After a hectic October in the free software world, in which we witnessed events including the launch of OpenOffice.org 2.0 and MySQL 5.0, I thought November would be quieter and that I’d be struggling to find material for this article. I couldn’t have been more wrong. If anything, even more has happened this month than in the last, so I have concentrated on the events I feel most are the most important and relevant. To start with, there have been new versions in five very major software packages. These are:
As well as the continuing discussion of OpenDocument adoption as an office format standard, and a new initiative on the patent front.
There are very few pieces of software, free or otherwise, that has made a more dramatic entrance than Firefox. It is just over one year old, in fact its birthday was celebrated this last month, and it can already boast about well over one hundred million downloads, and probably that number of installations. It has caught the imagination of a vast number of people and an enormous advocacy following.
Firefox has earned its laurels. It was well thought out and designed, and has all the hallmarks of a successful piece of software. The designers intentionally kept the interface both functional and simple, rejecting additions that made it more cluttered without any justifiable reason. It filled in a void of a free, secure, simple, tab based browser that followed standards. It used the free software model to ensure it became a cross platform product, thus allowing it to be installed on as near one hundred percent of machines in the world as to make no difference.
All in all, I have to say, Firefox is just brilliant.
All in all, I have to say, Firefox is just brilliant
After several years of stagnation, Microsoft has announced to the world that it is releasing its next version of Internet Explorer. The feature set of this new “IE7” seems to read like the feature set of Firefox. Whatever Microsoft may say they are obviously playing catch-up here, and are trying to stop the hemorrhaging of their user base to the truly free and superior product. The new feature set, performance improvement and general enhancement of the new Firefox will simply make their task harder.
During the first year, any enhancements (especially the few security issues that came to light), were dealt with quickly and professionally. A nice change from many proprietary products out there. Firefox is fast stealing the crown from Microsoft’s Internet Explorer as the browser with which all others will be measured.
Firefox 1.5 entered the world with such fanfare and shouting-from-the-roof-tops that it somewhat overshadowed another event that occurred the same day, namely the release of KDE 3.5.
It’s not possible to overstate what KDE, and its competitor/cousin GNOME, have done for the Linux desktop. A large misconception floating around is that The X-Windows System is a graphical user interface, or GUI for short. It is, in fact, an infrastructure in which a GUI can exist and is a very versatile infrastructure, but still just an infrastructure all the same. Prior to KDE and GNOME the graphical interfacing of GNU/Linux and other POSIX systems consisted of disjointed widget sets and window managers. Although this worked well enough to use applications, it fell short when significant interoperability and consistency was required.
There is no understating what KDE and its cousin GNOME, have done for the Linux desktop
KDE, and then GNOME took care of this. They not only provided a consistent look and feel between widgets on applications, as well as the window manager, they also provided a superior inter-process communication infrastructure between the applications themselves. At first they each just interacted with their own applications, but common sense and co-operation triumphed and they now interact well with each other. The friendly rivalry of KDE and GNOME can be compared to that of MySQL and PostgreSQL in that they spur each other on, and all benefit from that. The free desktops are now more than a match for the locked-in relatively non-versatile Microsoft Windows offering.
The new features that KDE have released into a waiting world further enhance the functionality and intuitiveness of an already fully featured desktop. And this isn’t just on the desktop either—there’s also KDE’s own browser Konqueror( which is now one of the few that passes the famous ACID2 test), an entire office suite, a number of graphical manipulation programs and a host of other applications. Full details can be obtained from the KDE 3.5 web page.
PHP is free software’s leading web scripting language. I believe, at over twenty three million sites, it could well be the world’s leading scripting language. As far as products go it is a quiet beast that beavers away in the background, without fuss, serving up web pages to the populace. All the while making life easy for the web developers who want to integrate pages with databases and the such. The arrival of the new release, PHP 5.1, is typical in behavior. It was released without fuss, without fan-fare, its purpose to be useful rather than glamorous.
PHP itself has more features than I can count. Its manual’s “Function Reference” sections goes up to number “CLXXI”—which an old Roman would know as one hundred and seventy one—amounting to several thousand individual functions. There seems to be a native function to do anything you could ever want a web server to do plus a whole lot more, and it is expandable and growing too.
The versatility and popularity of PHP means that the web-server scripting scene can never be cornered by any specific entity. For freedom to be denied in this area a proprietary product will need to absolutely trump this one. This is something that is unlikely anyway, and continuing improvements here make it virtually impossible.
The versatility and popularity of PHP means that the web-server scripting scene can never be cornered by any specific entity
Asterisk has the potential to become one of the most disruptive pieces of software in existence. Currently the telephone PBX industry seems to be currently dominated by expensive proprietary solutions from strictly proprietary companies. Asterisk is the upstart that offers a free software solution that runs on standard hardware, and thus can be installed for a fraction of the price. Although it seems this new product hasn’t achieved significant deployment in percentage terms yet, its deployment in number terms is increasing fast. I think we can expect some panicking soon from the dinosaurs resident in that industry.
Asterisk has the potential to become one of the most disruptive pieces of software in existence
The new release, version 1.2, is a natural part of the above process. In this bugs have been fixed, features added and performance optimized. It takes the product up a few more rungs of the ladder toward general acceptance. As deployments increase, and reliability is proven in more and more situations, we can expect Asterisk to become more and more a part of the telecommunications landscape. Also, as popularity increases so to can we expect the free software model to further take hold, and the more developers will work on it increasing its attractiveness further still.
This is very much a “watch this space” piece of software.
In last month’s newsletter I reported on MySQL’smove to release 5.0, and mentioned that any minute PostgreSQL will answer with an enhancement of its own. Sure enough, its here. PostgreSQL has for a long time now had as full a feature set and salability that can compete favorably with nearly all of its proprietary cousins. This new version pushes its position further to the front of the pack.
PostgreSQL 8.1 offers SQL language enhancements such as “Roles” as opposed to distinct “Users” and “User Groups” and the addition of “IN/OUT” parameters in stored procedures. The developers have also been hard at work improving salability and performance. One interesting addition is the “Two Phased Commit”. This, I admit, by itself sounds boring. However, it enables PostgreSQL to natively maintain ACID transactions (an industry standard) while replicating the database. This, with the other enhancements, make it on par with the best of the best for the really large database deployments, something that has not gone un-noticed by Sun Microsystems who are now offering support for PostgreSQL on its flagship operating system Solaris 10.
PostgreSQL is often compared to MySQL—and the friendly rivalry between them seems to benefit them both. MySQL scores well on performance with the small implementations such as small to medium sized web based solutions. PostgreSQL is more scalable and standards compliance for the larger deployments. As they both improve PostgreSQL is becoming faster at the low end, and MySQL is being able to cope with the larger databases, and each are showing improvements in their traditional strengths. This new version is a continuation of that trend, with PostgreSQL having as much, maybe more, features and advantages going for it as many expensive non-free offerings.
PostgreSQL has at least as many features and advantages going for it as many expensive non-free offerings
The saga of Microsoft versus the Open Document crowd is continuing, but for those who have missed previous episodes, here is a summary:
There exists an Office file format called OpenDocument. It was created by an organization known as OASIS to be a standard. This is the format used by the OpenOffice 2.0 suite by default, and most other office suite vendors have committed to support it. Most, in this case, does not include Microsoft. They have there own proprietary XML format which their new suite uses, and they will not compromise their holy-than-thou attitude by including anyone elses.
Various organizations, but in particular the Commonwealth of Massachusetts, decided that they did not want the access to their archived documents to be potentially encumbered by unfavorable licenses. And so they decided to standardize to the unencumbered OpenDocument format rather than Microsoft’s proprietary one.
There was then the expected objections, FUD-spreading, snarling and gnashing of teeth which came from Redmond, resulting in some extraordinary meetings and discussions. There were also some fairly nasty character assassinations. It is at this cliff hanger moment that the episode ended and we were left in suspense as to what would happen next.
What happened, in retrospect, was obvious, though I have to admit that I did not predict it. Microsoft have seemed to rename their format to “Office Open XML”, have produced a covenant that they claim will mean they will not sue anyone who uses their format, and have submitted it to the ECMA as a fast track to get it accepted as an ISO standard. Microsoft are not eager to lose their dominance of the Office market, and a theory circulating is that they see that maintaining control of the file formats is a way of doing that. I, for one, believe it to be true. However, I also believe it would be better for the world if the OpenDocument format is adopted as the standard. In my opinion, not only is it technically superior, but it is guaranteed to remain open to produce a level playing field in the market. Also, there is some doubt as to exactly what extent Microsoft’s covenant makes their proprietary formats open.
Stay tuned for the next thrilling installment...
It is nice to write a heading like that. Nowadays the patent news seems to be dominated by items such as “this stupid patent has been awarded here” or “that stupid patent is being used to extort money there” etc. But there has been a pleasant change.
The news that has got me smiling is that OSDL, IBM, RedHat and a number of others have got together to form a Patent Commons Reference Library that allows free software developers to develop without fear of infringing upon those patents. Only time will tell if this is useful or not, though at least an effort is being made to ensure software freedoms won by the community can benefit all without hindrance.
While researching and writing this article I've noticed a reoccurring theme. In many areas, proprietary solutions are more popular than the free software ones. However, the existence of the free solutions means that the proprietary ones cannot take advantage of their current popularity. The free solutions are a continuing threat to proprietary software vendors in such a way that should these vendors start overcharging, or providing a lousy service, or an inferior product, there is a free productready to take their business away. Also, unlike traditional competitors, free software never goes bankrupt, or disappears, or is bought out. Free software may not currently be as well deployed as the non-free, however the free-software model means that these checks and guarantors of value can never die.