Unfortunately, that means you don't get H.264 on Linux as its a proprietary codec that requires some form of (paid) licencing.
I mean, Firefox doesn't support H.264, but Microsoft will happily provide you with the capability of playing H.264 in firefox using a driver that leverages the OS capability... as long as you're running it on Windows.
I think you're partly right though, all the codecs should be implemented as drivers (or similar) and then you are technically using the OS-provided capability, once the correct codec is installed. But its not like the OS is providing the drivers directly, you'll haver to go get them from somewhere. As WebM is free, codecs for it will be freely available for all OSs.
I guess the problem comes for those OSs that are locked down, but then you'er always on to a loser - if Apple only supports H.264 on iPhone and Microsoft only supports (say) H.265 on WP7, and neither allows you to upgrade the video support, then you will never get a video to play universally.
At least there's no excuse for not supporting WebM by all manufacturers, and any who try to give one will quickly be found out by consumers.
As an analogy - look at the non-free 'internets', Microsoft tried to lock you into MSN, and AOL tried similarly. Look where they are now.
probably because most of the SoCs come with telephony chipsets baked in. The manufacturer could disable them, but considering the capability is there "for free", they might as well enable it and offer it as an added extra feature.
You don't have to use it you know, get a emergency-call-only sim (ie a PAYG sim with no credit) and use that.
its not so much the 'dual core-ness' of it but the general increase in power.
ARM says their next CPU design is going to be called the A15, not the A10 as you'd expect. This because, although ARM designs have incremental performance, the next version is so good that they decided to skip A10 through A14 tags.
The A15 BTW, has speeds upto 2.5Ghz, up to 8 cores, and virtualisation instructions. Although you might need a laptop battery to power the top-end version, the fact that there is a top-end spec suggests good stuff to come, even if the smartphones have only 2 cores throttled down to 1.5Ghz.
Phones are going to be the new PCs, I'm sure of it. We're still in the days of the 386 at the moment, it won't take long to revolutionise our lives.
To be fair to some of them, they allow plugins, some of which provide ones to display H.264,
now what will probably happen is Microsoft will bundle a H.264 viewer and automatically install it into your firefox directories to enhance your browsing experience with the "Silverlight+H.264 video" plugin as part of a Windows Update.
Personally, I don't think its going to be properly resolved until Facebook decides to back one codec or the other:)
The hyperlinks in Microsoft's blog post lead readers to data indicating that two-thirds of Web videos are using H.264, with about another 25% using Flash VP6
yes, but once Google updates Youtube to only use WebM, I guess that'll show 91% of all online video to be in WebM format.
alternative to the PC desktop. You're thinking in terms of how things were, not how things will be.
Look at those smartphones - not one uses Windows Desktop. Even MS says they're the future of computing devices, which is why they're so desperate to get market share.
Legacy apps - sure these will be around for a while, but will increasingly become virtualised (ie run on a server and accessed remotely via Citrix/RDP/screen scrapers even - just like the old mainframe apps that I used to write scrapers for so they could appear as nice Windows apps whilst still running unchanged). Even Exchange could be replaced with Google apps for Enterprise. That might be relatively primitive compared to the full exchange ecosystem, btu how many users actually uses all that Exchange-provided functionality, and do you really think Google is going to sit back and say "job done" with what they provide, or are they going to continually improve their offerings to the point where its a serious alternative. When, not if, do you think that will happen?
There will be years of support, but you're missing the point - there's still support behind COBOL apps and AS/400 servers, but the focus of new dev isn't with those systems is it. Pretty soon (if not already) the focus of new apps will be with more 'mobile' systems, and Windows will be considered a legacy system in itself.
So, you stick your head in the sand, stick your fingers in your ears, and keep thinking that nobody ever gets fired for buying Microsoft (you might have heard that phrase in the past)
cue comments regarding my goldfish being a better CEO than Ballmer.... no, what this means is that the other executives in their playroom don't have anyone with "I'm next" on their hat, so that (like my goldfish really) means they won't be able to decide who to choose as a replacement, and so Ballmer stays.
Of course, anyone who likes Android, Apple or well, anything that isn't Microsoft, should be overjoyed at this news. MS is a dead company, but like the proverbial dinosaur the neural impulses havn't travelled all the way from the tail to the brain yet - MS, being the dinosaur that it is, doesn't yet realise its day is over.
Mind you, one thing that comes to mind is the quality of Bob Muglia. The man that once said to the Seattle Times, with a straight face, "We're going to increase quality in Windows Vista by firing all our quality assurance people!"
Also, it appears this has been brewing for a few months. From minimsft blog: Somewhat trustworthy source: All BobMu's org's 10s will not see another review cycle. Terminated in groups of under 500 each month to avoid Warn. Can anyone confirm?
Friday, October 22, 2010 7:54:00 PM
Chances are the next CEO will be Kevin Turner, a bean counter like none other (apparently). Just the kind of guy to 'turn round' Microsoft's fortunes by basically reducing the number of beans to count. Or, as someone said on that msft blog: KT CEO - that will be a nice touch indeed - walmarting of Microsoft-- how appropriate
we do this - there's a spreadsheet (well, got to have something right) that gets updated whenever anyone dials in to a customer site, even if you did nothing. If you do something when dialled in, that gets logged too.
I don't know if failing to update it is a pink-slippable offence, but you will get a severe b*ll*cking if you fail to do it twice.
BTW, our customers are police, fire and ambulance control centres. Maybe that makes us different to the usual, but it's simple and works well. If we could get rid of the spreadsheet that's been used forever and I'm sure still is simply because "that's the way we've always done it" then it might be less of a chore. Does anyone have suggestions for a replacement?
The trick, BTW, is to start doing it and keep doing it. After a short period of grumbling about the 'overhead' people will get used to it, and like all good ideas, will accept it once their inertia is overcome.
no, but you'll be laughed at for having a single shared account that means anyone who logs in to perform some "support activity" (maybe after a few drinks, or just general brainfarts) cannot be determined after the event. This can be a good thing, depending on how bad your admins are (good for them, that is).:)
I guess the majority of it is either licensing deals (eg Microsoft paying to be allowed to put Facebook linkage in their phone 'social network hub'), or revenues from apps like Farmville that might pay a percentage of their revenues (farmville apparently makes $1m per day in revenue).
I can't think what else - maybe there's a lot more money to be made from selling your 'private' details than I thought possible!
Have you tried? Its one of those 'silent' technologies where a site allows OpenID auth and make no big fuss over it.
Google and Facebook are OpenID providers as well as clients. Client sites include Sourceforge, the telegraph newspaper, StackOverflow, and many others. Look out for the little logo next time you go to sign in somewhere.
Of course, some sites (like yahoo) are openID providers so you can use your yahoo id as an openid id, but do not let other openid ids login to their sites.
Its called OpenID and its a standard. Many sites use it, not only as a 'client' where you use your OpenID id to login, but also some sites use it as a 'server' too, where your account with them can be presented as an OpenID id to other sites. Obviously, some sites do it better than others - MyOpenID or Verisign for example has good security implementation.
Facebook is just another one of these sites, Facebook Connect apparently is the API Facebook provides to other sites to allow them to authenticate with your Facebook id.
Personally, I think if you're going to use a central account to use, you get it from a more reputable company that is more about authentication and less about content.
The biggest issue with merging is the "tree conflict" where a directory list is modified. You don't need to treat a directory the same as file as its just a list of files - and you can determine that from the change in the revision (as files in that directory that are modified, added or deleted are recorded, which is exactly what you'd get if you stored directories the same as files).
However, because you cannot tell if a file has been moved, you have to delete it and then add the 'new' file, and this can cause problems, especially if the new files are further modified. If you're the only one working on this directory then there's no problem, but if you rename a file that your colleague has modified... the merge will find that it is trying to update changes into a file that appears to have been deleted, losing those changes. Hence the notification that human intervention is required. This notification isn't so easy to interpret though so perhaps the way merge errors are handled could be improved.
Other systems do track renames so this relatively common issue isn't seen, however, if you work in Windows and rename the file (and modify it) in explorer and then try to commit, even git or mercurial will have difficulty determining that its the same file and not a real delete and add.
possibly not, Symbian devs have worked with C for years on all sorts of chipsets. They squealed more about rewriting their code in Java than cross-chip compiling.
Most Androids come on a few chip architectures anyway (or are they all ARM based?), and the API provided by Google would be standard so I can't see much problem with cross-compiling for all known Android devices. Hell, you could set up a build server in the app store so when you uploaded your code, it would auto-compile for all devices Google knows of. And it would optimise each for the particular platform - that would be a good option, IMHO.
People are writing code for Android using C++ and Qt, so they must be handling the cross-platform problem somehow.
compared to whatever else they'd put in - the problem that the short-term view of stupid researcher fails to spot, is that if you replaced Google with anything else, the spammers and SEOers would quickly learn to game those results too, so you'd end up with exactly what you had before.
Good for Google to start spotting these abuses, I wouldn't worry about losing the adsense results from the first page, as the first page will end up replaced with good results which are still full of adsense links.
oops - typo. "Sparse checkout". You can checkout a partial directory structure, update and commit as required.
So if you have a 12Gb source tree stored in your repo, you don't need to get it all out - only the parts you want to work on. Typically you checkout the root item only, then update the required directories to fetch them into your local drive as you need them.
The ultimate issue is that renames (or moves) are implemented as delete+addition operations. Maybe back in the day, that appeared to be ok, but now its obvious it's a large failing. That's the reason you have problems with merging - merging is fine for changes to existing files, its merging new files where you get the problem. (look up 'tree conflicts' in svn for more info).
The rest of the system works well (though there's still a lot of svn haters for the usual reasons - they have something 'new' or 'cool', they just hate it because its mature and established, they hate it because they last tried it when it was version 1.2 and still think its has those limited features), and there is excellent tooling.
Git and mercurial lose some features that enterprises like, spare checkouts for example is a killer feature for enterprises that don't work well with DVCSs simply because of their original design.
BTW, I like the svn branching system, I've used a lot of scms in my time and svn is the only one that really shows you all the branches easily laid out for you to view - no stupid layers of branch views over the core code. You do have to manage it a little then, and there are features in svn like being able to branch at any point that I think might be worthless - always branch at the root and have done with any sub-directory complexity.
At the moment, the current push for implementation is the next-gen working copy which should be easier on Windows virus checkers and faster. I would hope the tree-conflict merge issue be tackled next.
you're confusing the cap you get with the total cap the network has. Those new limits were imposed roughly the same time the iPhone and other smart phones appeared, became very popular, and started sucking up bandwidth like a vacuum compared to old phones that could-but-you-never-really-did use the internet.
imagine if you couldn't use your phone because the network was always full of other people's traffic? People would be irate if this happened (well, more so than on new year's eve for example).
There's a reason for cost-effective plans, and I'm sure the providers will increase the caps over time as they add more capacity, but until they give you more capacity than you need (not forgetting some people use it all, no matter how much you give them) then you'll have to put up with it.
They may also charge you excessive amounts for the extra usage, and that's a money-grabbing scam, but the fact that limits are there is not anything a sensible person should consider out of the ordinary.
Now, that your phone is sending 50Mb (fifty f***ing MB!) of data every day - that's shocking. That's truly shocking, how much xml crap does MS need to put in there? Have they forgotten that data is expensive and you can't treat as mobile phone like a desktop permanently connected to a Gb LAN? Software is so sloppy nowadays, I couldn't even think what 50MB of update/info data looks like.
it came across like the guy himself was a stupid, self-centred nerd-jerk.
I find managers are often just ignorant. (which isn't a criticism as they tend to know more about other stuff, but still have some idea that they should also know everything about the code/systems their staff work with and have too much input into the nuts and bolts when guidance and assurance are what's really needed from the boss).
Slashdot Mirror
Paying a nominal fee for universal support in Linux should not be problem for most people.
Its not about cash, think about the problem of putting (free) Nvidia or ATI graphics drivers into Linux and you'll see the true problem.
Besides, its all solved by having WebM!
Unfortunately, that means you don't get H.264 on Linux as its a proprietary codec that requires some form of (paid) licencing.
I mean, Firefox doesn't support H.264, but Microsoft will happily provide you with the capability of playing H.264 in firefox using a driver that leverages the OS capability... as long as you're running it on Windows.
I think you're partly right though, all the codecs should be implemented as drivers (or similar) and then you are technically using the OS-provided capability, once the correct codec is installed. But its not like the OS is providing the drivers directly, you'll haver to go get them from somewhere. As WebM is free, codecs for it will be freely available for all OSs.
I guess the problem comes for those OSs that are locked down, but then you'er always on to a loser - if Apple only supports H.264 on iPhone and Microsoft only supports (say) H.265 on WP7, and neither allows you to upgrade the video support, then you will never get a video to play universally.
At least there's no excuse for not supporting WebM by all manufacturers, and any who try to give one will quickly be found out by consumers.
As an analogy - look at the non-free 'internets', Microsoft tried to lock you into MSN, and AOL tried similarly. Look where they are now.
Hmm.
so the A4 is an A8 and the A8 will be an A9?
I guess this brain-hurting confusion is the reason why Steve has had to take medical leave. I think I need some too now.
well, its not like you're going to keep it for longer than a year or two anyway.
now that is awesome, thanks for the heads-up.
http://android-developers.blogspot.com/2011/01/gingerbread-ndk-awesomeness.html
probably because most of the SoCs come with telephony chipsets baked in. The manufacturer could disable them, but considering the capability is there "for free", they might as well enable it and offer it as an added extra feature.
You don't have to use it you know, get a emergency-call-only sim (ie a PAYG sim with no credit) and use that.
its not so much the 'dual core-ness' of it but the general increase in power.
ARM says their next CPU design is going to be called the A15, not the A10 as you'd expect. This because, although ARM designs have incremental performance, the next version is so good that they decided to skip A10 through A14 tags.
The A15 BTW, has speeds upto 2.5Ghz, up to 8 cores, and virtualisation instructions. Although you might need a laptop battery to power the top-end version, the fact that there is a top-end spec suggests good stuff to come, even if the smartphones have only 2 cores throttled down to 1.5Ghz.
Phones are going to be the new PCs, I'm sure of it. We're still in the days of the 386 at the moment, it won't take long to revolutionise our lives.
To be fair to some of them, they allow plugins, some of which provide ones to display H.264,
now what will probably happen is Microsoft will bundle a H.264 viewer and automatically install it into your firefox directories to enhance your browsing experience with the "Silverlight+H.264 video" plugin as part of a Windows Update.
Personally, I don't think its going to be properly resolved until Facebook decides to back one codec or the other :)
The hyperlinks in Microsoft's blog post lead readers to data indicating that two-thirds of Web videos are using H.264, with about another 25% using Flash VP6
yes, but once Google updates Youtube to only use WebM, I guess that'll show 91% of all online video to be in WebM format.
I wonder what Microsoft will say then?
hang on, I know Arnie's left office but surely I haven't slipped back in some timewarp to the 70s?
http://www.azlyrics.com/lyrics/deadkennedys/californiauberalles.html
obviously they missed the verse about restricting communications :)
alternative to the PC desktop. You're thinking in terms of how things were, not how things will be.
Look at those smartphones - not one uses Windows Desktop. Even MS says they're the future of computing devices, which is why they're so desperate to get market share.
Legacy apps - sure these will be around for a while, but will increasingly become virtualised (ie run on a server and accessed remotely via Citrix/RDP/screen scrapers even - just like the old mainframe apps that I used to write scrapers for so they could appear as nice Windows apps whilst still running unchanged). Even Exchange could be replaced with Google apps for Enterprise. That might be relatively primitive compared to the full exchange ecosystem, btu how many users actually uses all that Exchange-provided functionality, and do you really think Google is going to sit back and say "job done" with what they provide, or are they going to continually improve their offerings to the point where its a serious alternative. When, not if, do you think that will happen?
There will be years of support, but you're missing the point - there's still support behind COBOL apps and AS/400 servers, but the focus of new dev isn't with those systems is it. Pretty soon (if not already) the focus of new apps will be with more 'mobile' systems, and Windows will be considered a legacy system in itself.
So, you stick your head in the sand, stick your fingers in your ears, and keep thinking that nobody ever gets fired for buying Microsoft (you might have heard that phrase in the past)
cue comments regarding my goldfish being a better CEO than Ballmer.... no, what this means is that the other executives in their playroom don't have anyone with "I'm next" on their hat, so that (like my goldfish really) means they won't be able to decide who to choose as a replacement, and so Ballmer stays.
Of course, anyone who likes Android, Apple or well, anything that isn't Microsoft, should be overjoyed at this news. MS is a dead company, but like the proverbial dinosaur the neural impulses havn't travelled all the way from the tail to the brain yet - MS, being the dinosaur that it is, doesn't yet realise its day is over.
Mind you, one thing that comes to mind is the quality of Bob Muglia. The man that once said to the Seattle Times, with a straight face, "We're going to increase quality in Windows Vista by firing all our quality assurance people!"
Also, it appears this has been brewing for a few months. From minimsft blog:
Somewhat trustworthy source: All BobMu's org's 10s will not see another review cycle. Terminated in groups of under 500 each month to avoid Warn. Can anyone confirm?
Friday, October 22, 2010 7:54:00 PM
Chances are the next CEO will be Kevin Turner, a bean counter like none other (apparently). Just the kind of guy to 'turn round' Microsoft's fortunes by basically reducing the number of beans to count. Or, as someone said on that msft blog: KT CEO - that will be a nice touch indeed - walmarting of Microsoft-- how appropriate
we do this - there's a spreadsheet (well, got to have something right) that gets updated whenever anyone dials in to a customer site, even if you did nothing. If you do something when dialled in, that gets logged too.
I don't know if failing to update it is a pink-slippable offence, but you will get a severe b*ll*cking if you fail to do it twice.
BTW, our customers are police, fire and ambulance control centres. Maybe that makes us different to the usual, but it's simple and works well. If we could get rid of the spreadsheet that's been used forever and I'm sure still is simply because "that's the way we've always done it" then it might be less of a chore. Does anyone have suggestions for a replacement?
The trick, BTW, is to start doing it and keep doing it. After a short period of grumbling about the 'overhead' people will get used to it, and like all good ideas, will accept it once their inertia is overcome.
no, but you'll be laughed at for having a single shared account that means anyone who logs in to perform some "support activity" (maybe after a few drinks, or just general brainfarts) cannot be determined after the event. This can be a good thing, depending on how bad your admins are (good for them, that is). :)
I guess the majority of it is either licensing deals (eg Microsoft paying to be allowed to put Facebook linkage in their phone 'social network hub'), or revenues from apps like Farmville that might pay a percentage of their revenues (farmville apparently makes $1m per day in revenue).
I can't think what else - maybe there's a lot more money to be made from selling your 'private' details than I thought possible!
Have you tried? Its one of those 'silent' technologies where a site allows OpenID auth and make no big fuss over it.
Google and Facebook are OpenID providers as well as clients. Client sites include Sourceforge, the telegraph newspaper, StackOverflow, and many others. Look out for the little logo next time you go to sign in somewhere.
Of course, some sites (like yahoo) are openID providers so you can use your yahoo id as an openid id, but do not let other openid ids login to their sites.
apparently 9 million sites support it according to openid.net in 2009.
Yes, we already have SSO.
Its called OpenID and its a standard. Many sites use it, not only as a 'client' where you use your OpenID id to login, but also some sites use it as a 'server' too, where your account with them can be presented as an OpenID id to other sites. Obviously, some sites do it better than others - MyOpenID or Verisign for example has good security implementation.
Facebook is just another one of these sites, Facebook Connect apparently is the API Facebook provides to other sites to allow them to authenticate with your Facebook id.
Personally, I think if you're going to use a central account to use, you get it from a more reputable company that is more about authentication and less about content.
why is branching broken? Too simplistic? really.
The biggest issue with merging is the "tree conflict" where a directory list is modified. You don't need to treat a directory the same as file as its just a list of files - and you can determine that from the change in the revision (as files in that directory that are modified, added or deleted are recorded, which is exactly what you'd get if you stored directories the same as files).
However, because you cannot tell if a file has been moved, you have to delete it and then add the 'new' file, and this can cause problems, especially if the new files are further modified. If you're the only one working on this directory then there's no problem, but if you rename a file that your colleague has modified... the merge will find that it is trying to update changes into a file that appears to have been deleted, losing those changes. Hence the notification that human intervention is required. This notification isn't so easy to interpret though so perhaps the way merge errors are handled could be improved.
Other systems do track renames so this relatively common issue isn't seen, however, if you work in Windows and rename the file (and modify it) in explorer and then try to commit, even git or mercurial will have difficulty determining that its the same file and not a real delete and add.
possibly not, Symbian devs have worked with C for years on all sorts of chipsets. They squealed more about rewriting their code in Java than cross-chip compiling.
Most Androids come on a few chip architectures anyway (or are they all ARM based?), and the API provided by Google would be standard so I can't see much problem with cross-compiling for all known Android devices. Hell, you could set up a build server in the app store so when you uploaded your code, it would auto-compile for all devices Google knows of. And it would optimise each for the particular platform - that would be a good option, IMHO.
People are writing code for Android using C++ and Qt, so they must be handling the cross-platform problem somehow.
compared to whatever else they'd put in - the problem that the short-term view of stupid researcher fails to spot, is that if you replaced Google with anything else, the spammers and SEOers would quickly learn to game those results too, so you'd end up with exactly what you had before.
Good for Google to start spotting these abuses, I wouldn't worry about losing the adsense results from the first page, as the first page will end up replaced with good results which are still full of adsense links.
oops - typo. "Sparse checkout". You can checkout a partial directory structure, update and commit as required.
So if you have a 12Gb source tree stored in your repo, you don't need to get it all out - only the parts you want to work on. Typically you checkout the root item only, then update the required directories to fetch them into your local drive as you need them.
The ultimate issue is that renames (or moves) are implemented as delete+addition operations. Maybe back in the day, that appeared to be ok, but now its obvious it's a large failing. That's the reason you have problems with merging - merging is fine for changes to existing files, its merging new files where you get the problem. (look up 'tree conflicts' in svn for more info).
The rest of the system works well (though there's still a lot of svn haters for the usual reasons - they have something 'new' or 'cool', they just hate it because its mature and established, they hate it because they last tried it when it was version 1.2 and still think its has those limited features), and there is excellent tooling.
Git and mercurial lose some features that enterprises like, spare checkouts for example is a killer feature for enterprises that don't work well with DVCSs simply because of their original design.
BTW, I like the svn branching system, I've used a lot of scms in my time and svn is the only one that really shows you all the branches easily laid out for you to view - no stupid layers of branch views over the core code. You do have to manage it a little then, and there are features in svn like being able to branch at any point that I think might be worthless - always branch at the root and have done with any sub-directory complexity.
At the moment, the current push for implementation is the next-gen working copy which should be easier on Windows virus checkers and faster. I would hope the tree-conflict merge issue be tackled next.
you're confusing the cap you get with the total cap the network has. Those new limits were imposed roughly the same time the iPhone and other smart phones appeared, became very popular, and started sucking up bandwidth like a vacuum compared to old phones that could-but-you-never-really-did use the internet.
imagine if you couldn't use your phone because the network was always full of other people's traffic? People would be irate if this happened (well, more so than on new year's eve for example).
There's a reason for cost-effective plans, and I'm sure the providers will increase the caps over time as they add more capacity, but until they give you more capacity than you need (not forgetting some people use it all, no matter how much you give them) then you'll have to put up with it.
They may also charge you excessive amounts for the extra usage, and that's a money-grabbing scam, but the fact that limits are there is not anything a sensible person should consider out of the ordinary.
Now, that your phone is sending 50Mb (fifty f***ing MB!) of data every day - that's shocking. That's truly shocking, how much xml crap does MS need to put in there? Have they forgotten that data is expensive and you can't treat as mobile phone like a desktop permanently connected to a Gb LAN? Software is so sloppy nowadays, I couldn't even think what 50MB of update/info data looks like.
it came across like the guy himself was a stupid, self-centred nerd-jerk.
I find managers are often just ignorant. (which isn't a criticism as they tend to know more about other stuff, but still have some idea that they should also know everything about the code/systems their staff work with and have too much input into the nuts and bolts when guidance and assurance are what's really needed from the boss).