In a typical day, going from work, to school, and back home, riding the bus would add 1 1/2 hours to my day.
As it is, the subsidized bus system we have now just makes traffic worse by driving around almost empty, at 10 mph slower than the flow of traffic, making frequent stops right in busy traffic lanes.
True, 802.11b is cheaper, faster and has better range. But I think the point of Bluetooth was never range or lightning fast data transfer, it was designed for PANs. A technology like 802.11b is just not designed to be used with a mobile hands-free. That is why Bluetooth won't die.
Could you just have a very weak 802.11? Something that would work with any 802.11 device if you held it close enough, but use less power.
Whoah, you're right. Included subscription for first 3 years,.99 cents thereafter, and more importantly you can control the ReplayTV from the computer and transfer shows to the computer in DV, looks awesome! Too bad it's $500 for the 40 hour model though. But if they drop that significantly for Christmas I might just chuck out my TV tuner PC card after all.
The purchase price is one thing, but I'm not buying any more monthly subscriptions to anything, period. There's no reason to pay anybody more than $12/year to update an Internet TV-Guide. In fact I think that's about how much the paper TV Guide costs.
Don't most cordless (not cell) phones already use this same unlicensed spectrum? Seems to work well enough.
The article doesn't say how much of the university's voice traffic uses 802.11, vs how much is placed directly on the wired network. I don't see many cordless phones in offices or campuses, so I wonder what is the importance of 802.11 in this setup?
I'm guessing you live in California, where it is a BAD idea to spend $4 million ONCE to save $500,000 a MONTH.
Initial investment aside, there is the question of how much of that $500K/month will be eaten up by running the service in-house. Fortunately the article answers that question: "...expects to cut the costs of internal calls by $21 million over the next decade." So they're netting $175K per month - not $500K, but pretty darn impressive!
No, but there definetly is something wrong with forcing little kids to do it. And 99% of the time they arn't told it's optional. In fact, some kids do get in trouble for not saying it.
This doesn't need to be a big high-noon shootout. Just tell kids they're free to omit portions (or the entire pledge) as they please, and clarify that kids cannot be punished for not saying the Pledge. Problem solved.
because intelligent device managers will send a job to the hard drive that is predicting based on what is in memory what will be asked for next. the hard drive seeks and stores that in the drive cache for quick access to the data...of course, in the case of a failure to predict, it has to seek the drive, but prediction is fairly good.
Like I said, the OS has more information to make an informed guess about what will be accessed next than the disk controller does. The disk controller doesn't even know which blocks constitute a 'file,' and that knowlege alone would improve the accuracy of readahead caching.
also, data sent to the drive for writing has to wait some where while the spindle looks for where to write the data.
For that purpose, the onboard memory is more properly termed a buffer, and it doesn't need to be on the order of megabytes large. Especially since DMA reduces the need for quick response from the OS.
You care a lot if you're capturing a lot of lightly compressed video. That requires a fairly quick drive, I have noticed more dropped frames using Vdub on my 5400 rpm drive than my 7200 drive.
I thought this thread was about the drive's small 2MB onboard cache. How does a small cache hurt sustained write performance?
I would like to know why hard drives should have cache at all.
The OS already does file caching. The OS caches in main memory, which is both much larger than the on-disk cache, and also much quicker to access (over the memory bus instead of the ide/sata/scsi/etc). The OS can use more sophisticated caching algorithms, because it runs on the cpu, not just a little onboard controller. And to top it off the OS has more information about what applications are doing, since the application/OS interface is much richer than the OS/hdd controller interface.
No doubt a cacheless drive would suck at hard drive benchmarks, which intentionally rule out OS caching. But I wonder whether/how much difference on-disk caching really makes on a general WinStone-type benchmark.
But remember, companies like yours were the very last to get on the PC bandwagon. ("We have a mainframe. Use it!") Rigid top-down control is convenient, but there are downsides too.
If I opt for the $50/month subscription and CHOOSE to subscribe twice a year, every SIX months, then I'll pay only $100 and be able to download 600 songs.
Can you even sign up for a month every six months? When I was a member there was a minimum term in the contract to prevent people from doing that.
Most people are not familiar with 95% of artists on EMusic. These are not bands that get radio time (with some exceptions). That means you don't know what you're getting until you download a track or two, and 40 tracks / month ain't much for that. I imagine previews don't count, but 20 seconds of low-quality audio isn't the same - and that's IF there are previews of more than the best 2 or 3 tracks on the album.
I don't know ho emusic was doing before, but it will be interesting to see whether downgrading the service really improves their business.
I hate to say this, as I am the "linux guy" at work, and writing this on mozilla, etc, etc, but X is slower than Windows. Even under e.g. fvwm2 which is what I use. Just scroll up and down in any application, or move a window around. Windows is clearly more snappy.
I think it's partially because the Windows GUI is in the kernel.
It think it's also because X is based on messaging. Even when running locally, XLib serializes gui calls into messages, sends them through a socket, then the X server parses and interprets them. That is going to slow things down.
The right way to do it is to have programs dynamically link to a different implementation of the GUI library, depending on whether they'll be running locally or remotely. Calling into the local version of the library should not construct any "messages," just a trap into protected mode when it comes time to access the hardware. If running remotely, then the gui library is a proxy which constructs and sends messages, like X.
This could all be implemented while preserving Xlib compatibility, it just wouldn't use the X wire protocol. But honestly, who constructs their own X messages instead of using XLib?
As for this idea of using XML, I think it's nuts. We've got to get "parsing" out of the equation, at least when running locally.
Didn't Apple offer new quieter fans for those? I swear I read that you could exchange the fans for an updated quieter one.
You're right! I did a little web searching due to your post and found a whole website dedicated to the problem of noisy Macs. There's a long drawn out saga about people annoyed with it, but if you skip to the last paragraph it says there is (or was) an exchange program but it ended June 30 2003. DOH!
Er, third degree burns are the worst kind - when the skin is burnt straight through down into the deepest layers, usually resulting in charred meat and bone.
Oh, brother. Coffee is mainly water and it can only get so hot before it turns to steam. In fact some people use a special device called a "teapot" which whistles when (and not until) the water boils, when it's as hot as it can get. So anybody who has used a teapot has subjected themselves to the same mortal danger, yet surprisingly most have avoided serious injury.
I totally agree. The time for a (roughly) 5 GB format has come and gone. Needlessly large for documents and spreadsheets, too small for much music and video (at least when you're used to hard drives 20 times as big.)
DVD for data storage is going down as an example of how a format can fail when companies refuse to standardize the format.
What would Dell spend R&D money ON? Comparing Dell to Sun is silly. Sun = Dell + Microsoft + Intel. Dell is just the shipping and receiving department of DellWinTel.
Re:Were the copyright violations fixed? (yet?)
on
Mplayer Revisited
·
· Score: 1
Woudn't mplayer have to dump dvd support to come completely clean? I'd rather keep it the way it is. Not because I'm a crook, but I'd rather copy DVDs to my laptop's harddrive and use the drive bay to hold an extra battery when I fly... so shoot me.
I agree. Government policies that close doors to competition are bad. Linux might work in some situations, but not in others. There are plenty of good software packages out there to use, and plenty of specific packages for government, that wont exist in OSS until someone is paid (gobs of cash) to write them.
At the very least then, you would also argue against businesses, colleges, or govts standardizing on MS "solutions"?
Is it is socialistic and monopolistic for your boss to put Windows on all desktops instead of letting everybody choose their own? That's been the status quo for 10 years now. Personally I don't think it's wrong for organizations to weigh their options and settle on a single choice - in this case, Linux.
A lot of what's in this article is just unsubstantiated blather, like that OSS is more expensive to maintain, or that MA is going back to the dark ages by moving to linux. There really seems to be some bias there.
Slashdot Mirror
As it is, the subsidized bus system we have now just makes traffic worse by driving around almost empty, at 10 mph slower than the flow of traffic, making frequent stops right in busy traffic lanes.
Microsoft's first gen. bluetooth keyboard and mouse went on sale Nov 2002.
Are you under the impression that Apple was first to market with a bluetooth mouse and keyboard? Microsoft released theirs a year ago.
Whoah, you're right. Included subscription for first 3 years, .99 cents thereafter, and more importantly you can control the ReplayTV from the computer and transfer shows to the computer in DV, looks awesome! Too bad it's $500 for the 40 hour model though. But if they drop that significantly for Christmas I might just chuck out my TV tuner PC card after all.
The purchase price is one thing, but I'm not buying any more monthly subscriptions to anything, period. There's no reason to pay anybody more than $12/year to update an Internet TV-Guide. In fact I think that's about how much the paper TV Guide costs.
The article doesn't say how much of the university's voice traffic uses 802.11, vs how much is placed directly on the wired network. I don't see many cordless phones in offices or campuses, so I wonder what is the importance of 802.11 in this setup?
The point wasn't to have new or different services, but to save large truckloads of cash.
The OS already does file caching. The OS caches in main memory, which is both much larger than the on-disk cache, and also much quicker to access (over the memory bus instead of the ide/sata/scsi/etc). The OS can use more sophisticated caching algorithms, because it runs on the cpu, not just a little onboard controller. And to top it off the OS has more information about what applications are doing, since the application/OS interface is much richer than the OS/hdd controller interface.
No doubt a cacheless drive would suck at hard drive benchmarks, which intentionally rule out OS caching. But I wonder whether/how much difference on-disk caching really makes on a general WinStone-type benchmark.
But remember, companies like yours were the very last to get on the PC bandwagon. ("We have a mainframe. Use it!") Rigid top-down control is convenient, but there are downsides too.
I don't know ho emusic was doing before, but it will be interesting to see whether downgrading the service really improves their business.
I think it's partially because the Windows GUI is in the kernel.
It think it's also because X is based on messaging. Even when running locally, XLib serializes gui calls into messages, sends them through a socket, then the X server parses and interprets them. That is going to slow things down.
The right way to do it is to have programs dynamically link to a different implementation of the GUI library, depending on whether they'll be running locally or remotely. Calling into the local version of the library should not construct any "messages," just a trap into protected mode when it comes time to access the hardware. If running remotely, then the gui library is a proxy which constructs and sends messages, like X.
This could all be implemented while preserving Xlib compatibility, it just wouldn't use the X wire protocol. But honestly, who constructs their own X messages instead of using XLib?
As for this idea of using XML, I think it's nuts. We've got to get "parsing" out of the equation, at least when running locally.
I have a nice, new dual G4 powermac sitting in my office with a nice cinerama display, and it never gets used. It's just too loud.
DVD for data storage is going down as an example of how a format can fail when companies refuse to standardize the format.
What would Dell spend R&D money ON? Comparing Dell to Sun is silly. Sun = Dell + Microsoft + Intel. Dell is just the shipping and receiving department of DellWinTel.
Woudn't mplayer have to dump dvd support to come completely clean? I'd rather keep it the way it is. Not because I'm a crook, but I'd rather copy DVDs to my laptop's harddrive and use the drive bay to hold an extra battery when I fly... so shoot me.
Is it is socialistic and monopolistic for your boss to put Windows on all desktops instead of letting everybody choose their own? That's been the status quo for 10 years now. Personally I don't think it's wrong for organizations to weigh their options and settle on a single choice - in this case, Linux.
A lot of what's in this article is just unsubstantiated blather, like that OSS is more expensive to maintain, or that MA is going back to the dark ages by moving to linux. There really seems to be some bias there.