|
|
||
Solving the Kate Focus Problem© Copyright Darrell Anderson. Anyone who has followed this journal, or has read my previous entries, knows that I am fully dissatisfied with what the KDE developers did to Kate, the KDE high-end text editor. Without any survey or poll, without asking end-users, the self-serving developers removed the Project component from Kate to add a Sessions component. The problem for me and people like me is that I am not a developer and the sessions concept is useless to me. The closest I could come to such an environment is managing my web pages. Even then, the former project idea satisfied my needs. Far worse, however, is that the sessions idea crippled the multiple document interface (MDI) of Kate. Without the MDI, I would not consider using Kate, or any text editor. I will tolerate a tabbed interface, but I am a supporter of MDI programs. That is one reason why I continue to use Word 97 — the MDI. In KDE 3.5.x the Kate MDI is broken. The problem is that opening a document in Kate from outside of Kate (from the command line or from within Konqueror) will indeed select Kate, but the Kate window either never receives the focus or never rises to the front of the screen. Nothing I have tried has succeeded in forcing Kate to both receive the focus and rise to the front of the desktop. Usage is always an either-or dilemma and the KDE developers have had six opportunities ( KDE 3.5.1 through 3.5.6) to remedy this bug. They have not. There is a --use option that is supposed to solve the problems, but that option does not work. In my frustration and anger I began to brainstorm and eventually wonder whether I could use Kate from KDE 3.4.3 (version 2.4.1) rather than the version packaged with KDE 3.5.x (2.5.x). Some trial and error proved my idea a success. I now am able to use Kate 2.4.1 within the KDE 3.5.x desktop and with subsequent usage I notice no significant errors or issues. Kate now always opens with both the focus and rising to the front of the screen, and there no longer are any sessions. Projects are available. Using that older version also means that the small widgets (close and docking) in the side components of Kate are once again available. Just as pleasing, because the sessions model is no longer used, I no longer have issues with my most recently used (MRU) file list. Under the sessions model the MRU is specific to each session defined. I have shared my solution in a dedicated mini how-to here at my web site. Bear in mind that my solution is, fundamentally, a work-around. With each updating of KDE the work-around has to be re-implemented. Yet that is a small price to pay for eliminating the flawed and broken sessions model of Kate. To avoid some of the hassles I have written a shell script to help automate reinstalling the older version of Kate. I admit my solution means forever using the older version of Kate, unless Kate in KDE 4.x is repaired. Yet in Windows I continue to use older software that “just works” and I seem to never miss the “latest and greatest.” I only wanted an MDI text editor with syntax highlighting and some nice tools and the older version of Kate provides that. Since modifying KDE in this manner I have found some peace of mind with using KDE. Now my only hope is that the KDE developers stay out of my way. I never have liked the all-in-one packaging model of KDE. Kate and other text editors should not be a part of the KDE-base package. The base package should include only the libraries and necessities for the desktop. Kate should be packaged separately and if that had been the common sense way I had received Kate, I never would have had to experience the frustrations of a broken Kate because I could have maintained and used the older version. The concept of free software is supposed to be about choice. When end-users who do not possess the high-end skills to develop and manipulate source code are left at the whims of self-serving developers, then there is no longer choice but the same model currently embraced by developers in the propriety world. End-users are left in the cold. Yes, I have the choice to not use the software, or the choice to find a willing developer to make changes, or to learn to program, but those choices, in the real world, are essentially not available to most end-users. Although overall I like KDE, in the instance of Kate the developers simply have screwed up and they were wrong to impose significant design changes that many people did not and still do not want. Do the KDE developers care? I don’t know but after six opportunities to remedy the problem I believe end-users can judge the fruit. Finis. |
||