Re:Dead? It depends on your point of view
on
Motif's Not Dead
·
· Score: 1
Actually, software doesn't develop faster if it is done closed or open source. Too many factors go into this to make such a blanket statement. And while the CatBaz paper is a good read it is hardly conclusive proof of anything. You can probably make a good case for many types of applications but not all. It's just silly to evangelize it like it's the proof that open source does X Y and Z better than closed source. And every programmer knows that throwing more manpower/manhours at a project does not necessarily make the project progress any faster, often it can do the reverse (see Brook, Fred "The Mythical Man Month" and all the relates SE stuff).
I'm glad to see that the Linux community doesn't pull any punches even when the best and brightest members screw up. Gives me a little more faith in the public stances that they take.
""For 3dfx AMD K-6 3dNow! users, go into your config.cfg and daikatana.cfg files and change the line that says 'set gl_driver "opengl"' to 'set gl_driver "amd3dgl"'. You'll get DAMN GOOD performance. The game is running sweet here with a VooDoo 2 and AMD K6-2 450. Users without AMD, but with 3dfx chipsets, might want to use 'set gl_driver "3dfxgl"'. ""
My god.. you still ahve to do archaic Bullshit like this? I'm glad to see that some people are even further behind than Linux in user oriented design.
Re:Performance Hit (gaming)
on
WinDSL Coming?
·
· Score: 1
Oh, I think Valve and Cisco are working on the gaming side of the problem with their PowerPlay protocols. Don't have a link handy but I read about it in the recent PCGamer magazine issue.
Remember that the issue with bandwdth and latency is very different for Gamers and Downloaders. Gamers only really need about 2-4kbps of bandwidth for most games. What they really require is a low latency connection with minimal loss. Downloaders need a huge pipe with decent latency and tolerant levels of loss.
I'm sure you know this but it deserves repeating all the time.
I know it's not as sexy... but why do we seem to want to throw all our marbles into one basket (Mars). Wh not go to the moon and setup a base there first? While common sense says to take things one step at a time (especially when lives are at stake) everything I have ever read or seen about the mission to mars completely ignores any possibility of first setting up on teh Moon. Is there a reason (besides PR) for this?
People are in the mindset of lastname identity. Heh, the same last name you share with countless relatives and strangers.
There should be a new internet domain hierarchy for family domains, personal web pages, etc (didn't this happen a while ago?). That should fix part of the problem.
I'm really surprised people haven't thought about fixing the problem instead of the symptoms. The web lends itself to hierarchical organization and this isn't taken advantage of as much as it should.
Now for the people who use some big companies name for their website and run some sort of commercial venture... you're on your own. It's part of business to research before you step into possible legal troubles. Counterpoint: you're a big enough company to go after a domain name withlegal muscle... how come you didn't do that a long time ago? You missed the boat.
That said, I also would like to see the littles guys treated as equally as the big companies.
I'd really like to see things that MS avoided on purpose.
Linux - can read any multitude of existing file systems.
NT - well, uh, can I interest you in an NT file server and NT worskstation solution? You really don't need to talk to OTHER computers with foreign file systems... networking is not where it's at... other computers are danggerous (look at the G4, TANKS!) and full of securtiy holes... and they don't have any cool features... look here.. Solitaire while you wait for the next patch that will fix this very small insignicant security hole... no the Blue Screen is completly normal, just hit the reboot switch and let the system take 15 minutes to reboot.
Before going off on a knee jerk, "No it isn;'t so you lying cheap bastards!" reaction(I'm sure it's too late), you need to really look at the points listed. Some are factual(go ahead, dig down andadmit it) (or based on recently obsoleted facts), others true but twisted to have a very MS bent, and others are downright lies.
Perhaps it would be best for a very nice point to point bulleted feature list to be published and linked to. Have one listing of NT features and another of Linux.
Keep it up to date and be honest. Linux does have shortcommings, but show them that the shortcommings are NOT from the user community and that they are being rapidly addressed.
It seems they took the NT mentality and applied that to Linux. Pretty braindead thinking. Are IT sysadmins as totally clueless?
Any remotely experienced Linux user should be used to the constant patching that they 'get' to do when bugfixes are found, exploits secured, upgrades released, etc. ZDNet, however, assumed that all IT managers think like those trained in Windows. "If it's not called a 'service pack' then it's not important enough to install it" Bleh.
While I can see it being annoying to have to apply 21 patches, a person responsible for servers and security is PAID to do so. What do you think would happen if your corporate web server was hacked, and then you told the management that you didn't apply the fixes because it was inconvenient?!?
And honestly, if you're going to stick a total cluebie on a Linux or any *IX box, you're asking for trouble. It's still not for newbies, and someone trying to do something important in Linux should have a better clue than ZDNet assumes.
I'd think if ZDnet were an IT department it would have been 'let go' a while ago. Perhaps they should assume a little more professional pride and admit their foulups. Oh wait, people who get caught with their pants down make desparate excuses.
IF however, ZDNet is correcnt, and most Linux server admins are clueless gimboids then corporate management might do well to check in and see what security measures/patches etc. are being practiced in their organization.
What I find sad is that most of the CS majors(programmers) I have met through college ( a CS top 10 university) have so little experience with practical programming. They know next to nothing about software design, coding techniques, documentation, development cycles, etc. At least at my university there is little focus on programming style or project management.
I've found that books like Code Complete, Writing Solid Code, Effective C++ absolutely invaluable in shaping my view of programming. The problem, as I see it, is that bad habits form early and programmers, being the stubborn bastards that we are, are loath to change habits on account of someone else's advice. Mostly we seem to have to actually get bit by the problem before we wake up and take action.
Good programming practice is boring and requires intense discipline. BUT IT'S WORTH IT. I'm a believer. I've done it the hard way and the boring way and the boring way was magnitudes easier and quicker once the boring(spec'ing, documentation, design) work was completed.
There are exceptions but that's exactly the point, isn't it? It is the exception when someone (or some group) can hack code together that's on-time and robust without going through the boring stuff.
Users are the problem? Yeah, when they're brainwashed into it. I'll concede that the majority are probably tickled pink with new features and fancy graphics and slick animation... which wears off after the 5th crash in an hour. You pay for a lot of this software, complain complain complain.
I have some suggestions which have differing degrees of merit/practicality. And I'm sure they've been brought up before in other books and papers and articles.
- Write programs with the intention of porting. This forces many potential bugs that are hidden in one platform/compiler to rear their ugly heads.
- Have a more 'open' development process. For gaming companies this means that once you've achieved a certain level of completeness put some of the internal progress online for your sonsumers to view and track. This is already popular with.plan files and such but I've noticed some really nice touches like publishing internal "bug fix of the day" or inviting people to the company office to test and give feedback.
Obviously the OSS method invites peer review/monitoring. Big plus there.
- Complain to your university about it's curriculum/focus. Maybe you won't get any benefit from it immediately.. but where do you think your future co-workers will come from? You'll probably get some BS about "we're not here to train programmers we're here to educate computer scientists." (I can't begin to say what's wrong with that). Since CS majors, from what I've seen at job fairs and the like, are the single greatest pool of new programmers, it seems to make sense that you'd train them as PROGRAMMERS as well as computer scientists. Suggest a alternate curriculum to focus in more specific programming topics like specific languages, design, documentation, cross-platform dev, networking etc.
btw: Is there a program 'quality' standard or award? I'm not aware of any that applies to general computer software. Nor of any general download/ewview site that rates programs on someothing other than features. I am reminded of the console gaming market where crashes in titles seem a rarity. I think I rememebr one Nintendo (8-bit) game that crashed. (Granted that, compared to the PC market, they have very homogenous hardware constraints.)
Apple used to be very anal about who got the 'seal of approval' as well.
Anyways, I'm rambling. I hope I've made some sense.
Win95 and 98 are pretty easy to install, as long as you know what to do when a problem comes up. Stuff like unkown hardware is annoying and the mandatory endless rebooting gives you time to twiddle your thumbs.
NT4.0 Workstation was about the longest install I've ever experienced. My particular version insisted on copying everytigng to the HD from CD and then making 3 boot disks. Then it failed to initialize properly when restarted... Wooo... 4 hours after starting and deleteng/copying files - it finally installed itself. I didn't have any problems with my ISA devices; I count myself lucky. I don't want to _ever_ do an NT install again.
RH 5.x/6.0 was closer to 98/95 than NT. Of course I had to figure out how fdisk/Disk Druid worked... destroyed a few partitions before I figured it out. I've used DOS fdisk before plenty but for a newbie Linux partitioning has a long way to go.
SuSE6.1 - this one was rather confusing with the organization of the installation. Menu based is ncie to be able to go back. But I almost felt lost as I wasn't always sure if I'd done all the right steps. Partitioning was aboutas confusing as RH. Not too hard since I wasn't trying to protect any partitions.
BeOS 4.0/4.5 - Partition Magic b0rked on me for some reason so I went ahead and used the built in partitioning tool... Wow, this is _by far_ the best tool for partitioning that I've ever seen. Nice GUI with sliders and dropdown boxes for partition type... I recommend trying it just for the experience. Oh, installation takes about 4 clicks 10 minutes and a reboot.
I'd add that partitioning in any OS is 'dificult' to do safely and pretty much near impossible for a new user. However, if Linux or any other non MS OS wants to achieve more desktop popularity it's going to have to address this. So what if it's difficult in NT or 98? It's _already_ on the machine! You get to start the game down by a point (or 10).
That and monitor/video card selection are going to have to improve under Linux. Sure sure.. gripe about the OEM vendors... that doesn't change things.
WinLinux seems interesting, I think I'll try that next.
Slashdot Mirror
Actually, software doesn't develop faster if it is done closed or open source. Too many factors go into this to make such a blanket statement. And while the CatBaz paper is a good read it is hardly conclusive proof of anything. You can probably make a good case for many types of applications but not all. It's just silly to evangelize it like it's the proof that open source does X Y and Z better than closed source. And every programmer knows that throwing more manpower/manhours at a project does not necessarily make the project progress any faster, often it can do the reverse (see Brook, Fred "The Mythical Man Month" and all the relates SE stuff).
Ironic that if it is the next NeXT that this is the result of yet another Apple splinter group.
Quality design and product, user centered orientation, cutting edge technology... smacked around by 'inferior' but larger competitors.
I'm glad to see that the Linux community doesn't pull any punches even when the best and brightest members screw up. Gives me a little more faith in the public stances that they take.
""For 3dfx AMD K-6 3dNow! users, go into your config.cfg and daikatana.cfg files and change the line that says 'set gl_driver
"opengl"' to 'set gl_driver "amd3dgl"'. You'll get DAMN GOOD performance. The game is running sweet here with a
VooDoo 2 and AMD K6-2 450. Users without AMD, but with 3dfx chipsets, might want to use 'set gl_driver "3dfxgl"'. ""
My god.. you still ahve to do archaic Bullshit like this? I'm glad to see that some people are even further behind than Linux in user oriented design.
Oh, I think Valve and Cisco are working on the gaming side of the problem with their PowerPlay protocols. Don't have a link handy but I read about it in the recent PCGamer magazine issue.
Remember that the issue with bandwdth and latency is very different for Gamers and Downloaders. Gamers only really need about 2-4kbps of bandwidth for most games. What they really require is a low latency connection with minimal loss. Downloaders need a huge pipe with decent latency and tolerant levels of loss.
I'm sure you know this but it deserves repeating all the time.
I know it's not as sexy... but why do we seem to want to throw all our marbles into one basket (Mars). Wh not go to the moon and setup a base there first? While common sense says to take things one step at a time (especially when lives are at stake) everything I have ever read or seen about the mission to mars completely ignores any possibility of first setting up on teh Moon. Is there a reason (besides PR) for this?
Alright! Now maybe I can actually afford some decent PC-100 sdram. Aren't we soooo glad!!!
I hate these 'technology switchover's It's nice to see us getting better tech, but it's so inconvenient.
Great.. Now I'll just sit tight and watch the chip prices fall.
So basically they want pez.com to be the only site that comes up for a hit on a web search. Someone's looking to lose some friends.
(although I wish is was a bit easier to get to 'official homepages' of products/companies etc.)
Imagine if acting like a moron at Slashdot came back to bite you...
People are in the mindset of lastname identity.
Heh, the same last name you share with countless relatives and strangers.
There should be a new internet domain hierarchy for family domains, personal web pages, etc (didn't this happen a while ago?). That should fix part of the problem.
I'm really surprised people haven't thought about fixing the problem instead of the symptoms. The web lends itself to hierarchical organization and this isn't taken advantage of as much as it should.
Now for the people who use some big companies name for their website and run some sort of commercial venture... you're on your own. It's part of business to research before you step into possible legal troubles. Counterpoint: you're a big enough company to go after a domain name withlegal muscle... how come you didn't do that a long time ago? You missed the boat.
That said, I also would like to see the littles guys treated as equally as the big companies.
(And squatting is a no-no.)
I'd really like to see things that MS avoided on purpose.
Linux - can read any multitude of existing file systems.
NT - well, uh, can I interest you in an NT file server and NT worskstation solution? You really don't need to talk to OTHER computers with foreign file systems... networking is not where it's at... other computers are danggerous (look at the G4, TANKS!) and full of securtiy holes... and they don't have any cool features... look here.. Solitaire while you wait for the next patch that will fix this very small insignicant security hole... no the Blue Screen is completly normal, just hit the reboot switch and let the system take 15 minutes to reboot.
Before going off on a knee jerk, "No it isn;'t so you lying cheap bastards!" reaction(I'm sure it's too late), you need to really look at the points listed. Some are factual(go ahead, dig down andadmit it) (or based on recently obsoleted facts), others true but twisted to have a very MS bent, and others are downright lies.
Perhaps it would be best for a very nice point to point bulleted feature list to be published and linked to. Have one listing of NT features and another of Linux.
Keep it up to date and be honest. Linux does have shortcommings, but show them that the shortcommings are NOT from the user community and that they are being rapidly addressed.
It seems they took the NT mentality and applied that to Linux. Pretty braindead thinking. Are IT sysadmins as totally clueless?
Any remotely experienced Linux user should be used to the constant patching that they 'get' to do when bugfixes are found, exploits secured, upgrades released, etc. ZDNet, however, assumed that all IT managers think like those trained in Windows. "If it's not called a 'service pack' then it's not important enough to install it" Bleh.
While I can see it being annoying to have to apply 21 patches, a person responsible for servers and security is PAID to do so. What do you think would happen if your corporate web server was hacked, and then you told the management that you didn't apply the fixes because it was inconvenient?!?
And honestly, if you're going to stick a total cluebie on a Linux or any *IX box, you're asking for trouble. It's still not for newbies, and someone trying to do something important in Linux should have a better clue than ZDNet assumes.
I'd think if ZDnet were an IT department it would have been 'let go' a while ago. Perhaps they should assume a little more professional pride and admit their foulups. Oh wait, people who get caught with their pants down make desparate excuses.
IF however, ZDNet is correcnt, and most Linux server admins are clueless gimboids then corporate management might do well to check in and see what security measures/patches etc. are being practiced in their organization.
What I find sad is that most of the CS majors(programmers) I have met through college ( a CS top 10 university) have so little experience with practical programming. They know next to nothing about software design, coding techniques, documentation, development cycles, etc. At least at my university there is little focus on programming style or project management.
.plan files and such but I've noticed some really nice touches like publishing internal "bug fix of the day" or inviting people to the company office to test and give feedback.
I've found that books like Code Complete, Writing Solid Code, Effective C++ absolutely invaluable in shaping my view of programming. The problem, as I see it, is that bad habits form early and programmers, being the stubborn bastards that we are, are loath to change habits on account of someone else's advice. Mostly we seem to have to actually get bit by the problem before we wake up and take action.
Good programming practice is boring and requires intense discipline. BUT IT'S WORTH IT. I'm a believer. I've done it the hard way and the boring way and the boring way was magnitudes easier and quicker once the boring(spec'ing, documentation, design) work was completed.
There are exceptions but that's exactly the point, isn't it? It is the exception when someone (or some group) can hack code together that's on-time and robust without going through the boring stuff.
Users are the problem? Yeah, when they're brainwashed into it. I'll concede that the majority are probably tickled pink with new features and fancy graphics and slick animation... which wears off after the 5th crash in an hour. You pay for a lot of this software, complain complain complain.
I have some suggestions which have differing degrees of merit/practicality. And I'm sure they've been brought up before in other books and papers and articles.
- Write programs with the intention of porting. This forces many potential bugs that are hidden in one platform/compiler to rear their ugly heads.
- Have a more 'open' development process. For gaming companies this means that once you've achieved a certain level of completeness put some of the internal progress online for your sonsumers to view and track. This is already popular with
Obviously the OSS method invites peer review/monitoring. Big plus there.
- Complain to your university about it's curriculum/focus. Maybe you won't get any benefit from it immediately.. but where do you think your future co-workers will come from? You'll probably get some BS about "we're not here to train programmers we're here to educate computer scientists." (I can't begin to say what's wrong with that). Since CS majors, from what I've seen at job fairs and the like, are the single greatest pool of new programmers, it seems to make sense that you'd train them as PROGRAMMERS as well as computer scientists. Suggest a alternate curriculum to focus in more specific programming topics like specific languages, design, documentation, cross-platform dev, networking etc.
btw: Is there a program 'quality' standard or award? I'm not aware of any that applies to general computer software. Nor of any general download/ewview site that rates programs on someothing other than features. I am reminded of the console gaming market where crashes in titles seem a rarity. I think I rememebr one Nintendo (8-bit) game that crashed. (Granted that, compared to the PC market, they have very homogenous hardware constraints.)
Apple used to be very anal about who got the 'seal of approval' as well.
Anyways, I'm rambling. I hope I've made some sense.
I second that. My experiences are about parallel.
Win95 and 98 are pretty easy to install, as long as you know what to do when a problem comes up. Stuff like unkown hardware is annoying and the mandatory endless rebooting gives you time to twiddle your thumbs.
NT4.0 Workstation was about the longest install I've ever experienced. My particular version insisted on copying everytigng to the HD from CD and then making 3 boot disks. Then it failed to initialize properly when restarted... Wooo... 4 hours after starting and deleteng/copying files - it finally installed itself. I didn't have any problems with my ISA devices; I count myself lucky. I don't want to _ever_ do an NT install again.
RH 5.x/6.0 was closer to 98/95 than NT. Of course I had to figure out how fdisk/Disk Druid worked... destroyed a few partitions before I figured it out. I've used DOS fdisk before plenty but for a newbie Linux partitioning has a long way to go.
SuSE6.1 - this one was rather confusing with the organization of the installation. Menu based is ncie to be able to go back. But I almost felt lost as I wasn't always sure if I'd done all the right steps. Partitioning was aboutas confusing as RH. Not too hard since I wasn't trying to protect any partitions.
BeOS 4.0/4.5 - Partition Magic b0rked on me for some reason so I went ahead and used the built in partitioning tool... Wow, this is _by far_ the best tool for partitioning that I've ever seen. Nice GUI with sliders and dropdown boxes for partition type... I recommend trying it just for the experience. Oh, installation takes about 4 clicks 10 minutes and a reboot.
I'd add that partitioning in any OS is 'dificult' to do safely and pretty much near impossible for a new user. However, if Linux or any other non MS OS wants to achieve more desktop popularity it's going to have to address this. So what if it's difficult in NT or 98? It's _already_ on the machine! You get to start the game down by a point (or 10).
That and monitor/video card selection are going to have to improve under Linux. Sure sure.. gripe about the OEM vendors... that doesn't change things.
WinLinux seems interesting, I think I'll try that next.