I'm not an XML "expert", but I've used XML schema quite a bit the last few months and it seems much easier to read and far more powerful than DTDs. XML Schema has got enough bells and whistles that it's possible to create really hard to read schemas, but it certainly isn't "impenetrable".
The Voyage Home was far and away the best of the Trek movies. It actually had characters, plot, and humor instead of lots of flashing lights and people jumping out of their chairs on the bridge. Even Shatner managed to deliver a decent performance, which was pretty amazing for him.
The worst parts of "Voyage Home" were the "standard trek" parts: people running around at StarFleet headquarters like chickens with their heads cut off while "condition red" blares in the backround. (I've always been curious: on modern warships, does the "battlestations klaxon" keep sounding at maximum volume so no one can think, or do they just sound it for a few seconds and figure that everyone now has a clue what's going on?)
I had the same reaction... except I'm not using Tivo, I'm using a fairly decent VCR I bought back in the mid-80s. Still works fine, programming it is pretty easy, it can be set to record up to 8 shows, and if I'm busy and can't watch TV this week, just pop in a different tape. If it's a great show I want to save "forever", just pop in a different tape.
One problem I have with almost all of the massively multiplayer games out there is that there is defintely a point at which no new players can reasonably expect to join the game, because the long time players have gotten so many advantages over the newbies that the newbies can never expect to make up the difference
Since almost all EQ servers are "player vs. environment" the fact that older players are more experienced or have better gear is irrelevant to new players. You aren't competing against other players.
If you are killed, you get thrown to be back of the bus with the newbies.
When you die, you only lose a small amount of your experience. You certainly don't become level 1 again. In the worst case, if you had just "levelled", you would lose your level.
a newbie who starts today can never pass the leaders of today since those leaders will keep earning their way upward.
Again, the "newbies" aren't competing with other players. Everquest doesn't have anything to do with "passing the leader".
I think Linus' whole point is that what Linus controls is HIS source tree. By definition, he gets to decide what goes in. If he makes enough bad decisions, another source tree may end up becoming the "reference tree" for Linux... so be it.
All of these business models can be traced to having way more venture capital than you know what to do with. I worked for many years for a company that was privately owned. It was extremely well managed, because the owners weren't playing with someone else's money.
In comes some business yahoos (pardon the term) who don't know squat but have lots of financial backing. They conglomerate a bunch of profitable companies and manage to lose a ton of money... of course, not *their* money. Their only goal is to IPO, cash out, and leave the employees and customers holding the bag.
The difference here is that you don't have to put toast into your TV.
Seriously, the whole idea of controlling my house thru a computer network and then having lights, fridge, etc fail due to some software screwup sounds like someone desperately wants to sell me hardware and software. I can believe the M$ and Sun would love us to all require computers to open our garage doors or make dinner, but it's not where I want to go today, or tomorrow either.
Well, I have never seen a 98 box that worked properly for any significant period of time, whenever. An hour included.
In other words, you've never run Windows 9x. Well, at least we know how much to value your opinion.
Seriously, this is the kind of blind MS bashing that tends to hide the truly serious problems with Microsoft. It's so obviously wrong that people start thinking "oh yeah those anti-MS people they're just all full of it". Sort of like the statement in the Volvo article about how false collision alerts from the car cause drivers to ignore the real alerts (just to make this post very slightly on topic).
One argument I just thought of for open source: in closed source, the person who fixes the bug is often the only person who deals with the bug. In open source, several people may submit fixes, and someone integrates the fix into the product. Having more than one person look at a "fix" makes it less likely that the fix will actually introduce problems.
In a large system,
of course, luck cancels out and we can use statistics. A hand-waving argument
would go that, after a million hours of testing, we'd have found all the bugs with
an MTBF of less than a million hours, and we'd hope that the software's overall
reliability would be proportional to the eort invested.
The author is assuming that you start with n bugs and you systematically stamp them out, so the total number of bugs is constantly decreasing.
I saw an interesting study on bugs in some large system (IBM?) years ago that concluded that past some point, the number of bugs started going up again, because "fixes" were introducing more problems than they were solving.
The key to testing isn't the methodology or the tools you use, it's how you approach it.
Glenford Myers summed it up in "The Art of Software Testing":
A successful test finds bugs.
Yep, absolutely the reverse of what most of us think. And yet, consider: if you spend all afternoon testing and "prove your program is correct", what have you accomplished? Your program is no better than when you started!
If, however, you rip into your program, determined to hammer it and find bugs, you'll emerge with an improved product.
I agree. About the only use for ebooks that I can see is not reading "books", but magazines and newspapers. Books we may keep forever, but most of us try to get rid of newpapers and magazines ASAP.
Seems like an obvious application for ebooks, particularly if it was easy to save out the few articles that you care about. But for some reason (maybe no $$$ in it?) I never see this come up when people talk about ebooks.
Yeah, but their customer service is horrible. Simple stuff like monitor swaps under warranty take weeks. The individual people I dealt with seemed ok (one of them went out of his way to finally get me a monitor) but as an organization they seem to be really screwed up.
At least, this was true about 6 months ago when I got my computer. I would never deal with them again.
However, having an ion drive running continuously might play havoc with some experiments that are trying to do things like measure solar wind. Might also mess up micro-gravity experiments.
> It's none of my business if their business model
> depends on the majority of their customers not
> using all the bandwidth they're paying for.
That business model makes your phone and Internet service MUCH cheaper than it would otherwise be.
It's not that customers are "not using all the bandwidth they're paying for"... they are NOT paying for all that bandwidth. If they were, the service would cost more.
Slashdot Mirror
I'm not an XML "expert", but I've used XML schema quite a bit the last few months and it seems much easier to read and far more powerful than DTDs. XML Schema has got enough bells and whistles that it's possible to create really hard to read schemas, but it certainly isn't "impenetrable".
The Voyage Home was far and away the best of the Trek movies. It actually had characters, plot, and humor instead of lots of flashing lights and people jumping out of their chairs on the bridge. Even Shatner managed to deliver a decent performance, which was pretty amazing for him.
The worst parts of "Voyage Home" were the "standard trek" parts: people running around at StarFleet headquarters like chickens with their heads cut off while "condition red" blares in the backround. (I've always been curious: on modern warships, does the "battlestations klaxon" keep sounding at maximum volume so no one can think, or do they just sound it for a few seconds and figure that everyone now has a clue what's going on?)
I had the same reaction... except I'm not using Tivo, I'm using a fairly decent VCR I bought back in the mid-80s. Still works fine, programming it is pretty easy, it can be set to record up to 8 shows, and if I'm busy and can't watch TV this week, just pop in a different tape. If it's a great show I want to save "forever", just pop in a different tape.
Since almost all EQ servers are "player vs. environment" the fact that older players are more experienced or have better gear is irrelevant to new players. You aren't competing against other players.
When you die, you only lose a small amount of your experience. You certainly don't become level 1 again. In the worst case, if you had just "levelled", you would lose your level.
Again, the "newbies" aren't competing with other players. Everquest doesn't have anything to do with "passing the leader".
Well, there's option 3: Some people start using Win Show EQ. Sony bans them. People decide that they want to play EQ and don't use Win EQ. Sony wins.
Seems like the most likely outcome.
I think Linus' whole point is that what Linus controls is HIS source tree. By definition, he gets to decide what goes in. If he makes enough bad decisions, another source tree may end up becoming the "reference tree" for Linux... so be it.
The parent post about "no smaller hole" is the most cogent comment of all. This one ought to end the discussion right here: case dismissed.
All of these business models can be traced to having way more venture capital than you know what to do with. I worked for many years for a company that was privately owned. It was extremely well managed, because the owners weren't playing with someone else's money.
In comes some business yahoos (pardon the term) who don't know squat but have lots of financial backing. They conglomerate a bunch of profitable companies and manage to lose a ton of money... of course, not *their* money. Their only goal is to IPO, cash out, and leave the employees and customers holding the bag.
The difference here is that you don't have to put toast into your TV.
Seriously, the whole idea of controlling my house thru a computer network and then having lights, fridge, etc fail due to some software screwup sounds like someone desperately wants to sell me hardware and software. I can believe the M$ and Sun would love us to all require computers to open our garage doors or make dinner, but it's not where I want to go today, or tomorrow either.
In other words, you've never run Windows 9x. Well, at least we know how much to value your opinion.
Seriously, this is the kind of blind MS bashing that tends to hide the truly serious problems with Microsoft. It's so obviously wrong that people start thinking "oh yeah those anti-MS people they're just all full of it". Sort of like the statement in the Volvo article about how false collision alerts from the car cause drivers to ignore the real alerts (just to make this post very slightly on topic).
One argument I just thought of for open source: in closed source, the person who fixes the bug is often the only person who deals with the bug. In open source, several people may submit fixes, and someone integrates the fix into the product. Having more than one person look at a "fix" makes it less likely that the fix will actually introduce problems.
The author is assuming that you start with n bugs and you systematically stamp them out, so the total number of bugs is constantly decreasing. I saw an interesting study on bugs in some large system (IBM?) years ago that concluded that past some point, the number of bugs started going up again, because "fixes" were introducing more problems than they were solving.
This is exactly right.
The key to testing isn't the methodology or the tools you use, it's how you approach it.
Glenford Myers summed it up in "The Art of Software Testing":
A successful test finds bugs.
Yep, absolutely the reverse of what most of us think. And yet, consider: if you spend all afternoon testing and "prove your program is correct", what have you accomplished? Your program is no better than when you started!
If, however, you rip into your program, determined to hammer it and find bugs, you'll emerge with an improved product.
I agree. About the only use for ebooks that I can see is not reading "books", but magazines and newspapers. Books we may keep forever, but most of us try to get rid of newpapers and magazines ASAP. Seems like an obvious application for ebooks, particularly if it was easy to save out the few articles that you care about. But for some reason (maybe no $$$ in it?) I never see this come up when people talk about ebooks.
Yeah, but their customer service is horrible. Simple stuff like monitor swaps under warranty take weeks. The individual people I dealt with seemed ok (one of them went out of his way to finally get me a monitor) but as an organization they seem to be really screwed up. At least, this was true about 6 months ago when I got my computer. I would never deal with them again.
However, having an ion drive running continuously might play havoc with some experiments that are trying to do things like measure solar wind. Might also mess up micro-gravity experiments.
> It's none of my business if their business model > depends on the majority of their customers not > using all the bandwidth they're paying for. That business model makes your phone and Internet service MUCH cheaper than it would otherwise be. It's not that customers are "not using all the bandwidth they're paying for"... they are NOT paying for all that bandwidth. If they were, the service would cost more.