GNUstep I've posted my initial XInput patch to the list. You can check it there. Here's a silly screenshot.
A few of core files are compiled. The progress is slow but not that slow.
I'll put my modem in jail for a month to continue working on Maliwan and to make it compiled again. In case that I didn't come back, it means the project is dead.
2 new screenshots. A strange problem is, when the -drawRect has been invoked too frequentlly, back-art may be flickering. This is easily reproducable by setting the window to nonretained (and on my slow machine). Expensive view like clock can cause this during the animation (scaling the clock) and when adjusting the alarm arc.
See you next month!
I have no idea why I updated my system to Ubuntu while the only apps I am using these days are vim bash and Terminal.app. But that's how to manage boredom, I guess.
It is being discussed in GNUstep mailing list about using GNUspeech which was given to GNU long time ago to implement NSSpeechSynthesis. I went through the list and found this message. I have a feeling like, from its maintainer's attitude, GNUspeech (which is in Objective-C) should be forked and maintained by GNUstep developers instead. Although it would be nice if we can provided C API wrapper for the rest of the world.
Years ago I've painted stuffs for Kernelnewbies.org. Those are Penny and some t-shirt designs (they never made those shirts though, AFAIK). I do not interested in reading any kernel source code anymore (unless someone is insane enough to write kernel threads in Objective-C and tie DO to l4 messaging) and no longer staying in the channel. I think I am one of the reason why riel started the channel since I have been kicked around in #linux somewhere on MIT irc server for loading the channel with silly questions :( (Alan Cox and Riel are so nice to answer them) Hmm, I am missing Uncle Snafu.
Just a side note. I'm wondering if this is a good thing to tie desktop notification system to NS. I think it is better to try to extend NET_WM et al. in WMaker to manage stuffs and I don't think it is too ideal to use only X communication. There are many applications outside GNUstep world that should rely on these desktop managements and informations. Though it could be good in short term and that may worth it. But I still think it is better to focus on other things like unify configuration and decorations, eg. replacing WINGs with GNUstep (although a basic layer for importing and exporting configurations should be enough and simply use X messaging channels that already provided by WMaker, those that WPrefs is using). Thus, using NSWindow to do window decoration will be expensive because it will allocate memory in the backend for each and most of the time most part of it isn't visible since covered by the app. A custom window class mixing both types of windows should solve the problem though.
I'm prototyping a collection factory class. When factories form up a network, a virtual factory will be created to manage each network's production. This virtual factory can be mixed into other network and act like a traditional factory. Another thing is Pipe class will be rewritten to base on target/selector model. Now I don't know if I can finish all these by this year.
Just finished 2 movies. Waterboys and The Hole.
Well.. I wish I can be your Beagle 2, better than dogs.. and er.. from Mars.
I'm thinking of using xcf2 as a default icon format in garma. This will allow me to freely manipulate my icons on the fly like copositing and mixing a few pixmaps to construct an icon or mix two default icons to make another one, ie. a blank folder and images of video, documents, photos, etc. (May be I need another IconKit to wrap this up? If so I'll just wrap it around Maliwan). I am also thinking of drafting my own icon specs. I don't like a few things in Quentin's draft. Though, may be I'll just accept the draft if it can state it clear about the geometry of the 3d icons.
My iBook won't sleep when I close it. I don't know why but I like that, so now I can use it to warm me up in these winter nights.
Boring boring boring boring boring boring...
GNUstep Icon Spec.
The recent draft of the icon spec wants a square clickable area around the visible part of the icon. Which I don't agree with. This is confusing when a user wants to specify a region that select a few icons by dragging. A user may start selecting and accidently clicking into an invisible area around an icon she didn't want to select and then dragging it instead of doing range selecting she was expecting. My idea is, icon should be able to provide another layer specifying the clickable area, A single bit bitmap mask. Without this layer, App should extract the information from the alpha value.
ªÕÇÔµ¼ÁäÃéá¡è¹ÊÒÃµÑé§áµèà´ç¡ ¼ÁàµÔºâµÁÒã¹¤ÃÍº¤ÃÑÇªÑé¹¡ÅÒ§ ¾èÍáÁèà»ç¹¢éÒÃÒª¡ÒÃ ÇÑ¹æ¼ÁàÍÒáµèà·ÕèÂÇàµÃè ¡Ô¹àËÅéÒ àÁÒÂÒä»ÇÑ¹æ ¼Á¾ºÇèÒ ¼Ùé¤¹ÃÍºµÑÇ¼ÁÅéÇ¹áµè¾ÂÒÂÒÁ·Ó§Ò¹Ë¹Ñ¡ ¶éÒäÁèà¾×èÍª×èÍàÊÕÂ§à§Ô¹·Í§ ¡çà¾×èÍà«ç¡Êì ËÃ×ÍäÁè¡çà¾×èÍ¤ÇÒÁÃÑ¡
áµèËÅÑ§¨Ò¡·Õè¼Áä´éÃÙé¨Ñ¡¡ÑºÊµÍÅáÁ¹ ¼Á¨Ö§ä´éÊÓàË¹ÕÂ¡ÇèÒ¤ÓµÍº¢Í§ªÕÇÔµ·Õè¼Á à¤ÂáÊÇ§ËÒ äÁèä´éÁÒ¨Ò¡à§Ô¹·Í§ËÃ×Í¤ÇÒÁÊÓàÃç¨ã¹ªÕÇÔµ áµèà»ç¹¤ÇÒÁÊÓàÃç¨ã¹´éÒ¹¨Ôµã¨ã¹ÃÐ´ÑºªØÁª¹·Õè¼ÁãªéªÕÇÔµ ªØÁª¹·Õè·Ø¡¤¹ ÁÕ¤ÇÒÁàª×èÍÁÑ¹ã¹ÈÑ¡ÂÀÒ¾¢Í§µ¹àÍ§ ·Õè¨Ð¾Ñ²¹ÒºéÒ¹àÁ×Í§ ´éÇÂÈÃÑ·¸Òã¹¤ÇÒÁ´Õ ´éÇÂËÔÃÔâÍµÑ»»Ð ´éÇÂµÑµµÃÁÑª¬ÑµµµÒ ´éÇÂÇÔÃÔÂÐµèÒ§ËÒ¡ ¤ÇÒÁàª×èÍ·ÕèÁÕµèÍ¤ÇÒÁ´Õ áÅÐÈÑ¡´ÔìÈÃÕã¹¤ÇÒÁà»ç¹Á¹ØÉÂì ·Õè¡Ñ»µÑ¹ä¤àÍ´Ðà¤Â¾Ù´¶Ö§ àµÔÁàµçÁªÕÇÔµ¼Áä´éâ´ÂäÁèµéÍ§ áÊÇ§ËÒÊÔè§·ÕèÁÍ§äÁèàËç¹ áÅÐÅéÇ¹áµèµÔ´¤ÇÍ¹µÑÁàÊµ· ¤Ø³àÍ§¡çÊÑÁ¼ÑÊ¡ÒÃàµÔÁàµçÁ ·Õè¼Á¾Ù´¶Ö§ä´éàªè¹¡Ñ¹ ¤ÇÒÁ´Õ§ÒÁÊÇèÒ§äÊÇá¡èã¨µ¹àÍ§ áÅÐà»ç¹ÊÀÒ¾¸ÃÃÁ·ÕèÊÑÁ¼ÑÊä´é â´ÂäÁèµéÍ§ÃéÍ§¢ÍµèÍÈÒÅáµèÍÂèÒ§ã´ à¾ÕÂ§áµè¤Ø³àª×èÍÁÑè¹ã¹¤ÇÒÁ¨ÃÔ§ ã¹àËµØáÅÐ¼Å áÅÐã¹ÈÑ¡ÂÀÒ¾¢Í§¤Ø³à·èÒ¹Ñé¹
New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.
Keep up with the latest Advogato features by reading the Advogato status blog.
If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!