How can they then know how many such candidates that will fail to be release quality before hand?
Oh, that's easy.
It isn't the engineers but the marketing and business strategy people that are in charge of determining when it should be released. By definition (or fiat), it's ready when it must be.
As opposed to Debian, which is ready when it's, you know, ready. No matter how many years that takes;)
[See, I'm fair: I flame everybody a bit]
I think "the one true way" of release management is some sort of compromise. All you can guarantee is either shipping at some date OR shipping at some measurable quality level [this many features, this few bugs]. None of these are particularly exciting.
What works best in practice is probably a reasonable trade-off: we ship in this month, and cut features so that the features we do ship meet a minimal level of quality.
Do you know what 1000 Windows 2008 Server client and Exchange Server client licenses cost?
No, but last I heard the Microsoft licensing is so hard to figure out that the lawyers who can tell you how much this would cost are almost as expensive as the software itself.
Rather than writing a plugin for evolution, wouldn't it be better to write a mail server (i.e. pop or imap) with a lot of extensions that map onto what exchange can do, and then have an evolution plugin that talks to that?
In that way, you can do all the boring old mail operations from your mail client of choice and be happy, and use evolution for all the fancy non-mail stuff.
If there's a standard calendaring protocol, make the wrapper daemon compatible with both imap and the calendaring protocol. Or split it into two daemons (who talk together if need be).
(and of course provide a nice "Setup exchange account" in evolution that configures and runs these daemons; make it easy to use, of course.)
[for those of you who love design patterns, this'd be the adapter pattern, right?]
I know I'm going to pay everything my government has given me back in taxes, and maybe a bit more, but I think my life is better for it.
Where it not for free health and free university tuition---no, scratch that, I get paid to take an education---then I'm not sure where I would be today.
But I'm convinced that even with the tax breaks my parents would have gotten, they wouldn't have been able to afford both my hospitalization and my studies at a university [which claims to be the best at CS here in the country].
And I think that would have been a shame, since my ability to code is probably my best shot at providing value for others in our society.
So, were it not for Danish Socialism, society would have missed out on the (maximal) productivity of one of its members.
The analysis presented here is simplistic, and I'd suggest basing public policy on real science. But I suggest you all consider that "freedom=good, socialism=bad" isn't at all obviously evident either.
This is more evidence that DRM hurts the honest consumer.
As we all know, the pirates wait for the DRM-free... "collectors edition" release on The Pirate Bay.
Why do people continue doing it? Did they start when the economy was in a healthy growth period and then think "more DRM, more economic growth for us, it must obviously be causal".
(now there's a good application of "correlation is not causation" for you)
I have no issue with someone like Cox de-prioritizing their [P2P] traffic so that the people that just want their Vonage to work don't get squashed out.
I have a problem with my ISP giving my traffic lower priority based on the meaning of the bytes I'm communicating. All they should worry about is packet lengths and (maybe) QoS fields.
I'm fine with my bulk transfer being delayed during sparse bursts of interactive traffic from my neighbors: they get fast HTTP, I get all the pipe when they're reading the page (as opposed to downloading a new one).
That is, as long as we over time each get a fair share of the pipe up to the amount we're using individually: if they do a bulk transfer via HTTP, it should have the same priority as my bulk transfer via $PROTOCOL.
If they do a long-running VoIP stream, they better limit the bandwidth of that such that there's also room for my traffic.
But please, feel free to keep flaming people and bolstering your ego.
I didn't mean it as a flame, and none of the mods who cared about what I said saw it that way either.
Sorry if I came off that way.
ISTR from my statistics course that you could say something reasonably accurate from a surprisingly low sample size. OTOH, it didn't give as solid a theoretical foundation as I would have liked.
[...]If so, it will be the first version of Windows that makes computers run faster than the previous version.
Aside from how ridiculous that statement sounds to me ("Vista makes your computer run faster?")
Order your computer to perform some task, and note the time
Wait for the task to finish, observe the time
Compare the difference between the two times to a measured difference on another system
Conclude that one system makes your computer perform tasks ("run faster") than the other
Of course ${OS} ${VERSION} won't bump your CPU cycle frequency or increase your cache size.
But if one OS performs tasks in less time than another, if one thinks of "the computer" as an (OS, hardware) bundle, it does make the computer run faster.
This has nothing to do with "lessons learned from video games" and says everything about the power of conditioning.
FTFY.
Also, this is nothing new. As another poster has pointed out, the result says "Conditioning works. Stick a game controller in someone's hands. Conditioning still works."
A former girlfriend of mine was studying Marketing Economy, and I peeked in her textbooks. They had the law of effect (essentially "conditioning works") stated in them. I don't know in how much detail they explained it, though.
Unless you can explain to me what they marketing part of the study is, I'll prefer to say the conclusion as "conditioning works, marketing uses conditioning because conditioning works".
Could you please point to which of their inferences you think breaks down because of statistical problems caused by the sample size?
If no such problem exists, the sample size was fine.
I recall reading a set of guidelines for writing psych papers (discussing and critiquing an article). They said quite explicitly that complaining about sample size was about the cheapest shot available, so don't do it unless you can really back it up.
To the mods who think my parent is insightful: could you please spell out to me what the insight is? Because I haven't seen any problems with the sample size, only an unsubstantiated claim.
Society is better off if we don't prosecute crimes of low value or low impact, but rather leave it to citizens to work out between themselves.
I'm thinking that class action law suits exist for the case where one big guy illegally squeezes a little bit of blood out of a large number of small guys.
None of the small guys have an incentive to file suit (because each individual act is too small to be worth it), but if they all go together there's only one per-case overhead and so it will be worth it.
I'm not sure how I'm best served as the hotel owner (big guy) in that case. Oh well, I could just write the stolen soap off as the cost of doing business, and charge my customers a bit more. That's of course not just towards the honest customers, but it's the cheapest justice I can give them.
Yet the general assumption across different version number schemes is that once you climb to the top of the hill and blow the horn that can be heard all over the valley, and come storming down screaming "listen everybody"...
Then you have something important to say that most people want to listen to.
KDE made noise, but had a message only for the people on the hill, not everyone in the valley.
The real problem is that record companies exploit musicians. That's what we want to solve.
Forbidding corporations from holding copyrights will be ineffective: your employment contract will state that you give your employer permission to exercise all the holder-only rights, and promise not to exercise any of them yourself. The fact that you're still the copyright holder means nothing: the implications of that have been superseded by a contract.
You can try to squash the dodge by saying the rights can be signed away, but then you're playing Whack-a-Mole.
What's really needed is for musicians to have (and use!) more bargaining power. The musicians should be able to tell the record companies that they don't need the record companies and are able to do all the not-playing-the-music work on their own, and they should be able to follow up on that.
That's a bit difficult if the record company execs play golf with the radio broadcaster execs. Fortunately, there are no Internet execs, as long as there are ISPs who will sell you a pipe restricted only by the amount of bytes you send each second.
Slashdot Mirror
How can they then know how many such candidates that will fail to be release quality before hand?
Oh, that's easy.
It isn't the engineers but the marketing and business strategy people that are in charge of determining when it should be released. By definition (or fiat), it's ready when it must be.
As opposed to Debian, which is ready when it's, you know, ready. No matter how many years that takes ;)
[See, I'm fair: I flame everybody a bit]
I think "the one true way" of release management is some sort of compromise. All you can guarantee is either shipping at some date OR shipping at some measurable quality level [this many features, this few bugs]. None of these are particularly exciting.
What works best in practice is probably a reasonable trade-off: we ship in this month, and cut features so that the features we do ship meet a minimal level of quality.
Do you know what 1000 Windows 2008 Server client and Exchange Server client licenses cost?
No, but last I heard the Microsoft licensing is so hard to figure out that the lawyers who can tell you how much this would cost are almost as expensive as the software itself.
(rimshot)
[...] the Evolution MAPI plugin [...]
I'm stuck wondering about this.
Rather than writing a plugin for evolution, wouldn't it be better to write a mail server (i.e. pop or imap) with a lot of extensions that map onto what exchange can do, and then have an evolution plugin that talks to that?
In that way, you can do all the boring old mail operations from your mail client of choice and be happy, and use evolution for all the fancy non-mail stuff.
If there's a standard calendaring protocol, make the wrapper daemon compatible with both imap and the calendaring protocol. Or split it into two daemons (who talk together if need be).
(and of course provide a nice "Setup exchange account" in evolution that configures and runs these daemons; make it easy to use, of course.)
[for those of you who love design patterns, this'd be the adapter pattern, right?]
But as an Apple user, I don't have this problem.
Of course not: you don't have DRM-infested games, because you don't have any games. (photoshop is not a game)
What? Who's that on the phone? 1995? And they want their meme back? Tell 'em to take a hike ;)
Why can't we just do plain old regular teleportation?
http://xkcd.com/465/
Or maybe by "known valid source" you meant "the Sun and the stars";
Try Masters of Orion II :)
You want the government to give you everything?
I know I'm going to pay everything my government has given me back in taxes, and maybe a bit more, but I think my life is better for it.
Where it not for free health and free university tuition---no, scratch that, I get paid to take an education---then I'm not sure where I would be today.
But I'm convinced that even with the tax breaks my parents would have gotten, they wouldn't have been able to afford both my hospitalization and my studies at a university [which claims to be the best at CS here in the country].
And I think that would have been a shame, since my ability to code is probably my best shot at providing value for others in our society.
So, were it not for Danish Socialism, society would have missed out on the (maximal) productivity of one of its members.
The analysis presented here is simplistic, and I'd suggest basing public policy on real science. But I suggest you all consider that "freedom=good, socialism=bad" isn't at all obviously evident either.
This is more evidence that DRM hurts the honest consumer.
As we all know, the pirates wait for the DRM-free... "collectors edition" release on The Pirate Bay.
Why do people continue doing it? Did they start when the economy was in a healthy growth period and then think "more DRM, more economic growth for us, it must obviously be causal".
(now there's a good application of "correlation is not causation" for you)
Can you imagine the revolution society would undergo if "voice stress analyzers" actually worked?
O_o You're not seriously suggesting we'd have honest politicians, are you???
you can use your applications anywhere you have access to a web browser.
Apache + mindterm + sshd + screen + apps = win :)
what? sudo apt-get install it yourself!
I have no issue with someone like Cox de-prioritizing their [P2P] traffic so that the people that just want their Vonage to work don't get squashed out.
I have a problem with my ISP giving my traffic lower priority based on the meaning of the bytes I'm communicating. All they should worry about is packet lengths and (maybe) QoS fields.
I'm fine with my bulk transfer being delayed during sparse bursts of interactive traffic from my neighbors: they get fast HTTP, I get all the pipe when they're reading the page (as opposed to downloading a new one).
That is, as long as we over time each get a fair share of the pipe up to the amount we're using individually: if they do a bulk transfer via HTTP, it should have the same priority as my bulk transfer via $PROTOCOL.
If they do a long-running VoIP stream, they better limit the bandwidth of that such that there's also room for my traffic.
But please, feel free to keep flaming people and bolstering your ego.
I didn't mean it as a flame, and none of the mods who cared about what I said saw it that way either.
Sorry if I came off that way.
ISTR from my statistics course that you could say something reasonably accurate from a surprisingly low sample size. OTOH, it didn't give as solid a theoretical foundation as I would have liked.
You have a good point.
I'd have to know the details of the course to give a good answer to that.
[...]If so, it will be the first version of Windows that makes computers run faster than the previous version.
Aside from how ridiculous that statement sounds to me ("Vista makes your computer run faster?")
Of course ${OS} ${VERSION} won't bump your CPU cycle frequency or increase your cache size.
But if one OS performs tasks in less time than another, if one thinks of "the computer" as an (OS, hardware) bundle, it does make the computer run faster.
Is it really that ridiculous?
I didn't really want to bore everyone with mundane discussions of my career field ;-)
This is news for nerds. It wouldn't be boring to us, and we value precision.
Know and conform to the values of your peer group. Didn't you learn that in high school? :p
Stop it! You're being overly rational in a perfectly emotional debate.
"His inability to substantiate the claim may be due to the sample size."
Huh? His (parent(n)s) inability to substantiate the (implicit) claim that the sample size was too small may be due to the sample size?
That is, the sample size is just fine (like I suggested it might be)? And then you say that my post is a defensive whine?
*headsplode*
This has nothing to do with "lessons learned from video games" and says everything about the power of conditioning.
FTFY.
Also, this is nothing new. As another poster has pointed out, the result says "Conditioning works. Stick a game controller in someone's hands. Conditioning still works."
A former girlfriend of mine was studying Marketing Economy, and I peeked in her textbooks. They had the law of effect (essentially "conditioning works") stated in them. I don't know in how much detail they explained it, though.
Unless you can explain to me what they marketing part of the study is, I'll prefer to say the conclusion as "conditioning works, marketing uses conditioning because conditioning works".
Yes, they were given real sips of drinks.
So what? That's just a way to tap into a basic reward system.
If you play a game and the reward is the "you win" screen in the game, the reward is still rewarding.
The important part in conditioning is that a particular stimulus is followed by a reward.
<sarcasm>Good to see they're using a nice large group of test subjects.</sarcasm>
The article is here: http://www.jneurosci.org/cgi/reprint/29/4/1046.pdf
Could you please point to which of their inferences you think breaks down because of statistical problems caused by the sample size?
If no such problem exists, the sample size was fine.
I recall reading a set of guidelines for writing psych papers (discussing and critiquing an article). They said quite explicitly that complaining about sample size was about the cheapest shot available, so don't do it unless you can really back it up.
To the mods who think my parent is insightful: could you please spell out to me what the insight is? Because I haven't seen any problems with the sample size, only an unsubstantiated claim.
Society is better off if we don't prosecute crimes of low value or low impact, but rather leave it to citizens to work out between themselves.
I'm thinking that class action law suits exist for the case where one big guy illegally squeezes a little bit of blood out of a large number of small guys.
None of the small guys have an incentive to file suit (because each individual act is too small to be worth it), but if they all go together there's only one per-case overhead and so it will be worth it.
I'm not sure how I'm best served as the hotel owner (big guy) in that case. Oh well, I could just write the stolen soap off as the cost of doing business, and charge my customers a bit more. That's of course not just towards the honest customers, but it's the cheapest justice I can give them.
Yet the general assumption across different version number schemes is that once you climb to the top of the hill and blow the horn that can be heard all over the valley, and come storming down screaming "listen everybody"...
Then you have something important to say that most people want to listen to.
KDE made noise, but had a message only for the people on the hill, not everyone in the valley.
Dude, you're running three window managers!
I disagree.
The real problem is that record companies exploit musicians. That's what we want to solve.
Forbidding corporations from holding copyrights will be ineffective: your employment contract will state that you give your employer permission to exercise all the holder-only rights, and promise not to exercise any of them yourself. The fact that you're still the copyright holder means nothing: the implications of that have been superseded by a contract.
You can try to squash the dodge by saying the rights can be signed away, but then you're playing Whack-a-Mole.
What's really needed is for musicians to have (and use!) more bargaining power. The musicians should be able to tell the record companies that they don't need the record companies and are able to do all the not-playing-the-music work on their own, and they should be able to follow up on that.
That's a bit difficult if the record company execs play golf with the radio broadcaster execs. Fortunately, there are no Internet execs, as long as there are ISPs who will sell you a pipe restricted only by the amount of bytes you send each second.