Re:Will anyone gain anything from this? Not Linux
on
The End is Nigh for XP
·
· Score: 1
hey. is there a filter for "suse" ?:) i meant to say "suse is using something similar".
at least latest incarnations of opensuse, as i have little experience with older versions.
Re:Will anyone gain anything from this? Not Linux
on
The End is Nigh for XP
·
· Score: 1
there's bdiff (binary diff) and different implementations of it, that could be used to distribute only the changed part. actually, is using something similar, so updates in there are incremental.
as for openoffice.org, i've heard that the binary form of the software during two releases changes enough to make patch ineffective - it would be too close to a full download. don't know how true has this been for all of the releases:)
actually, no. they would have to prove that you broke it. of course, breaking off hw seals from a laptop would be a proof of breaking things, but, for example, you could install non-hp ram in a hp server, and keep warranty on all other parts. hp would void warranty only if the damage was caused by the 'alien' ram module (though we never got to such an actual situation and don't know how would proving that look like;) ). in case you are wandering, we got a written answer from local hp reps confirming this;)
now, this case here... heya, hp. i suggest your sales people to hid your support people in the head, as the sales people will have harder times ahead. oh, and you can ask pr people to join you. see - no expenses, a lot of free pr !!!~`111~! a very bad one, but hey...
good point, but there also have been some replies that covered my views a bit already (i'm not disagreeing with you, but i still despise showing off;) )
basically, of course, there is no strict limit (comparing a ring with a house does not seem practical) - so it always would be a bit subjective.
one factor would be "how much would i get if i tried to sell it" - difference would show how large a gap would be. note that it would be pretty huge to show a sign of stupidity - new car loses quite a lot of it's initial value after the first purchase. of course, then there are people who buy sweaty actors' suits, so this isn't exactly the best measure...
other could be "functionally, in durability and by other factors, what would an identical solution cost ?" or "how much does the device costs to make, and how much am i paying for the shiny status ?".
i remember a story about two stores in moscow (probably the story was exaggerated, but it kinda shows the point;) ) - the stores were one across the other, one was for common folks, other - for the new russians. you could get slippers in the first for 1 rouble, in the second - for 300. they were identical.
as for me... i can't think of anything that would have provided me value for being pricey. of course, i have spent money on non-essential things (ice cream !;) playing cards, wireless router), but all of the purchases have had a value of their own, and most of them have a decent resell price (well, except the ice cream).
i'd like to remind that you left out any overhead. depending on protocol, link quality, type of data transferred, it can get pretty high. i guess with wireless you have to account for somewhat higher latency and fluctuations in both signal strength and quality, so guesstimate the overhead on retransmitting even a bit more.
i haven't tested wireless speed and don't have access to any equipment right now, but it would be interesting to test different adapters, routers, protocols and use cases.
did the joke fly over my head ? hard to tell... leaving no place for overhead, but mentioning it, mixing up bits & bytes in every sentence... nah, this gotta be a good troll;)
senior attorney on Microsoft's internet safety enforcement team
that's one bloody scary profession... probably whole "internet safety enforcement team" is pretty well equipped with weapons and bulletproof jackets, going around and shouting all kinds of legalese.
especially check sections with "Compatibility Note". a couple of things from that document:
Several sections require the implementor to clone the behaviour of a proprietary product, where the behaviour to clone is not specified. For example: Section 2.15.3.6 page 2161, autoSpaceLikeWord95....
Compatibility Note Attributes like these have no place in an international standard, and are not needed for compatibility with existing documents. The correct way to achieve compatibility is through generic tags. For example: autoSpaceLikeWord95 should be replaced by a generic character-spacing attribute that takes a numeric value or set of numeric values.
note that most of the things listed in compatibility notes are done by opendocument, so it is not only capable of dealing with backwards compatibility issues, but it is much better prepared to do that and makes application developers' life so much easier.
please, read through http://www.grokdoc.net/index.php/EOOXML_objections. pay special attention to sections with "Compatibility Note". i hope after those documents you'll see what incredible low quality and intentional drawbacks the "format" has.
i acknowledge the problem of modified photos, but i can't think of a solution to this... i mean, drm-like solutions will fail (i hope we don't have to repeat why:) ). then there's a possible software that tests for image details that might be altered - but that's an uphill battle as it would be just a matter of re-running this software until all the supicious places are worked out.
so is there a way to actually detect that image is authentic ? maybe only by declaring procedures to prove this, which would include taking and processing photo in a controlled manner with a couple of witnesses...
odf actually is very much like that. if you unzip the file (as suggested here several times already;) ), you will see that content is quite separated from formatting. of course, hordes of direct formatting loving droids have a hard time grasping the concept of styles, but that's mostly they have never bothered to find out - most can handle styles quite well after just a short introduction, so i'd guess it's more about finding out about the concept than having unusable guis (btw, openoffice.org is sticking styles in the nose quite hard).
and dell will never sell preloaded ms windows then. somehow i doubt the quality of individual customer support or the operating system matters to dell that much.
on the other hand, distro vendor has to fight for the market share, thus their support is usually visibly better than dell or microsoft has ever provided.
on the other hand, there are cases when dell would get blamed heavily, but not because of a crappy distro support or quality - because of crappy or poorly supported hardware. problems with various bioses, firmwares and hw in general might be hidden or worked around with windows drivers, but those stick out much better when using linux systems.
since you asked... i personally would prefer a good quality ogg. flacs might be lossless, but i don't have equipment or hearing to distinguish a flac from 192 vbr ogg or something:)
of course, providing both could please also audiophiles, while allowing to keep diskspace requirement and download time lower.
There's unlimited supply and there is no reason why I tell you it was all a frame they only did it 'cos of fame - Who? EMI
Too many people had the suss Too many people support us An unlimited amount too many outlets in and out - Who? EMI
And sir and friends are crucified a day they wished that we had died We are an addition we are ruled by-none Never ever never
And you thought that we were faking that we were all just money making you do not believe we're for real or you would lose your cheap appeal?
Don't judge a book just by the cover Unless you cover just another And blind acceptance is a sign of stupid fools who stand in line like EMI
Unlimited edition with an unlimited supply That was the only reason we all had to say goodbye
Unlimited supply EMI there is no reason why EMI I tell you if was all a frame EMI they only did it 'cos of fame EMI I do not need the pressure EMI I can't stand the useless fools EMI unlimited supply EMI Hallo EMl goodbye A & M
(though looking at the second one it seems somewhat qeustionable;) ).
this alone is far from "a lot", but given amarok's overall market share and publicity (relatively low), i am sure there are more (which would constitute a lot in the end;) ). i must admit, i am not bothered enough to search, though...
there are a lot of other online stores that not only sell non-drm, music they also provide oggs, flacs, ability to re-download...
actually, one of them - magnatune - was recently integrated in amarok. after that, there have been several interested parties in such integration, all of them providing unencumbered tracks in free formats. an amarok developer is working on a generic music store interface, so it should be even easier to purchase such music and increase music available. of course, this unified interface is still some time away, but until that everybody can go to magnatune or any other shop that is not only selling non-drm stuff, but also providing it in open an high-quality formats:)
actually, there is a solution to this, that seems to be the best to me (though given current situation, it might be slightly utopian...).
dell could start by selecting only components that are supported by vanilla linux kernel or other opensource projects (like sane, cups etc). given their size, it shouldn't be too hard for them to convince single manufacturer in each are to get the drivers in a decent state (which is easier than before with the new linux kernel drivers initiative). this usually would require little more than releasing specifications. for faster results, reference hw could be provided to some key driver developers (that's usually cheap enough for the vendor:) ).
in such a case they could put _any_ distribution they want, and also advertise their computers as fully 'linux compliant'. consumers could just select whatever specification they want, without researching every single component for driver compatibility.
as for preinstalling - if full hw compatibility is achieved, that is much smaller concern, as now they can choose any one single distribution or several biggest ones, and preinstall those. that decision matters much less, though i believe slashdot would have a lengthy discussion on that;)
the first thing i thought about was some 'racing' or 'competition', as in "do you care a lot about being competitive in games". what's next ? "do you care about age of the character you are playing ?" - "do you care about sexual preferences of the character you are playing ?".
well, you are not supposed to change kernel just for the fun of it unless you know what you are doing (and this involves being informed about kernel changes:) ). most distributions don't just upgrade kernel, usually it's one kernel for the whole lifespan of a particular version (and this can be quite a long time with commercial 'enterprise' distros).
now, supporting distributions... on one hand, i can understand vendors - it is much easier to pick up to three defined environments and only care about single parameter -"what distribution is it". on the other hand, that promotes sloppy practices, inflexible solutions and (proprietary) functionality duplication and fragmentation, which makes life harder for their users.
for example, see hp asm package. it's a proprietary package that is supported only on few distributions, it has it's own semantics and interfaces... going lm_sensors way would make their solutions more flexible and easier to use. of course, there might be things that they either had to either develop inhouse or attempt to standardise, but this would both ease life for their users and reduce amount of work they must perform. summary - generic solutions and software should be the way to go:)
Slashdot Mirror
hey. is there a filter for "suse" ? :)
i meant to say "suse is using something similar".
at least latest incarnations of opensuse, as i have little experience with older versions.
there's bdiff (binary diff) and different implementations of it, that could be used to distribute only the changed part.
:)
actually, is using something similar, so updates in there are incremental.
as for openoffice.org, i've heard that the binary form of the software during two releases changes enough to make patch ineffective - it would be too close to a full download. don't know how true has this been for all of the releases
and, if normally some of the students would have seen the page, now whole world will see it. nicely done, eric !
openoffice.org is capable of using 3 widget sets - it's own vcl, gtk and qt
actually, no. they would have to prove that you broke it. ;) ). ;)
of course, breaking off hw seals from a laptop would be a proof of breaking things, but, for example, you could install non-hp ram in a hp server, and keep warranty on all other parts. hp would void warranty only if the damage was caused by the 'alien' ram module (though we never got to such an actual situation and don't know how would proving that look like
in case you are wandering, we got a written answer from local hp reps confirming this
now, this case here... heya, hp. i suggest your sales people to hid your support people in the head, as the sales people will have harder times ahead. oh, and you can ask pr people to join you. see - no expenses, a lot of free pr !!!~`111~!
a very bad one, but hey...
i hope you have taken time to complain to the vendor ;)
even if that's an old device, even if it is obscure - that is the only way to get a sane support in future...
good point, but there also have been some replies that covered my views a bit already (i'm not disagreeing with you, but i still despise showing off ;) )
;) ) - the stores were one across the other, one was for common folks, other - for the new russians.
;) playing cards, wireless router), but all of the purchases have had a value of their own, and most of them have a decent resell price (well, except the ice cream).
basically, of course, there is no strict limit (comparing a ring with a house does not seem practical) - so it always would be a bit subjective.
one factor would be "how much would i get if i tried to sell it" - difference would show how large a gap would be. note that it would be pretty huge to show a sign of stupidity - new car loses quite a lot of it's initial value after the first purchase.
of course, then there are people who buy sweaty actors' suits, so this isn't exactly the best measure...
other could be "functionally, in durability and by other factors, what would an identical solution cost ?" or "how much does the device costs to make, and how much am i paying for the shiny status ?".
i remember a story about two stores in moscow (probably the story was exaggerated, but it kinda shows the point
you could get slippers in the first for 1 rouble, in the second - for 300. they were identical.
as for me... i can't think of anything that would have provided me value for being pricey. of course, i have spent money on non-essential things (ice cream !
does that count as a sign of corruption of this world, if i can imagine actually somebody buying this thing ?
i'd like to remind that you left out any overhead. depending on protocol, link quality, type of data transferred, it can get pretty high.
i guess with wireless you have to account for somewhat higher latency and fluctuations in both signal strength and quality, so guesstimate the overhead on retransmitting even a bit more.
i haven't tested wireless speed and don't have access to any equipment right now, but it would be interesting to test different adapters, routers, protocols and use cases.
did the joke fly over my head ? hard to tell... ;)
leaving no place for overhead, but mentioning it, mixing up bits & bytes in every sentence... nah, this gotta be a good troll
that's one bloody scary profession...
probably whole "internet safety enforcement team" is pretty well equipped with weapons and bulletproof jackets, going around and shouting all kinds of legalese.
if http://www.grokdoc.net/index.php/EOOXML_objections is a variation of "Micro$oft is teh SUX0R", it's a pretty lenghty and detailed variation of that one...
especially check sections with "Compatibility Note".
a couple of things from that document
note that most of the things listed in compatibility notes are done by opendocument, so it is not only capable of dealing with backwards compatibility issues, but it is much better prepared to do that and makes application developers' life so much easier.
even with giving you a benefit of doubt i'd like to call the argument about backwards compatibility a bullshit.
s . pay special attention to sections with "Compatibility Note".
some interesting linked information is here :
http://www.sutor.com/newsite/blog-open/?p=1346
please, read through http://www.grokdoc.net/index.php/EOOXML_objection
i hope after those documents you'll see what incredible low quality and intentional drawbacks the "format" has.
...and this will be cracked in a couple of weeks.
:) ).
i acknowledge the problem of modified photos, but i can't think of a solution to this... i mean, drm-like solutions will fail (i hope we don't have to repeat why
then there's a possible software that tests for image details that might be altered - but that's an uphill battle as it would be just a matter of re-running this software until all the supicious places are worked out.
so is there a way to actually detect that image is authentic ?
maybe only by declaring procedures to prove this, which would include taking and processing photo in a controlled manner with a couple of witnesses...
odf actually is very much like that. if you unzip the file (as suggested here several times already ;) ), you will see that content is quite separated from formatting.
of course, hordes of direct formatting loving droids have a hard time grasping the concept of styles, but that's mostly they have never bothered to find out - most can handle styles quite well after just a short introduction, so i'd guess it's more about finding out about the concept than having unusable guis (btw, openoffice.org is sticking styles in the nose quite hard).
and dell will never sell preloaded ms windows then.
somehow i doubt the quality of individual customer support or the operating system matters to dell that much.
on the other hand, distro vendor has to fight for the market share, thus their support is usually visibly better than dell or microsoft has ever provided.
on the other hand, there are cases when dell would get blamed heavily, but not because of a crappy distro support or quality - because of crappy or poorly supported hardware.
problems with various bioses, firmwares and hw in general might be hidden or worked around with windows drivers, but those stick out much better when using linux systems.
since you asked... i personally would prefer a good quality ogg. :)
flacs might be lossless, but i don't have equipment or hearing to distinguish a flac from 192 vbr ogg or something
of course, providing both could please also audiophiles, while allowing to keep diskspace requirement and download time lower.
The Sex Pistols - EMI
There's unlimited supply
and there is no reason why
I tell you it was all a frame
they only did it 'cos of fame -
Who? EMI
Too many people had the suss
Too many people support us
An unlimited amount
too many outlets in and out -
Who? EMI
And sir and friends are crucified
a day they wished that we had died
We are an addition
we are ruled by-none
Never ever never
And you thought that we were faking
that we were all just money making
you do not believe we're for real
or you would lose your cheap appeal?
Don't judge a book just by the cover
Unless you cover just another
And blind acceptance is a sign
of stupid fools who stand in line like EMI
Unlimited edition
with an unlimited supply
That was the only reason
we all had to say goodbye
Unlimited supply
EMI there is no reason why
EMI I tell you if was all a frame
EMI they only did it 'cos of fame
EMI I do not need the pressure
EMI I can't stand the useless fools
EMI unlimited supply
EMI Hallo EMl goodbye A & M
just on amarok mailing list possibilities of further integration have been asked about by representatives of two of such music stores :
;) ).
;) ).
http://www.ind-music.com/
http://www.mp3ninja.com/
(though looking at the second one it seems somewhat qeustionable
this alone is far from "a lot", but given amarok's overall market share and publicity (relatively low), i am sure there are more (which would constitute a lot in the end
i must admit, i am not bothered enough to search, though...
there are a lot of other online stores that not only sell non-drm, music they also provide oggs, flacs, ability to re-download...
:)
actually, one of them - magnatune - was recently integrated in amarok.
after that, there have been several interested parties in such integration, all of them providing unencumbered tracks in free formats.
an amarok developer is working on a generic music store interface, so it should be even easier to purchase such music and increase music available.
of course, this unified interface is still some time away, but until that everybody can go to magnatune or any other shop that is not only selling non-drm stuff, but also providing it in open an high-quality formats
actually, there is a solution to this, that seems to be the best to me (though given current situation, it might be slightly utopian...).
:) ).
;)
dell could start by selecting only components that are supported by vanilla linux kernel or other opensource projects (like sane, cups etc).
given their size, it shouldn't be too hard for them to convince single manufacturer in each are to get the drivers in a decent state (which is easier than before with the new linux kernel drivers initiative). this usually would require little more than releasing specifications. for faster results, reference hw could be provided to some key driver developers (that's usually cheap enough for the vendor
in such a case they could put _any_ distribution they want, and also advertise their computers as fully 'linux compliant'. consumers could just select whatever specification they want, without researching every single component for driver compatibility.
as for preinstalling - if full hw compatibility is achieved, that is much smaller concern, as now they can choose any one single distribution or several biggest ones, and preinstall those. that decision matters much less, though i believe slashdot would have a lengthy discussion on that
remove the slash from the end of that link, it nicely messes things up
the first thing i thought about was some 'racing' or 'competition', as in "do you care a lot about being competitive in games".
what's next ? "do you care about age of the character you are playing ?" - "do you care about sexual preferences of the character you are playing ?".
well, you are not supposed to change kernel just for the fun of it unless you know what you are doing (and this involves being informed about kernel changes :) ).
:)
most distributions don't just upgrade kernel, usually it's one kernel for the whole lifespan of a particular version (and this can be quite a long time with commercial 'enterprise' distros).
now, supporting distributions... on one hand, i can understand vendors - it is much easier to pick up to three defined environments and only care about single parameter -"what distribution is it". on the other hand, that promotes sloppy practices, inflexible solutions and (proprietary) functionality duplication and fragmentation, which makes life harder for their users.
for example, see hp asm package. it's a proprietary package that is supported only on few distributions, it has it's own semantics and interfaces... going lm_sensors way would make their solutions more flexible and easier to use. of course, there might be things that they either had to either develop inhouse or attempt to standardise, but this would both ease life for their users and reduce amount of work they must perform.
summary - generic solutions and software should be the way to go