>...sometimes the guys working on Office don't even have all the information > on the secret APIs the OS folks come up with.
Having spent a little time (very little, fortunately) doing Windows app coding, I'd be incredibly surprised if they had all the information on the public APIs. Or the time to find anything.
I think much of the bloat in Office is because it's faster for the Office developers to re-invent the wheel than to search the Windows API's for things to reuse.
Offended? Hardly. I don't get offended by other people being dicks.
I'm just unimpressed with the article. Someone was trying way too hard to be clever, and I think it came out looking really shabby. That someone went ahead and published it anyways... Ah, well, it was worth about what I paid for it.
> Maybe we should have a bit more respect this time.
Well, the original quote itself was from the article. Which is one of the... oddest articles I've read from Wired. When you give something like that to/. as source material you're going to get some wildly inappropriate reactions.
For an article which is supposed to show the more "personal" side of things, the main thing I'm taking away from this is that the author is seriously fucked up. It's like the worst tabloid journalism combined with a Dvorak column. It certainly didn't do much to help Hans...
> Excuse me while I go selectively erase the mental image of Bill Gates in a French > maid uniform
Just imagine the grin on Melinda's face as she dresses him that morning:
"Oh, don't worry dear. It's what all the other CEO's are wearing".
c.
Re:~$ mv CommitAccess MergePrivileges
on
Linus on GIT and SCM
·
· Score: 2, Informative
> In any project where you have people who are going to fight about who gets commit > access, you'll just have a fight about who has the ability to merge into mainstream.
I really wish he would have addressed that question a little more directly, too.
I think the problem is that you're thinking about it from a classic centralized development model. I have some trouble getting my head around it, too.
Basically, from a truly distributed SCM perspective, there is no "mainstream". All branches are equal. Obviously this isn't quite the case with Linux, but bear with me here.
If you've got good code, what happens is that your changes get merged into more branches than bad code. The more popular your code, the more stuff gets built on top of it. If your code is good enough, eventually it gets into the "mainstream" simply by being an unavoidable dependency for other code.
Quality (or quantity) of code rules.
Politics are only an issue, then, if someone tries to bypass this process by skipping their changes right into the most "popular" branches. But this means they have to convince the owners of those branches to merge it. And while it's really hard to ignore changes from someone with commit access in a centralized SCM, ignoring someone in a distributed SCM is just a killfile away.
> This type of legislation is a cry for help on the part of the legislator.
Nope. This legislation is 100% in response to a very intense and focused lobbying operation by the United States. We're talking about threats from the movie industry, US Senators, the freaking Ambassador to Canada, you name it. Nobody in Canada wants this silliness, and it's not going to do a damn thing. If we're lucky, it's not even going to pass into law.
In reality, this is probably the thin edge of the wedge to try to force Canada to adopt DMCA-like stupidity; we've already for a FTA with the USA (which they ignore when convenient, but that's another story), so they can't sneak it in that way.
Canada, as you may have heard, is retaliating with a new Celine Dion album. And if the DMCA stuff happens, we've got an entire province of French Canadian adult contemporary crooner wannabes just lining up at Canadian Idol. Consider it fair warning.
> You just pull it out of the wall and blessed darkness and silence....unless the device notices power is missing and drops into a battery-backed blinkenlight standby. Like laptops, PDAs, etc.
> I'd really hope you would not be surprised that a function called "alloc"-something > would allocate memory on success.
Not at all. I am surprised that such a function would consider leaking memory as success. Apparently, I'm not alone in this surprise.
I know what point he was trying to make. I use that particular goto contruct myself occasionally.
But by using the functional equivalent of
int allocstuff(void) {
return !(malloc(100) && malloc(100)); }
as an example of good goto usage, the only point the author made is that excessive use of goto leads to unmaintainable and ineffective spaghetti code, even in 10 line examples.
Maybe my standards are too high or something (I doubt it; I code in C), but it seems reasonable that if you're going to bother writing a few lines of code to prove a point, you'll ensure your point isn't undermined by such blatantly obvious flaws.
> Did you notice how the "free(b)" call was after an unconditional return?
Yeah. I'm kinda wondering what sort of development shop this coder works in where "success" means you leak a couple hundred bytes (or so) of heap, but failures are cleaned up properly.
Should I just say "Microsoft" and save everyone else the trouble?
Reason #471 c will not have children: to avoid having to explain his accidental involvement in the infamous Jack Thompson Slash Fiction thread of 2007 AD.
More likely, she made a (really stupid) election campaign promise to do "something about the interweb" and someone (who gave her money) reminded her that she hasn't done a thing and there's probably an election call coming up shortly.
> What reasonable judge wouldn't go along with this line of thinking?
IIRC, in the NTP vs RIM case, something like seven out of the eight patents being disputed were rejected after re-examination and, even though the last one was still under review, the judge still refused to postpone an injunction.
Now, I'm thinking that we're not talking about a "reasonable judge" there, but defendants don't get a multiple choice form when the show up at the court:
Judge preferrence:
(a) reasonable
(b) techno-ignorant lunatic
(c) sleeps during trial
(d) I feel lucky
It's worth getting the patents re-examined, but after RIM vs NTP, no sane legal team would count on the PTO coming through on time.
... unless the lasers, smart mines and 'mechs are running Windows. Then it's p0wn3d, and the Russian mob gets to use an anonymous asteroid as a spam relay.
> he still remains generally opposed to infringing on privacy, big-government, and censorship.
That may be his personal tendencies, but it's safest to believe that when election time rolls around (and it's about that time) politicians are going to champion whatever it is they think will get them elected. I believe that's still mainly environment, defence/security, health care, and whatever Quebec wants, or did I miss some other over-hyped issue in the last few months?
Yup. I telecommute (AKA telework, 'cause the Canadian government can't just use the same word as everyone else) a couple days a week, everyone I work with knows about it, and effectively I think it's best to treat it as some kind of "closed door" day. Like I'm in the office getting work done, but I need to concentrate and can't be bothered. But telecommuting full-time... I couldn't do it, and I'd have a hard time working with someone who did.
Mind you, I mainly do development so just about everything I normally do shows up in commits, issue updates, and assorted e-mail. People see what's happening irrespective of when I'm working.
> Maybe the problem is that if you don't give a one-to-three year old a chance
Whatever we encourage our kids to play with, it's a good bet that the tinkering tendency should start young.
Which makes the "Modest Proposal" puzzling. I can't see any sane way in which "a one-year, co-op, certificate-granting program" for adults is somehow going to close a "self-taught engineer gender gap". "Self-taught" implies a certain minimal level of interest and personal motivation to do things without having to be guided into it by others.
I think it might be possible for a one-year program to produce some pretty decent techies (they won't be engineers, that's for sure), but equivalent to someone who has enough of a passion for it to have been working at it for, in some cases, all their lives?
Slashdot Mirror
> When are people going to accept that teenagers are sexual beings too.
Seeing as how these same lunatics haven't accepted that adults are sexual beings, I'm thinking "never".
c.
> ...sometimes the guys working on Office don't even have all the information
> on the secret APIs the OS folks come up with.
Having spent a little time (very little, fortunately) doing Windows app coding, I'd be incredibly surprised if they had all the information on the public APIs. Or the time to find anything.
I think much of the bloat in Office is because it's faster for the Office developers to re-invent the wheel than to search the Windows API's for things to reuse.
c.
Offended? Hardly. I don't get offended by other people being dicks.
I'm just unimpressed with the article. Someone was trying way too hard to be clever, and I think it came out looking really shabby. That someone went ahead and published it anyways... Ah, well, it was worth about what I paid for it.
c.
> Maybe we should have a bit more respect this time.
/. as source material you're going to get some wildly inappropriate reactions.
Well, the original quote itself was from the article. Which is one of the... oddest articles I've read from Wired. When you give something like that to
For an article which is supposed to show the more "personal" side of things, the main thing I'm taking away from this is that the author is seriously fucked up. It's like the worst tabloid journalism combined with a Dvorak column. It certainly didn't do much to help Hans...
c.
> Excuse me while I go selectively erase the mental image of Bill Gates in a French
> maid uniform
Just imagine the grin on Melinda's face as she dresses him that morning:
"Oh, don't worry dear. It's what all the other CEO's are wearing".
c.
> In any project where you have people who are going to fight about who gets commit
> access, you'll just have a fight about who has the ability to merge into mainstream.
I really wish he would have addressed that question a little more directly, too.
I think the problem is that you're thinking about it from a classic centralized development model. I have some trouble getting my head around it, too.
Basically, from a truly distributed SCM perspective, there is no "mainstream". All branches are equal. Obviously this isn't quite the case with Linux, but bear with me here.
If you've got good code, what happens is that your changes get merged into more branches than bad code. The more popular your code, the more stuff gets built on top of it. If your code is good enough, eventually it gets into the "mainstream" simply by being an unavoidable dependency for other code.
Quality (or quantity) of code rules.
Politics are only an issue, then, if someone tries to bypass this process by skipping their changes right into the most "popular" branches. But this means they have to convince the owners of those branches to merge it. And while it's really hard to ignore changes from someone with commit access in a centralized SCM, ignoring someone in a distributed SCM is just a killfile away.
c.
> This type of legislation is a cry for help on the part of the legislator.
Nope. This legislation is 100% in response to a very intense and focused lobbying operation by the United States. We're talking about threats from the movie industry, US Senators, the freaking Ambassador to Canada, you name it. Nobody in Canada wants this silliness, and it's not going to do a damn thing. If we're lucky, it's not even going to pass into law.
In reality, this is probably the thin edge of the wedge to try to force Canada to adopt DMCA-like stupidity; we've already for a FTA with the USA (which they ignore when convenient, but that's another story), so they can't sneak it in that way.
Canada, as you may have heard, is retaliating with a new Celine Dion album. And if the DMCA stuff happens, we've got an entire province of French Canadian adult contemporary crooner wannabes just lining up at Canadian Idol. Consider it fair warning.
c.
> You see that skyscraper? The one near your house?
Well, no. I see a grain silo. Not sure if that's high enough, and Jeff would be pissed if I dropped a laptop on a cow.
The cow might not be pleased, either.
c.
> You just pull it out of the wall and blessed darkness and silence. ...unless the device notices power is missing and drops into a battery-backed blinkenlight standby. Like laptops, PDAs, etc.
c.
> I hope those pesky dead people get sentenced to a bunch of community service.
A life sentence of pushing up daisies, I presume?
c.
> would allocate memory on success.
Not at all. I am surprised that such a function would consider leaking memory as success. Apparently, I'm not alone in this surprise.
I know what point he was trying to make. I use that particular goto contruct myself occasionally.
But by using the functional equivalent of
as an example of good goto usage, the only point the author made is that excessive use of goto leads to unmaintainable and ineffective spaghetti code, even in 10 line examples.
Maybe my standards are too high or something (I doubt it; I code in C), but it seems reasonable that if you're going to bother writing a few lines of code to prove a point, you'll ensure your point isn't undermined by such blatantly obvious flaws.
c.
> Did you notice how the "free(b)" call was after an unconditional return?
Yeah. I'm kinda wondering what sort of development shop this coder works in where "success" means you leak a couple hundred bytes (or so) of heap, but failures are cleaned up properly.
Should I just say "Microsoft" and save everyone else the trouble?
c.
> But my 85-year-old uncle probably will never own an iPod,
> and I hope we'll get him to own a Zune.
Not if his uncle has heard about this whole "squirting" thing Steve's got happening.
Reason #471 c will not have children: to avoid having to explain his accidental involvement in the infamous Jack Thompson Slash Fiction thread of 2007 AD.
> Jack wife is also a lawyer and she works for what is aparently a pretty well known firm in Miami.
That must make conversations around the water cooler a bit awkward.
"So, I heard your nutjob husband's legal theories were ruled unconstitutional and cost [US state] $[legal bill]..."
> a backbencher trying to catch some limelight
More likely, she made a (really stupid) election campaign promise to do "something about the interweb" and someone (who gave her money) reminded her that she hasn't done a thing and there's probably an election call coming up shortly.
c.
> What reasonable judge wouldn't go along with this line of thinking?
IIRC, in the NTP vs RIM case, something like seven out of the eight patents being disputed were rejected after re-examination and, even though the last one was still under review, the judge still refused to postpone an injunction.
Now, I'm thinking that we're not talking about a "reasonable judge" there, but defendants don't get a multiple choice form when the show up at the court:
Judge preferrence:
(a) reasonable
(b) techno-ignorant lunatic
(c) sleeps during trial
(d) I feel lucky
It's worth getting the patents re-examined, but after RIM vs NTP, no sane legal team would count on the PTO coming through on time.
c.
> Meanwhile, the children grow up and move to Vancouver.
Great. Toxic, flame retardant adults. Just what Canada needs.
c.
... unless the lasers, smart mines and 'mechs are running Windows. Then it's p0wn3d, and the Russian mob gets to use an anonymous asteroid as a spam relay.
c.
> he still remains generally opposed to infringing on privacy, big-government, and censorship.
That may be his personal tendencies, but it's safest to believe that when election time rolls around (and it's about that time) politicians are going to champion whatever it is they think will get them elected. I believe that's still mainly environment, defence/security, health care, and whatever Quebec wants, or did I miss some other over-hyped issue in the last few months?
c.
> when you have a free moment, delete your mbox file ...
>
> knock up a shell script which turns your machine into a spam-spewing zombie
See, that's the problem with Linux. You have to do all that extra work to get functionality which just plain works under Windows.
c.
The sad part is that you got modded as "Funny".
c.
Yup. I telecommute (AKA telework, 'cause the Canadian government can't just use the same word as everyone else) a couple days a week, everyone I work with knows about it, and effectively I think it's best to treat it as some kind of "closed door" day. Like I'm in the office getting work done, but I need to concentrate and can't be bothered. But telecommuting full-time... I couldn't do it, and I'd have a hard time working with someone who did.
Mind you, I mainly do development so just about everything I normally do shows up in commits, issue updates, and assorted e-mail. People see what's happening irrespective of when I'm working.
c.
> Maybe the problem is that if you don't give a one-to-three year old a chance
Whatever we encourage our kids to play with, it's a good bet that the tinkering tendency should start young.
Which makes the "Modest Proposal" puzzling. I can't see any sane way in which "a one-year, co-op, certificate-granting program" for adults is somehow going to close a "self-taught engineer gender gap". "Self-taught" implies a certain minimal level of interest and personal motivation to do things without having to be guided into it by others.
I think it might be possible for a one-year program to produce some pretty decent techies (they won't be engineers, that's for sure), but equivalent to someone who has enough of a passion for it to have been working at it for, in some cases, all their lives?
c.
> why do Bliar and Bush and co. call the Iraqi people that fight back 'insurgents'
I kinda thought it was a step up from calling local's who shoot at invading foreign soldiers 'terrorists'...
c.