Actually software progress is or atleast should be very measurable. There should be goals and milestones set periodically and if you meet them your on schedule if you miss them then something needs to be corrected (i.e. schedule , man power, etc. ).
I work large contracts all the time and thats how we measure progress.
Yeah but having a cable attached to my laptop isn't nearly as nice as wireless.... trust me sitting on the porch doing a little programming is a beautiful thing.
How about the fact that they are just human. Do you really honestly beleive microsoft engineers let bugs out intentionaly... negligence is along way from intent...
Besides people are going to mistake problems will get through even with months of testing. Its much harder but yes things will get through.
What if I don't want to pay over US$100 each time Apple releases a new version of OSX?
Don't upgrade if you don't want the new features... why do you feel it should be free? They need to make money too...
What if I just don't like Aqua?
Run X instead of aqua... or even better do what I do and run X next to aqua.
What if I like free software? (I know that most of OSX is open, but not all of it.)
You can compile almost everything (beauty of POSIX).
What if I want to develop software that runs on more than one OS?
Non gui stuff is portable. GUI stuff if done right is very easy to port between OS X and gtk. I've even seen projects trying to get gtk to use native aqua widgets (look at the pretty colors...)
I still haven't had a chance to try OSX. It might be wonderful for me, it might not.
I have, love it, wont use anything else on my desktop machine.
I would tend to agree I have some morons for managers but I also know alot of ill equiped IT people. As a developer I want my boxes to be up and running.... I don't want to have to tell IT how to fix it or how to install gcc correctly (and yes I've had to do it).
The point I was making about tinkering is that most of the time I want to sit down and have things work. I've spent many a happy day monkeying with my linux boxes but now with work and social life I don't have time to tinker to make things work. I want to sit down and have everything work perfectly out of the box and linux does not do that like OS X. I'm not saying linux is bad I'm just saying they don't have the hardware integration like Mac's and often don't run as smoothly. I'm not even suggesting it should be written for a particular hardware vendor.
You make it sound like IT departments know their ass from their elbow.... Most of the IT people I know I wouldn't let program my VCR let alone administer my box.
And as for the geek population... most of the time they want to tinker rather than get work done.
Wow... grouping ever one with aids into the "Slut" catagory, your a very accepting person aren't you. Your probably one of the people that says its a HOMO disease as well.
If you have unprotected sex just once your at risk, I don't care if its with a skanky girl or not. The fact of the matter beyond a blood test you can't tell if some is HIV positive or not. So to say yeah I know the person they are fine is just stupid.
Well this POS lime green clam shell is by far and away the best possible solution to the problems I needed solving. What requirements does it meet for me you ask?
1. Durable Laptop: I need something I can throw in a back pack and go. The clam shell design is awesome very rugged, no dongles all the ports built in.
2. Needs to be unix friendly: Not only is it a real unix. Its supported... on top of a very good community (Mac people as the article points out help each other.) I have a company that supports it, writes drivers for my hardware and packages them into a nice upgrader that runs each week.
3. Wireless networking: Built in standard 802.11b with tools to configure the base station built in.
The pluses on top of all this:
DVD support (if I had a superdrive I could burn them)
Office (Some times I get doc's it happens)
Photoshop (I know gimp but... its photoshop)
iTunes (really nice with my iPod.... yeah mac user)
Xserver running as a second desktop next to the quartz.
and yes I really like the way it looks;-)
I could go on but I digrees.... so you see this pos is perfect for me.
Thats the idea... try to layout your design and flow before whacking at it with code. Not enough coders put alot of thought into the design before they start the problem.
The trick is to have the requirements laid out. I know thats not possible some of the time but for the most part you should not be writting production code until most everything is nailed down. That said pathfinding (i.e. writting code to test theory) should be done before sitting down to write the real code.
Actually there are many... many different ways to measure software one of the ones we use is "McCabe Complexity" along side a handful of other metrics. Hell CMU came up with a rating system for software engineering groups called CMM that evaluates your process as well as your process to change your process (defect reduction and what not).
I'm not saying I agree with them but they are out there. I personally feel coding is a craft and not a science... but management doesn't like to hear that because it means results are less reproducable. Thats a whole other can of worms.
God help me if I have to use anything you design....
Its not so much documentation as thinking through a design before diving in. For some small things you can just do documentation and code at the same time but for very large chunks of code that need to work with other you need to lay the design down first before implementation. Its just the way things work. You need to be able to explain you ideas to co-workers/system engineers who ever needs to know.
Coding and engineering are 2 different animals and not enough programmers do the later. If your car failed as much as software does would it be tolerated? Its because of people with a mind set like yourself that throws stuff together and say "there it works"...
And don't even get me started on software testing....
Slashdot Mirror
Actually software progress is or atleast should be very measurable. There should be goals and milestones set periodically and if you meet them your on schedule if you miss them then something needs to be corrected (i.e. schedule , man power, etc. ).
I work large contracts all the time and thats how we measure progress.
2 bedroom condos for under 250K ... which "boston" do you live in?
I think its more of a way to give back to the BSD community rather than an alternitive to Free/Open/Net .
Yeah but I can put my all in one solution on my lap .....
Yeah but having a cable attached to my laptop isn't nearly as nice as wireless .... trust me sitting on the porch doing a little programming is a beautiful thing.
How about going to see live music and buying the album if you like it?
Armchair programming ... slashdot really has died.
I don't think he was really making fun of people but more pointing out what WILL really happen.
Minix .... no I got solaris for X-86 .... no wait BeOS.
... you kids these days with your "crazy" OS's ....
I give up
Mac OS .... superdrive .... mean anything to you?
Not just servers .... Its just like leaving a loaded gun lying around ... after a while something bad is bound to happen.
How about the fact that they are just human. Do you really honestly beleive microsoft engineers let bugs out intentionaly ... negligence is along way from intent ...
Besides people are going to mistake problems will get through even with months of testing. Its much harder but yes things will get through.
Good Question.
... why do you feel it should be free? They need to make money too ...
... or even better do what I do and run X next to aqua.
...)
What if I don't want to pay over US$100 each time Apple releases a new version of OSX?
Don't upgrade if you don't want the new features
What if I just don't like Aqua?
Run X instead of aqua
What if I like free software? (I know that most of OSX is open, but not all of it.)
You can compile almost everything (beauty of POSIX).
What if I want to develop software that runs on more than one OS?
Non gui stuff is portable. GUI stuff if done right is very easy to port between OS X and gtk. I've even seen projects trying to get gtk to use native aqua widgets (look at the pretty colors
I still haven't had a chance to try OSX. It might be wonderful for me, it might not.
I have, love it, wont use anything else on my desktop machine.
I would tend to agree I have some morons for managers but I also know alot of ill equiped IT people. As a developer I want my boxes to be up and running .... I don't want to have to tell IT how to fix it or how to install gcc correctly (and yes I've had to do it).
The point I was making about tinkering is that most of the time I want to sit down and have things work. I've spent many a happy day monkeying with my linux boxes but now with work and social life I don't have time to tinker to make things work. I want to sit down and have everything work perfectly out of the box and linux does not do that like OS X. I'm not saying linux is bad I'm just saying they don't have the hardware integration like Mac's and often don't run as smoothly. I'm not even suggesting it should be written for a particular hardware vendor.
Just my opinion.
You make it sound like IT departments know their ass from their elbow .... Most of the IT people I know I wouldn't let program my VCR let alone administer my box.
... most of the time they want to tinker rather than get work done.
And as for the geek population
Because not every one needs or even wants to ... if you have a fast machine its not going to make that much of a difference.
... then again thats why I own a mac.
I'm a computer nerd but often I'd rather do my work or other things than f*ck around with configs
for $200 I thought they where 733 celerons , 10 gig hd and 128 Meg of ram .... thats pretty cheap ... where can I get better?
Wow ... grouping ever one with aids into the "Slut" catagory, your a very accepting person aren't you. Your probably one of the people that says its a HOMO disease as well.
If you have unprotected sex just once your at risk, I don't care if its with a skanky girl or not. The fact of the matter beyond a blood test you can't tell if some is HIV positive or not. So to say yeah I know the person they are fine is just stupid.
Well this POS lime green clam shell is by far and away the best possible solution to the problems I needed solving. What requirements does it meet for me you ask?
... on top of a very good community (Mac people as the article points out help each other.) I have a company that supports it, writes drivers for my hardware and packages them into a nice upgrader that runs each week.
... its photoshop) .... yeah mac user)
;-)
.... so you see this pos is perfect for me.
1. Durable Laptop: I need something I can throw in a back pack and go. The clam shell design is awesome very rugged, no dongles all the ports built in.
2. Needs to be unix friendly: Not only is it a real unix. Its supported
3. Wireless networking: Built in standard 802.11b with tools to configure the base station built in.
The pluses on top of all this:
DVD support (if I had a superdrive I could burn them)
Office (Some times I get doc's it happens)
Photoshop (I know gimp but
iTunes (really nice with my iPod
Xserver running as a second desktop next to the quartz.
and yes I really like the way it looks
I could go on but I digrees
I'm sorry but if a programmer screws something up it can be just as deadly as building a house with 87 degree joints ....
...
Just because you can hit the delete key doesn't mean it solves the problem
Thats the idea ... try to layout your design and flow before whacking at it with code. Not enough coders put alot of thought into the design before they start the problem.
The trick is to have the requirements laid out. I know thats not possible some of the time but for the most part you should not be writting production code until most everything is nailed down. That said pathfinding (i.e. writting code to test theory) should be done before sitting down to write the real code.
Actually there are many ... many different ways to measure software one of the ones we use is "McCabe Complexity" along side a handful of other metrics. Hell CMU came up with a rating system for software engineering groups called CMM that evaluates your process as well as your process to change your process (defect reduction and what not).
... but management doesn't like to hear that because it means results are less reproducable. Thats a whole other can of worms.
I'm not saying I agree with them but they are out there. I personally feel coding is a craft and not a science
God help me if I have to use anything you design ....
...
....
Its not so much documentation as thinking through a design before diving in. For some small things you can just do documentation and code at the same time but for very large chunks of code that need to work with other you need to lay the design down first before implementation. Its just the way things work. You need to be able to explain you ideas to co-workers/system engineers who ever needs to know.
Coding and engineering are 2 different animals and not enough programmers do the later. If your car failed as much as software does would it be tolerated? Its because of people with a mind set like yourself that throws stuff together and say "there it works"
And don't even get me started on software testing
Oh my god ... you mean an actual process? What a crazy concept "think before you code."
Its called engineering.