OSDL: it’s not just for data centers and carrier-grade Linux anymore. Not as of this morning, that is, when OSDL issued a press release to announce the formation of a third initiative: the Linux desktop. Newsforge article, press release, Q & A. Trolltech has also joined OSDL and will participate in the Lab’s new Desktop Linux Working Group.
OSDL CEO Expounds on Desktop Initiative Announcement
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
We love Linux. We wish it can grow and take market share from Microsoft on the Desktop and the Server. Well, this is another step in the right direction. Once people will get accustomed to Linux at work, it will be a lot easier for them to install and use it at home. That’s why OSDL will focus on the enterprise Desktop. I wish I could travel in time to see what would be the situation and how will Linux look in a few years… say, by the time Longhorn decides to take form in expensiveware/bloatware.:) I like this term, Bloatware! Makes me think of windows…
:::PROUD TO LIVE IN THE BIRTHLAND OF LINUX:::
We love Linux. We wish it can grow and take market share from Microsoft on the Desktop and the Server. Well, this is another step in the right direction. Once people will get accustomed to Linux at work, it will be a lot easier for them to install and use it at home. That’s why OSDL will focus on the enterprise Desktop. I wish I could travel in time to see what would be the situation and how will Linux look in a few years… say, by the time Longhorn decides to take form in expensiveware/bloatware.:) I like this term, Bloatware! Makes me think of windows…
Let me first set this clear. The definition of bloatware is when you have a product which has a dispurpotionate amount of code vs the amount of features. A example of this would be if a company created a windows with a button in the middle of it, however, there was a million lines of code just to generate that button and window.
Windows according to that defintion isn’t bloatware. Hackware, sure, they create hacks to work around problems rather than addressing them thus creating an API that has numerous contradictions and ways of doing the same thing. One thing Microsoft does do correctly is the re-use of components throughout the system. Yes, there are downsides but there are also benefits as well. For better or worse, Microsoft has weighed up the pros and cons and have made their decision.
As for Trolltech joining OSDL, great news. If this means that KDE will become the centre piece to the linux desktop stratergy then I am all for it. One thing we must keep in mind is this, new programmers to the Linux world will come from the Windows world and Qt has enough similarities with win32 to make it easy to transfer to.
What there needs to be now are companies like IBM to demonstrate via “blog” the steps they used when porting Lotus Smartsuite to Qt/Linux and the possible benefits and pitfalls one may come accross. If they can give a running blogg then developers can actually track a real product in progress and understand the fully the magatude of any porting undertaking.
Qt is nothing like Win32. The former is a dynamic, object-oriented API. The latter is a static, procedural API. Qt isn’t even anything like MFC. The former is object-oriented, while the latter is obviously a wrapper over a C API. The former is based on signals/slots, while the latter is based on message loops. The former is a layout-manager based API, while the latter is a precise-positioning API. Other than the fact that they are both C++, there is little similarity between the two.
I’ve been working with the Qt code over the past few days, and I have to say that my respect for TrollTech and its development team has gone up yet another notch. Its easy to do C++ wrong (see MFC). There are lots of potential pits to fall into. Qt avoids most all of them. Resource management is clear, explicit, and rigerously adhered too. The source layout is clean, logical, and has just the right granularity (classes are neither small and trivial, nor bloated and complex). The inherent static nature of C++ is mostly avoided. Signals and slots make managing the complex interconnections between widgets much easier.
Its a great thing that Trolltech has joined the Desktop Linux initiative, because its a major selling point for the platform.
Qt is nothing like Win32. The former is a dynamic, object-oriented API. The latter is a static, procedural API. Qt isn’t even anything like MFC. The former is object-oriented, while the latter is obviously a wrapper over a C API. The former is based on signals/slots, while the latter is based on message loops. The former is a layout-manager based API, while the latter is a precise-positioning API. Other than the fact that they are both C++, there is little similarity between the two.
WOW! hold you horses. This is what I was told, “moving from win32 to qt isn’t that difficult, there are some similarities”. That is what I heard from a developer. I’ve never used it, hence, I cannot vouche for how good or bad it is.
Also, the fact I was trying to get at is the fact that qt is a well designed API from day one, it should prove to be very popular in the develop community and companies wanting to use solutions that minimise the development time required to implement a feature.
This is significant to KDE in the Linux desktop and enterprise. As a former Win32 programmer and current Qt one, I can only say kudos to Rayiner Hashem’s reply 🙂