MySQL has other problems as well. For example, if I want a column to be NOT NULL, I want any code that tries to insert a NULL into that column to fail. I don't want the engine to try to pick some default value for me. If I wanted a default, I would've added a default. That's why default constraints exist. By that same token, if I want a column to allow NULLs, I want to be able to put a NULL in the column. I don't want the current date/time instead of a NULL. If I define a column as auto-incrementing, I want to get an error if I try to insert something into that column. I don't want it to quietly succeed.
At my workplace, we've recently started using MySQL much more often, and have stumbled onto these "gotchas" you mention. Each time we hit a new one, someone spends up to several hours or more, trying to determine if it's our problem or mysql's problem. The natural inclination is to ask "what am I doing wrong?" and spend unneeded time staring at our code.
So, thanks for the the "gotchas" link. That will save us some time in the future, I'm sure.
There are those who try to learn what their customers want, and deliver it.
If this were a business relationship, I'd agree. In a business, if you *didn't* do what you describe, you'd go out of business.
But with Free (or free) software, there is no business relationship. No money exchanges hands. Users are not "customers" or "consumers" because they didn't give any money to the developers (or the mozilla.org organization -- with rare exception).
As such, developers are fully in their rights to blow off stupid comments.
And I say this *not* from a conceited or condescending developer point of view. I'm not a l337 developer.
But in my day development job, I do deal with customers from time to time. These people have contracts with us, they pay us money, and they expect us deliver X by Y date. Of course X, changes, and Y gets moved up.
In my humble opinion, a lot of X is stupid, but the customer wants it, so the customer gets it. I'm helpful. I'm nice. I make every effort to deliver.
A lot of my job in meeting the customers' demands consists of doing things that are decidedly not fun or interesting: writing documentation, creating flowcharts, dummying up boiler plate examples, etc. (We're a small shop and wear many hats).
We have to spend a good amount of time hand-holding our customers, who can be quite demanding. When some bug report can't be tracked down by our customer service or support folk, developers get pulled to investigate.
And, of course, there's the meetings. Oh God, the meetings.
I'm not complaining here, just observing. I know there are a lot of skilled developers who would love to have a good job right now, and I am truly thankful to be working. But that doesn't mean it's not frustrating at time.
My "development job" is a lot more than just writing code. But I like to write code, not all that other stuff.
So, at nights and on the weekends (Well, not recently. I've got a kid, soon to have another.) I would actually write code on little side projects. Not terribly useful to anyone besides myself. This is why I'm not a l337 developer. But I do it just because I want to, because I enjoy it. I enjoy writing things that do things.
This is what Free Software is about. Freedom of the developers. Users are nice, even desirable, but they are not customers and can make no demands on my time beyond what I'm already freely giving. I won't deride them, but I'm certainly under no obligation to meet thier demands for free.
If I want to 'deal with customers', I can just go to work, sit in my cubicle, and get paid to do it.
Free Software is not "Big Business" and I hope it doesn't become so, because then would start to look like my day job.
If money is made, fine. If users get good software, even better. But IMHO those are incidental, ancillary, indirect benefits. They may be good measurements of successful software, but they are not the *driving force* of free software. The driving force of free software is the software developer, and him creating the things he's interesting in creating.
The people actually creating all this free softare are mostly doing it for free, for Pete's sake. I haven't paid one red cent for linux, mozilla, scribus, evolution, gimp, gaim, vim, cvs, mysql, XMMS, apache, perl, bash, gcc, or any one of the huge array of software I have on my multiple systems. Nothing. I've gotten it all for free, thanks to the kindness and generosity of probably thousands of people.
They even help me out with problems from time to time, through email or support forums -- for free!
To me, it's humbling. As a *user*, I may be frustrated due to some bugs or incomplete documentation in a software package, but I really have no right to complain, unless I write a big fat check to pay for want I want, to make demands on others and expect to have those demands addressed.
Again, I'm not complaining, or thumbing my nose at users. I'm much more a user of Free Software than a creator of it.
If your roomba breaks, while it's getting fixed you bite the bullet and sweep the damn floor, just like you'd do if you vacuum quits tommorrow. If your fancy dish washing & collector robot breaks, you scrub the dishes in the sink - just like you'd do if you dishwasher died.
But brooms and vacuums and sinks and dishwashers are all designed to be used by humans, although some are much more efficient than others.
With home automation built around the idea that every task has a specialized tools not designed to be used by humanoids, and they break, humans can't easily step in and do the work without an alternate set of tools.
I'm trying to come up with an example to make my point, and realize its difficult becuase we don't have any reference of tools in the home that are *not* designed to be used by humans. I'll use something from outside the home: automobile manufacture.
I'm not intimately familiar with the process, but I've seen videos of these long automated assembly lines, full of non-humanoid robots, with wide swinging arms, and welding torches, etc. It's the only example I can think of because mass manufacturing has a longer history with robotic technology than household goods.
If some theoretical breakdown caused all those auto manufacturing robots to suddenly stop working, the factory would have to idle because the working environment is not designed with the assumption of humanoid operator or presence.
But if an giant electromagnetic pulse fried the circuitry of a fleet of humanoid robots that used dumb tools to build the cars, then humans could step in, use the dumb tools themselves, and fill the void immediately.
I realize the analogy isn't all that great, but if you reduce the example down to the much smaller scale of a home environment, I hope people can see what I mean.
I think when the day of robotic domestic servants finally arrives, their market will be driving by cost, not the fact that you can fill in for them when they break
I hope it is at least a consideration. People keep flashlights and batteries on hand for a reason -- things break.
House robots need to do just a few useful things..
Get x
Put x
Get paper from curb
Get beer from fridge
Get book from shelf
Get remote from table
Get phone from cradle
Put item into the trash
Put toy in child's room
Turn on/off light
Dogs can be trained to do this stuff. Why not robots?
Of course, it has to be able to keep itself charged.
Robots will become even more useful and desirable when they can start doing particalar tasks:
Wash dishes
Take out trash
Scrub toilet
Change cat box
Vacuum floor
Do laundry
Don't talk to me about Roomba. I'm talking about a _generally_ smart humanoid robots that is capable of using other dumb machines to accomplish a task. I wish IBM would spend research dollars on this rather than research how to play a better game of chess. (Not that don't like chess!)
This allows for two things:
1) If the dumbs are designed to be used by a human, then the work can done interchangebly by human or robot. Robots will be ultimately replaceable.
Suppose you have a really competent roobmba, that keeps the floor nice and clean. So much so that you no longer have a vaccuum, becuase the roomba is the tool for the job. When the roomba breaks, you are sol.
If you have a dedicated (potentially non-humanoid) robot for each dedicated task, we beging to lose control of our environment and become dependant on the robots.
But an intelligent humanoid robot can step in right now and start using the tool already available to to any number of tasks.
If the robot breaks, humans can step and clean the bathroom with the same tools the robots has been using. Or, you will only need a single set of redundant intelligence in case of failure.
2) By keeping the intelligence (and the expense of intelligence) in a humanoid form, we gain a lot by allowing the peripheral tools to remain dumb -- and therefore cheap and "the same as it has always been."
The Robotic Age will not look much different than the age we are in now. All of the same stuff will be in place and work the same way. It's just that there will this additional robot that does some of the work.
As voice recognition becomes more tenable, it would be nice to put that complexity in one place for consumers. Instead of having the microwave, tv, a/c and lighting system each having their own voice recognition system -- ("TV, turn off." and "AC, set temperature to 74 degrees.") We can have a single system in a robot that can respond to our voice commands and operate all the existing dumb systems in our current households.
In my imagination, robots like what I'm envisioning above will be significant purchases for households, on the order of a vehicle purchase. They would be financed. You would have one or two per house. They would be insured.
ESR specifically said he _wasn't_ bashing CUPS, per se. He was bashing poor usability as endemic of the open source community. Whether Fedora or CUPS is responsible to the bad usabilty he used as the example (or flashpoint of the rant) is beside the point.
I absolutely agree with him that this is the biggest remaining obstacle OSS must overcome before it has a chance of being accepted by the wider mainstream population. This really is something MS does better, and the OSS community would benefit greatly by listening to ESR on this.
The sticking point, though, is that creating a truly good user interface is hard and requires a different set of skills (and possibly training) that what joe coder has -- or possibly even interested in.
Lots of OSS projects have poor usability and limited documentation because documention and interface design requires a person to do something other than coding. Documenting is tedious, no matter how useful it is. Usability testing and interface design, IMHO, is not much better.
Developing a good interface requires exactly what ESR said: exert the mental effort to forget what you know and sit down at the system like a novice user who's never seen it before -- and watch a novice user in action!
In commercial software, this is expensive research. It's not something volunteer coders scratching itches on thier own pet projects want to do.
Fedora Core 2 is running the 2.6 kernel. I hear from first impressions that the performance improvements are nothing short of stunning, but I haven't tried it yet.
But should be still be using Google?
on
Google's Bigger Index
·
· Score: 1, Interesting
Until Linux is a complete entertainment package as well as a utility package, Linux will be hard pressed to take over the desktop.
[...] Game development will have to undergo some pretty radical changes before it will fit successfully into the OSS model and we continue to have the quality of games we have today.
BZFlag rocks, if you aren't familiar with it. Check it out asap.
Slashdot Mirror
Both bz2 files extract to the working dir..
[,,,]
You should always use:
tar xvjf foobar.tar.bz -C $extraction_dir
To prevent from this sorta thing
Useful only if you know IN ADVANCE that it dumps to cwd. Unless you do it as sop, which will give you a possibly confusing directory structure.
At the rate that Microsoft is applying for patents, I can imagine Microsoft being in a position like SCO--except with evidence on Microsoft's side.
I fully agree. I hope there will be a Linux Distro that guarantees 100% MS/Mono-free code. I can't believe we are inviting the vampire into our home.
What's next - Lucasfilm is cancelling Episode III!!?!?
We can only hope.
They would put a tape recorder up to the radio and capture the latest songs, then make copies for their friends.
And we liked it, too! We didn't need no new fandangled Internet to share our music. It was good enough for us, and it should be good enough for you.
Get a haircut!
MySQL has other problems as well. For example, if I want a column to be NOT NULL, I want any code that tries to insert a NULL into that column to fail. I don't want the engine to try to pick some default value for me. If I wanted a default, I would've added a default. That's why default constraints exist. By that same token, if I want a column to allow NULLs, I want to be able to put a NULL in the column. I don't want the current date/time instead of a NULL. If I define a column as auto-incrementing, I want to get an error if I try to insert something into that column. I don't want it to quietly succeed.
At my workplace, we've recently started using MySQL much more often, and have stumbled onto these "gotchas" you mention. Each time we hit a new one, someone spends up to several hours or more, trying to determine if it's our problem or mysql's problem. The natural inclination is to ask "what am I doing wrong?" and spend unneeded time staring at our code.
So, thanks for the the "gotchas" link. That will save us some time in the future, I'm sure.
Where's the damn the expansion slot? I don't care if Best Buy screws me on the rebate, I want an upgrade!
You have no way of knowing that it will have "none of the drawbacks".
Well, know GMail won't have at least one of the drawback's of HotMail -- it won't be owned by MicroSoft.
There are those who try to learn what their customers want, and deliver it.
If this were a business relationship, I'd agree. In a business, if you *didn't* do what you describe, you'd go out of business.
But with Free (or free) software, there is no business relationship. No money exchanges hands. Users are not "customers" or "consumers" because they didn't give any money to the developers (or the mozilla.org organization -- with rare exception).
As such, developers are fully in their rights to blow off stupid comments.
And I say this *not* from a conceited or condescending developer point of view. I'm not a l337 developer.
But in my day development job, I do deal with customers from time to time. These people have contracts with us, they pay us money, and they expect us deliver X by Y date. Of course X, changes, and Y gets moved up.
In my humble opinion, a lot of X is stupid, but the customer wants it, so the customer gets it. I'm helpful. I'm nice. I make every effort to deliver.
A lot of my job in meeting the customers' demands consists of doing things that are decidedly not fun or interesting: writing documentation, creating flowcharts, dummying up boiler plate examples, etc. (We're a small shop and wear many hats).
We have to spend a good amount of time hand-holding our customers, who can be quite demanding. When some bug report can't be tracked down by our customer service or support folk, developers get pulled to investigate.
And, of course, there's the meetings. Oh God, the meetings.
I'm not complaining here, just observing. I know there are a lot of skilled developers who would love to have a good job right now, and I am truly thankful to be working. But that doesn't mean it's not frustrating at time.
My "development job" is a lot more than just writing code. But I like to write code, not all that other stuff.
So, at nights and on the weekends (Well, not recently. I've got a kid, soon to have another.) I would actually write code on little side projects. Not terribly useful to anyone besides myself. This is why I'm not a l337 developer. But I do it just because I want to, because I enjoy it. I enjoy writing things that do things.
This is what Free Software is about. Freedom of the developers. Users are nice, even desirable, but they are not customers and can make no demands on my time beyond what I'm already freely giving. I won't deride them, but I'm certainly under no obligation to meet thier demands for free.
If I want to 'deal with customers', I can just go to work, sit in my cubicle, and get paid to do it.
Free Software is not "Big Business" and I hope it doesn't become so, because then would start to look like my day job.
If money is made, fine. If users get good software, even better. But IMHO those are incidental, ancillary, indirect benefits. They may be good measurements of successful software, but they are not the *driving force* of free software. The driving force of free software is the software developer, and him creating the things he's interesting in creating.
The people actually creating all this free softare are mostly doing it for free, for Pete's sake. I haven't paid one red cent for linux, mozilla, scribus, evolution, gimp, gaim, vim, cvs, mysql, XMMS, apache, perl, bash, gcc, or any one of the huge array of software I have on my multiple systems. Nothing. I've gotten it all for free, thanks to the kindness and generosity of probably thousands of people.
They even help me out with problems from time to time, through email or support forums -- for free!
To me, it's humbling. As a *user*, I may be frustrated due to some bugs or incomplete documentation in a software package, but I really have no right to complain, unless I write a big fat check to pay for want I want, to make demands on others and expect to have those demands addressed.
Again, I'm not complaining, or thumbing my nose at users. I'm much more a user of Free Software than a creator of it.
When will the satellite image be taken? I want to do some nude sunbathing in the backyard when it happens.
Don't bother. There wouldn't be anything interesting to see, anyway.
Bad-dum, Bum!
So quit the microsoft bashing.
You're new here, aren't you? This is slashdot. "News for Nerds. Stuff that matters. Bash Microsoft."
Asimov possibly explained it more eloquently than you.
Not possibly -- he did by far! I could not remember where I first read the idea. I read those probably 15 years ago. Thanks for reminding me.
Firefly appeals to the libertarian in all of us.
Firefly was one of the few shows I've watched that I really like and made it a point to watch.
Agreed. If Firefly returned as a series, I just might turn my tv back on.
If your roomba breaks, while it's getting fixed you bite the bullet and sweep the damn floor, just like you'd do if you vacuum quits tommorrow. If your fancy dish washing & collector robot breaks, you scrub the dishes in the sink - just like you'd do if you dishwasher died.
But brooms and vacuums and sinks and dishwashers are all designed to be used by humans, although some are much more efficient than others.
With home automation built around the idea that every task has a specialized tools not designed to be used by humanoids, and they break, humans can't easily step in and do the work without an alternate set of tools.
I'm trying to come up with an example to make my point, and realize its difficult becuase we don't have any reference of tools in the home that are *not* designed to be used by humans. I'll use something from outside the home: automobile manufacture.
I'm not intimately familiar with the process, but I've seen videos of these long automated assembly lines, full of non-humanoid robots, with wide swinging arms, and welding torches, etc. It's the only example I can think of because mass manufacturing has a longer history with robotic technology than household goods.
If some theoretical breakdown caused all those auto manufacturing robots to suddenly stop working, the factory would have to idle because the working environment is not designed with the assumption of humanoid operator or presence.
But if an giant electromagnetic pulse fried the circuitry of a fleet of humanoid robots that used dumb tools to build the cars, then humans could step in, use the dumb tools themselves, and fill the void immediately.
I realize the analogy isn't all that great, but if you reduce the example down to the much smaller scale of a home environment, I hope people can see what I mean.
I think when the day of robotic domestic servants finally arrives, their market will be driving by cost, not the fact that you can fill in for them when they break
I hope it is at least a consideration. People keep flashlights and batteries on hand for a reason -- things break.
House robots need to do just a few useful things..
Get x
Put x
Get paper from curb
Get beer from fridge
Get book from shelf
Get remote from table
Get phone from cradle
Put item into the trash
Put toy in child's room
Turn on/off light
Dogs can be trained to do this stuff. Why not robots?
Of course, it has to be able to keep itself charged.
Robots will become even more useful and desirable when they can start doing particalar tasks:
Wash dishes
Take out trash
Scrub toilet
Change cat box
Vacuum floor
Do laundry
Don't talk to me about Roomba. I'm talking about a _generally_ smart humanoid robots that is capable of using other dumb machines to accomplish a task. I wish IBM would spend research dollars on this rather than research how to play a better game of chess. (Not that don't like chess!)
This allows for two things:
1) If the dumbs are designed to be used by a human, then the work can done interchangebly by human or robot. Robots will be ultimately replaceable.
Suppose you have a really competent roobmba, that keeps the floor nice and clean. So much so that you no longer have a vaccuum, becuase the roomba is the tool for the job. When the roomba breaks, you are sol.
If you have a dedicated (potentially non-humanoid) robot for each dedicated task, we beging to lose control of our environment and become dependant on the robots.
But an intelligent humanoid robot can step in right now and start using the tool already available to to any number of tasks.
If the robot breaks, humans can step and clean the bathroom with the same tools the robots has been using. Or, you will only need a single set of redundant intelligence in case of failure.
2) By keeping the intelligence (and the expense of intelligence) in a humanoid form, we gain a lot by allowing the peripheral tools to remain dumb -- and therefore cheap and "the same as it has always been."
The Robotic Age will not look much different than the age we are in now. All of the same stuff will be in place and work the same way. It's just that there will this additional robot that does some of the work.
As voice recognition becomes more tenable, it would be nice to put that complexity in one place for consumers. Instead of having the microwave, tv, a/c and lighting system each having their own voice recognition system -- ("TV, turn off." and "AC, set temperature to 74 degrees.") We can have a single system in a robot that can respond to our voice commands and operate all the existing dumb systems in our current households.
In my imagination, robots like what I'm envisioning above will be significant purchases for households, on the order of a vehicle purchase. They would be financed. You would have one or two per house. They would be insured.
ESR specifically said he _wasn't_ bashing CUPS, per se. He was bashing poor usability as endemic of the open source community. Whether Fedora or CUPS is responsible to the bad usabilty he used as the example (or flashpoint of the rant) is beside the point.
I absolutely agree with him that this is the biggest remaining obstacle OSS must overcome before it has a chance of being accepted by the wider mainstream population. This really is something MS does better, and the OSS community would benefit greatly by listening to ESR on this.
The sticking point, though, is that creating a truly good user interface is hard and requires a different set of skills (and possibly training) that what joe coder has -- or possibly even interested in.
Lots of OSS projects have poor usability and limited documentation because documention and interface design requires a person to do something other than coding. Documenting is tedious, no matter how useful it is. Usability testing and interface design, IMHO, is not much better.
Developing a good interface requires exactly what ESR said: exert the mental effort to forget what you know and sit down at the system like a novice user who's never seen it before -- and watch a novice user in action!
In commercial software, this is expensive research. It's not something volunteer coders scratching itches on thier own pet projects want to do.
What will it be called?
gunix
goonix
googlenix
It will definitely run the 2.6 kernel. Where can I get the ISOs?
Fedora Core 2 is running the 2.6 kernel. I hear from first impressions that the performance
improvements are nothing short of stunning, but I haven't tried it yet.
Is Google becoming a task master for Big Brother?
Et tu, Google?
I think it may be worth pointing out that just because someone disagrees with your over-zealous point of view, does not make them a 'troll'.
You're on Slashdot, bud. Here, that is a troll.
I did the same. If you're getting a complaint about libstdc++.so.3 then do this:
/usr/lib/libstdc++.so.5.0.5 /usr/lib/libstdc++.so.3
ln -s
That was the problem, and after making that link, I am able to start nvu. Thanks!
Mod parent up!
I just downloaded it to my Fedora Core 1 machine, and couldn't getting it to run.
NOTE: The tarball unpacks into the current directory! It doesn't create a subdirectory for itself!
There is no 'install' script at all. Untarring is the installation, so put it where you want.
When I run it, it fails and reports that I'm missing two libraries, one I do have and one I don't.
I looked at the start script and it's got a Mozilla 1.7a path hardcoded in there. I'm running Mozilla 1.6, so that's probably part of the problem.
The script also seems to expect an installation of the MRE (Mozilla Runtime Environment) Is this it?
There's no README file, no LICENSE file, no docs of any kind. It'd be nice to have the dependencies identified.
Anyone else getting it running on Fedora Core 1? With or without Mozilla 1.7a? With or without the MRE/GRE?
Still hoping for the best...
I'm only getting 14.5kb/sec here. Please stop trying to download the code until I'm done.
bzFlag is just another one of those multiplayer shoot everything that moves game.
Yes, exactly. Isn't it great?
Until Linux is a complete entertainment package as well as a utility package, Linux will be hard pressed to take over the desktop.
[...] Game development will have to undergo some pretty radical changes before it will fit successfully into the OSS model and we continue to have the quality of games we have today.
BZFlag rocks, if you aren't familiar with it. Check it out asap.