[ Date Index ] [ Thread Index ] [ <= Previous by date / thread ] [ Next by date / thread => ]
Ben Goodger wrote: >> Check to see if your BIOS is also trying to set NumLock and then check >> the config of either X or your desktop manager (kdm, gdm or xdm). This >> isn't something that I've come across before but installing a new >> package is probably not the answer. > > The BIOS definitely leaves it on. Linux turns it off six or seven times > between GRUB and Metacity. That could be awkward to track down but I suspect it is a matter of configuration. It depends how much it annoys you but careful use of "grep" could be in order. >> 4. Why do Flash and Java plugins have such poor keyboard input? Is it so >> > difficult to make them accept it? (firefox) >> >> Possibly, yes. Again, Java is somewhat open but the JDK is still not >> under a free software licence and problems integrating Java into other >> systems are still down to Sun to solve - if they are interested in >> solving them. Free software developers simply do not have access to the >> code to solve the problems. > > But surely the mechanism to give them keyboard input is down to the plugin > system they are plugging into? Wrong direction. The plugin receives all input via the JDK. The "real" hardware sends interrupts to the OS. The OS sends messages to the open window. The window passes some messages to the JDK which passes the message onto the plugin. If an error occurs between the browser sending on the message and the plugin code receiving the message, the problem exists in the JDK. With free software browsers and plugins, it is relatively easy to check the values being passed and received. With proprietary plugins, nobody knows. >> That one is relatively easy. It is because Microsoft enter a private >> agreement with the monitor/graphics card manufacturer whereby the >> details of the hardware are declared to the MS developers. This allows >> the installer to have a reliable reference table, indexed on the >> machine-readable ID string for the device concerned. > > I thought this was a joke for a moment. But the tables are in the manual > for > the motherboard and all I need to do is enter the values, surely? Having > done that, X ought to work perfectly. If this was true, all hardware would work flawlessly with GNU/Linux from the moment of launch. Developers have access to these manuals and are free to enter the data - the fact that problems still occur must mean that either the data is incomplete or that the implementation behind the figures isn't discoverable. Some values are only true if the supplied drivers are used. >> Microsoft and Apple have non-disclosure agreements with the various >> manufacturers that allow their developers full access to these settings. >> >> Apple traditionally limited their risk by keeping their hardware >> specific to their systems. Microsoft obtain the data simply be being the >> monopoly player. Manufacturers know they won't sell their hardware >> without Windows operability. So Microsoft are the driving force behind >> the non-disclosure and Apple's previous usage of a limited range of >> hardware made their agreements less problematic. The core problem lies >> with the Microsoft monopoly - some manufacturers would be willing to >> offer their data to RedHat or Debian but are prevented by the agreements >> that Microsoft required to allow Windows support. > > > Oh, dear. I had no idea it was that bad... It's getting better in some areas, but worse in others. The battle continues.... > I do understand the reasons for proprietary software causing anger for > people trying to develop around it. However, saying "just don't use it" > isn't going to go down well to most people - if they can't access their > leet > flash sitez lol with linux, they'll simply stick to Windows. "Just don't use it" is a valid answer for some people (like me). If free software support for Flash was complete, I may have considered using Flash for some web content. As it is not, I decided not to make problems for myself. Avoiding problems is a rational solution for many people. > SVG and Javascript, although excellent, are not supported by IE, and until > they are not many people will use them for serious purposes or otherwise; That again depends - e.g. more rational Flash sites have non-flash alternative content. The same can be done for non-SVG content. > even ignoring their comparative lack of features, Flash is not going to be > replaced until IE supports SVG (or at least 70% marketshare of browsers do) > , and that means it's here to stay for the foreseeable future. True - however there is nothing to say that the proprietary implementation of Flash will remain static or predictable for the immediate future, let alone the foreseeable future. It makes it VERY hard to improve the free software flash support when the goalposts move with every release. Whether or not these changes are deliberately employed to crash non-Macromedia drivers, the result is the same. > Don't get me started on ATi, but the fact remains that proprietary software > is an unavoidable part of many people's computing, Our perceptions differ - I fail to see that Flash support is anything more than an annoyance. Doubtless others give it a higher priority (or we wouldn't have the support that is already available) but until more people want to see Flash support improved, things are likely to stay as they are. As with all voluntary effort, it is incumbent on those who care about an issue to do something towards the resolution of the issue because if you don't, who will? -- Neil Williams ============= http://www.data-freedom.org/ http://www.nosoftwarepatents.com/ http://www.linux.codehelp.co.uk/
Attachment:
signature.asc
Description: OpenPGP digital signature
-- The Mailing List for the Devon & Cornwall LUG http://mailman.dclug.org.uk/listinfo/list FAQ: http://www.dcglug.org.uk/linux_adm/list-faq.html