 20 connections for a regular page, and 23 connections for a frames page.� A difference of only 15%. But wait… the average user nowadays only views 3 pages on every webpage! Yes, now I hear you think… 7 connections against 3… wow a difference in connections, and therefore speed, of over 50%! Oh, and I haven’t even talked about the browser interpretation speed of frames.
12. Session problems
    in the more technical programming of frames we see a very annoying subject rise, namely the use of sessions within frames. Each server request for a file is done in a separate frame, and therefore each frame will receive their own session ID. This means that a very much used programming technique cannot be used anymore. This means frames are pushing the programmer to use different programming techniques, wow, that is what frames can do!� There are of course ways to work around this, but as usual they will all cost only more and more time to develop and to maintain.
13. Legal Issues
    if you would use a link in your website that loads a different webpage, the visitor might not be aware that he or she is viewing a different webpage. This might be interpreted wrong by the user which means the webpage is misleading. You might have serious legal issues with this.� In a case in Scotland in 1996 called the “Shetland Times Case” a judge agreed that the Times had presented at least a prima facie case of copyright infringement based upon� a United Kingdom’s law.

Well, as you can see, frames� have a big impact on your budget. If they are used, they will make your life miserable if you don’t implement them thinking of all these subjects, and maybe more. Again, when you have a big budget or a very specific reason to use frames,� I’d say� give it a try, but also take a look at Server Side Includes, AJAX and IFRAMES before you even think about using frames, they might solve your problem more easy and efficiently.

More articles about using frames in webpages:

• http://www.yourhtmlsource.com/frames/goodorbad.html
• http://www.useit.com/alertbox/9612.html
• http://itc.virginia.edu/desktop/web/frames_problems.html
• http://www.google.com/support/webmasters/bin/answer.py?answer=34445&topic=8522
• http://www.searchenginesbook.com/frames.html

Accessibility
All� information� managed by� the application has to be easy accessible. For instance, the user has to do some actions to get� the desired information. But also be able to find his or her own way through the application without reading a lot, or losing sight of� the target.

Information organisation
The application has to be prepared to help the user to prevent having similar information on different locations, so prevent having duplicates. This only makes users search more, and information being split onto more pages than is essentailly needed.

Customization
Take into account a possibility for every user to customize his or her profile. Using this profile to show information that is interesting for� each specific� user, helps in keeping a clear overview for� every user. Also this is a good way to� make the user find the application handy to use. You prevent the fact that the user is bothered with unnecessary elements.

Information linking
Create the possibility to see linked information all at one time (or just one click away). For instance:� I am making an appointment with Mr. Johnson at 14.00, and� I� cannot remember where he works. I should be able to find or see his address information at once. This would improve� user performance a lot and also make customers feel more satisfied.

These elements taken together will make� any user more satisfied. And eventually this is� a very� important element. If the user is satisfied, you will also be satisfied, and have the ability to� make even more users happy using your web application.

Off course there are a lot of different aspects to improve, but using these as a start will get you thinking of more. And to keep on improving at all times is� most important.

Page design

To let users know what task they are currently performing in a web application is probably one of the most important aspects when designing a page layout. This can be done by placing the title or subject of the current page in a very consistent location, and make sure the title or subject is very well readable. The next step is to make sure where to go, and what actions to do next to complete one task. A clear and consistent navigation is therefore absolutely necessary in page design. External links are absolutely forbidden, because this only distracts users from their current task and makes the web application look and feel more like a regular webpage. Also for regular web pages, content is written and built for indexing, in order to make sure a webpage can easily and accurately be found in search engines. This of course is not applicable for web applications, because they should never be found in search engines. Cross browser compatibility is another factor of page design which is very important in regular web pages, but not in web applications because web applications are specifically made for a few users, who can all be taught to use one browser. Of course when the user group of a web application expands to many thousands of users this topic will become more relevant to web applications.

Different pages of a web application need to be very light. Light in the form of being small in byte size and using only few images, so the application can be loaded very fast every time. This is especially important for the first page. The first page of a web application is a very important one, but whenever designing a web application it is a major issue what to exactly put on the main page. The main page can contain many different elements but is suppose to show the general statistics of an application which do not take too long to load.

�

Sugar CRM Mainpage

This is because users always have to go through this page before getting somewhere to perform the tasks they want. When a new user enters the system it is best to start a guided tour through the application or some sort of help function to get a user started. With these statistics a user can be guided to performing a relevant task which again leads us into the next subject of this article.

Task specific navigation

All information has to be ordered thoughtfully without any distractions to make sure someone only focuses on the current task. Tasks might also have a number of sub tasks available which need to be presented to the user during or after the main task has been completed. The most common tasks need to be placed on top. Whenever needed all tasks can also be shown in order of execution, to make it even more simple for users. During each and every step of these tasks it is also important to guide a user through this step. Sometimes this may seem unnecessary but in most cases user assistance is very important and can help users complete their task faster and can decrease communication with the supplier of the web application. To make sure all tasks will be completed and there will be no data corruption in any task, browser navigation elements should not be used at all. Users can not use these navigation items, because they are not made for use in a web application and they may return unknown or incorrect pages. Best is to hide these elements or teach the user not to use them.

Form layout

In web applications there is no search for information like on a regular website. The need for completing tasks by inserting data is probably most important and in order to do this the web application needs to have a very good form layout. In a good form layout users will run through the data insert form much quicker and find it easier to insert data. The use of scrolling is to be kept to a minimum, to make sure a user can oversee what is already filled in and what is left to be completed. Therefore another technique must be used to display only the needed elements. A wizard is a good example of data entry without lengthy forms. A step-by-step question and answer technique will work very good for most users, but can be annoying sometimes when internet connection fails or is slow. Another solution for this problem can be the display of elements organized in groups. This very much used technique works good for web applications even with a slow internet connection.

When we look at the form itself and the requirement to make sure a user can run through the form quickly and adequately we see that when having 2 columns the form is much shorter, but also much slower. Having 2 columns means showing the label of a field on the left and the input field itself on the right. Can you imagine the moving eyes of the user! From left to right and back, and so forth… Instead when using a top to bottom form with every field label just right above the input field itself users can move from top to bottom very fast because there is no need to look anywhere else. A disadvantage of this is that labels and fields are not easy to differentiate. This can be solved by using a different visual weight for important elements in the form. This means making the labels of all fields look bolder, but also making sure the primary action in a form attracts more attention than other actions. This is very user friendly because in this way the web application itself guides the user through the steps of a task.

�

Example of good form layout

Error messages

Errors must be shown with a very user friendly message, and possibly some form of advanced information. This advanced information may never distract users from the friendly message in any way. To continue helping the user complete the tasks, a solution for every problem must also be provided. When receiving an error stating the login information is incorrect, the application also has to provide ways of making sure the user typed the username and password correctly. This can be done for example by checking the status of the caps lock or num lock buttons. Also after these steps the user needs to be provided with an option to reset the password or have it sent to an e-mail address.

When inserting faulty data in a form, new information about this form has to be provided on the same page. This means that when a form field is filled in incorrectly, the same form will be shown with the inclusion of the error message about this field. This is a much better approach than using a different page to display an error message and then send the user back to the previous page without the error message.

The problem is eventually that something that is developed by different users can be sold by someone. Offcourse this is a problem for everyone that one person earns money on something more users have developed. Therefor open source can be protected by the GNU General Public License (GPL). This license can be used to ensure that all users of the open source will be able to use, alter and pass it through for free. GPL is also protecting the fact that open source software can not be sold or changed into a closed source software. The only benifit that is allowed using open source software when using the GPL is the service of this software.

GNU General Public License
During our research we have searched for the advantages and disadvantages of open source protected by the GPL. In this way we can compare the advantages and disadvantages of open source software and commercial software for companies using the two different license types taking the guidelines of both licenses to compare.

The advantages
To know why you have to choose for open source you first have to know the benefits. If there are no benefits you won’t even look at it. Open source has a lot of advantages that make it a powerfull tool for the business world. The most important benifits have been summed up using Linux as an example for some. Some of these advantages come from the interesting book Expanding Choices.

1. Escape vendor lock-in
   Frustration with vendor lock-in is a reality for all IT managers. In addition to ongoing license fees, there is lack of portability and the inability to customize software to meet specific needs. Open source exists as a declaration of freedom of choice.
2. Lower software costs
   Open source solutions generally require no licensing fees. The logical extension is no maintenance fees. The only expenditures are for media, documentation, and support, if required.
   Example of buying and implementing servers:
• Windows Server:
  win 2000:� 300,- + 0 clients
  upgrade 2003:� 778,- + 5 clients
  Implementation:� 640,- (8 hours)
• Linux Server:
  Linux: 0,-
  Upgrade: 0,-
  Implementation: 640,- (8 hours)
3. Simplified license management
   Obtain the software once and install it as many times and in as many locations as you need. There’s no need to count, track, or monitor for license compliance. Taking the example of the lower software costs, 5 windows 2003� servers cost about� 7100 euro and 5 linux servers cost about 3200 euro.
4. Lower hardware costs for some software
   In general, Linux and open source solutions are elegantly compact and portable, and as a result require less hardware power to accomplish the same tasks as on conventional servers (Windows, Solaris) or workstations. The result is you can get by with less expensive or older hardware. This is nowadays getting less and less important. Where you still can see the difference is when you compare MySQL and MS SQL. You see that MySQL can be installed on the same server as the web server. But MS SQL instead requires a seperate server specifically for MS SQL. Costs for a MS SQL server is about 4000 euro hardware and 6200 euro for the license. That compared with MySQL this is off course a really big difference.
5. Scaling/consolidation potential
   Again, Linux and open source applications and services can often scale considerably. Multiple options for load balancing, clustering, and open source applications, such as database and email, give organizations the ability to scale up for new growth or consolidate to do more with less.
6. Ample support
   Support is available for open source often even better then proprietary solutions. First, open source support is freely available and accessible through the online community via the Internet. And second, many tech companies are now supporting open source with free online and multiple levels of paid support. So for companies in need of Service Level Agreements this can be bought. But not all open source software has these possibilities.

Open source has a lot of advatages compared to closed source software and every day is growing because the community is growing, and people skills are growing.

Although this is growing this much and fast it does not mean that closed source will disappear in the end. To use open source you have to have another mentality regarding the expectations of the product, and the conditions of use, compared to closed source software. When choosing for open source software you have to know that it can be less fast that a closed software version, or that it can be less user friendly, or less reliable, or have less technical possibility’s. So to make a choice between open or closed source software is compare your needs and wishes with both products and balance what suits best.

Disadvantages
Today’s companies have some struggles with open source software. Also companies who sell closed source software are not really happy with the open source solutions. But this competition motivates both to develop more and more to try to distinct themselves from the other.
The following points are some important struggles what companies have regarding open source.

1. Quality software
   The quality of open source is still not perfect. If you put a open source software next to a closed source software you see that there are some differences in for instance design. Open source projects often don’t have enough possibility’s to set up a good design. Then it is more likely to find fauls in open source. This are often small things, but because most companies can not afford fauls, even if these are small it can be bad. Alltough If you take open source of today you see big progress in all of these aspects.
2. There is no guarantee that development will happen
   In other words you can not assume that there will be an ongoing developement of the product. Or even that the software will reach a usable stage. This is because if there are not enough fundings during the project start and development, the projects just “dies” when it runs out of fundings, programmers or needs of the users.
3. There may be significant problems connected to intellectual property4
   Because there are so many contributors to a open source project that eventually it is almost impossible to find out who owns it. This is becoming important now that countries are accepting software and algorithm patents. Like in mentioned in the article: “It is very difficult to know if some particular method to solve a software problem is patented, and so the community can be considered guilty of intellectual property infringement”. This subject is though pretty complicated, and the open source community is working hard, together with GNU to get it to work correctly and preserver the property rights correctly. To read more about it we refer to an MIT document on the internet
4. It is sometimes difficult to know that a project exist, and its current status
   This is because there is almost no advertising for open source projects. This is because there is often lack of financial funding by companies. These fundings are essential if you want to use good marketing.
5. Research and development does not exist.
   The struggle of open source is that there is almost no room voor research and development. And exactly that research and development creates possibility’s to grow, and develop new and interesting tools for the needs of the customers. At this moment most open source software tends to “follow” the path of the closed software.

Most disadvatages of open source can easily be refuted. There are a lot of alternatives and solutions developed by the opensoure comunity to create a better platform. Take for example the guarantee of development. Every person or company can motivate the open source comunity by getting involved. This could be enough to “motivate” a project to continue and ensure continuation of that particular project..
Intellectual property is a difficult point. But there are new licensing methods introduced that can become the solution for this problem. For instance “Creative Commons Licensing” is in a way working in that direction by protecting open source.
More and more interesting open source software are making it nowadays attractive to think about using or contributing to open source. So measuring your wishes and needs with the things open source can offer, will give a good answer to questions regarding open source software in this research.

Commercial use of open source
Commercial use of open source is not conform the GNU GPL definition. It is not allowed to sell an open source product. From the moment on you sell the open source product, or an alternated version it is not anymore open source. Also you are violating the GPL General Public License� license.
There are constructions possible for selling “open source” but eventually these solutions are not open source according to the GNU General Public License definition that says: “By contrast, the GNU General Public License is intended to guarantee your freedom to share and change free software–to make sure the software is free for all its users.”

To find out if companies really should use web applications, we will talk about the advantages and disadvantages of web applications, compared to client/server and desktop applications.

Advantages of web applications
There are a lot of notable advantages when using web applications. All these advantages have a higher purpose of serving the company which are categorised in this list into 2 different categories, namely one category specifying advantages to decrease the TCO (total cost of ownership), and a category specifying how business workflow can be increased through the use of web applications.

Decreased TCO 1: application installation not needed
The installation of an application can cost a company a lot of money, not only hiring the experts to install the application, but also the time that is needed to install the application on the company’s server which might interupt the company’s workflow. The installation of an operating system, including e-mail server software and a backup application might already cost around € 3000,- alone. The configuration of a complete network including the installation of client computers could also cost another € 3000,- or more, depending on the number of clients.

Decreased TCO 2: expensive hardware not necessary
Hardware upgrades are not necessary, because all available computers can still be used. When there is no hardware available yet, a very inexpensive computer of around € 300,- can be purchased. But when companies start using a new server/client application, new hardware like a company server and some client computers is usually necessary. The costs are really dependent but at least € 3000,- will be spent buying new hardware.

Decreased TCO 3: no more expensive software licenses
There is no need to buy any software licenses anymore because an online application is usually paid on a month by month basis. This means there is no need for buying any office applications, e-mail management software, data backup or any other software program. Of course all needed features then have to be available in an online version, which is not always the case. Licenses of various software applications can cost around € 100,- to € 600,- for every employee. When every employee needs around 5 applications, it is obvious that € 1000,- per employee is required.

Decreased TCO 4: automatic updating
Once in every time each application needs to be updated. usually the IT manager needs to apply these updates to every computer, keeping the employee from their computer for like 15 minutes. This will cost the IT manager around 2 hours for 5 computers. This is not a huge problem when an update is provided every month and there is only one application installed, but when we have 10 applications providing updates every week, the company needs to hire an extra IT manager!

When an online application needs to be updated, which is usually every month, it will be done during the night, probably in the weekend, and the update will be ready within 15 minutes for every user at once! Next to that, all applications are always up to date without having to follow the update cycle of every software company.
Decreased TCO 5: updating for free
Most updates are provided as a service when a software application is implemented. This is of course only for the term the software will be used, which is usually around 3 to 5 years. After this period, an application will be upgraded to the newest version which will cost money every time again.

When using a web application, there is always only one version of this application available. There will never be two companies using different versions of a web application because only one version exists. Therefore all updates are provided automatically withoud the need to pay extra every few years again.

Decreased TCO 6: company power usage decreased
When installing a server there will be a higher power usage in the company as a whole. Most servers will run 24/7 and need a high quality power outlet. Next to that there will be a special location for this server which needs a seperate airflow system and possibly an advanced backup system using a lot of power.

Decreased TCO 7: decreased size of IT department
All support of a web application goes through the web application itself. Usually 24/7 telephone and e-mail support is also available. Therefore the IT department can be decreased in size and knowledge. This will save most companies a lot of money every year. When there is no IT department available at all, this will greatly lower the cost of external IT managers.

Decreased TCO 8: migration without changing PC configurations
All computers, whether they have a Microsoft, Linux or any other operating system can be used to access a web application. Even all PDA’s and other mobile hardware can be used if only they have a browser. This means migrating local computers from using an old application to a new web application can be done within a second saving the time of migration.

Decreased TCO 9: information provided more easy, saving costly employee time
When all usable information is directly available to all employees, everyone can help a potentional customer. Every employee can find all information concerning the client within a few clicks. This concerns not only all employees in the building, but also all employees working from home or maybe even from another country.

Decreased TCO 10: no personal training sessions anymore
Users can be provided with an online training through the web application rather than going to a seminar or workshop somewhere to learn how to use an application. It is much more efficient to learn an application on the location where it will be used eventually. Also when something is not completely clear, everyone can induvidually repeat an online training session.

Decreased TCO 11: bugs and problems will be fixed on the server
When a problem occurs, someone from the software company will stop by to solve the problem. Usually within a few hours. When this problem can be fixed on the main server of the web application, first of all this will cost less time because the main server can always be accessed through the internet, and next to that, the problem will be solved for all other users at once, so less problems will occur, and less bugs will remain unsolved.

Decreased TCO 12: Centralised updates eliminate backward compatibility issues
Every web application only has one version, this means all old data will be available through the same program forever. When a new desktop application arrives, sometimes old documents have compatibility issues. To keep a backward compatibility there has to be some sort of software version control, which is never an issue in any web application.

Decreased TCO 13: a web application will work from everywhere
There is no need to buy different versions of applications for different devices. A web application will work on any device with a web browser available. Also the web application can be used interchangeable through different computers or mobile devices. This means of course a huge advantage in freedom of use and lower costs.

Increased workflow 1: backups, backups, backups
In a good web application backups are always available, and information can be restored with a single click or phone call. When a web application is used the data will be backed up automatically i.e. every day, and next to that can be backed up by every company itself at any given time, including an optional download of the backup. Restoring the backup is also not a problem, this can be done in the web application itself without any knowledge of backup applications. Backing up company data is often done inconsistently or might even be forgotten sometimes.

Increased workflow 2: first class protection against material damage
A web server on which the web application resides is located in a datacenter, a top safe environment with protection against fire, intruders, power loss and any other form of possible material damage. A datacenter can be compared to an advanced locked safe, or the black box inside a plane.

Increased workflow 3: first class protection against virtual damage
All company data is on the software side more safe than any application server in any company in the world. These web application servers are made to be protected against virusses and hackers. Their only task is to keep data secure, therefore professionals are working every day to make sure the security of data is not comprimised.

Increased workflow 4: access from anywhere in the world
24 hours a day, 7 days a week, employees or agents located at different locations, on the road or even overseas, all have direct access to current information.

Increased workflow 5: better sharing
When employees work together there will be more efficiency in the company, but this will only work when their software works with them, therefore web applications are the ideal software to make sure people can work together because everyone is actually still working in the same system, while being on a different location.

Increased workflow 6: real time online connections are very easy
Credit card verification is only one of the examples which is made easy due to the use of web applications. Also VoIP is one of the highly anticipated new features of web applications. Due to the fact that these web applications already completely run online there is a huge advantage in using web applications over regular server/client applications.

Increased workflow 7: website is maintained automatically
All company information can, if needed, directly be shown through the company’s website or can be used for other purposes where it can improve business. This means all information to be used only has to be entered once, to save more time and to minimize the possibility of errors.

Increased workflow 8: documentation always up to date
When having 10 software applications installed, a company will need a small archive to keep all documents delivered with these applications, in case someone might need it one day. When that day arrives, the documentation of the software is probably already outdated or maybe even lost. When this documentation is online, it can stay up-to-date and can always be found online, ready to print or read by anyone who needs it.

Increased workflow 9: try before buy
Usually there is an online demo available for companies to test a web application. When a new application needs to be implemented, companies can test a few of these applications themselves, before all salespeople come by to sell the product.

Increased workflow 10: platform independent
Whether a company wants to use a Mac, a PC or any combination of these and other devices, they can access the same software through the internet browser. These choices are usually neccesary due to other aspects i.e. the need for an apple workstation to do video editing while using Outlook on a Windows based device to keep an agenda or planning.

Disadvantages of web applications
Although not many, some of these disadvantages are pretty serious so we need to take a very close look at the disadvantages.

Disadvantage 1: authentication required
Logging in with a username and password is required everytime the user wants to use the application. Of course certain options are available to work around this problem so this is only a small problem, but the fact remains that with or without user input, a user has to be logged into the system first before it can be used. This is totally different than for example a desktop Word application.

The reason for this authentication is that one user can access the application from different locations on different computers or handhelds. Therefore the application cannot automatically determine who wants to log into the application from what device. This of course also provides some extra form of security because nobody can go and sit behind someone else’s computer and use the application in a wrong way.

Disadvantage 2: limited uptime
A professional internet connection is a must have, because whenever the internet connection fails, a web application cannot be used at all. A warranty of 99,6% uptime is industry standard. Still this means that every year the system might not be available for about 8 full hours. Of course dependent on the company’s wishes, this warranty can be extended to 99,9%. But this usually comes at a price most small companies cannot afford. In some cases however the uptime warranty can be only 95%. This means the web application is not reachable for one whole working day each month. This might be a serious problem for some companies. In order to seriously use a web application, companies need to make sure they have a connection with an uptime warranty of at least 99,6%.

Disadvantage 3: impossibilities
Some software cannot yet be executed in a browser, for example highly interactive photo edit software. There is no solution for this problem aside from the fact that when the need in the market for these kind of applications arises, the browsers will be ready and the internet connection speed will be ready too. The devellopment of photo edit software i.e. is also a very long and difficult process, so it will take some time before these applications arrive and are ready to be used in the mainstream market.

Disadvantage 4: trustworthy hosting provider
People are not yet confident with storing company critic information outside their own company. This is probably the biggest disadvantage of all. Companies are not well informed about the safety of their data, or simply do not trust the hosting provider. This problem might be overcome by time, but for now it is still a serious problem.

Disadvantage 5: service level agreements
When problems occur with company critic applications, it is always important to solve these problems as soon as possible. And with every serious software company a service level agreement (SLA) can be purchased. This is usually purchased on monthly fee, and can be very expensive depending on the quality of the service. This is no different in web applications, except for the fact that an SLA is absolutely necessary because only the software company can solve this problem, when with server/client applications it might in some cases be possible to have the IT department solve all problems.