A little while ago I did a tutorial on upgrading to Flash Player 10 Beta. I still recommend most Linux users do this, since it fixes a problem with drop down menus disappearing behind Flash content, however I noticed one issue today. The issue is, in fact, quite funny. When clicking on a CNN.com video, I was presented with this error message (I am on Ubuntu running Firefox 3 with Flash Player 10 Beta):

Interesting error message...

Hmmm… “This CNN.com feature is optimized for Adobe Flash Player version 8 or higher. You are currently using Flash Player 10″

Last time I checked 10 was greater than 8. Anyway, for anyone who watches lots of CNN videos, don’t upgrade to Flash Player 10 quite yet. If you already upgraded, don’t worry. Just install an older version like you would install version 10 beta and you should be fine again.

I checked most of the other major news sites and they seem to all be fine with Flash Player 10 beta.

I have always been a big fan of web (or cloud) applications, because they make it easy to switch freely between computers and between platforms. With web applications, your applications don’t tie you to any platform, leaving you free to choose whichever platform suits you best. Recently, though, a Linux.com article and a post on a Clipperz blog got me thinking about how open-source and the cloud go together, or don’t go together.

As I first began to think about it, it seems putting the two (open source and web applications) together would be a challenge, but I soon realized that they are, in fact, a natural fit for each other.

Perhaps the best example of open-sourcing a web application is Reddit, which recently open-sourced all of their code. The biggest argument against open-sourcing code with an application like Reddit is that everyone can copy what you have done. That is true, but anything based on, for example, the Reddit code must be released back as open-source, assuming the original code was released under a GPL-like license that requires this. So for an application (if you call it an application) like Reddit, open-sourcing seems to make a lot of sense. This basic logic can also be applied to most other more traditional web applications, such as word processors.

In the case of partially or fully businesses-oriented web application, a further step could be taken by selling a subscription update service for businesses that wanted to host the application themselves.

You could pretty much summarize the last two paragraphs by saying that the arguments for open source in web applications are essentially the same as the arguments for open source in traditional applications, but there is one more (or at least one more) reason for open-sourcing web applications:

One frequent concern with web applications is privacy. How do I know that Google is not reading everything I make in Google Docs? Done correctly, it should be completely possible to encrypt the data before it leaves the client’s computer such that Google (or <insert name of web application company>) cannot possibly read it. Even if a web application provider claims they have done this, though, how do you know for sure? The answer is that you really cannot - unless the application is open source. If it is open source, tech-savvy people will almost certainly go through the code and ensure that your data is safe, but if the application is closed source, you just have to trust the company.

So for these reasons, I see the aruments for open source in the cloud as even stronger than the arguments for open source on the desktop. I just hope that the web application developers agree.

Like it or not, one of the most important features of any Microsoft Office alternative is being able to read Microsoft Office files. With the recently released OpenOffice 3.0 Beta adding support for importing Office 2007 (AKA OOXML, AKA .docx) files, I decided to test how it would handle files created in Microsoft Office 2007 saved in both .doc (the older version) and .docx (Microsoft’s new format.)

To do the test, I used three different files, one with basic formatting (highlighting, font size, bold, italics, underline, etc.), one with changes tracked, and one with charts and pictures and that sort of thing. Each file was also saved both in .doc and .docx. Overall, particularly for a beta, I was very impressed with how OpenOffice handled the files. As you might expect, there are really no deal-breakers on the .doc side. On the .docx side, there are, unfortunately, some, but these will, hopefully, be resolved soon. (Keep in mind, this is still a beta version.)

Problems I noticed with .doc (older, but still the most common) files:

• Formatting changes did not appear to import correctly. For example, if you are tracking changes in Office and you make a piece of text bold, the text will be bold in OpenOffice, but there will be no indication that that is a change.
• Bullets do not always display correctly. For example, in Office there might be a little check mark for a bullet, but in OpenOffice that check mark might be replaced by a box.

Problems I noticed in .docx (newer, but not widely adopted yet) files:

• Text does not always wrap to the next line at the same point. For example, if a line ended with the word “cat” in Office, it might end with a different word in OpenOffice.
• Changes are not correctly imported. For example, if you had changes tracked in Office, they do not show up in OpenOffice. Comments are there, but there are weirdly inserted into the body of the document and totally mess up the formatting. Hopefully this is just a bug in the beta.
• Word Art, charts, and elements of shapes do not work correctly/at all. For example, Word Art, shapes, charts, and other similar objects may not show up correctly or at all in OpenOffice.
• Bullets do not always display correctly. See above.

Yes, there are still some issues, particularly in the .docx format, but these may well be resolved before even 3.0 is shipped. To put things in perspective, I will be following this up with a look at how Office handles ODF files. Oh wait, I guess I will have to wait until 2009 to do that. At least they are doing it.

Are you sick of drop down menus being hidden by Flash content? Flash Player 10 prerelease fixes that. Here is how you can install it: (This tutorial is for Ubuntu, however it should work on other distros.)

1. Visit this page. Click on “Download Plugin for Linux (TAR.GZ, 3.71 MB)” and save the file to your desktop.
2. Right click on the file and choose “Extract Here.”
3. Open up a terminal (Applications > Accessories > Terminal) and type in this code:
   cd Desktop
4. Then type in this code:
   cd install_flash_player_10_linux
5. Then type in this code:
   ./flashplayer-installer
6. Close all browsers, then press enter twice.
7. Type “y” and press enter, then type “n” and press enter.
8. Relaunch your browser, and your drop-down menu problems should be fixed.

Just about everyone, including the Mozilla developers, were amazed at the success of the Firefox Download Day. Just on the release day, Firefox 3 got over 8 million downloads. Perhaps more impressive, though, is the continued stream of downloads even two weeks after the release.

If you watch the download counter, you will already know that Firefox 3 has, at the time of writing, nearly passed 28 million downloads. This means that the average number of downloads per day over the last two weeks (since Firefox 3 was released) is almost exactly 2 millions downloads per day! Even if you don’t include the first day it still amounts to about 1.5 million downloads per day. That is an impressive number!

In the past few days, I am have been seeing about 1000 downloads per minute, according to the download counter, in the morning and around 500 per minute at night. I look forward to seeing if Firefox can keep it up at this pace. Even if it can’t, it has already done two impressive things, getting a ton of downloads in 24 hours and keeping up a steady stream of downloads over the course of two weeks. Great job Firefox!

A lot of focus is put on the Linux desktop. “Is Linux ready for the desktop?” “Will this be the year of the Linux desktop?” “Is the Linux desktop usable by regular people?”  All of this is, generally, asking the same thing. When is Linux finally going to be commonplace on the desktop? The average user’s experience is not all that matters, though. Developers must also be attracted to Linux before it will become commonplace on the desktop.

As shown by a recent Fortune article, Linux already is appealing to the developers, but it is important that Linux becomes not just appealing but irresistible for developers. Why is this so important? Because with developers come applications that have previously been Windows only. As developers switch to Linux, applications that used to be a reason why some could not use Linux as their only OS will become cross-platform or even Linux only.

Look at the Mac. Do you think it would have come as far as it has if Microsoft had not brought Office to the Mac? I don’t. Even though OpenOffice/NeoOffice would probably have worked for many of the people that bought Office for Mac, people are just more comfortable using the exact same application they are used to.

So how can Linux appeal to developers? There are a number of ways that come to mind. Of course, the best thing to do would be to ask a developer, but seeing as I am not a developer here are some ideas:

The most obvious, of course, is to make it easy to develop for Linux. There are other things too, though. Easy application distribution of free and commercial applications through systems like CNR is a definite plus. Also, providing a good introduction to programming on Linux would be helpful for those just switching to Linux.

Because of these sorts of advantages, hopefully more and developers will switch to Linux and bring users along with them.

Adobe Air is a technology that allows certain web applications to be run on the desktop, as if they were normal desktop applications.

According to the Adobe Air website, Adobe Air “lets developers use proven web technologies to build rich Internet applications that deploy to the desktop and run across operating systems.” Essentially what that means is that developers can use tools and techniques that are already familiar to them and create web 2.0 applications that will not only run inside a web browser, but also on the desktop, regardless of the operating system that desktop is running.

At the time of writing, however, Adobe Air for Linux is not feature complete, so not all applications will work perfectly.

One of the main barriers people face when moving from one operating system to another is applications that don’t run on their new operating system. This creates an uneven playing ground where the operating systems on top have a huge advantage over other operating systems, regardless of technical merit. Web applications started to solve this problem, because most web applications can be run on any operating system, but web applications have one major disadvantage: they require an internet connection. Air lets developers create web applications that will also run on the desktop, removing that one major disadvantage and allowing web applications to finally level the operating system playing field.

For Linux, if Adobe Air catches on, many new applications will become available for it, not because they were developed specifically for Linux, but because they were developed for any operating system. This means switching to Linux is likely to become very easy in the future.

For a guide on installing Air on Linux, see my tutorial “Adobe Air on Linux: What, Why, and How“.

Recently Adobe announced that it is joining the Linux Foundation, according to Fox Business. The Linux Foundation is “a nonprofit consortium dedicated to fostering the growth of Linux,” according to their website. When things like this happen, it is hard to tell if it is just a press release or if Adobe really intends to pay more attention to Linux, but we can hope they really do care about Linux.

Assuming this announcement is more than a press release, it could mean a number of things. For one thing, it could likely lead to a better Flash Player for Linux, which would be very nice. Another, perhaps more significant possibility, is that Adobe will bring their applications like Photoshop, Flash, Dreamweaver, etc. to Linux. If that happened, it would be a huge deal, because Adobe’s applications are almost all often considered #1 in their category (not that I am a fan, grr… monopoly grr…)

While it may seem crazy to some people (myself included) that Adobe would put all the effort into porting their applications to Linux, there is another way they could do it. In order to make Picasa for Linux, Google took the Windows version and packaged it with a custom version of WINE. With all the attention being put into Adobe applications on WINE already, something similar might work.

Adobe joining the Linux Foundation could just be a press release, but I am hopeful it means more than that for Linux.

It would seem as though there is little left to do in the world of word processing and office suites but fix bugs, ensure compatibility, and add tiny little new features to advanced features no one uses. That is not necessarily true, though.

Imaging you had only ever used a browser that browsed the internet and did nothing else. What else is there to add to that browser, you would think? But, now, if you had a browser that did nothing but browse the web, it would seem like a feature-free browser and you would switch right back to Firefox will all your extensions. The same thing could happen in office suites.

If you have not already figured it out, what I am saying is that extensions are likely to be the next big innovation (or copied innovation from browsers) in office suites. (After all, OpenOffice is already starting to put more focus on extensions.)

It is easy to think “oh yeah, extensions will be great,” but what kind of extensions would you really use in an office suite? Well, for one thing, there are always the time-wasting extensions. You could have RSS feeds, or a media player, or whatever just to waste time have fun. On the more useful side of things, you could have an extension that looks up words on some website and displays the result right inside the office suite. Another example would be an extension that could pull in data, such as stock prices, from a web site.

If extensions do take off, it will give OpenOffice another advantage over Office. Since OpenOffice is already moving in the direction of extensions and Microsoft is often slow to implement new features, it seems almost certain that OpenOffice will get a boost over Office from really supporting extensions. This is turn would give Linux a boost, because the more people who like OpenOffice, the more people who don’t have to worry about Office not existing for Linux.

As one blogger points out, Opera’s minimal success is difficult to understand, because Opera is filled with innovative and useful features, yet it has no where near the adoption of Firefox. The blogger’s explanation is that the user interface and experience is too different from Firefox and IE, and therefor not appealing to new users. While this may in part be true, I believe that openness is the key difference that prevents adoption. By openness, I mean the ability to customize the browser to be what you want it to be. Firefox accomplishes this on multiple levels. First of all, everyone is encouraged to write extensions (and a lot of people do.) These extensions are enough for most people to customize their browser to be anything they can think of. (Skins should be mentioned too, but Opera has those.) At the base of this ability of customize Firefox is that it is an open-source project. You can’t get any more customizable or flexible than that.

The author of the post mentioned above dismisses these differences as not things the average user would care about, but I think this misses a key point. Firefox is mostly spread through word of mouth, originating at tech-savvy users who care enough to do the research. These tech-savvy users tell their friends to use whatever browser they like and soon enough their friends are telling their friends and so on. In this way, some one who has no idea what an “extension” is or what “open-source” is may be using Firefox because it is open-source.

There is no reason this should only apply to Opera and Firefox. This same reasoning can be used with almost any piece of software. If you can think of any more examples of this, go ahead and post them in the comments!