hmm. actually, i don't see anything that critical in the linked page. actually, it even says :
"Editor's note: We had originally credited Paul Steed with starting this particular.plan war, but it seems we made a mistake. The original post was Myscha's. Sorry guys!"
for a new document you might try "text styles" - for most documents this might even be enough, and the list nicely fits in a small stylist window. there's also "automatic" mode, which seems to produce similarly short list, but i am unsure about criteria used for it:)
either simply drag it to the desired size (easier to do with oo.org window maximised), or hold down ctrl and doubleclick free space in stylist upper area (where style categories are located) - that will dock it in the last docked position.
i vaguely remember this being problematic in some cases - maybe that was with some widgetsets or something. you also might could try reversing ctrl holding.
in any case, trying the latest version is a good idea:)
Fix the style selection mechanism. I don't generally use around 100 styles in one document, and I don't need 15,746 different views of the styles. I just want a list of the dozen or so styles I actually care about.
well, there's "applied" view and "hierarchical" view, which both improve usability. though the bug where set view is forgotten after closing/opening stylist is annoying...
Provide commands to revert the formatting of selected objects/text to the default for the current character/paragraph/whatever style individually. The vague "Default" command on the menu is unhelpful.
well, that command should be named "default style" now;) in any case, it is a good idea to avoid direct formatting as much as possible.
lately there have been speed improvements regarding some spreadsheets.
if the document is not confidential, you could provide and i could test whether it is still problematic with latest dev snapshots (or maybe you can try that yourself)
why should a possibly future proof all-around office file format include such a tag _at all_ ?
why wouldn't it include flexible paragraph, table and other object layouting capabilities that are defined with generic values, allowing converting processes from these old formats to transfer the information hidden away in lineWrapLikeWord6 to generic formatting ?
that is only one of the many objections to this format, but it seems to make a point quite clearly - it's quality is not that good.
on the other hand, why was microsoft not participating in odf creation ? they could easily have put in there functionality they would have been interested in, if that was a sane solution, wouldn't they ?
on a more personal note, i'm unsure on how to view your position. at some point, i had an idea that you might be trying to act a as a vaccine for oss community, by bringing threats in the frontline faster and more agressively, thus forcing a creation of 'cure', but i'm not that sure anymore;)
i'm thankful for the involvement in things like midnight commander (though it could benefit from more active development...), but your latest actions seem to be too poisonous even for a vaccine. it's kinda telling that tag 'miguel' is slightly turning into a similar tag to 'dvorak'...;)
ufoai, widelands (though i can't get 11rc to run), openttd, warzone2100:)
most of these are remakes of the classics, but that's the great thing - those classics were awesome, but were abandoned or dragged into bad 3d mode. these remakes improve on those ideas and are really good.
yeah. and nvidia is fixing bugs only in the new drivers. woohoo. i am buying and recommending nvidia only still, but if ati would provide opensource drivers in kernel/xorg... that could change. not likely, given that we've heard all kinds of such vague promises before and nothing changed.
i'm not sure how those options are represented in the spec, but a lot of changes in oo.org have been done that do not include 'ww' or 'word' strings, but still are there for interoperability with old msoffice formats. something like "add spacing before thisandthat". would be interesting to find out how many of those are actually added only to workaround msword quirks...
How about not being able to recover in any obvious way if you rm -r a subtree instead of svn rm it?
i agree that local version control would ease some things considerably. but recovering ? what recovering ?
not worked much with svn, i decided to try that out. after removing a directory, 'svn status' showed that directory prefixed with '!' - probably meaning that i was missing it.
'svn up' nicely restored it, no complaints, no problems.
it's also bad enough that i still can't get full xen with vanilla kernel. yes, 2.6.23 will merge some things, but until i don't need additional patches to kernel (that are outdated and hard to obtain), i don't feel comfortable relying on sych a solution.
as for the management tools and overall openness, i can only agree. nowadays competition in many fields is only increasing, and projects/products have to compete for users. failing to do that successfully will push the project to some distant place while other projects overtake their once dominant position.
btw, for 'things to come' in a near future kernelnewbies changelogs were great. unfortunately, it seems that one for 2.6.23 is pretty much abandoned... http://kernelnewbies.org/LinuxChanges
actually, calculation does not take that much with modern cpus. what kills it, is a lot of small writes. more or less this means that most of these writes requires reading a bit of data from all of the drives. nasty. and if this continues for a prolonged period, once all caches are exhausted, it kinda deadlocks.
i would appreciate a common solutions in following areas : autoinstallation; centralised management; system configuration.
currently we have autoyast and kickstart, yast and a LOT of different tools across distros. and very few functional, userfriendly stuff for centralised management.
of course, all these things are opensource, but they are so much tailored to a single distro that porting them across involves quite a lot of work.
it is possible that desktop environments will get their act together faster than these projects, but that would be troublesome for servers where i don't want no damn desktop environment:)
what is a "native" app ? why doesn't oo.org qualify as a native one ?
and no, slowness on desktop is not only when running java apps or firefox. on a desktop with a couple of apps and a powerful machine this is not so noticeable, but interactivity with more applications open is not as good as i would like to.
additionally, single resource hungry process can make my mouse skip. arrrgh ! the joys of clicking in the wrong place. though i have heard this will be improved by moving xorg mouse control process in a separate thread or something:)
Slashdot Mirror
hmm. actually, i don't see anything that critical in the linked page. actually, it even says :
.plan war, but it seems we made a mistake. The original post was Myscha's. Sorry guys!"
"Editor's note: We had originally credited Paul Steed with starting this particular
for a new document you might try "text styles" - for most documents this might even be enough, and the list nicely fits in a small stylist window. :)
there's also "automatic" mode, which seems to produce similarly short list, but i am unsure about criteria used for it
either simply drag it to the desired size (easier to do with oo.org window maximised), or hold down ctrl and doubleclick free space in stylist upper area (where style categories are located) - that will dock it in the last docked position.
:)
i vaguely remember this being problematic in some cases - maybe that was with some widgetsets or something. you also might could try reversing ctrl holding.
in any case, trying the latest version is a good idea
styles panel (stylist) can be docked. i usually dock it at the right hand side. or have i misunderstood the problem ?
well, there's "applied" view and "hierarchical" view, which both improve usability.
though the bug where set view is forgotten after closing/opening stylist is annoying...
well, that command should be named "default style" now
in any case, it is a good idea to avoid direct formatting as much as possible.
lately there have been speed improvements regarding some spreadsheets.
if the document is not confidential, you could provide and i could test whether it is still problematic with latest dev snapshots (or maybe you can try that yourself)
hmm. actually, the device the article is about - Neo1973 - is pretty much open.
http://wiki.openmoko.org/wiki/Neo1973_Hardware
few things are closed for quite valid reasons, but hopefully with time this will have a chance to improve.
i had this on a laptop few years ago (it later regressed into "no sound until a suspend/resume cycle"
i could restore sound in both occasions by changing volume on master and pcm channels.
i'm too lazy to search for kernel bugzilla report
heh. try searchig for 'linux' on that site :)
i especially like the second product.
why should a possibly future proof all-around office file format include such a tag _at all_ ?
;)
;)
why wouldn't it include flexible paragraph, table and other object layouting capabilities that are defined with generic values, allowing converting processes from these old formats to transfer the information hidden away in lineWrapLikeWord6 to generic formatting ?
that is only one of the many objections to this format, but it seems to make a point quite clearly - it's quality is not that good.
on the other hand, why was microsoft not participating in odf creation ? they could easily have put in there functionality they would have been interested in, if that was a sane solution, wouldn't they ?
on a more personal note, i'm unsure on how to view your position. at some point, i had an idea that you might be trying to act a as a vaccine for oss community, by bringing threats in the frontline faster and more agressively, thus forcing a creation of 'cure', but i'm not that sure anymore
i'm thankful for the involvement in things like midnight commander (though it could benefit from more active development...), but your latest actions seem to be too poisonous even for a vaccine.
it's kinda telling that tag 'miguel' is slightly turning into a similar tag to 'dvorak'...
that's no bashing. can i get powerpoint alone ?
and, um, does it run on linux ?
ufoai, widelands (though i can't get 11rc to run), openttd, warzone2100 :)
most of these are remakes of the classics, but that's the great thing - those classics were awesome, but were abandoned or dragged into bad 3d mode. these remakes improve on those ideas and are really good.
yeah. and nvidia is fixing bugs only in the new drivers. woohoo.
i am buying and recommending nvidia only still, but if ati would provide opensource drivers in kernel/xorg... that could change. not likely, given that we've heard all kinds of such vague promises before and nothing changed.
i'm not sure how those options are represented in the spec, but a lot of changes in oo.org have been done that do not include 'ww' or 'word' strings, but still are there for interoperability with old msoffice formats. something like "add spacing before thisandthat".
would be interesting to find out how many of those are actually added only to workaround msword quirks...
heh. well, it's pretty hard to find out what she is trying to say, so let's try with subtitles - http://www.youtube.com/watch?v=WALIARHHLII
:D
nope. still can't grasp anything past "don't have maps"
i agree that local version control would ease some things considerably. but recovering ? what recovering ?
not worked much with svn, i decided to try that out. after removing a directory, 'svn status' showed that directory prefixed with '!' - probably meaning that i was missing it.
'svn up' nicely restored it, no complaints, no problems.
or have i misunderstood the problem ?
it's also bad enough that i still can't get full xen with vanilla kernel. yes, 2.6.23 will merge some things, but until i don't need additional patches to kernel (that are outdated and hard to obtain), i don't feel comfortable relying on sych a solution.
as for the management tools and overall openness, i can only agree. nowadays competition in many fields is only increasing, and projects/products have to compete for users. failing to do that successfully will push the project to some distant place while other projects overtake their once dominant position.
btw, for 'things to come' in a near future kernelnewbies changelogs were great. unfortunately, it seems that one for 2.6.23 is pretty much abandoned...
http://kernelnewbies.org/LinuxChanges
actually, calculation does not take that much with modern cpus. what kills it, is a lot of small writes.
more or less this means that most of these writes requires reading a bit of data from all of the drives. nasty.
and if this continues for a prolonged period, once all caches are exhausted, it kinda deadlocks.
one missing piece here is sdl - i'm not sure about all the modules, but there is for sure sdl_net.
actually (if i remember correctly), sled includes kde.
i really hope lenovo will support all packages coming with sled.
i would appreciate a common solutions in following areas :
:)
autoinstallation;
centralised management;
system configuration.
currently we have autoyast and kickstart, yast and a LOT of different tools across distros. and very few functional, userfriendly stuff for centralised management.
of course, all these things are opensource, but they are so much tailored to a single distro that porting them across involves quite a lot of work.
it is possible that desktop environments will get their act together faster than these projects, but that would be troublesome for servers where i don't want no damn desktop environment
i don't. i don't capitalise anything. ha !
what is a "native" app ? why doesn't oo.org qualify as a native one ?
:)
and no, slowness on desktop is not only when running java apps or firefox.
on a desktop with a couple of apps and a powerful machine this is not so noticeable, but interactivity with more applications open is not as good as i would like to.
additionally, single resource hungry process can make my mouse skip. arrrgh !
the joys of clicking in the wrong place. though i have heard this will be improved by moving xorg mouse control process in a separate thread or something
bleh. so many comments and nobody mentioned bad astronaut ?
:)
http://www.interpunk.com/item.cfm?Item=86168
cover picture is nice